De Vries - 2015 - Sustainability Science

SUSTAINABILITY SCIENCE
Sustainable development is becoming the guiding principle for the 21st cen-
tury. It is about quality of life: how to develop it and how to sustain it within
planetary boundaries. ‘Sustainability science’ has emerged recently as a new
academic discipline and is a growing area of both research and teaching.
Sustainability science seeks to:
r advance basic understanding of the dynamics of human-environment
systems and forge bridges between the natural and social sciences and
between science and policy;
r appreciate the variety of perspectives on sustainable development and
the variety of contexts for its design, implementation, and evaluation in
particular situations.
Bert J. M. de Vries has taught a course on sustainability science at Utrecht
University for many years, in connection to his research at the Netherlands
Environmental Assessment Agency (PBL). This textbook is based on that
course. The contents have been rigorously class-tested by his students. The
book provides a historical introduction into patterns of past (un)sustainable
development and into the emergence of the notion of sustainable develop-
ment. It systematically surveys the key concepts, models and findings of the
various scientific disciplines with respect to the major sustainability issues:
energy, nature, agro-food systems, renewable and non-renewable resource
systems and economic growth. System analysis and modelling are introduced
and used as integrating tools. Stories and worldviews are used throughout
the text to connect the quantitative and the qualitative and to offer the
reader an understanding of relevant trends and events in context. The reader
is explicitly invited to engage at a personal level into the interpretation of
what sustainable development means and what implications this has for
ideas and actions.
Sustainability Science is an ideal textbook for advanced undergraduate-
and graduate-level courses in sustainable development, environmental sci-
ence and policy, ecology, conservation, natural resources and geopolitics.
Bert J. M. de Vries is co-founder of the Institute for Energy and Envir-
onment (IVEM) at the University of Groningen in the Netherlands, where
he received his Ph.D. on sustainable resource use. Since 1990, he has been a
senior scientist at the Netherlands Environmental Assessment Agency (PBL,
formerly MNP and RIVM). He has been actively involved in modelling and
scenario construction for the Intergovernmental Panel on Climate Change
(IPCC). Since 2003, he has also been a Professor of Global Change and
Energy at Utrecht University in the Netherlands. His research expertise
and publications are in resource and energy analysis, modelling and policy;
climate and global change modelling; and complex systems modelling for sus-
tainable development. He has co-edited several books, including Perspectives
on Global Change: The TARGETS Approach (Cambridge University Press,
1997) and Mappae Mundi: Humans and Their Habitats in a Long-Term Socio-
Ecological Perspective (2002).
Downloaded from Cambridge Books Online by IP 132.248.9.8 on Thu Jul 09 23:27:23 BST 2015.
http://ebooks.cambridge.org/ebook.jsf?bid=CBO9780511794469
Cambridge Books Online © Cambridge University Press, 2015
Advance Praise for Sustainability Science
“Achieving some sort of sustainability will be THE focus of global societies in the
21st century. To be successful, our leaders will need a perspective of centuries, the
full breadth of scientific insights, system thinking skills, great cultural sensitivity and
an awareness of spiritual values. All of these are offered in this wonderful, unique
text, which will be useful for decades.”
– Dennis Meadows, co-author of The Limits to Growth
“This textbook is one of the first truly all-encompassing introductions to sustainab-

ility science. It is methodical, clearly written and well-illustrated, truly a pleasure
to handle and to read. It sets a standard for the discipline and solidly educates the
generation of students that will most directly have to deal with the challenges of
creating a sustainable Earth system.”
– Sander van der Leeuw, Dean, School of Sustainability, Arizona State
University
“In this important new book Bert de Vries has adopted a systems approach to
examining all the issues that collectively amount to the determinants of sustainab-
ility. It is an excellent, comprehensive and up-to-date text dealing with not only
the underlying biophysical science but also human behaviour. His use of interest-
ing examples throughout makes it both instructive and enjoyable to read. I highly
recommend it.”
– Brian Walker, CSIRO Ecosystem Sciences, Australia
“Bert de Vries’ Sustainability Science is particularly welcome as it breaks ground

in a new field, which so far lacks a proper systematic treatment. No wonder! The
challenge is overwhelming: the book covers a series of disciplines and fields – geo-
graphy, social and economic sciences, physics, chemistry and biology – using a sys-
tems description and system dynamics as the main tool. De Vries not only succeeds
in this overwhelming task but spices up the text with multiple excursions into his-
tory, philosophy, literature, not to forget the key issue of ethics. Justice and how we
would like a future world to look is always present. Bert’s book is impressive, rich
and inspiring.”
– Lars Rydén, Centre for Sustainable Development, Uppsala University
Sustainability Science
Bert J. M. de Vries
Utrecht University
cambridge university press
Cambridge, New York, Melbourne, Madrid, Cape Town,
Singapore, São Paulo, Delhi, Mexico City
Cambridge University Press
32 Avenue of the Americas, New York, NY 10013-2473, USA
www.cambridge.org
Information on this title: www.cambridge.org/9780521184700
C Bert J. M. de Vries 2013

This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2013
Printed in the United States of America
A catalog record for this publication is available from the British Library.
Library of Congress Cataloging in Publication Data

Vries, Bert de, 1948–
Sustainability science / Bert J. M. de Vries. – 1st ed.
p. cm.
Includes bibliographical references and index.
ISBN 978-1-107-00588-4 (hardback : alk. paper) – ISBN 978-0-521-18470-0
(pbk. : alk. paper) 1. Sustainable development – Research. I. Title.
HC79.E5V735 2012
338.9′ 27072–dc23 2012024935
ISBN 978-1-107-00588-4 Hardback

ISBN 978-0-521-18470-0 Paperback
Additional resources for this publication at www.cambridge.org/devries
Cambridge University Press has no responsibility for the persistence or accuracy of

URLs for external or third-party Internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
“In bytsje minder as men lêst,
Dat bikomt jin fierwei ’t bêst.”
(A little bit less than you crave for,
Makes you enjoy so much more.)
Fries spreekwoord
Contents
Preface page xv
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Roots 1
1.2 Sustainability Science 4
1.3 Sustainable Development Is About Quality of Life 7
1.4 Guidelines for the Reader 9
Appendix 1.1 United Nations Decade of Education for Sustainable
Development 12
2 The System Dynamics Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.1 Introduction 14
2.2 The World Car System 15
2.3 System Dynamics: The Basics 22
2.3.1 What Is a System? 22
2.3.2 Stocks and Flows 24
2.3.3 Feedback Loops 28
2.3.4 An Illustrative Simulation Experiment 33
2.4 System Dynamics Modelling 35
2.4.1 The Rules of the Game 35
2.4.2 Archetypes 40
2.4.3 An Example: Modelling Car and Public Transport Use 43
2.5 Structure, Space and Time 46
2.6 Summary Points 47
Appendix 2.1 Integral-Differential Calculus 50
3 In Search of Sustainability: Past Civilisations . . . . . . . . . . . . . . . . . . . 54

3.1 Introduction 54
3.2 The Beginnings: Two Environmental Tales 56
3.2.1 The Hohokam People in Arizona 56
3.2.2 Easter Island 57
3.3 Emerging Social Complexity: State Formation 60
3.3.1 Early Mesopotamia: Urban Centres and Their Elites 60
vii
viii Contents
3.3.2 Egypt: The Nile and Its Rhythms 61

3.3.3 South Asia: The Indus-Sarasvati Civilisation 63
3.3.4 The Aegean and Mesoamerica: The Role of Ecological
Diversity 64
3.4 Empires 66
3.4.1 The Roman Empire 66
3.4.2 Other Empires: China, India and Russia 72
3.5 Mechanisms, Theories and Models 75
3.5.1 Mechanisms 75
3.5.2 Theories and Models 78
4 The World in the Past 300 Years: The Great Acceleration . . . . . . . . . . 84

4.1 Introduction 84
4.2 The World in the Last Three Centuries 84
4.2.1 Accelerating Growth: Population and Economic Activity 84
4.2.2 Social-Cultural Changes 93
4.3 Accelerating Impacts: The Natural Environment 96
4.3.1 The Source Side 96
4.3.2 The Sink Side 100
4.3.3 Experiencing Change 102
4.4 Earth System Analysis: Regimes and Syndromes 104
4.4.1 Social-Ecological Regimes 104
4.4.2 Syndromes and Archetypes 109
5 Sustainability: Concerns, Definitions, Indicators . . . . . . . . . . . . . . . . 117

5.1 Introduction 117
5.2 Global Change: The Scientific Worldview 117
5.2.1 The Scientific Worldview: Earth 117
5.2.2 The Scientific Worldview: Life 118
5.2.3 The Scientific Worldview: Society 120
5.3 Rising Concerns 122
5.3.1 Early Concerns: Managing Common Inheritance 122
5.3.2 The Environmental Movement 123
5.3.3 Our Common Future? 125
5.4 The Notion of Sustainable Development 129
5.4.1 Prelude: Categories of Goods and Services 129
5.4.2 Interpretations and Definitions 133
5.5 An Indicator Framework for Sustainable Development 136
5.5.1 From Principle to Action: Indicators 136
5.5.2 A Sustainable Development Indicator System (SDIS) 138
5.5.3 Quality-of-Life–Oriented and Aggregate SD-Indicators 139
6 Quality of Life: On Values, Knowledge and Worldviews . . . . . . . . . . 146

Contents ix
6.2 Quality of Life and Values 146

6.2.1 Needs and the Quality of Life 146
6.2.2 Capabilities and Satisfiers 149
6.2.3 Values and Their Measurement 152
6.3 The Cultural Theory 155
6.3.1 Stories from the Himalayas and Bali 155
6.3.2 Four Perspectives 158
6.4 Worldviews: Ways to See the World 164
6.4.1 Values and Beliefs: Four Worldviews 164
6.4.2 Worldviews in History 168
6.4.3 Mechanisms of Social Change 172
6.5 Worldviews in Action 174
7 Energy Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

7.1 Introduction: The Essential Resource 179
7.2 Basic Energy Science: Thermodynamics 184
7.2.1 Classical Thermodynamics: The First and Second Law 184
7.2.2 Energy Quality: The Potential to Do Work 190
7.2.3 Energy Forms 194
7.3 Movement in Space and Time: Mechanics 195
7.4 Stories 201
7.4.1 Fuel-Efficient Stoves for People in Darfur 201
7.4.2 The South Nyı́rség Bioenergy Project 202
7.5 Energy Conversion 203
7.5.1 Elementary Processes 203
7.5.2 The Energy System 206
7.6 Energy Futures 208
Appendix 7.1 The Steady-State Mass-Energy Equation 213
8 On Knowledge and Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

8.2 Models in the Natural Sciences 216
8.2.1 The Scientific Method 216
8.2.2 Models and the Modelling Process 219
8.3 Strong and Weak Knowledge 221
8.4 Complexity 227
8.5 Metamodels and Organising Concepts 232
8.6 Science in the Age of Complexity 238
Appendix 8.1 A Brief History of Complex Systems Science 242
9 Land and Nature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244

9.2 Earth: Soil Climate Vegetation Maps 245
9.3 Stories 253
x Contents
9.3.1 Forest Fires 253

9.3.2 Biodiversity in South America: The Trésor Project 254
9.3.3 Mining in Papua New Guinea 255
9.4 Land Cover Change and Degradation 256
9.5 Ecosystem Dynamics: Population Ecology 262
9.5.1 Population Ecology: Logistic Growth 262
9.5.2 Population Models: Prey-Predator Dynamics 265
9.6 Food Webs: The Stocks and Flows in Ecosystems 272
9.6.1 Food Webs and Their Representations 272
9.6.2 Stability and Resilience 274
9.7 Catastrophic Change in Ecosystems 277
9.8 Biodiversity and Ecosystem Services 283
9.9 Nature and Sustainable Development 288
Appendix 9.1 Prey-Predator Models and Stability Analysis 293
Appendix 9.2 Catastrophic Change and Bifurcations 294
10 Human Populations and Human Behaviour . . . . . . . . . . . . . . . . . . . 296

10.1 Introduction: The Image of Man 296
10.2 Demography: Human Population Dynamics 299
10.2.1 Modelling Population 299
10.2.2 Driving Forces 303
10.3 Evolution: Our Biological Roots 308
10.4 Homo Economicus and Its Critics 312
10.4.1 Consumers and Producers 312
10.4.2 Games, Dilemmas and Cooperation 314
10.5 Simulating Human Behaviour 320
10.5.1 Introduction 320
10.5.2 Cellular Automata Models 321
10.5.3 Interaction: Networks 325
10.5.4 Multi-Agent Simulation: Behavioural Variety 329
Appendix 10.1 Models of Economic Decision Making 334
Appendix 10.2 Replicator Dynamics 336
Appendix 10.3 Network or Graph Theory 336
11 Agro-Food Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

11.1 Introduction: The Human Habitat 339
11.1.1 Land and People 339
11.1.2 Anthromes 341
11.2 Agricultural Systems 344
11.2.1 Agro-Food Systems in the World 344
11.2.2 Food: Needs and Consumption 347
11.2.3 Food: Resources and Potential Supply 348
11.2.4 More Food: Can It Be Supplied Sustainably? 349
11.3 Stories from the Real World 357
11.3.1 Nomads in Mongolia 357
11.3.2 Can and Should Rural France Be Saved? 358
Contents xi
11.4 Land Use and Cover Change 359

11.4.1 Land Use Changes and Its Causes 359
11.4.2 Modelling Land Use and Cover Change 360
11.5 Towards a Global Industrial Agro-Food System 363
11.5.1 Diversity in Transition 363
11.5.2 The Global Agro-Food System 366
11.5.3 Markets, Scale and Innovations As Driving Forces 369
11.6 Perspectives on Food and Agriculture 372
Appendix 11.1 Income and Price Elasticity 377
12 Renewable Resources: Water, Fish and Forest . . . . . . . . . . . . . . . . . 379

12.1 Introduction: Lakeland 379
12.2 Renewable Resources 384
12.2.1 Renewable Resource Use: An Archetypical Model 384
12.2.2 Model Extensions and Management Principles 387
12.3 Water Resources 391
12.3.1 Water Availability and Use 391
12.3.2 Water for Irrigation: A Case Study and a Model 398
12.4 Stories 400
12.4.1 The Canadian Fish Drama 400
12.4.2 European Union Fisheries Policy in Senegal 400
12.5 Fisheries and Forests 401
12.5.1 World Fisheries 401
12.5.2 Fisheries Models: Strategies and Interactions 404
12.5.3 Fisheries Models: Behavioural Variety 408
12.5.4 World Forests 412
12.6 Interactive Modelling for Sustainable Livelihood 414
12.7 Perspectives on Water, Fish and Forest 416
Appendix 12.1 The Simple Population and Renewable Resource
Model 421
Appendix 12.2 Resource Use in the Simple Model 421
Appendix 12.3 Modelling Different Harvesting Strategies 422
Appendix 12.4 The Geonamica Software 423
13 Non-Renewable Resources: The Industrial Economy . . . . . . . . . . . . 425

13.1 Introduction: The Industrial Regime 425
13.2 Non-Renewable Resource Chains: Extraction 427
13.2.1 Biogeochemical Element Cycles 427
13.2.2 Classification 430
13.2.3 Availability, Exploration and Extraction: Two Models 431
13.3 Elementary Resource Economics 436
13.3.1 Supply Cost Curves 436
13.3.2 Innovation: The Learning-By-Doing Mechanism 439
13.3.3 Optimal Depletion: The Resource Curse and Resource
Security 442
13.4 Stories 445
xii Contents
13.4.1 Oil and Power 445

13.4.2 The Promise of Gold: Tambogrande 446
13.5 Resource Chains: Material Use and Efficiency 447
13.5.1 Assessment Methods 447
13.5.2 Dematerialisation: The Intensity-of-Use Hypothesis 449
13.5.3 Richer and Cleaner? 453
13.6 Stories 455
13.6.1 Pearl River Estuary, China 455
13.6.2 Water As a Commodity: Ban on Bottled Water in an
Australian Town 455
13.6.3 Organotin Compounds As Antifouling Agents 456
13.7 The Sink Side: Environment and the Industrial Economy 457
13.7.1 Resource Chains: The Sink Side 457
13.7.2 Enduring Environmental Problems 459
13.7.3 Persistent Chemicals 463
13.8 Perspectives on the Industrial Economy 465
Appendix 13.1 The Crustal Abundance Geostatistical (CAG) Model 469
Appendix 13.2 The Logistic Growth Life Cycle Model 471
14 Towards a Sustainable Economy? . . . . . . . . . . . . . . . . . . . . . . . . . . 472

14.1.1 An Archetypical Model 472
14.1.2 Substitutability, Technology and Optimality 477
14.2 Theories of Economic Growth 479
14.2.1 Classical Theories 479
14.2.2 Economic Growth Theory 481
14.2.3 The Role of Technology, Learning and Behaviour 485
14.3 Source and Sink Constraints in the Economy 492
14.3.1 Structural Economics: The Input-Output Formalism 492
14.3.2 Resource Efficiency and Pollution Abatement: Economic
Mechanisms 496
14.4 Economic Growth and Sustainable Development 501
14.4.1 GDP and the Need for a Better Indicator 501
14.4.2 Beyond Models: Welcome to the Real World 503
Appendix 14.1 A Simple Behaviour Model of Saving 511
Appendix 14.2 Evolutionary Models of Producers and Consumers 511
Appendix 14.3 Input-Output Tables 512
Appendix 14.4 Gross Domestic Product 514
15 Outlook on Futures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 516

15.2 Outlooks 517
15.2.1 Sustainable Futures: Urban, Rural, Global 517
15.2.2 The Scenario Approach 519
15.2.3 Sustainable Development in a Scenario Frame 521
Contents xiii
15.3 Scenarios for a Sustainable World 522

15.3.1 Four Stories 522
15.3.2 Growth within Limits . . . 526
15.3.3 . . . But Is It Sustainable? 529
15.4 An Agenda for Sustainability Science 533
Glossary 537
References 561
Index 585
Preface
This book is the outcome of eight years of teaching the course Sustainability Science
for students of the M.Sc. in Sustainable Development at Utrecht University. Its
aim is to give a broad overview of what the sciences have to say about sustainable
development. To this purpose, it offers a mixture of concepts, theories, models and
facts and an invitation to the student to become a critical, independent thinker and
to act accordingly.
The book can be used at the B.Sc. level as part of an introductory course or at
the M.Sc. level as context for other courses and M.Sc. theses. For most chapters,
a high school or college background should be sufficient, but some capacity for
abstract thinking is needed. As an introduction into the concept of sustainability and
sustainable development issues, it is useful for people in NGOs, government agencies
and business who are interested in framing the discourse from multiple perspectives
and different scientific disciplines. It can help them to make better decisions for life
on a finite planet.
The content is based on three personal convictions. The first one is that human-
ity faces a transition period in which many ideas, habits and expectations will be
challenged and scientists should, therefore, offer an integrated perspective on how
developments are connected and may unfold. I follow others in using the term sus-
tainability science to summarise this effort. The second conviction is that all scientific
disciplines can and should contribute to the content of sustainability science. The
conceptualisation of sustainable development as a guiding principle for the 21st cen-
tury is still fragmented, and this should change. We need a new science, one that
uncovers the unity of science and mobilises understanding and offers context. One
that uses the novel ways to access information (Internet, Wikipedia, etc.) and to
engage in the real world (stories, simulation games, etc.). A third conviction is that
sustainable development can best be defined in terms of quality of life and that the
pluralism in people’s values and ideas about what quality of life is and in their cir-
cumstances should be acknowledged explicitly. It implies the framing of sustainable
development as a global challenge within local diversity, capacity and contingency.
How to realise this ambitious goal is not self-evident. The first part of the
challenge is to find observations, concepts, theories and models that are relevant
in context. I first introduce system dynamics and influence diagrams and simulation
models as an inherently integrating language and toolbox. Next, I summarise history
xv
http://dx.doi.org/10.1017/CBO9780511794469.001
xvi Preface
insofar as it is relevant for understanding sustainable development: rise and decline

of past civilisations, the transition to the industrial era and the emergence of the
idea of sustainable development. Subsequently, I introduce worldviews, as combina-
tions of values and beliefs, in order to communicate and appreciate the diversity in
the notions of sustainable development and quality of life. I offer epistemological
reflections on the nature of scientific knowledge in order to appreciate the inherent
complexity and uncertainty of the issues being dealt with. Some basic natural science,
notably on energy, is introduced in Chapter 7.
In the second part of the book, I discuss the subsystems in which the concerns
about (un)sustainable development are most pressing: natural ecosystems, agro-
food systems, renewable and non-renewable resources and the economy-technology
system. Each chapter covers what I consider important concepts, theories and models
and their link with the observations in the real world. Each chapter has a natural
connection with specific scientific disciplines, but ecology and geography are core
disciplines. There are a few stories in each chapter to illustrate and complement the
theory and the models and invite the student to engage with real-world experiences
and situations. At the end of each chapter, five statements about the subsystem are
offered to the reader in order to practice worldview pluralism. A few quotes are
given to provoke ‘out-of-the-box’ thinking.
There are many different angles from which one can look at development, qual-
ity of life and sustainability. Moreover, the flood of data, theories and models keeps
growing. Therefore, I can and do claim neither completeness nor representativeness.
No doubt, the treatment of some topics is biased because I was trained in physics and
chemistry and because I know certain persons and books and do not know others.
I am aware that the treatment of some topics is incomplete or inadequate. Ethics,
psychology, political science and subdisciplines in geography and economics, for
instance, remain underexposed. Besides my limited knowledge, it reflects the inher-
ent complexity of these disciplines and their still limited input in the sustainability
discourse. Also, engineering details and legal aspects are hardly addressed. As to the
narratives, I do not yet offer a balanced mix of stories. In particular, stories about
the many social, economic and technological initiatives and innovations for a more
sustainable world are missing. I hope that the book is a start for a more comprehen-
sive, web-based ‘atlas for a sustainable world’, in which the latest scientific insights
merge with relevant stories from all over the world and, together, point at the right
actions. The website www.sustainabilityscience.eu is a first, simple start towards this
goal.
This book could, of course, only be written by building on the thoughts, feel-
ings and intuitions of many other people. Some of them are voices from a distant
past or from people I never met, but much of it is easily traced. My promoter Jan
Kommandeur offered the inspiration and courage that helped me to follow my own
path. With my colleagues at the University of Groningen, I spent many years in
the 1970s and 1980s in an atmosphere of great intellectual curiosity and freedom.
In those years, I also joined the Balaton Group, founded by Donella and Dennis
Meadows. Ever since, the annual meetings have been a source of inspiration, friend-
ship and joy. A considerable part of the ideas in this book have taken shape during
exchanges with Balaton Group members. Donella Meadows taught me the virtue
of combining clear systems thinking and compassionate engagement. From Dennis
Preface xvii
Meadows, I learned the usefulness and fun of communicating insights in the form
of games. Aromar Revi nurtured my search for cosmic dimension and wisdom. I
refrain from mentioning the many other members of the group – but they know.
In the 1990s, I had the privilege to work in a rather unique group of Global
Change Modelling pioneers: the TARGETS and IMAGE teams at RIVM (later
MNP and still later PBL – the English name turned out to be more sustainable:
Netherlands Environmental Assessment Agency). I thank Klaas van Egmond and
Fred Langeweg for the opportunities to follow uncharted paths, and the many col-
leagues for the instructive and pleasant collaboration. Another source of inspiration
and insight were the meetings and discussions as part of writing the book Map-
pae Mundi, with Joop Goudsblom, on the 250th anniversary of the Hollandsche
Maatschappij der Wetenschappen. Since 2003, I occupy for one and later two days a
week the chair Global Change and Energy at the Copernicus Institute for Sustain-
able Development and Innovation at Utrecht University. The chair is financed by
PBL, and I am most grateful for the permission I got to spend time on the teaching
and writing that has led to this book.
As to the actual writing of the book, I received useful and constructive input
from many people: Hans Deuss, Tom Fiddaman, Peter Janssen, Eric Lambin, Sander
van der Leeuw, Erik Lysen, Evert Nieuwlaar, Martin Patel, Charles Redman,
Max Rietkerk, Lars Ryden, Mark Sanders, Isak Stoddard, Yoshi Wada and Bob
Wilkinson. I would also like to acknowledge a few persons in particular. The con-
structive and humorous criticisms of Jodi de Greef on anything I wrote about com-
plexity and sustainability have protected me from an overdose of sincerity. The
chapter on worldviews got a major boost from the inspiring discussions with Klaas
van Egmond, my colleague at Utrecht University since 2008. The cooperation with
Markus Brede since 2006 brought an intellectual depth to parts of the book that
would otherwise have been absent. Cristina Apetrei has been of tremendous help
with comments and layout, and her practicality and determination kept me afloat in
times of despair. My brother Joop gave with great precision the right books at the
right moment, my son Tom applied his management skills for me to figure out what
the message of the book is and my daughter Marieke assisted with cover design.
I owe much inspiration to the silence and darkness, the skies and clouds, and the
mountains, trees and humans of the Vallespir. Finalement, Annelize, sans toi j’avais
eu ni le courage ni l’espace de persévérer dans un monde où, comme Proust l’a dit,
‘les forts, . . . ont seuls cette douceur que le vulgaire prend pour de la faiblesse’.
I suffer from the injustice and violence in the world, the destructiveness of the
relentless pursuit of ‘stuff’, the lack of awareness among large numbers of people
and the cynicism and hypocrisy among many members of the powerful and wealthy.
But I also discern and enjoy the germs of rejuvenation, the search for meaning, the
genuine compassion that can be seen all over the world. I am grateful that I had the
opportunity to work on this book. I am grateful to share with my students and others
what I have learned up to now about the road forward.
1 Introduction
1.1 Roots
When you open a newspaper, any newspaper, there is a big chance you will encounter
the term sustainable development. Introduced to a broader public in the 1980s with
the publication of the UN’s report Our Common Future, sustainable development
has become common vocabulary. The word ‘development’ is commonly used to
indicate growth, not only in quantity, but primarily in quality. The word ‘sustainable’
refers to something that can or should last. The idea of sustainable development
reflects one of the leading aspirations of humankind in the 21st century, not unlike
the idea of socialism in the early 20th century. It has become a modern equivalent of,
and complement to, the Declaration of Human Rights, formulated shortly after the
devastating Second World War. Civil society organisations have pushed sustainable
development forward; respected business and government leaders now hail it as the
foremost challenge for the 21st century.
Inevitably, such an aspiration or ideal accommodates a large variety of explan-
ations, objectives and proposals. These are intertwined with personal and collective
values and perceptions, which are in turn rooted in millennia of developments
shaping human experiences, knowledge, technical skills and social arrangements.
Given the human population’s continous growth and its use of the planet as a
source of resources and a sink of waste, humanity needs an ongoing dialogue that
slowly converges to a widely shared vision on the theory and practise of sustainable
development.
The word sustainable has been known in European languages since the early
Middle Ages. It is rooted in the Latin verb sus-tenere, sub meaning ‘up from below’
and tenere meaning ‘to hold’. In the physical sense, the verb to sustain is equivalent
to bearing, or carrying the weight of something to keep it from falling by support
from below. However, early on, the word had a meaning beyond a simple mechanical
act, as is already evident in the words of the Roman philosopher Seneca (3 BCE–
65 CE): ‘The society of man is like a vault of stones, which would fall if the stones
did not rest on another; in this way it is sustained’.
One of the oldest and most common connotations of the verb to sustain is to
keep a person, a community or the spirit from failing or giving way, to keep it at the
proper level or standard. It can be active, as ‘to support (life)’ and being capable
2 Introduction
and willing to go on. It can also be passive: ‘to undergo’ or ‘to endure’ and is then
equivalent to bearable or defensible. Which of the two meanings apply depends on
the role, attitude and circumstances of the actor. As he or she may succeed or fail,
the verb to sustain reflects the human condition: ranging from willpower, duty and
pride to fate, pain and suffering. As early as 1290 CE, this connotation was known
in the English language.
A closely related connotation stems from the archetypical notion of some force
or god, which ‘keeps the world running’. In Chinese and Indian cosmology, the
forces sustaining the world reflect a dynamic equilibrium between opposites; in
Greek cosmology, it is Atlas who kept the earth and the heaven separated. In this
sense, the verb to sustain gets a transcendent connotation, as in Milton’s words:
Whatever was created, needs

to be sustained and fed.
This is even more outspoken in ‘sustenance’. The word sustenance has become equi-
valent to nourishment, food and more generally the means of living or of sustaining
life – without any specification of ‘that which sustains’. This usage, as in Tennyson’s
verse:
Water is one part,

and that not the least of our sustenance,
comes close to the ecological meaning, which the word sustainable has acquired in
recent times.
An English equivalent of the verb to sustain is ‘to last’, meaning to go on existing
or to continue. Interestingly, it used to be associated with performance and duty.
The verb ‘to endure’ is an English equivalent of to sustain and is rooted in the
Latin verb durare. It is common in other European languages. In German, the word
dauerhaft is the common word for sustainable, with nachhaltig as a synonym. The
Dutch equivalents are duurzaam and houdbaar. In French, the word durable is
most common – and is also used in English as a synonym of lasting or permanent.
The words soutenable and viable are also used in French as synonyms to indicate
something that is bearable, can survive or is feasible.
Present-day usage of sustainable refers to an act, a process or a situation, which is
capable of being upheld, continued, maintained or defended. It has a largely active
disposition, in the context of sustainable resource use or management. The word
sustainability expresses the presence of such a capacity and is a recent coinage. The
words rooted in durare suggest a more passive connotation than those rooted in
sustenere.
The word development comes from the des meaning ‘undo’ and veloper meaning
‘to wrap up’ in old French and is possibly of Celtic origin. In present-day use, the verb
to develop means to (help) strengthen and enlarge. In particular, it is a progression
from earlier to later stages of a life cycle or a process from simpler to more complex
stages of evolution. It is about growing by degrees into a more advanced or mature
state. Development is considered to be broader than quantitative growth. It involves
maturing, ripening or bringing from latency to or towards fulfilment and fullness. It
refers to a dynamic process of (causing to) grow and differentiate along lines natural
to its kind, of improving the quality and of (causing to) become more complex or
1.1 Roots 3
Box 1.1. Sustainability science. ‘A new field of sustainability science is emer-

ging that seeks to understand the fundamental character of interactions between
nature and society . . . . [it] needs to move forward along three pathways. First,
there should be wide discussion within the scientific community regarding key
questions, appropriate methodologies, and institutional needs . . . Second, science
must be connected to the political agenda for sustainable development . . . . [and]
third (and most important), research itself must be focused on the character
of nature-society interactions, on our ability to guide those interactions along
sustainable trajectories, and on ways of promoting the social learning that will
be necessary to navigate the transition to sustainability’ (Kates et al. Science
292(2001) 641–642).
‘Sustainability science is not yet an autonomous field or discipline, but rather
a vibrant arena that is bringing together scholarship and practice, global and
local perspectives from north and south, and disciplines across the natural and
social sciences, engineering, and medicine’ (Clark and Dickson PNAS 100(2003)
8059–8061). In the same article, the link with a new kind of science is emphasised:
‘Post-Normal Science has been developed to deal with complex science related
issues. In these, typically facts are uncertain, values in dispute, stakes high, and
decisions urgent, and science is applied to them in conditions that are anything
but “normal”’.
intricate. Development is an evolution from simple to complex in terms of technical

and managerial skills and of social-cultural connections and institutions.
The concept of sustainable development supposedly combines the ideas of a
process or situation that can be continued and one that is growing in complexity
and maturing towards ‘natural’ fulfilment. Introduced in the 1970s, it was applied
initially with reference to an ecological or environmental desire, target situation or
state. A measure or indicator of sustainability is then the difference between the
actual and the desired situation and the timepath towards it. Often, the desired or
target situation is related to some reconstructed preindustrial ‘natural’ situation and
serves as a reference. Practitioners often held the view that ‘sustainable development’
could thus be given an objective interpretation.
In the 1990s, the interference of social scientists and notably economists made
it clear that formulating such a desired or target situation for a sustainable devel-
opment trajectory cannot be legitimised solely on ecological-environmental criteria.
First, which indicators for decision making are chosen? Should not economic and
social aspects be included as well? Second, if there is agreement on the choice of
indicators, the desired or target indicator levels must be the outcome of a societal
negotiation process and be open for renegotiation. Even if a command has to be
obeyed under penalty of complete and irreversible loss and, therefore, should be
considered an unconditional moral obligation – Kant’s categorical imperative – it
may still have to be renegotiated in the light of new information, a shift in values
or new and other pressing needs.1 Accordingly, economists argue that the quest for
1 Examples are an unconditional interdiction of slavery and a complete ban on entering Antarctica.
4 Introduction
sustainable development can and should be founded on welfare economics and soci-
etal cost-benefit analyses. Other social scientists brought their own and different
observations, concepts and theories. In any event, it is now widely acknowledged
that the interpretation of sustainable development includes the subjective, value-
laden elements that are inherent in the words sustainable and development. The
inputs from a variety of scientific disciplines have led to a flux of ideas, concepts
and observations, which together lead to a new branch of science in the making:
sustainability science.
1.2 Sustainability Science

This book has also the word science in the title. The word science comes from the
Latin noun scientia and the associated verb scire meaning ‘to know’. Originally, it
may have come from the Latin verb scindere meaning ‘to cut, divide’. Thus, an old
connotation related the word already to the act of distinguishing, that is, of divid-
ing in the mind. The word science has undergone a gradual evolution in European
history. In the 17th century, it became separated from more practical and artistic
knowledge – and practises and skills – then became understood as theoretical truth.
Gradually, science got the meaning of a ‘body of regular or methodical observations
or propositions . . . concerning any subject or speculation’. The overlap with philo-
sophy became smaller, and science evolved into an expanding set of compartments
of systematised knowledge about particular objects. When it concerns objects in the
physical world and its phenomena, science is natural science – but the word natural
is often omitted. The focus is on general truths and laws as well as on particular
methods of enquiry. However, the first and common meaning of science still is ‘a
state of knowing: knowledge as distinguished from ignorance or misunderstanding’
(Encyclopedia Britannica).
Connecting sustainable development to science stems from the idea that sus-
tainable development is an aspiration that can and should be realised only on the
basis of scientific knowledge. Does it justify the establishment of a new branch of
science? Yes, for several reasons. An overriding reason, in our view, is that human-
ity is confronted at an ever larger scale and rate with the consequences of its own
success as a species. The complexities and uncertainties of the human adventure are
becoming such that insights from all the sciences are needed to deal with them if
we are serious about the aspiration to improve quality of life (development) that
endure (sustainable). Understanding and acting upon the causal mechanisms and
behavioural responses across several time- and space-scales is a great challenge for
an increasingly fragmented science. A transdisciplinary approach is called for, in
which the quantitative and the qualitative, the natural and the social and also theory
and practise (or science and policy) are reconciled and creatively combined. Such
an integrating and synthesising approach deserves the name sustainability science.
This new science started with research and education on change processes across
all scales in space and time. Known as Global Change Science or Earth System
Analysis, the focus is on large clusters of observations and phenomena, and new sci-
entific branches develop along the borders of classical disciplinary sciences. Much of
it emanated from the environmental sciences, which branched into many directions
such as atmospheric and ocean science, marine biology, human ecology, ecological
1.2 Sustainability Science 5
economics and so on (Boersema and Reijnders 2009). Gradually, it became clear

that there were rather general laws underneath the phenomena in the different
fields of study. Also, the emphasis shifted from description of states to understand-
ing changes in states. System theory and the methods of system analysis and system
dynamics became the most coherent expression of this insight (Bossel 1994). The
new field increasingly expanded beyond the environmental sciences. The time has
come to reinforce the unified approaches and unifying tendencies in science and to
liberate the study of real-world processes from the confines of artificial, 19th-century
boundaries between the scientific disciplines. Sustainability science is perhaps the
most clear and desirable expression of this endeavour.
What is a meaningful definition and working program for sustainability science?
During the last few decades, scientists from different backgrounds have formulated
attempts at founding statements. The following are a good start (Kates et al. 2001;
Clark and Dickson 2003):
r sustainability science is about understanding the dynamics of evolving, coupled
social-ecological systems (SES);
r sustainability science is transdisciplinary: solutions to the problems have to
acknowledge that the world is/becomes more integrated, more complex and
more uncertain;
r the focus of sustainability science is on the interactions between the resource
system (earth/life sciences), its users and the governance system (social sciences);
and
r sustainability science is problem-driven to manage complex coupled SES in order
to have them deliver what people value.
This book uses the term sustainability science to indicate the efforts from all corners
of science to construct a framework for understanding and acting in relation to
(un)sustainable development. The aim is to provide foundations for and essential
concepts and methods in such a sustainability science.
In sustainability science, human individuals and societies are studied in a
biosphere context. The key notion here is that of SES, which are defined as
integrated human–nature systems with reciprocal feedbacks and interdependences
(www.resalliance.org). It appears to be a ‘natural’ unit of study. We reflect expli-
citly on the ways and limitations of a transdisciplinary approach and pay attention
to models and the process of modelling. We introduce system dynamics and, more
broadly, complex systems science as one set of methods and techniques (‘tools’) for
integration. As part of the transdisplinarity, we consider explicitly the great variety
and richness in people’s values and beliefs about the world and about sustainable
development. The idea of worldview pluralism is applied in various places as a way
to deepen personal understanding and engagement.
The problem-driven nature of sustainability science is controversial. If one
accepts it as a constitutive element, two questions arise: Which problems should
be addressed and where does the overlap begin with practical policy and man-
agement insights and skills? Because there are very different appraisals of the
problems humanity faces in the 21st century, the first question raises the issue of
worldview pluralism. Natural scientists formulate the problem largely in terms of
human-induced changes at a planetary scale. Some of these processes have become
6 Introduction
well known, such as the ozone layer depletion (‘ozone hole’) and anthropogenic
climate change (‘global warming’). Others are still less known but possibly equally
serious (Rockström et al. 2009). These ‘problems’ were already addressed one year
after human beings set foot on the Moon, in the report Man’s Impact on the Global
Environment (SCEP 1970): ‘the concept of the earth as a “spaceship” has provided
many people with an awareness of the finite resources and the complex natural rela-
tionships on which man depends for his survival . . . Some . . . have warned of both
imminent and potential global environmental catastrophe. Theories and specula-
tions of the global effects of pollution have included assertions that the building
of CO2 from fossil fuel combustion might warm up the planet . . . the possibility
of . . . particles emitted into the air . . . lower[ing] global temperature . . . the effects
on ocean and terrestrial ecosystems of systematically discharging . . . heavy metals,
oil, and radioactive substances; or nutrients such as phosphorus’. It was repeated and
supported with a system dynamics computer model in the report Limits to Growth
(Meadows et al. 1971) to the Club of Rome. Since then, the world has in terms of
population and economic activity largely followed the ‘business-as-usual’ path in that
report (Turner 2008). Nevertheless, there are still widely divergent interpretations
and appreciations of what happened during the intermediate forty years.
The second question points to the demarcation between science and technique,
ep©sthmh and tecnh. Sustainability science differs from sustainably managing SES,
although the two have a fruitful and creative connection. Knowledge of how the
world works and how we think the world works is only one, and not necessarily
an essential or sufficient, ingredient of acting upon the world. This book follows a
pragmatic course. First, it considers models of systems important in understanding,
assessing and communicating about sustainable development. Therefore, Chapters 2,
8 and 10 deal rather extensively with the construction and use of models and with
some analytical and modelling tools. Aware of sometimes large, or even impreg-
nable, limitations of (mathematical) modelling, Chapters 7, 9 and 11–13 present
stories and case studies in order to bridge the concrete and the abstract, the data and
the theory. Finally, Chapters 6, 10 and 12 introduce other ways of engaging people’s
insights and participation, such as value surveys, participatory modelling and sim-
ulation games. In this way, the scientific insights presented provide a context for
action – technical, political or other – without explicit discussion of the myriad
options for action.
Box 1.2. Exactness. Don’t expect a Unified Theory of SES or Social-Ecological

Systems. It does not exist. Do not expect exactness either . . . precision is not
always what we are looking for:
Joachim went to the famous Museum of Natural History where, he was told,
one could see many dinosaurs.
Seeing one of them, the provost standing nearby told him ‘This dinosaur is very
old, sixty-five million and fifteen years’.
‘Very old indeed’, replied Joachim, ‘ . . . but how do you know its age with such
precision’?
‘Well’, the man replied, ‘when I came here to do this work, it was sixty-five
million years old and that was fifteen years ago. . . . ’
1.3 Sustainable Development Is About Quality of Life 7
1.3 Sustainable Development Is About Quality of Life

We have spoken about the roots of the words sustainable and development. But
what is sustainable development? The word sustainable is used in many different
contexts, and not always sensibly: sustainable cities, sustainable traffic, sustainable
water, sustainable livelihoods, sustainable banking, sustainable technology – and
sustainable growth. The French newspaper Le Monde celebrated the start of the
‘Semaine du développement durable’ in 2005 with a special supplement saying:
‘Let’s forget the words “sustainable development”, because they provoke indiffer-
ence . . . and, worse, they make people smile. Let’s take care of its contents, though:
producing what people need without destroying their environment. . . . ’ They are
not the only ones who give this advice. Yet, this book does not follow the advice and
sticks to the notion of sustainable development as the core concept. The objective is
to deepen its content both in the abstract and the concrete.
For the moment, we use the most widely known definition of sustainable devel-
opment, given in the UN’s World Commission on Environment and Development
(WCED) report Our Common Future (1987):
Sustainable development is development that meets the needs of the present

generation without compromising the ability of future generations to meet their
own needs.
Within this definition, the goal of sustainable development expresses a quest for
developing and/or sustaining qualities of life. Framing sustainable development in
terms of quality of life introduces the subjective and objective dimensions of human
well-being and invites a truly transdisciplinary approach. It has an intergenerational
and an international dimension: people should act here and now in such a way that
the conditions for a (decent/high) quality of life elsewhere and later are not eroded.
Did we only shift the problem – because: What is quality of life? And which
qualities, for whom and for how long? Should these be experienced or imagined
qualities of life, for the lucky few or for all, for humans only or for all species, for our
children or for the next seven generations or even longer?2 Another question poses
itself: Is it development towards a static situation, a world that can be sustained
because there is no change? Such a blueprint approach is seductive, as the many
utopias in literature testify (Achterhuis 1998; de Geus 2003). But science tells us that
everything is always and everywhere in flux and that development is an evolutionary
path of success and failure.
Quality of life is an experience that stretches out over large domains in space and
time, in our individual life space (Figure 1.1a,b).3 In first instance, I as an individual
person relate to it in the here and now: material and immaterial well-being. Do I have
enough to eat? Do I have shelter? Can I avoid or cure diseases? Can I have sex and
experience love? Can I learn or apply skills? Can I communicate and relate? What
we experience as quality of life is, through our actions and emotions, our beliefs
2 Most people will adhere to an anthropocentric view, but some people want to extend it to all life or
even the planet as a whole. Such an ecocentric view can in theory be defended – but the defense is,
in practise, always by a human being.
3 Of course, each individual lives in a larger, social-cultural context and the two schemes in Figure 1.1
are flat-world simplifications of a complex reality.
8 Introduction
a)
space
Elsewhere and later

world
nation
region
town
community
neighbourhood
family
time
Here and now next next next next climate
week year election generation change
b)
world
religions
globalization poverty in
Africa
(my) (my)
church sports-
club
(my) neighbourhood
company skills,
play relationships
(un)employ- terrorism
I food, sex
ment rate job/income
(opportunities) me
myself recreation
(opportunities)
traffic health
congestion
housing air/water
trade quality
tropical
balance interest
forests
rate
immigrants
climate
change
Figure 1.1a,b. Simplified schemes of time and space in the perception of sustainable develop-
ment issues: (a) the space and time scales in which we experience the world; (b) items in the
centre-periphery of our daily consciousness.
and thoughts, also something of others and elsewhere, of past and future (Figure
1.1b). Is there food for the whole family? Can I pay my children’s school fees? Will
there be riots in town? Will my husband’s job disappear next year? Can I still enjoy
last week’s celebration or forget last year’s insult? Will there be a good harvest this
year? Clearly, quality of life is not easily condensed in one or a few quantitative
measures.
We have seen that development is growth towards a more complex and mature
state. Growing up, maturing as a human individual implies a widening of one’s
perspective on what life is about and on what ‘the good life’ can or should be. It
can coincide with a deepening of experiences, from the sensate and emotional to the
mental and spiritual. Thinking and acting can become more inclusive, with caring
for the larger scale and the longer term.The motivation for inclusive thinking can be
an expression of biological pragmatism and functionalism or of solidarity with wider
kin and of tribal ‘commandments’ and social norms for the benefit of the group.
Perhaps, it arises out of spiritual yearning.
Whatever the motives, in the manifestations of inclusive thinking, the aspiration
of sustainable development is recognised as more than a scientific quest: It is an
ethical and transcendent endeavour. It challenges us as individuals and groups in
how we manage our needs and wants and how we organise and manage resource use
that goes with it. Ethics enters the discourse, with questions about fairness, solidar-
ity, justice, egoism and altruism. Transcendence appears when we inquire into the
meaning and dignity of human life. I can give many references to books by philo-
sophers, artists and mystics who have struggled with these questions and attempted
to express it in words (Elgin 1993). It is probably better if you reflect on your own
life to see if it makes sense for you. Subsequent chapters offer openings for such
reflections.
In a practical sense, we must link sustainable development to the aspirations of
human beings for a good quality of life while respecting the plurality of worldviews.
Science can contribute here with adequate beliefs about the world and its workings
and with methods and tools to make those beliefs effective in action. It can also
teach how to complement intuition with rationality in managing uncertainty. It can
inspire feelings of respect and joy in the face of the world’s rich diversity, beauty and
complexity. And, finally, it invites us to reflect and give knowledge and mind their
proper place in our lives.
Immanuel Kant (1724–1804) offered a guide in the search for sustainable devel-
opment when he summarised his philosophy as the search for an answer to three
questions: ‘What can I know? What ought I to do? What may I hope for?’ In the
present context, these questions can be reinterpreted. What can science tell (and
what not)? What, then, is an individual’s responsibility and duty? And which dreams
and destinies can he or she expect? This book is primarily about Kant’s first question:
What can I know?, but it opens a panorama on answers to the other two questions.
1.4 Guidelines for the Reader

Set up as a textbook and an introductory guide, this book is intrinsically transdis-
ciplinary: The student is not expected to become an expert in a particular domain
or method but, instead, is invited to learn about and appreciate the different per-
spectives on (un)sustainable development. Not only are those perspectives following
the different scientific disciplines, but they also reflect the different individual world-
views. In this way, the book prepares the student for an independent yet personal pos-
ition on ideas and actions regarding sustainable development. The most important
10 Introduction
word in this respect is probably context. After studying the book, the student should
be able to put into context what is read or seen in journals, newspapers, books or
on TV or the Internet about ecological, economic and social aspects of sustainable
development. Equally important, the reader should feel better equipped to engage
with his or her own values and beliefs in the aspiration for a more sustainable world.
There are four threads in this book. The first is a historical and qualitative
description of past developments and civilisations in order to get an idea of ‘where
we are and what the problem is’. It is based on reconstructions by archaeologists and
historians and comprises Chapters 3 and 4. In most chapters, there are also ‘relevant
(hi)stories’.4 These are personal accounts of people living through particular events,
which illustrate concepts or models discussed in the chapter. Of course, it is merely
a selection of the numerous stories about (un)sustainable development around the
world.
The second thread centres around the notion of worldviews and is the focus
of Chapters 5 and 6. I briefly examine the historical background of the concept
of sustainable development. Next, I discuss the objective and subjective aspects of
needs and capabilities and the variety in value orientations and beliefs that describe
and explain the difference in views about sustainable development and quality of life.
After this groundwork, I offer a theory and a framework that explore the centrifugal
forces behind unsustainable developments, which are applied throughout the book
to categorise and understand the different views people have on important issues
such as population, resources, technology and economic growth.
The third thread is the systems approach and the methods and techniques that
have been developed over the years in order to understand a system’s behaviour
over time. Chapters 2 on system dynamics and the thematic Chapters 7 and 9–14
use simple simulation models to explore basic mechanisms of change. These models
are constructed with software packages such as Stella R
, VensimR
and NetLogo R
(Mathematical details are set apart in the appendices for the interested student). To
appreciate the role of mathematical models in sustainability science, there is a brief
orientation on the philosophy of science, in particular on the nature of knowledge
and models and on uncertainty and complexity.
A fourth thread summarises the major insights and findings of the scientific dis-
ciplines that shed light on (un)sustainable development. Each scientific discipline
contributes to the search for a sustainable world. It may be argued that ecology
and geography are the core of sustainability science, but the natural and engin-
eering sciences and the economic and social sciences may be of equal importance.
Largely following the reductionist-empiricist paradigm, later chapters survey obser-
vations (‘facts’), concepts, methods and theories (‘laws’) in contemporary science
with respect to environment and development. What is the input from classical
thermodynamics and mechanics and, broader, energy science (Chapter 7)? What
insights do we gain from ecologists and demographers who study nature’s evolution
and species populations and what is the image of man in the economic and social
sciences (Chapters 9 and 10)? What does science tell us about the dynamic processes
4 We will also use the words narratives or, sometimes, case studies or anecdotes. The website
www.sustainabilityscience.eu is collecting stories with relevance for sustainability science issues.
behind resource exploitation and environmental degradation (Chapters 11–14)? Of

course, these areas are far too large for a single book or a single mind to com-
prise. Therefore, the focus is on what are considered the most important concepts
and insights, illustrated with data and stories and connected through models and
worldviews.
The book’s fifteen chapters include the cross-reference of ideas and concepts
within chapters, in order to enhance coherence, integration and the benefit of cross-
fertilisation. At the end of each chapter is a list of useful books for further reading
and of selected websites that offer, as of 2011, relevant data, models, reports and
papers. At the end of the book, there is a list of acronyms and a glossary that
briefly explains keywords. Most chapters can also be read on their own, particularly
those on rather well-defined topics such as agro-food systems and renewable and
non-renewable resource use.
One consequence of the integration is that the reader may encounter parts that
are already familiar to him or her. There are two options then, apart from get-
ting annoyed or angry and stop reading. The first one is to skip those texts. The
second one is to read it again with the objective to refine one’s insights through
the discovery of new links and vistas and new connections to personal values and
beliefs. ‘Creativity is the power to connect the seemingly unconnected’ (think-
exist.com/quotes/william_plomer/). I invite you to choose the second option.
There is more to learning than the mind. Learning requires discipline and patience.
The Persian Sufi-Master Idries Shah says: There are some things which you have
to do for yourself. These include familiarizing yourself with study-materials given to
you. You can only really do this – and thus acquire real qualities – if you suspend the
indulgence of desire for immediate satisfactions. An attitude of empathy, of love is
needed as well. As the Persian poet Rumi said:
If you’re in love,
You need no proof.
If you’re not in love,
What good is a proof?
SUGGESTED READING
Suggestions for a research program from a Global Change perspective.

Clark, W., and N. Dickson. Sustainability science: The emerging research program. PNAS
100 (2003): 8059–8061.
A sequel to the outline started in Kates et al. (2001).
Clark, W., ed. Sustainability Science: A room of its own. Special Issue PNAS 104 (2007):1737.
A first outline of what sustainability science should address. Since then, quite a few papers have
been published on this topic.
Kates, R., et al. Sustainability Science. Science 292 (2001): 641–642.
The journal PNAS now has a special section on sustainability (see websites list).
Perrings, C. Future challenges. PNAS 104 (39) (2007):15179–15180.
An evaluation of the 1972 Limits to Growth report on the basis of historical data since then.
Turner, G. A comparison of the Limits to Growth with 30 years of reality. Global Environ-
mental Change 18 (2008): 397–411.
12 Introduction
USEFUL WEBSITES
r www.igbp.net/ is the site of the International Geosphere-Biosphere Program (IGBP) and

at www.igbp.net/documents/resources/science-4.pdf one can download the report Global
Change and the Earth System (Steffen et al. 2003).
r www.pnas.org.proxy.library.uu.nl/site/misc/sustainability.shtml is the website of the Sus-
tainability Science section of the Proceedings of the National Academy of Sciences of the
United States of America (PNAS), with special issues on a series of topics (land change,
health, food, climate change, poverty and others).
r www.awakeningearth.org/ is the site of Duane Elgin, who discusses sustainable develop-
ment in the perspective of the evolution in human consciousness.
Appendix 1.1 United Nations Decade of Education for Sustainable

Development
In December 2002, Resolution 57/254 on the United Nations Decade of Educa-
tion for Sustainable Development (2005–2014) was adopted by the United Nations
General Assembly. UNESCO was designated as lead agency for its promotion
(portal.unesco.org/education/en). The rationale for this Decade is stated as:
Education as the foundation of sustainable development was reaffirmed at the

Johannesburg Summit, as was the commitment embodied in Chapter 36 of Agenda
21 of the Rio Summit, 1992. The Plan of Implementation establishes the linkages
between the Millennium Development Goals on universal primary education for
both boys and girls, but especially girls, and the Dakar Framework for Action on
Education for All. The creation of a gender-sensitive education system at all levels
and of all types – formal, non-formal and informal – to reach the unserved is emphas-
ized as a crucial component of education for sustainable development. Education is
recognized as a tool for addressing important questions such as rural development,
health care, community involvement, HIV/AIDS, the environment, and wider eth-
ical/legal issues such as human values and human rights.
There is no universal model of education for sustainable development. While
there will be overall agreement on the concept, there will be nuanced differences
according to local contexts, priorities and approaches. Each country has to define
its own priorities and actions. The goals, emphases and processes must, therefore,
Table 1.1. Key action themes of the UN decade of

education for sustainable development
Quality education Cultural diversity

Overcoming poverty Indigenous knowledge
Gender equality Media & ICTs
Health promotion Peace & human security
HIV/AIDS Governance
Environment Climate change
Water Biodiversity
Rural development Disaster reduction
Sustainable consumption Sustainable urbanisation
Sustainable tourism Corporate responsibility
Human rights Market economy
Intercultural understanding
Appendix 1.1 United Nations Decade of Education for Sustainable Development 13
be locally defined to meet the local environmental, social and economic conditions
in culturally appropriate ways. Education for sustainable development is equally
relevant and critical for both developed and developing countries.
The list of key action themes covers an impressive twenty-three items, revealing how
broad the topic of sustainable development is being conceived today (Table 1.1). It
not only reflects the understanding that ‘everything depends on everything else’,
but also that many different stakeholders are involved. Full of good intentions
and large aspirations, these global initiatives show, on the one hand, the emer-
gence of a global consciousness of the human predicament – and, on the other
hand, the risk of creating worldwide bureaucracies that become as much part of the
problem as of a solution. Whatever the judgment, they provide the setting for this
course.
2 The System Dynamics Perspective
2.1 Introduction
To develop a proper framework for sustainability science, we must learn to think
and model across disciplines and in terms of complex systems. In the last couple of
centuries, many mathematical methods and techniques have been developed, which
have become the hallmark of the natural sciences and their successes. Since the 1960s,
the evolution of these methods and their applications became closely linked because
of the advent of the computer and its rapid advances in performance. Mathematical
modelling has become standard practise in the natural and engineering sciences, and
increasingly in the life and social sciences. Scientists in many disciplines now have
access to simulation software and models.
The roots of system theory go back to the early days of the Enlightenment,
in natural science as well as in social science (Richardson 1991). Its present-day
form emerged in the 1930s, inspired by insights in biology and ecology. In the
second half of the 20th century, system analysis got its specific content from control
engineering (cybernetics) and electro-mechanical engineering, which emphasises
‘the larger picture’ by shifting the focus from events to behaviour to structure. System
analysis has been applied in a variety of engineering, management, economic, social
and resource contexts.
This chapter gives a brief introduction into the theory and practise of system
analysis and of modelling and simulation of systems. These concepts and methods
are used throughout the book. System analysis is an excellent method to concep-
tualise and, to a certain degree, simulate the dynamic behaviour of systems. It is
inherently transdisciplinary in its search for general or universal principles govern-
ing systems. The classical method in system analysis is the mathematical language
of calculus, for example, integral-differential equations. Calculus is an essential tool
in understanding behaviour of systems over time. For more complex systems, it has
limitations, even for the skilled mathematician. However, rapidly expanding sets of
computer simulation tools provide novel and complementary ways to model and
examine system behaviour. Before I give a more formal introduction, I illustrate the
system’s view with the example of the world car system, which is of great importance
from a sustainability perspective.
14
Crude oil & minerals Car parts manufacturing

RD&D/innovations
‘Resources’ ‘Production’
‘Macro-economy’
Oil refinery & Car manufacturing
Employment
minerals processing ‘Services’
Car sales, insurance,

financing... Trade
Road construction
‘Infrastructure’ ‘Consumption’
Car use: forensic, Repair, maintenance

Traffic control, police,
business, tourism...
environmental regulation...
Car recycling
Road accidents,
medical/legal services
Emissions, pollution
Figure 2.1. The car system and related subsystems in society.
2.2 The World Car System

Many of us use a car for a variety of purposes. In industrialised regions, mobility
is an intricate part of life: commuting to work, visiting family and friends as well
as allowing for leisure and business trips. Between 1970 and 2002, the number of
passenger-kilometres (pkm) travelled in Europe increased from 1,600 to almost 4,000
billion per year – or more than 10,000 kilometres (km) per person per year. Here, we
restrict the analysis of the transport system to the automobile. This means omitting
other transport modes such as trucks, trains, ships and airplanes. Also, the bicycle,
motorcycle and bus are not considered, although they are still very important in
many low-income regions. How does an integrated systems perspective on the car in
the context of (un)sustainable development look?
Figure 2.1 shows a diagram of the system illustrating key variables and their
relationships. Such a diagram is the first step in a conceptual model. The boxes
represent important subsystems and the arrows indicate material, monetary and
informational interactions. The graph in Figure 2.2 shows the growth of the world
fleet of registered cars since 1960. The total number of cars in use exceeded 700
million in 2010 – more than one for every tenth person on earth. This total is expected
to increase to 1.25 billion by 2020. The rate of manufacturing is now around 70 million
per year; the rate of scrapping is about 15 million per year. The production and use
of this ‘vehicle stock’ has enormous large-scale and long-term impacts, both directly
and indirectly.
The private car’s positive aspects for people of all professions and classes are
well-known: more mobility and social contacts, better access to schools and medical
services, more and cheaper travel and tourism, to mention the most important ones.
The private car has thus satisfied a variety of needs and desires – for jobs, status,
education, health and so on – and has created new ones. In high and medium
income countries, the majority of people are now dependent on the car for their
jobs, shopping and entertainment. In the United States, more than 50 percent of
800
Other
700
Brazil/Russia/India/China
Registered cars (mln)
600 Japan
500 France/Germany/UK
Canada/USA
400
300
200
100
0
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
Year
Figure 2.2. Fleet of cars in the world for several regions, 1960–2010. Data are in million of
registered cars (source of data: TEPD 2010).
households have two or more cars, and households spend on average almost one-
sixth of their income on their cars and its use. Average mileage, measured as the
number of km’s driven per vehicle per year, of U.S. households has stabilised in the
last decades to about 33,000 km per year, with almost one-fourth used for journey
to work (TEPD 2010). The average mileage of households in Europe and Japan is
lower, as they are in other world regions, but the trends point in the same escalating
direction.
The car system also has negative sides that make its present form unsustainable.
Initially, these were hardly noticed, because in the early 20th century, cars often
replaced horse carriages, solving in some urban areas the huge problem of horse
manure in the streets and its odor. It also relieved the less obvious problem of
the large amount of land needed to feed the horses. Alternatives such as the electric
tramway lost out to the car in the early stage of transportation change, partly because
Box 2.1. The world motor vehicle industry. The International Organization of
Motor Vehicle Manufacturers’ (OICA) website, www.oica.net, gives data on the
economic importance of car manufacturing. The motor vehicle industry (cars,
vans, trucks and buses) has become a powerful player in the world economic
system. It is one of the major employers: More than 8 million people are directly
engaged in it while another 40 million are indirectly employed in making vehicles
and parts. In 2005, car manufacturing comprised 5 percent of worldwide manu-
facturing jobs and 3 percent of the world GDP. These numbers do not include
the employment and expenditures for road infrastructure.
The automobile industry is also a major innovator, investing almost €85 billion
in research, development and production. It is also a major contributor to govern-
ment revenues around the world, contributing over €430 billion in 26 countries.
As the OICA site says, ‘If auto manufacturing were a country, it would be the
sixth largest economy’.
of active interventions by the automobile industry. The disadvantages of the car

system have become more visible since the second half of the 20th century: notably
impacts on health from accidents and air pollution and the destruction of landscapes
and ecosystems. With the ‘oil crises’ of 1973 and 1979, the dependence on finite
oil resources and the risk of geopolitical conflict became another concern, although
they had always been around (Yergin 1990). In the 1990s, the growing evidence
that carbon dioxide (CO2 ) emissions from burning oil contribute to climate change
became another negative side effect.
There are millions of deadly and non-deadly traffic accidents every year.1 In
2005, there were between 50 and 250 fatalities per million inhabitants in the medium-
to high-income regions (Figure 2.3a). In emerging economies, the number of cars
increases rapidly, and so do traffic casualties.2 Although there is a clear trend towards
declining numbers of fatalities in high-income countries – which sometimes means
less people killed but more people wounded and associated medical costs – most
low-income regions in the world still report increases in the number of fatalities in
absolute and relative terms.
Another health-related side effect is air pollution, not only from cars but also
from motorcycles and trucks. This has effectively been reduced in the high-income
regions, but urban populations in Europe are still exposed to concentrations of
nitrogen- and sulphur-oxides and other particulate matter above the recommended
World Health Organization (WHO) levels (Table 2.1). Urban pollutant levels in
India and China, especially particulate matter concentrations, are significantly above
the recommended values (Table 2.1 and Figure 2.3b). They pose already severe
health risks and are expected to increase further. One other side effect is noise
pollution. About 55 million people in the large agglomerations in the EU-27 are
exposed to long-term average noise levels above 55 decibel (dB), with 85 percent of
them from road traffic noise (EEA 2010). In the Netherlands, nuisance from traffic
noise is amongst the three biggest environmental problems perceived by citizens.
The car has a large impact on landscapes and spatial planning in the form of
land use for roads and parking space (around 1–2 percent) and, indirectly, because
of the fuels and materials needed to operate the system. Indeed, the car has largely
structured the way in which modern urban areas are built and inhabited. In Europe,
the existence of ancient city centres has moderated the rate and extent of adjustment,
but elsewhere urban areas are almost completely determined by the car system logic
and requirements. This causes additional pressure on land.3 In many parts of the
world, traffic congestion is a major problem, with loss of quality of life and of working
hours. More roads may diminish the congestion, but at increasing cost and further
landscape deterioration.
Another strain on the system’s sustainability is oil use. In 2000, an estimated
77 EJ was used for transport, nearly half of world crude oil production and the
1 The numbers in this section are from the OECD Factbook (www.oecd.org; tab, Statistics).
2 Absolute numbers are misleading because it depends on where a country is in the industrial trans-
ition. For instance, the OECD-countries and India had, on average, the same number of road
fatalities per million inhabitants (97) – but in India it climbed between 1997 and 2007 with 55
percent, whereas in the OECD-countries, it declined in this period with 27 percent.
3 It adds to the direct and indirect costs of car use: Parking costs are now in the order of 2–3 €/hr in
urban areas in Europe.
Russian Federation
Turkey
Slovak Republic
Korea
Poland
Hungary
Mexico
Estonia
Greece
Czech Republic
Slovenia
Belgium
Portugal
United States
Denmark
OECD total
Canada
Austria
France
Italy
Ireland
New Zealand
Finland
Spain
Australia
Chile
Luxembourg
Germany
Norway
United Kingdom
Netherlands
Sweden
Japan
Switzerland Per million inhabitants
Isreal Per million vehicles
Iceland
0 100 200 300 400 500 600 700 800 900

Number of road fatalities
Figure 2.3a. Road fatalities per million vehicles and per million people in 2005 (source of
data: OECD Factbook 2007).
equivalent of 2,200 billion litres of gasoline.4 About three-quarters of this was for
road transport. Therefore, fossil fuel use for transport is largely responsible for the
fast depletion of oil fields around the world and the geopolitical implications. From
2003 to 2004, the United States spent an estimated $50 to 100 per ton of oil used
(7.5 to 15 $/bbl) on military expenditures to defend oil supplies from the Middle
4 EJ is a unit of energy and the equivalent of 1018 Joule. See Section 7.2.1 for details on energy units.
Urban PM10 concentrations in 2000
" "
" " ""
" "
" " " " " " "
" """ """" " " " "" "
" " """" " " "" """""" " " " "
"
"
"
"
"" "" " " " " """ """""
"" "" " " """ """ """" " " "" " "
""" "" "" "" """ """""" "
"""" " """ " " "" """"
" " """"" "
"""" "" """
""" "" " """
""
""" "
"" " "" " " """""""""" " "
""""""
"" " " " " " "
"""" "
" " "
" " " """ """"""""" """"
"" """""
""
"""""""""
"" """""""" " " " """ " """"""" "" " "
" " """ " " " " "
" " " " """"""" """"
""
"
""
""
""
"""
" """ ""
""""""
"""
"
" ""
""" "
""""""
""""" """
"" "
"
"""
"""
""
""" """"
" ""
"" " """ "" " " """ "
" " " " "
" " " "" """ "" "" """ """
" "" "" """" " "
"" " " "" "" ""
"
"""""" """ """" """ ""
""""" """"
""""" """
"" """" ""
""" "" " "
" " " """""""" """ "
"" """ " " " " "
""""""""" " ""
" " "" " """ "" " "
" " " """""" " "" " " " " "" """"
" ""
"
"" "
"""""
" "
" """"""
" "
"""""
"
"""""""
"""" "" " "
" " "
"""""" " " """ " " "
" "
""
""""""""""" " "" "
""" "
"" "
""" """""" ""
"" """"""" " " """"" "" " """""" "
"
""" " " " "" "
" "" """ " "" " "
"""
" " " " " """""" " " """""""
" " ""
"""" """ "" " """"""" """""" """""" " " " "
""" " " """" """ """"
"" "
""
""
"" " ""
"" "" """ " " " """ """ " "
" " " " """ """ " "" "" """ """ " "" """"""""""" " " "" " " ""
""
""
""
"
"
"
"""
"
"
" " ""
"
" " "" " " "" """
"" " "" """
""""""""" "" "" "" " " " """" "" "" "
" "" """" """" """" "" " "" " ""
""" "" "
""""
"
" ""
""
" "
"
""
"""""
"
" " " " " " """" """"" "" """ "" """" " " " "" " """ " " "
" "" " " "" """" """" """ ""
" "" " ""
""
" ""
"""""
""
""
"""
"
""
"" "
"
" ""
" " "" "" """" "
"" " "
" """" " " "
" "
"""""" " " """""
"
" "
"" "" """"
""""""" " ""
"""
""
""
"""
""
"""
"""
"
"""""""""""
""""
"
"" " " " "" " " """ " " ""
"""" " "
" "" " " "
""
""
"
"""
""
""
""""""" "" " "
" """" " " """ "
""" " ""
"
"""""""""""" ""
""
""""
" " "" " " " " """""""""
"" "
"
" """
"
""
"
"
"
"
"""
"
"
"""
"
"""" " "
" ""
""" "
"
"" " " "" "
"
"
""
"
" " " " " " "" """ "" " " " " " "" """
""" """
""
"
"""
"""" """
"
" """"
""" " " "" "
"" ""
" "" """ "" """" "" "" " " "" " " " " """""" "" " "
" "" "" """ " " "" " " """" "" ""
""""
" """
" """""" """""" " "" "" " """ "" "
" "" " """ " " " " " """ " """" """"""" """
"
"""" """"""""
"
""
"
"
""""
"" " "" ""
""""
""""" " """ """"""" " """"""
""
""""""""
"" " """ ""
"
"""
""
" """""""
" " "
" " """ ""
"
"""""
""
"""""""""""""" "" "" """ ""
""
""
"
""" "
"""""" "
""""
" " """ "" "
"" ""
" "
"
"" " " " "
"
" " " """"" "" " ""
"""
"""
"
" " "
" """
" " " " "
" "
" "" "" " "
"""""" " " "" " " "" "" "
""""""" ""
""""" "" """"" " """ "
""" " " """
" " " " " "" """ """ " ""
"
""
"" "
"""" " """ ""
" """ " " """ "
""" " " " "
"
"""""
""
""""" """
"
" "" ""
" "
" "
"" "" "" " """"
" "" " """" "" ""
" """ " " " "" " " ""
" " " " "" "
""
""""
"""
""""
" """ ""
""" """"
" " """" " "" " " "" " "" "
" "" "" ""
" "" " " " " """ """""" """" """ " "
" "" " " ""
"" " "" " " """ " ""
"""
""" "" " " """ """
" "
""" " """
"
" " " """
"" " " " " " ""
Urban PM10 concentrations (ug/m3) ""
"""
"" " " ""
" """ """ " ""
" " "
"
" ""
"
"
"" " "
"" " " " " " " " "
"" " "" " " " " "" ""
" < 20 "" " " " " " """"" """"""
"" " " " "" """" " " " "" " "
" " """
" "" " ""
"" "" " "
"" "
" 21 - 30 " "
"
" " " " "
"
"" "
" "" " "
" """ " "
" """ " "" " "" "" " "
" " " """ " """
"" "" " "
" 31 - 50 " """ """"" """"" " "
" " " "" """
"
"""
"""""""
"" " " "
" "" "" """" "" "
" "" " "" "" ""
" " "
" "
" 51 - 70 " " " "" ""
" """ ""
"
"
"
" " " """ "
" " " "" """
""" " "" ""
" > 70 ""
"" " ""
" " " "" ""
" "
"" "
" "
"
"
Figure 2.3b. Air quality in terms of Particulate Matter (PM10) concentrations in the cities of the world. GUAM concentration and population data
are coupled to city locations (x,y). The values are aggregated to an average value for IMAGE-model 0.5o × 0.5o grid cells. The cells are converted
into square points. The darker the colour, the worse urban air quality (source of data: PBL). (See color plate.)
19
Table 2.1. Air quality in large cities in Western Europe and India. WHO guideline values are
40 μg/m3 (annual mean) for NO2 , 20 μg/m3 (24-hr mean) for SO2 and 20 μg/m3 (annual
mean) for PM10 (source of data: PBL)
Population NO2 SO2 PM10

City (million) (annual mean) 24-hr mean annual mean
Europe Paris 9.32 43 8 21

London 7.65 44 4 25
Berlin 3.45 22 4 27
Milano 3.29 55 8 54
Athens 3.07 32 11 41
Madrid 3.01 43 11 29
Rome 2.70 41 2 –
West Midlands (UK) 2.30 30 3 23
Greater Manchester 2.28 43 2 23
India Greater Bombay 12.59 19 7 77
Calcutta 11.02 53 9 237
Delhi 8.42 57 9 432
Madras 5.42 6 5 96
Hyderabad 4.34 29 6 178
Bangalore 4.13 61 7 173
Ahmedabad 3.31 23 15 231
Pune 2.49 53 31 340
Kanpur 2.03 20 9 413
Lucknow 1.67 33 16 391
East.5 Producing an estimated 14 percent (passenger cars) to 23 percent (all trans-

port) of global CO2 emissions, the transport system is a major contributor to the
enhanced greenhouse effect (OECD/IEA 2009). The car system is also an avid
consumer of mineral resources, notably iron, aluminium and plastics. In fact, recyc-
ling car parts has become a major industry as car scrap is an increasingly valuable
resource.6
Many analysts agree that the car system is not sustainable in its present form and
that a transition towards a sustainable transport system is needed. Such a transport
transition should address the previously mentioned adverse consequences. But there
are many ways to Rome, as the saying goes. The appeal of the private car and its
accessibility with rising incomes will continue to stimulate rapid growth of the car
fleet in low-income regions. Cross-country and time-series data suggest a logarithmic
relationship between income and car ownership and use, although the curve has a
high branch for U.S.-style countries with low population density such as New Zealand
and a low branch for European-style countries (Figure 2.4).7
5 One bbl is another unit of energy, namely, the equivalent of one barrel of 159 litre of crude oil.
Depending on the type of crude oil, a bbl is equivalent to approximately 6.1 GJ (1 GJ = 109 Joule).
6 A recent analysis, for instance, estimated that automobiles in use in the USA contain 19 million
tons of aluminium, which is equivalent to almost one year of world production (Buckingham/USGS
2010).
7 Characteristically, the countries with exceptionally low car density are the city-states Hong Kong,
Macao and Singapore – densely populated places with intense public transport.
900
AboveAvg
BelowAvg
Within30%Bound
600
Cars per Thsd
300
0
0 10000 20000 30000 40000 50000 60000
Current GNI (2007) I$

Figure 2.4. Car ownership in number of cars per thousand inhabitants as a function of income
(GDP/cap in 2007 I$) for 131 countries in 2007. A logarithmic curve fits fairly well (R2 = 0,72).
Amongst the high-income countries, the United States, Canada and Australia are more than
30 percent above this curve and countries like Hong Kong and Singapore fall more than 30
percent below this curve. The middle band are mostly countries in northern Europe (source
of data: TEPD 2010).
Most efforts to deal with negative impacts focus on car technology. Sustainability
is interpreted within an engineering context as more efficient fuel use and lower
specific emissions. There is still much room for efficiency improvements and there
are good prospects for new engine and new fuel types. Biomass-based fuels supply
a growing fraction of demand in countries like Brazil and the United States. The
electric/hybrid drive and fuel cell can replace the internal combustion engine, but
their introduction requires system changes and behavioural adaptations. There are
still many uncertainties and controversies about costs, efficiency and environmental
aspects of these novel techniques. Innovative alternatives have to compete with
a mature technology – the combustion engine – which is currently produced at
astonishingly large scale and low cost.
Any forecast has to consider the system as a whole. Predictions are difficult
because it is a complex system with growth-promoting as well as stabilising and
counteracting forces. Traffic congestion, space limitations and the cost of infrastruc-
ture slow down the growth of car use in most densely populated regions. A penet-
ration rate towards the U.S.-lifestyle number of 800 cars per thousand inhabitants is
hard to realise in densely populated parts of Europe, China and India. With rising
income, there is also less willingness to accept traffic casualties and urban air and
noise pollution. Oil availability and the growing need to reduce emissions can also
put a brake on growth. Dependence on finite and unevenly spread oil resources is
expected to grow in almost all official scenarios, and so is the oil price (OECD/IEA
2009). As was shown in the 1970s and 1980s, oil price is an important determinant:
The hike first led with some delay to increasing fuel efficiency, and the subsequent
fall stopped, if not reversed, the trend with the advent of the sports utility vehicles
(SUVs). Mounting concerns about climate change and a subsequent imposition of
carbon taxes will have great impact. Another uncertain part of the transition concerns
the modal split, such as the role of transport modes other than the car (e.g., faster
trains and air transport). Without regulation to internalise health expenditures, oil
depletion, climate change and so on in the transport cost, the transition will be slow
and difficult. These are the ingredients for an integrated system analysis of the car
system and its (un)sustainability. Let us now have a closer and more formal look at
systems.
2.3 System Dynamics: The Basics
2.3.1 What Is a System?

What is a system? One definition is:
A system is an interconnected set of elements that is coherently organised around
some purpose.
Therefore, the three key attributes are: elements, (inter)connections and purpose
(Bossel 1994, Meadows 2008). It is important to explicitly define a system boundary.
In this definition, a heap of sand or an arbitrary group of pedestrians is not a system.
But all around us are examples of (sub)systems:
r a car, a house or an airplane, making up connected parts in order to provide
mobility, shelter or movement;
r a forest with its connected elements such as trees and animals, and the purpose
to sustain their metabolism;
r a marketplace where buyers and sellers exchange fish, fruits and flowers accord-
ing to a set of conventions and relationships;
r a football team or a school, with rules being the connections, which permit to
play the game or teach the children;
r a country with its people and the customs, contracts, laws and institutions that
keep them more or less together around a social-cultural identity.
These systems are organised around a purpose or function, whether it is simple and
straightforward or complex and elaborate. A characteristic of these systems is their
integrity or their wholeness. They are ‘more than the sum of their parts’, as the saying
goes.
Often, the elements are the easiest part to observe and classify. A car has an
engine and wheels; a house has windows and doors; a forest has trees and animals;
a market has vendors and buyers; a football team has players; and a country has
individual people. Element identification depends on scale or resolution and, in
turn, on the objective of the analysis. Car engines and tree leaves, for example, are
intricate subsystems with their own elements, connections and purpose.
Interconnections or relationships are a more difficult affair. They represent the
physical, monetary and informational flows, as well as the laws and rules that govern
these flows. In a forest, these are, for example, the connections between the soil and
the roots, but at a smaller scale, they are the laws governing diffusion and reaction
processes. In a market, buyers and sellers are connected in a process of comparing
and bidding from which a price and a transaction result. In football teams, the players
are connected via their actions and signals within the constraints of the rules of the
game. In the car system, the drivers are connected to the road system via traffic
regulatons and other drivers as well as to car manufacturers via car performance and
built-in safety devices. Evidently, in more complex systems such as organisms and
institutions, the relationships are often hard to observe and classify – and it may be
even harder to find the underlying laws and rules.
The third item, purpose, or goal, is usually not included in the analysis of ‘inan-
imate’ matter in physics and chemistry. Scientific inquiry has increasingly gone to
the micro-scale of atoms and molecules with their physical connections and appar-
ent absence of purpose. Most natural scientists will not consider purpose a system
property. After all, what is the purpose of a piece of rock, a lake or a volcano? It
is only in a more holistic, phenomenological perspective that the notion of purpose
makes sense.
The purpose of living organisms is undeniable, but it is still difficult to grasp
system property. It manifests itself in all kinds of behaviour and in a variety of
settings such as the car driver going home, the wolf trying to catch deer, or human
beings protecting themselves against the cold. In a broad sense, the purpose is to
survive and to sustain. The individual and, in a larger context, the species tries to
maintain a certain structural integrity against the forces of decay and, in order to do
so, evolves. The same can be said about human institutions: They sustain and evolve
themselves. For inanimate artifacts, it is less ambiguous what is meant with purpose.
Tools and appliances are designed with a clear purpose that is usually evident in a
given context. For example, think of an oven, bicycle or radio. Or, think of the car
system: The purpose of the car is usually to transport people or goods, whereas the
purpose of the larger road system is to facilitate individual car movements.8 Even in
technology, however, purpose becomes less obvious and quite diverse with generic
machinery such as computers. The human-designed objects – including such things
as the Internet – resemble organisms in their complexity. Artificial life may be less
distant than we think. From the opposite side, the natural sciences are uncovering
the complexity of macromolecules and microorganisms into such depth that the
boundaries of what an organism is become ever more obscure.9
The purpose of a system is often even less visible than the interconnections
between the elements. In fact, it is only meaningful to talk about purpose within
a cognitive understanding, which makes such a purpose contextual and anthropo-
centric. It helps to ask oneself, what would happen if the purpose of a system is
changed? For instance, if a car is used to live in, the connections between its ele-
ments change and one may end up with something else – a house, possibly? Changing
the purpose of a football game or a stock market would create a new system with
novel behaviour. Playing with changes in relationships, rules or purposes is precisely
what makes fiction novels sometimes so interesting.
8 Of course, individual car drivers and the designer of the car system consider rules and have purposes
that are quite different from these purely functional rules and purposes.
9 The fading divide made between living (‘organic’) and non-living (‘inorganic’) has a long history.
It reflects a fault line that for a long time was thought to exist between the ‘inanimate’ objects of
physics and chemistry and the élan vital of organisms with their intuitively obvious purpose.
Figure 2.5. An influence or causal loop diagram (CLD),

picturing Balzac’s vicious circle (Richardson 1991).
2.3.2 Stocks and Flows

Our interest in systems is because we are interested in change within the context of
sustainable development. There are various ways in which one can look at change
in systems. Classical calculus – integral-differential equations – has become the
language par excellence to study change and to analyse the dynamics of real-world
processes. It has become an indispensable tool in the natural and engineering sciences
because it uses the language of mathematics, which implies more consistency and
transparency as well as simplifications and limitations.
System dynamics practises an open enquiry into the object system, starting with
a conceptual model and a focus on qualitative relationships: A influences B or,
even stronger, A causes B.10 The resulting schemes are influence diagrams or causal
loop diagrams (CLDs) that represent what is thought to be the essence of the
dynamic structure of the system under consideration. For example, nowadays a
French resident travels on average 40 km a day, whereas around 1840, residents
averaged only 4 km a day. The French writer Balzac noticed 160 years ago a paradox
about travelling: “I’d need rest to refresh my brain, and to get rest, it’s necessary to
travel, and, to travel, one must have money, and, in order to get money, you have to
work, create, etc.: I am in a vicious circle (cercle vicieux), from which it is impossible
to escape” (Balzac 1850:32, quoted in Richardson 1991). Such an observation can
also be translated into a CLD. Balzac’s account then looks like the loop in Figure 2.5.
The arrows indicate a presumed cause-effect relationship. This simple observation
with diagram has an important message to convey: sometimes we are caught in a
loop. If it goes in the wrong direction, as in this case, one speaks of a vicious circle.
The opposite is the virtuous circle. The construction of such conceptual models from
observations on cause-effect sequences is at the core of system dynamics.
The next step is to formalise the conceptual model into equations and search
for solutions. Most real-world systems exhibit nonlinearities, delays and feedbacks,
and analytical solutions are hard to come by. The use of mathematical techniques
(calculus) often leads to extreme model simplifications and, increasingly, simulation
is used to analyse otherwise intractable systems. One method for simulation that is
easily accessible is via dedicated software package for system dynamics simulations
10 We will usually speak of system dynamics as an equivalent of system theory, system analysis and,
more generally, the theory of dynamical systems.
(see the list of suggested software packages at the end of this chapter). It allows
the modeller to do model experiments and explore the dynamics without advanced
mathematical skills and too severe simplifications.
An important distinction in system dynamics is between stocks and flows. Stocks
(or levels) are referring to the content of reservoirs or compartments in which
something is stored. They are the equivalents of integrals in calculus and usually
indicated as state variables in system theory.11 They represent accumulated ‘stuff’
(things, individuals, matter, etc.) within a given system boundary. For example, water
in an underground aquifer, metal in a mineral deposit, phosphates in a lake, carbon
in the atmosphere, biomass in a forest, and so on. Or, think about them in relation
to the number of cars in the world (Figure 2.1), people in a town, students in a
classroom, television sets in a retail shop or money on a bank account. Sometimes,
also nonphysical variables are treated as stocks in the sense that they accumulate
over time. One can simulate, for instance, happiness or hunger as something that
increases or decreases over time and serves as persistent indicator of the state of the
system. But such a variable can also be considered a stock characteristic or property,
like temperature or density. Changes in a stock are called flows (or rates or fluxes).
Their equivalent in calculus are derivatives.12
In ecological and environmental science, the stocks are those parts of natural
ecosystems that provide services for human use, such as fertile soils, animal pop-
ulations, forest biomass or forested area and aquifers (source function). They can
also be air, water and soil compartments for disposal (sink function). In agriculture,
relevant stocks are grain stores and livestock. In engineering and economic science,
the stocks refer to the goods in an inventory or, more broadly, the capital stocks in
the form of machinery and other productive facilities. In finance, it refers to a sum
of money or a fund, which tends to become the most widely known meaning since
the advent of stock markets.13 Stocks can be lumped together in aggregate stocks,
as in ‘economic capital’, ‘financial capital’, ‘natural capital’ or even ‘human capital’
and ‘social capital’, although there is a risk that such aggregate variables become
meaningless.
The spatial and temporal scale have to be consistent across the definition of a
stock and its associated flows. If physical interactions in space are considered, stocks
are usually defined as densities (individuals per unit area). The stock of cars, for
example, has to be linked to the surface area (ha) where they are parked or drive
around if the objective is to make sense of the traffic system. If physical interactions
are weak or absent, as in monetary or informational exchanges, variables are often
expressed on a per person or per unit cost basis. The temporal aspect, in particular
the choice of the relevant time interval, is important because one may only have
information on changes in the stock during certain periods. Such a change per
period represents the net flow. The chosen time interval in a simulation influences
the outcome. Measurements in stock-flow diagrams are sometimes indirect because
information about flows is the only way to estimate the size of the stock. An example
11 See Appendix 2.1 for a brief introduction on integral-differential calculus.

12 Also, flows can be considered as variables that change over time. In calculus, this is the second
derivative.
13 The word stock in English also refers to a document that confers share of ownership, but with a
volatile value – as in the stock market.
is how demographers estimate medieval populations in European cities from baptism

and funeral data as proxies of birth and death rate (Braudel 1979).
To do model simulations, stocks identified in an influence or causal loop diagram
must be quantified and, therefore, be assigned a unit of measurement. Cars are
counted in numbers and roads and railways in km length. Elephants in a natural
park can, as a stock, be measured in number of individuals or in biomass-weight,
depending on the purpose of the study. Identifying a clear definition and unit of
measure can be difficult, notably with aggregate stocks of great heterogeneity such as
pollution, machinery or information. Scientists usually look for quantities that can be
measured and related to another, more interesting but difficult to measure variables –
such a substitute quantity is called a proxy. For instance, a proxy of economic capital
is the accumulated investments in manufacturing. Conceptualisation, quantification
and measurement are related but not the same. It is possible to conceptually enter
love and hate in a model, for example, to facilitate thinking about their relationship
and dynamic. They can be quantified as normalised variables, that is, fluctuating
between 0 (absent) and 1 (full intensity presence). To actually measure them is next
to impossible and one has to use proxies.
The level of a stock is determined by an inflow and outflow. A widely used
example, which also serves as an analogue, is the bathtub. It can be written math-
ematically in the form:
[WaterInBathtub]
= WaterInflowRate – WaterOutflowRate
t
with WaterInBathtub the stock variable, which should be given an initial value in a
simulation. WaterInflowRate and WaterOutflowRate are the associated flows. The
time interval t is the period over which the change in the state variable due to
the inflow and outflow are considered. This type of equation is called a difference
equation and, for t approaching zero (t→0), a differential equation (see Appendix
2.1). As long as the inflows equal the outflows, the level in the stock does not change
and the system is in a state of dynamic equilibrium.14 It is also called a stationary or
steady state. If the sum of the inflows exceeds the sum of the outflows, the stock level
rises and vice versa. It sounds trivial, yet it is often an essential and revealing first
step in system analysis.
Stock levels can only be influenced or managed by changing the inflows and
outflows – one cannot change their level directly. In the example of the bathtub,
this is clear. But there are many other, less evident situations. For instance, the
gradual emission of chemical compounds has raised their concentration in coastal
sea sediments and this can, at best, slowly disappear through natural mechanisms. As
a result, a system with stocks has inertia. The stock level cannot change or be changed
faster than the maximum difference between inflow and outflow. Stocks are a kind
of memory and are the source of delays. Sometimes, depending on the function, this
may not pose any problem, as with the slow refilling of a glass of beer for a patient
customer. But inertia may be a serious problem in situations when fast change is
desired or needed – think of the rate at which people can get out of a cinema in case
of fire. Inertia is experienced as an usually unexpected and undesired delay and is
14 If inflow and outflow are also zero, the system is in a state of static equilibrium.
a) 100
Male
91
Female
82
73
64
55
Age
46
37
28
19
10
1
15 10 5 0 5 10 15
Population (mln)
Figure 2.6a. Population pyramid for India in 2000: number of male (left) and female (right)
persons in a particular (1-year) age cohort (source of data: PBL).
often not adequately perceived and interpreted. The fact that in 2000 more than half
of India’s population was comprised of people less than twenty-six years old implies
a large and almost inevitable growth of the population in the first two decades of the
21st century – given the slow changes in reproductive behaviour and the probable
rise in life expectancy (Figure 2.6a). This is known as the demographic momentum.
It is in this sense that one can say that a moving oil tanker, extensive infrastructure
or social movement has a large momentum – it has much kinetic, economic or social
energy that is difficult to stop overnight. Similarly, it is the age distribution as well as
other characteristics, such as rent/lease or ownership, of dwellings and cars that are
important qualities in an assessment of energy savings potential and implementation
rate (Figure 2.6b). In environmental issues, for example, the climate change issue,
misunderstanding this can cause great confusion.
b) 8
<1945 1945-1969
1970-1979 1980-1989
Number of dwellings (mln)
1990-1999 ≥ 2000
6
0
1971 1975 1980 1985 1990 1995 2000 2005
Year of census
Figure 2.6b. Dwellings in the Netherlands: distribution according to age classes and type
(source of data: VROM/ABF 2009).
Box 2.2. Understanding climate change. An illustration of the difficulty to

appreciate stock-flow delays is provided by Sterman and Booth Sweeney (2002).
A widely supported hypothesis amongst scientists is that emission of carbondiox-
ide (CO2 ) into the atmosphere leads to warming of the earth surface (‘greenhouse
effect’). The natural mechanisms to remove CO2 from the atmosphere are slow
and a CO2 -molecule remains on average for about 100 years in the atmosphere.
Because the emission rate exceeds the removal rate, the concentration rises. This
causes a change in radiative forcing and warming.
Focusing on the climate system, Sterman and Booth Sweeney investigated
whether the (American) public understands the physical facts. To this purpose, a
group of graduate students were presented descriptions of greenhouse warming
drawn from the nontechnical reports of the Intergovernmental Panel on Climate
Change (IPCC) and subsequently asked to identify the likely response to various
scenarios for CO2 -emissions. The only understanding required for this task is the
basic stock-flow dynamics of the climate system and, in particular, the fact that at
present the rate of influx of greenhouse gas is twice the removal rate. It was found
that the students’ performance was poor: ‘Many believe temperature responds
immediately to changes in CO2 -emissions or concentrations. Still more believe
that stabilizing emissions near current rates would stabilize the climate, when in
fact emissions would continue to exceed removal, increasing GHG concentrations
and radiative forcing’.
2.3.3 Feedback Loops

There are a few elementary processes, which are basic in many systems and are
referred to as positive and negative feedback loops. Their essence is that a stock
influences, via a series of signals and decisions, the inflow and outflow rates of
the stock and hence its own level. The first one is the positive (or reinforcing or
amplifying) feedback loop (PFL). It is a dynamic growth process in which the rate
of change of a quantity is positively proportional to that quantity.15 Examples of
such processes are an autocatalytic chemical reaction (explosion), an unconstrained
population growth or the money on a bank account that ‘grows’ with the rate of
interest. Another example is a capital stock producing an output that is partly added
to the existing stock – it is considered the engine of economic growth in economic
theory. Such exponential growth processes go toward infinity for large t(ime). In
other words, infinity (∞) is the attractor, fixed point or equilibrium point for t→∞.
Of course, this is impossible for a physical system on a finite planet.
The PFL can for a stock variable X in its simplest form be phrased in mathem-
atical terms as a difference equation:
Xt+1 = Xt + a · Xt = (1 + a) · Xt (a > 0) (2.1a)
with a discrete time step of one between the events and a fractional (per unit)
growth rate a (a > 0). Such a discrete formulation is often appropriate, because
15 There is also the possibility that a system (organism) responds to an anticipated or projected future
trajectory, in which case one speaks of a feedforward loop (Rosen 1985). It can make system
behaviour significantly more complex.
many processes consist of a series of separate events. The amount of money on a

bank account with regular interest being added is a perfect example of continuous
exponential growth. Unconstrained population growth and a reproducing capital
stock are less perfect examples, because the change events in these processes are
usually not regular and can only in the aggregate be described by equation 2.1a.
There are also many processes in which the rate at which a quantity changes is
negatively proportional to that quantity. These are called a negative (or stabilising or
balancing) feedback loop (NFL). An example of such a decline process is radioactive
decay. The mathematical formulation is the same as for the PFL but with a negative
growth rate a. In its simplest discrete form it is the function Xt+1 = (1 + a)Xt with the
parameter a being a fractional (per unit) decline rate (−1 < a < 0). The equilibrium
state to which it eventually will go for large t(ime) may differ from zero. If the system
takes on the value K for for t →∞ in equilbrium, one can write it in mathematical
form as:
(X − K)t+1 = (X − K)t + a · (X − K)t = (1 + a) · (X − K)t ( − 1 < a < 0) (2.1b)
The system drives the stock variable towards the state X = K where, once reached,
the stock remains constant. K is the attractor. An example is the opening of a
sluice after which the high water level will exponentially decline to the low level.
Mathematically, the attractor represents a steady-state and as such a system state
that can intuitively be associated with a sustainable state. Examples are when the
birth rate of a population equals the death rate or the influx of a pollutant is equal
to the outflux. In the car system, it can be the state at which the purchase rate equals
the discard rate and the number of cars per thousand inhabitants remains constant,
also known as market saturation (Figure 2.5). In a management context, K can be
interpreted as a goal or target. If the stock is disturbed, the NFL will tend to restore
the system state to K.
The classical mathematical approach to describe exponential growth and decay
processes is in the form of a differential equation:
dY
= bY (2.1c)
dt
with dt an infinitesimally small timestep (t→dt→0), b the rate of change parameter

and Y the state variable.16 This equation can be solved analytically by taking the
integral over time of (dY/dt)/Y = b. This yields:
Yt = Y0 ebt + c (2.1d)
with c an integration constant. Exponential decline can be similarly expressed.17 The

advantage of an analytical solution is that the formulas and rules of calculus can be
used to explore the behaviour of the system.
16 In system theory, one usually speaks of the state equation, with Y the state variable. We denote the
state variable with Y in a differential equation. It is the same as the state variable X in a discrete
equation. See Appendix 2.1 for a brief introduction on the equivalence of continuous and discrete
equations. For a negative loop and a non-zero attractor K, Y should be replaced by Y-K. It can be
shown that b in equation 2.1d equals ln(1+a) in equation 2.1a.
17 For exponential decline, b < 0 and c = K.
a) 1200
900
Mln people
600
300
0
1700 1750 1800 1850 1900 1950 2000
Year
Figure 2.7a. Growth of the number of people in India, 1700–2000 (source of data: PBL).
Exponential growth and decline processes can be characterised by the doubling

and halving time, respectively, provided that the growth rate parameter a is constant,
which is not usually the case. The doubling time (DT) in exponential growth process is
the number of timesteps during which the state variable doubles. Doubling happens
in T timesteps if:
XT = 2X0 = (1 + a) · XT −1 = (1 + a)(1 + a)XT −2 · · · = (1 + a)T X0 (a > 0) (2.1e)
Thus, (1 + a)T = 2, from which it is seen that DT = log(2)/log(1 + a) with a > 0. The
rule of thumb that the doubling time is roughly seventy divided by the percentage
growth rate (DT ∼ 70/a%). It is easily checked. For instance, the number of cars in
China and India grows with about 10 percent/year, and it will, therefore, double in
about seven years.
The exponential decline process can be characterised by the number of timesteps
in which the stock halves: the halving time (HT). Similar to the doubling time, the
halving time is derived from the condition that XT = X0 /2 at time T. Therefore,
HT = log(0,5)/log(1 + a) = −log(2)/log(1 + a) with a < 0. The same rule of thumb
is valid. For example, if the number of traffic accidents decreases with 2 percent/year,
it will be almost halved after thirty-five years.
There are many real-world processes that follow, at least for a while, an
exponential growth path. Usually, the growth rate parameter a is not constant.
Figure 2.7a,b shows three examples: the growth of the human population in India
between 1700 and 2000, the number of Internet users since 1990 and the global wind-
power capacity installed since 1996. It should be noted here that there is a difference
between the examples. Population growth happens because each woman can give
birth to children so change depends directly on the population size. Internet users
and wind turbines do not breed young ones. Their growth is driven by more complex
phenomena, in which reinforcing feedback loops via, for instance, word-of-mouth
infrastructure networks and expected high profits are operating. There are also many
examples of exponential decline, such as the decay of the two radioactive elements
b) 2000 200
Mln Internet users
GWe windpower installed
1500 150
GWe installed
Mln users
1000 100
500 50
0 0
1990 1995 2000 2005 2010
Year
Figure 2.7b. Growth of the number of Internet users (upper) and of windpower capacity
installed in the world (lower), 1990–2009 (source of data: World Bank and www.ewea.org).
14
C and 99 Tc (Figure 2.8). The latter is an ingredient of high level waste from nuclear
reactors that is presently generated worldwide at a rate of about 12,000 tons per
year.18
Often, time-series consist of combinations of growth and decline. The curve in
Figure 2.9 depicts the number of casualties in traffic in the Netherlands between
1950 and 2005, with the government target for 2020. Initially, it increased steeply
in association with exponential growth in car ownership in the 1950s and 1960s.
It was driven by other, interconnected PFLs: population, affluence and technology.
Increasingly, the large and rising number of traffic deaths was felt to be unacceptable
and political targets were formulated and countermeasures were introduced. This
was an NFL with the implicit goal of zero accidents. With the obligation of safety
belts in new cars (1971–1975) and helmets for motorbikes (1975), a significant decline
set in despite a further growth of the number of vehicles and traffic. In 2009, the
number of traffic deaths was down to 720, some 5 percent below the target for 2010.
Policy has been very successful – as far as deadly accidents are concerned, as it
does not count the wounded – but the rate of decline slows down because a further
reduction costs proportionately more policy effort due to other negative feedbacks
such as aging and risk perception changes (Adams 1995).
Real-world systems will have both positive and negative feedback loops. The
relative dominance of positive and negative feedbacks will then determine whether,
in the time period considered, the system exhibits growth or decline or a combination
of both. A simple but characteristic dynamic process, in which both a PFL and an
NFL operate, is the logistic growth process. In system dynamic terms, the outflow
rate is a function of Y and approaches the inflow rate for Y approaching K. This
18 Such waste contains many different elements, most of which are very radioactive and have halving
times (half lives) between 5 years and 15 million years. The graph is representative for one of the
more important long-lived elements in nuclear fission reactor waste, technetium 99 Tc, with a halving
time of 211,000 years.
100
99
Tc activity (%)
14
C activity (%)
% remaining 80
60
40
20
0
0 120000 240000 360000
Time in years
Figure 2.8. Decline of the radioactive element of 99 Tc (technetium) and 14 C (carbon).
process can be formulated as the differential equation for exponential growth with
the growth rate b* approaching zero for Y ≈ K:

dY ∗ Y
=b Y =b 1− Y (2.2a)
dt K
Effectively, the growth rate itself is part of a negative feedback. It is easily seen that
the formula is the one for exponential growth for Y ≪ K. There are two attractors,
that is, Y-values for which dY/dt = 0, namely Y = 0 and Y = K. At Y = K, there
is a steady-state at which inflow cq. growth rate equals outflow cq. decline rate. The
functional solution to equation 2.2a is found by integrating and gives for the state
variable Y as a function of time:
K
Y= (2.2b)
1 + e−b(t−t0 )
with t0 the value at which Y = 1/2K (Figure 2.10). It is an S-shaped form, reflecting
a transition from positive feedback to negative feedback, hence its other name
4000
Traffic death
Policy target
3000
Number of people
2000
1000
0
1950 1970 1990 2010
Year
Figure 2.9. Number of casualties in traffic in the Netherlands, 1950–2007 (source of data:
CBS).
Value of X UK population dX/dt per 1,6yr
80 0.3
60 0.225
Mln people
40 0.15
20 0.075
0 0
1700 1810 1920 2030
Year
Figure 2.10. Example of a logistic growth process and estimates of the United Kingdom
population since 1750. The dotted curve shows the growth rate (source of data: Maddison
2006; Schandl and Krausmann 2007).
sigmoid curve. The value of the attractor K is associated with the carrying capacity
because it is the maximum value to which the variable Y (0 < Y < K) can grow.
The logistic growth equation is used in many branches of science to describe widely
different phenomena, such as population or biomass growth and market penetration
of products. We later discuss this equation in more detail.
In complex natural and societal processes, more and more interdependent pos-
itive and negative feedback processes take place than in the logistic growth model.
In environmental systems, there is a mix of natural growth and decay processes
and interfering processes induced by humans. In a technical and management/policy
context, the negative feedback forces are an essential control mechanism and the
equilibrium value is the desired or target value of the state variable. A technical
example from everyday life is the thermostat. It is set at a desired value, after which
the heating installation will start delivering heat at a rate that is proportional to the
difference between the measured and the desired temperature. Provided it func-
tions well, the rate of heat production will decline to zero when Tmeasured approaches
Tdesired . Another example is a market with the price of a good or service – pre-
sumably the same and known throughout the system – as an information signal. If
price goes up, demand tends to go down. Producers shut down the plants with the
highest marginal unit cost, which lowers the price and, in turn, stabilises or increases
demand again. Often, oscillations will occur and the demand-supply pattern looks
like a cobweb – which is why this simple model of the price mechanism is known
in economics as the cobweb model. Price regulation can be introduced to stabilise
prices again and prevent windfall profits or stimulate use.
2.3.4 An Illustrative Simulation Experiment

Two simple model experiments illustrate the phenomena of delay and inertia. A
car manufacturing company is selling cars to customers at a constant rate. To be
prepared for a change in demand, a desired inventory of ten days of perceived
sales is maintained. A larger inventory would be too costly, a smaller one too risky.
a)
40 4000
customer demand
orders to car manuf
30 3000
Rate (cars/month)
inventory demand ratio
Number of cars
car fleet
20 2000
10 1000
0 0
0 30 60 90 120 150 180 210 240 270 300
Month
Figure 2.11a. The customer demand, order rate of cars and the ratio of inventory and demand
in response to a 10 percent increase in demand in year 25 (timestep t = 1, Runge-Kutta-4).
The output of cars is based on orders for new cars, which is calculated from sales
and the difference between desired and actual inventory. If demand suddenly goes
up and the manufacturer responds instantly, there will be a sudden jump in cars
ordered; however, all customers will still be served because the inventory buffers the
change. Real-world people will not and cannot respond instantly and mechanically.
They decide on the basis of past experience and forward expectations and, therefore,
postpone or adjust their response in line with a rather complex mix of goals (minimise
cost and maximise revenues, keep good relations with customers and suppliers, and
so on). Therefore, we add to the model:
r perception delay: the manufacturer uses some average of past sales instead of
yesterday’s sales for this decision;
r response delay: the inventory level is not restored immediately and completely,
instead only a fraction of what would be needed is ordered;
r manufacturing and delivery delay: once ordered, it takes time before the new
cars are actually produced and arrive at the sales/inventory location.
Some delays are strategic and/or informational whereas other delays are of a more
physical-technical nature. How does the system respond with delays to a change in
demand?
Let us assume that the car market is initially in equilibrium, with a fleet of
3,600 cars, an average lifetime of 15 years and a monthly delivery of 20 new cars.
What happens if in year 3 (month 36) demand suddenly jumps from 20 to 25 new
cars/month? It turns out that a perception delay of 2 months diminishes the spike in
orders significantly and a response delay of 2 months adds to a further smoothing
of the spike. Both can be seen as good management. However, a delivery delay of
2 months causes fluctuations in orders of a factor 2 and reverberate in the system for
more than 3 years (Figure 2.11a). Another finding is that it takes more than 25 years
(month 300) before the system is again in equilibrium. Experiments with different
parameter settings show that the system is easily destabilised.
b) 2 4000
spec fuel use

(Specific) fuel use (normalized)
delayed spec fuel use

1.5 car fleet fuel use 3000
Number of cars
car fleet
1 2000
0.5 1000
0 0
0 30 60 90 120 150 180 210 240 270 300
Month
Figure 2.11b. The fuel use of the car fleet in response to a sudden doubling of fuel efficiency
in year 72 (timestep t = 1, Runge-Kutta-4).
This simple model run shows the importance of system delays in a changing
environment. What is said here for cars is also valid for the emission of pollutants
that are gradually broken down or for a chain of beer users, distributors and manu-
facturers (see the famous Beer Game, beergame.mit.edu/). It is valid for commodity
chains in general: apples and pears, chemical and steel plants and products, and for
financial products and services as the financial crises show. The core idea is that the
world is permanently in a state of dis(turbed) equilibrium.
A second experiment illustrates the inertia of a stock. The car market in the
model has a complete turnover in fifteen years. Suppose the fuel use of new cars
suddenly drops with 50 percent to half its value. Such a sudden change is implausible
to say the least, but it is instructive for this model experiment. If the drop takes place
in year 5 (month 60), it takes the full 15 years until year 20 (month 240) before the
average fuel efficiency has come down to the value of the new car released in year
5 (month 60) (Figure 2.11b). It is recognised in the historical data: The average fuel
use km/litre of passenger cars in the United States and Europe doubled between
1978 and 2008, but individual new car performance went up much faster. Again,
what is said for cars is true for long-lived substances in air, water and soil, and for
other capital goods with long lifetimes, such as dwellings, office buildings and electric
power plants with on average lifetimes of more than thirty years. The lesson from
a sustainability perspective is: decisions about emissions and investments throw a
large shadow into the future and that cleaning up environmental compartments or
making a capital stock more resource-efficient will necessarily take time.
2.4 System Dynamics Modelling
2.4.1 The Rules of the Game

Making a dynamic model of a system of interest is as much an art as a science.
It starts with genuine interest and careful observation. Although disciplined and
Box 2.3. A definition of sustainability in the vein of system dynamics is the one
proposed by the ecological economist Daly, in the form of three rules:
r ‘Its rate of use of renewable resources do not exceed their rates of regeneration;
r its rates of use of nonrenewable resources do not exceed the rate at which
sustainable renewable substitutes are developed; and
r its rates of pollution emission do not exceed the assimilative capacity of the
environment’ (Meadows et al. (1991) 209).
Because of the nonlinearity of the exponential and logistic growth functions,
a small change in one variable can easily cause big changes in the same or other
variables over time. This key feature of nonlinear reinforcing loops, famous now
under the name tipping point, was already known long ago as is shown in this old
English proverb:
For lack of a nail, the shoe was lost;

For lack of a shoe, the horse was lost;
For lack of a horse, the rider was lost;
For lack of the rider, the battle was lost;
For lack of the battle, the kingdom was lost –
And all for the lack of a horseshoe nail!
formal analysis is always part of it, qualitative stories and accounts are important
too. If people talk about (un)sustainable development, they often come up with
a story they have heard or read or experienced. Seldomly, such a story ‘tells the
whole story’ and usually it is anecdotal. Yet, such narratives do often reveal essential
insights about what (un)sustainability is about. Moreover, a good story tells it in the
words of those involved and in the form of concrete, specific events. This makes
stories a precious part of sustainability science. This book uses them on various
occasions, as experiences to be investigated for their concreteness and lessons. In
a system dynamic modelling setting, they are in combination with statistical data
the basis for the first steps. These can be summarised with the following guidelines
(Bossel 1994; Sterman 2000):
r Develop a clear statement of the problem to be addressed, that is, of the purpose
of the model, and choose the time horizon of interest;
r make qualitative and quantitative descriptions of the system’s behaviour in the
past, and hoped-for or feared behaviour in the future, using a.o. the stories and
accounts;
r develop hypotheses about the dynamics driving the system’s behaviour and
express them as an influence diagram or a causal loop diagram (CLD).
At this stage, one should have a conceptual model that forms the basis for more
formal modelling. The next steps are to identify stock variables, that is, the relevant
inflows and outflows and those variables that change as a consequence of these flows
(or fluxes). The value of a stock variable is referred to as its level and of a flow as its
rate of change. The identification of the variables and their interrelations and their
representation in a stock-flow CLD is important. The notion of causal loop is not
unproblematic, but here we stick to a rather straightforward interpretation: It is when

a change in a variable (stock, flow) A causes changes in another variable B, which
in turn changes A. Be aware of the notation: A connection between two variables
(stocks, flows) is indicated with a + sign if a positive (negative) change in variable A
causes a positive (negative) change in variable B. If the reverse happens: a positive
(negative) change in variable A causes a negative (positive) change in variable B, we
use a – sign. If a loop has only + signs, it is always a positive (reinforcing) feedback
loop. If there are an uneven number of – signs, it is a negative (stabilising) feedback
loop.19 In Figure 2.12, a few simple CLDs are shown, all representing combinations
of positive (growth) and negative (decline) feedback loops, with the exception of
fossil fuel depletion because its formation rate (inflow) is assumed to be zero. We
will meet these elementary CLDs again in the thematic Chapters 10–14.
If the system is complex, one can eliminate some detail and divide the system
into subsystems to make analysis manageable. An important next step is to identify
the structure of the decisions or relationships that govern the rates of change, which
will create feedback loops amongst the stocks. In the process, the conceptual model
is adjusted and, hopefully, converging to a more definitive form. It can then be
translated into a mathematical simulation model. Variables are given values from
available observations and data and one can start experiments with the model in
order to determine typical dynamic behavioural modes. This is usually done by sim-
ulation experiments and by comparison with other systems thought to be analogues.
Table 2.2 gives a list of stocks and associated qualities and flows relevant in a
sustainability context. The examples are from all scientific disciplines. In the pro-
cess of stock and flow identification, one should consider their properties.20 For
instance, the ocean and the atmosphere are simulated in climate models as layers
of different temperature. A stock of mineral or fossil fuel resources is divided into
size/quality classes according to grade, location and cost. A stock of cars can be
classified according to age, purchase price, power and energy efficiency. A stock
of people can be distributed over age and gender classes (Figure 2.6a). Whether
and how to disaggregate stocks into classes or categories depends on the objective
of the analysis. A functional disaggregation is often useful because stocks provide
certain goods and services, such as how mineral deposits provide metals, ground-
water reserves provide water, forests provide timber, and fertile soils provide food.
Similarly, machinery delivers goods and females deliver offspring. The stock prop-
erties are usually an important determinant of its productivity, that is, unit of good
or service delivered per unit of stock.21 Stock productivity is usually influenced by
the inflow and outflow rates. Indeed, a key question in sustainable resource use is
whether resource productivity declines and, if so, how it can be increased or at least
maintained and at which effort/cost.
Disaggregation into (size/quality) classes should be done up to the level that is
still functional for the analysis. Age distribution in a population may be important
in considering long-term consumption and savings trends or the health situation.
19 Another convention is to indicate the positive loop (+) with the s of same and the negative loop (–)
with the o of opposite.
20 Also, flows can be considered as variables that change over time. In calculus, this is the second
derivative.
21 The reverse is called the resource intensity, that is, unit of stock per unit of output.
Figure 2.12. Examples of causal loop diagrams of basic dynamic processes: population growth,
capital stock growth, forest growth and fossil fuel depletion.
The age distribution (or vintage, in economic jargon) of economic capital influences,
amongst other factors, the potential rate of innovation. Distributions across space
are often crucial in analysing land change processes. For example, a distinction
between rural and urban populations can be important because of different birth
Table 2.2. Examples of stocks, associated flows and stock qualities
Stock Associated flows

Units of quantity Units of quantity/time Stock properties
Mineral deposits and fossil fuel Discovery rate Mineral ore grade
deposits (stocks in tons, flows Exploitation/production Depth/extension
in tons/yr) rate Reservoir characteristics
Distance from consumers
Forest: standing biomass (stocks Re/afforestation rate Species characteristics
in ton or area, flows in ton/yr Harvest/exploitation rate Age cohorts
or area/yr)
Agricultural soil (stocks in area Sedimentation rate Organic matter content
or depth, flows in area or Erosion rate Yield/productivity
depth change per yr) Crop output/harvest rate
Agricultural inputs
Groundwater occurrences Replenishment rate Depth/extension
(stocks in ton or m3 , flows in Withdrawal rate
ton/yr or m3 /yr)
Population size Birth rate Age cohorts
(animals/livestock) Death rate Metabolism
(stocks in number, flows in Hunting/slaughter rate Position in foodweb
number/yr)
Population size (humans) Birth rate Age cohorts
(stocks in number, flows in Death rate Health/morbidity
number/yr) Migration rate Family size
Income
Industrial/service capital stock (Expansion/replacement) Age cohorts (vintage)
size (factories, offices etc.) investment rate Lifetime
(stocks in €, flows in €/yr) Depreciation/demolition/ Productivity
scrapping rate
(Retrofit) investment rate
Household capital stock size (Expansion/replacement) Age cohorts (vintage)
(houses, cars etc.) investment rate Lifetime
(stocks in €, flows in €/yr) Depreciation/demolition/ (Average) specific energy
scrapping rate use
(Retrofit) investment rate
and death rates and different income (growth) rates. Spatial gradients in resource
quality or pollutant accumulation can be important in exploitation or restoration,
but it may be irrelevant in other analyses. As a general rule, disaggregation should be
done parsimoniously because it makes the model more complex and more difficult
to parameterise. It is usually best to start simple and add detail and complexity as
understanding advances.
A basic ingredient of models are relationships between variables. These are
usually derived from experiments, observations and statistical data on the one hand
and intuition and ‘educated guesses’ on the other. Sometimes there is a direct causal
relationship that can be put into a formal model. For example, the water level in the
bath has a simple causal relationship with the inflow and the average annual fuel use
in car travel is causally related to the distance driven, although the relationship is not
necessarily known exactly. Often, comparison of datasets suggests a relationship for
which no obvious causal mechanism is available. Such a correlation can be considered
Box 2.4. Causality. The notion of causality is strongly present in the natural
sciences. It implies that, even with uncertainties in the state description, the exist-
ence of a (one-directional) causal linkage exists and constrains the possible future
states; in other words, successive consistency. The inclusion of human beings with
conscious goal-seeking behaviour may suggest that a system develops towards an
end – a teleological explanation. However, such goal-seeking behaviour has to
be rooted in past experiences. In this sense, it does not contradict the initial-state
assumption in system science that the future state of the system has no influence
on process dynamics.
A causal chain is a way of representing a hypothesised sequence of causally
linked events. The events follow from the behaviour of interrelated and inter-
acting variables. It is important to distinguish between causality and correlation.
‘ . . . one of the central issues in the quantitative study of two variables: is there a
relationship between them? . . . The statistical term for such a relationship or asso-
ciation is correlation, and the measure of the strength of that relationship is called
the correlation coefficient . . . It must be stressed from the outset that correlation
does not imply causation. It is only some appropriate theory that can provide a
hypothesis that might lead us to believe that there is a causal relationship between
two variables’ (Feinstein and Thomas (2002) 71–72).
as a phenomenological ‘law’ that connects two observable quantities in ways that

are not understood at the level of the underlying system dynamics. A well-known
example in transport economics is Zahavi’s hypothesis, which states that the average
travel time budget (TTB) and travel money budget (TMB), that is, the time (hr/day)
and money (€/yr/cap) spent on travel by the average person, is remarkably constant
across cultures and over time (Schäfer et al. 2010). But correlations usually have no
direction of causality and no explanatory mechanism. What a modeller hopes to find
are regularities, where two variables that are far apart causally have a consistent
relationship due to the system’s equilibrium tendencies. From that perspective, a
regularity like Zahavi’s law has to be examined by building a model of individual
decision making that yields the law as an outcome. Indeed, proposing hypothetical
mechanisms to explain observed regularities and subsequently testing them with
(new) data is the core task and challenge of every modeller.
2.4.2 Archetypes
One of the rules mentioned in the preceding list is to identify typical dynamic modes
of behaviour. We have already seen exponential growth and decline, and logistic
growth. Senge has listed in his book The Fifth Discipline (1990) a series of such
generic dynamic mechanisms. He refers to them as system dynamics archetypes
or templates and applies them in strategic management and in policy formulation
and implementation. Constructing archetypes recognizes and exploits the fact that
systems that appear on the surface to be quite different may exhibit similar behaviour
because they share a common feedback structure. The archetypes can be used as
building blocks for a larger, more comprehensive understanding of system behaviour.
Figure 2.13. System archetypes: escalation (left) and eroding goals (right) (redrawn from
Senge 1990).
Usually, they are constructed and discussed in the form of conceptual models, but
they have an analytical equivalent in the form of coupled differential equations.
They offer a filter that can clarify complex processes – not more and not less.
A first archetype are the physical and social Limits to growth. A PFL causes
a (stock) variable to grow exponentially, but at some point counteracting forces
start to operate and slow down the positive feedback. Such growth-weakening or
stabilising forces start to operate when some condition in the system triggers an
until-then hidden mechanism. For instance, food shortages or declining resource
quality act as physical limits and rising prices or public dissatisfaction as social limits
to growing population. The onset of limits may happen suddenly due to the very
nature of exponential growth, and more so if there are nonlinear thresholds in the
way the system responds to such growth. The logistic growth phenomenon represents
the essence of this archetype. In a context of business or political competition, the
Limits to growth archetype can manifest itself as a brake on successful operation. A
way to avoid it is to anticipate the possible consequences of the rapid growth and
respond before the limitations start operating – a recipe for sustainable development.
A second archetype is Escalation (Figure 2.13 (left)). The positive feedback con-
sists in this case by setting one’s goals or targets for (the growth of) a (stock) variable
via comparison and competition with someone else who has the same ambition. The
action of a person or organisation or country A is driven by the difference between
the own situation and the situation of the other. The other is chosen as the yardstick
for comparison or the opponent for competition.22 Such a mechanism is beneficial,
if the desired goal or target benefits also the larger system. It may be the basis for a
successful expansion of an economy or penetration of a technology and an important
instrument in policies aiming at sustainable development. It may also be harmful,
22 The word benchmark is used to indicate a shared point of reference or standard by which others
may be measured or judged.
Box 2.5. Arms race. A famous example of the escalation archetype is the arms
race model. In 1919, it was first formulated by Richardson to understand one of the
causes of the First World War. The first assumption is that the arms expenditures
of nation A depend on the perceived expenditures on arms of nation B, and vice
versa. This fuels the escalation in arms expenditures. The second assumption is
that there is a ‘pacifist’ tendency in each nation to spend on ‘butter, not arms’,
which counteracts the reinforcing spiral of armaments expenditures. This model
has been tested for several historical situations – no evidence of an arms race was
found for the Greece-Turkey conflict but the India-Pakistan interaction suggests
an arms race (Dunne et al. 1999). Extensions of this admittedly oversimplistic
model have been proposed. It has heuristic value as a metaphor for corporate
business in capitalist economies. For instance, an extensive review of McKinsey
& Company of pharmaceutical sales force effectiveness stated (In Vivo, October
(2001) 74): ‘the leading pharmaceutical companies have driven that [phenomenal]
growth by engaging in an increasingly intense commercial “arms race” to shift
share to new, more efficacious therapies’. A British journalist called the first
decade of the 21st century one of ‘a consumption arms race’.
if the original goal or target negatively influences the larger system or gets lost and
means become ends. A crucial issue in escalation is to evaluate the measure used for
the judgment of each other’s actions and the assumptions and delays involved.
Two other archetypes describe the phenomenon of resistance against change,
which is a common feature in (social) systems: ‘ . . . resistance is a response by the
system, trying to maintain an implicit system goal. Until this goal is recognized, the
change effort is doomed to failure . . . Whenever there is “resistance to change”, you
can count on there being one or more “hidden” balancing processes . . . It almost
always arises from threats to traditional norms and ways of doing things. Often these
norms are woven into the fabric of established power relationships’ (Senge (1990)
88). The first one rests on the idea of Eroding or drifting goals (Figure 2.13 (right)).
When a person, organisation or country A has a certain goal it wishes to achieve,
indicators are formulated to measure the performance of the plans and actions in
reaching the goal. If a gap is observed between the desired and the actual situation,
one should take corrective action: Improve the actual situation until the goal is
met and the gap is closed. Unfortunately, such action takes time and may go against
ingrained habits and interests. It is tempting to go for an easier solution: lowering the
goal. Possible solutions to such erosion are an understanding of what are the driving
forces behind the goal and the allocation of the goal and its monitoring outside the
system.
A related archetype is the phenomenon of Shifting the burden (to the intervenor).
To manage a system towards a desired goal usually turns out to be difficult because
there are so many unanticipated and unintended side effects. Analysis of the problem
may point at a ‘fundamental’ solution, which is rather drastic and uncertain in its
consequences. There are usually also lower-risk strategies, which are less drastic and
easier to implement. Such symptomatic solutions and their inevitable side effects
may draw all the energy away from the real solution. In the process, the real cause
Box 2.6. Eroding goals and shifting the burden are occurring in more or less
outspoken form in environmental issues during stages of denial and controversy.
In 2005, the U.S. government decided to subsidise the production of ethanol and
biodiesel from corn in order to reduce the dependence of the U.S. transport sector
on imported oil. Most of the alternative fuel is produced from corn and ethanol
production increased sevenfold between 2000 and 2009 to almost 40 billion litres.
The aggregate worth of the subsidy for the biofuels industry was an estimated $92
billion between 2006 and 2012 (www.globalsubsidies.org). The effectiveness of
this policy from an energy security and environmental point of view is doubtful.
One interpretation is that the real problem – addiction to cheap oil – is addressed
by an ineffective solution, which does benefit special-interest groups. The biofuels
industry becomes stronger and the symptomatic solution may weaken attempts
to address the real problem through a fundamental solution, like raising the price
of oil. It is in this sense that subsidy can create addiction. Another example of
eroding goals or shifting the burden is development aid. Although this is surely
not the whole story about development aid, it is true that aid may hinder more
structural solutions to problems of poverty and corruption (Moyo 2009). Often,
the burden is shifted by blaming an external intervenor.
of the problem may move out of sight – the burden has shifted. What is needed is a
step back in order to formulate a more comprehensive perspective on what was and
is going on.
2.4.3 An Example: Modelling Car and Public Transport Use

In this last section, I return to the car system and construct a conceptual model of a
few important loops. The car system has gained a dominant position, not in the least
because the costs of pollution, noise, accidents and congestion are not or only partly
internalised. Also the opportunity costs of road area and the hidden subsidies on
parking and fuel are at best partly accounted for in the costs of car driving. It makes
other modes of transport less attractive, not only economically but also regarding
other quality of life aspects (health, safety, esthetics). In the longer run, it causes
people to move to suburbs, which further increases the need for vehicles, erodes
the tax base for mass transit systems and makes urban centres even less attractive.
In the low-density suburbs, neither walking and bicycling nor public mass transit
transportation systems are attractive options and the car mode becomes a lock-in. I
focus in this complex process, which has unfolded in the United States to its extreme,
on two subsystems only: the interaction between car ownership, travel time and road
construction, and the interference between the car and public transport (Sterman
2000). Figure 2.14 shows CLDs that operate in the transport system. I indicate three
stocks explicitly: roads, car fleet and public transport fleet. Each of these causes
inertia in the system.
There are four loops in the CLD. The first one is the road-congestion loop. It
portrays the incentive to construct new road capacity in response to increasing travel
time due to congestion and the subsequent public pressure to do something about
it. It is in essence a Limits to Growth archetype. Given a desired travel time, the
44
Figure 2.14. A causal loop diagram (CLD) of parts of the transport system. The four loops are indicated in bold (based on Sterman 2000).
system would experience an increase in road capacity over time. Because there is
a long delay in road construction and because desired travel time may change due
to external factors,23 one can expect the system to show overshoot and undershoot
behaviour (Figure 2.12).
Congestion decreases the attractiveness of car travel, for most people. Once the
actual travel time is at or even below the desired travel time, average car speed will
go up and the number and length of car trips will probably go up too (car use loop).
This will increase travel volume. It will also induce people to buy more cars, which,
in combination with increasing population and economic activity in the region, is
another drive towards larger traffic volume (car ownership loop). As a consequence,
the system will start to again exhibit more congestion and longer travel times.
Finally, there is a fourth public or public transport loop. If the actual travel time
by car exceeds the desired travel time, the attractiveness of car driving goes down.
People will look around for alternatives and choose the available bus, tram, metro
or train or a combination of these and the car (or even walking or bicycling). It
increases the use of public transport, which, at least in the short term, leads to lower
service levels due to overcrowding, seat capacity problems and the like. This exerts
a downward pressure on use (stabilising), but it also tends to increase utilisation rate
and, therefore, decrease the cost and, hopefully, the price of a trip (reinforcing). In
the longer term, the response is to invest in additional public transport capacity. If this
translates into better coverage and service levels, the public transport attractiveness
increases relative to car driving.
Clearly, quite different pathways are possible, as the large differences in trans-
port infrastructure between European cities and American cities show. A lock-in for
public transport at a low level is sketched by Sterman (2000) and labelled the Mass
Transit Death Spiral. Because public transport systems have rather high fixed costs
(infrastructure, personnel), the operating company is faced with a budget deficit
once the number of passengers drops. If ticket prices increase and service and fre-
quency levels reduce in response, the number of passengers drops even further and a
downward spiral sets in. Constructing more roads reinforces the negative spiral. The
mechanism is particularly hard to overcome if the public transport service degrades
because of car congestion. In many urban areas in the world, people are accepting
ever longer car and bus travel times in a dynamic process that is characteristic of an
eroding goals archetype (Figure 2.13).
A very different process can unfold, too. The public transport authority plans
for and construct new infrastructure. It takes time, as with road construction. It
needs public hearings and construction permits. The construction of buslanes in
the city of Utrecht took more than ten years from inception to construction. It
also requires upfront investments for which loans and, therefore, the prospect of
steady revenues are needed. At this point, failure often sets in due to political and
institutional obstacles – for instance, lack of financing due to creditor’s desire for
high and fast profits. But if the local authorities and the public or private operator
succeed in overcoming these obstacles – for example, by attracting more people – a
positive loop starts. The utilisation rate goes up, revenues go up and there is room for
23 One can think here of the availability of audio/video, mobile phones and GPS-navigation, which
influence the perception of time spent in the car and the options to control it.
new infrastructure such as bus lanes and greater schedule/route coverage, and other
ways to attract passengers such as increased reliability, cheaper tickets, electronic
payment, additional comfort and so on. Such a positive loop can be reinforced by
popular concern about local air pollution and inner city destruction and by high
parking fees as the price of limited space goes up. These dynamic processes create
a lock-in effect for public transport at a high level. It can be observed in many
European cities.24 This conceptual scheme, simple as it may be, highlights some of
the behavioural responses in the transport system that are overlooked in purely
technical analyses. It explains why such systems often show oscillatory behaviour
over time and why well-intended policy measures are sometimes ineffective or even
counterproductive. It also reveals some of the intricacies of operating a ‘public goods’
system, an issue of great concern in sustainability science.
2.5 Structure, Space and Time

One last word about concepts. In media accounts of what happens in the world,
there is usually a focus on events. For instance, it is observed that in an area in the
Mediterranean oak trees die prematurely – an event. It is often followed by more
systematic observations and measurements. For instance, the tree die-off is traced
to declining groundwater levels, which in turn are caused by intensified exploitation
for kiwi and citrus in nearby regions. Thus, the behaviour (policies, mechanisms)
behind the event operates within the wider system of farming and groundwater.
But why and how does it operate? A combination of experiment and theory has to
investigate the underlying rules and relationships in the system and propose a system
structure. As it turns out, kiwi and citrus farming is driven by perverse subsidies and
the need to compete with other producers, which favours short-term exploitation
of groundwater over strategies that are more sustainable in the long-term. System
dynamics modelling aims at understanding system behaviour and structure. Structure
will often be influenced by changes outside the system: the system’s environment.
Let us look at another example. In early 2008, the steep increase in the price
of some staple foods on the world market is an event, which triggered all kinds
of behaviour: price controls and rationing to prevent riots, speculation by large
international companies, export quota and food aid initiatives. Behind these are
structural trends such as chronic underinvestment in agriculture, which may have
been reinforced by policies like food subsidies and trade policies. An essential part
of systems thinking and systems understanding is to consider the larger system, in
which long-term trends and large-scale changes can be investigated as the drivers of
behaviour and as the background of events. In short: inclusive thinking.
Events, behaviour and structure all have to be considered at various scales in
space and time. The choice of scale is an important part of the modelling activity
and usually directly related to the modeller’s objectives. In space, every item can
be investigated at a micro, smaller scale and at a macro, larger scale – its grain or
resolution. Everyday observations are usually in a range of 10−3 to 106 meter (m)
and can be indicated as phenomenological and coarse-grained. There are also many
24 The attempts in many European cities to introduce bicycles, with electronic ticketing and road use
privileges, is an illustration of the use of system dynamic insights.
phenomena at much smaller than the human scale, such as biological processes in
living cells (<10−3 m). Assuming a nested and hierarchical organisation of existence,
observations and analyses at such a more fundamental and fine-grained level are
necessary to understand the human scale processes. In science and technology, it is
the domain of bio- and nanotechnology. There is also a level of observations and
analysis at much larger than the human scale, for instance, about some global change
processes (>106 m) or stellar nebulae (>1014 m). These can also be studied at a fine-
grained and at a coarse-grained level, depending on the objectives of the analysis.
As to time, every event can be investigated at an earlier, previous time inter-
val and at a later, posterior time interval – the time scale or time horizon. It is
important to realise that only the here and now is lived reality. All else is memory,
imagination and mental construct. But in commerce and science, time has been
abstracted and objectified: It is a line with events on it as markers. We call the
time elapsed in-between events a period. Everyday experience deals with periods in
the range of 1 to 108 seconds (sec). As in space, there are coarse-grained and fine-
grained observations. The coarse-grained ones make sense for slower processes, for
instance, demographic change (>108 sec) or geological phenomena (>1013 sec). The
fine-grained descriptions are needed if one wishes to consider the faster processes,
such as some chemical reactions (∼10−3 sec) or electromagnetic and atomic particle
phenomena (10−6 sec).
Table 2.3 summarises some space-time scales for various relevant domains. The
categorisation is indicative, because most domains have their specific ranges in space
and time, and there can be changes in the categorisation itself, as with technology or
politics. With spatial scale, the choice of timescales has to be commensurate with the
timescales inherent to the system. If one is interested in the long-term future of the
car system, for instance in relation to climate change, one is naturally interested in
the macroscale in space and the slow variables in time. At the macroscale, there are
clusters of phenomena such as ocean warming/cooling, glacial epochs and tectonic
shifts (Turner et al. 1990). To understand the corresponding systems, however, one
often has to go ‘down’ to the faster processes at the smaller scale, such as the urban
road construction or vegetation change. This, in turn, may lead the analysis to yet
higher resolution in space and time, but there is a logical end to this for a given
objective. For instance, the daily trajectory of an individual car driver or the hourly
features of a large storm or a forest fire are hidden in more aggregate descriptions.
In the background, lurk questions like the universality of the various laws and the
degree to which ‘history matters’ in complex nested dynamic systems.
2.6 Summary Points

This chapter introduced system thinking as a tool to look in an inherently transdiscip-
linary way at the world. The world car system is used to show the many interrelated
aspects of a system. Points to remember from this chapter:
r the essence of system analysis/dynamics is to see the world as ensembles of

interrelated and interacting elements;
r a system description contains: system boundary, elements, interconnections and
purpose or goal (although on the latter there is difference of opinion);
Table 2.3. Phenomena at coarse-grained and fine-grained scales and

at different scales in space and time
Phenomenological coarse-grained Fundamental fine-grained
Temperature, pressure Atom/molecule dynamics

Phase transition Minimisation of molecular energy
Tsunami (storm surge), cyclone Atmospheric particle dynamics
Spiral cloud Star dynamics
Forest; ecosystem Trees; plants and animals
Markets Individual transactions
Fast, local Intermediate Slow, global
Earth: day-night and Erosion and salinisation Earth:

seasonal fluctuations (wind/water); River orbital forcing, tectonic
formation/change and forcing; mineral
sedimentation ore/fossil fuel formation
Weather: El Niño/Southern and Climate:
thunderstorm, tsunami Northern Atlantic glaciation, global warming;
(storm surge), cyclone; Oscillation atmospheric circulation
particle dispersion, patterns
Ecosystems: Habitat change, succesional Life: evolution, species
photosynthesis, change, forest dynamics, extinction
metabolism most ecosystem service
dynamics
Population: Age distribution Genetic change
birth, death
Economy: consumption/ Capital goods lifetime, Economy:
production, market labour-capital substitution, system paradigm;
prices and transactions business cycles capital stocks (buildings,
canals, rail/roads, etc.)
Technology: R&D induced change, Major technological waves
incremental innovations, process/product (Kondratiev)
marketing substitution, breakthrough
innovations
(Re)sources: Mineral and fossil fuel Sinks: substitution
production process, mining/depletion; processes, most
market price long-term environmental environmental sink
sink (greenhousegases, (pollution) dynamics
ozone layer) dynamics
Political: events, elections, Value changes in population, Formation of nations,
office time of most CEOs change in governance long-term historical/
and members of structure (laws, customs, cultural change, empires
government etc.)
r key notions in system dynamics are stocks, flows, feedbacks, delays, inertia and
causal loop diagrams (CLDs);
r the behaviour of systems over time can be understood by identifying the relevant
stock variables and the inflows and outflows for a chosen system boundary and by
considering both the energy/material aspects and the behavioural/social aspects
of the system elements and their interactions;
Suggested Reading 49
r two elementary modes of system behaviour are exponential growth (posit-

ive feedback) and exponential decline/growth towards an attractor (negative
feedback);
r real-world systems constitute of interacting positive and negative feedback
loops, with the logistic growth as one simple but frequently used model;
r the important steps in the construction of a system dynamics model are: formu-
late the problem, make a conceptual (stock-flow) model, collect relevant data,
implement the model in equations/software and run and test the model;
r some phenomena are apparently quite different but turn out, from a system
perspective, to share generic dynamic properties, and can be represented in
archetypical models; and
r system dynamic understanding and modelling skills are helpful for the remainder
of this book.
Systems thinking is a discipline for seeing the ‘structures’ that underlie complex
sitiuations, and for discerning high from low leverage change. (Senge (1990) 69)
One of the central insights of systems theory, as central as the observation that
systems largely cause their own behaviour, is that systems with similar feedback
structures produce similar dynamic behaviours, even if the outward appearance of
these systems is completely dissimilar. (Meadows (2008) 51)
Wo aber Gefahr ist, wächst das Rettende auch.
– Friedrich Hölderlin (1802), from the poem Patmos
When the people of the Earth
All know beauty as beauty
There arises ugliness
Lao Tze, Tao Te Jing
SUGGESTED READING
An introduction into dynamic models: a gradual build-up from elementary models to more
sophisticated ones, with an emphasis on population-environment issues.
Bossel, H. Modeling and Simulation. Ltd./Vieweg, 1994.
A practical introduction to system dynamics modelling, with emphasis on natural systems and
policy intervention.
Ford, A. Modeling the Environment. Washington, DC: Island Press, 2009.
An elementary introduction in system thinking, in clear language and with examples to practice
it in a sustainability context.
Meadows, D. Thinking in Systems. Vermont: Chelsea Green Publishing, 2008.
A historical investigation into the sources of system thinking in the social sciences.
Richardson, G. P. Feedback Thought in Social Science and Systems Theory. Philadelphia:
University of Pennsylvania Press, 1991.
Introduction to mathematical modelling, from simple to advanced.
Robinson, J. Ordinary Differential Equations. New York: Cambridge University Press,
2004.
An introduction into system thinking applied to management issues and extensive discussion of
the nature and role of mental models in decision making.
Senge, P. The Fifth Discipline – The Art & Practice of The Learning Organization. New York:
Doubleday Currency, 1990.
The most thorough and comprehensive book on how system thinking can and should be
applied in a variety of applications, from physics to management, with ample sample models
and discussion of the underlying mathematic and simulation features.
Sterman, J. Business Dynamics – Systems Thinking and Modeling for a Complex World.
Boston: McGraw Hill, 2000.
USEFUL WEBSITES
r www.sutp.org/index.php?option=com content&task=view&id=426&Itemid=72&lang=
en offers a Sourcebook on Sustainable Urban Transport for developing cities in a truly
integrated fashion.
r www.iseesystems.com/ is a site about the Systems Thinking software packages StellaTM
and iThink.
r thesystemsthinker.com/systemsthinkinglearn.html is a site on the background and con-
tent of systems thinking.
r www.systemswiki.org/index.php?title=Category:Model is useful to explore and run mod-
els, if the simulation software package is available.
r www.calculusapplets.com/growthdecay.html is a site where you can play interactively
with various growth functions: type in the expression for an exponential or logistic func-
tion and explore its behaviour.
r www.aw-bc.com/ide/idefiles/navigation/main.html is a site that permits exploration of dif-
ferential equations interactively, for instance, exponential growth and decay and logistic
growth models.
r www.metasd.com/sdbookmarks.html is a site operated by Tom Fiddaman with many
environment-energy-economy links and presentations and a system dynamics model
library.
r www.systemdynamics.org/ is the site of the system dynamics society with an explanation
of the most important concepts.
SOFTWARE PACKAGES
r StellaTM (www.hps-inc.com) is the earliest and simplest-to-use simulation software pack-
age for system dynamics modeling.
r VensimTM (www.vensim.com) is similar to StellaTM , has a free demo download, is some-
what less simple to use and has in its professional version interesting model analysis
features.
r forio.com/simulate/e.pruyt/ is a freely downloadable system dynamics simulation soft-
ware package. The site is operated by Erik Pruyt (Delft University) and has a variety of
models and simulations.
r www-binf.bio.uu.nl/rdb/grind.html is a freely downloadable software package to analyse
ordinary differentiual equations, with application in ecology. The site is operated by Rob
de Boer (Utrecht University).
r www.simulistics.com/products/gallery/one.htm is the site of the commecial simulation
software package Simile. The site has a model gallery.
r www.wolframalpha.com/ is a site that gives detailed information on any mathematical
function.
Appendix 2.1 Integral-Differential Calculus

Exploring sustainable development, we are primarily interested in change. Systems
analysis postulates that every system, given a well-defined boundary and initial and
boundary conditions, can be described in the form of 1..n state variables at time t:
= X1 (t ), X2 (t ) · · · Xn (t )
X (A2.1a)
X is a vector in n-dimensional state space. Observations of the system during a

period T provide a dataset X that represents system states at the moments in time
{t0 , t1 . . . T}. If the intervals between two observed states are the same and equal to
t, then the change of the system can be described with a set of discrete difference
equations of the form:
Xi (t + t ) = Fi (X1 (t ), X2 (t ) · · · Xn (t ); ; t ) i = 1..n (A2.1b)
with X the state variables and a set of constitutive and environmental parameters.
The Fi express how each state variable depends on itself and the other state variables
and on external parameters () and time (t). If we assume that the interval t
becomes ever smaller (t → 0), the set of difference equations becomes a set of
(simultaneous) differential equations:
dXi
= Fi (X1 , X2 · · · Xn ; ; t ) i = 1..n (A2.1c)
dt
with the d indicating an infinitesimally small change and Fi a set of analytical func-
tions. The representation of all possible states of the system is called phase space.
Without exogenous or autonomous change and only two state variables, equation
A2.1c becomes:
dX1 /dt = F (X1 , X2 )
dX2 /dt = G(X1 , X2 ) (A2.2a)
with F and G functions. dXi /dt are called (time-)derivatives. The system is in steady-
state or stationary state, when dXi /dt = 0 for i = 1,2. Such a state is called an attractor.
It is often depicted qualitatively as a ball in a landscape, in which the lowest point
is the attractor because that is where one intuitively expects the ball to move to
(Figure A2.1). In an attractor, the rate of change is zero and the tangent is, therefore,
a horizontal line. Wherever the rate of change has an extremum, the system is in a
state that can flip either way due to a minor perturbation. The landscapes a and b
in Figure A2.1 indicate the difference between a stable and an unstable attractor.
The landscape c represents the more complex systems one can expect in real-world
systems.25 The attractor is found for F(X1 , X2 ) = 0 and G(X1 , X2 ) = 0 in equation
A2.2a. Upon perturbation, the system will move in first approximation according to
the laws of simple positive/negative feedbacks towards the nearest attractor.
For simplicity, assume that the functions F and G are linear in the state variables.
Equation A2.2a becomes a set of linear first-order ordinary differential equations
(ODEs):
n

dXi /dt = αi j X j i = 1, 2 (A2.2b)
j=1
and the steady-state occurs for:

dX1
= α11 X1 + α12 X2 = 0 (A2.2c)
dt
dX2
= α21 X1 + α22 X2 = 0
dt
25 The situation is often even more complex: the landscape itself changes in response to the movements
of the ball – one of the topics in evolutionary modelling.
Figure A2.1. Landscapes with attractors. The upper graphs represent the system state as a
ball in a (gravitational) forcefield. The lower graphs show the corresponding phase space
diagrams of position X and rate of change in position (velocity) dX/dt.
The curves of X2 as function of X1 in phase space are called isoclines. The point at
which the two isoclines intersect indicates the attractor. For the system according
to equation A2.2c, the two linear isoclines are shown in Figure A2.2. This simple
graph can tell much about the system behaviour over time – at least, inasfar as the
model is correct. Points on the isocline dXi /dt = 0 are system states for which Xi is
constant and Xnon-i is not constant. Inspection of the different system states show
that the phase space is divided in four separate parts, in each of which the system has
a certain dynamic tendency indicated with the arrows. Thus, for any (initial) state
Figure A2.2. Representation of the two

isoclines in phase space. The arrows indic-
ate the dynamic tendency of the system.
The intersection of the two isoclines is the
steady-state. This particular representation
is drawn on the assumption that α ij > 0 for
all i,j.
of the system, we can derive how the state variables will evolve over time. At the
intersection of the two isoclines is the attractor of the system {X1 *,X2 *}. It is the
state where both X1 and X2 do not change – a sustainable state in the sense that
the net change of the state variables is zero. If no intersection point is found within
the permissible domain of Xi -values, there is no finite, meaningful attractor.
An important question is whether a steady-state is a stable or unstable equi-
librium. Stability analysis is an important part of integral-differential calculus. The
coefficients α ij can be written in matrix form and represent the so-called Jacobian
matrix A. In the general description of equation A2.2a, it is defined as:

∂F/∂X1 ∂F/∂X2
A=
∂G/∂X1 ∂G/∂X2
with ∂ indicating partial derivation. Under the assumption of linearity (equation
A2.2b) it becomes:

a a12
A = 11
a21 a22
The diagional elements α ii are equivalent to the growth/decline parameters and
determine the internal feedback processes of Xi . The non-diagonal elements repres-
ent interactions between the Xi . The nature of an attractor can be seen from the value
of the trace and the determinant of the Jacobian matrix. Matrix algebra can be used
to solve equation A2.2. It gives the characterictic (or eigen) equation and values. The
solutions are sums of exponential functions or products of oscillatory functions with
exponentials. If one or both eigenvalues are positive, there is growth over time. If
both are negative, there is decline over time. If there are complex eigenvalues, there
are oscillations and the real part of the solution determines whether the amplitude
increases, remains constant or decreases over time.
Only the simplest phenomena can be described by linear models. Most real-
world phenomena are more complex and have to be approximated with nonlinear
models. To explore system behaviour, one technique is to expand the state equation
around the equilibrium state and retain only the linear terms. Such a linearised sta-
bility analysis is only valid near the equilibrium state, but that is the most interesting
state anyway. The dashed curves in Figure A2.2 indicate the isoclines for a nonlin-
ear system. In that case, the linear curves are the linearised isoclines. Refer to the
suggested literature for further study.
3 In Search of Sustainability: Past Civilisations
3.1 Introduction
Early human groups were completely dependent upon their natural environment.
They were confronted with changes of catastrophic immediacy such as earthquakes,
and slow but no less impactful issues, such as changing courses or drying out of
rivers and changes in climate and vegetation. These changes presented threats as
well as opportunities and risks as well as challenges. Populations responded with
outmigration into new areas with better opportunities or with attempts at increasing
control of plants and animals. Indeed, the capacity to adapt to a variety of environ-
mental situations may well be the most remarkable characteristic of the homo faber
sapiens: ‘All mankind shares a unique ability to adapt to circumstances and resolve
the problems of survival. It was this talent that carried successive generations of
people into many niches of environmental opportunity that the world has to offer –
from forest to grassland, desert, seashore and icecap. And in each case, people
developed ways of life appropriate to the particular habitats and circumstances
they encountered. Farming, fishing, hunting, herding and technology are all expres-
sions of the adaptive talent that has sustained mankind thus far’ (Reader (1988)
7–8).
With the broad brush of Big History, one can distinguish several regimes in
the existence of homo on earth. The first one was the fire regime, in which the
control of fire was the determining feature (Goudsblom 1992). The transition from
hunting-gathering to agriculture and herding was the second large regime shift.
This transition is called the agrarianisation process, a better name than agricultural
revolution because it was a rather slow process with local characteristics. Humans
gradually expanded into nearly all corners of the earth, a phenomenon we call extens-
ive growth, and created increasingly larger opportunities for more efficient exploit-
ation of resources, the equivalent of intensive growth. It led to stronger interactions
between humans and their natural environment, with a more sedentary lifestyle and
new techniques and forms of social organisation. In particular, urbanisation did allow
activities and exchanges that prompted relationships and innovations that might not
otherwise have occurred.
54
3.1 Introduction 55
Prior to 15000 (Before Present) BP1 , almost the entire interior of southwest Asia
was covered by a desert-like steppe. Around that time, vegetation started to become
richer across the Fertile Crescent as a consequence of climatic change. Wild cereals
and grasses spread and were followed by oak-dominated woodlands. It offered
local foragers a diverse array of wild plant-foods that could provide the necessary
ingredients for a healthy human diet at relatively little energy expenditure. The
prevailing current view is that agriculture originated at least 12000 BP in the so-called
Levantine Corridor, near the Jordan valley lakes, in the form of the domestication
of cereals and pulses. Within a millennium, there was also domestication of pigs,
goats and sheep. But there were also early agro-pastoral economies outside the
Levant, for instance in the semi-arid areas surrounding the Iranian Plateau around
10000 BP. There is increasing evidence of very early, probably independent origins
of agriculture in China and in other places in Asia and the Americas.
Recent data from pollen diagrams seem to confirm the view of Sahlins (1972)
that the life of early hunter-gatherers may have been more pleasant than that of the
early farmers. Why then did the shift from gathering wild plant-foods (foraging) to
crop production (horti-/agriculture) and from hunting to protective herding and rais-
ing livestock (pastoralism) happen? One theory is that agriarianisation was in part
a response to deteriorating conditions for foraging. It initiated a series of changes.
People became less mobile, birth-spacing was reduced and women had more chil-
dren. The resulting higher fertility rates reinforced the formation of settlements and
the need for food. Mortality rates probably went up, too, because of higher incidence
of diseases due to rising temperature and more intense human-animal contact. The
net result was a slow but persistent population growth. The causal loop diagram
(CLD) in Figure 3.1 shows elements of the agrarianisation process.
In characteristic economic reasoning, Weisdorf (2005) suggests that hunting
and foraging had originally a higher labour productivity (food provision per day)
than growing crops, but that it started to decline due to population growth and
a subsequent increase in competition for food and depletion of the food resource
(game, vegetation). As a consequence, farming became in some places at the margin
more productive than hunting-gathering. Changes in organisation and skills did
play a role as well. For instance, exogenous improvements in food procurement
technology may have triggered the shift to farming and created room for nonfood-
related skills and the seed of economic growth.
This chapter explores the rise and fall of human civilisations in order to under-
stand better the mechanisms of (un)sustainable development. The focus is on the role
of the natural environment and the interaction with social and economic develop-
ments in both growth and decline of states and civilisations. Historians, philosophers
and social scientists have thought and written extensively on ‘the rise and fall’ of
civilisations. In the last decades, natural science methods and data collection and
processing equipment have enormously expanded the data on sediments, ice cores,
1 BP: years Before Present. The reference for Present is the year 1950 CE: Christian Era. Some
scientists prefer the consistent use of CE and BCE: Before Christian Era. In older texts, one still
finds BC: Before Christ, and AD: Anno Domini. Dating is often based on carbon isotope analysis,
which is a statistical procedure with a considerable margin of error.
Figure 3.1. A causal loop diagram (CLD) sketching some major dynamic factors in the early
agrarianisation process. Intensified food procurement was the determining positive feedback.
tree rings and so on. It allows reconstruction of present and past climate and vegeta-
tion characteristics and, in some regions, it has been possible to derive rather detailed
reconstructions of people’s diets and diseases from the environmental conditions.
In combination with archaeological and historical analyses and the use of com-
puter simulation methods, understanding of past human-environment interactions
has deepened significantly.
3.2 The Beginnings: Two Environmental Tales
3.2.1 The Hohokam People in Arizona2

Early in the 3rd millennium BP, maize farmers settled in the river valleys in places
suitable for floodwater farming in central and southern Arizona. The large empty
surroundings provided a sustainable flow of wild food and material resources and
the localised nature of accessible surface water and farmland stimulated settlement.
Known as the Hohokam people, these early farmers depended on the nutrients
brought down in flood periods by the two rivers traversing the Phoenix lowland
basin. Around 1250 BP, they developed irrigation agriculture, which tied households
together into community networks with ritual and belief systems. By 800 BP, larger
villages with platform mounds began to appear and canal systems linked communities
into larger networks. Hundreds of canals led the water and sediments to the fields,
thus regenerating soil fertility and enabling the use of the productive potential of
their immediate surroundings. The labour investments in canals and population
growth reinforced living in settlements. Social institutions and community networks
2 This account is largely based on Redman (1999).
evolved in order to maintain the canals and to spread the risks of food shortages.
But around 550 BP, the Hohokam culture had disappeared. Why and how?
The Hohokam way of life – no domesticated animals and modest use of wood –
tended to spare the vegetation. However, despite a way of life apparently based
on principles of sustainability, archaeological evidence suggests that the increase in
population led to a scarcity of protein-rich food such as fish and rabbits. The first
factor at work was climatic change. Tree ring analyses indicate a rather high climate
predictability in the period 1250 to 900 BP without extreme floods or droughts.
Communities developed along the feeder canals that brought the water downstream
to the distribution channels. Population, organisation and trade all increased. Tree
ring evidence suggests that in subsequent centuries flood variability became larger
but these communities were resilient enough to handle droughts and floods through
storage, repair and trading. However, after 700 BP, climate became apparently more
erratic and the frequency of floods and droughts increased. More labour was needed
to repair and maintain irrigation structures. Crop losses in the valleys led farmers
to overplant in the good fields, reduce fallow periods and start using marginal soils.
This further lowered soil fertility and productivity.
Intense irrigation may have been a second cause of trouble. Because the
Hohokam remained in the same location, their activities unavoidably caused envir-
onmental changes such as increased runoff volume and velocity, subsequent soil
erosion and canal silting. Although they maintained soil fertility through various
conservation methods and by supplementing local food with goods brought in
by exchange systems, these climatic changes put additional pressure on the sys-
tem. Investing more labour in order to extract the maximum from the land made
the system probably more, not less vulnerable to climatic extremes. Socio-political
changes towards more centralised control and ceremonial activities may have been
a third factor in weakening resilience. Their disappearance, it seems, was due to
a dynamic interplay of environmental and social forces working on different time-
scales. For several other groups of people in the southwestern United States, such
as the Anasazi (‘the Ancient Ones’), similar past histories have been reconstructed
(Diamond 2006).
3.2.2 Easter Island3

In 1722 (Christian Era) CE on Easter Sunday, the Dutch admiral Roggeveen
‘discovered’ a small island in the Pacific Ocean that he gave the name ‘Easter Island’.
It has become an archetypical example of how a human population on a finite island
with a fragile environment can overexploit its environment and as a consequence
collapse. When Roggeveen set foot on the island, it had a society ‘with about 3000
people living in squalid reed huts or caves, engaged in almost perpetual warfare
and resorting to cannibalism in a desperate attempt to supplement the meagre food
supplies on the island’ (Ponting 1991). Scattered across the island were more than
600 massive stone statues of, on average, some 6 metres (m) high and weighing more
than 50 tons. The inhabitants of the island seemed incapable of carving and moving
statues and indicated they had no knowledge of how to do so. It was a mystery how
3 This account is largely based on Ponting (1991) and Diamond (2006).
they had ever reached such an outlying place: They only had small leaky canoes. It
was an even bigger mystery how they could ever have erected such huge statues in
large numbers.
Many explanations have been offered, some much more speculative than others.
By now, there is enough archaeological and palynological evidence to reconstruct a
plausible storyline (Diamond 2006). The first human occupants of Polynesian origin
arrived to the island around 900 CE. The land mass of about 180 km2 has a gentle
topography and a mild climate, with fertile soils of volcanic origin. But the island’s
windiness, little rain by Polynesian standards and relative isolation made it less than
an attractive place for humans. Estimates of human population numbers during the
heyday range from 6,000 to 30,000 (33 to 165 per km2 ); therefore, 15,000 inhabitants
might well have been possible. The inhabitants lived on a diet of mainly sweet
potatoes and chickens, which was nutritionally adequate and not demanding in terms
of labour. The dense and diverse vegetation – amongst them palm trees probably
related to the huge Chilean wine palm – provided trunks for canoes, construction
material for household goods, pole and thatch houses, firewood for heating and
cooking and for cremations, as well as delicious oily nuts.
Structured like other Polynesian islands, there were chiefs and commoners,
organised in rival clans. The combination of plenty of free time for ceremonial
activities and enduring competition and conflict between the tribal clans led to the
construction of the huge stone statues. The statues (moai) were hewn out of the rock
in quarries in the centre of the volcanic island, and transported over the still visible
roads to the coastal areas, where they were erected on top of stone platforms (ahu).
This technical tour de force absorbed not only enormous amounts of peasant labour –
obsidian stone axes being the only tools – but also required prodigious quantities of
timber rollers as there were no draught animals.
In the end, the increasing numbers and cultural ambitions overexploited the
limited resources and collapse followed. Deforestation started upon arrival of the
first settlers and by 1700 CE most forest had disappeared. Soils deteriorated. Land
birds and open-ocean fish also disappeared from the diet, because of combinations
of deforestation, overhunting and predation by rats. Chickens then became more
important, and defensive chicken houses were erected. Social and ceremonial life
came to a standstill. With no more statues being built, belief systems fell apart and
with it the legitimacy of social organisation. Slavery, war and cannibalism followed
suit. Social mechanisms probably played an important role. In one interpretation,
the chiefs and priests, claiming relationship with the gods and promising prosperity,
‘ . . . buttressed that ideology by monumental architecture and ceremonies designed
to impress the masses, and made possible by food surpluses extracted from the
masses . . . As [their] promises were being proved increasingly hollow, [their] power
was overthrown around 1680 by military leaders . . . and Easter’s formerly complexly
integrated society collapsed in an epidemic of civil war’ (Diamond 2006). Were the
people on Easter Island stupid or evil? A comparative analysis of more than eighty
Pacific islands suggests that they were, in fact, unfortunate: Easter Island is highly
susceptible to erosion and thus amongst the most environmentally fragile islands in
the Pacific Ocean.
The Easter Island history may have had quite some precursors elsewhere. One
can read a similar sequence of events in the Megalithic memorials and burial mounds
Box 3.1. Peasants, priests and soldiers. The rise of social organisation can be
interpreted in terms of the risks early agrarian communities were facing. Priests
fulfilled a mediating role between ordinary people and the extrahuman world.
They also induced the self-restraint required for a farming life of hard work and
for the exigencies of food storage and distribution: ‘ . . . rites conducted by priests
helped to strengthen the self-restraint which could keep people from too readily
drawing upon their reserves’ (Goudsblom et al. (1996) 42). Harvest feasts and
sacrifices were social institutions to manage the pressures of frugality. Priests
were resource-managers avant-la-lettre – not a strange idea when one knows
about the rules in Christian and Buddhist monasteries.
In many societies, another alliance proved more durable: between peasants and
warriors. The priest-led, religious-agrarian regimes were probably first, but came
almost everywhere in competition with warrior-led, military-agrarian regimes.
Arguably the most crucial force behind the latter was the bonding of warriors
and peasants: ‘The warriors needed the peasants for food, the peasants needed
the warriors for protection. This unplanned – and, in a profound sense fatal –
combination formed the context for the great variety of mixtures of military
protection and economic exploitation that mark the history of the great major-
ity of advanced agrarian societies . . . wherever in agrarian societies rural settle-
ments developed into city-states which were subsequently engulfed by larger
empires, the priests became subservient to the warriors’ (Goudsblom et al. (1996)
59). The emergence of a warrior class, that is, of professional killers and pilla-
gers, should be seen as one stage in the monopolisation of violence and cannot
be explained solely in terms of their discipline, equipment and organisational
skills.
constructed some five to eight thousand years ago on the islands of Malta, Corsica
and the Orkneys. The Easter Island tale teaches us several lessons. First, it is pos-
sible to give an evocative and persuasive account of past events, even if evidence is
scarce and controversial. Some scholars argue that a collapse involving starvation,
major warfare and cannibalism is neither supported by the 18th-century journals nor
by the scientific evidence (Boersema 2002). An alternative narrative, with appar-
ently strong empirical evidence, is that Polynesian settlers arrived rather late and
that Polynesian rats played a major role by preventing forest regeneration. In this
reconstruction, there was a rather smooth transition from a ‘rich’ Moai culture to
a ‘poor’ bird-culture and the real collapse was caused by the raid on the island in
1862 by Peruvian slave traders. A second lesson is that, even in these early primitive
communities, both economic (resource) and social-cultural (organisation, rituals)
aspects played an important role in rise and decline. Third, it is possible and useful
to construct an explicit model of what happened in order to debate the different
hypotheses. Following the renewable resource economics approach of the 1950s,
Brander and Taylor (1998) have constructed a simple analytical model of an Easter
Island-like population that harvests and consumes a renewable resource (food) and
some ‘other good’. The positive link between net population growth rate and harvest
rate causes a ‘feast and famine’ pattern of population overshoot and endogenous
resource degradation and has become a widely used metaphor for unsustainable
development.4
3.3 Emerging Social Complexity: State Formation
3.3.1 Early Mesopotamia: Urban Centres and Their Elites

From the earliest stages, human groups have shown signs of self-organisation. In
many cases, this process gradually intensified into stratification between the rulers
and the ruled, differentiation into castes and guilds, and the emergence of towns
and trade markets. Tribes and chiefdoms evolved into one dominant authority in
larger regions, the state. The early state had a ‘very strong, usually highly central-
ised government, with a professional ruling class, largely divorced from the bonds
of kinship . . . highly stratified and extremely diversified internally, with residential
patterns often based on occupational specialisation rather than blood or affinal rela-
tionship . . . The state attempts to maintain a monopoly of force . . . while individual
citizens must forgo violence, the state can wage a war; it can also draft soldiers, levy
taxes, and exact tribute’ (Flannery 1972). Mythical and legendary charters, war and
terrorism were amongst the methods in which states enforced legitimacy, displayed
power and exerted control. States emerged all over the world, under a variety of
environmental conditions, and they are still a vibrant reality in the 21st century. The
state represents in its different appearances the most complex organisational form
by which human groups resolve the perennial tension between the individual and
the group, between competition and cooperation.
The earliest state has probably developed around the urban centres in the plains
of southern Mesopotamia. Already in the 6th millennium BP the city of Uruk exten-
ded more than 100 hectares (ha). The rural peasant families produced the food such
as wheat, barley, vegetables, milk, meat and the flax, wool, hides, dung, reeds, clay for
nonfood needs such as clothing and housing. The early towns facilitated all kinds of
exchange as part of a ‘tributary economy . . . dependent to a significant degree on the
mobilisation of tribute, in the form of goods or the labour used to reproduce them,
from producers to a political elite’ (Pollock 2001). Hierarchy and status started to
replace egalitarian forms of organisation. Building temple platforms and serving as
priests, even if on a voluntary basis, became a source of social and political prestige
and of material gain. As time progressed into the Uruk Period (6000 BP), the elite
divorced itself increasingly from the material forms of production and appropriated
ever larger portions of the surplus food and goods. Institutions for political control
emerged and long-distance expeditions were set up to procure exotic goods.
There is general agreement that environmental deterioration, mainly in the form
of salinisation, played a role in the collapse of the later Ur dynasties. Agricultural
productivity declined continuously from the middle of the 5th millennium BP, with a
shift from wheat to more salt-tolerant barley. Deforestation as a result of woodcutting
for domestic and industrial hearths, with further destruction by goats, was another
detrimental force. But social responses were an inherent part of the decline.
4 A nonlinear model of the underlying dynamics may explain the collapse of ancient societies with
the concept of sunk costs (Janssen et al. 2003).
Figure 3.2. Possible factors in the rural-urban dynamics in Ubaid and Early Uruk society in
Mesopotamia.
The rise in centralised control with its policy of maximising the surplus was a
major force behind the declining agricultural productivity and environmental dam-
age (Redman 1999). When food supply further deteriorated, probably in combin-
ation with climatic change, increasing numbers of debt-ridden peasants fled to the
cities. To feed the growing urban populations, increasing tribute had to be paid. This
intensified the flight to the towns and out of the region, with a further reduction in
food supply. A vicious circle of extraction and control evolved in order to sustain
the centres of power and prevent social revolt (Figure 3.2). At some point, the elite
structures collapsed and gave way to more local production modes that were geared
again to local needs and resources.
3.3.2 Egypt: The Nile and Its Rhythms

During the 7th and 6th millennium BP, episodes of severe drought, dwellers of
the desert drifted towards the Nile valley in search of water, food and fodder. In
subsequent millennia, the basis for Egyptian civilisation was laid. Nile water levels
have always mirrored climatic events throughout the region. Long-time historical
water level records in combination with archaeological finds show that episodes of
high Nile flood discharge are correlated with greater rainfall in East Africa and a
warmer climate in Europe, whereas low levels correlate with cooler temperatures
and drier conditions in East Africa. The Nile river is Egypt’s artery and at the root
of its economic prosperity and environmental fragility.
Extreme water levels had large consequences for human populations in Egypt.
Very high floods led to the destruction of dikes and houses and prevented the plant-
ing of crops. Very low discharge levels reduced the cultivable area as natural irriga-
tion failed. A variety of responses to such risks have emerged: building, maintaining
and repairing dikes, irrigation canals and drains. It led to a centralised administrative
20
15
Mln inhabitants
10
0
-4250 -3500 -2750 -2000 -1250 -500 250 1000 1750
Year BCE-CE
Figure 3.3. Estimated population in Egypt. In 2007, the population amounted to 80 million
(source of data: Turner et al. 1990).
state. Droughts in the Nile Valley around 5200 BP may have been responsible for the
unification of Egypt and the rise of the Old Kingdom. Early documents show admin-
istrators shifting from wheat to the more salt-resistant barley to combat increasing
salinisation resulting from irrigation. One millennium later, around 4200 BP, low
Nile River levels with catastrophic droughts probably caused the collapse of the Old
Kingdom. They triggered a series of civil uprisings, widely recorded during this time:
‘The Old Kingdom . . . was a time of tremendous royal power . . . and it saw a big
increase in population . . . This placed great reliance on maximising the use of the
land flooded and fertilised each year by the inundations. Around 4250–4150 BP, the
same prolonged dry period that caused such problems for the Ur III kings in Iraq
brought a series of consistently low floods and precipitated half a century of famine.
This helped pull a declining order. The monarchy was overthrown . . . ’ (Wood 1999).
Social order was regained by the end of the Ninth Dynasty, about 4000 BP, with
the administrative centre now at Thebes. The rapid recovery can be interpreted as a
sign of the strength of Egyptian civilisation after so many centuries. During the New
Kingdom, the population of Egypt as a whole increased again, possibly to as many
as 5 million during the Eighteenth and Nineteenth Dynasties (3489–3136 BP). Up
to half of the population may have lived in cities such as Memphis and Heliopolis,
the former one being perhaps the world’s first city with more than 1 million people.
Since then, the population has fluctuated between 2.5 to 5 million people, until it
started around 1800 to surge to its present value of more than 80 million people
(Figure 3.3).
Figure 3.4 shows a CLD with possible mechanisms of social-ecological evolu-
tion. As ruling elites were depending on the tax revenues from farmers, a series of
extremely high or low Nile water levels were critical for governance. Farmers were
confronted with starvation, urban craftsmen and artisans died in massive numbers5
5 It is reported that less than 10 percent of the weavers in Cairo survived the 1200–1201 CE famines
(Hassan 2000).
Figure 3.4. Environmental change induced forces of decline in Egypt: (a) famine can cause
a positive feedback loop towards disintegration, (b) accelerated by subsequent tendencies
towards disintegration. Investment in adaptive capacity and restoring ruling class legitimacy
(c) can halt or reverse the decline.
and the king saw his income shrink rapidly. One can imagine a bifurcation in such
periods of economic distress. If irrigation channels and dams had been maintained
and enough food had been stored for the ‘seven bad years’, as recounted in the
Bible, famines could be avoided, which in turn would make it possible to sustain
ruler legitimacy and military support (loop C). If, on the other hand, the infrastruc-
ture had been neglected and urban administrators, merchants and other nonfood
producers tried to stick to their ‘good life’ habits, a vicious cycle started (loops A
and B). The legitimacy of the ruling class came under threat. Rivals would emerge,
denouncing royal excesses such as construction of large monuments. Because of
declining revenues, the king could pay neither for military and political support nor
for large agricultural projects that would redistribute food and secure against future
droughts. The ‘power range’ of the central government – an estimated 400 km with
the available transport capacities – decreased and the state disintegrated in smaller,
provincial units. External enemies saw opportunities to invade the country. It is quite
probable that such downward spirals happened during the periods of low Nile dis-
charge around 4200, 3100 and 2700 BP. The outcome of such a disintegration process
was not necessarily disastrous: the newly emergent, smaller units may actually have
been more resilient.
3.3.3 South Asia: The Indus-Sarasvati Civilisation

The first signs of an agrarian civilisation in South Asia emerged in the valleys of
the Indus river and the – now largely dried up – Sarasvati river. Ever since British
explorers discovered the famous sites of Harappa and Mohenjo-Daro, the civilisation
that flourished here some 4,000 years ago has been called the ‘Indus civilisation’.
Its origin can be dated back to ancient settlements in Baluchistan around at least
7000 BP. In the past decades, remains of this civilisation have been found in a much
larger area than originally thought, possibly the size of Western Europe.
Apparently, this civilisation has known three phases. In the Early Phase, 5100–
4800 BP, it still wore the traces of the early farming communities and pastoral camps.
Box 3.2. Priests, no warriors? There are – and will be – different interpretations
of the Harappan settlements’ social complexity. It is clear that in the mature stage
these settlements were part of a larger political and administrative framework.
Some have speculated about priest-kings not unlike those in Sumer and Akkad –
but Chakrabarti writes that ‘An Egyptian or Mesopotamian kind of kingship need
not be envisaged in the Indus context. In later Indian history, the king . . . was a
much more humble figure . . . he does not strut around in sculptural reliefs tower-
ing above ordinary mortals and cutting the heads of his enemies . . . he functioned
within the well-formulated concept of the royal duty of looking after the well-
being of his subjects. . . . It is futile to look for the remains of a royal palace . . . for
the simple reason that there will not be any way of identifying it, going by the
later Indian examples. Priesthood is far more sharply visible . . . The concept of a
yogin, one who sits in meditation, is writ large . . . Remains . . . unmistakably imply
the services of priests – priests of a type that a practising Hindu would engage for
performing his household rituals even today’ (Chakrabarti (1997) 123). Such a
view suggests much more continuity between this mysterious past and the Indian
present – and more diversity in the behaviour of early priest-kings – than pre-
viously thought. This issue is still an important item in the debate on Indian
identity.
The Mature Phase, 4800–3900 BP, is the period of at least a dozen large cities. A
variety of crops were grown, including rice and cotton, and there were extensive
metal mining and smelting activities. The culture was characterised by sophisticated
architecture, a distinctive own writing style and fire worship. A remarkable feature
was, apparently, the absence of personality or ruler cults such as those that emerged
in contemporary Mesopotamia, Egypt and China. There is evidence of cultural inter-
action with Rajastan, Central India and Maharashtra and of trade with populations
in Mesopotamia through ports such as Lothal.
The Late Phase was around 3900–3400 BP, during which it mysteriously disap-
peared. Until recently, it was thought that the decline came rather abruptly – the
causes being a mixture of climatic change with more frequent droughts and changes
in the Saraswati river bed, more or less gradual invasion by northern Aryan nomads
and/or internal cultural erosion. However, new data suggest that a combination of
adverse changes in rainfall patterns and a massive earthquake caused river courses to
change. The important rivers in the region – the Sutlej, the Yamuna and the Sarasvati
or Ghaggar-Hakra – have changed course several times in the last four millennia.
Whatever the precise mechanisms, there is increasing evidence that, between 4200
and 3800 BP, a global change in climate took place with a particularly fast change
around 4200 BP known as the ‘4.2 K Event’ (Costanza et al. 2007). This can explain
why several early civilisations declined simultaneously in this period. The downfall
may have been accelerated by a collapse in trade caused by the synchronicity.
3.3.4 The Aegean and Mesoamerica: The Role of Ecological Diversity

The civilisations of the Euphrates-Tigris, Nile and Indus river basin were centred
around a large river. The fertile soils and irrigation potential was a determining
factor, it is thought, in the way they developed and were organised into large, cent-
ralised ‘hydraulic civilisations’. Some civilisations emerged under different circum-
stances. One example is the Aegean Sea region, with a very different biogeographic
situation. It covers mainland Greece in the north and west, the coast of Asia Minor
in the northeast and the island of Crete in the south. In between are hundreds of
large and small islands. It is an intensely researched and interesting area with regard
to state formation.
The oldest civilisation in the region is Crete, an island of about 8,000 km2 of
largely marginal mountain country. It had established itself by the end of the 5th mil-
lennium BP. Characteristic features of this culture were the large palaces with their
peculiar orientations, the mountain peak sanctuaries and the conspicuous absence
of fortified citadels or defensive walls and earthworks. The written script, institu-
tionalised trade- and sacred rituals gave powers of organisational and psychological
control to the classes of priests and kings. The palaces were destroyed a few centuries
later, probably by earthquakes, and rebuilt. The ‘New Palaces’ were destroyed viol-
ently or abandoned around 3400 BP, probably due to conquest by the less trade- and
more war-oriented Mycenaean people in the southern part of Greece.6 Mycenaean
cities collapsed in around 3100 BP, due to invasions and possibly climatic change or
natural disasters.
Against the resulting background of petty lords practising small-scale redistribu-
tion as part of a strategy of ironing out localised shortages, the later Greek civilisation
emerged and dominated the southern Aegean Sea for some centuries. The south-
ern Aegean consists largely of coasts and islands with a large ecological diversity.
People experienced interannual harvest fluctuations that were large, unpredictable
and asynchronous in these highly diverse environments. In such a situation, trade
with neighbouring settlements was an important buffer against food shortages. This
ecological diversity probably contributed to the emergence of a social ranking with a
prominent role for merchants and for symbolic transactions. It is tempting to relate
this early landscape-inspired evolution to the well-known features of the later Greek
city-states: politically autonomous units resisting unification and an extensive trading
network and colonies all along the Mediterranean coasts.7
The reconstructions of another civilisation, the Maya in Mesoamerica, also sug-
gest an important role for the biogeographical situation. The Maya were amongst
the larger and most advanced civilisations in Meso-America. It lasted from 4000 BP
(Early Pre-Classic) to 500 BP (Late Post-Classic). The Yucatán Peninsula was its
heartland. It had significant gradients in elevation, soil thickness, rainfall and water
accessibility. Three river basins on the peninsula provided transition zones – or
ecotones – which created and sustained cultural diversity.
A possible relationship between environment and social organisation is sug-
gested from a comparison between Yucatán and the highland valleys of Oaxaca
and Mexico. The mountains in the highland valleys are natural barriers that obstruct
6 Possibly, a huge volcano eruption on Thira – modern Santorini – did wipe out the Minoan civilisation
on Crete in a complex interplay of factors each operating on different time-scales: flood waves,
interruption of trade, climatic change and erosion of dominant belief systems and invasions.
7 An interesting network-based perspective on the dynamics of Aegean civilisation is developed by
Knappett et al. (2008) (§10.6). The Bronze Age maritime network is used to explore the balance
between local resource cultivation and trade.
communication and have limited potential for agriculture. Their agricultural systems
tended to be uniform and hierarchical: ‘a continual preoccupation with comparat-
ively little horizontal complexity – the vast majority of parts are alike, and they tend
to stay that way. The replication of uniformity is usually accomplished by means of a
strong vertical differentiation – a powerful hierarchy manages most affairs’ (Blanton
et al. 1993). The Yucatán lowlands, on the other hand, are much vaster and unboun-
ded. There were many interacting socio-political units with permeable boundaries
and large effort went into regulating flows of things and people across the boundaries.
People in these circumstances tended to employ a wide and diverse spectrum of pro-
duction techniques. It led to diversity and horizontal complexity, with coordination
in matters such as access to resources, allocation of time and labour and exchange of
products. Economic institutions developed early and provided a consistent basis for
survival. The evolutionary problem was the ‘organisation of diversity’: how to keep
the components in these horizontally complex systems doing their jobs even though
they have very different roles and interests? This problem is still with us.
Why did Mayan civilisation decline? Archaeological evidence clearly shows that
Mayan farmers had to cope with a capricious and fragile environment. It explains
their elaborate calendar system with a superior ability to prognosticate patterns of
cyclical environmental changes. It also explains why the intensive farming necessary
for an ever-growing population could not be sustained. Climatic change seems to
have caused population shifts between the three river basins in the Early and Middle
Pre-Classic period. Later, elevated cities were abandoned in the period of great
droughts (1050–1250 BP). Only in one river basin are there signs that human activity,
notably deforestation with subsequent larger river discharge volumes, played a role
in the decline. Yet, the general view is that population outstripped the available
resources. Deforestation and erosion degraded the resource base and, at least partly
as a consequence, warfare intensified and made the situation worse. Climatic change
and short-term egotism of the elite probably were the final blows (Diamond 2006).
3.4 Empires
3.4.1 The Roman Empire8

Some state-based civilisations managed to further develop their technical and organ-
isational skills, towards population sizes and areas of control that led to the epitaph
‘empire’. Whereas the word state refers to political organisation and government,
the word empire is etymologically rooted in rule and authority. Empires are char-
acterised not only by autocratic rule but also by a territory far beyond the one of
the original rulers or nation. The word was used first for Rome after it had been
weakened by civil wars and Augustus seized power in 27 BCE. Many books have
been written on the rise and fall of the Roman Empire. The focus on the human-
environment nexus is rather recent (van der Leeuw 1998). This section presents
some narratives and insights about the mechanisms that led to growth and decline of
this and other empires. The Roman Empire is of special interest, not only because
it is a precursor to the Western or European civilisation, but also because much is
8 This section is largely based on van der Leeuw and de Vries (2002).
3.4 Empires 67
Box 3.3. Environmental degradation: a record of collapse. The pioneering

Central Petén Historical Ecology Project uncovered the environmental history
of a Maya site in lowland Guatemala (Redman (1999) 141–145). In this region, a
swidden – or milpa – agricultural system with a three- to five-year fallow period
can support population densities in the order of 25 people/km2 . How did the
Maya manage to sustain in some places and times up to 250 people/km2 ? Examin-
ing sediment core from lake bottoms, a significant increase in phosphorus and
silica deposition during the Maya occupation was found. Accelerated phosphorus
deposition points at more phosphorus in the soil due to human activity (waste,
food, disintegration of bodies and stoneworks) and to more soil erosion. Similarly,
silica deposition increased severalfold during the period, which archaeologists
think had the highest population – another indicator of increased erosion rates.
Apparently, the Maya appreciated the tropical forest ecosystem well enough to
thrive for centuries by managing land clearance, control water flows and transport
food to cope with localised shortages.
However, the lake sediment data indicate that ‘the high forest that prevailed
in much of [the] region was largely removed by the farming and settlement build-
ing activities of the Mayas as early as 3000 to 4000 years ago. This resulted in a
shift toward more open vegetation . . . with the maximum deforestation between
1000 and 2000 years ago. The basic drain on the land . . . increased to the point
where the system was no longer sustainable. . . . By the end of the 10th century
AD, most of the large settlements of the Mayan uplands and southern lowlands
had been abandoned or at least seriously depopulated’ (Redman (1999) 145).
Coincident was a relatively dry period, which put crop productivity under addi-
tional pressure in an ecosystem already strained by human activities. There are an
increasing number of environmental history reconstructions on the local/regional
scale, which give empirical and model-based insights on how human activities
interfered with long-term and short-term environmental change (Dearing 2006).
The reconstructions may help to understand better the present situation.
known about it over a long period and across several geographical scales. Besides,
many traits of (pre)modern society were already part of Roman society, such as
rapid colonisation of most or all of the known world, an elaborate military and civil
organisation to control the Empire, an urban base and major infrastructure.
Let us first have a look at the biogeographical situation. The Mediterranean basin
has relatively mild winters, hot summers and precipitation in spring and autumn.
Most Mediterranean landscapes were suitable for human habitation and had become
entirely dependent on interruptions of the natural cycles by humans by the time the
Romans colonised the region. Without human intervention, landscapes would have
rapidly changed due to erosion, growth of scrubland (garrigue) and recolonisation
by forest and become unsuitable for the human activities that had shaped it. The
variety of environments in Mediterranean landscapes lend themselves to the small-
scale cultivation of different crops, together with herding, rather than to the large-
scale farming of homogeneous crops. Already long before the advent of the Romans,
the cultivation of cereals in combination with that of olives and vines made it possible
for farmers to harvest more than one crop. This natural resilience through diversity
increased when the original territory of the Roman Republic, which was relatively
homogeneous, expanded with the conquest of other provinces. It distributed the risk
of food shortage. Moreover, Roman occupation added inventions in the areas of food
storage, treatment and trade. It implied a shift from local environmental resilience
to national/regional social-ecological resilience, which is an essential characteristic
for civilisations to be sustainable.
When the Republic, and later the Empire, expanded, the Romans imposed with
a varying degree of success a commercial system of exploitation on the existing,
highly varied ways in which the local populations exploited their lands. Exploitation
was initially organised in a very dense pattern of farms of variable size from before
the advent of the Romans, or in country houses (villa) and large, highly efficient
farms (latifundia). As time went on, the different exploitation systems became more
dependent upon each other and had larger impacts on the landscape. In a next
stage, the agricultural landscape started to change profoundly under the influence
of mass-production for the export of agricultural products. It meant an increase in
scale – larger farms, larger installations, larger bovines. Many smaller farms were
either abandoned or became part of the larger ones that had grown out of the first
Roman establishments in the region, located on the best land and with the most direct
access to the markets. The forests on many hilly slopes were cut down and had to be
terraced or otherwise protected against erosion. It made these landscapes extremely
dependent on human intervention and vulnerable to erosion in its absence.
The commercialisation and rationalisation of agriculture were made possible by
effective administration of rural areas (taxes and markets). An important role in
this was played by the organisation of the land in a system of square-mile blocks
that were laid out and mapped by a professional class of surveyors.9 Because this
‘square’ administrative point of view often prevailed over any practical considera-
tions, the resulting organisation of the landscape was regularly at odds with the nat-
ural dynamics of an area. For instance, in the Tricastin in southern France the roads
and ditches were all laid out in north-south and east-west directions, whereas the
natural drainage of the area generally went from north-east to south-west. When the
number of soldiers drastically declined, there was insufficient manpower to keep
the ditches open. The drainage scheme fell apart, and the land became neglected and
ultimately began to erode (van der Leeuw 1998). Such a mismatch between the local
reality and the practises of the exploiter can be an important cause of decline, which
was also experienced during European colonisation in the 16th to 20th centuries.
In parts of northern Africa, the local situation created another pattern of colonial
exploitation. More and more simple Roman farmsteads emerged between the 1st
and 3rd centuries CE. It pushed the desert pastoralists farther and farther away
from the coast and reduced the grazing land available to them. Eventually, it forced
them to begin herding camels instead of sheep and goats. Their impoverishment
increased the contrast in wealth. It led to more frequent raids on the rich Roman
9 The land was divided into square blocks measuring one by one mile. Mile comes from milia,
thousand – one mile being a thousand steps with a Roman step being two of our steps: left-right. The
imposition of order and uniformity as an instrument of control is a characteristic of the hierarchical
state (Scott 1998).
3.4 Empires 69
Box 3.4. Views of nature in European Antiquity.10 How nature was seen
depended in antiquity – and in many places still – on its actual or potential
threat to humans. Nature, with its unpredictability and threats, was in no way
always a friend. Therefore, it was permitted, and even a duty, to fight it, as can
be seen in a variety of cultural motifs. The disappearance of forests was seen as a
sign of civilisation, at least in the dominant view.
Clearing forests for agriculture, wood, firewood and animal grazing was the
main cause of deforestation. Timber demand was another one. Thirgood (1981),
quoting Strabo, mentions the mountains in southeastern Spain ‘covered with
thick woods and gigantic trees’, being cut for shipbuilding. North Africa was
another important timber-producing region for the Romans, leading to temporary
depletion of Moroccan forests. Wars had a devastating effect as it accelerated the
felling of trees to be used for the warships and made people flee into the mountains
and abandoned their land. Wild animals were another aspect of nature. Heracles
battled with near-invincible mythical wild animals, becoming a symbol of courage.
The Roman Emperors wanted not only slaves but also wild animals to be taken
from the conquered lands and to be emblems of total dominance over man and
animal. More intense hunting brought some species to extinction. The enormous
demand for animal skins, from as far as the Baltic and northern Russia, further
increased the pressure.
The Pax Romana acquired a specific connotation through the mass killing of
wild animals. Thousands of wild animals were slaughtered in the venationes during
the games or ludi. These took place in every garrison town, but Rome had by
far the largest. With the inauguration of the Colosseum in 70 AD, some 5,000
animals were ‘used’ in a few days time. It must have given rise to huge transport
problems. Only recently, it has been acknowledged that the grand scale of these
killings may have led to the extinction of some of these species. The games may
indirectly have contributed to the expansion of agriculture in the Mediterranean
by strongly reducing the threat from wild animals – an idea that fits into the view
that humans should ‘civilise’ the world and was, therefore, approvingly supported
by scholars until recently.
settlements until, eventually, they were fortified to protect the colonists. Again, it
preceded similar developments in European colonies almost two thousand years
later.
Between the 4th century BCE to the 1st century CE, the surface of the Roman
Empire increased from less than 1,000 to more than 267,000 km2 , and the number
of Roman citizens – excluding members of other nations and slaves – from less than
1 million to more than 4 million. After that, the expansion stalled.11 How did the
10 This box is based upon a contribution from Jan Boersema, Vrije Universiteit Amsterdam, in de
Vries and Goudsblom (2002).
11 The average population density of Italy – about 20 inhabitants/km2 – was probably of the same order
of magnitude as that of other parts of the Mediterranean world. In contrast, for Gaul at the time of
Caesar, the equivalent number was about 9 inhabitants/km2 – implying about 5.7 million people for
all of Gaul, including Gallia Narbonensis.
Romans manage to conquer such large areas so quickly and durably? Part of the
answer is organisational talent and technology: They were able to establish and con-
trol an appropriate infrastructure, which depended on the spread of towns. Urban
areas were easier to control and could via their administration and commerce be con-
nected directly to Rome’s own. It was essentially an exploitative, pyramidal pattern
of administration, based on links with the centre of different form and intensity and
with a variable degree of Roman impact on the various components of its territory.
There are at least three other elements in how the Romans managed to hold
together an Empire the size of the Mediterranean basin. First, the location of Rome
at a central position in the Mediterranean basin gave it an advantageous position
for trade and governance. Second, the road system with the major roads and the
many (sub)regional roads provided dependable and efficient means to move people
around and thus to exchange information and ensure administrative and military
control. It also deliberately broke through the unity of the tribal territories, and
rightly was the pride of the Empire. The third element was a trade system, which
greatly exceeded previous commercial activities.
Yet, in the second century of its existence, the Empire started to show signs of
crisis. Roman authors used to blame invasions, epidemics, incompetent Emperors
or, in a more structural vein, the weak and overstaffed administrative infrastructure
and the heavy burdens of maintaining a government and an army over such a wide
territory. Lead poisoning, soil exhaustion, famine and possibly climatic change were
mentioned, too.12 Most of these factors were real and did contribute to the difficulties.
One plague epidemic lasted from 165 CE to about 180 CE, with the loss of up to
a third of the population in certain areas. The 3rd century saw a rapid succession
of emperors lasting between a few months and three years. There were always one
or more provinces in revolt, independent or out of control. The emperor Diocletian
(284–305 CE) restored a semblance of order and his reign initiated half a century of
relatively stable government and relative wealth. The administration and the army
were reformed into larger, more complex and more powerful institutions; the Empire
was divided into an Eastern and a Western part with a separate administration,
different languages – Greek and Latin, respectively – and different cultures. The
provinces were subdivided into smaller ones. People were conscripted to perform
duties in the maintenance of infrastructure and many other activities and taxes were
increased. It was not durable: The structure of the Roman Empire collapsed in
395 CE. The periphery increased in independence as the core began to lack the
means to impose itself. Cities retracted within their walls, and the flow of traded
goods and of people to and from areas farther from Rome began to dry up. Tribes
outside the Empire were attracted by its riches and fame and increasingly raided
the areas where wealth could be found. The Western Empire was more vulnerable
than the culturally and linguistically homogenous Eastern Empire and collapsed in
less than a century. The Eastern Empire was able to maintain itself for another
millennium.
12 In areas where these data can be compared – principally in southeastern France – there are no
indications of major changes in climate between about 400 BCE and about 600 CE. At the time
of the Roman expansion, precipitation may have been slightly lower and average temperatures
somewhat higher than either before or after that period.
3.4 Empires 71
Many books have been written about the causes of the decline. The theories tend
to reflect the fashion of the day. There certainly was not one single cause. The decline
was a long, protracted sequence of mutually connected causes and consequences.
One determinant was the unbalanced financial situation. Money was important in
sustaining the empire and conquest of new territory provided it: ‘In 167 BC the
Romans seized the treasury of the King of Macedonia, a feat that allowed them
to eliminate taxation of themselves. After the Kingdom of Pergamon was annexed
in 130 BC, the state budget doubled, from 100 million to 200 million sesterces.
Pompey raised it further to 340 million sesterces after the conquest of Syria in
63 BC. Julius Caesar’s conquest of Gaul acquired so much gold that this metal
dropped 36% in value . . . By the last two centuries BC, Rome’s victories may have
become nearly free of cost, in an economic sense, as conquered nations footed the bill
for further expansion’ (Tainter 1988). After Augustus became emperor in 27 BCE,
there were no longer any rich territories to be conquered in the periphery of the
Empire. This amputated the income of the State, a problem well known to modern
governments. Most of the emperors between Augustus and Diocletian complained
of fiscal shortages and/or had to resort to new kinds of taxes. Faced with crises or
ambitious to leave their mark on history, they resorted to the debasement of the
currency. Inflation must have demanded a heavy toll.
While revenues declined, ever-larger sums of money went towards the main-
tenance of the administration and the army, bread and games (panem et circenses)
for hundreds of thousands of Roman citizens, and the maintenance of the physical
infrastructure (roads, wharves, etc.). Written documents and archaeological evid-
ence give the impression that the structure of taxation was to blame, as well as the
increasing independence of large latifundia and the loss of administrative control
over the countryside. There was no money left for crisis management, and the low-
margin and highly decentralised economy was not able to generate additional value
and provide a flexible source of income to the core.
A second cause of the disintegration was organisational. The Empire suffered
from insufficient capacity to share and negotiate information between the different
members of society. The Romans managed to enhance integration and reduce risks
by setting up a road network, control local elite groups and organise a bureaucracy
and a monetary system and regular flows of goods to all parts of the Empire. It was
directed towards facilitating the flow of goods, energy and information throughout
the system, notably from the periphery to the centre and vice versa. In the process,
the centre became more and more dependent on interaction with the periphery.
However, people in the periphery learned Roman ideals, techniques, goods, attitudes
and concepts, and Rome lost its advantage over the periphery in terms of information
processing. The political and cultural entities in the periphery came to deal with
Rome on a more equal basis and became more difficult to control. Fragmentation
was inevitable.
A third and related cause of decline stems from the accumulation of long-
term risks as a consequence of frequent and intense intervention of people in their
environment. Many of these risks resulted from short-term solutions to problems.
For instance, deforestation provided new agricultural land in order to produce more
food, but in the long term, food production suffered from erosion, which was then
solved by increasing food imports. The long-term risks were less easily understood
and perceived and were also considered less relevant because they were far away in
time and/or space. These changes in ‘risk spectrum’ differed from region to region,
but their temporal coincidence caused regional and local stagnation and ultimately
blocked the expansion of the Empire as a whole. The exact domain in which the
stagnation – climate, soil exhaustion, social or economic difficulties or invasions –
first manifested itself is less relevant than the inability to cope with all the issues
simultaneously and effectively. In retrospect, one may wonder why innovation did
not emerge to solve these constraints. It is tempting to speculate what would have
happened if the Empire had, for example, found new sources of energy and adopted
steam engines . . .
Viewed as a self-organising system, the Roman Empire showed a capacity for
conceptualisation and information processing that was in combination with its nat-
ural diversity at the essence of its resilience. However, it had to expand for its survival
and increasingly suffered from the inherent limitations in correctly handling complex
physical and informational networks and observing and interpreting environmental
change. The resulting societal crises triggered local overexploitation in the already
highly human disturbance–dependent Mediterranean basin. The lesson is that the
resilience of social structures is determined by their capacity to innovate in response
to the changes in circumstances that they themselves generate. The slow degrada-
tion of the Pax Romana of the 1st and 2nd century CE seems highly relevant for the
interpretation of the rise and decline of later Empires and of the past Pax Brittanica
and the vanishing Pax Americana.
3.4.2 Other Empires: China, India and Russia13

The world has seen several other empires. In China, one speaks of an empire since
the unification under the king of the State of Ch’in in 221 BCE. The caliphates
and dynasties in the Middle East and the southern Mediteranean between the 7th
and 15th centuries, and the later Mughal Empire in India (16th–19th centuries) and
Ottoman Empire in Turkey (14th–20th centuries) were all rooted in Islamic culture.
Although Spain and England had less autocratic imperial rule than these empires,
historians do speak of the Spanish Empire (15th–20th centuries) and British Empire
(16th–20th centuries). I briefly discuss three empires/periods where the system had
come so close to its carrying capacity that it could no longer sustain the number and
welfare of its people: China, India and Russia.
Chinese history has been influenced greatly by peoples coming in from the large
Eurasian steppes. The basin of the Wei River, the largest tributary to the Huang
(Yellow) River, in particular has been a corridor in and from which the earliest
recorded states have come into being. It had a natural outlet into the large Huang-
Huai-Hai plain, covering some 350,000 km2 and largely built up from 400–500-metres
(m)-thick sediment layers from the Huang River (Zuo Dakang 1990). The area has
hot summers and cold winters, and the distribution of settlements up to 2000 BP,
mostly in the hilly areas between the mountains and the plain, suggest the inhospit-
able nature of the plain as a human habitat. Yet, it was here in these northern valleys
and plains and much earlier than in southern China that state development started.
13 The accounts of China, India and Russia are based on chapters in de Vries and Goudsblom (2002).
3.4 Empires 73
China has a diverse and peculiar geography. From the southwest to the northwest
it is surrounded by high mountains, in the north by the cold Mongolian grasslands
and mountain ranges, and in the northeast, east and southeast by seas. The natural
environment has always pushed its imprints on humans: ‘Frequent major and minor
river shifts have affected the geomorphology of the plain; it is recorded that during
the past 3000 to 4000 years, the Huang river has broken its banks 1593 times’ (Zuo
Dakang (1990) 473). People practised a mix of hunting and slash-and-burn farming.
The king sent out farmer-soldiers to colonise and hunting was a military activity. The
word for hunting and agriculture was the same, t’ien. Environmental and climatic
change may be one reason that war was a most prominent feature of life in these
parts from the early Xia and Shang period (4200–3100 BP) onwards (Elvin 1993).
There was a close connection between sacrifice, war and agriculture with elaborate
rituals related to droughts. The king had unique divine powers and divined personally
about war and agriculture. The legitimacy of his power depended on his success in
manipulating the weather and war spirits.
Only after the unification of the northeastern part of present-day China under
the Ch’in Dynasty (221–206 BCE), soon succeeded by the Han Dynasty (206 BCE–
220 CE), one can speak of an empire. Famines remained a recurrent phenomenon
because of a variety of factors, many of them linked to the endemic warfare, such
as field abandonment resulting from corruption and violence and destruction of
the farming population and harvests in military campaigns.14 At the same time, the
introduction of iron farming implements and animal husbandry made agriculture
more productive and increased the food supply. These higher yields led to higher
population growth. During the Ch’in-Han period, the larger part of the estimated
60 million inhabitants of China lived in northern China and most cultivable land was
used. Millions of people were forced or sent out to cultivate grasslands and forests.
The surging construction of military and civil works – such as the Great Wall –
required massive amounts of timber and brick and led to large-scale deforestation
and subsequent desertification and increased flooding. The large differences in access
to food supplies were another source of tension.
According to Elvin (1993), one of the early scholars on the environmental his-
tory of China, the social structure of power was the most important single factor
controlling what happened to the natural environment. A most intriguing aspect of
the ancient Chinese civilisations was the coexistence of a political philosophy that
put conservation of a well-ordered nature at its centre, with a political reality that
led to warfare in the ruthless pursuit of personal wealth.15 Disturbance of the natural
order was thought to bring catastrophe and seen as a sign of moral decline – yet, ‘in
order to mobilize resources and build up populations to man the armies that were
engaged in warfare that was rapidly shifting from a ritualized combat to merciless
14 After the decline of the Roman Empire, the gradual fragmentation of power, administration and
land ownership became the start of a similar period in European history, with centuries of almost
endemic feudal warring. Yet, there are also some remarkable differences, for instance, in religious
practise and technological development.
15 Of course, the tension between a society’s ideology, on the one hand, and the reality of greed and
power, on the other hand, are found throughout all civilisations. There is ample historical evidence,
for instance, that a similar mismatch between the stated (Christian and chevaleresque) ideals and
everyday feudal practises existed in medieval Europe.
destruction . . . statesmen turned to economic development . . . [which became] both

the consequence and the cause of intensified warfare’ (Elvin (1993) 17–18). In prac-
tise, this meant limited access for commoners to non-agricultural land, violation of
nature conservation regulations and the creation of an advantageous state mono-
poly. People were forced to settle down and become a source of taxes and soldiers.
Taxes on mountains, marshes, dams and reservoirs were under the control and to the
benefit of the Imperial Court. In the official response to the famines and disasters,
one recognises the ‘powerless wisdom’: the rulers were morally obliged to put the
people’s welfare first and many administrative rules and techniques were introduced,
but often only revolt or the fear for it would bring practical relief.
In later times, the Chinese population has known more prosperous periods,
but reconstructions of the size of the human population indicate large fluctuations
resulting from recurrent clashes between food needs and food supplies and, relatedly,
sequences of war and peace, order and chaos (Turner et al. 1990). Judgment on the
causes of these events is partly a matter of focus and interpretation. Whereas Elvin
(1993) summarises Chinese reality as ‘fine sentiments, dosed action’, Yates (1990)
finds it hard to imagine that the Chinese population could have grown to the present
size without the technical and moral foundations for a system of state involvement
in providing for peoples’ basic needs.
Several other historical accounts reinforce the impression that large populations
of humans were living close to the environment’s carrying capacity and went through
periods of severe food shortages and associated violence and hardship (Revi et al.
2002). The fertile lands of the Indian subcontinent led to rather large population
densities long before European colonisation started, with probably more than 100
million people around 1500 CE. Until the 17th to 18th centuries, the population of
100–150 million people lived at a rather low-level equilibrium between food needs
and food supplies. Pressure on the land was high because of a variety of socio-
cultural factors. Variation in rainfall – sometimes as part of larger climatic/monsoon
changes – occasionally triggered a cycle of bad harvests, famine and disease. With
the decline of the Mughal Empire and the advent of the East India Company and
British colonialism, these cycles appear to have been negatively influenced by high
land taxes due to the pressures for profit in far away Britain, to lack of reinvest-
ment in land maintenance and to frequent wars, to mention the most important
ones. The incidence of famine increased, disease epidemics often followed. Only by
the late 19th and early 20th centuries did better nutrition and hygiene and more
effective relief measures lead to a decline in mortality. Indian history illustrates
that the more recent environmental history can only be understood in a larger
geo-political context.
When Czar Peter the Great called himself Emperor of all of Russia in 1721, it
was after three centuries of rapid territorial expansion (Revi et al. 2002). Although
the territory was vast, its population was small: some 15 million people around
1700 CE or an average density of only three to four people/km2 . The peasantry in
the upper Volga region with poor forest soils heavily relied on subsistence farming
on small plots of land. The severe climatic conditions of Russia required more fodder
to be stored than the farmers could provide, which made cattle herding a risky and
unprofitable venture. There was no potential for expansion and production for the
market. Getting children was also risky, so population growth was modest.
The situation changed when the vast uncultivated territories of the southern
steppes became accessible for cattle breeding to Russian peasants. Food shortages
were solved by cultivating new land and not by increasing productivity. Population
growth went up in line with the expansion of cultivated area. As a result, the pop-
ulation of European Russia had increased by the second half of the 19th century to
49.6 million in 1885 and 67.3 million in 1900, and its birth rate approached its biolo-
gical maximum. In many regions of European Russia, there was rural overpopulation
and scarcity of cultivable land reached dramatic proportions. In combination with
large cereal exports to Western Europe, there were serious food shortages. Millions
of Russian peasants were forced to leave their villages out of poverty. Unable to
improve farming practises and failing the resources to migrate to new regions, they
started searching for seasonal work in towns and other regions. By the end of the
1890s, their number had reached 9 million. It is an indicator of rural overpopulation,
which was estimated at 23 million in 1901 by a special commission under the gov-
ernment of the Russian Empire. It is hardly surprising that 670 peasant riots were
observed in the European part of Russia in the years 1902–1904.
One way to alleviate overpopulation was to resettle Russian peasants from the
European part of the country in the southern areas of Siberia and Central Asia. By
the end of the 19th century, millions of Russians were already living in these vast
and barely populated territories. Many more millions followed, but outmigration
remained below 5 percent of the population and did not solve the much larger
overpopulation. Thus, one can discern the social-ecological contours of a period,
which, in the collective memory, is primarily associated with the Russian Revolution.
3.5 Mechanisms, Theories and Models
3.5.1 Mechanisms
The sociologist Elias distinguishes in his book The Civilizing Process (1969/2000),
three dangers threatening human groups:
r the extra-human world: droughts and floods, wild beasts and pests, earthquakes
and volcanic eruptions;
r inter-human relationships: hostile neighbours, invading warriors; and
r intra-human nature: mismanagement due to negligence, ignorance, lack of self-
restraint or discipline.
Initially, the first of these dangers was dominant and the basic options were to adapt
or migrate. With the rise in social complexity, the second and third danger became
ever more important. Development was a process of continuous attempts at control,
in the form of conquest, unification, exploitation and adaptation in the physical as
well as the mental realms of life (Figure 3.5). This ‘domestication of the wild’ was part
of the transition to a settled life. It created new dependencies and subsequent needs
for further control. For instance, once an irrigation canal has been constructed, it has
to be maintained and protected and the food system becomes dependent on experts
and on soldiers. It also necessitated more and new forms of cooperation, which often
proved to be more advantageous than war and competition between smaller units
(Wright 2000). This in turn created social arrangements and hierarchical institutions.
Figure 3.5. The control and dependency loop. Responses to natural hazards have led to control
mechanisms, which in turn brought new dependencies and risks.
Environmental historians understandably placed more emphasis on the role of

the natural environment as a necessary resource for human life than social scientists
tend to do. The latter’s neglect is vocally expressed by Reader in his book Man on
Earth: ‘As city states have expanded into nations, kingdoms and empires, civiliza-
tion has repeatedly become obsessed with the acquisition and wielding of power;
an activity that has left millions of people dead in the streets and dominated the
pages of history so much that the significance of the substance bases upon which
civilizations are founded have all but ignored’ (Reader 1988). The environmental
histories in previous sections highlight a few points about the human-environment
interaction:
r human populations tended to grow towards the maximum number (‘carrying
capacity’) that could be sustained by the regional resource base;
r this occurred at a rate different from the degradation rate of the environment,
and could, therefore, usually not be stopped at the moment people became
aware of the consequences;
r when the resource base got overexploited as a result, the struggle for the stagnant
or declining (food) surplus intensified;
r the resulting tensions could turn outward, in the form of outmigration and
conquest, or inward, in the form of oppression and sacrificial rituals or technical
and social innovations and changing environmental practises; and
r the capability, or lack of it, of the elite to understand, anticipate and act for the
good of the larger system became a more important or even dominant factor in
whether a civilisation could sustain itself.
Most of the time, a mix of all these sometimes postponed, sometimes reverted, and
sometimes accelerated the processes of decline and collapse. In system dynamics
language, developments were driven by positive and negative feedback loops. The
positive loop was manifested in growing size and density of human populations and in
productivity increases as a function of accumulated experience and its manifestation
in technical artefacts. A reverse, negative loop set in when certain thresholds in the
use of environmental resources were exceeded and the resource base got degraded
as a consequence. The social and cultural dynamics became a more integral part with
the rise of economic and social complexity.
For instance, the ornithologist Diamond (2006) lists human-induced environ-
mental change, climatic change, external enemies, external friends causing depend-
ence, and endogenous cultural dynamics as the main determinants of decline. The
last one brought him to the sobering reflection that ‘ . . . we have to wonder why
the kings and nobles failed to recognise and solve the[se] seemingly obvious prob-
lems undermining their society. Their attention was evidently focused on short-term
concerns of enriching themselves, waging wars, erecting monuments, competing
with each other, and extracting enough food from the peasants to support all those
activities’ (Diamond (2006) 177).
Some archaeologists describe past processes of social organisation in system
dynamics terms (Flannery 1972). Increasing differentiation and specialisation of
subsystems (segregation) and increasing linkages between the subsystems and the
highest-order controls (centralisation) are seen as driving forces of the rise of special-
purpose institutions like the military, and the over-ruling of lower-order controls by
higher-order controls as in centralised irrigation schemes. Such systemic phenom-
ena could develop into socio-cultural ‘pathologies’, causing stress that might erode a
system’s resilience and accelerate disintegration. Numerous histories illustrate how
such pathologies, in interaction with natural environmental dynamics, caused stag-
nation and decline. Other archaeologists have emphasised social interactions such
as warfare, competitive emulation, cultural absorption, innovation dispersal and the
exchange of goods and information as part of the evolution of social-ecological sys-
tems (Renfrew and Cherry 1985). For instance, warfare – itself often induced by
resource scarcity – was a perennial phenomenon in the early civilisations of the
Middle East and Greece with sometimes devastating impacts. Competitive emu-
lation, that is, imitation amongst competing elites, may have been a factor in the
decline of some cultures, with the statues on Easter Island but also the famous medi-
eval towers of San Gimignano and present-day skyscrapers in Middle Eastern oil
states as examples.
The evolutionary unfolding of social complexity that came with agrarianisation
manifested itself in various ways:
r social stratification, as a way to combine technical and organisational skills with
social coherence;
r specialisation in certain skills, in order to permit and sustain higher and more
diverse production; and
r trading as one of the strategies to reduce risks of famine and war.
An urban-rural divide started to develop, with large variations in population dens-

ity. A large and stable food surplus, and an adequate means and infrastructure for
distributing it, were a precondition for the rise of urban settlements. It required
capital goods such as tools, terraces, dams and canals, granaries, roads and transport
equipment that were susceptible to deterioration or destruction. The rise of social

complexity was thus largely a sequence of cause-and-effect chains in order to con-
trol the natural and human environment. With it came new risks and vulnerabilities.
The new techniques and practises introduced unknown long-term environmental
impacts and started or accelerated detrimental feedback loops. Often, they became
visible only by the time the system came under internal or external stress. Learning
to cope with long-term and more erratic events is difficult. The human perception
and interpretation of what is going on is always simplified and incomplete. As long
as change is within the variations experienced before and stored in the collective
memory, such simplified understanding may still be adequate. For change outside
these variations – many environmental processes operate on centennial or millennial
scales – the capacity for adequate response is less. This limited response effective-
ness had important consequences for how people in ancient civilisations dealt with
environmental change – as it will have in the 21st century.
3.5.2 Theories and Models

Many theories have been proposed to explain the decline and collapse of states
and empires. Most spectacular are the Big Catastrophes, such as earthquakes, vol-
canic eruptions or climatic change. They dismiss the possibility that ‘ . . . the world
ends, Not with a bang but a whimper’, in the words of T.S. Eliot. Another set of
explanations point at population pressure per se as the cause of famine and col-
lapse. The number of people grows exponentially and food availability does not,
so people inevitably run against food limits, as the English scholar Malthus (1798)
proclaimed. Such (neo-)Malthusian theories invoke resource overexploitation and
mismanagement as co-determinants. The Danish economist Boserup introduced in
her book Population and Technological Change (1966) a feedback that invalidates
Malthus’ theory: Increasing population pressure forces people to produce more food
by putting in more labour, usually at lower labour productivity. The transition to
agriculture was thus a response to scarcity and its outcome in the form of improved
technical and organisational skills was ‘progress born out of necessity’ (Wilkinson
1973).16
A third set of theories focuses on internal political processes: overstretch in
material and cultural terms, erosion or lack of spiritual values. For instance, the
British historian Toynbee states in his book Mankind and Mother Earth: ‘Man is a
psychosomatic being, acting within a world that is material and finite . . . But Man’s
other home, the spiritual world, is also an integral part of total reality; it differs from
the biosphere in being both immaterial and infinite; and, in his life in the spiritual
world, Man finds that his mission is to seek, not for a material mastery over his
non-human environment, but for a spiritual mastery over himself’ (Toynbee 1976.
It seems prudent to be aware of the degree to which theories big and small are a
reflection of dominant scientific fashions, schools, disciplines and individuals.
16 This is also the view, though from a different angle, of the French agronomist Gourou (1947):
‘ . . . des populations denses sont généralement douées d’une civilisation supérieure: elles ont en
effet su résoudre les problèmes économiques, techniques, sociaux et politiques posés par les fortes
densités’.
Tainter concluded in his book The Collapse of Complex Societies (1988) from
the available literature that explanations of the collapse of civilisations can be cat-
egorised into three groups:17
r resource- and environment-related changes, fully exogenous or partly endogen-
ous, in the sense of human-induced;
r interaction-related changes in the form of conquest or other, less dramatic forms
of invasion; and
r internal changes in socio-political, cultural, and religious organisation and world-
view, diminishing the adequacy of response to external events.
In his view, the key concepts are energy and complexity (Tainter 1988, 2000). First,
socio-political systems require physical energy sources to maintain the investment
in infrastructure that keeps the members of the society acting together. Second,
the increase of social complexity is a strategy to solve societal problems such as
diminishing productivity. Initially, such a response has a high return but in due course,
the returns may decline to zero or even become negative. The first cooperative
efforts to store food or share irrigation channels were very effective, whereas adding
another organisational unit to an already centralised administration may have only
costs and no benefits. If social and political complexity increases in an attempt to
solve problems and spread risks in a variable and uncertain environment, it may
at some point start to be counterproductive and merely serve to reproduce the
institutions necessary to deal with the complexity. This is a vicious cycle that leads
to an ineffective bureaucracy and, ultimately, collapse.
In this interpretation, the infrastructure costs of integrating Spain, Gaul and
other territories into the Roman Empire were relatively low because it was delib-
erately based on incorporating existing infrastructures of local origin. But around
the 3rd century CE, the imperial organisation had reached a point of declining mar-
ginal returns. Further expansion and the tapping of new sources of energy became
unprofitable. Once expansion had halted, the Empire lacked sufficient reserves of
energy to maintain the existing infrastructure when it came under pressure. Dis-
integration became inevitable, either as a planned devolution or as a more or less
gradual breakdown. The strategy of Diocletian and his successors in the 4th century
to cut the Empire into two, and later four, parts can in retrospect be considered as
one of the failed attempts at avoiding collapse by simplification and devolution. The
Eastern part, the later Byzantine Empire, was an example of a successful devolution,
based as it was on a planned simplification of society.
This theory raises a number of questions. For instance, there is the problem
that (social) complexity is not easily measured and that the provenance and role of
innovation is only implicitly addressed. The ecologist-historian Turchin has proposed
in his book Historical Dynamics (2003) an ethno-kinetic theory of rise and fall of
agrarian states that is based on population ecology and social dynamics. His source
of inspiration is the 14th-century Arab sociologist avant la lettre Ibn Khaldun, who
developed a theory of political cycles from studying the history of the contemporary
Arab world. He observed that human individuals cannot live outside a group and
17 This classification matches the categories of extra-, inter- and intra-human change introduced by
Elias.
that cooperation and a communal way of life are necessary ingredients of human
life. Different groups have different abilities for concerted action, which led Ibn
Khaldun to introduce the concept of asabiya. It can be translated as ‘group feeling’
or ‘collective solidarity’.18 In his view, life in the desert tends to increase asabiya,
whereas life in the civilised urban centres, with their luxuries and intra-elite conflicts,
leads to its diminution. Thus, a human group initially has a large collective solidarity
or asabiya and lives austerely in great ethnic unity. It can manifest itself in an outward
expansion and conquest of the urban centres.
Using data and theories from a variety of social sciences, Turchin offers a formal
framework for state growth and decline. Asabiya is a state variable, which tends to
grow towards some fixed value, but at a rate dependent on the territorial size. Its
growth rate is high when the territory is still small and there is a large sense of solid-
arity, for instance, in the form of ethnicity. Territorial area, the other state variable,
tends to grow also logistically towards some fixed value. In the model, interaction
between individuals is a key element and the centre-periphery gradient in wealth
and skills and solidarity increases until the downfall sets in. Such a mathematical
model of a complex phenomenon should not be seen as prediction in any sense;
its purpose is to yield insights into the relative importance of phenomena. But it
incorporates the notion of cultural identity as a driver of change, which is against
most expectations back on the scene in early 21st-century world history.
Box 3.5. Colonists, Inuit and the church.19 Danish colonisation of Greenland
started in 1721, but a Norse colony in west Greenland had already been an outpost
of Europeans throughout the period 985–1500 AD. There had been contacts
between North American hunters and European farmers. It seems that the Norse
community managed quite well for the first 150 years, with a maritime-terrestrial
economy with small animal herds, seasonal hunting of walrus, polar bears and
seals and occasional trading with Iceland. However, they were living on a knife
edge, with great skills being required to survive the long cold winters. There
is ample evidence that the Little Ice Age fluctuations in temperature, sea ice
conditions and faunal resources put the communities under stress from 1270 AD
onwards. Moreover, the southward migration of the indigenous Inuit peoples –
in response to climatic change – and the decline in trading relations with Europe
further complicated their subsistence strategies. The western settlements seem to
have collapsed rather suddenly around 1350 AD.
In this story, it has been argued, one should not treat the human response to
climatic stress as a minor and dependent variable. The Inuit peoples survived
these harsh times. The Norse farmers had several options to adapt, for instance,
orienting themselves more towards the oceanic resources and de-emphasising
pasture and cattle-rearing. They also would have benefited from Inuit practises
and technology related to boats, fishing gear and clothing. ‘Rather than exploring
the possibilities of new technology and searching out alternative resources, Norse
18 Asabiya produces ‘the ability to defend oneself, to offer opposition, to protect oneself, and to press
one’s claims’, as stated by Ibn Khaldun and quoted by Turchin (2003).
19 This paragraph is largely based on McGovern (1981) in de Vries and Goudsblom (2002). For another
account, see Diamond’s Collapse (2006).
society in Greenland seems to have resolutely stuck to its established pattern,

elaborating its churches rather than its hunting skills’ (McGovern (1981) 425).
Whence this conservatism and loss of adaptive resilience in the face of rising
economic costs and declining returns?
Under the influence of Iceland, the mediaeval church in the Norse communities
in Greenland had become more powerful spiritually as well as materially in the
12th century. Between 1125 AD and 1300 AD, spectacular church construction
had taken place, small communities building amongst the largest stone structures
in the Atlantic Islands. Economic, political, religious and ideological authority
appears to have come into the hands of a lay and clerical elite. If a society such as
this is confronted with increasing fluctuations in resource abundance and use, it
has to invest in additional data collection and improve its interpretation of these
data in order to survive. This constitutes an overhead cost, which is often resisted
by the population and has to be enforced by military force or by ideology.
The elite-sponsored expensive elaboration of ceremonial architecture and
ritual paraphernalia in Norse Greenland may indicate the successful ideological
conditioning of the population. Administrators may not only have declined Inuit
superior technology but also have sustained erroneous beliefs – for example, that
lighting candles had more impact on the spring seal hunt than more and better
boats. As stresses mounted, elite groups may have pushed up the necessary over-
heads in their obsession with conformity and the suppression of dissenters and
detached themselves farther and farther from the phenomenal world, adding the
pathology of hypercoherence to the pathology of auto-mystification.
3.6 Summary Points

What do these lessons from history tell us about the present and future? Do people
inevitably run against food limits, as Malthus believed, or are there always timely
feedbacks that prevent collapse, as Boserup argued? It is probably a complex mix.
The natural environment has always played an important if not decisive role in
the growth and decline of human societies. Yet, environmental determinism is not
permitted for all but the simplest human societies. Several civilisations during the
agricultural regime were able to sustain themselves more or less continuously for
periods up to 1,000 years or 35 generations. In view of the recent duration of the
industrial era – say, 300 years – this is impressive. The relatively stable climate during
the Holocene and the cyclical agro-ecological patterns and social-religious beliefs
that dominated human societies are part of the explanation. Many people in those
civilisations will have experienced their societies, if not their individual lives, as
sustainable. Such a stability is hard to imagine in the early 21st century, and perhaps
we should search the past not only for the causes of unsustainability but also of
sustainability. Other points to remember from this chapter:
r development was primarily a dependency-control spiral of outside and inside
opportunities and threats. The mechanisms behind growth and decline trans-
itions are a positive feedback loop of expansion, reinforced by population dens-
ity, followed by negative feedbacks such as overwhelming complexity and envir-
onmental constraints;
r the decline/collapse of several past civilisations can be assigned to overexploita-

tion of the resource base in an environmentally fragile environment with (large)
natural variability;
r with agrarianisation, the interaction between humans and the natural environ-
ment became more intense and socio-cultural processes gained in importance;
r with development, the spectrum of environmental risks widened from nearby
and short-term to indirect and long-term – another element in the dependency-
control spiral;
r environmental deterioration eroded the economic-financial basis of the ruling
elites, creating social tensions and other disturbances such as sectarian violence,
invasion by hostile neighbours and dependence on other states;
r short-term thinking of societal elites aggravated the impending crisis in the form
of greed and hoarding, ineffective rituals in the form of sacrifices or monuments
and resource-demanding oppression or warfare.
The history of past civilisations teaches us some lessons that are the more important
now that humanity has no more escape routes into large and unsettled areas and
some of the environmental changes become global and irreversible.
War has always been part of life. The quest for living space and for power and
exploitation has taken many forms. This is how Bertold Brecht remembered his
brother who left for the Spanish Civil War:
Mein Bruder war ein Flieger,

Eines Tages bekam er eine Kart,
Er hat seine Kiste eingepackt
Und südwärts ging die Fahrt.
Mein Bruder ist ein Eroberer,
Unserm Volke fehlt’s an Raum,
Und Grund und Boden zu kriegen, ist
Bei uns alter Traum.
Der Raum, den mein Bruder eroberte
Liegt im Guadarramamassiv,
Er ist lang einen Meter achtzig,
Und einen Meter fünfzig tief.
– Bertold Brecht, 1937
Complex societies, it must be emphasized again, are recent in human history.

Collapse then is not a fall to some primordial chaos, but a return to the normal
human condition.
– Tainter, The Collapse of Complex Societies, 1988
SUGGESTED READING
Overview of the IHOPE Dahlem conference with accounts of past civilisation developments.
Costanza, R., L. Graumlich, W. Steffen, C. Crumley, J. Dearing, K. Hibbard, R. Leemans, C.
Redman, and D. Schimel. Sustainability or collapse: What can we learn from integrating
the history of humans and the rest of nature? Ambio 36 (2007) 522–527.
Useful Websites 83
Models, maps and myths about the role of the natural environment in ancient civilisations.
de Vries, B., and J. Goudsblom, eds. Mappae Mundi – Humans and Their Habitats in a Long-
Term Socio-ecological Perspective. Myths, Maps, and Models. Amsterdam: Amsterdam
University Press, 2002.
Review of scientific data and theories.
Dearing, J. Climate-human-environment interactions: Resolving our past. Clim. Past. 2 (2006)
187–203.
In-depth account of a number of collapse events in human history (Easter Island, Maya, Iceland
a.o.).
Diamond, J. Collapse – How Societies Choose to Fail or Succeed. New York: Penguin Books,
2006.
An accessible, archaeological investigation into the role of the natural environment in ancient
societies.
Redman, C. L. Human Impact on Ancient Environments. Tucson: University of Arizona Press,
1999.
USEFUL WEBSITES
r www.pages-igbp.org, about Past Global Change research.
r www.igbp.net is the site of the International Geosphere-Biosphere Program (IGBP)
research projects.
r www.ihdp.unu is of the International Human Dimensions Program (IHDP) research
projects.
r www.ncdc.noaa.gov/paleo/ is a website with lots of information and links to palaeo-
reseach.
A simple STELLA R
model, with coconut palm trees and statues as state variables, gives
insights into the dynamics and the sensitivity of the outcome for the assumptions. To
run the Easter Island model in Stella, visit www.iseesystems.com/community/downloads/
EducationDownloads.aspx.
4 The World in the Past 300 Years: The Great
Acceleration
4.1 Introduction
Although past civilisations are a source of imagination and insight, present-day
concern about (un)sustainability is anchored largely in the exponential growth of
population and economic activity in the last few centuries. These growth processes
are part of what is known as the Industrial Revolution. Industrialisation occurred
in many places throughout Western Europe at roughly the same time, although
with locally specific features. It is rooted in the commercial and trade capitalism of
medieval Europe (Braudel 1979). A series of events and trends since 1700 mutually
interacted and boosted manufacturing and trading of goods in a successful mixture
of science, technology and capitalism. It reinforced the process of European colonial
expansion. The European ‘offshoots’ in America and Australia underwent similar
transformations as Europe. A collectivist form of industrialisation took place in
Russia after the revolution. The larger part of the human population, however,
still lived a traditional agricultural life at modest levels of population growth and
economic output until the middle of the 20th century. Only after 1950, they started
to experience similar processes of change.
This chapter explores in some detail the important changes that came with
the Industrial Revolution and modernity. Perhaps Churchill was right when he said,
‘The further backward you look, the further forward you can see’. In any event, some
knowledge and understanding of what happened in the last 300 years is essential for
an interpretation of our present situation and an exploration of what a sustainable
future might look like. For that reason, I briefly survey demographic and economic
trends, aspects of governance and globalisation, and socio-cultural trends. In the
last section, two generic concepts are introduced as descriptive tools that can put
the changes in a different perspective. The topic is vast, so I refer the reader to the
suggested literature for further reading.
4.2 The World in the Last Three Centuries
4.2.1 Accelerating Growth: Population and Economic Activity

The first major change to be mentioned is the number of human individuals. Between
the year 1000 and 1820, the world population increased from an estimated 268 to
84
Table 4.1. Key economic indicators 1913–1998a
Labour
Income Avg life productivity Govt
(GDP 1990 expectancy Working (GDP 1990 expenditures
I$/cap) at birth (yr) hours (hr/yr) I$/hr) (% of GDP)
Western Europe 3,473 46b 2,553–2,624 3.12e 12

1913 4,594 67 1,926–2,316 5.54 29.8
1950 17,921 78 1,428–1,664 28.53 45.9
1998
USA 1913 5,301 47b 2,605 5.12 8
1950 9,561 68 1,867 12.65 21.4
1998 27,331 77 1,610 34.55 30.1
Japan 1913 1,180 44b 2,588 1.08 14.2
1950 1,926 61 2,166 208 19.8
1998 20,084 81 1,758 22.54 38.1
Russia 1913 1,488 32
1950 2,834 65
1998 3,893 67
China 1913 552 24
1950 439 41
1998 3,117 71
India 1913 599 24
1950 619 32
1998 1,746 60
Latin America 35
1913 2,554c 51 2,042d 2.48d
1950 5,795 69 1,841d 7.87d
1998
Africa 1913 – 24
1950 852 38
1998 1,368 52
a Source: www.theworldeconomy.org/MaddisonTables/Maddisontable3–9.pdf
b 1900
c Forty-four Latin American countries (incl. Mexico)
d Brazil
e Weighed average for twelve Western European countries
1041 million (0.16%/yr). Since 1820 it increased almost sixfold to nearly 7000 mil-
lion (0.95%/yr) (Table 4.1; Figure 4.1).1 Perhaps most remarkable was the large
emigration out of Europe towards the Americas, Oceania and Africa (Figure 4.2).
Because the rapid growth happened in Europe and its offshoots, the geographical
distribution of the population over the world changed, too. If European Russia and
the European offshoots elsewhere are included, the European population increased
from an estimated 15 percent of world population in the year 1000 to 23 percent by
1820. Since then, the fraction has declined to 19 percent in 2000. Asian populations
made up about two-thirds of world population between 1000 and 1820. The popula-
tion of China reached an estimated high 38 percent by 1820, after which it declined
1 The numbers in this paragraph are from Maddison (1995, 2006) and www.ggdc.net/. Most numbers
cited in the chapter have rather large uncertainties, especially those before 1900 and for non-
European regions.
86 The World in the Past 300 Years: The Great Acceleration
8000
Oceania
SE Asia
East Asia
6000 South Asia
fUSSR & M. East
Mln people
Europe
4000 Africa
Latin America
North America
2000
0
1700 1760 1820 1880 1940 2000
Year
Figure 4.1. Global trends in human population (Source of data: Maddison 2006).
to about 21 percent in 2000. The population in the rest of Asia rose from about
29 percent by 1820 to 37 percent in 2000.
The different population growth rates between Europe, its offshoots and the rest
of the world result from the process known as the demographic transition (§10.3). The
first stage occurred in the 19th century when a spectacular decrease in the mortality
rate happened in several European countries. It has led to a phenomenal rise in
life expectancy at birth, from less than forty years in 1820 to more than seventy-six
years by 1990. It was the consequence of ongoing improvements in food quantity
and quality and of better sanitation, hygiene, medicine and medical practice. In
the second stage, the birth rate started to decline but the delay with the mortality
rate decline caused a period of rapid population growth. In the third stage, where
most industrialised nations are now, population sizes stabilise or even decline. The
transition can be mathematically described with a logistic growth curve (§2.3).
Most countries in the world are in one of these stages of the demographic
transition. For instance, estimated average life expectancy was still below thirty-five
years in 1900 in Spain and Russia but had risen above sixty-two years in 1950. In many
other countries, including Mexico, Brazil, China, India and most African countries,
life expectancies were still below fifty years in 1950. Since then, however, almost all
went into the transition and as a result experienced a rapid growth in population size.
In the fifty years since 1960, many countries in Asia, Latin America and Africa –
Philippines, Mexico, Nigeria, to mention a few – experienced a growth of their
population that is equivalent to what the United Kingdom experienced in the 150
year period between 1750 and 1900 (Figure 2.10). In other words, three times faster.
These high growth rates in an increasingly connected world put an enormous strain
on these countries, not only in terms of food and infrastructure but also in terms of
employment and governance.
During these centuries, the process of urbanisation led to ever more geograph-
ical concentration. A bit dependent on how it is defined, the fraction of humans
living in cities increased from 3 percent in 1700 to more than 47 percent in 2000.
Historical population density
1700 A.D.
a)
2000 A.D.
b)
inh/km2
<1 6 - 10 31 - 50 101 - 500
1-5 11 - 30 51 - 100 > 500
Figure 4.2.a,b. Map with population densities 1700 and 2000 (Source: HYDE/PBL). (See
color plate.)
The urbanisation process is still going on, at rates higher than population growth. It
leads in the low-income regions of the world to megacities of unprecedented size.
Not only do they rapidly become new centres of economic activity and technology,
with a burgeoning middle class – but also places where millions of people live in
poverty and with inadequate public services such as water, sanitation and electricity.
Congestion and air pollution are another increasing problem (Figure 2.3b). Qual-
ity of life is deteriorating, and signs of unsustainable developments, such as food
riots and political conflict, are most visible in these megacities. The major attractor
is employment and income opportunities, and the only stabilising feedback is a
India
China
30000
Avg Total 30 Europe
United States of America
Total Former USSR
Avg Total Latin America
$/yr/cap
20000
10000
0
1820 1840 1860 1880 1900 1920 1940 1960 1980 2000
Year
Figure 4.3. Global trends in income in 1990 (International Geary-Khamis $/yr/cap; source of
data: www.ggdc.net and www.conference-board.org/economics/database.cfm).
slowdown in immigration from rural areas. Although the stories about London,
Paris and New York in the 19th century picture similar dismal circumstances, there
are also differences: The 21st-century metropoles have not 1–2 but 12–15 million
inhabitants, and they have extensive communications with the outside world and
overexploited natural resources in the surrounding areas.
The second and related large change concerns economic activity. Average
income has risen thirteenfold since 1820 (Table 4.1; Figure 4.3).2 Maddison dis-
tinguishes five growth ‘phases’: 1820–1870; 1870–1913; 1913–1950; 1950–1973; and
1973–1992. These phases represent identifiable segments of the economic growth
process in the world. Notably the periods 1870–1913 and 1950–1973 were times of
high growth rates. Long-term analyses reveal oscillations – the Kondratiev cycles,
after the Russian economist who first identified them – which more or less coin-
cide with the four growth phases. It appears that science-driven technology is the
dominant force behind these growth cycles: They are at least partly explained by
the interaction between waves of innovations and investments in capital-intensive
industries (Sterman 1986; Tylecote 1992).
Because economic growth since 1820 has been very uneven across the world,
income distribution did change drastically. Before 1820, differences in average
income amongst world regions were less than a factor 2. By 1950, incomes in Western
Europa and in particular its offshoots in North America and Australia had increased
fivefold to tenfold since 1820. In the same period, average incomes in Asia rose with
at most 40 percent and in Africa with an estimated 50 percent. After World War II,
the economic take-off of countries such as Japan and Korea led to a sixfold increase
in average income between 1950 and 1992. Other Asian countries follow, but African
countries lag behind (Figure 4.3).
2 GDP is defined as the total of monetary transactions in an economy. GDP per inhabitant is known
as income and is a flow in €/yr/cap. Time-series are usually corrected for inflation. For international
comparisons, the US$ can be converted to international dollars I$ in order to account for differences
in purchasing power between countries (PPP; see pwt.econ.upenn.edu/php_site/pwt_index.php).
GDP/cap is not an adequate measure of well-being, but one of the best if not only ways to picture
long-term trends in human activity (§14.4).
Box 4.1. Mechanisms behind income distribution. A number of scientists have

applied network theory in order to get a better understanding of the causes of
income inequality. Boucheaud and Mézard (2000) apply the ‘directed polymer’
problem in physics to economics. They assume that the world is inhabited by
agents with a few simple characteristics: The agents can exchange wealth through
trading and the amount of money they earn or spent is proportional to their
wealth. This can be formulated in a stochastic dynamic equation for the wealth
Wi (t) of agent i at time t. It can be shown that, if assuming that all agents exchange
with all others at the same rate, the system ends up in a situation where a small
number of agents has a large share of the wealth. This is observed in the real
world and known in economic science as the Pareto power-law distribution.
The authors explore the effects of income and capital tax, suggesting that the
former may decrease income inequality and the more so if redistributed. The
latter can, surprisingly, cause a lower spread of wealth if combined with income
tax and without redistribution. If the agents have exchanges according to some
social network configuration, again a Pareto distribution is found. It is concluded
that wealth tends to be very broadly distributed when exchanges are limited
and that favouring exchanges and, less surprisingly, increasing taxes seem to be
efficient ways to reduce inequalities.
Of course, many other factors determine income distribution. There is, for
instance, a systematic positive relationship between asset holdings and the return
on these holdings, which explains why household wealth distribution is much
more concentrated than income distribution. The existence of some extremely
wealthy households is possibly explained by the disproportionate large number of
entrepreneurs in the top income class. Also, the increasing role of unique talents
to be exploited on an increasingly global market for information and services is
giving rise to large income and wealth differentials.
In preindustrial societies, most people lived a subsistence life in economies

dominated by agriculture and commerce, and only a small part of the population
belonged to the rich upper class. The large income inequalities within these societies
declined since the early 19th century, particularly in Europe and its offshoots. But
the differences in income growth have created significant disparities in income and
wealth between countries. Between 1950 and 1992, income disparities did not change
much, and world income distribution became even slightly more equal between 1970
and 2000 (Barro and Sala-i-Martin 2004).3 Since 2000, they appear to increase again.
The extremes are drifting apart: Since 1960, the income ratio between the twenty
richest and the twenty poorest countries in the world has increased from five to
twelve. The acclaimed benefits of globalisation for the poor are apparently undone by
the opportunities for the rich to accumulate more wealth. The estimates of how many
people live in absolute poverty – which is based on a poverty line established by the
World Bank and presently set at $1.25 US/cap/day – indicate a decline as percentage
of the world population but a stable or even increasing level in absolute numbers.
3 If within region income differences are included in the analysis, world income inequality increased
rapidly and continuously since 1820, with a pause between 1910 and 1929. Thereafter, it decreased
(Bourguignon and Morrison 2002).
Trends in population and economic activity are key indicators of the processes
of change. However, the rise of modernity was an extremely complex process with
many interlinked phenomena, which are not easily disentangled from the whole
and for which often no adequate quantititive indicators are available. One such
phenomenon is what economists call structural change, that is, the gradual shift in
sectoral shares in GDP and employment from agriculture to manufacturing and then
to services.
The fraction of the population employed in agriculture declined in the last cen-
turies all over Europe and so did agriculture’s share in GDP (Figure 4.4). This change,
with important consequences for natural resource use, is well documented for the
present high-income countries in the world.4 The surplus of labour from agricul-
ture was absorbed by industrial activities, and the share of industry in employment
and GDP rose. Key manufacturing sectors were initially mining, ironworks, textiles,
ceramics, machine tools, food processing and base chemicals, and later transport
(ships, trains, automobiles), electrical equipment, plastics and electronics. A next
stage of development started in the 1960s, in the industrialised regions with a shift in
GDP and employment to the services sector. It represents the advent of the service-,
the information- or the experience-economy, depending on what is seen as the most
crucial component. Whether the low-income countries will follow this trend is as yet
unclear.
Another series of trends can be characterised with a single word: globalisation.
The spreading of people and goods and the dissemination of ideas across the con-
tinents started millennia ago. The Roman Empire and later the Asian, Arab and
African Empires had important trade routes. From the 17th century onwards, the
process accelerated as part of European expansion and colonialism. Apart from
the stagnation period of 1910–1945, world trade tonnage has grown exponentially.
Seaborne international freight in tons loaded increased more than fourfold between
1880 and 1910, with a further tenfold increase between 1945 and 1980 (Chisholm
1990). As fraction of Gross World Product (GWP) exports made up about one sixth
in 2000, a fourfold increase since the early 20th century.
The degree to which countries are connected in the world economy through
trade (im- and exports) varies considerably. Table 4.2 gives some key indicators for
a few countries for the year 2009 – a snapshot view because trade flows can fluctuate
significantly.5 The four richest economic countries/blocks (USA, EU-27, Japan and
Korea) account for 35 percent of world merchandise and 43 percent of commercial
service trade. China is rapidly gaining a larger role. For all these countries, trade flows
as fraction of GDP are above 27 percent. The interconnectedness and specialisation
is clearly visible in the numbers. In Saudi Arabia, Nigeria and other fossil fuel or
mineral-rich countries, more than 85 percent of merchandise exports are fuel/mining
4 Different sector classifications and names are used. For instance, agriculture, forestry and mining are
also called primary sector, with manufacturing the secondary and services the tertiary sector. There
are discussions about the correct of measuring sector shares. For instance, if changes in productivity
are considered, the share of the service sector has risen much less (Teives, Henriques and Kander
2010).
5 In the Netherlands, for instance, there was a surge in exports during the wave of globalisation
between 1870 and 1913, with more than 55 percent of GDP exported. It dropped to less than
20 percent in the 1930s but was again above 50 percent by the end of the 20th century.
UK USA
100% 100%
Fraction (%) 80% 80%
Fraction (%)
60% 60%
40% 40%
20% 20%
0% 0%
1820 1913 1950 1998 2008 1820 1913 1950 1998 2008
Year
Year
GERMANY JAPAN
100% 100%
80% 80%
Fraction (%)
Fraction (%)
60% 60%
40% 40%
20% 20%
0% 0%
1913 1950 1998 2008 1913 1950 1998 2008
Year Year
FSU/RUSSIA CHINA
100% 100%
80% 80%
Fraction (%)
Fraction (%)
60% 60%
40% 40%
20% 20%
0%
0%
1913 1950 1998 2008
1913 1950 1998 2008
Year
Year
Services
Mining, Manufacturing, Construction and Utilities
Agriculture, Forestry and Fishing
Figure 4.4. Structural economic change: shares in employment for six countries. The United
Kindgom and the United States were leading; France (not shown) was almost identical to
Germany (Source of data: Maddison 2006, www.ggdc.net and www.conference-board.org/
economics/database.cfm).
91
Table 4.2. Trade characteristics for some countries/regions in the world in descending order of income,
2009a
Korea, Saudi Russian

2009 USA Japan EU-27 Republic Arabia Federation
Population (mln) 307 128 499 49 25 142

Income (PPP US$/cap, 2009)b 46,437 32,433 30,543 27,195 23,421 18,938
Income (current US$/cap, 2009) 46,437 39,714 32,851 17,095 14,535 8,673
Frac Trade pc (2007–2009) in 27 30 30 109 110 58
Income
Share Merch Ex/imports in 11 5 17 3 1 2
world trade
Frac ServTrade in TotTrade 23 19 27 16 16 17
2009 Mexico Brazil China India Vietnam Nigeria
Population (mln) 107 194 1,332 1,155 87 155

Income (PPP US$/cap, 2009) 14,341 10,429 6,838 3,275 2,956 2,149
Income (current US$/cap, 2009) 8,146 8,116 3,744 1,134 1,052 1,092
Frac Trade pc (2007–2009) in 67 24 51 45 150 61
Income
Share Merch Ex/imports in 2 1 9 2 1 0
world trade
Frac ServTrade in TotTrade 7 20 11 29 9 4
a Source: WTO.
b Purchasing power parity (PPP) values are corrected for differences in purchasing power between countries.
products. More than one third of merchandise exports in Brazil and many African
countries is agricultural product. India and EU-27 rank with 29 percent and 27
percent highest in the fraction of commercial service trade in total trade. Commercial
service trade in low-income countries is nearly all in travel and transportation, which
is an indication of the importance of tourism.
The increase in global trade has many and, according to economists, mostly
beneficial consequences. Its net effect in facilitating the penetration of a market-
oriented capitalism into the farthest corners of the world is not easily assessed. Trade
and growth are not directly or simply connected (Helpman 2004). From an ecological
sustainability perspective, the use of fossil fuel in the ocean bulk and container ships
and the, sometimes on purpose but often unintentional, dissemination of bacteria,
plants and animals across the continents are among the negative side effects (Crosby
1993; Diamond 2006).
The migration flows of people across the globe increased significantly in absolute
numbers. In the 19th and early 20th century, there was a massive emigration flow
out of Europe as a consequence of famines, unemployment and wars. There were
also massive migration flows within countries, notably within Russia and China. In
the past, migration was caused by rural overpopulation and famines such as the
infamous Irish Potato Famine in 1845–1852. In some cases, it was the result of
deportations and wars. There is still large-scale migration but different. Emigration
flows to North America and Australia and New Zealand were in 1998 twice the
size of the flows in 1900, yet they no longer originated from Europe but from Asia
and Africa. In Europe, the flow changed sign in this period, from emigrants to
immigrants (Maddison 2006). Present-day migrants are mostly economic migrants,
that is, people looking for better opportunities to earn a living. It is still for the larger
part a consequence of rural overpopulation and natural disasters and wars. In the
early 21st century, the flow of refugees is estimated to be several tens of millions of
people. Yet, the interregional flow of migrants is still a small fraction of the labour
force.
Global capital flows have increased enormously with the deregulation trend
since the 1980s. Capital, and with it knowledge, is no longer an immobile production
factor in the 21st century. Its mobilisation was instrumental in the explosive growth of
economic activity. International financial flows have reached unprecedented levels,
but only about 10 percent of the transactions are related to the ‘real economy’ of
trade and production, the remainder being essentially speculative dealings (Dicken
2009). In the first decade of the 21st century, global financial system has become one
of the most important factors in the rate and direction of economic growth. In its
unregulated ‘casino capitalism’ form, it contributes to unsustainable development
because of its relentless search for the highest returns and its inherent instability
resulting from speculative and herding behaviour. In the process, the traditional role
of the state in creating and controlling money flows is being eroded.
4.2.2 Social-Cultural Changes

Equally important as the numbers about growth are the more difficult to quantify
social, cultural and political changes, which were enablers as well as outcomes of the
above-sketched change processes. The common view is that the philosophical and
societal basis of western civilisation was laid in the 18th-century period of the Enlight-
enment. It was then that the idea of modernity emerged: the gradual replacement
of feudal, political and administrative power by commercial interests and market
forces; the emergence of democratic forms of governance with separation between
church and state; and the empiricist-rational approach to the world expressed in
science and technology. The United States Constitution of 1787 and the French
Revolution that began in 1789 are two of the hallmarks of these changes.
Cultural and political developments are visible in a society’s institutions. In the
course of evolution, human beings have learned to organise themselves in groups to
get the benefits of cooperation (Wright 2000). Past and present institutions, foremost
the nation-state but also armies, churches, ministries, corporations, sport clubs and
so on, evolved out of the early tribal groups in order to benefit from cooperation
and coordination and restrain behaviour that is harmful for the group. They are
cornerstones of societal as well as their own stability and continuity, in their role of
repositories of learning and customs and of conflict mediators. They have proven to
be the most durable structures in the human world: flexible yet resilient and able to
sustain themselves for centuries.
An example is the Catholic Church, possibly one of the longest surviving institu-
tions. Monastic orders such as the Cisterciënzers have survived wars and revolutions
for more than eight centuries. Nation-states, as political and ethnic/cultural units tied
to a particular geographical territory, have succeeded to maintain a certain coherence
for many hundreds of years. Some multinational corporations have been sustaining
an essentially unchanged identity for many decades.6 Evidently, such institutions

need to have the capability of self-renewal and adaptation, in combination with a
degree of inertia and opportunism. A hierarchical structure may be a precondition,
provided it does not become too rigid.
The other side of institutional resilience is resistance to change. So how was
industrialisation possible against the vested institutional interests? There are no easy
answers, but economic historians suggest some clues (Heilbroner and Milberg 2005).
Before 1700, economic life in Europe was largely traditional and had no incentives to
expand. There were all kinds of rigidities impeding the production and use of goods,
services, labour and capital: ‘gild regulations, monopolies, an excessive schedule
of holy days (ninety-two in France in 1666), sumptuary legislation, monasticism,
settlement laws, price controls, and taboos and religious sanctions on economic
behaviour or even on the study of science and technology’ (Jones (2003) 96). These
were large social and religious barriers to industrialisation, often actively defended by
elites and institutions with their preference for privileges and command-and-control
(‘conservatives’).
With the advent of merchant capitalism and later industrial capitalism, the real
force of the market mechanism in combination with science and technology became
apparent. The organising entrepreneur, the enabling capitalist and the industrial
labourer came to dominate economic life. Economic growth and innovation were
the outcome of personal ambitions to grasp opportunities and reap large monetary
rewards. There was no central control, coordination was done by the ‘invisible hand’
of the market and democracy was the political reflection – at least that was the myth.
Industrialism generated new forces of resistance, notably socialism in response to the
human sufferings of early industrialism and romanticism as opposition against the
new rationalism and commercialism. In the process, European society created some
of its best social science and art. But the capitalist system and its liberal ideology
have turned out to be resilient, as they fulfilled expectations of a better quality of
life, adapted to all kinds of change and economically outperformed the 20th-century
state-led command-and-control economies.
Partly in conflict and partly in association with these changes, the role of the
(European) nation-state in life did increase, not decrease. With the rise in social
and economic complexity, collective arrangements became more important in order
to protect citizens with legal codes of conduct and their enforcement against fires
and floods, epidemic diseases, external enemies – and against each other. With the
rise of European colonialism since the 16th century, the state served as protector
of exploitation and trade in the colonies. In the 19th century, state interventions
broadened as a consequence of scientific and technological developments and of
enlightened ideas about societal progress. Sanitation and sewage systems, medical
care and schooling, clean air and water, social security and income redistribution all
became to some extent state responsibilities.
6 According to the Forbes 100 list of United States companies, most had gone bankrupt or merged
between 1917 and 1987. Eighteen companies were still in the elite group: “These eighteen com-
panies . . . were grand-champion survivors, weathering the storms of the Great Depression, World
War II, the inflationary 1970s . . . ” (Beinhocker 2005). Interestingly, almost all of these companies
underperformed in terms of stock market value.
With this came the need for larger state revenues. In ancient and medieval times,
random confiscations and political arbitrariness by the elite were quite common in
order to finance their luxuries and rivalries. This misinterpretation of what is in
the interest of the state was aptly expressed by the French King Louis XIV when
he tried to secure money for his wars in Parlement: ‘L’etat, c’est moi!’. Gradually,
state revenues have become organised in administered trade and taxation schemes
(Jones 2003). In Europe, government expenditures as fraction of GDP increased
in the course of the 20th-century from an estimated 8–18 percent to 40–52 percent
(Maddison 2006; Table 4.1). Part of it is for the delivery of public goods and services
like infrastructure and health and education. Another part is redistribution from the
rich to the poor and unemployed – social security arrangements that became the
hallmark of the 20th-century welfare state. Elsewhere, the fraction of government
expenditures in GDP is smaller: less than 40 percent in the United States and Japan
and 20–30 percent in Africa.
By the end of the 20th century, the role of the state became less prominent in
the wake of the fall of communism and the rise of neoliberalism. New societal actors
such as the large and increasingly multi-/transnational corporations (MNCs/TNCs)
and a variety of Non-Governmental Organisations (NGOs) enter the public arena
and create, in combination with the Internet, an ever-more interconnected and volat-
ile world. At the same time, the American-European postwar model of economic
growth and its modernist worldview are challenged by the emerging economies of
China and other countries. The pressure on finite resources increases, while the
expectations of hundreds of millions of people keep rising. Governments find it
more and more difficult to balance expenditures and revenues, with the resulting
problems of economic debt and inflation and social unrest. The deregulated financial
system, with the right to create money, owes its existence partly to this tendency.
These are the ingredients of economic (un)sustainability.
Perhaps the least tangible and yet most important and enduring change has to do
with values. Social scientists have attempted to grasp the cultural aspects of change,
with an important role for the distinction between individualism and collectivity
(Durkheim) and for religion (Weber). More recently, extensive empirical data have
been collected in the World Value Survey, a large investigation of attitudes, values
and beliefs around the world in four waves (1981–1983, 1989–1991, 1995–1997 and
1999–2001). The results have been analysed and presented by Inglehart and Welzel
(2005) as a (revised) Theory of Modernisation. The empirical data, it turns out,
suggest two crucial value dimensions: traditional versus secular/rational, and survival
versus self-expressive.
If the data are confronted with indicators of socioeconomic development, such
as income and economic sector shares, it is found that the process of industrialisation
coincides with a clear value shift from traditional to secular/rational. The shift correl-
ates with the employment share of industry (Figure 4.4). A process of secularisation
and rationalisation but also bureaucratisation and centralisation sets in, during which
people become less dependent on the caprices of nature and science, and technology
takes the place of religion as a source of authority. Social life gets organised in the
uniform way of mechanised mass production, with rigid social classes and uniform
standards. It appears this is the price people willingly pay for the enlarged control
of technology over nature.
With the further rise in income that happened in the advanced economies of
Europe and its offshoots in the second half of the 20th century, there is another
shift from survival to self-expression values. Survival needs are largely met and
people’s values move towards self-expression. This change coincides fairly well
with the employment share of the service sector (Figure 4.4). Existential secur-
ity concerns are replaced by the desire for personal autonomy, self-fulfillment
and other forms of self-expression. Keywords in the postindustrial society are the
provisions of the welfare state, the intellectual mobilisation through advances in
information and communication technology (ICT), and liberation from the social
constraints of family and community. In the context of environmental or, more
broadly, sustainability concerns, it is worth mentioning that issues such as environ-
mental protection, recycling, education, gender and income equality are amongst
the postindustrial self-expression values. Whereas industrialisation liberated people
from the constraints posed by the natural environment with secularisation of author-
ity, postindustrialisation led to an emancipation from authority (Inglehart and Welzel
2005).
The value survey results indicate distinct clusters in the two-dimensional space.
For instance, Protestant Europe ranks high on both dimensions, whereas low-income
countries in Africa and South Asia rank low on both. Former communist countries
rank high on secularisation but low on self-expression, an obvious legacy of commun-
ism. Catholic and Protestant countries show marked differences. While Protestant
societies rank high on interpersonal trust, which is positively correlated with local
control, Catholic societies tend to put higher value on hierarchical, centralised con-
trol and rank lower on the secularisation dimension.
It seems that socioeconomic development tends to propel people all over the
world from a survival-oriented traditional worldview towards a more secular and
rational worldview with emphasis on expression of the individual self. Yet, the
survey results also show convincingly that the cultural traditions of individuals and
nations do play an important role, too. Unlike some social scientists have argued,
modernisasion is not the same as Westernisation and the United States is not the
model for the cultural changes happening around the world. Cultural modernisation
results from socioeconomic developments, but it is also influenced by the traditional
cultural heritage and may also be reversed if the economy collapses.
4.3 Accelerating Impacts: The Natural Environment
4.3.1 The Source Side

There were not only economic and social but also natural obstacles to industrialisa-
tion. However, the natural environment posed less and less resistance to the control
over natural that science and technology brought. The growth of population and
economic activities influenced the natural environment, slowly but steadily, but it
hardly raised widespread concern about the consequences for human populations:
‘restraint in the interest of the biosphere was rarely given any consideration at all
until the emergence of a more or less worldwide environmental movement in the
1960s’ (McNeill (2007) 302).
Box 4.2. The European Miracle. Many books have been written about the reas-
ons for Europe’s ascendancy in the last 300 years, amongst them The European
Miracle by Jones (2003). It is a study of an historian on Europe’s comparative
advantage in becoming the world’s first and leading industrialist region. Jones
argues that environment, market and state have to be analysed to understand the
reason. Europe’s industrialisation is of much earlier origin than the rise of Britain
and cannot be attributed mostly to the exploitation of colonies. In Jones’ view
it is a mix in which the relatively small-scale, political decentralisation, markets
for land and labour and rather small income differences makes the difference
with the steppe imperia (Mongol, Moghul, Ottoman), which were command hier-
archies imposed upon customary agriculture. Also, European ‘governments’ had
more concern about public goods such as disaster relief and incentives for private
investment. ‘A relatively steady environment and above all the limits to arbit-
rariness set by a competitive political arena do seem to have been the prime
conditions of growth and development. Europe escaped the categorical dangers
of giant centralised empires as these were revealed in the Asian past. Beyond that,
European development was the result of its own indissoluble, historical layering’
(Jones (2003) xxxvii).
The economist Maddison (1995) has a different emphasis in the list of factors
that have favoured Europe’s development: rational scientific investigation and
experiment, the ending of constraints on property transactions, more trustworthy
and regular financial institutions and taxing, and the system of nation-states and
family traditions, which induced more modest population growth than elsewhere.
The most visible changes had to do with land (Figure 4.5a,b). Agriculture was and
still is the greatest force of land cover and land use change. Next come forest cutting
for timber, expansion of urban lands and mining. Between the year 600 CE and the
early 13th century, the population in Europe went from 18 million to 76 million while
the land area covered by forests and swamps decreased from roughly 80 percent to
50 percent in this period. Similar changes happened in other densely population
regions of the world. In 1700, roughly 2–3 percent of the global land surface was
cultivated. In 2000, it was an estimated 37 percent.7 Cropland expanded in the last
three centuries from 3–4 million km2 to 15–18 million km2 , largely at the expense
of forested lands. Global forest area declined from an estimated 53 million km2 in
1700 to 43–44 million km2 today. Deforestation accelerated in the 20th century with
large-scale cutting of tropical forests for cropland and grazing land. In high-income
regions, cropland area hardly changes anymore or even declines. Wood production
also stabilises since 1990. However, water use, irrigated area, production of crops
and meat and the catch of fish continue to grow (Figure 4.6a,b). At present, an
7 Most of the date cited are based on Lambin and Geist (2006) and Klein Goldewijk (2004). It
should be noted that there are considerable uncertainties, due to different definitions and limited
data. Besides, not only areas but also quality, e.g., of forests, has to be considered (§ 11.3 and
§ 12.2).
a)
Figure 4.5a,b. Map with land use in 1700 and 2000 (Source: HYDE/PBL). The upper (a) shows
the fraction of a gridcell used as cropland; the lower (b) shows the fraction used as pasture.
estimated 25 percent to 39 percent of the net biomass production on earth is under

human control (Haberl 1997).
The development perspective on nature is that of a resource, but the resources
have come under increasing pressure. One fishery after another is depleted through
overexploitation and habitat degradation. World fish catch from open fisheries has
been in decline since the 1990s and sustaining supply for an increasing demand
is only possible by large-scale expansion of aquaculture. Non-renewable resources
b)
Figure 4.5 (continued)
such as high-grade mineral ores and low-cost oil and gas deposits are depleted to
the extent that newly discovered ones have lower ore grades and are farther away
and/or deeper and more fragmented. Their exploitation causes additional burdens
on the environment, as is evident from, for instance, the exploitation of tar sands in
Canada as a source of oil. There are also more insidious changes going on. In many
parts of the world, soil fertility has declined because of a combination of agricultural
and pastoral exploitation and the natural forces of water and wind erosion.
a) Wood Production (Mm3/yr) Crop area (Mha)
Crop Production (Mt/yr) Water use (km3/yr)

Units 4000
2000
0
1950 1960 1970 1980 1990 2000 2010
Year
Figure 4.6a. Global trends in wood and crop production, crop area and water use since 1950
(Source of data: FAO/PBL). (See color plate.)
4.3.2 The Sink Side

The changes on the source side show up on the sink side of system Earth in the
form of emissions to air, water and soil and subsequent interference with natural
element flows across the planet. The natural carbon (C), nitrogen (N), sulphur (S)
and phosphorus (P) flows are to an even larger extent, and at an ever higher rate,
influenced by human activities (Figure 4.7). Exponential growth in the combustion of
fossil fuels and the change in land use and production of cement resulted in an average
9.3 billion tons of carbon emitted into the atmosphere annually in the period 2006–
2009. This is responsible for a continuous rise of the atmospheric CO2 -concentration
from an estimated 284 parts per milion on a volume basis (ppmv) in 1832 to 389 ppmv
Irrig area (Mha) Total fish prod (Mt/yr) Aquacult fish prod (Mt/yr)
b)
300
200
Units
100
0
1950 1960 1970 1980 1990 2000 2010
Year
Figure 4.6b. Global trends in total fish production, aquaculture fish production and irrigated
area (Source of data: FAO/PBL). (See color plate.)
TgC all TgC all update GgN from N2O

GgS/10 from SO2 10*kgCH4 GgN/10 from NO2
Update 2010
12000
Emission rate
10000
8000
6000
4000
2000
0
1890 1910 1930 1950 1970 1990 2010
Year
Figure 4.7. Global trends in the emissions of the greenhousegases CO2 , N2 O and CH4 and
the acidifying gases SO2 and NO2 since 1890 (Source of data: PBL).
in 2010. In combination with the rise in other greenhousegases such as methane, this
is thought to pose serious risks of disruptive climatic changes in the course of the
21st century (IPCC 2007). The combustion of fossil fuels is also an important source
of SO2 - and NOx -emissions into the atmosphere and a cause of acidification of lakes
and soils. These emissions have been reduced drastically in the last decades, which
illustrates the effectiveness of technological change in combination with regulatory
policies. The emission profile of ozone-depleting CFCs, which caused depletion of
the stratospheric ozone layer, is another illustration of how effective policy cannot
completely eliminate, but can at least significantly mitigate harmful impacts (Haas
et al. 1993; Miles et al. 2002).
The food system is another important link in this part of the story. Until the
20th century, the provision of nitrogen had been done by natural fertilisation such
as planting clover. In 1908, Haber en Bosch invented an industrial process by which
nitrogen, which constitutes 78.1 percent of ambient air, reacts with hydrogen to form
ammonia (NH3 ). This can be converted into nitrates and nitrites, which became
the feedstock for the large-scale production of industrial fertiliser. In this way, food
output increased and now sustains a larger world population than the British scholar
Malthus had ever contemplated to be possible. However, one of the consequences
is the eutrophication and algae growth in many surface waters all over the globe
(Bouwman et al. 2009). Another element in the food system on the source and sink
side is phosphorus (P). It is an essential ingredient for plant growth and naturally
occurs in the form of phosphates. Phosphates are industrially produced and more
than two-thirds are used in agriculture in fertilisers. Its widespread use is contributing
to eutrophication.8 Finally, biodiversity has been declining in most places – a trend
that is difficult to measure and has complex causes and consequences.
In a system dynamics frame with stocks and flows, the preceding trends are unsus-
tainable if following the definition outlined in Section 2.3. The rate of exploitation
8 It is also a source side issue: Phosphorus reserves are small and highly concentrated, and shortages
could pose a severe limitation to future food production.
Box 4.3. Experiencing landscape changes. The changes in the natural environ-
ment influenced also the way in which it was experienced. Many authors, from
ancient times onwards, have lamented the loss of once beautiful landscapes. Trav-
ellers in the 18th and 19th centuries in the Mediterranean judged the landscape
in many places degraded – or at least a badland as compared to what they had
imagined. Today, there are still alarming reports about degradation and desertific-
ation in the Mediterranean. However, it is a complex region with a long history of
natural and anthropogenic changes, but even so ecological historians like Grove
and Rackham (2001) find the evidence unconvincing and speak about the Theory
of the Ruined Landscape of Lost Eden.
Just as the Greek landscape of 2000 Before Present (BP) was romanticised in
some later paintings and writings, so did some authors romanticise the 19th land-
scape of industrialising Britain. An example is the German Prince von Pückler-
Muskau who was in Yorkshire in 1827: ‘With the falling dusk I reached the big
factory town Leeds. The wide space which it occupies upon and amidst several hills
was covered by a transparent cloud of smoke. A hundred red fires shone up from
it, and as many tower-like chimneys emitting black smoke were arranged amongst
them. Standing out delightfully in the scene were five-storey high huge factory
buildings in which every window was illuminated by two lights, behind which the
industrious worker finds himself engaged until deep in the night. In order to add
a touch of romance to the bustle of enterprise, two old gothic churches arose high
over the houses, on the spires of which the moon poured its golden light while, in
the blue firmament, with the vivid fires of the busy people underneath, it seemed
to repose in majestic rest’ (Sieferle (1997) 165–166). It should be no surprise that
fact, fiction and dream are often mixed in the interpretations of the causes and
consequences of past and present landscape changes.
of fisheries, forests and soils exceeds in many places the natural regeneration rate.
Non-renewable resources like rich metal ores are used at much higher rates than
renewable substitutes are produced, although technology may provide versatility.
But, crucially, fossil fuels are extracted much faster than nature accumulates them
and than present-day society is producing renewable substitutes like wind and solar
power. The assimilative capacity of sinks is exceeded in many places in the sense that
pollutant loads are only partly and slowly broken down at rates below the inflow rates.
This is sometimes obvious, as with large oil spills and aerosols. Sometimes it is known
but next to invisible, as with PCBs and other toxic chemicals. All of these trends in
the use of sources and sinks for human development cannot keep growing beyond
the planetary scale and will inevitably stabilise in the course of this century. An
important question in sustainability science is how such a process might evolve and
what the options are to stay within manageable risk zones (Rockström et al. 2009).
4.3.3 Experiencing Change

To give numbers about the changes is one thing, to understand how the changes were
experienced quite another. Landscape degradation, resource depletion and environ-
mental pollution were observed and discussed already in antiquity. In London, air
pollution from coal burning was already mentioned as a problem in the 13th century.
Impending resource scarcity was another and more important reason for concern.
Medieval Europe was faced with wood scarcity in several places and periods. Early
in the 18th century, the threat of wood shortage in parts of Germany led to the notion
of ‘nachhaltende Nutzung’ (sustainable use), as opposed to the common practise of
clearcutting (van Zon 2002). In the same period, intensified logging forced Japanese
rulers to use a mix of regenerative forestry and wood imports from tropical regions,
and by the end of the century, regulation of consumption and plantation forestry
were introduced (Totman 1989). Also, later concerns about depletion were heard
time and again. Famous is the warning about the exhaustion of Britain’s coal reserves
made in 1865 by the economist Stanley Jevons. Alarms about overfishing were regu-
larly expressed in the first half of the 20th century (Schrijver 2010). Yet, the changes
during the Industrial Revolution were in the main seen as improvements: Nature
was domesticated and made more productive. Wild nature was seen as evil, ruthless
and selfish. The American pioneer was the symbol of entrepreneurial optimism. A
telling statement of Freud made in 1929 is given by Lowenthal (1990): ‘ . . . a country
has attained a high state of civilization when we find . . . everything in it that can
be helpful in exploiting the earth for man’s benefit and in protecting him against
nature. . . . ’
It was in the second half of the 19th century that doubts arose about the accept-
ability of the large, ruthless exploitation of Nature such as the near-extinction of
the American buffaloe and the Siberian fur-bearing animals (Scott Taylor 2011;
Ponting 1991). An influential book was Man and Nature (1864) by the American
diplomat Marsh. He deplored the harmful impacts of human activity, which were
the consequences of greed and ignorance. It was the beginning of the conservationist
movement. Yet, in the vein of Enlightenment and Darwinism, these first environ-
mentalists felt that the changes were largely beneficial and that the forces of Nature
exceeded those of man so much that serious harm was inconceivable. It was also,
and conveniently, believed that technology and state regulation would remedy the
excesses. This attitude prevailed far into the 20th century and much evidence of
environmental damage was ignored. An undercurrent of ecologists developed a dif-
ferent view, rooted in a mixture of classical philosophy and Romanticism, which saw
the equilibrium and stability of the climax ecosystem as the ideal. Nature provides
a moral order and is imbued with balance, integrity, order, harmony and diversity.
Advanced industrial man degrades Nature and its beneficence.
Pollution of air and water became a genuine issue when the first signs of negative
impacts on human health showed up. But until late into the 20th century, action was
usually only taken when the harm or damage was local and evident, as with smog
or stench, and when the affected upper-class citizens could mobilise political sup-
port. This changed gradually with the rise in the standard of living. In the 1960s, the
upcoming middle class in the United States and Europe expressed growing concern
that pollution had negative health impacts. Also, rising noise levels, reduced visib-
ility and nasty smells were increasingly considered unacceptable. Still, opposition
and subsequent (calls for) action were largely local, where people could link the
experience to the cause.
While the risks of a local and short-term nature were increasingly recognised and,
though with a delay, addressed, the large-scale long-term risks remained unnoticed
and ignored. It was only with reports such as the Limits to Growth to the Club of
Box 4.4. Population overshoot and collapse. How fragile the situation was in
medieval times is illustrated in the historical reconstruction of population num-
bers in medieval France between the year 1000 CE and 1800 CE (Figure 4.8).
The graph shows France’s population and average Northern Hemisphere tem-
perature relative to the period 1961–1990 (Mazoyer and Roudart 1997). Detailed
analyses of agricultural practices and techniques reveal an increase in potential
population density since the Carolingian era. A farmer in Mediterranean Europe
needed in 900 CE about 16 hectares (ha) of which two-third were pasture to feed
a five-person family (Mazoyer and Roudart 1997). This could sustain a popula-
tion of 20–30 persons/km2 . In Mid- and Northern Europe, these numbers were
34 ha and 8–15 persons/km2 , respectively. This gradually rose to 30–80
persons/km2 around 1250 CE.
In the 14th century, a series of famines and wars set in, possibly related to rather
sudden drops in temperature, but there was a remarkable recovery. The fall in
population liberated large amounts of productive land and better fertilisation
strategies and tools, such as the use of nitrogen-fixing clover and the horse-drawn
plough, led to major yields on especially the northern soils. The population quickly
recovered to a new plateau with a density of up to 160 persons/km2 . This period
in European history is a dramatic illustration of a cycle of expansion and decline
where both natural and social forces are simultaneously at work.
Rome (Meadows et al. 1971) and the Blueprint for Survival from The Ecologist (1972)
journal that local and regional resource depletion and environmental degradation
were framed in a larger, global context and considered outside the circles of scientific
experts and environmental activists. It introduced a more systemic analysis and
emphasised the linkages between the various causes and consequences. One of the
challenges of sustainability science is to merge these and subsequent scientific insights
about the global and long-term prospects with the local potential and willingness to
act in accordance with these insights.
4.4 Earth System Analysis: Regimes and Syndromes
4.4.1 Social-Ecological Regimes

The previous historical accounts provide a necessary background for an interpret-
ation of the present situation in terms of (un)sustainability.9 But is it possible to
go beyond the anecdotical and the case studies to distinguish some larger, generic
patterns? One way to look at human history is as a series of transitions from one
social-ecological regime to the next (Fischer-Kowalski and Haberl 2007). A transition
is a dynamic process in which a system changes from one relatively stable equilib-
rium to another (§8.6). After a take-off, important system variables first accelerate
and then decelerate to a new, relatively stable state. I have already introduced the
9 There are a vast array of books on this topic, and we refer to the suggested literature for more
in-depth historical accounts.
mln people
20
food shortages
famines
pests
10 wars
0,6oC below 1000-1998 average Northern

Hemisphere temperature (IPCCTAR WGI:29 2001)
Revolution Crise Reconstruction Rebondissements

agricole de lo crise
Annee
600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800
Figure 4.8. Population of France between 600 CE and 1800 CE. The population data are
based on Mazoyer and Roudart (1997). The dotted curve indicates the average Northern
Hemisphere temperature relative to the period 1961–1990 (IPCC 2001).
demographic transition; however, the focus is now on a transition of one social-

ecological regime, that is, a particular pattern of society-nature interactions that is
sustained in a dynamic equilibrium for a certain period, to the next (Schandl et al.
2009). It comprises societal (culture) as well as biophysical (nature) elements.
The transitions from one regime to another are, in a world history context, pro-
cesses of deep structural change with revolutionary consequences. The old regime
is more or less absorbed and transformed into the new regime. Each new regime
brings an expansion of the anthroposphere within the biosphere. The first regime is
commonly associated with the domestication of fire: People transformed a destruct-
ive natural force into a regularly available source of energy, with important con-
sequences for culture and nature (Goudsblom 1992). The second great transform-
ation brought about by humans was the domestication of plants and animals: the
process of agrarianisation (§3.1). The resulting agriarian regime became the domin-
ant one during the Holocene period in almost all regions of the world.
In agrarian societies, unlike in hunter-gatherer groups, people actively and delib-
erately intervene in the ways in which natural process convert sunlight into plant and
animal biomass. They use the land to convert forest into agro-ecosystems in order to
get food. The metabolism is completely dependent on incoming solar energy, at an
average density of 100 W/m2 and with strong fluctuations. There is always a delicate
balance between food requirements for a growing population and food supply in a
finite environment. Pre-industrial agrarian societies were thus in dynamic equilib-
rium, with high birth and mortality rates and life expectancy below forty years, on
average. Improvements in yields and the capacity to feed people happened slowly
and discontinuously and human populations were still dependent on the vagaries
of the natural environment. In particular, climate change may have triggered or
exacerbated existing vulnerabilities and risks and indirectly caused famines, epi-
demics and wars (Figure 4.8).
Around 1700, there were the first clear signs of a new social-ecological regime.
It has been documented quite well for the United Kingdom of Great Britain and
Northern Ireland (UK) (Schandl and Krausmann 2007). Crude death rates started
to go down and life expectancy went up. To feed the growing population, agriculture
had to adapt. Cropland and intensely used grassland expanded. The feudal insti-
tutional framework was restructured into a market-oriented system of landlords,
tenants and agricultural labourers. This led to greatly improved yields because of
increase in the size of land holdings and of innovations in strategies and tools, and
until 1830, per capita food output could be sustained. Nevertheless, population star-
ted to outgrow food production capacity and shortages were relieved by importing
food from the ‘New World’ and Russia. Around 1900, Britain imported more than
60 percent of its food, which was the equivalent of an area the size of the UK cropland
area.
The expansion of the nutritional basis was one element of the transition to
the new regime and the first escape from the solar-based agrarian society. Because
not only land productivity but also labour productivity increased, there was a large
surplus population in rural areas. Their migration to urban centres provided the
labour force for industry. The appearance of transport networks (canals, trains, and
so on) and the energy to operate them were an essential precondition. This, in turn,
evolved on the basis of the science-technology-resource complex of coal mining,
iron production and steam engines.10 Primary energy use per person, in the range of
20–70 GJ/cap/yr and more than half biomass-based, had been more or less constant
for a long period, but in the industrial era, it started to rise rapidly towards a new
plateau with values in the range of up to 150 GJ/cap/yr by 1900.
The further liberation from the constraints of the agrarian regime and a new
stage in the industrialisation process was the intrusion of oil and electricity into all
economic and social activities. It was made possible through the internal combustion
engine and the electric motor. In combination with a manifold of other inventions
and innovations, not only in the United Kingdom but all over Europe and the United
States, it had profound impacts on individual and societal life. Agriculture became
decoupled from the solar flow constraints with the advent of road and electric power
networks, the industrial production of fertiliser and the replacement of animals
by tractors. It permitted a threefold increase in yields since 1950. The decentralised
nature of the new energy technologies accelerated also in industry the replacement of
physical labour by machines. Average working hours started to decline and human
labour was increasingly engaged in operational skills and information processing
(Table 4.1).11 Urbanisation rates started to rise. At the end of this process, primary
energy use per capita had made another jump, to 120–200 GJ/cap/yr or three- to
fourfold the level of agrarian societies. In fact it increased much more because the
10 It can be argued that these inventions have been induced by the wood scarcity in England – a widely
used example to argue that scarcity problems generate their own solution.
11 There are rather large discrepancies in the data on working hours per year between the data of
Maddison and other sources such as the ggcd-website and other authors who use different sources
and probably definitions.
a)
1000
Netherlands Sweden
Italy Spain
GJ/cap/yr
100
10
1800 1850 1900 1950 2000
Figure 4.9a. Graphs of primary energy use per person in GJ/cap/yr since 1800 for four
European countries: Netherlands, Sweden, Italy and Spain (Gales et al. 2007).
energy was used more efficiently (§7.3). In useful energy units, the increase was more
than twentyfold!
The regime change can be seen quite well in the reconstructions by historians
of the use of energy carriers in European countries. Defining food and fodder-
based muscle power of men and working animals, firewood, wind and water and
peat as traditional energy carriers, most European countries experienced a trans-
ition to less than 50 percent of primary energy carriers in the form of traditional
ones between 1850 and 1950 (Gales et al. 2007). The relative shares of traditional
energy carriers reflected local circumstances. While in Sweden, for instance, firewood
dominated, muscle power prevailed in The Netherlands and Spain. After 1950, an
acceleration of the transition to fossil fuels took place. The graphs of primary energy
use per person per year in four European countries show this transition from an
agrarian level of 20–50 GJ/cap/yr towards an industrial level of 100–200 GJ/cap/yr
(Figure 4.9a).
A widely used formula to decompose primary energy use in a country is to rewrite
energy use as the product of population, income (GDP/cap/yr) and energy-intensity
(E/GDP):
GDP E
E =P· · GJ/yr (4.1)
P GDP
Figure 4.9b shows per capita energy use (E/P) and energy-intensity (E/GDP) for four
European countries between 1800 and 2000. There is a clear trend towards lower
energy-intensity in these four European countries, which reflects primarily the net
effect of improvements in the efficiency with which primary energy carriers are used
in economic production.
The high-income regions entered a new, fourth social-ecological regime in the
second half of the 20th century. It is exemplified by the large share of the services
sector (Figure 4.3). In Northwestern Europe, the United States and Japan, the
b)
300
GJ/$ of economic output (GDP) Netherlands Sweden
Italy Spain
30
3
1800 1850 1900 1950 2000
Figure 4.9b. Graphs of energy-intensity in GJ/$ since 1800 for four European countries: Neth-
erlands, Sweden, Italy and Spain (Gales et al. 2007).
service sector output consisted around the year 2000 for more than one-third of
public services and for more than one-third of financial services, the remainder
being wholesale and retail trade, hotels and restaurants, transports and posts, and
communication (Teives Henriques and Kander 2010).12 In other words, three out of
five jobs are about dealing with people, symbols and information. It is in line with
a cultural change towards postindustrial self-expression values, which emphasise
personal autonomy and individual choice and are conducive to more participatory
forms of democracy (§4.1). Once basic needs such as food and shelter are met, the
demand for mass-produced goods such as household and transport equipment start
to dominate. When this becomes saturated, the richer segments of society spend
an increasing part of their income on private and public services.Yet, the long-term
significance of this transition is still difficult to interpret.
Evidently, the rapid development and spread of ICT-based applications through-
out society, with the Internet as its for now supreme representation, is a key ingredi-
ent in the transition to the postindustrial world (Figure 2.7b). Measured as useful
energy per unit of economic output, there are signs of a clear decoupling between
energy use and economic output in the United States since 1970. This may be inter-
preted as the turning point at which information became more important than phys-
ical work as a factor of production (Ayres and Warr 2005). Some other indicators
also suggest a decoupling of monetary and physical flows.
Another noteworthy phenomenon is the productivity gap between manufactur-
ing and services sectors, as observed first by Baumol (1967). Those parts of the
service sector where the human element is crucial – such as entertainment, edu-
cational and personal care services – cannot realise the high rates of productivity
12 Data are for 2005 and based on constant prices.
increase that happen in manufacturing.13 This makes economic growth in the postin-
dustrial economy different from growth in the industrial era and is one explanation
of the decline in economic growth rates with rising income. It is counteracted by
replacing expensive services with do-it-yourself activities, mechanisation and robot-
isation. These mutually interacting and reinforcing changes influence the modern
world in ways that are hard to interpret and predict.
The data on sectoral changes in emerging economies such as China and India
have to be interpreted with caution, because the informal economy, that is, the trans-
actions outside of organised labour unions and tax collectors, is large and growing.
It consists of traditional service sector activities and may transform itself into a mod-
ern service sector without fully becoming part of official statistics. It is an important
buffer against economic ups and downs for the millions of people with insuffient
or inadequate skills in countries with weak governance and an increasingly global
labour market.
Evidently, the contours of the new regime are not yet clearly visible, and the
demarcation between industrial and service jobs and activities is rather arbitrary and
vague. Are we witnessing the emergence of the noösphere, an expanding network
of human consciousness and mental activity as the next stage in the evolution of the
biosphere? Is the fulfillment of human desires gradually disconnected from material
stocks and flows? Are we experiencing the first signs of a sustainability transition?
Hopefully, you know at the end of this book what such a transition could be like and
what your role can be.
4.4.2 Syndromes and Archetypes14

A fragmented, disciplinary approach does not work in sustainability science. A
broader perspective transcending disciplinary boundaries is needed. One framework
to investigate global change phenomena in such a broader, problem-oriented setting
is the syndrome approach.15 The word syndrome comes from ‘running together’,
in particular, the simultaneous occurrence of signs of disease. Here, it refers to a
typical co-occurrence of different complex natural and social dynamic phenomena
or symptoms. They are causally connected and constitute the elements of a syn-
drome. The basic assumption of the syndrome approach is the existence of clusters
of interrelated symptoms.
Examples of social symptoms are positive trends in individualisation and inform-
atisation, in mobility of people and transport of goods and information, and in
rural-urban migration and urban sprawl. Natural symptoms can be source-based,
such as trends in use and depletion of energy and other resources and in resource
use productivity. Sink-oriented natural symptoms are changes in biodiversity and
in accumulation of waste and pollutants in soils, water and air. Several graphs
13 In fact, the notion of (labour) productivity as output per unit of (labour) input is a flawed one,
because it homogenises widely divergent ‘outputs’ on the basis of a divergent ‘input’, namely, wages.
14 I would like to acknowledge the contributions of Marcel Kok and Paul Lucas, both at PBL, for this
paragraph.
15 The syndrome approach was originally proposed by the German Advisory Council on Global Change
(WBGU 1994) and later conceptualised and developed at the Potsdam Institute of Climate Impact
Research (PIK) (Schellnhuber et al. 1997; Petschel-Held et al. 1999).
Table 4.3. Overview of the global change syndromesa
Syndrome name Short description of the mechanism
Utilisation syndromes
Sahel Syndrome Overcultivation of marginal land
Overexploitation Syndrome Overexploitation of natural ecosystems
Rural Exodus Syndrome Environmental degradation through abandonment of
traditional agricultural practices
Dust Bowl Syndrome Non-sustainable agro-industrial use of soils and water
Katanga Syndrome Environmental degradation through depletion of
non-renewable resources
Mass Tourism Syndrome Development and destruction of nature for recreational ends
Scorched Earth Syndrome Environmental destruction through war and military action
Development syndromes
Aral Sea Syndrome Environmental damage of natural landscapes as a result of
large-scale projects
Green Revolution Syndrome Environmental degradation through the adoption of
inappropriate farming methods
Asian Tiger Syndrome Disregard for environmental standards in the context of rapid
economic growth
Favela Syndrome Environmental degradation through uncontrolled urban
growth
Urban Sprawl Syndrome Destruction of landscapes through planned expansion of
urban infrastructures
Disaster Syndrome Singular anthropogenic environmental disasters with
long-term impacts
Sink syndromes
High Stack Syndrome Environmental degradation through large-scale dispersion of
emissions
Waste Dumping Syndrome Environmental degradation through controlled and
uncontrolled waste disposal
Contaminated Land Syndrome Local contamination of environmental assets at industrial
locations
a Lüdeke et al. 2004.
presented earlier in this chapter are representative of certain symptoms. A region

may be vulnerable to a syndrome when it is exposed to a particular, causally con-
nected set of symptoms. Its disposition signifies the probability that the syndrome
occurs. It depends on the region’s exposure and sensitivity whether and how the syn-
drome affects the region. Exposure depends on exposition factors. These are rather
short-term events that may activate a syndrome if the disposition is high. They can
arise from within the system (endogenous) or from the outside (exogenous) as with
natural catastrophes and extreme political or economic shocks. The sensitivity of
a region for the event and intensity of a syndrome is a complex pattern of agro-
ecological, economic and socio-cultural factors. The final impact depends also on
the capacity of people in the region to respond. Such a coping capacity or capability
can prevent or mitigate the consequences of a syndrome.
Table 4.3 shows a list of situations that qualify as syndromes. Based on the dom-
inant aspect of the human-nature interaction under consideration, a distinction can
be made between utilisation syndromes, development syndromes and sink syndromes.
Syndromes are often occurring in interaction, for instance, when an operating
syndrome triggers or reinforces another syndrome. In later chapters, several stories

describe some of the listed syndromes in more detail.
Building upon the syndrome approach, a similar framework has been developed
under the heading of archetypical patterns of vulnerability. It was developed as part
of the 4th Global Environmental Outlook (GEO) in response to a request from
governments to show how the environment provides challenges and opportunities
for development (UNEP 2007). While GEO is a global assessment, its strong regional
focus bridges the gap between a coarse global overview, on the one hand, and insights
from local case studies of sufficient relevance for countries and regions, on the
other.
The concepts of vulnerability and risk play a key role. Similar to syndromes, vul-
nerability is considered a combination of exposures, sensitivities and adaptive capa-
cities in a social-ecological system. Examination of local and regional situations and
trends shows that there are certain mechanisms in human-environment systems that
create specific, representative patterns of the interactions between environmental
change and human well-being (Kok and Jäger 2007). These patterns of vulnerabil-
ity tend to have characteristic space- and time-scales at which human development
takes place and is confronted with constraints. They can be regarded as archetypical
or generic, because they describe the common element in case studies from widely
different regions of the world. The details may be quite different, but their essence
is the same.
An archetypical pattern of vulnerability has two components. One is the vul-
nerability creating mechanisms, and the other is the spatial distribution of typical
combinations of the mechanisms. Their investigation in GEO-4 is guided by five
questions (Kok et al. 2011):
r What are the main exposures, key vulnerable groups and their sensitivities that
together define the pattern of vulnerability?
r What are the basic vulnerability creating mechanisms that constitute this pattern
of vulnerability?
r In what form and where does this pattern manifest itself?
r How can future changes within the human-environment system affect the human
well-being situation for the vulnerable groups?
r What are the opportunities – individual responses or policy responses – to cope
with and adapt to future changes?
The first two questions address a description and formalisation of a specific pattern
under investigation. For instance, income, population pressure, access to markets,
soil quality and water availability are identified amongst others as determinants
of food security – determinants meaning rather complex elements or subsystems.
Answering the third question includes quantification and statistical analysis based
on the information from the answers to the previous two questions. It generates a
set of proxy indicators for the most important determinants of vulnerability creating
mechanisms. For instance, population pressure is measured as population density
and soil quality as an index for water erosion sensitivity. The next step is a cluster
analysis, that results in one functional and one spatial characterisation of a pattern
of vulnerability. The functional one consists of a specific constellation of indicators
that are labelled vulnerability profiles, while the spatial one contains the spatial
Figure 4.10. Condensed influence diagram of the pattern of vulnerability of smallholder farm-
ers in drylands. The boxes indicate important vulnerability determinants. The arrows show
hypothesised relationships and their direction of influence. The thin lines represent the indic-
ators that are used as proxy quantification of the determinants (redrawn from Kok et al.
2011).
distribution of the profiles. The profiles and the differences and similarities between
the profiles give insight into the factors that create the vulnerability in a specific
cluster. The resulting vulnerability patterns provide an answer to the fourth and fifth
question, as an entry point for identifying opportunities to reduce vulnerability and
directions for policymaking.
An illustration is given for the vulnerability of small-holder farmers in dry-
land ecosystems, where humans have to sustain food production in an environment
of erratic rainfall.16 The hypothesised relationships between determinants of food
security are represented in an influence diagram, with a focus on food production,
soil and water resources, and income (Figure 4.10). For those in the dashed boxes,
quantitative indicators have been collected at the grid-cell level: water availability,
soil fertility and degradation, population and infrastructure density, and income.
The cluster analysis is used to identify correlations in indicator space. Six of such
vulnerability profiles are represented in Figure 4.11. Each profile is associated with
grid-cells in space, and these are shown for each of the eight profiles in Figure 4.12.
It is seen, for instance, that the resource poor regions with people in severe poverty –
and characterised by high infant mortality – are largely located in a small band south
of the Sahara Desert. A subsequent, more rigorous categorisation of vulnerability
patterns in drylands indicates that it is indeed possible but not simple to identify
generic patterns of potentially unsustainable developments in fragile regions (Sietz
et al. 2011).
Why is it useful to identify and study such archetypical patterns? In many places
on Earth, people feel powerless in the face of forces that threaten their quality of
16 Another worked out case are forest ecosystems, which exemplify the conflicting claims on land:
exploitation for human use in agriculture and for (fire)wood versus conservation to prevent loss of
biodiversity.
Figure 4.11. Six typical vulnerability profiles in drylands worldwide as arising from the cluster
analysis. The indicators are along the x-axis. The indicator values of the respective cluster
centres, from low to high and normalised (min/max), are along the y-axis (Kok et al. 2011).
Figure 4.12. Distribution of the eight vulnerability profiles in drylands worldwide according
to the vulnerability profiles in Figure 4.11 (Kok et al. 2011). (See color plate.)
life and appear to be inevitable. Many assessments are done all over the world to
understand the causes of deteriorating quality of life and unsustainable development
trends. Unfortunately, the insights gained and lessons learned often remain local or
regional in their impact, due to the specifics of the situation. If archetypical patterns
can be identified, such local and regional assessments can provide an analogue and a
remedy for other places in the world. People in one situation may then learn how to
respond, adapt and heal from insights about what happened elsewhere. It provides
generic insights in the dynamics that create vulnerable situations for people and
can be used as a basis for further analysis of opportunities and responses to reduce
vulnerability.
4.5 Summary Points

In the past three centuries, an exponential growth has taken place in the number of
human beings and their activities. The historical background to the changes since
1700 are the advent of industrial capitalism, the scientific and technological inven-
tions and the transition to fossil fuels. Industrialisation started in Europe and its
offshoots. Population growth rates started to slow down in the 20th century, as part
of the demographic transition. Economic growth has been exponential throughout
the period, with worldwide increases in income and a change from agricultural to
manufacturing and later to service sector activities. In the second half of the 20th
century, industrialism went global and expectations of hundreds of millions of people
soared. There are still large income inequalities within and between countries, which
is one of the threats to a sustainable future. There are a some other key points to
remember:
r The changes took (and take) place in complex interaction with less easily quanti-
fied social-cultural changes. The nation-state got new roles, ranging from colonial
rule to welfare provision. An extensive global trade network evolved. Values
changed during the process of industrialisation, notably from traditional and
survival-oriented to secular and self-expressive. The dominant ideas and values
of modernity and the politics and organisation of industrial capitalism mani-
fest themselves in the rise of science and technology and empirical rationalism,
in innovations and markets as engines of economic growth, and in democratic
institutions.
r Industrialisation led to massive and accelerating exploitation of natural sources
and sinks, with unintended and unanticipated consequences. Past and anticip-
ated depletion and degradation of natural resources is another, large threat to a
sustainable future. The ‘Western’ development model itself may turn out to be
unsustainable.
r As individuals, we need to interpret the situation in a long-term global perspect-
ive while appreciating its momentary acuteness and local diversity. Frameworks
such as the transition to a new social-ecological regime and archetypical pat-
terns of (un)sustainable development can provide the broader perspective that
is necessary, if the transition we are in is to fulfill the promise of a good quality
of life for all while staying within the planet’s biophysical boundaries.
. . . until recently, ecological regimes have been losing ground vis-à-vis . . . two
regimes in particular: the money regime and the time regime. Both have, each
in their own way, turned people’s attention away from the natural environment, and
from ecological issues, toward more purely social aspects of the anthroposphere.
– Goudsblom, 2002: 370
Feed the poor

tell the truth
make water-places
for the thirsty
and build tanks for a town –
you may then go to heaven
after death, but you’ll get nowhere
near the truth of our lord.
And the man who knows Our Lord,
he gets no results.
From: Speaking of Siva (Vs. 959).
Ed. A.K. Ramanujan. Penguin Classics 1973
“La biosphère est tout autant (sinon davantage) la création du Soleil que la manifest-
ation de processus terrestres. Les intuitions religieuses antiques de l’humanité qui
considéraient les créatures terrestres, en particulier les hommes, comme des enfants
du soleil étaient bien plus proches de la vérité que ne le pensent ceux qui voient
seulement dans les êtres terrestres la création éphémères . . . ”
– Vernadsky, La Biosphère, 1928:51
SUGGESTED READING
Overview of the IHOPE Dahlem conference with accounts of past civilisation developments.
Costanza, R., L., Graumlich, W. Steffen, C., Crumley, J. Dearing, K. Hibbard, R. Leemans,
C. Redman, and D. Schimel. Sustainability or collapse: What can we learn from integrating
the history of humans and the rest of nature? Ambio 36 (2007) 522–527.
Models, maps and myths about the role of the natural environment in ancient civilisations.
de Vries, B. de, and J. Goudsblom, eds. Mappae Mundi – Humans and Their Habitats
in a Long-Term Socio-ecological Perspective. Amsterdam: Amsterdam University Press,
2002.
Economic history of the Industrial Revolution.
Heilbroner, R., and W. Milberg. The Making of Economic Society, 12th ed. New York: Pearson
Prentice-Hall, 2005.
A natural science overview of Global Change phenomena during the last century.
IGBP Science 4 2001 www.igbp.net/documents/IGBP ExecSummary.pdf.
One of the best documented quantitative studies on long-term trends in population and economy,
with many data.
Maddison, A. The World Economy – A Millennial Perspective. Paris: OECD, 2001.
Although a bit outdated, this is still a landmark study about Global Change and the human-
induced transformations.
Turner, B. et al. (Eds.) The Earth as transformed by human action – Global and regional
changes in the biosphere over the past 300 years. Cambridge: Cambridge University Press,
1990.
USEFUL WEBSITES
Historical and Country Statistics

r www.pbl.nl/en/themasites/hyde/index.html is a site under the authority of the Nether-
lands Environmental Assessment Agency (PBL), which presents the HYDE (History
Database of the Global Environment), that is, (gridded) historical time series of popula-
tion and land use.
r www.gapminder.org is a site with large amounts of data on the world and the countries
of the world, with a nice and flexible interface.
r gcmd.gsfc.nasa.gov/ is a Global Change Master Directory with info, data and links.
r nasadaacs.eos.nasa.gov/index.html is the Earth System Science Data and Services site,
operated by NASA.
r www.ggdc.net/databases/index.htm is a database at the Groningen Growth and Devel-
opment Center of the University of Groningen, with extensive data on economic history
and performance.
r www.conference-board.org/data/economydatabase/index.cfm is a site of the Conference
Board Total Economy Database.
r www.ilo.org is the site of the International Labour Organization (ILO), with data and
information on global labour and employment related topics.
r www-histecon.kings.cam.ac.uk/history-sust/energy.htm is the site of History and Sustain-
ability, with an overview of available historical datasets on energy use and other variables.
Reports and Analyses

r www.unep.org/publications/ is the site of UNEP, with a.o. the four Global Environmental
Outlook (GEO) reports.
r www.igbp.net is the website of the IGBP with a.o. the Executive Summary of Global
Change and the Earth System: A Planet Under Pressure.
r www.iiasa.ac.at is a site on various global change issues (population, land use, energy) at
the International Institute for Applied Systems Analysis (IIASA).
r www.wto.org is the site of the World Trade Organization (WTO) with trade statistics,
amongst others.
r www.gatt.org/ is a site where the WTO and the economic views it defends are criticised.
r www.millenniumassessment.org/en/index.aspx is a guide to the various reports published
as part of the Millennium Ecosystem Assessment.
r www.ipcc.ch gives access to all reports of the Intergovernmental Panel on Climate Change
(IPCC).
5 Sustainability: Concerns, Definitions,
Indicators
5.1 Introduction
Historical accounts are given from a particular vantage point. Ours is the emerging
field of Sustainable Development research in a Global Change context (Turner et al.
1990; Costanza et al. 2007). It explores human development and the planetary bound-
aries within which it unfolds (Rockström et al. 2009). The dominant frame is given by
the natural sciences and their concepts, methods and theories. Their findings repres-
ent the prevailing outlook on the world in Western, and increasingly other, societies
and are at the core of university curricula associated with Modernism. It is, therefore,
instructive to briefly introduce the history of the ideas behind this worldview and its
essential tenets. How does modern man see earth, life on earth and man himself after
these 300 years of scientific advances and economic growth? Such a reflection also
makes apparent the limitations of this worldview. Clearly, science and technology
have liberated us to a large extent from the controls of Nature. There is no need
anymore for Gods who have to be begged for help with prayers, virtues and sac-
rifices. At the same time, there is growing discontent with the scientific, modernist
outlook. A primarily natural science interpretation of problems of development and
sustainability and their solutions meets with opposition and critique. Chapters 6, 8
and 10 explore these critiques in more depth.
Against this background, the emergence of environmental concerns and the
heightened interest in the notion of (un)sustainable development has to be under-
stood. The second part of this chapter briefly recounts the pedigree of the concept
and emphasises the historical context and changes. I abstain from a rigid defini-
tion and instead, in the last section, focus on indicators of (un)sustainability and
(un)sustainable development that are proposed in order to make the concept more
operational.
5.2 Global Change: The Scientific Worldview
5.2.1 The Scientific Worldview: Earth

The history of the Earth is one long process of change. It has become popular know-
ledge that Earth dynamics are measured in geological time-scales of hundreds of
117
118 Sustainability: Concerns, Definitions, Indicators
millions of years. Every year new and better reconstructions of all sorts of ‘physiolo-
gical variables’ in the earth’s system and its subspheres (atmosphere, lithosphere,
cryosphere, hydrosphere) are becoming available, such as atmospheric concentra-
tion of carbon dioxide, temperature and sea level over the last hundreds of millions
of years.1 Some processes are so slow that humans hardly notice. Others move very
fast, even when measured in terms of human time-scales (Table 2.3).
This scientific view of our planet is rather recent. In 1654, the Irish prelate
Ussher declared that, according to extensive research, creation had taken place on
26 October in 4004 BC at 9 AM. But at the end of the 18th century, the Frenchman
Buffon widened the perspective by his statement that the seven days of creation
were actually seven eras of 74,832 years in total (Boorstin 1985). In his influential
book Principles of Geology, first published in the 1830s, the Scottish geologist Lyell
argued that the earth around us can be understood as the outcome of small changes
of geological processes at work during very long periods of time. Whereas Newton’s
timeless laws of mechanics and Linnaeus’ elaborate classification of plant and animal
species were still consistent with traditional belief in biblical cosmology, these new
discoveries and theories made such a reconciliation increasingly difficult.
Among the many discoveries and theories in the earth sciences, two 20th-century
theories stand out. Wegener was, with his book Die Entstehung der Kontinente und
Ozeane (1915), the first to propose in some detail the hypothesis that the contin-
ents were moving across the earth crust (‘continental drift’). Initially dismissed on
physicochemical grounds, this radical idea had developed by the 1960s into the
theory of plate tectonics and is considered to be one of the important conceptual
revolutions of the 20th century. It led to a whole new understanding of geological
observations. For instance, what is known today as the continent of Europe has,
during the last 300 million years, not only been subjected to strong temperature
and sea-level fluctuations but also drifted thousands of kilometres across the earth’s
crust.
Milankovitch published his Kanon der Erdbestrahlung und seine Anwendung
auf das Eiszeitenproblem in 1941. In this book, he attributed – although he was not
the first one to do so – ice ages to changes in the angle between the Earth’s axis
and its orbit around the Sun and the shape of the Earth’s orbit around the Sun. At
first, meteorologists banished this idea and some even insinuated that Milankovitch
had confused astrology with astronomy, which is considered a fatal reproach in the
physical sciences. However, by the end of the 1960s, the results of deep-sea sediments
and theoretical models of celestial mechanics and climate confirmed that these so-
called Milankovitch-cycles had been one of the important causes of the change in ice
cover. The Earth temperature has continued to fluctuate in the last couple of million
years within a range of about 10◦ C.
5.2.2 The Scientific Worldview: Life

Ideas about life in ancient civilisations and in medieval times were embedded in
religion, but they also reflected the practical insights from farmers and herders and
from physicians and medical doctors. The developments in European science in the
1 See the list of suggested websites for such time-series.
last few centuries have also revolutionised ideas about life. Natural history emerged
as a separate branch of science in the 18th and 19th centuries, with the observations
of the micro-world with the help of microscopes, the durable taxonomy of species
by Linnaeus (1735) and the accurate observations framed in a romantic-holistic
Weltbeschreibung by Von Humboldt (1814).
But ideas about life were never the same after 1859 when Darwin published
his theory of evolution in the book, On the Origin of Species by Means of Natural
Selection, or the Preservation of Favoured Races in the Struggle for Life. On the
basis of his own, and other people’s observations and speculations, Darwin argued
that populations adapt to (changes in) the natural environment through a hereditary
process in which more apt individuals have higher survival rates. Darwin’s investig-
ations were the beginning of modern biology and the idea of natural selection is still
the essence of modern evolutionary biology. It is now scientific canon that in the
past millions of years species have come and gone – the debate is about whether the
causes behind the changes were evolutionary (‘survival of the fittest’), revolutionary
(natural catastrophes) or both.
The investigations of how species interact with their environment and with each
other led to the science of ecology. It unravelled how plants and animals form
patterns of mutual competitiveness, cooperation and parasitism, which give rise to
complex behaviour of populations over time. The 20th century gave rise to a whole
series of discoveries about the mechanisms behind evolutionary processes, both at
the macro- and micro-level. Building upon the experiments by the Austrian monk
Mendel in the 1850s, the crucial role of information (coding) and its transmission in
evolutionary processes became understood. The science of genetics was born.
The life sciences have demonstrated convincingly that change and interdepend-
ence are everywhere. The idea of changeless creation has slowly but irresistibly
given way to the concept of dynamic life history. With the centrality of Earth gone in
the Copernican revolution, the same happened now to the human species: Human
beings are no longer divine creations at the apex of the world but descendants of
primitive life.2 These changing views increasingly anchor life in the physical sciences,
uncovering the material roots of life and the biological roots of human life. Philo-
sophers such as Bergson tried to stem the tide in the early 20th century with the
concept of élan vital (vital force), but the fence between organic and non-organic,
life and non-life had been pulled down forever. In the 1960s, the chemist Lovelock
proposed that organisms not only had adapted to their environment, but that life
had created itself the physicochemical environment in which it could flourish (the
Gaia hypothesis).3 It brought down the fence between geology and biology. With
the advent of complex system science and the cognitive and artificial life sciences,
another fence is eroding: the one between matter and mind. For some, it reveals the
unity of science. For others, it is the ultimate disenchantment of the world, to use
Weber’s expression.
2 At the same time, most people in modern societies will consider human beings to be at the apex of
earth history in terms of complexity and consciousness.
3 One finding that affirms the Gaia hypothesis is that the oldest bacteria used free energy flows to
build structures from hydrogen and carbon, and in this way brought oxygen into the atmosphere.
The Gaia hypothesis appeared in a more esoteric and holistic gown as Mother Earth – and was
vehemently rejected by many scientists (Schneider and Boston 1991).
20000 BCE 10000 BCE
lower estimate
higher estimate
millions
7000
6000
5000
4000
3000
2000
1000
inh. / km2
0 0
1
2000 BCE 0 2000 CE 2
4
8
16
Background: 1990 situation
Figure 5.1. The size of the human population during the last four millennia, with an uncer-
tainty margin. For illustrative purpose, the estimated temperature profile during the Holocene
is indicated (Feynman and Uzmaikin 2007).
These changing ideas had slow but thorough implications for society. With the
advent of geology, earth history was no longer the domain of theologians. With
the theory of evolution, their domain further eroded and adherence to the scientific
worldview became a sign of enlightenment. People extended their range of action and
information beyond their own part of the world. Universalism gained an advantage
over ethnocentrism. Spirits and gods no longer played a prominent role. Obser-
vations and the search for mechanisms came in its place and the ‘mechanisation’
and ‘mathematisation’ of the worldview began. In the 1960s, the full implications of
the scientific paradigm became clear, with books like Chance and Necessity (1970)
by Monod and The Selfish Gene (1976) by Dawkins. Witnessing the debates about
Creationism and Intelligent Design, this change in orientation is still going on.
5.2.3 The Scientific Worldview: Society

Homo sapiens evolved from earlier homines and experienced a slow but persistent
growth in numbers during most of the Holocene, which was a period with an excep-
tionally stable climate (Figure 5.1; Feynman and Ruzmaikin 2007). Only in the last
few centuries, the human population started to grow exponentially in numbers and
has become so successful that it is an identifiable force in global change: welcome
in the anthropocene and the anthroposphere. Countless books have been written
on what is the newness of homo sapiens, the wise and knowledgeable man. Is it his
language, his religion, his tools – as homo faber or the working man? Or is it his
playfulness – as homo ludens or the playing man? Or is it his consciousness, or even
more: his self-consciousness, that makes the difference? Obviously, the (self) image
of man has changed immensely in response to the scientific investigations of the last
centuries.
The study of individual and societal dynamics has not yet delivered laws in
the way they are understood in the natural sciences. The quantitative is seldom
reconciled with the qualitative and formal models are at best a complement to the
case studies and anecdotes that prevail in the social sciences. If one can speak of
empirical laws based on observations, they are probabilistic and correlative in nature.
There are theories around which schools form. They adopt the name of the
originator or most ardent protagonist and the essence is communicated through the
‘classical works’ of its greatest thinkers. In Western science, these are associated
with the names of Freud, Adler, Jung and Skinner in psychology and of Comte,
Durkheim and Weber in sociology. In economics, the separate schools of thought
are usually indicated with the names of their proponents: Malthusianism, Walrasi-
anism, Keynesianism, and Schumpeterianism. Or there is a historical distinction,
as between the classical and the neo-classical theory. Formation of theories and
schools is a normal and fruitful part of the scientific process. However, the differ-
ence is that in the natural sciences one theory survives albeit in adapted form after
a period of contest and competition on the basis of evidence and the others dis-
appear into oblivion. Natural selection seems an apt metaphor. Due to the greater
complexity of the object of investigation and the more ambiguous and probabilistic
nature of evidence, this process is much slower in the social and economic sciences.
(In Chapters 8 and 10, I come back to this.) For this reason, it is hard to summarise
how social scientists view man on earth.
An important question in this situation is, which social science is considered
relevant and legitimate? In his book, Le Phénomène Humain (1955) the French priest
and scientist Teilhard de Chardin advanced the idea of a cosmic unfolding towards
a collective consciousness, for which he used Vernadsky’s notion of the noösphere.
In his cosmology, there are three dimensions: the infinitely large, the infinitesimally
small and the infinitely complex. He viewed the increase in complexity as a process of
unfolding and of interiorisation, in which each human individual partakes. Eastern
philosophers had long been venturing similar ideas and these emanated again in the
1960s. For instance, Shri Aurobindo describes in his book, Le Cycle Humain (1972),
his view of humanity’s evolution towards an ever higher consciousness and Western
author Capra connected 20th-century discoveries in physics with ancient Eastern
philosophy in his book The Tao of Physics (1975).
However, such speculative and metaphysical views are not considered legitimate
by most scientists. Since the Enlightenment, most social scientists have embarked on
a research program and methodology along the lines of the highly successful natural
sciences.4 They favour a view of man and society that considers the experiences of
the senses as the most authentic source of knowledge and falsifiability as its hallmark.
Their investigations are along the lines of positivist thinkers such as Saint-Simon,
Laplace, Comte and their followers. It shows up in the metaphors: The human psyche
4 Throughout this book, the term social sciences includes economic science, unless indicated otherwise.
seen as a steam engine or a computer, society considered as a self-organising net-

work. It is also evident in the notions borrowed from the natural sciences to describe
behavioural and social phenomena: exchange processes on a market as a chemical
equilibrium, innovation as evolution and political expansion as reaction-diffusion
(§8.2). This book takes the position that the natural science paradigm will ever more
penetrate the social sciences with the advent of new computer and communication
tools and modelling applications (§10.5). This is not to say that this venture will
succeed. I respect and listen to those who argue that the path of empirical reduction-
ism, even in the new clothes of complex system science, is not solving sustainability
problems. I leave this topic for further reflection in the next chapter.
From the previous summary, it can be concluded that sustainability has to be
found in an ever-changing world. This paradox can be resolved by considering sus-
tainable development as a search for the preservation of quality of life in a continuous
play between change and stability. I proceed with a brief history of how the notion
of sustainable development rose to prominence.
5.3 Rising Concerns
5.3.1 Early Concerns: Managing Common Inheritance5

Although you may think the talk about resource depletion and environmental pol-
lution is a recent phenomenon, as previous chapters have indicated, there have been
warnings of overexploitation of local and regional resources throughout the ages.
When the exploitation of sources and sinks became more intense with the advent
of industrialisation, it became increasingly obvious to all those involved (produ-
cers, consumers and governments) that regulations were needed to prevent outright
destruction. Amongst the earliest examples of interstate cooperation to manage
natural resources are the early 19th-century river commissions in Europe and the
late 19th-century agreements on seal hunting and fisheries. The conventions on
transboundary rivers addressed primarily the principle of free and nondiscrimin-
atory navigation through issuing permits and enforcing navigation rules. It was in
the immediate interest of industrialisation and its river- and canal-based transport
needs. The seal hunting led to an arbitration between the United States and the
United Kingdom about the right to unilaterally declare ownership and regulation of
resources, the motive being the preservation of seals as an economic asset.6 When
steam engines led to a rapid increase in fishing capacity, genuine concerns about
overharvesting some valuable fish species led to the first fisheries treaties. In 1902,
the installation of the International Council for the Exploration of the Sea (ICES)
was tasked to coordinate, among other things, scientific research on fisheries. It
was one of the first signs that scientific information was to play a role in resource
management.
After World War I, several attempts at international law-making took place
under the authority of the League of Nations. One of the topics was ‘the exploitation
5 This paragraph is to a large extent based upon Schrijver (2010).

6 Regarding the motives behind agreements, Schrijver (2010) gives the example of a 1902 convention
about the protection of nature and wildlife with the telling name The Convention for the Protection
of Birds Useful to Agriculture.
of the products of the sea’. Three considerations emerged in the debates (Schrijver
2010). First, the focus on commercial interests despite the increasing evidence that
those cannot and should not be separated from biological interests. Second, the
necessity to make exploitation subject to international regulation in view of the non-
territorial character of fisheries, particularly of migratory species. Third, the unwill-
ingness to accept that the principle of the ‘freedom of the seas’ is inadequate with
respect to finite amounts of resources, more so when their exploitation intensifies.7
All three observations are still relevant in the early 21st-century negotiations about
resource management. Despite many obstacles, there were some successes during
the interwar years, such as a multilateral convention on whaling (1931) and one on
the preservation of fauna and flora, notably in Africa (1933).
During these years, there were also international deliberations about the access
to raw materials, which became increasingly important for industrialised nations.
In first instance, the focus was on commodity regulation in order to reduce the
fluctuations in the prices of primary products. A more principled debate unfolded
over the ownership and control of natural resources (rubber, metals, fossil fuels,
phosphates, etc.). In essence, the rich countries of North America and Europe tried
to enforce the rules of the ‘free market’ in the form of principles that forbid restriction
on raw material exports, give equal rights to foreigners regarding natural resource
development and ask for the prevention of excessive price increases. The selfish
attitude of the then colonial powers and rich countries has not changed much since
then, but the negotiating power of many resource-owning countries has increased
significantly.
After World War II, the United Nations (UN) had the restoration and main-
tenance of peace and security as its first goal. Economic growth and expansion of
trade were considered the best way to reach it. A secure resource supply was one
of the preconditions. It was in this context that UN organisations got involved in
debates about the timber shortages and forest restoration in Europe and about put-
ting Middle East oil under UN control. In 1949, a conference was organised, where
for the first time, scientists discussed the world resource situation and concluded
that it was possible through more efficient use and new techniques ‘to support a
far greater population than exists today, at a much higher level of living’ (Schrijver
2010). However, in the 1950s and the 1960s, the international arena was increasingly
dominated by the Cold War rivalry between the (capitalist) West and the (commun-
ist) East and by the efforts of countries in the Third World to get control over their
resources in a post-colonial world. Resources for development was the foremost
concern, not depletion or environment.
5.3.2 The Environmental Movement

The book Silent Spring (1962) by Rachel Carson was a wake-up call when it appeared.
It dealt with the pollution of air, water and soil in the United States as a consequence
of rapidly growing industrial and consumer activities – in short, of economic growth.
The population in the United States and, later, postwar Europe had become rich
7 The principle was proposed by the Dutch jurist Grotius, with the argument that it is impossible to
occupy infinite air and water spaces. It was an opposition to claims to sovereignty over the oceans
by Portugal, Spain and other countries.
enough and the negative side effects had become visible enough to start worrying
and questioning. In the 1960s, the destructiveness of industrialism became even
more apparent, with media reporting on massive pollution of air, water and soil and
on the health consequences for people. In the eyes of the protesters, environmental
destruction was seen as one more manifestation, besides the repression of women and
non-Europeans, of the inherently exploitative worldview of science- and technology-
based industrial capitalism. Opposition was directed against the imperialist war
in Vietnam, the reckless expansion of nuclear weapons and nuclear power, and
the disrespectable degradation of environment and nature. Modernity had lost its
innocence and industrialism its enchantment.
Many local environmental initiatives emerged, which were organised by citizen
groups and gradually institutionalised into a variety of Non-Governmental Organ-
isations (NGOs) representing civic society. In retrospect, the protests rescued the
capitalist system by forcing it to adapt to the new ‘environmental scarcity’ – as it
was rescued by the socialist movement in an earlier era by ameliorating labour
conditions. The centralised communist systems suppressed all opposition, including
warnings of environmental deterioration, and paid dearly for it as later developments
have shown.
Scientists played an important role. The largely local actions in civil society were
given a global context with two publications in the early 1970s. The first one was
the report Limits to Growth (Meadows et al. 1971) to the Club of Rome, which
used a computer simulation model to depict in quantitative detail the possible future
trajectories of the human population, its food and industrial output and its resource
base. The two main conclusions were:
1. If the present growth trends in world population, industrialisation, pollution, food

production, and resource depletion continue unchanged, the limits to growth on
this planet will be reached sometime within the next one hundred years. The most
probable result will be a rather sudden and uncontrollable decline in both population
and industrial capacity. 2. It is possible to alter these growth trends and to establish
a condition of ecological and economic stability that is sustainable far into the future
(Meadows et al. 1971).
The report was met with excitement in some European countries and with thorough
skepticism among most scientists (Meadows 2006). The response of economists was
particularly vehement. In their view, the model was nonsensical because the crucial
feedback of prices in markets was not incorporated. This critique was also expressed
by (economists in) organisations such as World Bank and the Organisation for
Economic Co-operation and Development (OECD).
A second influential report was The Blueprint for Survival, originally an issue
of the journal The Ecologist and later published as book. It had a dramatic message
about the irreversible disruption of life support systems on Earth and the breakdown
of society, if current trends persisted. The authors expressed the hope that ‘Man will
learn to live with the rest of Nature rather than against it’. In subsequent years,
several more global analyses were made, some also based on computer simulation
models. One of these was the The Global 2000 Report to the President: Entering
the 21st Century, by the Millennium Institute (Barney et al. 1980). It informed the
United States President Carter that:
If present trends continue, the world in 2000 will be more crowded, more polluted,
less stable ecologically, and more vulnerable to disruption than the world we live in
now (1977) . . . Despite greater material output, the world’s people will be poorer in
many ways than they are today. For hundreds of millions of the desperately poor,
the outlook for food and other necessities of life will be no better. For many it will
be worse . . . At present and projected growth rates, the world’s population would
reach 10 billion by 2030 and would approach 30 billion by the end of the 21st century.
These levels correspond closely to estimates . . . of the maximum carrying capacity
of the entire earth.
It shows the danger of trend extrapolation: In 2010, most forward projections of

population in 2100 were below 10 billion people.
The Blueprint for Survival publication preceded the first large UN Conference
on the Human Environment in Stockholm with a few months. This conference was
one of the first signs that environment was recognised as a global concern. At this
meeting, the United Nations Environment Program (UNEP) was founded. It gave a
boost to environmental science and many environmental problems were identified
in scientific detail and subsequently solved or at least mitigated. In retrospect, the
ozone layer depletion and the acidification by sulphur and nitrogen compounds are
good examples. Apparently, environmental problems could be solved at limited cost
and the industrial growth paradigm did not have to be questioned.
However, a number of environmental problems were more difficult to tackle.
Examples are the depletion of open-access resources such as fisheries and ground-
water reservoirs, the degradation of agricultural soils, the cutting down of tropical
forests under the pressures of wood demand and the emission of greenhousegases
that cause climate change. Although conferences were organised and conventions
were signed about these issues, the causes are more deeply ingrained in the social
and economic habits, ideas and institutions of industrial society. There was – and is –
no clearcut scientific identification of and an equally straightforward policy response
to these problems.
Against this background, the more comprehensive notion of sustainable devel-
opment in its modern connotation became prominent in the 1980s. It reflected the
growing awareness that what was at stake was more than local environmental degrad-
ation. The core of the problem went beyond the environmental sciences and had to
do with the interconnectedness and speed of the changes – the very idea of develop-
ment. Therefore, the quest for sustainable development became the expression of
the need and desire to face the world problématique in a larger, holistic perspective
and to reconsider the very roots of modern industrial society with its market-driven
innovations, consumerism and income inequalities. This broader, systemic perspect-
ive was also the essence of the 1971 Limits to Growth report.
5.3.3 Our Common Future?

The launch of sustainable development as the new, overarching concept came with
the publication of the report Our Common Future by the UN World Commission on
Environment and Development (WECD), also named the Brundtland Commission
after its chairperson. Its main conclusion was that ‘humanity has the ability to make
development sustainable to ensure that it meets the needs of the present without
compromising the ability of future generations to meet their own needs’. It gave a
long list of actions and policy directions in order to achieve the newly stated goal of
sustainable development, noting that a world in which poverty is endemic will always
be prone to ecological and other catastrophes. Its impact was reinforced by the two
oil price crises of 1973 and 1979–1980 that confronted the industrial economies with
their dependence on oil, and the Chernobyl nuclear disaster in 1986.
Upon reading this more than twenty-year-old report, one is struck by the relev-
ance of the observations and recommendations. It has been criticised for being too
optimistic in its assessment of the physical resource base and the ecological absorp-
tion capacity (Duchin and Lang 1994). In retrospect, it can also be criticised for its
overconfidence in the willingness and capacities of governments to act on behalf
of the poor and to cooperate internationally. The fall of communist regimes, the
rise of neoliberalism and unregulated financial capitalism and the ethnic-nationalist
reaction to globalism – they all significantly changed the prospects for sustainable
development. None of them had been anticipated.
In the aftermath of Our Common Future, many meetings were organised and
many reports were published on sustainable development. The UN and members
of the UN system such as the United Nations Environment Programme (UNEP),
United Nations Development Programme (UNDP), Food and Agricultural Organiz-
ation (FAO) and the World Health Organization (WHO) took the lead. In 1992, the
UN Conference on Environment and Development (UNCED) was attended by tens
of thousands of people. In the outcome, the World Commission on the Environment
and Development (WCED) message of balancing environmental protection and
promoting development was prominently present. Two conventions were signed,
one on climate change and one on biodiversity. The Commission on Sustainable
Development (CSD) was installed with a broad mandate to promote and monitor
sustainable development and in particular Agenda 21, the international action plan.
Ever more groups from society got involved in what became a global movement.
Labour unions, youth organisations, women and indigenous people’s movements –
they all became stakeholders in the Earth’s future. These groups, broadly referred to
as NGOs, were rather loosely organised and represented a wide variety of often local
and specific issues, viewpoints and initiatives. During the 1990s, sustainable develop-
ment also appeared on the agenda of other, sometimes long-established world fora
such as the World Energy Council (WEC) on energy, the General Agreement on
Tariffs and Trade (GATT) and later World Trade Organisation (WTO) on trade,
WHO on health, UNDP on development and the United Nations Population Fund
(UNFPA) on population.
By the late 1990s, global environmental change had become firmly established
as agenda topic in national and international institutions and meetings. A crescendo
built up towards the new millennium with the Rio+5 meeting in 1997 and the
declaration of the eight Millennium Development Goals (MDGs) in 2000. One of
its manifestations was the first Global Environmental Outlook (GEO), published by
the United Nations Environment Program (UNEP) in 2000. It gave a voice to those
within the UN who feared that development and environment were not in balance:
Two over-riding trends characterize the beginning of the third millennium . . . the
global ecosystem is threatened by grave imbalances in productivity and in the
distribution of goods and services. A significant proportion of humanity still

lives in dire poverty . . . if present trends in population growth, economic growth
and consumption patterns continue, the natural environment will be increasingly
stressed . . . environmental gains and improvements will probably be offset by the
pace and scale of global economic growth, increased global environmental pollution
and accelerated degradation of the Earth’s renewable resource base (UNEP 2000).
The evaluation of the situation was influenced by, amongst other factors, the dis-
closure of the serious environmental degradation in the former communist Soviet-
Union and Eastern European countries. UNEP’S first Global Environmental Out-
look (GEO) ended politely with a message of hope: the trends towards environ-
mental degradation can be slowed and economic activity can be shifted to a more
sustainable pattern.
Some important events during the period 1972–2005:
r 1972: UN Conference on the Human Environment (Stockholm)

r 1980: The International Union for the Conservation of Nature (IUCN) presents
the World Conservation Strategy
r 1982: Adoption of the UN Convention of the Law of the Sea (UNCLOS)
r 1987: The Montreal Protocol on Substances that Deplete the Ozone Layer opens
for signature
r 1988: World Meteorological Organization (WMO) and UNEP set up the Inter-
governmental Panel on Climate Change (IPCC)
r 1992: UN Conference on the Environment and Development (UNCED), also
called the Earth Summit, leads to the Rio Declaration and Agenda 21 (Rio de
Janeiro)
r 1992: Convention on Biological Diversity (CBD) and UN Framework Conven-
tion on Climate Change (UNFCCC) open for signature
r 1994: UN Convention to Combat Desertification (CCD) open for signature
r 1995: World Food Summit, organised by the UN FAO (Rome) and a ‘five-year
later’ summit in 2000
r 1997: First World Water Forum, organised once every three years by the World
Water Council
r 2000: UN Millennium Summit and the declaration of the MDGs (New York)
r 2002: World Summit on Sustainable Development (WSSD) (Johannesburg)
r 2005: Ratification of the Kyoto Protocol.
In 2002, the World Summit on Sustainable Development was organised in Johan-

nesburg by the UN Commission on Sustainable Development (CSD) around five
issues with the acronym WEHAB: Water and sanitation, Energy, Health, Agricul-
ture, Biodiversity. In line with the trend of the 1990s to engage business, there was a
growing emphasis on the implementation via partnerships in the form of agreements
between governments with and between other societal stakeholders (public-private
partnerships). The involvement of even more groups and interests made effective
decision making more difficult. When the challenges ahead and the required costs
and efforts became better understood, the divergent stakes and viewpoints of the
UN member states surfaced.
Box 5.1. Celebrations go global. One of the many signs of globalisation, in the
spirit of global villages and universal human rights, are the many UN-initiated
celebrations. The more than one hundred religious festival days of medieval
Europe are being replaced by days commemorating the new world culture: The
UN has now twenty-six days in the year to be celebrated (www.unac.org). Quite a
few are on sustainable development related aspects, amongst them World Water
Day (March 22nd), World Health Day (April 7th), World Environment Day
(June 5th), World Day to Combat Desertification and Drought (June 22nd),
World Population Day (July 11th), International Day for the Preservation of the
Ozone Layer (September 16th), World Habitat Day (1st Monday of October) and
World Food Day (October 16th). Also, there is, of course, United Nations Day
(October 24th) and days for many other good intentions: International Women’s
Day (March 8th), World Press Freedom Day (May 3rd), World Refugee Day
(June 20th) and World AIDS Day (December 1st). The round of worldwide
celebrations started with a citizen’s initiative: In the 1970s, United States citizens
started Earth Day (April 22nd) as a demonstration for a more sustainable world.
The UN celebration days can help create consciousness of global issues and
remind the UN every year that there is work to do.
Also, an undercurrent of different views and interests had started to surface. In

the words of Bob Dylan in 1963: ‘There is a battle outside and it is ragin’ . . . For the
times they are a-changin’ . . . ’ The world followed in words, but not in actions the
WCED recommendations. With the fall of the Iron Curtain and the proclamation of
The End of History, a wave of neoliberal capitalism and Information and Commu-
nication Technology (ICT) inspired optimism set in and worries about environment
and poverty became less fashionable in the postmodern world of riches and fun that
emerged. In 1997, the financial crisis in Asia and the ‘dotcom’ stock market bubble
in 2000 signalled the search for short-term gain and adventure. In 2001, after the
9/11 terrorist attack in New York, there was a sudden concern for the consequences
of the global and unregulated capitalist expansion. Financial crisis management and
anti-terrorism came to dominate political agendas and the willingness and trust to
act for a better long-term future declined. In hindsight, in the first decade of the 21st
century, the quest for sustainable development lost the momentum and coherence
needed to make a sustainability transition happen.
Still, most scientists engaged in Global Change research are convinced that the
accelerating changes brought upon the biosphere by the human species constitute a
serious threat for the well-being of future human populations. They base their view
on scientific observations and interpretations of change processes at all scales. The
latest warning is from a group of ecological economists, who in 2009 concluded in a
paper on planetary boundaries that:
. . . the planet’s environment has been unusually stable for the past 10,000 years.
This period of stability . . . has seen human civilizations arise, develop and thrive.
Such stability may now be under threat. Since the Industrial Revolution, a new era
has arisen . . . in which human actions have become the main driver of global envir-
onmental change. This could see human activities push the Earth system outside the
stable environmental state of the Holocene, with consequences that are detrimental
or even catastrophic for large parts of the world (Rockström et al. 2009).
These are broad sweeping statements that are not easily validated or falsified. There
are still many uncertainties about the long-term and large-scale consequences of
past and present developments. Therefore, scientific findings have to be stated in
probabilities and risks, not in terms of predictions and facts. One reason for this,
and scientists should acknowledge it, is that a necessarily biased evaluation of the
scientific facts will happen whenever there are large and as yet irreducible uncertain-
ties and complexities. We have entered the world of postnormal science (Funtowicz
and Ravetz 1990). It is one of the important reflections in sustainability science, and
we come back to it in Chapters 6, 8 and 10. Let us now look more closely at how
these concerns gradually shaped the notion of sustainable development.
5.4 The Notion of Sustainable Development
5.4.1 Prelude: Categories of Goods and Services

The previous sections teach two things. First, according to modern science,
everything is connected and in change. Nothing happens in isolation, which is why
we need a system’s view. Second, resource management is inherently a collective
affair. We interact with each other through the use of environmental sources and
sinks, which are part of the public space – or, more solemnly, of the ‘common herit-
age of mankind’. Therefore, it makes sense to introduce two notions from economic
science: excludability and rivalry. When a good or service loses its use value upon
‘consumption’, it is called a rival (or rivalrous) good.8 Food is a good example: Once
consumed by me, it is no longer useful for you, at least not as food. If a good or
service can be used again and again and also simultaneously by many consumers, it
is called a non-rival good. An example is a beautiful sunset, at least as long as one
does not feel bothered by other tourist-consumers. If it is possible to prevent people
from consuming or enjoying a good, it is called excludable. Most private goods are. If
one cannot exclude another person from enjoying it and the owner cannot enforce a
meaningful form of payment, the good is non-excludable. The same beautiful sunset
is an example. It gives aesthetic and other pleasures that are difficult to exclude
people from.
Rivalry is a continuum. In principle, my enjoyment of the sunset does not inter-
fere with yours. In practise, this is only true within certain bounds. Your wish to enjoy
a natural scenery or follow a high-quality class in sustainability science may be valid
and honoured if the number of visitors or students is limited. If it exceeds a certain
threshold, user expectations and actual use start to interfere: The roads are clogged,
so you only reach the natural park after sunset, and the classroom is overloaded with
noisy students. Congestion makes the good or service rival. Excludability is also a
continuum. In principle, anything that can be appropriated can be made exclusive,
as some dictators and super-rich people demonstrate. But it may be difficult and not
8 We will speak of goods but actually refer to the broad array of goods and services, which can satisfy
needs and desires and for which to a smaller or larger extent material stocks and flows are involved.
Figure 5.2. Four categories of goods and services, based on the dimensions of rivalry and
excludability, with examples from the perspective of the individual in society.
worthwhile. Nobody can be excluded from fresh air near the seashore, UV-radiation
from the ozone hole, or a less stable climate.
Based on these two dimensions, economists have introduced four categories
(Figure 5.2). If I own a pond with fish in it, I can prevent others from catching and
eating my fish. The fish are a rivalrous and excludable private good. If the pond is
not yet appropriated by any individual, either by force or with societal sanctioning,
its fish are called a common pool resource (CPR): It is rivalrous and non-excludable.
The word stems from the commons in historical documents that refer to land that
is not used by a single person or household during the whole year, but by several
persons or households for parts of the year. There are equivalent terms in other
European languages, and the common use was usually subject to some kind of rule.
The water in a river is non-rivalrous and non-excludable: It is a public good.9 Fresh
air and security in the streets are public goods and services too, as it is neither
possible nor necessary to prevent people from using them.
In the real world, there are hardly any absolutely non-rival or non-excludable
goods and services, but some come close. Examples are given in Figure 5.2 for illus-
trative purposes. How to value a good or service in practice is context-dependent.
Most resources have to some extent the characteristics of rivalry and excludability.
This shows up in the ownership and user right arrangements of the people who use
the resources. Such arrangements reflect historically grown rules and practises. In
Europe and its offshoots, individual property rights have become strongly embedded
9 Of course, this is not true for use of river water for irrigation, where the upstream user leaves less
water for the downstream users. This is a well-known CPR situation for which historically solutions
have been found (§ 12.3).
Table 5.1. Five modes of provision of goods and servicesa
Frame 2: Frame 3:
Frame 1: creation rules of the subjective
Modes of provision constitution of goods provision dispositions
Self-provision No exchange and no Allocation to producer a-personal

commensuration
Informal provision Exchange (direct) and Social norms personal
commensuration
Market provision Exchange (direct) and no Price mechanism, on the impersonal
commensuration basis of preferences and
endowments
Professional provision Exchange (indirect or direct) Expert knowledge impersonal
and commensuration
Public provision Exchange (indirect) and Political decision making impersonal
commensuration
a Based on Claassen, 2008.
in law, although other possibilities as in cooperatives exist. Many resources, in a

broad sense, are appropriated as private goods, the owner claiming the right to
exclude others. The owner can also be the state, and state ownership of resources is
usually seen as an essential part of state sovereignty. Because resources often extend
over more than one private owner or sovereign state, there are often conflicts that
have to be resolved. In early times, the issue was often settled by violence – and
still is in some regions. Nowadays, governments offer provision of public goods on
the basis of political decision making and laws, rules and regulations. For common
pool resources, governments negotiate and implement collective arrangements and
regulatory frameworks. For private goods and services, the market is the dominant
system of provision of goods and services, based on prices. Other modes of provision
are autarchic, informal and professional and all had and have their place in the
resolution of (resource-related) disputes (Claassen 2004; Table 5.1).
Regarding sustainable development, the sovereign nation-state is the most
important actor and stakeholder and this explains the important role of the higher-
level political entities such as the U.S. Federal Government, the EU Commission
and the UN in framing policies. Many resources are shared by more than one state,
and legal arrangements are needed to settle user rights – leaving aside the option
of conquest. Rivers and coastal fisheries and the disputes about mineral and fossil
fuel resources have already been mentioned. The situation is most problematic with
regard to those resources that are outside any national jurisdiction, such as ocean
fisheries, rich mineral ores on the ocean floor, and the atmosphere. The overexploit-
ation of these global commons can only be prevented by treaties and conventions
that are signed by enough countries to perform effective monitoring and punishment
of offenders.10 Chapters 10–12 come back to this in the context of social dilemmas.
A variety of collective resources are positioned within the framework of rivalry and
excludability in Table 5.2.
10 Other ‘resources’ that can be considered global commons are the genetic resources of living organ-
isms and the cultural resources of indigenous peoples, but both rivalry and excludability are more
ambiguous in these cases.
Table 5.2. Four categories of goods and services, based on the dimensions of rivalry and excludability
and seen from the perspective of the nation-state
non-excludable excludable
outside territory interterritorial territorial
non-rivalrous Public goods ‘Club’ goods

peace and security navigable rivers army-owned estates
outer space
the atmosphere intercountry rivers and seas state-owned nat’l parks
polar regions wild seals (Greenland) mineral resources
(arctic/antarctic) (territorial)
migratory species marine resources (territorial sea) forests (territorial)
‘high sea’ fisheries
mineral resources
(ocean floor)
rivalrous Common pool resources Private goods
(CPR)
Box 5.2. Externalities. Economists call the effects of economic activities that are
not, or not adequately, taken into account in the market allocation externalities.
For instance, a firm benefits from well-educated employees or an efficient infra-
structure, but does not pay for it – at least not directly. Inhabitants suffer from
air pollution from a local factory, although most of them will not benefit from its
employment or profits – at least not directly. The externalities are often related
to non-excludable goods or ‘bads’. Those who benefit from positive externalities
cannot be made to pay for it by those who provide it and those who suffer from
negative externalities have no user right or claim which permits them to punish
the polluter or be compensated. Obviously, in all these situations, the collective,
a government of some sort, has to mediate. How it should do this is the topic of
a large part of environmental economics. Solutions are a redistributive tax – the
inclusion of the damage in the prices (‘internalising’), assigning property rights
and negotiating a workable arrangement among stakeholders. Each solution has
its pros and cons. Sometimes there is also an issue of rivalry. For instance, if a
bus lane is constructed near your house, you benefit without effort or payment –
apart from tax money. It is a public service. But if many people start using it, its
use becomes rivalrous – and possibly cheaper. Similarly, if a noisy and polluting
factory is built near your house, you share in the ‘bads’ – although you may gain in
other ways. It is a public disservice. You may respond to it with in-house airfilters,
by moving to another place or starting a lawsuit. In the case of climate change
from past carbon emissions, the problem is de facto because it is imposed upon
people as a public bad and investigating the capability and cost of adaptation
have become the dominant response.
5.4.2 Interpretations and Definitions

The list of large international meetings organised since the 1980s shows how envir-
onmental concerns gradually expanded to also include development issues and then
evolved into the more comprehensive notion of sustainable development. The Rio
Declaration in 1992 begins with the statement, ‘Human beings are at the centre
of concerns for sustainable development. They are entitled to a healthy and pro-
ductive life in harmony with nature’. In 2002, the Johannesburg Declaration built
on this aspiration and expressed the commitment of world leaders ‘to build a
humane, equitable and caring global society cognizant of the need for human dignity
for all’.
Thus, sustainable development has evolved into an ethical guiding principle and
leading aspiration of humankind in the 21st century, not unlike the idea of socialism
in the late 19th and early 20th centuries and the 1948 Declaration of Human Rights
in the late 20th century. Such principles and aspirations do not necessarily have to
be defined very precisely to be effective. They should be open for reappraisal and
adjustment in the light of new facts and experiences. At the same time, however,
there is a clear need to operationalise the idea of sustainable development into
feasible objectives and measurable targets. This demands a process of convergence
towards a shared notion of what sustainable development is about and how it should
effectively be implemented. This book hopes to contribute to this process. Therefore,
I present a closer look at a few definitions.
In first instance, the notion of sustainability emerged in circles of nature conser-
vationists and emphasised the aspect of human intrusions into natural ecosystems. In
this case, the outlook in the World Conservation Strategy (1980) of the International
Union of Conservation of Nature (IUCN) was outspoken bio- or eco-centric. The
emphasis on the rapid degradation of biodiversity was also the major concern in
the next strategy report, Caring for the Earth (1990). The Dutch Stichting Natuur
en Milieu (1981) presented guidelines that reflect more an environmental than an
ecological view:
Requirements for a sustainable society are:

r avoidance of risks at the macroscale;
r use of renewable energy sources as much as possible;
r limitation of the use of non-renewable resources to a minimal share of the
available resources, and
r consumption of renewable resources that does not exceed the amount that is
sustainably produced by nature.
Energy was often a major element. In the late 1970s and early 1980s, it reflected the
widespread concerns about air pollution, risks of nuclear power – accidents, handling
and storage of radioactive waste and proliferation of material for nuclear weapons –
and energy supply security. Oil prices were at an all-time high, and the Chernobyl
disaster came five years later. Concern about climate change from greenhouse gas
emissions only entered in the second half of the 1980s.
The definition offered by the Dutch National Institute for Public Health and
the Environment (RIVM) in an influential report Zorgen voor Morgen (Care for
Tomorrow) represented more clearly the environmental scientist:
‘Sustainable development requires that exploitation of and pressures on reservoirs

(both abiotic and biotic) is such that the powers of regeneration are not exceeded
and that flows are not influenced in an undesirable manner’ (Zorgen voor Morgen,
RIVM 1988).
This definition used explicitly the systems terminology (§2.4). For many environ-
mental NGOs, the ecological notion of carrying capacity was important: Sustainable
development is ‘improving the quality of human life while living within the carrying
capacity of supporting ecosystems’ (IUCN-WWF 1991).
Already early on, there was a more anthropocentric orientation on sustain-
able development. It emphasised unjust and unfair income and wealth distribution
and was prominent among poverty and development oriented NGOs and religious
institutions. An example is the statement by the World Council of Churches in
1976:
A sustainable society requires

r attention for the long term;
r to keep open as many options for future generations as possible;
r continuous redistribution of global energy resources; and
r a “contract” with nature in order to reach a just and responsible use of the
harvest.
With the Our Common Future report in 1987, the anthropocentric interpretation
became even stronger. It proposed to include explicitly socio-economic aspects, as
is shown in one of the most widely quoted definitions:
Sustainable development is development that meets the needs of the present gen-
eration without compromising the ability of future generations to meet their own
needs (WCED 1987).
The report argued that developments going on in the world, both in the North
and the South, were environmentally, socially and economically not sustainable.
Warning that behavioural changes at all levels would be needed, it described the
characteristics of a sustainable society:
r a political system that ensures the effective participation of citizens in decision
making;
r an economic system that generates value added in an intrinsic and sustainable
way;
r a social system that is capable to resolve tensions caused by inharmonious devel-
opments;
r a production system that keeps the natural base intact;
r a technological system that searches continuously for new solutions;
r an international system that ensures sustainable structures for trade and finance;
and
r a management system that is flexible and can correct itself.
This is about as broad and ambitious as the aspiration for a better world can be.
Another difference in emphasis started to influence the definition and the
debate: economy versus ecology. When environmental and other economists got
involved, they stressed the benefits of economic welfare that were in their view
undervalued:
‘[sustainable development] requires the maximisation of the net benefits of economic
development in such a way that services and the quality of natural resources are
maintained’ (Pearce 1990:24).
The World Bank joined with the World Development Report 1992, Development and
the Environment, in which it observed that more than one billion people lived in acute
poverty and had insufficient access to natural resources. It also advocated the theory
that economic growth caused environmental problems with adverse consequences,
but that they could only be solved by ‘resource-efficient economic growth’. It sum-
marised what it called the prevailing paradigm of worldwide economic growth, based
on the belief that:
r the negative effects of economic growth can be reduced substantially, and
r economic development is possible without repeating the industrial countries’
mistakes (‘leapfrogging’).
This requires:
r policy and investment to be directed towards the more efficient use of our
resources;
r replacement of scarce resources; and
r application of environmentally benign technologies.
The three P’s: People, Planet, Profit adage emerged in this period as a short name
for the insight that neither environmental degradation nor poverty can be elimin-
ated without profitable business. Economists, notably at the World Bank, invented
in this period the broadening of the notion of capital. Besides economic capital,
there is natural (or ecological) capital, human and social capital, and cultural cap-
ital to be considered. These concepts underline the durability and inertia of many
processes and activities, but they are difficult to operationalise and remain rather
vague. With the economist’s outlook came a more positive attitude towards techno-
logical solutions and business involvement, and a more practical and action-oriented
orientation.
The economy versus ecology debate has not ended. ‘Mainstream’ economists –
many of them from the less industrialised countries – emphasise the need for
economic growth and poverty reduction. Most ecological economists argue that
economic growth will not relieve environmental pressure and, instead, jeopardise
quality-of-life improvements and exacerbate unsustainability. The debate reflects
the broader tension between anthropocentrism versus ecocentrism and between a
focus on Man versus a focus on Nature. Among the followers of a more economic
orientation, a distinction can be made between those who emphasise equity concerns
and those who focus on efficiency. The former group stresses that more equitable
access to resources and income distribution and the elimination of extreme poverty
are preconditions for sustainability. Many poor people have no alternatives other
than using their resources in an unsustainable way, a situation that is aggravated
by exploitative trade and aid patterns. This view is prevalent among development
NGOs and UN organisations like UNDP and implies that the quest for sustainable
development should focus on breaking the poverty-overexploitation cycle. This asks

for better governance and institutions and internationally coordinated trade and aid
regulation. The latter group, with a focus on efficiency, equates sustainable develop-
ment in essence with environmental protection and argues that economic growth is
a precondition, in the sense of affordability, for sustainable development. Measures
and policies should not impede economic growth and be done in the economically
most efficient way. Prices must be market-based ‘optimal’. It is the prevailing view
among the business and government elites. Of course, there are multiple shades of
grey in between.
5.5 An Indicator Framework for Sustainable Development
5.5.1 From Principle to Action: Indicators

We have seen how attempts to make sustainable development an operational and
action-oriented concept have led to more comprehensive frames and definitions. The
representation in terms of economic, natural, human and social capital, the three P’s
and the Nature, Society, Wellbeing, Economy in the ISIS methodology (AtKisson
2008) are examples. The evolution of definitions shows divergence in emphasis on
Nature, on income and on poverty. How to connect the broad principles and goals
to local and global action? One answer is: through indicators. There is vast literature
on (performance) indicators in the management literature. However, I confine the
discussion here to specific sustainability indicators (Kuik and Verbruggen 1991; UN
1996; Parris and Kates 2003).
Actions at the individual, group or (inter)national policy level happen with a
certain degree of rationality. Figure 5.3 represents a system’s perspective on the
decionmaking process (§2.4; Meadows 1998). Suppose that you wish to change some
situation in a desired direction and that this requires interference with some state
variable X. Your perception of the situation (XP ) may differ from the actual situation
(X), due to delays, filtering and so on. If the desired situation is XG , then your
action will be based upon the difference between XG and XP . The discrepancy
between goal and perceived reality drives the desire and action to change. This
simplified representation suggests that 1) a goal (target, objective or purpose) has
Figure 5.3. A system perspective on decision making and the role of indicators.
to be formulated, and 2) quantitative and qualitative descriptions of the goal and

associated indicators should be formulated for operational action and monitoring.
First, an example at the personal level with you as house-owner. The situation:
You use E units of energy per year and pay P €/year for it. You consider the
situation undesirable and set the goal to reduce E by half in two years time. Your
prime indicator is P, although your goal is stated in terms of E. Your perception
may be somewhat wrong, for instance, because last year was exceptionally cold.
Nevertheless, your task is simple: After two years, you check whether your energy
use has halved as a consequence of all the actions you have undertaken.
Suppose, you live with four other people in one house together with the same
energy use E and energy bill P. You have the same goal of halving energy in two
years – but you want the other four inhabitants to do their share. You cannot exclude
them from using energy and their energy use competes with yours in the sense that
your effort to reach the common goal increases when one or more of the others
don’t care. It is a public good or common pool resource situation. You negotiate
with the others first about your goal, because some fellow inhabitants may think it
is nonsensical, impossible or both. Then, you negotiate about the distribution of the
burden, as there will be different preferences regarding technical options like buying
efficient lightbulbs or behavioural options such as wearing a sweater more often –
each with their specific indicators such as cost.
Here is another example, but now at a town/region level. The province of Utrecht
is among the most sustainable regions in the Netherlands according to the Triple
P Monitor, designed and implemented by employees of a Dutch bank.11 A score is
given to each of the forty administrative regions along the dimension social (people:
societal coherence), economic (profit: income generating capability) and ecological
(planet: quality of nature and culture). The scores are based on a dozen properties,
each in turn derived from forty-five characteristics. The results are shown in maps,
one for each of the three P’s and one for the aggregate.
The exercise uses statistical data in first instance. This gives it an objective
appearance. However, the data have to be weighed in order to construct a single
aggregate indicator for each of the three P’s. This is particularly difficult when certain
characteristics are ambiguous. Once, in a discussion about sustainability indicators,
someone from the Netherlands suggested the number of policemen per thousand
inhabitants as an indicator. She meant it as positive because the higher police count
meant more safety (better). Another participant from an African country said it
is negative in his country, because the higher the police count, the more trouble
(worse). Another problem with approaches like the Triple P Monitor is that it omits
dynamic and spatial relationships that tie regions to their past and to each other.
A last and related problem is that ends and means are to a certain extent reversed:
The available data are the basis for evaluation (the means), whereas the goal of
the exercise, namely the promotion of sustainable activity patterns, remains largely
implicit (ends). Therefore, such indicator maps are certainly useful, but they do not
resolve the issues surrounding sustainability or sustainable development indicators.
There are some practical considerations about the construction and use of sus-
tainable development indicators (SDIs). The essential role of an SDI is to guide the
11 www.uu.nl/SiteCollectionDocuments/Corp UU%20en%20Nieuws/triple p monitor.pdf.
decision-making process after a goal is agreed upon. In first instance, the SDI can be
based on statistical data such as trends over time and distributions across space. This
limits them to what can be and actually is measured and collected. In other words,
SDIs provide quantitative information, but possibly not about what we value or what
we consider important. But the advantage is that there can be precision about what
is measured and about how it is measured, even though measurement errors and
deliberate falsification cannot be excluded.
Information and, thus, SDI are more than statistical data. Data in time and space
have structure that is at least partly captured in scientific models. One might say that
indicators are ‘on the outside’ of scientific models. One disadvantage of demanding
structure is that it limits SDI further to what can be and actually is modelled. It
also runs the risk that an incomplete and partly value- and discipline-biased model
determines the goal – or means determining ends. But an advantage is that SDI
can be related to past trends and to other SDIs so that unrealistic ideas about their
evolution are exposed.
How an indicator is presented is important for its (lack of) usefulness. For
instance, the installed capacity of wind turbines in the world can be seen as an
indicator of the progress towards a sustainable energy future (Figure 2.7b). But one
can also present the same information in a different format. If windpower capacity
is shown as a fraction of total installed capacity, the message is that wind turbine
capacity is still only about 4 percent of the world total. If the electricity generated
is used as indicator, it tells you that windturbines generate less than 2 percent
of world electricity. If the subsidies for wind turbines are shown as an indicator,
one realises that growth may not stop overnight if subsidies are abolished. This is
not to say that installed capacity of windturbines is a bad SDI – it only warns for
simplistic interpretations. A few additional model-based indicators already make it
more accurate, in particular if the important causal relationships are also shown. But
there is clearly a trade-off between completeness, on the one hand, and transparency
and ease-of-use, on the other.
5.5.2 A Sustainable Development Indicator System (SDIS)

A satisfactory single indicator for sustainable development, as goal and performance
measure, will probably never be found. A single indicator will always fall short in
view of the diversity of issues and actions considered in sustainable development.
Besides, its construction will always need a subjective and value-laden weighing of
the components. It may be better to speak of a sustainable development information
system (SDIS). One can make a list of items and conditions for such an SDIS. First,
its construction is an interdisciplinary effort, with inputs from physics to philosophy.
It must be possible to quantify or find adequate proxies and assign units to the
indicators in the SDIS: physical stocks and flows to describe ecosystems and societal
metabolism, monetary units to describe economic transactions and, finally, other
units to describe the complexities of human society.12 Conversions into units per
capita or in units per land area or normalised with respect to a given year or reference
situation can make the SDIS more informative.
12 A widely used framework are the country-based economic input-output (I-O) tables in combination
with satellite accounts for natural resource stocks and flows (§ 14.4).
Second, the SDIS should be balanced and address essential dimensions of sus-
tainable development of a (sub)system such as social equity and sufficiency, eco-
nomic efficiency, and environmental and ecological sustainability. In short, they are
about quality of life. The SDIs in the system must be acceptable for the scientist
and accessible for the politician. They communicate well, thanks to good visual and
verbal representation, such as interactive views, maps, animations and so on. They
are also cheap, making their construction and maintenance a relatively simple task
for the statistical bureaus of countries in cooperation with scientists.
The elaboration from SDIs to SDISs can start with the inclusion of simple
and transparent models to provide context. For example, the use of resources can
be connected to the known reserves, the rate of emissions can be linked to its
accumulation rate in the environment and available food per capita can be indicated
in relation to a healthy nutritional diet. The system boundary, indication of mass
and monetary flows and of property rights and damage across boundaries must be
well defined. Distributional impacts across a population are shown in the form of
distributions across income scales and location.
Clearly, the construction of an SDIS requires the art of compromise, and the
quantitative has to be complemented with the qualitative. One can raise the expect-
ations about a SDIS in two directions. The first direction is preferred by the scientist,
the second direction by the politician. The first one tries to link the SDIS stronger
to scientific models, preferably models with good reputation among scientists and
legitimacy among politicians. This is especially true for SDIs related to the Global
Change phenomena, such as the causes and effects of resource availability and deple-
tion, land productivity and degradation, environmental quality, biodiversity and so
on; such an embedding has great value added. It also gives flexibility in the sense
that new insights and changing interests are easily accommodated by creating and
adding new model-based indicators. As previously said, a danger is that models are
often intransparent to all but the experts, and they are discredited when the integrity
of the expert scientists’ are in doubt or when there are more models with conflicting
outcomes. The preferred solution to this is to indicate explicitly the uncertainties
involved (PBL 2010a).
The second direction to go is to see the construction of an SDIS primarily as an
ongoing and participatory process among the group of stakeholders involved (Bossel
1998; Atkisson 2008). Such direct involvement tends to focus on local situations and
to bring in personal experiences and insights. This gives the SDIS more legitimacy
and more support for action. Disadvantages are that the participants may lack rel-
evant scientific knowledge, their interest and attendance may be volatile, and the
focus on their own local quality of life can give a bias to the short-term and the incid-
ental. As such, a subset of worldviews may come to dominate the process and the
outcomes. The challenge is to widen the circle towards quality of life in an inclusive
worldview (see Figure 1.1).
5.5.3 Quality-of-Life–Oriented and Aggregate SD-Indicators

The notion of quality of life preceded the concerns about sustainable development
and there have been earlier attempts at constructing quality-of-life indicators. The
underlying intention was to improve the life of human beings, with the corresponding
goals being conceived in the spirit of 19th-century utilitarianism, Roosevelt’s New
Deal and the UN Charter on Human Rights in the 20th century. There is now
an enormous and growing amount of data available on the social and economic
characteristics and trends of countries, as can be seen from the suggested websites
at the end of this chapter. These are used to construct quality-of-life indices from
national averages on infant mortality rates, life expectancy, energy use per person,
household size, number of persons per physician, literacy level of women and so on.
These are gradually extended to include aspects of environmental and ecological
sustainability. This approach is in agreement with an emphasis on quality of life as
the essential ingredient of sustainable development.
One early example of a quality-of-life index is the one introduced by Forrester
(1973) in the early simulation models of world dynamics. He defined it as: ‘Quality of
life (QL) is a measure of performance of the world system . . . computed as a quality
of life standard (QLS) multiplied by four multipliers derived from material standard
of living, crowding, food, and pollution’ (1973). Its value over time was calculated
from the model simulation. For such aggregate, normalised indicators is the coupling
to a dynamic model almost a necessity, because it is the change and not the value
per se that matters.
A benchmark of setting targets in combination with indicators was the World
Bank report, A Better World for All (2000). It set goals for the period 1990–2015
for a number of issues: reduce the proportion of people living in extreme poverty
by half13 ; enroll all children in primary school; and make progress towards gender
equality and empowering women by eliminating gender disparities in primary and
secondary education by 2005. Also, the other goals focused primarily on health,
notably child and maternal mortality. Only the last of the seven objectives was about
the environment: implement national strategies for sustainable development by 2005
so as to reverse the loss of environmental resources by 2015. In essence, these are also
the earlier mentioned goals accepted by the UN as the Millennium Development
Goals (MDGs):
r Eradicate extreme poverty and hunger;

r Achieve universal primary education;
r Promote gender equality;
r Reduce child mortality;
r Improve maternal health;
r Combat major diseases; and
r Ensure environmental sustainability.
Each of these goals has its own set of indicators, ranging from the local to the global.
They reflect the worldview of Modernism explicitly or implicitly and cover only part
of the sustainable development spectre.
The last category worth mentioning are aggregate indicators that try to com-
bine the three P’s. The best known is the Human Development Index (HDI), which
13 The goal was from 30 percent to 15 percent, and it was 25 percent in 2000. Since then, the definition
of extreme poverty has been adjusted. The revised estimates suggest a drop from 41.7 percent in
1990 to 25.7 percent in 2005. See www.socialanalysis.org for a critical discussion of the measures
used, for instance, extreme poverty.
was proposed in the 1990s as a complement or, for some, an alternative to the
most widely used indicator of welfare, namely the Gross Domestic Product (GDP)
per capita (§14.5). The HDI is the geometric mean of a life expectancy index, an
education index and an income index and thus takes into account health, educa-
tion and standard of living aspects of quality of life. The HDI is slowly gaining
recognition as a more adequate measure than income, particularly for low-income
countries.
Within a Global Change framework, there are a number of aggregate indicators
that have become prominent in the sustainable development discourse. The large
datasets on Global Change can best be considered as signals of the State of the
World.14 Most of these data are given as (global) maps, and I refer to this chapter’s
suggested websites for more information. These data are not easily aggregated – just
think about how you would construct an indicator such as ‘average global surface
temperature’ or a single index of land degradation. Many data are related to spe-
cific environmental problems, such as ozone layer depletion, land erosion and soil
acidification. For most of these, one can speak of indicators, as there are regional
or global targets either already accepted or under negotiation, which are part of a
SDIS. There are many data available on resources, but their more regional or local
features make it less amenable to aggregate indicators other than (world) market
price and reserve-production ratio.
There has been one explicit attempt to include other forms of capital than
economic/financial capital into the economic equation: the Genuine Savings Rate for
countries (World Bank 1997). The rationale is that natural capital is destroyed, often
reinforced by subsidies, for short-term gains and is not compensated by other forms
of capital, for example, human capital. The inclusion of natural resources such as oil
and forests as ‘natural capital’ on the balance sheet indicates that development in
some countries was and often still is unsustainable. It implies a downward correction
of net economic growth with several percentages, resulting in negative values. It
suggests that economic growth is only sustainable when the total amount of capital
is sustained – with the implicit assumption that the various forms of capital are
substitutable (§14.3).
A second aggregate indicator that tries to take into account the negative parts
of economic growth is the Genuine Progress Indicator (GPI). This later developed
into the Index of Sustainable Economic Welfare (ISEW) for countries. Its main
assumption is that GDP and income (GDP/cap) is a false measure of well-being
because it counts many undesirable or harmful activities as positive. Therefore,
those activities – such as pollution abatement but also car accidents and military
adventures – are to be substracted from the official GDP figures. It has been found
that most affluent countries are beyond an optimum income since the 1980–1990s,
and that further GDP growth is no longer a net increase in well-being measured as
ISEW. But this so-called ‘threshold hypothesis’ is controversial and the outcomes
depend, of course, on the weighing factors applied.15
14 Under this heading, with the subtitle Progress Towards a Sustainable Society, the Worldwatch
Institute has published an annual report since 1984 (Brown et al.).
15 It is, therefore, instructive to do an interactive exercise with the ISEW on the site of Friends of the
Earth, www.foe.co.uk/community/tools/isew/templates/storyintro.html.
A third aggregate indicator is the Ecological Footprint (EF).16 It has more out-
spoken environmental roots. It expresses environmental impacts in terms of the
equivalent space used and is calculated by adding up the cultivated land required
for growing crops, pasture for grazing animals, forests for timber production, the
space related to fishing grounds, the land use for the built environment (housing,
transportation, infrastructure and power stations) and other space requirements. In
addition, fossil fuels are converted into equivalent space via the biologically product-
ive area needed to sequester enough CO2 to avoid its increase in the atmosphere.
It received wide publicity through a 2002 paper stating that the human population
was now using 1.2 Earth – a clear case of ecological overshoot (Wackernagel et al.
2002). The EF is also calculated for countries. It is severely criticised by scientists for
its lack of scientific consistency and the neglect of other externalities than negative
environmental ones (Opschoor 2000; Grazi et al. 2007). However, it has acquired a
solid position among large groups of people as a concept that links the individual to
the state of the planet.
5.6 Summary Points

The scientific worldview teaches us that everything is in flux and that natural selection
is the mechanism of change. Sustainable development is a search for the preservation
of quality of life in a continuous play between change and stability. Regarding the
emergence and use of the concept of sustainable development, there are a couple of
points to remember:
r Since the 1950s, there have been warnings about resource supply limitations to
development. In the 1970s, pollution of the environmental sinks became another
concern. In the 1980s, development and environment came together in the UN
report Our Common Future in the term sustainable development.
r In the course of the 1990s, the ambiguity and plurality of the concept became
clear in divergent emphases on nature versus culture and on efficiency versus
equity. A common denominator is to have a balance between the three P’s:
people, planet, profit.
r Many sustainability issues derive from the collective (public) character of
resources and the interaction between resource users. This has been formal-
ised in the notions of excludability and rivalry, and associated categories of
public goods (PG) and common pool resources (CPR).
r To implement actions to reach a goal, one needs indicators. Sustainable Devel-
opment Indicators (SDI) are constructed on different scales and at different
aggregation levels. Some aggregate SDIs are the Human Development Index
(HDI), the Genuine Savings Rate, the Index of Sustainable Economic Welfare
(ISEW) and the Ecological Footprint (EF).
16 An indicator constructed and organised along similar lines is the Water Footprint (www.
waterfootprint.org). Also, the Human Appropriation of Net Primary Production, as a measure
of human interference in natural photosynthesis processes, is in this category (Haberl 1997).
Useful Websites 143
The optimist believes we live in the best of all worlds.

The pessimist fears she’s right.
Listen, O Lord of the meeting rivers,

things standing shall fall,
but the moving ever shall stay.
Basavanna, Speaking of Shiva
SUGGESTED READING
Lumley, S., and P. Armstrong. Some of the nineteenth century origins of the sustainability
concept. Environment, Development and Sustainability 6 (3) (2004): 367–378.
A comprehensive and accessible report on indicators, based on a 1996 workshop. The report
is available via the Sustainability Institute (www.Sustainer.org) or www.biomimicryguild
.com/alumni/documents/download/Indicators and information systems for sustainable
develoment.pdf.
Meadows, D. H. Indicators and Information Systems for Sustainable Development. Report to
the Balaton Group. Sustainability Institute, 1998.
Treatment of concepts of excludability, rivalry, commons and public goods.
Perman, R., Yue Ma, J. McGilvray, and M. Common, eds. Natural Resource and Environ-
mental Economics. Harlow, UK: Pearson Education Ltd, 2003.
A comprehensive overview of the history of international debates and decision about resource
management, and in particular the role of the United Nations.
Schrijver, N. Development without destruction – The UN and global resource management.
Bloomington: Indiana University Press, 2010.
A conceptual natural science approach to Global Change.
Schnellnhuber, G., P. Crutzen, W. Clark, and J. Hunt. Earth system analysis for sustainability.
Environment 47 (8) (2005) 10–27.
A good overview of some of the major issues in Global Change science from a natural science
perspective. The report can be downloaded from www.igbp.net/page.php?pid=221.
Steffen, W., and P. Tyson. Global change and the Earth system: a planet under pressure.
IGBP Science 4 (2004).
USEFUL WEBSITES
Organisations
r www.earthsummit.info/ is an excellent list of earth system science related websites.
r www.unu.edu/unupress/rio-plus-5.html is the site of the UN University (UNU).
r www.unep.org/ is the site of UNEP.
r www.undp.org/mdg/ is the site of the UNDP.
r www.fao.org/wfs/index en.htm is the site on the 1996 World Food Summit.
r www.who.int/en/ discusses the activities of WHO.
r www.globalpolicy.org/ is the Global Policy Forum (UN-based and related to Security
Council).
r www.iucn.org is the site of the International Union for Conservation of Nature (IUCN).
r www.iisd.org is the site of the International Institute for Sustainable Development (IISD).
r www.wbcsd.org is the site of the World Business Council on Sustainable Development
(WBCSD).
r www.worldenergy.org/wec-geis/ discusses the activities of the World Energy Council
(WEC).
r www.globelaw.com/ is about international environmental law and agreements.

r www.eoearth.org/ is an Internet encyclopedia with brief treatments of many for sustain-
ability science relevant topics.
Reports and Conventions

r www.theecologist.info/key27.html gives access to all the papers in the Blueprint for Sur-
vival 1972 issue of The Ecologist.
r www.millenniuminstitute.net/resources/elibrary/index.html contains models and papers
of the Millennium Institute, including the Global 2000 Report to the President published
in 1980.
r www.un-documents.net/wced-ocf.htm gives online access to the 1986 report Our Com-
mon Future of WCED.
r www.johannesburgsummit.org/html/documents/wehab papers.html is the site with info
on the World Summit on Sustainable Development in Johannesburg in 2002.
r www.un.org/millenniumgoals/ is a site with information on the Millennium Development
Goals (MDG).
r www.undp.org/mdg/basics.shtml is the site of the Millennium Development Goals
(MDGs).
r publications.worldbank.org/ecommerce/ is an overview of the World Bank’s publications.
r hqweb.unep.org/geo/index.asp is the site of UNEP’s Global Environmental Outlooks
(GEO).
r https://www.cia.gov/library/publications/the-world-factbook/appendix/appendix-c.html
is the site where you can find the CIA World Factbook with a long alphabetical list of
international environmental agreements.
r www.cites.org is the site of the Convention on International Trade in Endangered Species
of Wild Fauna and Flora (CITES).
r www.cbd.int is the site on the Convention on Biological Diversity.
r www.unccd.int is the site of the UN Convention on Combat Desertification.
Global Change Statistics and Sustainable Development Indicators

r www.homethemovie.org/ is the site of the movie Home by Yann Arthus-Bertrand. It is
the Story of Creation retold in the age of science, with a compelling call to stop destroying
and become sustainable.
r www.iisd.org/publications/pub.aspx?id=607 is an initiative of the International Institute
for Sustainable Development (IISD) to construct and maintain a database on sustainable
development indicators.
r gcmd.gsfc.nasa.gov/ on global change indicators is the Global Change Master Directory,
operated by the Goddard Space Flight Center, with info on Earth science data and
services.
r www.ciesin.org is the site of the Center for International Earth Science Information
Network with a vast amount of online available data on natural and social phenomena.
r www.lgt.lt/geoin/ is a site on geo-indicators, that is, magnitudes, frequencies, rates, trends
of geological processes and phenomena.
r www.epa.gov/ is a site of the US Environmental Protection Agency (EPA) with data with
brief descriptions on ozone depletion, climate change and other topics.
r www.gapminder.org is a site with vivid graphical tools to explore a large set of data.
hdrstats.undp.org/en/tables/default.html is the site of the Human Development Index
(HDI), operated by UNDP. On the site hdr.undp.org/en/statistics/ Human Development
Reports can be downloaded and HDIs are given for countries and years.
r en.wikipedia.org/wiki/Human_Development_Index#2008_statistical_update is a wiki
site that contains HDI-data for the last few years and an extensive list of other indicators.
r data.worldbank.org/data-catalog is the site with access to large datasets, for instance on
the World Development Indicators (WDI).
Useful Websites 145
r www.worldbank.org/html/opr/pmi/envindic.html is a site on environmental performance

indicators.
r www.inequality.org/facts.html features indicators and data on income inequality.
r mdgs.un.org/unsd/mdg/ gives an overview of Millennium Development Goals (MDG)
indicators.
r esl.jrc.it/dc/ is an interactive dashboard to explore the MDGs, with a data-based assess-
ment of the MDG goals for all countries and for separate years.
r www.beyond-gdp.eu/download/bgdp-ve-gpi.pdf and dieoff.org/page11.htm report on the
Genuine Progress indicator (GPI), an attempt to come up with a more adequate measure
of progress and well-being.
r www.foe.co.uk/community/tools/isew/templates/storyintro.html is the interactive
Friends of the Earth website on Measuring Progress, with an explanation of the Index of
Sustainable Economic Welfare (ISEW) and the ISEW 1960–2000 as compared to GDP
for a number of countries; it also offers a sensitivity analysis for the ISEW for the United
Kingdom.
r www.isis.csuhayward.edu/alss/geography/mlee/geog2400/2400ISEW.htm is time-series
for the Index for Sustainable Equitable Welfare (ISEW) for several countries.
r www.neweconomics.org/projects/happy-planet-index is a website on The Happy Planet
Index. An index of human well-being and environmental impact.
r www.footprintnetwork.org/en/index.php/GFN/ is the site of the Global Footprint Net-
work (GFN), with an extensive briefing on the footprint definition and results and the
option of an interactive evaluation your own footprint at ecofoot.org/.
r www.ecologicalfootprint.com/ and www.bestfootforward.com/footprintlife.htm are sites
where you can calculate your proximate ecological footprint.
r www.waterfootprint.org/?page=files/home is the site of the Water Footprint Network
(WFN), along the lines of the global footprint network and definitions and results about
water footprint, virtual water trade and other topics.
r wwf.panda.org/about_our_earth/all_publications/living_planet_report/living_planet_
report_graphics/ is a site operated by the World Wildlife Fund (WWF), with links to
GFN and WFN.
r www.uu.nl/SiteCollectionDocuments/Corp UU%20en%20Nieuws/triple p monitor.pdf
gives a description of the Triple P Monitor sustainability maps for the Netherlands.
r www.fwrgroup.com.au/isismethod.html is a site that introduces various methods to con-
struct and explore interactively in a group (indicators for) sustainable development (ISIS
methodology).
6 Quality of Life: On Values, Knowledge
and Worldviews
6.1 Introduction
The scientific worldview cannot give meaning in itself to our lives and it cannot
resolve the ethical questions surrounding sustainability issues. The scientific ‘facts’
about the world, important and accurate as they may be, have to be complemented
with what people value and believe. It is time to have a look at the more subjective,
personal side of the quest for sustainable development. Are there empirical data and
theoretical concepts about the subjective side of sustainability?
Previously I have stated that sustainable development is about quality of life. But
what is quality of life – are we merely shifting the problem? Sustainable development
is to act here and now in such a way that the conditions for a (decent/high) quality of
life elsewhere and later are not eroded. But for whom and for how long? Throughout
history, individuals have struggled to realise their idea of ‘the good life’, by exploiting
environmental opportunities and cooperating with and oppressing others. Since the
dawn of civilisation philosophers have reflected on what ‘the good life’ entails. What
can we learn from them?
Evidently, quality of life relates to our needs in a broad sense and to the means
to satisfy those needs. What we experience as and conceive of as needs are related
to our values. These, in turn, are interwoven with our beliefs. This chapter investi-
gates the notions of needs, values and beliefs in order to discuss worldviews as a
way to organise the complex field of (un)sustainable development. It is meant as a
theoretical framework that is applied to a variety of issues in subsequent chapters. I
do not claim scientific rigour or completeness. The objective is to provide practical
context and guidelines.
6.2 Quality of Life and Values
6.2.1 Needs and the Quality of Life

The most widely used definition of sustainable development uses the word needs
twice (§5.3). Most discourses on development are about needs, wants and desires.1
1 Often, the words ‘needs’, ‘wants’ and ‘desires’ are used more or less as synonyms. They are indeed
close in daily language. Desire is usually associated with personal impulses and longing. Wants
146
Quality of life results from the fulfilment of needs. It is part of perennial wisdom that
those needs have material and immaterial aspects. In science, this insight has been
rediscovered and associated with the needs hierarchy proposed by Maslow (1954).
He argued, partly as a rebuttal of behaviourism, that people behave and use their
resources in such a way that their physiological needs (food, shelter and so on.)
are met first. When these ‘basic needs’ are sufficiently satisfied, people will orient
themselves towards higher needs: safety, belonging and being loved. Even higher
on the needs hierarchy are esteem and self-actualisation.2 Or in Bertold Brecht’s
famous phrase: ‘Erst kommt das Fressen, dann kommt die Moral . . . ’. Maslow’s
theory is a holistic approach that offers the prospect of unlimited opportunities for
personal growth, provided that basic needs can be fulfilled.
The different ways in which the notion of needs is approached reflect this hier-
archy. Basic needs are defined almost exclusively in terms of nutritional require-
ments by organisations such as the Food and Agriculture Organization (FAO). Food
deprivation, or in common language, hunger, are quantified in great detail and in
relation to the location, gender, age and so on of the individuals concerned. Similarly,
the World Health Organization (WHO) focuses on the prevalence of various diseases
and its determinants. The extensive statistics on the state of health of people provide
numerous ‘objective’ indicators such as the number of HIV/AIDS patients, access to
medical services and safe water and expenditures on health education. All these data
suggest the possibility to define and measure quality of life in the sense of (conditions
for) the fulfilment of basic needs for food, shelter, safety and bodily health.
Yet, even the ‘lower’ physiological needs have less tangible and measurable
aspects. Food requirements are reflecting genetic and climatic circumstances and
too little or too much food can affect quality of life negatively. The prospect of
starvation or disease may diminish quality of life as much as an actual food shortage
or illness. Even for the ‘lower’ needs, then, an objective definition of needs and of
their fulfilment or deprivation in relation to quality of life is difficult. For the ‘higher’
needs, in a complex society where needs are constructed increasingly in interaction
with social exchange and technology, it is even harder.3
Economic science ‘studies human behaviour as a relationship between ends and
scarce means, which have alternative uses’ in the 1935 description of Robbins. In
economic textbooks, quality of life is addressed primarily as a decision problem of
an individual person, who must choose between alternatives in a situation of limited
resources and opportunities. Alternatives to satisfy needs – or ends – are ranked
in order of their utility for the individual. The rational choice is to satisfy those
needs that maximise utility/benefit or minimise regret/cost for a given income. No
normative statements are made about neither ends nor outcome, because ethics is
declared to be outside the domain of (welfare) economics.
There are two objections to this narrow view of quality of life. First, people’s
behaviour is often not according to the presumed economic rationality. It involves
are also seen as rather personal. Needs refer to more universal aspects of life. Therefore, I will
consequently use the word need.
2 How strict people adhere to such a hierarchical sequence is a matter of dispute. Immaterial needs
as expressed in religion, art, culture and trade are existential, too (James 1902).
3 The notion of mimetic desire has been introduced by Girard to point out that needs and desires are
socially constructed, notably by imitation.
148 Quality of Life: On Values, Knowledge and Worldviews
risk and probability assessments, contextual judgments and social imitation and
comparison. Second, people’s quality of life depends on more than the goods and
services one can buy with money. Therefore, some economists prefer the more
comprehensive notion of well-being over welfare. But usually, individual income or
expenditures expressed in monetary units is used as the prime indicator of quality of
life. An example is the notion of a poverty line – the $1.25 a day per person of the
World Bank – which is established from the expenditures needed to buy a basket of
commodities for survival. Adding indicators of work and living conditions from, for
instance, the World Development Indicators gives a more complete picture.
Economists are not denying the social and other aspect, but argue that people’s
‘revealed preferences’ as expressed in their purchases are more trustworthy than the
‘stated preferences’.4 Conversely, social scientists emphasise that quality of life is
more a ‘subjective’ socio-cultural construct than an ‘objective’ state. ‘Human needs
and wants are generated, articulated and satisfied in an institutionalised feedback
system. They do not appear from thin air but are created by the social interactions
that comprise the civic community’ (Douglas et al. 1998:259). It is, therefore, prob-
lematic to construct a quality-of-life index from statistical data at an aggregate level
and to engage the individuals whose quality of life is assessed only indirectly via their
economic decisions. To overcome this problem, economists are analysing people’s
behaviour in more depth and on the basis of surveys and experiments.5
A clear choice in favour of an experiential, subjective approach to quality of
life is made in the expanding field of happiness research. People are asked directly
whether they are happy and content with their life or not – the Subjective Well-Being
(SWB) approach (Veenhoven 1991). Happiness is then described as ‘the subjective
enjoyment of one’s life as-a-whole’ and correlated with quality of life. The previously
mentioned World Values Survey (Inglehart and Welzel 2005) and the European
Values Study are broader but use a similar approach.6 This and other research
shows that for the United States and the United Kingdom, the fraction of happy
people did hardly change over the last half century despite a threefold increase
in income. The scientific literature suggests seven factors that really matter for an
individual’s experience of ‘being happy’: family ties, financial situation, work, social
environment, health, personal freedom and philosophy of life, in order of importance
(Layard 2005). Income plays only a partial and indirect role. This finding, though,
may be valid for European and American society only, and even there it may change
because of, for instance, a growing sense of crisis or number of immigrants.
Summarising, there is evidence that well-being is a function of the ‘object-
ive’ social-economic situation (income, family size, education level, health situ-
ation and so on) but also of subjective experiences of a person’s situation (health,
job, marriage, community, religion and so on). The objective approach is means-
oriented and tends to focus on observable resource opportunities and constraints.
4 An example is a construction model of health indices across countries, in which functional limitations,
self-reports of health, and a physical measure are interrelated to construct health indices (Meijer
et al. 2011). The authors find that ‘health indices correlate much more strongly with income and net
worth than self-reported health measures’.
5 Much research is done in emerging fields like behavioural and experimental economics, where game
theory, laboratory experiments and simulation models are combined (Gintis 2005).
6 The results are usually presented at the country level, which may be misleading because subgroups
within countries – for instance, young people or people living in large cities – may differ more from
other subgroups than from similar subgroups in other countries.
Box 6.1. Measuring utility. Economists disagree on how to measure utility.

Some consider utility to be immeasurable and an ordinal utility function the best
possible way to describe the ordered preferences of an individual. Others think
that a utility function is more than merely an abstract concept and that an empir-
ical, cardinal utility function can be constructed (van Praag and Frijters 1999).
The issue becomes even more difficult if one takes interactions among individu-
als, items without a market price, uncertainty and intergenerational aspects into
account. Because there is more to life than money, economists have invented the
contingent valuation method to assess the value of those parts of utility that have
no market price. Presuming the existence of fixed and individual preferences,
people are asked to indicate how much money they are willing to pay in order to
maintain a particular (environmental) situation, or to accept as a compensation
for its loss. The outcomes are called stated preferences and are different from the
revealed preferences, which are derived from how people spent their money. This
approach can, at least in theory, account for the non-market aspects of quality of
life. One of the critiques is that respondents give answers that are strategic or in
other ways biased and do not consider real-world constraints.
The subjective approach is ends-oriented with a focus on experienced quality of

life. The means-oriented approach acquires a certain objectivity at the cost of
excluding the experiential aspect. The ends-oriented approach has to be derived
from questioning individuals with the associated methodological problems. It seems
attractive to combine both.7 Thus, quality of life can best be evaluated with objective
and subjective indicators. Let us have a brief look at two approaches that combine
both.
6.2.2 Capabilities and Satisfiers

Not only realised but also non-realised options contribute to the experience of ‘the
good life’. It is good to know that there is medical help in case of an accident,
police in case of robbery and the countryside in case of stress. This is captured in
the Capability Approach (Nussbaum and Sen 1993). Capabilities are what a person
might wish and is capable to achieve. They represent his opportunities and the set
of options from which he can choose. They reflect the potential to fulfill human
needs and the freedom to choose which ones to fulfill. A person’s set of capabilities
depends on the means (income, resources and so on), as in a utilitarian or welfare
approach, but also on personal characteristics (creativity, for example) and on social
and environmental arrangements (community, access rights and so on) (Figure 6.1).
A realised option is called a functioning: something a person manages to do or be in
leading a life. The capabilities of a person thus reflect the alternative combinations of
functionings that he or she can realise. Poverty is seen more as capability deprivation
than income deprivation.
The capability approach connects the subjective experience of a good quality of
life – freedom to choose ends – with the objective resource-oriented aspects – means
to realise ends. An unstable climate, polluted drinking water or a corrupt police force
7 Both are admittedly anthropocentric – but who can speak on behalf of (some) other living beings?
Figure 6.1. Possible causal direction from resources, capabilities and subjective well-being
(Robeyns and Van der Veen 2007).
confine a person’s capabilities and limit his or her freedom of choice and, therefore,
the quality of life he or she can realise. It has an intuitive plausibility in concrete
situations. For instance, the existence of a public transport system is a capability:
Every citizen may decide to use it or not. Therefore, capabilities are connected to
alternative options that can be provided in different ways and in relation to their
public or private character (government, markets, and so on), as shown in Table 5.1.
But operationalisation of the concepts of capabilities and functionings is difficult.
An attempt at concrete implementation is the formulation of basic needs and the
use of the available statistics to explore the extent of satisfaction (Nussbaum 2000;
Table 6.1).
Another approach to quality of life is the Human Scale Development (HSD)
theory (Max-Neef 1991). It emerged in the 1980s in Latin America in response
to postwar ‘developmentalism’ and monetarist neo-liberalism. It starts from the
postulate that development is about people and not about objects and that the
Table 6.1. Quality-of-life indicators: from objective-material to subjective-immaterial. The rows

indicate, in ascending order, kewords associated with the various aspects of quality of life. The lowest
row indicates the source
Transcendence Control over Ethical

one’s orientation
environment
Freedom Play Psychological Political
needs freedom
Identity Other species Reproduction Gender
equality
Sahasrara Creation Affiliation Coexistence Community
(Crown) life
Ajna (Throat) Idleness Practical reason Adaptability Job security
Vishuddha Participation Emotions Security Political
(Throat) stability
and security
Anahata (Heart) Understanding Interacting Senses, Freedom of Family life
imagination, action
thought
Manipura (Solar Affection Doing Bodily integrity Effectiveness Material well
Plexus) being
Svadhisthana Protection Having Bodily health Subsistence Health
(Sacral)
Muladhara Subsistence Being Life Existence Climate and
(Base) geography
Chakra wheels Max-Neef Max-Neef Nussbaum Bossel (1998) Economist
(1991) (1991) (2000) (2005)
best development process is one that allows the greatest improvement in people’s
quality of life. Quality of life depends on the possibilities people have to adequately
satisfy their fundamental human needs. These needs are understood as a system
of interrelated and interactive needs without a hierarchy. There are many possible
classifications, and none is final or ultimate.8 Existential needs can be categorised in
the four classes of being, having, doing and interacting. Along with the preceding
three are axiological needs (Table 6.1). Human needs are met in a dynamic process
of social construction of satisfiers, on the one hand, and physical means, on the
other. A satisfier is anything that contributes to the actualisation of human needs.
Satisfiers include space, social norms and practises, and organisational and political
structures.9 They render needs historical and cultural. Economic goods are their
material manifestation.
Several classes of satisfiers can be distinguished. Pseudo-satisfiers are items or
behaviours that give a false sense of satisfying an axiological need, usually via propa-
ganda, advertisements and other forms of persuasion. Examples are certain medical
treatments (‘placebo’) for subsistence and protection; status symbols and chauvinistic
nationalism for identity; and formal democracy for participation. Inhibiting satisfiers
satisfy a given need excessively and at the expense of other present and future needs
satisfaction. They usually originate in customs, habits and rituals. Examples are eco-
nomic competitiveness in the name of freedom, which often inhibits the needs for
subsistence, protection and affection, or religious fundamentalism providing identity
at the expense of understanding and freedom. Some actions destroy the possibility
to satisfy needs in the name of another need. For instance, the arms race, national
security doctrines and censorship and bureaucracy claim to satisfy the need for pro-
tection, but often destroy the need for subsistence, affection, participation, identity
and freedom. Conversely, increasing regulation due to crowding and congestion is
often experienced by the individual as a reduction in capabilities and thus in quality
of life.
Social scientists have long been aware that needs and their satisfaction influ-
ence other needs in terms of effectiveness and over time. In his book Social Limits
to Growth (1977), Hirsch observerd that the need for and value of certain items
and behaviours are desired and valued for their very scarcity (‘positional goods’).
Economists use the notion of reference drift for the phenomenon that the needs of
an individual influences and is influenced by the needs of others (van Praag and
Frijters 1999; van Praag and Ferrer-i-Carbonell 2004; Layard 2005). Needs and their
satisfaction also change over time as a consequence of habituation, new experiences
and knowledge and novel goods and services. This phenomenon, called preference
drift or hedonic treadmill, means that a person needs an increasing income in order
8 This differs from the needs theory of Maslow in which a hierarchy is hypothesised. Max-Neef and
colleagues suggest that only the need of subsistence, that is, the need to remain alive, overrules other
needs (Max-Neef 1991:17). The matrix of needs should be constructed by individuals participating
in a group; experiments with groups in Argentina, Bolivia, Great-Britain and Sweden have been
performed and led to significant variations.
9 There are also dissatisfiers: those rules, structures, etc., that prevent the fulfilment of needs. They
may come from oppressive power structures, but they may also stem from increasing regulation
because of crowding and congestion. Experienced by the individual as a reduction in capabilities
they are, if well designed and implemented, the least harmful for the total of individuals. It explains
the perennial attempts at escape from such dissatisfiers.
to remain satisfied because he is, in retrospect, less happier with the income gain
than he had expected. It is a growth-promoting feedback loop. The elusive character
of more income and subsequent consumption is one of the drivers of ever-increasing
consumption (Jackson 2009). Table 6.1 summarises categories that have been pro-
posed in order to come to grips with the notion of needs in relation to quality of life.
They have different origins, but all contain the idea of ‘lower’ and ‘higher’ needs.
6.2.3 Values and Their Measurement

Quality of life is linked to what people value. The U.S. Academy of Sciences identified
in its report Our Common Journey: A Transition Toward Sustainability (1999), four
grand values in people’s collective aspirations: peace, freedom, development and
environment (Leiserowitz et al. 2006). Values about what is to be sustained and what
is to be developed were listed and have been crystallised since then in more concrete
form in, amongst others, the Millennum Development Goals (MDGs). But what are
values? The notion of value is a complex one and definitions abound in the social
science literature, mostly suggesting that values express a belief about a desired end,
which guides individual action (Dietz et al. 2005; Hitlin and Piliavin 2004). This book
uses the following definition (Aalbers 2006):
a value is a prescriptive conviction about desirable behaviour and goals, in particular
in a longer-term perspective.
Values tend to change only slowly, at both individual and societal levels, although
sudden changes (catharsis, revolution) cannot be excluded. Values can develop and
be expressed freely upon reflection – at least, this is the case in most societies and
situations and appreciating such freedom can be regarded an important value in itself.
Values are manifested in the choices people make among different alternatives and
are also an expression of their moral principles.
Around both personal and societal values, there is often a tension between a
desired situation and valued behaviour – an ideal – and the actual situation and beha-
viour. The ideal is usually a more or less shared standard or norm within a group, to
which individuals may aspire and thus respond to positively, but that in practise is
only partially or not at all realised. Modern-times advertising is a good example of
the construction of consumerist ideals and its subsequent exploitation. Sustainable
development as a collective ideal is another example. These ideals are values crys-
tallised in ethical codes of conduct and, as such, an important cultural dimension.
There is a vast social science literature on this topic. The brief discussion merely
serves as a background to the conceptual framework presented later in this chapter.
There are various ways to identify and measure values. Most widely known and
applied is the survey approach, in which people are questioned directly about their
values. The ethnographic method makes use of in-depth interviews.10 But because
it is labour-intensive, the survey approach dominates empirical research on values.
Many surveys are organised to find out what people value in life, in the public domain
as well as in the private sector. The data are used for evaluation of social change and
policy and for targeted marketing. One finding is that preferences for technology
10 Laboratory experiments is a third method. Such an experimental approach gives interesting new
insights, particularly with the progress in ICT, but it has limited external validity and generalizability.
Conservative
[non-worldly] ideals
Caring faithful (13%)
Conservatives (17%)
Socially minded (11%)
Collective Individual
Other Balanced (28%) Self
Hedonists (8%)
The World My world
Broad minded (7%)
Materialists (7%)
Professionals (9%)
Progressive
[worldly] achievement
Figure 6.2a. Value orientations constructed from the answers of respondents in the Dutch
population. The two major axes are identified as rather robust, but the names for the extremes
are somewhat ambiguous (adapted from Vringer et al. 2007).
and frugality correlate well with a ‘me’ versus ‘we’ dimension. The tension between
the individual and the collective manifests itself in questions about the environment
as a public good. For instance, a large majority of EU citizens say to make efforts to
protect the environment but doubt the effectiveness as long as others (individual and
corporations) do not do the same. In the World Value Survey, a majority says that
spending money on themselves and their families is one of life’s greatest pleasures,
but nearly as large a majority agrees that consumption threatens human cultures
and the environment, that less emphasis on money is a good thing, and that gaining
more time for leisure activities or family life is their biggest goal in life (Leiserowitz
et al. 2006).
As part of the Sustainability Outlooks at the Netherlands Environmental Assess-
ment Agency (PBL), extensive value surveys were organised between 2003 and 2006
in order to investigate the relationship between people’s values, beliefs and beha-
viour. Data on several thousands of Dutch citizens have been analyzed over longer
periods of time and ranked in a multi-dimensional value space. Using factor ana-
lysis, eight value clusters or value orientations have been identified (Aalbers 2006).
On the basis of interviews and in combination with datasets on socio-economic vari-
ables (age, gender, income and so on), the value clusters have been given an ‘identity’
and a name: Caring (14%), Conservatives (15%), Hedonists (10%), Luxury Seekers
(10%), Business-like people (8%), Cosmopolitans (9%), Engaged people (13%) and
the middle group of Balanced people (21%). The last category includes people who
are not aware of or do not express an outspoken position. A proximate distribution
of the Dutch population over the clusters is shown in Figure 6.2a.
To give: achieve for others

Benevolence
Relations
Other-oriented Stagnation
Self-transcendence Conservation
Social-oriented Conformism
We Order
Big, unlimited world Small, limited world

Universalism Conformity
Freedom Family life
Progress Self oriented

Openness to change Self-enhancement
Find your own way Individual oriented
Freedom Me
To take: achieve for self
Achievement
Performance
Figure 6.2b. Value orientations, based on the survey outcome in Figure 6.2a, in the broader
setting of value survey outcomes.
The value space can be analyzed in terms of pairs of opposing values. A prin-
cipal component analysis suggests the existence of two dimensions: the contrast
giving (upper) versus taking (lower) and the contrast small world (right) versus big-
world (left) (Figure 6.2b). Correlations between value positions and socio-economic
variables such as age, gender, income and education turn out to be weak, albeit
in the expected direction (Aalbers 2006).11 The vertical dimension is close to the
contrast religious versus worldly. The horizontal axis can be associated with an ori-
entation on the own local community versus the world as a whole or, from a more
individual stance, as the contrast between self and the other(s). It also reflects the
tension between a more egocentric and a more social attitude.
Within this framework, Caring people value immaterial aspects of life and Lux-
ury Seekers value the material ones. Conservatives share an immaterial oriented
lifestyle with the Caring people, but tend to focus more on the local ‘own’ situ-
ation. On the opposite side, one finds Cosmopolitans and Business-minded people,
for whom the money, travel and opportunities of the Big World are important
ingredients of quality of life. Hedonists are small-scale and self-oriented, in combin-
ation with a materialistic attitude. Engaged people are their opposites, with a political
and idealist perspective and a focus on global-scale issues. The totality reflects the
variety of value orientations existing in the population as a whole.
Comparative research on the basis of different and more extensive sets of values
suggest that the two dimensions are universal in the sense of context-independent
11 On the website of the Dutch public survey company, TNS-NIPO, you can do an interactive session
that will tell you to which value orientation your worldview belongs. The questionnaire can be found
at the site www.tns-nipo.com.
Box 6.2. Characterisation of worldviews. Since the term sustainable develop-

ment came into being, people have expressed different interpretations (§5.3).
Four divergent images or views of nature, man and technology can be charac-
terised with keywords: Technocrat-adventurer, Manager-engineer, Steward and
Partner (Table 6.2; de Vries 1989). There are no strict boundaries, of course. The
difference between the views in essence on how one sees man’s position in Nature
and on the degree of control man can or may have about his own fate. Note that
a person’s view on science and technology is one of the crucial elements, because
technology is to such a large extent a linchpin between the possible and the prac-
tical, between heaven and earth. Words like technocrat, steward and partner,
have (temporarily?) gotten out of fashion. However, other words with similar
connotations take their place.
and trans-situational.12 The relative position of countries, however, is found to be

different. Comparison with other empirical surveys indicate plausible correlations
with properties such as lifestyle, self-control and egoism. For instance, Business-
minded people are most active and in control of their lives, whereas Caring people
score highest on affection and group-orientation. To find our way in this complex
mix of values, beliefs and behaviours and its relation with social-cultural change,
I introduce two frameworks for analysis. The first one is the Cultural Theory, a
social science theory that offers a pluralist framework of individual and collective
perspectives in life.
6.3 The Cultural Theory
6.3.1 Stories from the Himalayas and Bali13

Himalayan villagers parcel out their transactions with their physical environment to
four distinct solidarities, each of which is characterised by a distinct management
style. Agricultural land, for instance, is privately owned whilst grazing land and
forests are communally owned. But grazing land and forests do not suffer the ‘tragedy
of the commons’ because transactions in their products are under the control of a
commons managing institution. Villagers appoint forest guardians, erect a ‘social
fence’, that is, a declared boundary, not a physical construction, and institute a system
of fines for those who allow their animals into the forest when access is forbidden
or take structural timber without first obtaining permission. If the offender is also a
forest guardian, the fine is doubled. If children break the rules, their parents have
to pay up. There is fragmentary evidence that such regimes have occurred since at
least 600 years ago.
12 The Rokeach Value System is the most widely known (Rokeach 1973). People are asked to rank
two sets of eighteen values each, one on end values and one on instrumental values. The values
are portrayed along two dimensions: self-enhancement versus self-transcendence and openness to
change versus conservation (or traditionalism) Schwartz and collaborators (1994) have extended the
method into the Schwartz Value Survey, which lists fifty-six items to be rated by respondents on a
scale between 7 and –1. The list with the ten values suggested by Schwartz to have validity across
all cultures: hedonism, power, achievement, stimulation, self-direction, universalism, benevolence,
conformity, tradition and security.
13 The story on Himalayan villagers is taken from Thompson in de Vries et al. (2002).
Table 6.2. Sustainable development according to four worldviewsa
Technocrat-Adventurer Manager-Engineer Steward Partner
Computer machine Garden ‘wilderness’

Pioneer planner Caretaker participant
‘frontier economy’ ‘mature ecosystem’
competitve hierarchy cooperative solidarity
Exploitation control Management harmony
courage-creativity order care adaptation
struggle planning frugality sprituality
Anthropocentric ecocentric
power-over-others power-over-yourself
(technological) (material) (just) (spiritual)
progress prosperity welfare development
science fiction religion
technopolis ecotopia
engineering sciences ‘shallow’ ecology ‘deep’ ecology
economics
a de Vries 1989.
Although they may seem informal and lacking any legal status, these commons
managing arrangements work well in the face-to-face setting of a village and its phys-
ical resources. Drawing on their ‘home-made’ conceptions of the natural processes
that are at work – their ethnoecology – the forest guardians regulate the use of these
common property resources by assessing their state of health, year by year or season
by season. In other words, these transactions are regulated within a framework that
assumes, first, that one can take only so much from the commons and, second, that
you can assess where the line between so much and too much should be drawn. The
social construction inherent to this transactional realm is that nature is bountiful
within knowable limits. It is the myth of Nature Perverse/Tolerant (Table 6.3).
In recent years, when forests and grazing lands have suffered degradation for
a variety of reasons, not the ‘tragedy of the commons’, villagers have responded
Table 6.3. Characterisation of the four cultural perspectivesa
Worldview: Individualist Hierarchist Egalitarian Fatalist
Nature is: Benign perverse, tolerant ephemeral capricious

People are: self-seeking malleable: deeply caring and sharing inherently
flawed but untrustworthy
redeemable
Resources are: infinite because they finite and to be finite and to be what can be
are dependent on managed managed with grabbed
human genius efficiently based respect and
on control and frugality
knowledge
Fairness is: who puts in most rank-based what matters is non-existent
gets most out distribution by equality of result,
trusted, not opportunity
long-lasting
institutions
a Thompson et al. 1990.
by shifting some of their transactions from one realm to the other. For instance,
they have allowed trees to grow on the banks between their terraced fields, thereby
reducing the pressure on the village forest, and they have switched to the stall-
feeding of their animals, thereby making more efficient use of the forest and grazing
land and receiving copious amounts of manure, which they can then carry to their
fields. In other words, transactions are parceled out to the management styles that
seem appropriate and, if circumstances change, some of those transactions can be
switched from one style to another.
Because they are subsistence farmers whose aim is to remain viable over gener-
ations – rather than to make a ‘killing’ in any one year and then retire to Florida –
their transactions within their local environment can be characterised as low risk,
low reward. However, during those times of the year when there is little farm work to
be done, many villagers engage in trading expeditions or in migrant labour in India.
Trading expeditions are family-based and family-financed, and highly speculative:
high risk, high reward. Therefore, a farmer’s individualised transactions, when added
together over a full year, constitute a nicely spread risk portfolio. The attitude here,
and particularly at the high-risk end of the portfolio, is that ‘fortune favours the
brave’, ‘who dares wins’ and ‘there are plenty more fish in the sea’. Opportunit-
ies, in other words, are there for the taking. The idea of nature here is optimistic,
expansive and non-punitive: Nature Benign. Thus, the farmers have various ways to
accommodate their interests.
On the island of Bali, there is an intricate relationship between rice cultiva-
tion and social organisation (Reader 1988). Wet rice cultivation is labour-intensive,
because mechanisation is difficult to implement. It also has a high and lasting pro-
ductivity, mainly because of the benefits of controlled flooding on the nitrogen
balance (Lansing 2006). In this situation, a form of social complexity has arisen
with a strict caste-related hierarchy and an important role for priests, who derive
authority from high standards of behaviour. Such hierarchies can be a strong, moral
organizing force in society (Dumont 1984).
Yet, Balinese society also has distinctly egalitarian features: ‘The Balinese tend
to dislike and distrust people who project themselves above the group as a whole,
and where power has to be exercised they tend to disperse it very thinly . . . For the
Balinese there is no difference between a person’s spiritual and secular life . . . The
devils and demons [in religion] . . . provide forceful reminders of the obligation to
restrain selfish instincts for the benefit of social harmony . . . Rice cultivation is the
ultimate expression of the Balinese readiness to follow the edicts of some greater
authority’ (Reader (1988) 65). The complex religion of the Balinese is an example of
the functional significance of religion in human ecology. One is also reminded of the
role that monks and monasteries have played in the periphery of large civilisations.
The more individualist outlook on life is left to the trading communities, often for-
eigners. Nowadays, the rapidly growing tourist industry provides new opportunities
for the population.
These field observations indicate that groups of people represent enough
diversity to follow a variety of strategies in the face of different threats and oppor-
tunities. It is in this context that social scientists propose institutional arrangements,
that do not exclude any of the voices while accepting contradictory defintions of
problems and solutions and encouraging constructive argumentation. Such ‘clumsy
Figure 6.3. The four perspectives identified in Cultural Theory (clockwise): the hierarchist,
the egalitarian, the individualist and the fatalist (based on Thompson et al. 1990).
institutions’ is one of the solutions for sustainable resource management (Verweij

and Thompson 2006).
6.3.2 The Four Perspectives of Cultural Theory

One way to understand the diversity and complexity in social-ecological systems is
through the framework proposed in Cultural Theory (Thompson et al. 1990, 1992,
1997). It offers an interesting heuristic for an analysis of the controversies and risks
about sustainable development issues such as novel technologies, climate change and
biodiversity preservation (Schwartz and Thompson 1989; Adams 1995; Rayner and
Malone 1998). Combining insights from cultural anthropology and ecology, Cultural
Theory distinguishes four perspectives on how people experience and interpret
reality, and organise themselves and manifest rationality, solidarity and fairness.14
These perspectives – one can also speak of visions or myths – are categorised along
two axis (Figure 6.3):
r the degree to which individuals behave and feel themselves part of a larger group
of individuals with whom they share values and beliefs (the group axis), and
r the extent to which individuals are subjected to role prescriptions within a larger
structural entity (the grid axis).
The resulting four perspectives are related to their position along these two axes:
the hierarchist (high on both), the individualist (low on both), the egalitarian (high
14 Other, similar categorisations have been proposed, and I do not claim to be complete. For instance,
an investigation of business cultures across the world also identified similar distinctions (egalit-
arian versus hierarchical, adventurous/enterprising versus anxious/xenophobic, individualist versus
collectivist, masculine versus feminine and short-term versus long-term oriented) (Hofstede 1984).
in ‘group’, low in ‘grid’) and the fatalist (low in ‘group’, high in ‘grid’).15 Whereas
the hierarchist favours control and expertise in order to guarantee stability within a
world of limits, the individualist is convinced that nature has inherent stability and
abundance. The egalitarian emphasises the fragility of nature and the possibility of
irreversible destruction. The fatalist experiences the world as determined by pure
chance. The differences are graphically depicted in Figure 6.3 with the mechanical
metaphor of the rolling ball. Let us have a closer look.
Each perspective can be associated with a certain way of structuring the rela-
tionship between man and nature and between fellow men, each with characteristic
judgements about quality of life and technological and environmental risks. Seen
from the individualist perspective, the past is full of human success stories, the future
full of promise. The debate about ‘limits to growth’ is seen as exhibiting a lack of
trust in the ingenuity of humans and in the resilience of nature. It is precisely the
existence of limitations that challenges people to surpass themselves, as happened so
often in the past. Individualists see markets and prices as the only efficient and fair
organisational scheme and have confidence in the human capacity for adaptation.
Prominent and cherished values are competitiveness, success, achievement, cour-
age, adventurism and its rewards of wealth and power. Individualists may, however,
come into conflict with their own convictions. As soon as successful new areas are
exploited and new markets are conquered, a loss of exclusiveness and the necessity
of coordination and regulation emerge to avoid the worst excesses.
The individualist. ‘False bad news about population growth, natural resources, and
the environment is published widely in the face of contradictory evidence . . . Bad
news sells books, newspapers, and magazines . . . the cumulative nature of expo-
nential growth models has the power to seduce and bewitch’, according to Julian
Simon (Science 208, 1980). ‘The ultimate constraint upon our capacity to enjoy
unlimited raw materials at acceptable prices is knowledge. And the source of
knowledge is the human mind. Ultimately, then, the key constraint is human
imagination and the exercise of educated skills.’
Many administrators, managers and engineers have confidence in the results of

measurements and risk calculations and have concern for as well as confidence in
good economic and environmental management. They are hierarchists. On the basis
of what experts say about the constraints and the costs and benefits, an optimum
technical-economic course is determined. Seen through the eyes of the hierarchist,
chaos and anarchy are just around the corner and the world is perpetually in need
of management. This is the world of bureaucrats and politicians who understand the
need for and subtleties of institutions, committees, and negotiations. They appreci-
ate and support legitimacy of big government and big business and their procedural
rationality. Stability, reliability, integrity and planning skills are appreciated. But
they too face the shortcomings and contradictions of their view. The hierarchical
organisation is intent on securing its own continuity; hence, it is not easily mobilised
for policies that go beyond accepted risks or challenge the status quo. The inherent
15 The fifth worldview, that of the hermit, is beyond the area defined by these two axes. The emphasis
tends anyway to be on the three ‘active’ worldviews: hierarchist, individualist and egalitarian.
delays and buffers resulting from procedures and consultations may have a stabilising
effect, but they also tend to paralyse commitment and creativity and cause cynicism
and loss of credibility. Also, in their struggle for greater power, hierarchist organ-
isations come across limits in the form of corruption, suspicion and indifference.
The hierarchist. The Commission on Global Governance, in their report Our

Global Neighbourhood (1991), addresses a variety of problems related to unpre-
cedented change on a global scale such as the New Arms Race, Widespread
Violence, Persistent Poverty, and Environmental Threats. It observes that due to
globalisation of capital, media and technology, ‘states retain sovereignty, but gov-
ernments have suffered an erosion in their authority. In the extreme case, public
order may disintegrate and civil institutions collapse in the face of rampant viol-
ence’. They continue: ‘The world needs to manage its activities to keep the adverse
outcomes [of environmental impacts] within prudent bounds and to redress cur-
rent imbalances’ (p. 11). A solution is offered, ‘The concept of global governance:
There is no alternative to working together and using collective power to create
a better world’. The United Nations should play a leading role here.
The third perspective is the egalitarian one which is characterised by communitarian

aims and values and absence of structured authority. Nature is perceived as a fragile,
connected, complex whole. Human knowledge of the earth is limited. The basic
premise is equity between man and nature and among people, as expressed in the
image of people as stewards or partners. A fair distribution of benefits and costs
between rich and poor – the context for socialism – and between us and posterity – the
context for environmentalism – is considered a precondition for sustainability, and
so are justice and security. Egalitarians are risk-averse and favour the precautionary
principle as management strategy. As with the other perspectives, egalitarians face
limitations in maintaining theirs. They always run the risk of individualism and
hierarchy arising within their ranks. The solidarity and dogmatism of their radical
positions is often accompanied by a loss of creativity and effectiveness. They need
a future with environmental disasters and a present peopled by acquisitive and
powerful enemies in order to keep the group together.
The egalitarian. In her book Staying Alive, Vandana Shiva (1989) describes devel-
opment as a new project of Western patriarchalism. It is rooted in reductionist sci-
ence: ‘As a system of knowledge about nature or life reductionist science is weak
and inadequate; as a system of knowledge for the market, it is powerful and prof-
itable’. The paradigm of Western science is itself the problem: ‘ . . . development
is a permanent war waged by its promoters and suffered by its victims . . . and sci-
entific knowledge, on which the development process is based, is itself a source of
violence . . . this transformation of nature from a living, nurturing mother to inert,
dead and manipulable matter was eminently suited to the exploitation imperative
of growing capitalism’. Modern science destroyed the old metaphor: ‘The nur-
turing earth image acted [no longer] as a cultural constraint on exploitation of
nature’ (p. 7).
Many of the world’s poor are fatalist in their outlook on life. They try to cope with
everyday life’s contingencies within the limited freedom of manoeuvre that they
have. People and nature are both capricious, and all one can do is to make the
best of it. Climate change and other environmental threats only get significance
for them if they are actually hit by the consequences. Fatalists are passive and yet
part of the fourfold scheme, because they are needed by the adherents of the other
perspectives. They provide legitimacy for the hierarchists as potential members of
the lower organisational ranks. For the individualist, they are the necessary com-
plement to their success and give the American dream its appeal. And throughout
history, they are again and again the potential converts and followers for egalitarian
movements.
Cultural Theory claims that the interpretation of past events and the anticipa-
tion of future events are filtered through the rationalities and solidarities associated
with the four perspectives. Religious and market fundamentalism, ethnic and nation-
alist movements, and environmental radicalism can all be understood against this
background. Of course, people rarely express these perspectives in their extreme
form and one should not give in to the temptation to make caricatures. In fact, one
of the critiques of Cultural Theory is that real-world individuals manifest contex-
tual behaviour and that it is unclear how adherence to a particular perspective in a
particular context develops. Also, the relation between perspective and power mat-
ters. The resources at the disposal of those who adhere to and support a particular
perspective influence the degree to which they can change the world in a desired
direction. Thus, it is important to distinguish the perspective, that is, how one thinks
the world works, and the management style, that is, how one would act if one were in
power.16
Cultural Theory interprets social-cultural change from the continuous, dynamic
interplay between the adherents of the four perspectives (Thompson 1992, 1997).
Individuals alter their perspective when it is no longer reconcilable with their experi-
ence. Collective institutional change happens whenever larger groups of people start
to doubt the correctness and adequateness of the dominant perspective. Such change
may be triggered by surprise experiences, new information, a desire for balance and
by other – for instance, demographic – changes. Therefore, extreme positions are
counteracted by reactions from one or more of the other perspectives. As a result,
societal development unfolds dialectically. Some of the switches between perspect-
ives are indicated in Figure 6.4. Excessive hierarchism leads to bureaucratisation and,
at some point, the system collapses and liberalisation and privatisation processes take
over. Excessive individualism leads to marginalisation of the less successful people
who then resort to fatalism, which in turn feeds egalitarian movements in a process
of radicalisation. The social-cultural dynamics cause societal oscillations in which
excessive swings to either side are corrected.
The four perspectives are a decor against which developments in the world can
be interpreted, not as law-like truths but as the manifestations of human beings in
their individual and social variety. As such, it is a qualitative, rich theory of complex
16 This distinction has been applied by running computer model with different sets of assumptions on
how processes are simulated and on which interventions are made (Rotmans and de Vries 1997; de
Vries 1998).
Figure 6.4. Two important mechanisms in the dynamics of social-cultural change: the inter-
action between hierarchist (king, state) and individualist (merchant, market) in the form of
bureaucratisation and privatisation, and between egalitarian (sect) and fatalist (commoner)
in the form of radicalisation and marginalisation (based on Thompson 1992).
adaptive systems. For instance, in recent times, the excessive hierarchism in the
former Soviet Union ended with the collapse of the collectivist state in 1989–1991.
Growing support for the individualist perspective, as proclaimed in Fukuyama’s
book The End of History and the Last Man (1992), led to a victory of neoliberalism.
It accelerated financial deregulation, abolishment of trade barriers and selling out of
nationalised firms to globally operating private capital funds. In combination with
the Information and Communication Technology (ICT) revolution, the unfairness
of global capitalism started to generate counterforces that culminated in the Seattle
and Genoa revolts against globalisation and the 2001 terrorist attack on the World
Trade Center in New York. In the new century, millions of less successful, excluded
and marginalised individuals shed their fatalism and resort to radicalism and reform
in egalitarian movements. Some are fundamentalist and (ultra)nationalist, others
environmentalist or modernist in their outlook. The tensions between hierarchy
and individualism and the connections between fatalism and egalitarianism are a
Box 6.3. Mr. López the entrepreneur. There is a story about an entrepreneur
that illustrates the shift in prevailing perspective. One way to sustain an egalitarian
regime is with ceremonies in which those who accumulate inordinate amounts of
land or other property are compelled to give it away. However, a smart individual
can always bypass such a system and create structural change.
In the 1890s, Mr. Lopez subverted the old mechanism of wealth levelling in
the Mexican village of San Juan Guelavı́a. Traditionally, only rich people would
be designated as major-domo, that is, the one who has to sponsor for the village
fiestas and festivals and gains prestige from it. However, with help of the clergy,
Mr. López forced the village council to designate less wealthy people – who then
could not refuse – and offered the insolvent sponsors a loan with their land put up
for collateral. By the eve of the Mexican Revolution, Mr. López owned most of
the community’s best land: By 1915, his large family owned 92.2% of the arable
land and an even larger part of the irrigated land and, with the support of the
church, strongly opposed any censure. Thus, equal access to strategic resources
had disappeared by the perversion of a traditional regulatory mechanism – an
example of the vulnerability of egalitarian cooperative regimes to individualist
interference.
recurrent theme in the political-economic history of Europe and its offshoots. In

the postindustrial and globalising 21st century, its cultural dynamic can be expected
elsewhere, too.
As in many social science theories, the statements of Cultural Theory are hard
to test and validate or falsify empirically. It is applied in energy policy, management
of hazardous wastes, household consumption styles and other areas, and provides a
novel and inspiring way of looking at sustainability issues. Its greatest contribution is
to offer a consistent way to uncover the value-laden construction and interpretation
of ‘facts’ and to reveal the inherent value pluralism and hence uncertainty and rich-
ness in ‘solution space’. Incorporating cultural perspectives into the sustainability
discourse clarifies the diversity and ambiguity in the use of the word sustainable
and the clashes between people – think of the controversies about animal rights,
forest clearance for new roads, genetically modified organisms (GMOs) and sub-
sidies for renewable energy to mention a few. In the words of Thompson (1990), ‘A
fourfold vision is the key to understanding the contradictory visions of the future
that are generated by our social interactions . . . Single visions – economic man, sci-
entific rationality, any consciousness that has to insist that all others are false – are
pathological’. It points at the necessity of building alliances that bridge different
values and beliefs and at the merits of pragmatism: ‘My view is that, if environment-
alists aren’t willing to engage with big business, which are among the most powerful
forces in the modern world, it won’t be possible to solve the world’s environmental
problems’ (Diamond 2006: 17). In other words, we should avoid a battle between
utopians and their claims to a particular view and implementation of sustainability.
In the next section, I introduce another framework to operationalise such diversity
and pluralism and apply it in the thematic chapters later on.
Worldview pluralism (
C Stefan Verwey 1978).
6.4 Worldviews: Ways to See the World
6.4.1 Values and Beliefs: Four Worldviews

As dicussed, it is possible to get a rough idea about people’s values by questioning
them. A person’s values in association with particular needs gives rise to certain
expectations about the person’s ideas and behaviour – for instance, in relation to
development, environment and sustainability. If someone prefers nature walks over
air travel or being employed in a multinational corporation rather than in an envir-
onmental NGO, one suspects a link with his value orientations. But the links are
complex. The nature lover flies to an African wildlife park and the MNC-employee
is active in a human rights NGO in the evening hours.
I do not treat the many social and cognitive science theories in this field and
simply introduce our own definition of the notion of worldview:
A worldview is defined as a combination of a person’s value orientation and belief

system, that is, of his view on how the world is best understood.
Therefore, a worldview combines a person’s interpretation of and beliefs about the

outer and inner world and his value orientation.17 A worldview has a prescriptive
17 I use the word ‘beliefs’ to indicate that people build up cognitive structures according to which they
interpret the world around them. I use it more or less as an equivalent to interpretation, model or
and a descriptive aspect (Rokeach 1973). The prescriptive part is about the relevance
of certain things and themes over others, that is, the value orientations previously
discussed. They are individual, subjective and imply a certain, preferred quality of
life. The descriptive part consists of the belief system about how things work. It is a
mental map or model, that is, a more or less causal interpretation of the world. It can
be rather objective in the sense of shared and formalised. It frames how people think
they can improve or maintain an aspired quality of life. In practise, prescriptive and
descriptive aspects are interrelated, as previous examples show.
One can try to identify and measure beliefs in the same way as values, by
questioning people. Indeed, it is difficult to distinguish the value part from the belief
part in the outcomes of social surveys because values are embedded in cognitive
mental processes. Some researchers have tried to extract belief systems from answers
to questionnaires or behaviour in participatory workshops. An example is an analysis
of the outcomes of the World Value Survey (§4.2) over the period 1989–2004, in
order to identify popular support for the transition to a market economy in India
and China (Migheli 2010). They posed questions about the role of competition,
government intervention, income inequality, hard work and fairness. The questions
were about values (such as, do you consider competition harmful?) as well as beliefs
(such as, are larger income differences an incentive for economic growth?). The
answers may imply a causal relation as, for instance, in the belief that competition is
good because it provides an incentive to work hard or the conviction that hard work
is irrelevant because one’s success depends on luck and connections.18
It was found that Indian people are, on average, less prone to support the char-
acteristics of a free and competitive market and blame poverty less on laziness and
lack of willpower of the poor than the Chinese. In both countries, there is a change
towards preferences for a less competitive market. Perhaps, the author suggests, this
is a reaction against the policies currently undertaken by the governments, because
more competition means more uncertainty as compared to a strongly regulated or
centralised economy. It is also found that people expect the government to inter-
vene in order to insure a minimum level of well-being for the population and that
the preference for a market economy weakens when people have confidence in the
government.
A second example is about a participatory exercise organised in New Zeal-
and, with the aim to make a shared conceptual mental model of the health system
(Cavane et al. 1999). Such a shared model, in the form of a causal loop diagram
(CLD), is supposedly helpful in finding solutions for persistent problems (Morecroft
and Sterman 1992; Vennix 1999). Two groups of staff from the Ministry of Health
participated in a series of workshops, during which clusters of issues were identified
and subsequently translated in a conceptual model. One group consisted of policy
managers, the other of clinicals (medical/health practitioners). The authors – and
mental map. Some people prefer the word mindset or, in a scientific context, paradigm. I do not use
words like attitude, norms or motivation. They are close to values, but it is outside the scope of this
book to treat such refinements.
18 The author considered also the role of socio-economic variables, which are known to influence
preferences: religion, education level, city size, age, gender and the state of health Besides, the
different histories of India and China since the 1950s and the uncertainties of the present transition
will also play a role.
most participants – found the CLDs useful in illustrating the roots of different
worldviews, for instance, whether one looks at the issue from a micro- or a macro-
perspective. The exercise illustrates the different ways in which individuals and
groups understand their responsibility and tasks: ‘The policy group looks to the
effectiveness of the whole sector, while the clinicians look to the delivery of a ser-
vice’ (Cavane et al. 1999). It was evident that the values of people as expressed in
their jobs are linked to their professional backgrounds and the beliefs and models
acquired during education and work.
The beliefs that guide behaviour are based upon a lifelong experience with par-
ents, neighbours, community leaders and institutions and all kinds of formal and
informal education. It is a mixture of ‘personal knowledge’ and ‘community know-
ledge’. The assumption of economic rationality is overly simple. Clearly, behaviour
can be made more ‘rational’ by a more informed evaluation of the costs and benefits.
But these ‘costs’ and ‘benefits’ are assessed not only in terms of money, but also
of prestige, relationships, new or lost opportunities and so on. Even if we have an
idea about a person’s values and beliefs, or worldview – for instance, about how
to achieve a more sustainable energy system or to stop the destruction of tropical
forests – one can expect only a probabilistic relation with such a person’s actions.
Worldviews are not predictors of behaviour. Almost identical behaviour may have
different backgrounds. Sometimes there are valid reasons why behaviour cannot be
consistent with the worldview.19
Thus far, worldviews have been associated with individual persons. When a large
majority of people adheres to a certain worldview, it gets social and historical sig-
nificance. This is manifested in philosophical enquiries and historical events. An
elementary and ancient idea is that the human mind tends to operate by thinking in
opposites. Many religions and myths describe creation as the separation of primor-
dial substance into opposites: light and darkness, heaven and earth, male and female.
Modern cosmologies reiterate a similar view. The opposites are not merely mental
images but are a fruitful source of psychic energy (Jung 1960). Table 6.4 summarises
some of the great philosophers’ investigations in terms of opposites. The two main
dimensions that came out of social surveys (Figure 6.2) show a remarkable similarity
with the two dimensions found in philosophical works. Combining the two gives the
four worldviews that are depicted in Figure 6.5.
The vertical dimension is associated with immaterial and spiritual values on the
upper side and with material and secular values on the lower side. It is the opposition
religious versus worldly, giving versus taking and mind versus matter.20 This vertical
dimension can be traced back to the ancient Upanishad scriptures in India and to
the writings of Plato. It reflects the controversy between realism and nominalism
in medieval Europe, and it is, in the contrast between idealistic and materialistic,
found in the work of the 19th-century philosopher Hegel. The sociologist Sorokin
(1957) uses the words Ideational and Sensate for it. This vertical dimension is also
recognised in the quality-of-life elements in Tables 6.1 and 6.2.
19 Often, the reasons given are invented or imagined because the respondent does not give priority to
the implied behavioural change or cannot resolve underlying conflicting values and ideas.
20 It is less relevant here whether the spiritual orientation is seen as an a priori and pre-existing quality
‘from above’ or as a quality that is emerging ‘from below’ as part of human evolution. There is room
for both views.
Table 6.4. Interpretation of the vertical and horizontal axis in Figure 6.2b
and 6.5a
Source universal ↔ particular
Surveys (Fig 6.2) Big World Small world

The others The self
Weber Collective Individual
Sorokin One single truth Pluriform truth
Universalism/Active Singularism/Ascetic
Steiner Objective reality Subjective understanding
Pauli/Jung Non-locality; Wave Locality
Particle
Levinas/Wilber The Other I
We
Schwartz Group values Individual values
Kant Laws Maximes
Rules for personal behaviour
Source immaterial ↔ material
Surveys (Fig 6.2) Giving Taking
Religious Worldly
Plato Ideas Senses
Mind Body
Hegel Idealistic Materialistic
Sorokin Ideational Sensate
Steiner Spirit Matter
Jung Archetype–spiritual Instinctive-material
Wilber Interior Exterior
Fromm Being Having
a Van Egmond and de Vries 2011.
mental / spiritual
immaterial
Absolute Subjective
Idealism Idealism
B1 B2
universal particular
A1 A2
Modernism: Postmodernism:
Objective Subjective
Materialism bodily Materialism
material
Figure 6.5. Representation of the four worldview quadrants as combinations of the value
survey outcomes (Figure 6.2) and philosophical reflections (Table 6.4). The letters A1, A2,
B1 and B2 refer to the names of scenarios that are discussed in Chapter 15.
Box 6.4. Beliefs and scientific knowledge. Beliefs about the world is knowledge.
How do we acquire it and what makes it legitimate? Regarding knowledge, there
is a gradual change from skills, as the ability to act according to rules, to knowing
how, which implies skills-in-social-context, and from there to expertise, which
includes the even higher abstraction level of not only knowing but also reflecting
and adjusting the rules. Empirical-reductionist science is mostly about expertise
and relies on a universal scientific truth, which can be uncovered. But in the
worldview frame, science cannot claim the Truth nor the method to uncover it.
Complex system science is causing a paradigm shift in the Modernist worldview,
partly thanks to the new tools of computers and Internet. The path of inner
reflection of, for instance, Gurdieff’s self-remembering and recounted in personal
accounts of the beyond are also legitimate ways to find truth. It has a long history
and many appearances and leads according to its practitioners to a holistic science
(Wilber 2001). It represents an aspect of the vertical dimension in Figure 6.5.
Therefore, the world of humans is not only one of plural values, but also of plural
beliefs about how to acquire knowledge. It explains much of the present-day
debates about the legitimacy of scientific claims.
The horizontal dimensions can be given a broadly philosophical interpretation,

too. Depending on the scale level, it is the opposition we versus I, collective versus
individual, global versus local, universal versus particular, and whole versus fragmen-
ted. This tension between ‘The World’ and ‘my world’ is also reflected in the group
axis in Cultural Theory, where the hierarchist is positioned opposite of the indi-
vidualist (§6.3). Each of us has an individual existence, yet we are in innumerable
ways connected to the larger material and mental world around us. The left big-
world side is interpreted as shared existence in the collective realm. The right small-
world side is associated with the non-shared existence in the individual realm. For a
collectivity to exist, a certain level of uniformity in values and beliefs is required.
In contrast, the differentiation of individuals creates and sustains diversity of value
and beliefs. From an epistemological point of view, the horizontal dimension is,
therefore, about whether a person adheres to an objective, universal truth or lives
primarily a subjective, personal truth.
6.4.2 Worldviews in History

If worldviews can be given an historical interpretation, how would they look? Let
us look again at Figure 6.5. The upper right quadrant is labelled Subjective Idealism,
because it represents combinations of non-materialistic and individual orientations.
Needs are satisfied by mastery of the sensate needs, and there is a diversity in the
interpretation of truth. The material world is perceived as an incarnation one has
to be liberated from. Nature is experienced as divine mystery. Big government and
corporations are distrusted, science is considered unacceptably rational. In concreto,
the focus is on personal growth, on the ‘management of greed’ and on self-reliance
and conflict resolution at the community level. The Gandhian concept of swaraj is
an expression of this worldview. Another icon of this worldview is Schumacher’s
book Small is Beautiful (1973), originally titled Buddhist Economics, which makes a
plea for small-scale solutions and individual responsibility: ‘I have no doubt that it is
possible to give a new direction to technological development, a direction that shall
lead it back to the real needs of man, and that also means: to the actual size of man.
Man is small, and, therefore, small is beautiful’ (Schumacher 1973). Its darker side
are the fundamentalist sects that embrace the single subjective truth of the leader.
In the upper left quadrant, Absolute Idealism, a majority in society adheres to
the view that there is a universal objective truth. This Truth is not to be uncovered
by personal introspection or the rational mind, but should be understood from the
scriptures and consolidated in churches and religious prescripts. Its essence is imma-
terial and collective. The institutions are hierarchist and demand order and stability,
based on uniformity and obedience. Churches and their dignitaries are visible and
outspoken representatives. Also, the State is an expression of this worldview inasfar
as governments work for the well-being of their citizens, as reflected in, for instance,
the Declaration of Human Rights. The demarcation between church and state, theo-
cracy and democracy, is not sharp – and continuously in dispute. Widespread poverty
and ecological degradation are according to this worldview caused by egoism and
moral permissiveness and the remedies range from repentence to prayer and ‘good
works’. In a more worldly, political context, these ills are considered a consequence
of oppressive structures and lack of social solidarity and are to be resolved by more
adequate rules and institutions.
The diagonal opposite of the Subjective Idealism worldview is Objective Mater-
ialism or Modernism. It has become the dominant worldview in European and
American ‘Western’ society, at least among the elites. It combines a belief in the
existence of a universal objective truth with a material-secular value orientation and
cosmology (§5.2). The world can be understood and managed according to scientific
and utilitarian principles and with the societal objective of ‘the greatest happiness for
the greatest number’. In the Theory of Modernism, it is said that: ‘At the level of the
individual, modernization is characterized by an increased openness to new experi-
ences, increased independence from authority, belief in the efficacy of science, and
ambitions for oneself and one’s children’ (Easterlin 1983; cf. §4.3). The convergence
in the science- and technology-driven modern world towards a uniform playing field
and political systems is well described in Friedman’s The World Is Flat (2005). The
European welfare state but also capitalism, socialism and communism are manifest-
ations of Modernism, although with divergent interpretations of rights and duties
towards the collective. The underlying idea is that each individual person has a
vote in what the collective ends should be (democracy). Academia, government
organisations and corporate bureaucracies are the institutional representatives.21
The scientific and technological success in Modernism is not only the root of much
progress but also of many sustainability problems, and yet its advocates point at
science- and technology-driven innovations as the main source of solutions.
The fourth worldview combines a material-secular orientation on life with a
focus on the individual. It is the world of Subjective Materialism or Postmodernism.
The belief in universal truths and the collective becomes a widening array of convic-
tions, lifestyles and material manifestations of personal identity: ‘ . . . in postmodern
times we are moving out of an age of uniformity, collectivity and universality and
21 Corporations wish a ‘level playing field’ in the form of uniformity and standardisation on the global
market. Small and medium enterprises are often in a different, and dependent, position in this arena.
Box 6.5. Universalism as a hierarchist utopia. Universalism is an attempt of the

mind to control the world. Universal knowledge, in contrast to knowledge from
practical experience, tends to simplify the world in order to make it manageable.
This can have dramatic consequences. ‘Radically simplified designs for social
organization seem to court the same risks of failure courted by radically simplified
designs for natural environments. The failures and vulnerability of monocrop
commercial forests and genetically engineered, mechanized monocropping mimic
the failures of collective farms and planned cities’ (Scott 1998:7). This observation,
inspired by anarchism, reflects the evils of radicalism to the left side in Figure 6.5.
The state and other hierarchical institutions can easily become obsessed with their
utopias and become the enemies of human well-being. It has not disappeared with
the fall of communism because ‘large-scale capitalism is just as much an agency of
homogenization, uniformity, grids, and heroic simplification as the state is, with
the difference being that, for capitalists, simplification must pay’ (Scott 1998:8).
The challenge is to recognise the value in locality and diversity and to balance it
with the genuine merits of universal insights, procedures and social institutions.
into one characterized by individuation, fragmentation and difference’ (Hall et al.

1996). Postmodernism is also an expression of feelings of anxiety and insecurity:
There are no anchors for meaning and consolation, neither in the collective to the
left, nor in the soul above. Sustainability issues are narrowed down to issues of indi-
vidual survival, with a fatalist touch and a local or national orientation. Huntington’s
book Clash of Civilizations (1997) is one of the books that identified a shift towards
Postmodernism in the 1990s.
The tensions between the diagonal opposites in Figure 6.5 are reflected in philo-
sophy and history. The opposition between upper left and lower right manifests itself
in the rejection on the one side of the physical, notably the sexual, aspect of human
life by the church – for instance, of Darwin’s thesis that men descended from apes –
and extreme forms of hedonism and body culture on the other. The diagonal from
upper right to lower left mirrors the opposing views of science, with its emphasis on
observation and falsification on the one side, and the religious convictions of indi-
viduals practised in esoteric and mystery schools on the other. The four quadrants
together form the playing field where the prevailing values and beliefs of a people
and a period are manifested. The scheme is meant to make sense out of complex
phenomena, not as a comprehensive Theory of Everything.22
If the preceding, somewhat caricatural, positions are not only individual but also
social realities, it should be possible to interpret history in terms of the four world-
views and, more importantly, to use it as a framework to explore sustainable
22 The worldviews can be associated with the different forms of goods and services provision (§5.3).
Markets claim to be a universal allocation mechanism, primarily in the material domain (A1–A2),
where prices can be attached to tangible goods and services. Public and professional provision is
largely oriented towards collective goals, in governments and corporations (A1–B1). Self-provision
and informal provision are by choice or by necessity an orientation towards the individual needs
(A2–B2). Each mode has its preferred goods and services and its specific rules and dispositions.
development pathways (Van Egmond and de Vries 2011). Certainly, it should be

possible to trace the extreme manifestations in history. Let us make a small detour
and examine this hypothesis for Europe. Early Christianity can be positioned in the
Subjective Idealism upper right quadrant. It emerged in reaction to the preceding
Roman sensate and ‘carpe diem’ culture of the lower right worldview. When more
individual people got inspired by it, early Christianity got more followers and was
forced to institutionalise – it became the Church of Rome. The personal and mystical
was replaced by uniformity in religious beliefs. The 14th-century Arab thinker Ibn
Khaldun made a similar observation about historical change (Turchin 2003; §3.5).
In his view, change comes from small human groups that have a small territory but
a large collective solidarity, or asabiya as he calls it. This gives them a large capacity
for collective action and the ‘ethnic momentum’, shared conviction and willpower
to conquer other peoples and to expand. In the process, their ideas become more
widespread and are consolidated and claimed to be universal. At that point, society
has arrived in the upper left quadrant.
Throughout the European Middle Ages, the worldview of Absolute Idealism was
manifest in the dogma’s of the Roman Catholic Church. The great cathedrals with
their orientation towards God and made by innumerable anonymous devotees are,
literally and metaphorically, its highest expression. The physical world and body
were repressed, either by sanctifying ascetism or by ruthless persecution. Truth
was radically monopolised and deviant convictions were forced underground. The
repression led to the social discontinuities of the Reformation and the religious wars
of the 16th and 17th centuries. Small groups of committed Protestants were at the
root of a new inspiration, that transformed into the rationale and ethics of Euro-
American capitalism according to the sociologist Weber. Other cultures and religions
have experienced similar developments – for instance, the periodical repression of
sufism in Middle-East Muslim countries.
The outward expansion in the direction of the upper left quadrant reached its
zenith with the global exploration by the then-leading Catholic countries Portugal
and Spain. However, the seeds of other orientations on the world had already entered
Europe with the translation by Arab thinkers of the work of Aristoteles and other
Greek philosophers and were reinforced when Europeans came in contact with
previously to them unknown cultures outside Europe. The inquisitive minds of the
early scientists, with their rationalism and empiricism, created scientific theories
and technical inventions that initiated a shift towards a more secular universalism.
In the process, it eroded the religious claims. Modernism slowly emerged as the
new worldview that foreshadowed the great changes of globalisation and industrial
revolution in the 19th and 20th centuries. It provoked a number of conflicts between
Church and Science. The life of Galileo Galilei is one of the best-known illustrations.
By 1900, science and technology had become successful enough to leave in its turn
little or no room for diversity or dissent. It coincides with a value shift from traditional
to secular/rational as part of the industrialisation process (§4.3).
This emphasis on materialistic-secular values has increasingly manifested itself
as a denial of the more immaterial, spiritual-oriented worldviews. Social life became
organised in the uniform way of mechanised mass production, with rigid social classes
and uniform standards. Science took the place of religion as a source of author-
ity. Liberalism and socialism were both expressions of this worldview in 19th- and
Box 6.6. Sustainable development in Postmodernism. The transition in the

early 21st century from a science- and society-oriented, highly productive Mod-
ernist worldview to a sensate and individual oriented Postmodern worldview car-
ries the seeds of the unfolding social-ecological crises. Science and technology can
provide solutions for the great changes ahead. But a radical shift towards a Post-
modernist worldview erodes the necessary legitimacy of science and technology
to effectively introduce solutions. The proposed solutions reflect widely differ-
ent value orientations, as, for instance, in the debate about nuclear power and
GMOs, and continuous societal conflict results. Although scientists still adhere
to the ‘immutable laws’ of nature, their recommendations and solutions succumb
in ineffective and polarizing societal debates. The resulting incrementalism and
localism may fail to solve the sustainability problems addressed in this book. The
balance with the universal and the immaterial has to be restored if we are to find
a sustainable path for the world.
20th-century Europe. Romanticism was a reaction to it, with a temporary return to

the values of subjective idealism and their expression in visual arts and music, but
it lost momentum. As part of the changes towards a more secular worldview and
its manifestation in industrial capitalism and colonialism, there was a trend towards
uniformity and collectivity at the expense of the individual. State communism and
Nazi fascism in 20th-century Europe brought this worldview to its extremes, with
disastrous consequences.
In recent decades, Modernism is visibly making room for Postmodernism of the
lower right quadrant. The seeds for this change were already sown in the early 20th
century, when science discovered fundamental uncertainty and complexity and the
‘Brave New World’ turned out less utopian than expected. Its claim to universal truth
started to weaken in favour of social constructivism, a move that was reinforced
by the disenchantment with the great utopias of liberalism and socialism. In the
Postmodern worldview, there is room for diversity and pluralism and a shift in focus
from globalisation to the smaller scale of the individual and the local group. It
coincides with the transition to postindustrial values from survival to self-expression
(§4.3). The Internet, with its pros and cons, is becoming the ultimate expression of
this shift towards the right-hand side of Figure 6.5. Postmodernism has its dark sides,
too. Extreme manifestations are the hedonism and consumerism of the shopping
malls and the luxury elites, and the cynicism about other aspirations than the pursuit
of individual material goals. States increasingly lack the legitimacy to formulate and
implement societal goals in a complex and interconnected world. Fragmentation and
disintegration are one consequence, as in the falling apart of the former USSR and
former Yugoslavia and in the separatist movements in parts of Europe.
6.4.3 Mechanisms of Social Change

The historical interpretation of worldviews suggests several mechanisms of societal
change along the lines discussed previously for the Cultural Theory. One mechanism
of social change is a reinforcing feedback, which tends to drive the expression of a
worldview to its extreme (Figures 6.4 and 6.5). Initially, a worldview is supported
because of the original, authentic values and beliefs it proclaims. If more people
adhere to it and follow its interpreters, the ascendant worldview seeks legitimacy
by the introduction and consolidation of institutions that extend and rationalise its
domination. It is the period of growth and is a move to the left in Figure 6.5. Author-
ities, representing the new worldview and its institutions, reinforce the process.
People with less honorable intentions join and, gradually, other worldviews become
disrespected and suppressed. Evil behaviour unfolds from good intentions. Fun-
damentalist rabbis, priests and imams demand the faithful to live according to The
Book and avoid, convert or eliminate the non-believers. States and corporations con-
trol individuals by classification and standardisation, by administrative language and
procedures, and by political and managerial censorship. At its height, the imperial
powers of Rome or Britain were a dense conglomerate of vested interests – govern-
ment, army and business. ‘What’s good for General Motors is good for America’
was a characteristic statement during the United States industrial-military complex
at its peak.
The power and diffusion of ideas, in the upper two quadrants, is a powerful
societal force in itself. It is the force of messianic vision and tribal asabiya. Its worldly
equivalent is the spread of goods and skills in the lower two quadrants. It is the force
of efficiency, economies of scale and zero-sum game benefits from cooperation.
Products that are successful on the market or skilled labourers who start a successful
enterprise become in a positive feedback process large traders and corporations,
with the same processes of expansion, consolidation and standardisation as in the
formation of religions and churches.
Inevitably, the monopolisation of truth and of power creates counterforces.
Although these can be suppressed for long periods, with methods ranging from
torture to Internet censoring, at some moment in history they will erupt. It is the
period of decline. It is an action-reaction mechanism that heralds large-scale social
change. The more extreme the caricature of the worldview and the longer it is
sustained against these forces, the more violent and catastrophic the reaction tends
to be. One can distinguish two kinds of reactive forces against the legitimizing force
of the dominant worldview (Castells 1997). One is conservative, as it wishes to go
back to an imagined better past. The other is progressive in its embracement of
a transformation into an imagined better future. Castells labels the conservative
reaction as a resistance identity, which is taken up by those who are in devalued
positions and conditions and stigmatised by the logic of domination. The progressive
reaction is associated with a project identity and carried on by those who seek to
redefine their position in society and in the process seek transformation of overall
social structure. The green and feminist movements are examples of the latter,
while Islamic and Christian fundamentalism and nationalism in Eastern Europe are
expressions of the former.
It is individual persons who are behind the centrifugal, outward forces that turn
worldviews into their own caricature and cause their collapse. Why do individuals
move to one of the corners of the worldview scheme of Figure 6.5? One underlying
reason is the search for identity. To identify oneself with certain convictions and
associated social roles and material and immaterial benefits is a deep psychological
need. It also serves the innate desire for control. The benefits are immaterial, as with
Box 6.7. Values and ethics. Values are associated with good and bad, with ethics
and morals. Daly (1973) proposed a hierarchist spectrum of values, ranging from
ultimate worldly means to ultimate immaterial ends. Values have been linked
to ‘lower’ and ‘higher’ human needs (Maslow 1954; §6.2). In fact, all religious
teachings point at a value hierarchy, often in association with Good and Evil,
God and the Devil. In La Divina Commedia, Dante recounts his upward journey
through the realm of the afterlife: Hell, Purgatory, and Paradise. In the Bhagavad
Gita, Krishna talks about the three qualities of prakritri, the sanskrit word for
nature or the basic energy from which the mental and physical worlds take shape
(Easwaran (1985) 177–178):
‘Sattva – pure, luminous, and free from sorrow – binds us with attachment
To happiness and wisdom.
Rajas is passion, arising from selfish desire and attachment . . .
Tamas, born of ignorance, deludes all creatures
through heedlessness, indolence and sleep . . .
Those who live in sattva go upwards;
Those in rajas remain where they are.
But those immersed in tamas sink downwards.’
positions of status and power or getting protection and privileges, and material, in the
form of money and the sensate pleasures you can buy with it. In the left quadrants,
identity is strongly connected to a collective identity. To the right, identity is more
individualist, less connected and more fragmented. The centrifugal move is also a
flight from the center with its existential demands. The individual fears the freedom
inherent to the centre, where he or she has to take full responsibility for his or
her existence. Media, in its old and new appearances, often support the centrifugal
forces by paying attention primarily to extreme views rather than to reconciliation.
Of course, people mature during their lives and the social and cultural setting are
important determinants, too.
What does this have to do with sustainable development? I put forward as
a working hypothesis that sustainability can only be found in the middle, that is,
through an acknowledgement and synthesis of worldview pluralism. Then, the con-
ceptual framework of worldviews and their dynamics can help to identify the diverse
positions with respect to sustainability issues and to resolve the conflicting tenden-
cies. It is in line with the notion of ‘clumsy institutions’ advanced in Cultural Theory
(§6.3). In a policy context, it implies a search for robust solutions. In the thematic
chapters, this is clarified with concrete examples.
6.5 Worldviews in Action

In Chapters 7 and 9–14, I discuss a series of themes that are relevant for sustainability
science: biodiversity, demography, agriculture, water and minerals, and the economy.
Each chapter surveys the ‘state of the world’ in terms of historical trends and models.
In this pivotal chapter, I argue that the search for sustainable development involves
‘objective’ science and ‘subjective’ perspectives and I offer a conceptual framework
6.5 Worldviews in Action 175
Table 6.5. Possible descriptions of extreme positions in the quadrant space of Figure 6.3
Values achievement, income, security, individual well-being, social order, stable

competition, health, tradition, wisdom, and fair
adventure, (local) identity awareness, institutions, equity,
efficiency, cooperation, peace, human
rationality, utility compassion, rights
respect for nature
Motivation material success, material comfort, personal immaterial success,
worldly status worldly power, development, mental and
bodily pleasure inner peace spiritual status
Beliefs the world operates the world consists of ‘there is more the world operates
on corporate and impulsive, between heaven on collective and
state power emotional and earth . . . ’. shared, religious
hierarchies. individuals. Self-sufficiency in and/or social-
Global ‘open’ Be a realist; there material and cultural ideals.
markets and are no universal spiritual affairs in a Governments
innovations are truths and community setting primary task is
the way to opportunism is social justice and
progress and natural environmental
prosperity integrity
Indicators when does Chinese personal income, number of local state budget for
economy surpass (local) crime rate, shops and artisans, education and
the US economy? tax levels market share social security,
Top-500 firms . . . organic food church attendance
Policy stimulate support regional support and facilitate regulation and
mechanisms international suppliers and local small-scale international (UN)
markets; create markets; initiatives binding
global protectionism if agreements
competition on necessary for
level-playing field economic or
social-cultural
reasons
that allows exploration of the notion of sustainable development in its full width.
It is applied in the thematic chapters in the remainder of the book. At the end of
each thematic chapter, I do not present a single judgment, recipe or indicator for
sustainable development. Instead, I present tables with perspective-based statements
on how the situation is valued and interpreted from the different worldviews. You
as reader are invited to reflect on these statements on the basis of your own values,
beliefs and actions – and your own indicators.
In practice, each chapter ends with a representation of worldviews in the quad-
rant space in Figure 6.5, in the form of a table (Table 6.5). The black dots in Table 6.5
indicate four extreme positions. Each position manifests itself in an individual person
in the form of values, beliefs and actions. Most individuals are never fully or always
in one of the extremes. Positions will depend on age, gender, income, education and
so on. Therefore, the descriptions I give in the thematic chapters are best considered
incomplete caricatures. If a single individual were to hold such an extreme position,

it may influence the local situation, such as in a family or office, but not the outside
world. Such a person would provide a mooring point in the worldview space. It is a
useful exercise to find out which persons you view as representative of a particular
worldview.
If many individuals adhere to such a worldview, it does have an impact on the
larger world. It becomes a dominant view that is an active force in shaping human
society and history. I propose that, for a sustainable development path, the dominant
worldview should be a balanced ‘integral’ worldview. People should ‘include’ other
worldviews in their life. In contrast, if the dominant worldview in society is on or
over the edge in Figure 6.5, the situation is unsustainable. An intriguing question,
then, is which are the signals (‘indicators’) of an extreme position (‘overshoot’) and
which are the balancing forces to be expected (‘feedback’)?
The thematic chapters that follow are mostly about the material aspects of sus-
tainability. Is nature disappearing? How and why are human populations growing in
number? Are fertile soils degrading? Are groundwater and fisheries soon depleted?
Will there be an oil peak within a decade? Is chemical pollution making us ill? To
answer such questions requires a scientific assessment of the situation – but science
cannot give definitive answers in complex situations with large uncertainties. Neither
the individual, nor the collective can expect unambiguous recipes for action in a com-
plex world. Besides, to take action implies also certain goals (‘ends’) and allocation
of available resources (‘means’), implying choices that will not be value-free. To
be sure, science can give answers with near-certainty to many questions (§8.3). But
there are also many questions to which science has yet no or at best probabilistic
answers.23 As to sustainability issues, the individual person has to reflect on what
are for him or her the constituents of quality of life and what is valued in life. A next
reflection is on what is believed about the world, from the simple to the complex.
Each position suggests certain answers and behaviour. The search for a sustainable
world will be a play and a battle of perspectives – and you are part of it.
6.6 Summary Points

I consider sustainable development to be about quality of life. Against this back-
ground, the following points are important:
r Quality of life has an objective side and a subjective side, that are both part
of the expression and satisfaction of human needs. Food and health indicators
have rather strong objective components. Income often serves as a proxy for the
more subjective ‘higher’ needs.
r Concepts such as capabilities and satisfiers emphasise that needs, and with it
quality of life, are at least partly social constructs.
r Values are important in a person’s assessment of quality of life. They can be
identified with surveys. A rather universal set of value clusters or orientations is
found for individual persons.
23 In some forms of Postmodernism, the ignorance of scientists is abused, using scientific controversy to
defend certain interests or obstruct acquisition of relevant knowledge and transparancy See Proctor
and Schiebinger’s (2008) discussion on agnotology and the creation and use of ignorance.
r The Cultural Theory is one social science theory that explicitly looks at sus-
tainable development from different cultural perspectives with their own forms
of fairness, solidarity and rationality. It uses the dimensions of group and grid,
rooted in ecology and anthropology, to categorise four perspectives.
r Also, beliefs about how the world ‘works’ are important if one wishes to under-
stand how people see sustainable development. Values and beliefs together
make up worldviews, in my definition.
r Based on value surveys, discourse analysis and historical reflections, I propose a
two-dimensional framework to organise and apply worldviews in the analysis of
past and present developments and their (un)sustainability. The forces that give
rise to radicalisation and subsequent demise of worldviews can be understood
as a search for identity, power and control and associated feedback mechanisms.
r Sustainability problems can in this broader framework be seen as the one-sided
dominance of one worldview at the cost of the others. Their resolution is in
balance and synthesis.
People (and not only managers) only trust their own understanding of their world
as the basis for their actions.
—de Geus, in Morecroft and Sterman 1992
To be without some of the things you want is an indispensable part of happiness.

—Bertrand Russell, The Conquest of Happiness, 1968
. . . only a small portion of existing concrete reality is coloured by our value-

conditioned interest and it alone is significant to us.
—Weber 1904 [1949:76] quoted in Swedberg 2003
. . . the irresistible tendency to account for everything on physical grounds corres-

ponds to the horizontal development of consciousness in the last four centuries, and
this horizontal perspective is a reaction against the exclusively vertical perspective
of the Gothic Age. It is a manifestation of the crowd-mind, and as such it is not
to be treated in terms of the consciousness of individuals . . . we are at first wholly
unconscious of our actions, and only discover long afterwards why it was that we
acted in a certain way. In the meantime, we content ourselves with all sorts of ration-
alized accounts of our behaviour, all equally inadequate . . . we overestimate physical
causation and believe that it alone affords us a true explanation of life. But matter
is just as inscrutable as mind.
—C.G. Jung, Modern Man in Search of a Soul, 1933
. . . belief systems are not merely beliefs – they are the home of the ego, the home
of self-contraction. Even a holistic belief, like the web-of-life, always houses the
ego, because beliefs are merely mental forms, and if the supramental has not been
discovered, then any and all mental constructions house a tenacious ego . . . the web-
of-life is just a concept, just a thought. Ultimate reality is not that thought, it is the
Witness of that thought. Inquire into this Witness. Who is aware of both analytic
and holistic concepts? Who and what in you right now is aware of all those theories?
—Wilber 2000:95/117
When all things began, the Word already was. The Word dwelt with God, and what
God was, the Word was.
—John 1:1
SUGGESTED READING
An excellent comprehensive introduction into the philosophical background to this chapter.

Tarnas, R. The Passion of the Western Mind – Understanding the Ideas That Have Shaped Our
World View. New York: Ballantine Book/Random House Publishers, 1991.
A description of and simulation experiments with a world model, exploring future pathways for
the world depending on which worldview is dominant.
Rotmans, J., and B. de Vries. Perspectives on Global Change – The TARGETS Approach.
Cambridge: Cambridge University Press, 1997.
de Vries, B., and A. Petersen. Conceptualizing sustainable development: An assessment meth-
odology connecting values, knowledge, worldviews and scenarios. Ecological Economics
68 (2009) 1006–1019.
A series of short articles with opposing views on 21 environmental issues in the United States.
Easton, T. Taking Sides: Clashing Views on Environmental Issues. 11th ed., Contemporary
Learning Series. New York: McGraw-Hill, 2006.
Thompson, M., R. Ellis, and A. Wildawsky. Cultural Theory. Boulder, CO: Westview Press,
1990.
van Egmond, N., and B. de Vries. Sustainability: the search for the integral worldview. Futures
43 (2011) 853–867.
USEFUL WEBSITES
r changingminds.org/ is a site with access to disciplines, techniques and explanations of
social science topics.
r cep.lse.ac.uk/events/lectures/layard/RL030303.pdf is de weergave van de Lionel Robbins
lectures on happiness door Layard in 2003.
r www.tns-nipo.com is the site of TNS-NIPO where a questionnaire can be filled in to find
out about your own value orientation.
r ec.europa.eu/public_opinion/index_en.htm is the website of the European Commission
on public opinion research and in particular the Eurobarometer.
r www.worldvaluessurvey.org/ on the World Value Survey with extensive information on
the survey set-up, outcomes, and so on.
r www1.eur.nl/fsw/happiness/ is a database of literature on happiness research, operated
by Veenhoven of Erasmus Universiteit Rotterdam.
r www.sinus-sociovision.de/ and www.motivaction.nl are two other sites on measuring
values, mostly for marketing purposes.
r www.awakeningearth.org/ is the site of Duane Elgin on the role of consciousness in
envisioning a sustainable future.
r www.6milliardsdautres.org/ is a site with thousands of interviews with people all over the
world, who answer essential questions about their lives, such as what they value most in
life and what Nature means to them.
7 Energy Fundamentals
7.1 Introduction: The Essential Resource

Energy in the form of heat and work has always been crucial for human beings.
Wood usage as a source of heat goes as far back as the control of fire. Wind and
water have been used for millennia, but the most important source of mechanical
energy until a few centuries ago was the labour of domestic animals and human
beings (peasants and slaves). This all changed with the discovery of fossil fuels as a
source of high-density chemically stored energy.1
Coal was already known as a fuel in early dynastic China and ancient Rome.
It was used for heating in medieval Europe but was considered inferior to wood
because of its foul smoke. Moreover, it was harder to obtain in most places. Under
the pressure of wood shortage in 17th-century England, coal became a more common
fuel. Chimneys protected people from the immediate effects of smoke. By 1700, the
fraction of coal in energy use in the United Kingdom exceeded already 50 percent
(Figure 7.1). With the invention of the steam engine, its use got another boost. Like
the use of coal, the principle of the steam engine had been discovered long before,
in both the Chinese and Roman empires. At those early stages, no practical uses
were found for it. In 18th-century England, however, circumstances were highly
propitious. Initially, the steam engine was developed as a device for pumping water
from the coal mine shafts, which made the coal more easily accessible. In turn,
the cheaper coal was used to power other steam engines, some of them propelling
the ships and locomotives that transported coal to the growing number of users.
This further lowered the cost and expanded the market. In 1850, coal supplied over
90 percent of energy use in the United Kingdom. It was one of those positive
reinforcing feedback processes that ignited the industrial revolution. But more was
to come.
Following similar dynamics but faster, crude oil, and oil products like gasoline
and fuel oil, and later natural gas entered the energy scene in a series of supply and
demand innovations and expansions. These two other members of the hydrocarbon
1 The energy use density, that is, energy flow per unit of area, in agricultural societies depended on
the solar influx which is ∼1320 W/m2 at the top of the Earth’s atmosphere and at full insolation
∼1000 W/m2 at the Earth’s surface. Because insulation strongly fluctuates during day and year, it
has, on average, an energy density of about 250 W/m2 at the surface.
179
Muscle Firewood Traditional Wind Water

Coal Oil Natural Gas Electricity
100%
90%
80%
Share in total energy supply
70%
60%
50%
40%
30%
20%
10%
0%
1700 1750 1800 1850 1900 1950 2000
Figure 7.1. The energy transition in England/Wales 1561–1995. The graphs show the fuel
shares in total primary energy use (Warde 2005). Similar transitions took place in other
European countries and in the United States.
family were also known in antiquity. In the ancient Mediterranean world, bitumen
pools were known, but the substance was mainly used for construction. The Chinese
used bamboo to drill holes and transport gas and burn it to evaporate brine. Once
its potential as a source of energy was discovered in 19th-century America, its use
rapidly increased. It took oil less than eighty years to gain a 50 percent market
share in the United Kingdom (Figure 7.1). Worldwide, the same happened and by
1965 oil had a greater market share than coal (Figure 7.2). The fraction of biomass-
based traditional fuels, mostly wood and crop/animal residues, dropped worldwide
from 90 percent to 20 percent between 1850 and 1950, although its absolute use
kept increasing. These changes make up what has been called the energy transition.
Besides, coal, and later and at much larger scale, oil and gas, were and still are used
as feedstocks to produce, amongst others, cement, fertiliser and plastics. Official
forecasts expect the dominant role of fossil fuels to remain for the decades to come,
while traditional biomass-based fuels are increasingly replaced by modern ones. The
role of renewable sources for electric power remains modest in these forecasts.
The energy transition was in essence a technological transition. On the supply
side, it was mining and drilling, transporting and refining, and military support where
and when needed. On the demand side, there was a non-ending series of user appli-
cations like steam ships and trains, the automobile and electric power generation and
the myriad of electricity using appliances. Four out of the five long-term (Kondratiev)
waves in economic activity are directly linked to energy, namely, hydropower, steam
power, electrical power and motorisation based on oil.2
2 The fifth and last of these waves, called ‘computerisation of the economy’ by Freeman and Louca
(2002), also deals directly with energy, notably electricity, but at a more subtle, informational level.
Coal Oil Gas

Hydro Nuclear Other NonThElec
Trad&Mod.biofuels
80%
Share in global primary energy use
60%
40%
20%
0%
1850 1890 1930 1970 2010 2050
Year
Figure 7.2. The energy transition in the world 1850–2010. The graphs show the fuel shares in
global primary energy use (source: PBL). The expectations up to 2035 in the World Energy
Outlook 2010 are also indicated (IEA 2010). It should be noted that nuclear and other
nonthermal (NTE) options are counted in electricity equivalents.
Fossil fuel, the new source of energy, made the industrial regime fundamentally
different from the earlier agricultural regime (§4.5). Fossil fuels are accumulated
solar energy from a remote past, contained in geological formations and concen-
trated by geological processes in particular locations, from where they can be made
available through human effort and ingenuity. They have a high energy density,
unlike the diffuse flows of solar energy and its derived flows (waves, water, wind),
and are not directly connected to the flow of solar energy, unlike plants and animals
and wood. Our food is now as much or more derived from oil than from sunlight.
The fossil fuel incorporated in an ‘industrial’ tomato is several times its calorific food
content.
In advanced industrial societies, humans now have control over energy flows at
levels up to 10,000 times the levels in pre-industrial societies, which is truly no small
achievement. But there are several reasons why the present state is unsustainable.
First, there are great disparities in energy use and access. The richest 10 percent live
at levels of 20 kW/person and more, which is the equivalent of having permanently
some 26 horses or some 200 slaves at one’s disposal and not counting the energy
needed to feed them. At the same time, a large fraction of the world population still
lives in a biomass-based society at levels below 2 kW/person.3 The distribution is
3 This level is up to twice as high if the traditional biomass use is included. Its end-use efficiency is,
however, in most cases significantly lower than for fossil fuels, such as in cooking and space heating.
Commercial energy
Energy use (kW/cap) 20 Idem, incl. traditional

biomass
10
0
-30% 0% 30%
Number of people (% of total)

Figure 7.3. Energy pyramid: distribution of energy use in kW/cap. The width of the pyramid
shows the fraction of the world population within an energy use class (kW/cap) (source: World
Development Indicators 2004).
shown in the energy pyramid in Figure 7.3 and is a reflection of the similarly shaped
income pyramid. It is a concern in sustainability science, because energy is crucial for
the provision of basic needs like food and medical services, and a persistent shortage
may cause serious social and economic imbalances.
Second, fossil fuels are finite: Incapable of growth or reproduction at a rate
meaningful to humans, therefore, they are irreplaceable and non-renewable. They
are also geographically unevenly spread. The affluent countries with high per capita
energy consumption are now dependent on, although some prefer the word addicted
to, finite oil and gas resources that are largely controlled by other countries. As the
past century has shown, this brings serious geopolitical tensions that will aggravate
when high-quality deposits get depleted and users have to switch to more expensive
alternatives (Yergin 1991; §13.2). Third, the large-scale exploration, production and
combustion of fossil fuels has caused and still is causing environmental damage in
many places. Locally and regionally, coal mining, oil leaks and tanker accidents
and emission of sulphur- and nitrogen-oxides contribute to water and air pollution
and ecosystem degradation. Globally, the emission of carbon dioxide (CO2 ) and
methane (CH4 ), two important greenhouse gases, has already led to an increase in
radiative forcing and is causing, with near-certainty, changes in climate variables
such as temperature and precipitation.
In daily life, most people experience energy largely in the form of payments
at the petrol station and the monthly bill for electricity and gas. If one broadens
the horizon, the link with the big world is evident in the news items about OPEC-
revenues, Middle East oil and Russian gas power, offshore oil leakages, construction
of wind turbines and nuclear power plants, agricultural crops to fuel cars and so on.
To understand the role of energy in society in more depth, an introduction into
energy science is needed. In particular, two classical branches of physics, thermody-
namics and mechanics, have to be studied.4 The former teaches the basic principles of
4 A brief introduction in the basic concepts and laws of chemistry would also be in place in a sustain-
ability science textbook. Here it is considered outside the scope of this book.
Box 7.1. Free energy flow density as a measure of complexity. The astrophys-
icist Chaisson (2001) has proposed an interesting link between energy and com-
plexity. Organisms can be viewed as dissipative structures: Ordered objects whose
structure can be maintained thanks to a steady input of high-quality energy. The
free energy flow density () necessary to sustain such a non-equilibrium structure
is a measure of complexity. It can be expressed in erg per second per gram or,
more commonly, in mW per kg (1 erg = 107 joule). The free energy flows are
on earth supplied almost exclusively by solar radiation. It is possible to make
estimates of the -values for different systems on earth and the results are shown
in Figure 7.4. It seems that ever since the Big Bang, which started off our story of
the universe, complexity measured in this way has been continuously increasing.
After early life forms like the archaebacteria in the heat gradients on the ocean
floor, an important step towards higher complexity was made with the advent
of photosynthesis, the biological mechanism to convert solar energy into chem-
ically stored free energy (∼0.1 W/kg-biomass). Most animals have -values in
the range of 1 to 10 W/kg-biomass, depending on their activity. Birds operate
at an order of magnitude higher than reptiles, which gives this measure some
intuitive appeal. The human body dissipates free energy at a rate of about 1
W/kg-biomass, but if you are bicycling it is a factor ten higher – so complex is
bicycling in this definition. An astounding increase in free energy density flows
happened in the industrial age, with -values over 100 W/kg-biomass per person.
If measuring the free energy density per unit of mass of equipment, going from
organic to technological systems, there is a further increase to -values in the
order of 1,000–100,000 W/kg. The most extreme densities are realised in military
equipment and microdevices and nanodevices.
heat and work flows at the macroscale or phenomenological level. The latter explains
the elementary phenomena of mass and motion. Why bother about the basic laws
and models of thermodynamics and mechanics in sustainability science? First, there
are some natural and technical phenomena that can only be understood on the basis
of these laws and models. For example, the forces of water and wind in an erosion
process or the heat exchange in heating and cooling processes and equipment. They
are essential in even the simplest climate models. Understanding them facilitates an
elementary assessment of natural processes and energy technologies. Second, the
very notion of sustainable development is at least partly built upon concepts and
models developed in the natural sciences. For instance, every process needs a source
of work in order to sustain and develop against the dissipating forces of friction.
The notion of energy quality is crucial here. Third, these concepts and models have
often served as analogues or metaphors to investigate and discuss the behaviour of
more complex systems. For instance, the notion of sustainability is often associated
with equilibrium and related concepts such as stability and attractors. Elementary
knowledge of mechanics is helpful to grasp their essence. Harmonic and other types
of oscillators were introduced in ecology and economy as hypothetical models of
population and market dynamics and such analogies had large heuristic value des-
pite their oversimplification. Finally, sustainability science can benefit from the long
Pentium chip
military flighter plane 1990s
military flighter plane 1960s
Boeing 747
World War I Liberty plane
standard car USA 2000
motorway during rush hour NL
standard car USA 1978
Daimler engine 1900
Otto engine 1880
technological (wo)man
industrial (wo)man
European 1700 CE
bicycling human
birds
early agriculturalist
mammals
early fire-using gatherer-hunter
Australopithicene forager
sleeping human
reptiles
agricultural crops
trees
photosynthesis
hayfield
bacterial life
earth at genesis
geoth. heatflow at ocean bottom
Free energy flux density (mW/kg)

Figure 7.4. Estimated values of the free energy flow density in a number of systems (based on
Chaisson 2001).
history in the natural sciences, where phenomenogical laws exist alongside more
fundamental microdescriptions of dynamic processes.
7.2 Basic Energy Science: Thermodynamics
7.2.1 Classical Thermodynamics: The First and Second Law

Physics and chemistry have their own perspective on and contribution to sustainable
development. That is the topic of this chapter, though oriented more towards physics
than to chemistry. The word energy is derived from the Greek ενεργ εια (activity),
which is composed of the preposition εν, in, and the word εργ oν work. It is a funda-
mental quantity of nature that is transferred between systems or parts of systems. It
is ‘stored’ in a variety of forms and associated with movement, combustion, heat and
the capacity to do work. It is also widely used in more metaphorical and symbolic
sense, referring to a dynamic social, psychological or spiritual force or quality and to
the capacity of acting or being active.
It is important to get some familiarity with the units of energy stocks and flows.
The energy stored in systems is a stock. Examples are the content of heat in a cup
of tea or the kinetic energy in a moving car, as well as the chemical energy stored in
a barrel of oil. The basic unit of an energy stock or quantity is the joule (J), which
is about one quarter of the traditional unit of heat or the calorie. The calorie is the
amount of heat needed to heat one gram of water one degree Celsius in temperature.
The joule is such a small unit that one usually measures it in thousands: kilojoule
(kJ), millions: megajoule (MJ) or billions: gigajoule (GJ). For instance, the amount
of energy stored in one m3 of natural gas is the equivalent of about 36,000.000 joule
or 36 MJ.5
What makes the world go round are not energy stocks but energy flows. Ocean
water evaporates by the heat flow of the sun, a cup of tea cools down as heat energy
flows into the environment and the barrel of oil is taken out of the ground in order
to burn it and use the heat flow of the combustion gases. Similarly, it is the flow
movement of falling water that drives the turbine. The energy flow into a system is
defined as the quantity of energy that crosses the system boundary per unit of time.
The basic unit of energy flow is the joule per second (J/s) or watt (W). One hour has
60 × 60 = 3,600 seconds, so if you burn one m3 of natural gas in the course of one
hour, the energy flow equals 36,000,000/3,600 = 10,000 W or 10 kilowatt (kW). In
energy science, we do not speak of energy flows but of power. If the power is electrical
power, the unit kWe is often used. A widely used unit for an amount of energy is
the kilowatthour (kWh), which is defined as the energy content of delivering 1 kW
of energy flow during one hour (1 kWh = 3,6 MJ). Table 7.1 gives some values for
energy stocks and flows associated with familiar processes and systems.
Classical thermodynamics is the part of physics that deals at the macro-scale level
with energy and matter and particularly with conversion of energy from one form
into another.6 Boiling water for a cup of tea, using the microwave oven, switching on
the central heating system, cooling food in the refrigerator, lighting a room, riding a
bicycle or driving a car are all processes in which one form of energy is converted into
another. The conversion processes are described in scientific equations and models
that have evolved during centuries of experimentation. This introduction touches
only upon a few aspects of heat and mass transfer processes of relevance for the
notion of energy and the transition to a sustainable energy system.
First, some basic concepts. Let us look at a system consisting of a given amount
of a gas, a liquid (and its vapour) or a solid. The system boundaries are well defined
but can move during the change processes under consideration, as is, for instance,
5 The word ‘stored’ is not completely adequate, in the sense that it is stored in the 1 m3 of natural plus
the 10 m3 of air and, as part of, the 2 m3 of oxygen that together give on combustion a heat flow of
36 MJ.
6 One refers to classical thermodynamics to distinguish it from the micro-approach in statistical
thermodynamics.
Table 7.1. Energy stocks and flows associated with some processes and systems. Values are
approximations, because more specific system and process data are needed for precise data. See for
general unit conversions www.digitaldutch.com/unitconverter/ or similar sites
Energy Energy
content content
Process System MJ kWh System MJ kWh
Combustion 1 kg of coal 29.3 8 1 kg munic solid waste 9.4 2.6

1 kg of crude oil 42.1 11.7 1 barrel of crude oil 5750 1597
1 litre gasoline (petrol) 34.7 9.6 1 litre bioethanol 23.4 6.5
1 m3 natural gas (CH4 ) 38 10.6 1 m3 hydrogen (H2 )
Potential (lift/fall) 1,000 kg water at 100 m 1 0.3 75 kg body 270 m up 0.2 0.06
on stairway (Eiffel)
Kinetic (move) 20 ton truck at 90 km/hr 0.3 0.1 small car at 120 km/hr 0.016 0.005
Humans One day body heat 8.6 2.4
Animals 6 hr horse work 16.2 4.5
Food avg calorie intake/ 12 3.4 1 yr 1 soft drink 0.23 0.06
adult/day can/day
Heat/cold 500 litre water to 50◦ C 105 29.2 1 kg of ice at 0◦ C ? ?
(bath)
the case if a gas in a cylinder with a piston expands. The system boundaries may or
may not be permeable for energy and matter. A system is said to be (Figure 7.5):
r open if the if the system boundaries are permeable for energy and matter;
r closed if the system boundaries are impermeable for matter but permeable for
energy;
r isolated if the system boundaries are impermeable for energy and matter.
Isolated systems only exist in the laboratory (apart from the universe in its entirety).
The Earth is a closed system, with a constant mass and a nearly constant solar energy
inflow and equally large heat outflow. Other examples of (near) closed systems are
underground nuclear waste deposits or the cooling circuit in a car engine. Most sys-
tems of interest are open systems. The mass of an open system changes if the inflow
of matter into the system differs from the outflow. If both flows are equal, the system
Figure 7.5. Isolated, closed and open systems.
is said to be in a steady state or stationary state, that is, in a dynamic equilibrium.

Such a state has often been evoked to illustrate that a sustainable state is not a
static, frozen one. In economic science, where the notion of equilibrium penetrated
19th-century thinking, this point was already made by Marshall in his book Prin-
ciples of Economics (1890) and again by Daly in his book Steady-State Economics
(1973).
For a (near) isolated system, there is a certain quantity that does not change,
despite all kinds of changes in the system. This remarkable observation has been
confirmed over and over again. The quantity is called energy and the observation
is called the First Law of Thermodynamics. It expresses the conservation of energy
principle. It is also stated as, energy cannot be created or destroyed (first order
perpetuum mobile is impossible).7 There is another quantity that can never decrease
for an isolated system. It is called entropy and this remarkable observation is called
the Second Law of Thermodynamics. It expresses the natural tendency towards
disorder. It is also stated as, it is not possible to design an engine which, working in a
cycle, produces no effect other than the transfer of heat from a colder to a hotter body
(second order perpetuum mobile is impossible). In everyday language, you cannot
extract work from a heat source without another heat sink of a lower temperature.
Since the 19th century, the two laws of thermodynamics have been formulated
and reformulated. One formulation is in the form of a state equation: a system in
equilibrium is characterised by the state variables internal energy (U), entropy (S),
volume (V), masses (mi ), pressure (P), temperature (T) and chemical potentials
(μi ) for a system with i = 1 . . . N substances. The chemical potentials refer to the
chemical substances and their concentrations in the system. If a state variable is
proportional to the size of a system, such as U, S and V, it is an extensive variable. If
it is independent of system size, such as P, T and μi , it is an intensive variable. Often,
a system description is in specific terms, that is, the extensive variables are divided
by the system size.8 The same letter is used but in lowercase, for instance, u = U/m
for internal energy per unit mass. In this formulation, the two laws are:
I. The internal energy U of an isolated system is constant. For closed systems, the
internal energy U of the system only changes if energy crosses the boundary and
for open systems if energy and/or matter cross the boundary;
II. The entropy S of an isolated system can never decrease, tending towards a
maximum in an equilibrium state.
Both laws are empirical theorems, that is, they cannot be proven deductively and
they are valid only because all conclusions drawn on its basis are in agreement with
experience.
Entropy is, unlike energy, not a conserved quantity. It is created continuously.
And it is difficult to grasp. Real-world processes occur via a continuous sequence of
non-equilibrium states. Non-equilibrium means that there are internal disturbances,
for instance, a difference in temperature or pressure across the system. It generates
7 It would be better to speak of conservation of mass-energy, in order to include the conversion of

mass and energy in nuclear reactions.
8 The measure of size is usually mass m. But it may also be volume, as in the expression for density
ρ = m/V.
Box 7.2. The notion of thermodynamic equilibrium. By the end of the 19th
century, the laws of thermodynamics were well formulated and confirmed by
experiment. They are statements about systems in equilibrium, because only in
equilibrium are the intensive variables (pressure P, temperature T or concentra-
tion μ) homogeneous in space and constant in time. As soon as this equilibrium
is disturbed, for instance, by heating or pushing it from the outside, an adequate
description of the system would require a huge amount of variables – for instance,
the pressure and temperature differences across space and over time. Thermody-
namics is actually thermostatics: The system jumps from equilibrium state X0 at
t0 to equilibrium state X1 at t1 .
In a thought experiment, one can imagine to go in steps from the initial to
the final equilibrium state by briefly bringing the system in contact with the
environment, then isolate it again and wait until equilibrium is reached. I will
call such a process pseudo-dynamic or quasi-static. If the steps are infinitesimally
small, at any given moment, an infinitesimally small gradient in an intensive
variable (P, T or μ) is enough to keep the change process going. The overall
process would be reversible, that is, one could reverse from the final to the initial
state along exactly the same path. In other words, no macro-order is lost to micro-
disorder. Unfortunately, it has to be infinitely slow, an unpractical condition. After
all, it is only a thought experiment.
A thermodynamic equilibrium is only a static situation at the phenomeno-
logical level. At the micro-level of atoms and molecules, there is continuous
movement of mass and exchange of energy. Because the number of particles is
huge (≈1023 ), statistical laws can be applied successfully, as in statistical ther-
modynamics and mechanics. Thermodynamic equilibrium is actually a dynamic
equilibrium at micro-level.
The Law of Chemical Equilibrium provides an illustrative example of inter-
ference with dynamic equilibria. It was formulated in the 1860s by Guldberg and
Waage and expresses the insight that chemical equilibrium is a dynamic and not a
static condition because chemical reactants change in both directions at the same
speed. Another important observation about chemical equilibria is the Principle
of Le Chatelier, dating from the 1880s: If one of the factors determining the equi-
librium changes, the system undergoes a change such that, if this had occurred
by itself, it would have introduced a variation of the factor considered in the
opposite direction. In common language, the system in equilibrium responds to
an intervention with an attempt to undo it. This finding that real-world processes
tend to restore the equilibrium state upon being disturbed had a large influence
on the subsequent development of the life and economic sciences. Economists
considered it such a basic feature of markets that it became the foundation of
standard economic theory (§10.4).
internal frictions and shows up as heat. Such dissipation of energy at the microscopic
level implies a loss of order. It can only be avoided when the process is infinitely
slow. In statistical thermodynamics and mechanics, the second law has gotten an
interpretation in terms of probability: Entropy is proportional to the probability
of microstates, that is, atomic and molecular configurations (position, momentum,
mass) at the microlevel.9 I refer the interested reader to the suggested literature at
the end of this chapter.
One formulation of the second law is that ‘things go in a certain direction’ in the
universe – there is an ‘arrow of time’. Imagine a system brought in contact with the
environment, such as a hot cup of coffee in a cold room or salt thrown into the water
when boiling potatoes. In the state of being in contact with the environment, as in
these examples, mass and energy interactions take place until equilibrium is reached
and the state variables no longer change. The second law states that such a process
always goes into a direction in which the differences in temperature, pressure and
chemical potentials in a system tend to equilibrate. Have you ever seen a cup of hot
coffee getting hotter spontaneously or the solved sugar spontaneously reappearing
as a sugar cube?
How then is the order possible that we see in organisms, humans or societies?
The answer is that the emergence and sustenance of such complex highly-ordered
systems is only possible at the expense of an increase of disorder in the environment.
Were the earth an isolated system, life would be (nearly) impossible. However, the
earth is not: The solar energy influx represents an energy source from which work can
be extracted before it dissipates into ambient-temperature heat – in photosynthesis
and in photovoltaic cells, for instance.
If you search for it yourself, you get a big discount (

C Stefan Verwey 1999).
9 In Boltzman’s formula for an ideal gas: S = k log W, with k Boltzman’s constant and W the number
of possible microstates. Entropy has been linked to information theory by Shannon and Wiener:
processes that lose information are analogous to the processes that gain energy.
7.2.2 Energy Quality: The Potential to Do Work

What does all this mean for sustainable development? If energy can neither be
created nor destroyed, as the first law states, then the energy content of the universe is
constant – which is, at first sight, good news. It even raises the question why mankind
should economise on energy use and it suggests that we should stop speaking of
energy consumption and spending energy. But there is more to it. Although human
beings have always appreciated heat as a way to make their habitat more accessible
and comfortable and to perform tasks like cooking and metal smelting, it really is
work that matters most. Farming requires ploughing, irrigating, terrace building –
traditionally done by human and animal muscle power, and in some places by wind or
water. Manufacturing requires moving around materials and products, compressing,
cutting, separating and so on – processes which could only be done at large scales
and rates after the introduction of the fossil-fuel based steam engine and later steam
turbine. Transport means moving people and goods around – traditionally done with
animals but nowadays almost completely based on the internal combustion engine.
Finally, communication increasingly means moving electrons around.
It is here that the second law comes in. It relates that not all the internal energy
of a system is available as work. But classical thermodynamics is a bit of a strange
science. It can only make statements about systems in equilibrium, that is, systems
in which no change and no outside intervention takes place. In equilibrium, the state
variables have fixed values. What we are interested in are processes from one state
to another. Let there be two well-defined equilibrium states 1 and 2 of a system.
Changes in the state of a systemfrom 1 to 2 are caused by interaction between
the system and its environment. Such interactions are controlled interventions in
experiments, like adding a certain amount of heat Q while fixing the volume V. Or
everyday interactions between a system and its environment; for instance, lifting
a box from the floor up on the table or boiling an egg. There are in principle
infinitely many ways to bring the system from state 1 at time t to state 2 at time
t + t. Thermodynamic analysis wants you to be precise in the description of the
equilibrium states 1 and 2. The derivation of the basic equation for a steady-state or
stationary state process is (Appendix 7.1):
Q = WCV + ṁ · (h2 − h1 ) + g(z2 − z1 ) + ṽ22 − ṽ21 /2 J

(7.1)
In this formula, a system change is considered from state 1 to state 2. During

this process, Q, W and m are the heat, work and mass flows respectively across
the system boundary. h is the specific enthalpy, that is, enthalpy per unit mass (in
J/kg), g the earth gravitation constant, and z and v the position and the velocity
with respect to a reference level. If you can apply this formula, you have a basic
understanding of elementary energy conversion processes such as evapotranspira-
tion, salinisation, steamturbines, refrigerators and hydropower plants. I refer to the
suggested literature for those who accept the challenge.
This equation allows the introduction of a very important notion: energy quality.
You probably intuitively grasp that one kWh of heat in your bathtub differs from
one kWh used to power your laptop or charge your battery. So let us look again at
equation 7.1 and assume that the heat flow Q is from an infinitely large reservoir
of temperature Th to another infinitely large reservoir of temperature Tl . The ther-

modynamic formulation of the second law in terms of entropy change is:

δQT
S = + Sirr J/K (7.2)
T rev
In words, the entropy change equals the heat flow at temperature T divided by the
temperature plus the irreversible entropy production.10 The entropy is produced
whenever a process is not infinitely slow and stems from the different forms of
friction if the process proceeds at finite velocity – think of the turbulence that you
create stirring your cup of coffee. Because the two reservoirs are infinitely large and
the only entropy change comes from irreversible entropy production, you have:
−QT −QT QT QT
h l l h
Sirr = ShighT − SlowT = − = − J/K (7.3)
Th Tl Tl Th
in joule per degree Kelvin (J/K). How much work can be extracted from this
heat flow? The first law reflects that the sum of heat and work is conserved, so
upon rewriting QTh = QTl + W, you get for an infinitely slow, reversible process
(Sirr = 0):
QT
Tl

h
Wrev = QT − QT = QT − · Tl = 1 − · QT J (7.4)
h l h Th Th h
You may wish never to read this again, but it is an extremely important result.
It states that the maximum – because reversible – work that can be extracted from a
heat flow between a heat source and a heat sink amounts to the fraction (1 − Tl /Th )
of that heat flow. You now have an expression for the intuition that heat has a lower
quality than other forms of energy: Only part of it can be converted into work! In
qualitative terms, it expresses the fact that heat is a less ordered form of energy than
potential and kinetic energy in force fields.
Now, for the last step, suppose that the low temperature reservoir is the Earth,
at ambient temperature T0 and with other intensive variables p0 and μi0. If the heat
flow happens at a finite rate and there is also a stationary mass flow in and out of the
control volume, we write the equivalent of equation 7.1 for entropy:
QT QT
0 h
Ṡirr = ṁ(s2 − s1 ) + − J/K (7.5)
T0 Th
with s the specific entropy, that is, entropy per unit mass, (in J/kg/K). This can be
rewritten into (Appendix 7.1):
QT QT
0 h
Ṡirr = ṁ(ϕ2 − ϕ1 + pot−en + kin−en ) + − J (7.6)
T0 Th
Why is this equation so important? Because it makes it possible to calculate
the maximum amount of work that can be extracted from a system that is brought
reversibly into equilibrium with the Earth environment. For this, we define a new
state variable: specific exergy ϕ. The equation tells that the maximum extractable
10 Note that in thermodynamics, temperature T is always in degrees kelvin or K, with T = 273 + [temp
in ◦ C].
work equals the change in specific exergy and in macroscopic position and velocity,
and the fraction (1 − T0 /Th ) of the heat entering the system. In the real world, the
work W that can be extracted in a change from state 1 to state 2 will always be less
because of the irreversible entropy production T0 dSirr/dt . The following definition is
used for the newly defined state variable:
The exergy of a system is defined as the maximum amount of work that can be
extracted from it against a well-defined earth environment.
To calculate it, we specify the earth environment. We indicate it as (T0 , P0 , μ0i ), with
T0 ambient temperature (usually 273 K or 0◦ C), P0 ambient pressure (usually 1 atm)
and μ0i the chemical potentials of the reference substances and their concentrations
in the earth environment. In everyday life, you are particularly interested in two
processes: fuel combustion and heat transfer. Combustion (‘burning’) of carbon
fuels is the process in which chemically stored energy in the reactants (‘fuel’ and air)
is released in the form of heat in the products (mostly water and carbon dioxide).11
Heat transfer is the process in which a system changes temperature.12 If bringing a
fuel reversibly into equilibrium with (T0 , P0 , μ0i ), then the full exergy content can
be extracted as work. But burning in an open fire, a boiler or a furnace is never
reversible. The chemical energy is converted into heat of a temperature T – and only
the fraction (1 − T0 /T) is available as work (equation 7.6).13
The higher the temperature of the combustion products, the more work can be
extracted. Given that work is what we want, it gives us a ranking of energy stocks
according to quality. Table 7.2 lists the quality of energy and matter in a system in
terms of the exergy content. The sun can be considered a black body of 6000 (degree
Kelvin) K according to Planck’s law. Therefore, solar radiation has in theory about
95 percent work potential and is a high-quality energy source. Nuclear energy in
fission material and chemical energy in fossil fuels or biomass can be converted into
a heat flux of high temperature – usually in the range of 500◦ C to 1500◦ C – and,
therefore, they represent stored high-quality energy. On the other hand, using the
20◦ C difference in temperature between the surface of the ocean and the ocean
water deep down will produce much less power per unit of heat flow. And the water
of 60◦ C in a radiator also has a very low work potential.
It is also possible to calculate the quality of a mineral ore or metal, by calculating
the exergy from the difference in chemical potential between the ore or metal and
the earth environment. Each piece of metal represents work potential in an exergy
analysis. Almost pure metal represents a 100 percent work potential, because most
metals in nature occur only at low concentrations (<1%) (Table 7.2). Nature did
work for us in accumulating certain minerals in rich ores! Vice versa, work potential
is lost if pure metal is dissipated in the environment – as with lead in gasoline or
titanium in paint – and it will require work to recover and concentrate it again.
11 Chemical reaction kinetics is another branch of natural science that provides simple examples
of positive (‘autocatalyic’) and negative feedback loops. It is also becoming a research area in
complexity science (Nicolis and Prigogine 1989; Solé and Goodwin 2000).
12 Heat transfer can also happen without temperature change, for instance, in phase transitions (evap-
oration of water, smelting of ice and so on).
13 The exhaust gases drop in temperature during the heat transfer process. The logarithmic mean of
entrance and exit temperature turns out to be a good approximation for use in the formula.
Table 7.2. The quality of different forms of energy. The quality index indicates the fraction
(in %) of an amount of energy that can be converted to work by bringing it reversibly into
equilibrium with earth environment (after Wall 2001; www.exergy.se)
Form of energy/matter Quality index (%)
Extra superior Potential energy, e.g., high water reservoir 100

Kinetic energy, e.g., waterfalls 100
Electrical energy 100
Pure element, e.g., diamond 100
Superior Nuclear energy in nuclear fuel almost 100
Metals (commercial quality) almost 100
Sunlight 95
Chemical energy in fossil fuels 95
Element mixes, e.g., steel, alloys, plastics 90
Rich mineral ore deposits 50–80
Hot steam 60
Poor mineral ore deposits, e.g., bauxite 20–50
District heating 30
Water in radiator at 50◦ C 15
Inferior Waste heat 5
Valueless Heat radiation from the earth 0
There is still one problem to be resolved. Nature does not give any clue about
the absolute value of internal energy U, nor of entropy S or exergy Ф. Thus,
we have to choose it ourselves, as a convention.14 The choice is a reference situ-
ation R, where it is simply agreed that the internal energy in that situation equals
zero. An obvious choice for R is the equilibrium with the ambient temperature,
pressure and atmospheric concentrations (T0 , P0 , μ0i ). The standard or reference
value for temperature is 0◦ C (273.15 K) and for pressure 1 atm (101.325 kPa).
In chemical reactions, the standard enthalpy of formation href (h = u + Pv) of a
chemical compound is then defined as the change in enthalpy h when the compound
is formed from the stable compounds in their natural concentrations (CO2 and H2 O),
the reactants and products all being in a given standard state The specific exergy
of an important class of compounds, namely fuels, is within about 5 percent equal
to the combustion enthalpy of the fuel, because the entropy term is rather small.
Therefore, the fuel enthalpy is often used in exergy analysis (Table 7.1).
One last reflection on the relevance of the Second Law. Every real-world con-
version process will degrade the quality of the energy in dissipative processes such as
friction. Macroscopic order, in the form of, for instance, potential or kinetic energy,
is converted into microscopic disorder in the form of heat. To ‘sustain’ such a pro-
cess, there has to be a continuous supply of high-quality energy from the system’s
environment. This happens across gradients in the intensive variables such as tem-
perature, pressure or concentrations. The larger the gradient, the larger the driving
force, the more irreversible the process – and the larger the dissipative work. If you
14 We do this also, in a more accessible way, for potential energy in the earth gravitation field: Without
some convention of a reference height, it is not possible to calculate E = mgh. The same holds for
the kinetic energy.
Box 7.3. The Colosseum, slaves and containerships. The energy needed to
deliver energy has always been a concern for societies. ‘All our societies require
enormous flows of high-quality energy just to sustain, let alone raise, their com-
plexity and order (to keep themselves [ . . . ] far from thermodynamic equilib-
rium) . . . [and] after a certain point in time, without dramatic new technologies
for finding and using energy, a society’s return on its investments to produce
energy [ . . . ] starts to decline’ (Homer-Dixon 2006:54–55). An example is the
Colosseum in Ancient Rome (Homer-Dixon 2006). Construction began under
Emperor Vespasianus between 72 and 75 CE and inaugurated in 80 CE by his
son Titus, with a hundred days of games and some ten thousand beasts being
killed. An estimated million ton of raw material have been moved for its con-
struction. An estimated 185 TJ of energy was needed. Over 75 percent of it
was used to feed the thousands of oxen engaged in transporting materials. The
remainder powered the human labourers: Over 2,000 people working 220 days a
year for 5 years. During this period, some 55 km2 in the largely agrarian economy
was needed to deliver this solar energy.
Compare these numbers with the largest container ships under construction.
Such ships measure 400 m in length and over 12 m in depth and can carry 15,500
containers of 36 m3 content or 56,2000 m3 of freight. Their normal speed is
46 km/hr, so the kinetic energy of only the freight is equivalent to 46 GJ at an
average freight density of 1,000 kg/m3 . With a 10 percent speed reduction, fuel use
drops with a quarter, which means about 250,000 ton/yr less of CO2 -emissions.
The next round of upscaling is to ships of 18,000 containers, which are expected
to be profitable because they carry more freight but at lower speed. The builders
claim a further reduction in CO2 -emission, to 3 grams per ton-km of freight. At
present, Rotterdam Harbour can handle 100,000 containers/week and is among
the few harbours that are deep enough for such ships (NRC 18 March 2011). The
trend is an even larger scale of operations in the name of lower cost and lower
emissions.
drive a distance of 110 km in a car in two hours at 55 km/hr (∼15 m/s), you need
about 5 kWh to overcome air resistance. If you drive the same distance in one hour at
110 km/hr (∼30 m/s), you need 20 kWh or four times more (MacKay 2009). But large
driving forces are desirable because they make processes go fast, and we humans
are in a hurry because we do not live forever. Doing things means loss of potential
work – and doing things fast means more loss of potential work. Or the other way
around, doing things slowly reduces loss of potential work. Of course, doing things
infinitely slowly makes no sense, so there is some optimum between what we want
to accomplish and how much work potential we are able and willing to use for it.
7.2.3 Energy Forms

Energy can be categorised in several ways. One distinction is between transitional
energy and stored energy. Stored energy is energy that is available in a mass or a
position in a force field. Stored energy can be converted to another form of stored
energy and the process takes place via transitional energy: heat Q and work W.
As discussed, energy is around in different forms: internal energy and poten-

tial and kinetic energy in force fields. The internal energy U is the energy of the
constituent atoms and molecules in the form of rotational, vibrational, translational
and chemical bonding energy of atoms and molecules (equation 7.2). It is mani-
fested as sensible heat (temperature related), latent heat (phase change related,
for instance, when water vapour condenses as raindrops) and chemical and nuclear
energy. Changes in internal energy is what matters. It manifests itself in everyday
experiences like burning your fingers on a stove or lighting a match. Potential and
kinetic energy in a gravitational or electromagnetic field are two other energy forms
(equation 7.4). A lifted arm or object represent potential energy in the earth grav-
itational field. Water flowing downhill, a moving bus or a metal rod spinning in a
fluctuating magnetic field in an operating electromotor represent kinetic energy.
Table 7.3 lists the major energy forms and their conversion processes and associated
appliances.
The essential source of energy is solar electromagnetic radiation, which via pho-
tosynthesis is converted to chemically stored energy. It can be transformed in kinetic
energy of movement, for instance, in the human or animal body (metabolism). The
chemical energy from photosynthesis has been accumulated in geological time in
fossil fuels, which upon combustion deliver thermal energy. At high temperatures,
this heat is used in internal combustion engines and in steam/gas turbines to pro-
duce kinetic energy (moving piston, rotating shaft). This is either used directly for
moving mass or converted into electricity in an electric power generator. If the heat
comes from a geothermal or a nuclear source, it is similarly used to generate power.
In these processes, part of the kinetic energy is inevitably converted into thermal
energy (heat). Thermal energy is also stored or released in phase transitions (evap-
oration/condensation, melting/freezing). Solar radiation is directly converted into
electricity, in photovoltaic (PV) solar cells, or into heat, in solar heat absorption
collectors. Chemical energy is directly converted into electricity in fuel cells.
Electricity has become the dominant and most versatile form of energy available.
Electric power is converted into electromagnetic radiation of different wavelengths
in an increasing amount of Information and Communication Technology (ICT)
appliances and other devices in household and other equipment. Electricity is also
used directly in industrial processes such as electrolysis in aluminium smelters. It
mostly ends up as heat at ambient temperature and is used directly for room heating.
The biggest disadvantage of electricity as compared to chemical energy in fuels is
that it is not easily stored. Devices like batteries and flywheels do store it in the form
of chemical or kinetic energy, but they are relatively inefficient and costly.
Because work is the high-quality energy form of genuine interest to us, the con-
version routes of energy are very important and at the core of energy technologies.
Understanding the causes of exergy losses and the potential for reducing those losses,
that is, increasing the exergetic efficiency by extracting work from high-temperature
heat, is an important part of the search for sustainable energy systems.
7.3 Movement in Space and Time: Mechanics

Change is an elementary phenomenon in life. It is most visible in movement of
matter. It is useful to look first at a simple, well observable and controllable system:
a mass m at the end of a spring. It is the classical model of linear elasticity, associated
196
Table 7.3. Energy conversions. Each column shows how a particular form of energy can be converted into another one that is indicated in the
corresponding row to the left
FROM:
kinetic/
thermal chemical gravitational electric electromagnetic nuclear mechanical
TO:
thermal heat flow (heat exothermal resistance heating solar heat collector fission/fusion friction in
exchanger), phase reaction, (nuclear movement,
transition (boiling, combustion reactor) inelastic
melting etc.) collision
chemical endothermal chemical reaction electrolysis photosynthesis ionisation radiolysis
reaction, (battery
thermolysis charging)
gravitational mass displacement electric elevator lifting a mass
electric thermoelectricity Battery, fuel cell transformator solar (PV) cell röntgen dynamo
generator
electro- thermal radiation chemolumi- electrolumi- photolumines cence radioactivity axe on stone
magnetic (light bulb), nescence fire nescence (laser display) (sparks)
thermocouple, flies (LED-lamp),
MHD-generator radio/tv,
LCD-display
etc.
nuclear particle gamma reactions charged
accelerator particle
reactions
kinetic/ Expansion, heat metabolism, falling mass electric motor, solar sail particle emission gearbox
mechanical engine, (internal muscle power, hydroturbine flywheel (nuclear
combust, steam/ fire arms bomb)
gas turbine)
1
Position x
-1
Figure 7.6. Graphical presentation of the mass on a spring. In the left, it is kept in stretched
position, When left loose, it moves upwards to the middle and then the right position.
with the 17th-century British scholar Hooke. Change is about potential and kinetic
energy in a gravitational field (equation 7.1).
Assume that the position of the mass is indicated with x(t) and that initially it
is at rest in position xR . If you pull the mass, it starts an oscillating movement. This
happens under the influence of two forces: the gravitational force Fg and the spring
force Fs . The gravitational force is given by Fg = mg with g the gravitational constant
in m/s2 , and the spring force is approximated by Fs = −kx with k the spring constant
in kg/s2 . In the rest position, the gravitational force pulls the mass a small distance
down – say xR , at which the two forces are equal: Fg = Fs and m.g = k.xR or xR =
m.g/k. Let us normalise and define xR = 0. Suppose you pull down the mass to a
position – p at time t = 0. At this point the upward spring force will equal Fs = kp and
the potential energy in the gravitational field has dropped with an amount of mgp.
The system is in equilibrium because your hand exerts a downward force equal to
Fs − Fg . Letting loose, the mass will start to move upward because of Fs – Fg = kp. Its
velocity v increases and the kinetic energy is 1/2mv2 . The law of conservation of energy
implies that the sum of potential and kinetic energy remains constant. Therefore,
when the mass passes in its upward movement the point x = 0, the potential energy
it had at x = p, equal √ to mgp, has been converted into kinetic energy. Hence, at
mgp = 1/2mv2 or v = 2gp on passing the point xR 15 . The further you pull the mass
down, the greater the speed at which the mass goes past the point xR .
Using the formal language of mathematics, one can approximate the movement
of a mass on a spring with two differential equations:
dx/dt = v
dv/dt = −kx (7.7a)
in which the small d indicates an infinitesimally small change. The first one says that
the velocity v equals the rate of change of displacement and the second one that the
rate of change of velocity, that is, the acceleration, is negatively proportional to the
15 This is not completely correct, because a small part of the energy is needed to lift the spring upwards
in the gravitational field and equal to mg(p-xR ).
distance from the point x(0) = xR = 0. Substituting the first equation into the second
yields a single second-order differential equation of the form:
d2 x/dt 2 = −kx (7.7b)
which can be solved analytically16 . I will not derive analytical solutions and instead
use the simulation software package Stella R
to explore the system behaviour over
time. The harmonic oscillations in Figure 7.7a represent the position of a mass, x(t),
over time for the reversible (ideal) frictionless spring. Plotting the velocity of the
mass, v(t) = dx(t)/dt, as a function of the position x(t), one gets the phase diagram
(Figure 7.7b).
Of course, you know that the mass will not keep oscillating forever. There will
be friction, which shows up as a dampening force proportional to the velocity. This
yields an additional term in the expression for the restoring force:
dx/dt = v
dv/dt = −kx − rv (7.8)
with r the friction coefficient. This description is more realistic because it incorpor-
ates the phenomenon that any process is irreversible and more ordered energy is
dissipated into heat (molecular motion) in agreement with the second law (Sirr in
equation 7.2). The speed of the mass slowly diminishes over time and ends in the
equilibrium point x(0) = XR = 0, called the attractor (Figure 7.7c–d). One can also
apply an external force on the mass, for instance, by pulling it at a constant fre-
quency ω and amplitude q. Now, the natural and the forced movement interact and
the position x(t) over time becomes erratic (Figure 7.7e–f). If someone would show
you this graph, you would have a hard time to discover how simple the underlying
system actually is!
What happens if the spring constant is itself a function of the displacement from
equilibrium: k = k(x)? Suppose it is found that the spring constant behaves according
to k = (x2 – 1). This implies that the system behaves like a mass at the end of a spring
only for x > 1 or x < −1. For −1 < x < 1 the value of k is negative and the system
operates in this domain not with deceleration due to friction but with acceleration.
In system dynamics terms, there is a positive or amplifying feedback in this domain.
Such a system is described by the two differential equations:
dx/dt = v
dv/dt = −(x2 − 1)x − rv (7.9a)
The system has three attractors, two stable and one unstable. A steel spring
placed between two permanent magnets is a bistable oscillator described by these
equations (Bossel 1994).17 If there is no dampening (r = 0), the system oscillates
between the two attractors, as shown in the trajectory in time and in the phase dia-
gram in Figure 7.7g–h. The damped system moves towards one of the two attractors,
but it is hard to say in advance to which one. A slight uncertainty or change in the
initial state, that is, x at t = 0, or in one of the parameters can make the system
16 The general solution for a class of such equations is by separating variables (x, left; t, right) and
integrating both sides. I refer to the suggested reading for details.
17 Written as a single second-order differential equation, the model is known as the Duffing equation.
a) 1.5 b) 1.5
Position and velocity
0.75 0.75
Velocity
0 0
-1.5 -1 -0.5 0 0.5 1 1.5
-0.75 -0.75
-1.5 -1.5
0 4 8 12 16 20 24 28 32 36 Final
Timestep Position
c) 1.5 d) 1.5
0.75 0.75
Velocity
0 0
-1 -0.5 0 0.5 1
-0.75 -0.75
-1.5 -1.5
0 4 8 12 16 20 24 28 32 36 Final
Position
Timestep
e) f) 1.5
1.5
0.75 0.75
Velocity
0 0
-1.5 -0.5 0.5 1.5
-0.75 -0.75
-1.5 -1.5
0 4 8 12 16 20 24 28 32 36 Final
Position
Timestep
mass position
mass velocity
Figure 7.7a–f. Movement of a mass at the end of a spring over time (state variables: mass
position x and mass velocity v; left) and in phase space (right). The middle graphs simulate
a situation with friction; the lower one with friction and exogenous forcing (timestep 0,01;
Runge-Kutta-4).
g) 3 h) 2
2
Position and velocity 1
1
Velocity
0 0
-2 -1 0 1 2
-1
-1
-2
-3 -2
0 10 20 30 Final
Position
Timestep
i) 3
j) 2
2
1
1
Velocity
0 0
-2 -1 0 1 2
-1
-1
-2
-3 -2
0 10 20 30 Final
Position
Timestep
mass position
mass velocity
Figure 7.7g–j. Bistable oscillator: movement of a mass at the end of a spring over time (state
variables: mass position x and mass velocity v; left) and in phase space (right) when the spring
constant is itself a function of the displacement. The upper graphs show an undamped situation
without friction; the lower graphs with an external forcing (timestep 0,01; Runge-Kutta-4).
move towards a different equilibrium state – it is extremely sensitive to the initial

conditions. The term (x2 – 1) introduces a nonlinearity. This usually leads to multiple
equilibria and critical dependence on the initial conditions. These simple technical
devices show complex behaviour.
If this system is also subjected to an external force of constant frequency ω and
amplitude q:
dx/dt = v
dv/dt = −(x2 − 1)x − rv + q · cos(ωt ) (7.9b)
its behaviour becomes irregular and even chaotic for certain parameter domains
(Figure 7.7i–j). Chaotic means in this context that the system – the same steel spring
but placed now in an oscillating, not a permanent magnetic field – behaves determ-
inistically but is nevertheless unpredictable in the sense that its future can only be
known if one would know the initial state with infinite precision. Note that these
4.4 Stories 201
Box 7.4. Strange attractors and butterflies. A famous example of extreme sensit-
ivity for initial conditions are the equations named after the meteorologist Lorenz.
He applied the basic equations in fluid dynamics in order to model the movements
in a closed box with smooth interior and filled with gas, which is heated at a con-
stant rate:
dx/dt = a(y − x)
dy/dt = bx − y − xz
dz/dt = xy − cz (7.10)
with x, y and z the coordinates in space. It is a simplified three-dimensional
description of atmosphere processes. The behaviour of {x(t), y(t), z(t)} turns out
to be highly sensitive to the initial conditions and, therefore, becomes highly
unpredictable or chaotic. The system will also tend towards one of two particular
patterns over time, a phenomenon called a strange attractor. The model has
important implications for climate and weather prediction, because it shows that
a fully deterministic system can exhibit quasi-periodic change and abrupt and
apparently random change. It has become popular with the question for weather
forecasters, ascribed to Lorenz, whether the flap of a butterfly’s wings in Brazil
can set off a tornado in Texas.
equations describe the movement of a mass in space and time (potential and kinetic
energy) and not the associated heat flow (friction).
These simulations show that rather simple nonlinear equations (‘models’), which
may or may not be accurate descriptions of real-world systems, can exhibit complex
behaviour over time. Their use as analogues is useful in the sustainability discourse,
provided that one understands the idealising assumptions. In particular, the invest-
igation of multiple equilibria in ecosystems and economic systems is a new area of
research. Elementary chemical processes, for instance, chemical kinetics, provide
more interesting and complex examples, but this is considered beyond the scope of
this book. I refer the reader to the suggested literature.
7.4 Stories
7.4.1 Fuel Efficient Stoves for People in Darfur18

In the fall of 2005, Berkeley scientist Ashok Gadgil was asked by the U.S. Gov-
ernment to try to find a solution to a grave problem facing Darfuri families in
displacement camps: Women had to walk as long as seven hours, three to five times
per week to find firewood. In response, he and his colleagues and the women of
Darfur designed the Berkeley-Darfur Stove (BDS). The project is a good example
of sustainable livelihood improvement, as it solved several problems simultaneously.
Because the BDS uses half as much firewood as traditional cooking methods, it limits
harmful emissions that contribute to global warming. Users save $300 per year in
18 This text is provided one of the project leaders, Ashok Gadgil. For more information, see www
.darfurstoves.org
fuel costs. Over the five-year life span of the stove, this savings is approximately
$1,500. Reduced excursions for firewood collection translate into rest and relief for
the exhausted women in the displacement camps and time that could be spent with
family or on income generating activities. As to safety, the stove reduces the need
for women to leave the camps in search of firewood and it thereby reduces their
exposure to sexual assault and violence. It also contributes to health because it limits
toxic pollutants released that cause respiratory diseases such as pneumonia.
The BDS was developed by scientists and engineers at the Lawrence Berkeley
National Lab, with evaluation and feedback from women in Darfur. They brought
together the world’s best minds in engineering to develop a simple, locally appro-
priate technology and to provide ongoing technical support to the field partners.
The BDS is well adapted to the local situation. It is specifically tailored to the windy
climate, the sandy terrain, the pot sizes and the cooking style of families living in
the displacement camps in Darfur. The local participation of the women in Darfur
provided feedback at every step of the process, ensuring the stove design fits their
needs. The local employment and low cost due to its manufacturing process are also
beneficial. The BDS starts out as sheet metal pieces stamped out in India. These
‘flat-kits’ are shipped to Sudan, where they are assembled by the Sustainable Action
Group, a Sudanese NGO affiliated with the project partner Oxfam America. The
total cost to fabricate, ship and assemble each stove is $20.
The Darfur Stoves Project and its partners produced over 15,000 stoves by the
end of 2010. Each year they are in the field, these 15,000 stoves save Darfuri women as
much as $2 million in fuel expenses and offset more than 20,000 tons of greenhouse
gases, which is equivalent to taking >4,000 average-sized cars off the road in the
United States. An estimated 300,000 families are in need of a fuel-efficient stove
and the goal is to distribute a BDS to each family. This project teaches at least two
lessons for a project to succeed: You have to ‘walk with the people’, and there are
always winners and loosers and you have to identify them.
7.4.2 The South Nyı́rség Bioenergy Project19

The South Nyı́rség region in the northeastern part of Hungary was traditionally an
agricultural area, renowned for its tobacco and apple production. Since the transition
in 1990, local agriculture has gradually lost its importance. Apple export to the Soviet
market has stopped, tobacco production has dropped and other agricultural activities
are hindered by the relatively low quality of soil. Today, most of the landowners
want to plant trees. Trees grow on lower-quality soil and there are subsidies for
afforestation. There are about 80,000 hectares (ha) of privately owned forest and
about 70,000 ha of state-owned traditional forests in the South Nyı́rség region. Most
of the new plantations include fast-growing species such as black locust or poplar,
and the produced wood is typically used for pellet production.
The Project was initiated in 2003 by local governments and entrepreneurs
because of high unemployment and outmigration of skilled workers. A logical
choice was biomass-based power generation. Hungary was already in the process
of European Union accession, and it was expected that regulation would support
19 This text has been written by Zoltán Lontay, initiator and chief engineer of the project.
renewable energy production. A company was founded that established strategic

cooperation. Project development took more than four years. The local government
of Szakoly village offered 20 ha, which had to be requalified from agricultural to
industrial land, and dozens of permits had to be obtained. Connection to the net-
work had to be negotiated with the owner of the local public network, the German
company E.ON. The major challenge was finance. In 2007, the banks agreed to give
a loan of 85 percent of the €55 million investment – thus, 15 percent equity had to
be collected from elsewhere. Another company was formed to finance a complex
structure with four shareholders, who each had their own interests. Given the rather
unfriendly legal, regulatory and economical environment, the close relationship with
the local society turned out to be crucial.
In December 2007, the first greenfield biopower plant in Hungary started com-
mercial operation after an official ceremony and blessing by the priests of Szakoly
village. The fuel is woody biomass, delivered on a contract basis with twenty to
thirty small, local suppliers and the state-owned forestry company of the region.
The latter covers 10–20 percent of the fuel demand of 180–200 kt/yr (wet). Thin
branches, trunks, bark, vegetation from flood areas of rivers and trees from old
orchards are chipped, sorted on size and used in the power plant. Research is started
on how to utilise the ashes. Obstacles are the inefficiency of the state-run subsidy
program (farmers receive subsidies a year after plantation) and the fluctuations in
the agricultural prices (when the price of traditional agricultural products go up, the
landowners switch back to them).
Economically, the situation is still precarious. The first years of production were
below expectation but suppliers and operators are on a learning curve. The financial
crisis in 2008 made the banks more demanding and project revenues are too small
for economic operation because of a low feed-in tariff at €100/MWh, which is well
below the Central European level. The government says that a higher tariff increases
the electricity price and is, therefore, not possible for commercial and social reasons.
External benefits of small-scale renewable projects such as lower unemployment
payments, higher taxes and avoided emissions are apparently not considered.
Economic and energetic performance will be improved with the use of waste
heat in a greenhouse, which is expected to produce 2,500 t/yr of vegetables that meet
the European ‘bio’-standard and to employ 80–100 people. The project demonstrates
the opportunity for a paradigm shift in rural development. A national plan shows
that the construction of 50–70 similar plants in the 5–20 MWe capacity range can
help Hungary to reach the 2020 European Union targets for renewable energy
production, avoid up to 3 Mt/yr of CO2 emissions and employ over 10,000 people.
The project shows that growing biomass for fuel supply is an interesting energy
option for (parts of) Europe.
7.5 Energy Conversion
7.5.1 Elementary Processes

Thermodynamics can only make statements about systems in equilibrium. There-
fore, many of the laws and models describing real-world near-processes or non-
equilibrium processes are empirical or phenomenological laws. Here I briefly discuss
a few such processes. It is not the intention to be complete, and I refer to the Sug-
gested Reading and Websites sections for more information. The first process is
heat flow or heat transfer. Heating air in a room with a stove is a heat transfer pro-
cess. No work is involved: You simply produce a flow of heat Q from burning gas,
oil, coal or wood. However, as soon as the air has a temperature higher than the
ambient temperature, Nature strives to equilibrate the two – the consequence of the
Second Law. Therefore, you not only heat the room but also the surroundings. The
heat transfer from the room with an average air temperature Troom to the surround-
ings where the air has an ambient temperature T0 , is approximated by Fourier’s
Law:

Q Troom − T0
=k·A· W (7.11)
t t
It states that the heat flow out of the room is proportional to the temperature
difference between inside and outside, to the heat conductivity k (in J/m2 /K) and to
the size A (in m2 ) of the interface between room and surroundings (t is the time
period considered).20
This relationship is essential knowledge if you want to insulate your house or a
manufacturer wants to reduce process heat losses. The equation tells you that you
can reduce the heat flow and your energy bill by:
r putting on a sweater so T
room and hence the difference Troom – T0 can be lower
because the difference Troom – T0 bothers you less;
r reducing the size of the interface A – that is why living in a smaller house or
having smaller windows costs less energy; and
r reducing the heat conductivity k by insulating roofs, walls, floors and windows,
usually called energy conservation or efficiency improvement.
The third option is now widely applied and new buildings have to comply with ever
stricter standards. With t = 1 second, the k-values are in the range of 0.11 W/m2 /K
for a 30 cm glasswool insulation layer to 2 and 6 W/m2 /K for double and single glass
windows, respectively. Innovations lead to ever-lower heat conductivity values, but
it takes time before they are cost-competitive and accepted by house builders and
owners. Insulation is very effective: Replacing a single-pane window with the latest
HR-glass lowers k from 6 to 2 W/m2 /K – or 2 kWh per m2 of window on a cold day.
Also, it reduces noise levels. Putting glass mineral wool under the floor or against a
standard roof gives similar reductions.
There are more drastic solutions to make energy use in buildings more efficient.
You may remember that one loses a lot of work potential (exergy) by burning fuel
and use the heat in the exhaust gases to warm water to 60◦ C. If a small engine – a gas
turbine, for instance – is put between the exhaust gases and the water heater, work
can be extracted. Such a scheme is called cogeneration or Combined Heat and Power
(CHP). It is one of the options for decentralised electric power supply, in combination
with renewable sources like PV-solar. There are drawbacks: The small scale makes it
rather expensive and the coupling with room heating confines electricity production
20 In a more precise description, the temperature difference is indicated as a gradient across a distance
d and the heat flow is given in W/m.
to cold days. However, cogeneration and CHP is already widely applied in industrial
processes and increasingly in large buildings.
Another interesting option is the heat pump. It operates as a refrigerator: A fluid
absorbs heat from an outside reservoir (river/groundwater, air) as it evaporates and
delivers the heat inside your house when the vapour is compressed and condenses.
In other words, it is a refrigerator, but now interest is in the hot, not the cold side.
The heat exchanger in the rear of the refrigerator, or outside of the window for the
air conditioner, is now the heat source (radiator). Electricity ‘pumps’ the outside
heat to a higher temperature. The number of heat pump installations grows rapidly.
The second process to have a brief look at is the diffusion and discharge of mass flows.
Mass flows, and their diffusion in air, water and soils are ubiquitous in biological
and environmental as well as in technical processes. They are the core of micro-level
understanding of processes such as changes in soil nutrients, dispersion of pollutants
and (bio)accumulation. Here also there are two empirical laws that formulate the
change in terms of a gradient. The first one is Fick’s (first) law about the diffusion of
a substance across a concentration gradient. In equation form:
m (c − cl )
= −D · A · h kg/s (7.12)
t x
with m the mass, A the diffusion surface, ch and cl the high and low concentration,
respectively, and D the diffusivity which is the equivalent of conductivity in Fourier’s
law (equation 7.11). If concentration is replaced by osmotic pressure, it describes
the process in which osmotic pressure difference is used to extract work from the
difference in salt concentration at the outflow of river into the sea. A Norwegian
pilot plant uses this salinity gradient to generate 2–4 kW of power, at 1 W/m2 and
flow rates of 10–20 litre/s. It is a truly renewable resource, but the membranes are
still very expensive.
A similar relationship is used to describe the discharge of a fluid in the soil. For
instance, water flows in an aquifer are determined by the pressure drop per unit
length and are approximated with Darcy’s law:
V k · A (Ph − Pl ) 3
=− · m /s (7.13)
t μ x
with V the volume, k the permeability, A the discharge surface, μ the viscosity and
Ph and Pl the high and low pressure levels. Here, the inverse of viscosity is the
equivalent of conductivity. It is the basic equation for the calculation of the energy
needed to pump up water – or oil or gas – from an underground reservoir.
These laws, and other ones like Ohm’s law for electrical current and the Gaus-
sian dispersion model for air pollutants, express the second law observation that
Nature tries to reduce the difference in intensive variables (temperature, pressure,
concentration) between two (parts of) systems. It is a natural negative-feedback
process. These laws are never exact in the real world, their relevance is in describing
the underlying process and being often a good first approximation or ‘rule-of-thumb’
(§8.6). Interestingly, these laws came into existence well before they were scientific-
ally understood at the micro-level and were based on the intuition that inhomogen-
eities spontaneously tend to dissolve at a rate proportional to the gradient, that is,
on the Second Law.
This section ends with one of the most crucial conversions in modern society:
the generation of electricity from chemical energy (Table 7.3). It serves as an illus-
tration of the difference between an energy and an exergy analysis. In theory, the
chemical energy in a fuel can be converted into electricity with close to 100 percent
exergetic efficiency.21 In practise, most electricity is generated in a heat engine at
much lower efficiency due to the Second Law. Bringing a heat reservoir of high T –
for instance, exhaust gases from a furnace or engine, although this is not the infinite
heat reservoir from the theory (equation 7.6) – irreversibly into equilibrium with the
earth environment implies a loss in work potential (Table 7.2). But the higher the
exhaust gas temperature expanded in a turbine and the lower the temperature of
the at ambient river or sea water (condenser) or air (cooling tower) to which heat
is exported, the lower the exergy losses. There are various technologies to extract
work from a heat flow, such as the steam turbine and the gas turbine. The turbines
drive a generator in an electric power plant. The system can be operated as a closed
circuit as in a water-vapour Rankine-cycle steam turbine, in a closed-cycle gas tur-
bine or in an open-circuit open-cycle gas turbine. Increasingly, large-scale electric
power generation is done in Rankine-cycle operation coupled to a gasturbine in
a so-called Steam And Gas Generating-Combined Cycle (STAG-CC) plant. The
exergetic efficiency can be quite high. For a gas turbine, inlet temperature of 1200K
or more, over 75 percent of work can in theory be extracted. Present-day STAG-CC
power plants reach exergetic (chemical-to-electric) efficiencies up to 60 percent. In
comparison, most power plants around 1900 worked with much lower temperatures
and had efficiencies of 5–10 percent. One can increase the energetic efficiency by
using the low-temperature heat to heat offices or dwellings (district heating) or for
use in industrial processes (the previously mentioned CHP). The other option to
generate shaft work are the internal combustion (Otto and diesel) engines and Stirl-
ing engines. The efficiencies are lower and they are used primarily in cars and ships
and for cogeneration.
7.5.2 The Energy System

Energy features high on the sustainability agenda, because it is largely based on finite,
non-renewable resources and it causes a series of environmental problems. Massive,
and still increasing, amounts of fossil fuel are extracted, transported, processed
and then made to react chemically with the oxygen in the air (‘burnt’) to generate
power, heat and combustion products. In this way, one gets useful energy or energy
services such as heating houses and offices, running vehicles and industrial ovens,
and powering electricity using devices from aluminium smelters to streetlights and
from vacuum cleaners to televisions and computers. Together, they constitute the
‘energy chain’ from source to sink.
In addition to the energy use side, the energy system comprises energy sup-
ply, transport and distribution (Figure 7.8). This is what is usually meant with the
‘energy industry’. Its major elements are coal mines, oil refineries and electric power
21 In an electrochemical or fuel cell, the conversion rate to electricity has in theory no inherent limit-
ation. But there is still a gap between theory and practice, and fuel cells are not yet a reliable and
cheap enough option for large-scale application.
Figure 7.8. The energy system and its subsystems: energy services, energy conversion and
energy supply.
plants. It is a very large and highly concentrated industry. The economic value of the
transactions in the oil and gas industry amount to an estimated 2,000 billion €/yr. In
the electric power industry, it is in the order of 1,000 billion €/yr. Their combined
sales are about 5 percent of gross world product (GWP). The energy industry is very
capital-intensive. In Europe alone, in 1971–2000, annual investments were in the
order of 80–120 billion €/yr. Some of the most influential technological transitions
have been in the energy field. From a decision-making and policy perspective, the
large energy companies are in a more powerful position than the millions of dis-
tributed energy users. This is an important issue in the transition to non-depletable,
non-carbon energy sources.
The energy flow densities in electric power plants and other parts of modern
energy systems are extremely high. For instance, the fuel and heat flows in large
power plants and the oil and gas flows in large pipelines are in the order of W/m2
of throughput surface. This is possible because of the high energy density of fuels
(20–30 GJ/m3 ). Energy density is one of the big differences with renewable sources
of energy. They are all based on solar energy fluxes with values of at most 1.2 kW/m2 .
The consequence is that these resources are, on the one hand, abundant and ‘free’,
apart from the equipment to harness it. On the other hand, it implies the use of
space. In Europe, an average 100–200 W/m2 can be harnessed from the sun.22 With
PV-solar installations, present-day installations produce in the order of 3–4 W/m2 .
If it is converted to plant biomass, it will be difficult to produce in Europe at values
above 1 W/m2 . Offshore wind turbine parks deliver at 2–3 W/m2 , and tidal power
delivers in the same range. Hydroelectric dams can deliver up to 10 W/m2 of lake
area. These values are low as compared to the average oil- or gasfield or coalmine.
22 The numbers are based on MacKay (2009) and some other sources.
For instance, the Groningen gasfield has for many years produced at a rate of 80
billion m3 /yr or 90,000 MW. If 1 percent of the province of Groningen is needed for
its exploitation, the density is 3,000 W/m2 .
In electric power generation, there is a tendency towards ever larger scale in
order to reduce costs – the capacity of nuclear power stations is in the order of 1,200
MWe for a single plant and possible nuclear fusion plants are envisioned to be much
larger. Large-scale fossil-fuel and nuclear systems need centralised social-technical
management in view of the huge upfront investments, high-tech skills, safety meas-
ures, waste disposal and so on. The capital stocks last for four or more decades and
their dismantling and waste management may last many more. Responsible opera-
tion is only possible in advanced societies. A renewable energy system of spatially
dispersed wind and PV-solar parks is naturally rather decentralised, although very
large windturbine parks are being planned and solar power plants in American,
Chinese or North African deserts could cover huge tracts of land. They face differ-
ent constraints in the form of competition with other uses of land and the need for a
reliable grid with storage options. Thus, the energy transition away from fossil fuels
will rely on more efficient provision of energy services and on some mix of land-
intensive and intermittent (sun/wind) renewable energy and possibly sufficiently
safe forms of nuclear energy. Unless a yet-unknown, miraculous energy source is
found, fossil fuels and notably natural gas will play an important role as transition
fuel (§12.3).
7.6 Energy Futures

Thanks to tremendous improvements in efficiency, there has been a steady decline
in the energy needed to provide one unit of energy services (Figure 4.9b). But
because the number of and variety in energy services did and still does increase
exponentially, total energy use keeps growing (Figures 7.2 and 13.8). As it is largely
derived from burning fossil fuel, emission of CO2 and other gases did increase as
well and is expected to continue to do so (Figure 4.7). Thus, the system operates
under the increasingly tight constraints of depletion (‘source’) and environmental
pollution (‘sink’).
How will the future energy systems in the world unfold? Past trends can be
extrapolated (Figure 7.2), but how and with which arguments? Most energy experts
agree that demand for energy services will continue to increase and that fuels for
transport and electricity are the two major carriers. That is where agreement stops.
Table 7.4 illustrates the differences in views with a series of statements about the
problems, the solutions and the mechanisms for a sustainable energy future. I apply
the worldviews to contrast different possible directions, using statements from vari-
ous institutions (§6.3). As the reader, you are invited to examine your own values
and beliefs on the basis of these statements.
The preferred values can be summarised with the keywords availability, afford-
ability, reliability and (environmental) sustainability. The controversies are mostly
about markets and prices versus governments and standards, about the risks of
energy dependence, and about the costs and risks of particular technological options.
Clearly, there are serious differences in the evaluation of what is possible and desir-
able. If any of the four worldviews becomes dominant, the serious shortcomings
7.6 Energy Futures 209
Table 7.4. Worldviews and energy futures.
ENERGY
Statement 1: World energy use can and should be stabilised at the present level of about 400 EJ/yr.
Impossible. People need I don’t know and I I don’t know but I will It is necessary in view of
and want energy. don’t care, as long as try in my own environmental risks. It
Low-income regions I can get my gasoline situation. I think and is also possible and
have legitimate and electricity feel it is necessary if challenging. Poor
aspirations for cheaply and reliably. we are to survive in people still need more,
high-energy lifestyle. the long term. so the rich have to use
less.
Statement 2: The world will remain heavily reliant on fossil fuels for the next forty years (CEO Shell, 2010).
Quite possible, if political Why not? There are Irresponsible. Oil Impossible. Shell assumes
alliances are made with abundant fossil fuels companies only a doubling of demand
the oil-rich states. and human-induced think about their between 2010 and 2050.
Climate issues can be climate change is profits. Why do they It would mean massive
solved with CCS, plus still unproven and not invest their huge coal burning – and
unconventional oil and maybe acceptable or profits in renewable dangerously high
gas and coal reserves even beneficial. energy? carbon emissions.
are outside OPEC.
Statement 3: Nuclear power is neither safe nor clean. In order to save the future of our planet, we must
continue to fight the expansion of nuclear power (Greenpeace 2011).
Irresponsible. Industry Nonsense. It is Correct. The The precautionary
needs power and probably the only life-threatening principle should be
nuclear power is way out for the problems of guiding. Stricter
probably the only world energy radioactive regulatory and safety
serious, cheap and system, given the contamination, measures are needed,
zero-carbon option threat of climate proliferation and as the accidents make
available. change and fossil waste disposal are clear.
fuel depletion. not solved!
Statement 4: Governments should support energy efficiency and technologies such as the electric and
hydrogen car.
In principle, energy There is no need for The technical but also There are still enormous
markets should be these fancy options, the lifestyle aspects efficiency gains
deregulated – energy because there is no of energy efficiency possible, but price
prices should do the energy shortage. It should be more incentives should be
job. For infant is a waste of tax intensely promoted. complemented with
high-tech support may money. But high-tech will regulation and R&D
be useful. never be a real support.
solution.
Statement 5: Sufficient clean power can be generated in the world’s deserts to supply mankind with enough
electricity on a sustainable basis (Desertec Foundation).
It is an ambitious project, Is this a serious Another technocratic An ambitious project, that
but if it can be proposal? It makes megalomania. should be considered as
financed – why not? Of us more dependent Renewable energy is one part of a global
course, there are some on unreliable states needed, but not in renewable energy
technological and and in the end I will this highly system. It may also
political challenges, but have to pay for these centralised form. solve water problems
that’s OK. big projects. and provide much
needed employment.
of such an option would show up sooner or later – as the opponents are all too
eager to point out. One can expect that different directions will be taken in differ-
ent countries. For instance, France is already over, with 70 percent of its electricity
generation dependent on nuclear power. The collective experiences will teach us,
hopefully, about the sustainable middle road without too catastrophic and irrevers-
ible damage to future mankind’s options for a good quality of life.
A value survey about the pros and cons of wind power in Denmark is an illus-
tration of this point.23 It revealed two very different profiles. The Profile of the
Nay-sayer is characterised by statements like: Renewable energy cannot solve our
energy problems; wind turbines are unreliable and dependent on the wind; wind
energy is expensive; wind turbines spoil the scenery; and wind turbines are noisy.
They represent the Not In My BackYard (NIMBY) attitude. The Yes-sayer, on the
other hand, expresses quite different values and beliefs: Renewable energy is very
much an alternative to other energy sources; climate change theory must be taken
seriously; wind energy is limitless unlike fossil fuels; wind energy is non-polluting;
and wind energy is safe. These people are in favour of wind farm cooperatives,
small-scale and self-reliance.
Similar differences are found with respect to nuclear power. Some people do
not accept the large risks and the centralised structure of a nuclear energy system,
whereas others see it as a necessary step in the best of possible worlds. Also, the
role of oil and gas depletion and dependence is quite differently viewed by people.
Exhausting the high-quality oil and gas resources as fast as our population does now
causes the risk of an irresponsible switch to coal and nuclear energy, in order to
avoid disruptive energy shortages. At the same time, and accelerated by a switch
to coal, there is already too much carbon released in the atmosphere to prevent at
least some change in climate. Only moderation in energy use, drastic improvement
in efficiency and rapid and ongoing introduction of renewable energy sources offer
the prospect of a sustainable energy transition. Will this road be taken? Unlike the
apparent consensus on energy futures in the 1960s, it is not clear how the energy
systems in the world will look like by 2050. Because of the long lead times of oil
production and conversion plants and power plants, there is quite some inertia in
the system (§2.3). But for the period after 2030, it is next to impossible to make
predictions about the details of the envisaged energy transition.
7.7 Summary Points

Energy is an essential aspect of development, as source of food, as driver of the
industrial economy, and as cause of environmental damage. From a sustainability
science perspective, the following points about energy are to be remembered:
r In agrarian societies energy use was almost exclusively based on solar energy.
With industrialisation, there was a regime shift to fossil fuels;
r Classical thermodynamics gives us the equations to formulate the energy quality
in terms of work potential and to understand energy conversion processes. It
also teaches us that, although people cannot destroy energy itself, people do
destroy its quality in combustion and dispersion processes;
23 www.windwin.de/images/pdf/wc03041.pdf.
r the concept of equilibrium, important in the sciences as an ‘ideal’ and/or ‘natural’

situation, is rooted in classical thermodynamics and mechanics;
r classical mechanics describes elementary processes that are often used as meta-
phors and shows that apparently simple processes and models can exhibit com-
plex behaviour;
r in the real-world of non-equilibrium processes, empirical/phenomenological
laws are used. Their essence is that nature tries to annihilate gradients (in tem-
perature, pressure, concentration);
r production of work (mechanical, electrical) is a crucial process in the provision
of energy services and has consequences (depletion, emissions) that are at the
core of some sustainable development concerns; and
r increasing the efficiency of energy conversion processes is of utmost importance
to sustain or expand the energy flows required for present-day societies. The
second pillar of a sustainable energy system is the introduction of renewable
energy resources.
On peut bien expliquer l’arbre par ses causes, par sa structure, par les mécanismes
qu’il met en jeu, les échanges qu’il entretient avec son environnement, etc. Mais le
comprendre, non : il n’y a rien à comprendre, et c’est pourquoi aucune théorie ne
saurait remplacer le regard, la simplicité du regard.
– André Comte-Sponville, L’amour la solitude, 2000:31
Translation:
You can very well explain a tree in terms of its causes, its structure, the mechanisms
it calls into play, the exchanges with its environment, etc. But understanding it, no:
there is nothing to understand, and that’s why no theory can ever replace looking,
simply looking.
I sell here, sir, what all the world desires to have – power.
– Boulton, 18th-century British industrialist, quoted in
Heilbroner and Milberg, 2004:66
SUGGESTED READING
Introductory text on Life Cycle Assessment.

Baumann, H., and A.-M. Tillman. The Hitch Hiker’s Guide to LCA. An Orientation in
Life Cycle Assessment Methodology and Application. Lund, Sweden: Studentlitteratur AB,
2004.
An introductory textbook on theory and practice of energy analysis.
Blok, K. Energy Analysis. Amsterdam: Techne Press, 2006.
An introduction into dynamic models: a gradual build-up from elementary models to more
sophisticated ones, with an emphasis on population-environment issues.
Bossel, H. Modeling and Simulation. Wiesbaden: AK Peters Ltd./Vieweg, 1994.
A detailed discussion of exergy and its applications.
Gong, M., and G. Wall. On exergy and sustainable development – Part 2: Indicators and
methods. Exergy International Journal 4 (2001) 217–233.
An introduction into the scientific aspects of energy use and energy conversion technologies
with much practical info and sample calculations.
MacKay, D. (2009). Sustainable energy – without the hot air. UIT Cambridge (download at
www.withouthotair.com).
Introduction on sustainable development from a technical and design perspective.
Mulder, K. Sustainable Development for Engineers. Sheffield, UK: Greenleaf Publishing, 2006.
One of the early natural science explorations of complex non-equilibrium phenomena.

Nicolis, G. and I. Prigogine. Exploring Complexity – An Introduction. New York: Freeman &
Company, 1989.
An instructive and extensive introduction into the history of energy in all its forms.
Smil, V. Energy in World History. Boulder, CO: Westview Press, 1994.
A rather advanced textbook on the mathematics of nonlinear dynamics.
Strogatz, S. Nonlinear Dynamics and Chaos – With Applications to Physics, Biology, Chem-
istry, and Engineering. Boston: Addison-Wesley Publishing Company, 1994.
A detailed discussion of exergy and its applications.
Wall, G., and M. Gong. On exergy and sustainable development – Part I: Conditions and
concepts. Exergy International Journal 1(3) (2001): 128–145.
A comprehensive textbook on modeling environmental processes, with background and
examples of natural science principles.
Wainwright, J., and M. Mulligan. Environmental Modelling – Finding Simplicity in Complexity.
London: John Wiley& Sons, Ltd., 2004.
USEFUL WEBSITES
The Wikipedia presents for most elementary processes good descriptions. There are many
sites that explain basic concepts in thermodynamics and mechanics.
Specific Techniques, Processes and Models

r www.convert-me.com/en/, www.onlineconversion.com/energy.htm and www
.digitaldutch.com/unitconverter/ are three sites for unit conversion.
r www.exergy.se/ is the site where Wall offers courses on exergy analysis and exergy
economics, with extensive documentation.
r www.ornl.gov/sci/roofs+walls/insulation/ins 01.html is the Oak Ridge National Lab/
United States site with information and models on heat insulation.
r www.e-calculator.nl/ is a site (in Dutch) that can be used to estimate your annual
energy use. There are a number of energy calculator sites in English, for instance,
energyabacus.com/.
r www.carbonify.com/carbon-calculator.htm is a site that can be used to estimate annual
carbon emissions (carbon footprint).
r webphysics.davidson.edu/physlet_resources/thermo_paper/ is an interactive demonstra-
tion of the expansion of a gas in a cylinder.
r www.aw-bc.com/ide/idefiles/navigation/toolindexes/9.htm#9 has several interactive mod-
els on mechanical systems, such as the harmonic oscillator, the mass-and-spring and the
forced vibration, and also on electrical systems.
r www.aw-bc.com/ide/idefiles/media/JavaTools/nlhcrate.html has an interactive model on
cooling.
r There are many energy models around, for instance, the Message model developed
at IIASA. www.iiasa.ac.at/Research/ENE/model/extensions.html, the TIMER model
developed at PBL at themasites.pbl.nl/en/themasites/image/model_details/energy_
supply_demand/index.html and the GET model at Chalmers University.
r There are also many studies about the future of the energy system, for instance, the
annual World Energy Outlook (WEO) of the International Energy Agency (IEA) at
www.worldenergyoutlook.org/.
r www.eia.gov and www.iea.org are both sites with many data and news on energy.
r www.ewea.org/index.php?id=180 is the site of the European Wind Energy Association
(EWEA).
r www.iaea.org is the site of the international Atomic Energy Association for info on
nuclear energy but also other energy issues.
r www.supersmartgrid.net is the site about the supersmartgrid for Europe.
Appendix 7.1 The Steady-State Mass-Energy Equation 213
Figure 7.A1. The control volume as starting point for the steady-state energy equation.
Appendix 7.1 The Steady-State Mass-Energy Equation

We define a control system with volume CV (Figure 7.A1). Suppose that in a steady-
state process a mass m1 is entering a control volume at time t and a mass m2 is leaving
the same control volume at time t + t (Nieuwlaar and De Ruiter 2010). Because
mass is a conserved quantity, you have:
m1 + mCV (t ) = mCV (t + t ) + m2 kg (7.A1)
For t → 0, we get dmCV /dt = dm1 /dt − dm2 /dt or in physics notatio ṁCV = ṁ1 −
ṁ2 .
For this process, the first law requires that the heat exchange Q in the time
interval t with the system must equal the change in the internal energy of the
control volume Ecv plus the work W done. If you indicate the specific energy of
a mass unit with e (in J/kg), then equation 7.1 becomes for the control volume
considered:
Q + m1 e1 + ECV (t ) = ECV (t + t ) + m2 e2 + W J (7.A2)
If one considers the volume work done by the inflowing and outflowing mass, PV,
then this equation becomes:
Q = ECV (t + t ) − ECV (t ) + m2 e2 + m2 P2 v2 − m1 e1 − m1 P1 v1 + WCV J (7.A3)
with WCV the work done minus the volume work done.
The specific energy e consists of the internal energy u, which is an intrinsic char-
acteristic of mass m, plus potential and kinetic energy in macroscopic gravitational
and other fields. For a system with mass m and velocity ṽ, these are in the earth
gravitation field:
1 2
e pot = mgz and ekin = mṽ J (7.A4)
2
with z the distance above the earth surface, g the earth gravitation constant
(9,8 m/s2 ). Using now the formula for specific enthalpy h = u + Pv, rewrite
equation 7.3:
Q = [ECV (t + t ) − ECV (t )]
+ m2 h2 + gz2 + ṽ22 /2 − m1 h1 + gz1 + ṽ21 /2 + WCV J

(7.A5)
You have written this balance for the change in a period t. As previously, it can be
rewritten as a rate equation for t → 0:
Q = ĖCV + WCV + ṁ2 · h2 + gz2 + ṽ22 /2 − ṁ1 · h1 + gz1 + ṽ21 /2 J (7.A6)

The dot indicates the first derivative of time. In words, the heat transfer to the system
equals the change in the internal energy of the control volume plus the work done by
the control volume plus the difference in the internal energy of outflow and inflow. If
you now assume that the mass flows and the state of the control volume are constant
in time, you get the basic equation for a steady-state or stationary process:
Q = WCV + ṁ · ((h2 − h1 ) + g(z2 − z1 ) + ṽ22 − ṽ21 /2) J

(7.A7)
If you can apply this formula, you are able to understand elementary energy conver-
sion appliances such as turbines and heat exchangers. I refer you to the Suggested
Reading for those who accept the challenge.
The steady-state equation can now be written in an entropy form by combining
equation 7.5 with equation 7.A1. It yields an expression for the work flow that can
be extracted when the system is brought reversibly into equilibrium with the Earth
environment:
⎡ ⎤
(h1 − T0 s1 ) − (h
22 − T02s2 )

T
W = ṁ · ⎣ +g(z1 − z2 ) + ṽ1 − ṽ2 /2) ⎦ + 1 − l · QT − T0 Ṡirr J (7.A8)
Th h
with Th = T0 . This formula reflects that the maximum work (Ṡirr = 0) you can
extract on earth from the system equals the changes in the state variable h–T0 s, in
gravitational and kinetic energy and in the reversible work from the heat flow. Call
ϕ = h–T0 s the specific exergy of the mass. In thermodynamics, it is in more general
form formulated with help of a new state variable: the free enthalpy or Gibbs free
energy defined as G = U + PV – TS or, per unit of mass, g = u + Pv – Ts. The formula
can be extended to include more than on substance and chemical reactions.
8 On Knowledge and Models
8.1 Introduction
Sustainability science: The word science suggests pursuit of ‘scientific knowledge’.
But what is scientific knowledge? Let us have a closer look at what the acquisition
of scientific knowledge is in practice. Suppose you are concerned about air pollution
and set up an experiment to measure the concentration of substance X in a well-
defined area. The measuring tool is itself a specimen of scientific development. The
result of your experiment is a series of concentration values at given location p and
time t, c(p,t). Building upon atmospheric physics and chemistry, you interpret the
results in terms of dynamic cause-and-effect processes. Such a description, framed
in the formal language of mathematics, is called a scientific model.
You realise that it is actually the impact of air pollution that matters, so you
decide to explore impact on the forest in the area. With the help of ecologists, you do
additional experiments and extend the model. The concentration values c(p,t) are
now inputs to descriptions of the various trees in the forest. They are a measure of
the exposure of the simulated trees to external factors. Because the tree dynamics
are relatively slow, longitudinal experiments have to be set up (>5 years). The result
of these experiments are an indication of the sensitivity of the various trees for the
particular exposure c(p,t).
Unfortunately, you cannot rely on such solid laws in this field as in atmospheric
science. Estimates of tree sensitivity are based on controlled laboratory experiments
and fields surveys, but they are only partly transferable to your field situation. The
trees in the forest differ in age and in location-dependent parameters such as soil
and water access. There is also interference with other species and other pollutants.
Besides, the forest may have varied responses to different or prolonged exposure
c*(p,t), which falls outside your measurement domain. It will take much effort and
many years before you have valid scientific knowledge about the impact of air
pollution on the forest.
One evening, you meet a friend who argues that the real issue is whether the
measured air pollution has a negative effect on the health of the people living in
the area. Taking up the challenge, you ask some medical scientists to engage in
a longitudinal research project to measure the health situation of people in the
area and, for comparison, of people in another area with negligible air pollution.
215
Although there is substantial medical knowledge about how the air pollutant affects
the physiology of the human body, your long-term experiment is confronted with
large uncertainties. For one thing, the group of people followed in the experiment
is not constant because people move in and out of the area. There is also a large
variety, both somatic and psychic, in the population samples. For instance, some
individuals are more sensitive to exposure, while others are better able to cope with
the effects. It is difficult and ambiguous how to deal with this variety. Often, ad-hoc
research strategies have to be designed. The experience has surely made you less
naı̈ve about ‘scientific knowledge’.
The ordeal is not over yet. At an environmental economics conference, an eco-
nomist argues that one does not need to know the impacts of the air pollutant con-
centration in great detail; it is more important to know at what cost it can be reduced
below some level, which is considered or negotiated as ‘acceptable’. Recognising
the appeal of this argument, you step into another research project with econom-
ists to estimate the options and costs to reduce the concentrations. You calculate
the emission reductions needed for a ‘safe’ concentration level and some economist
colleagues identify and rank the emission reduction options according to the cost
per unit emission reduced. In their view, polluters respond mostly or solely to price
incentives so a policy should focus on the proper tax levels. They quote several policy
analyses and behavioural surveys to proof their point. Other economist colleagues
disagree and quote scientific evidence in favour of emission standards for equipment.
Policy should enforce stringent standards in order to induce technological options
that reduce emissions at much lower cost.
During the deliberations, you discover that some of your colleagues’ relation-
ships with government officials and entrepreneurs seem to play a role in their con-
victions about the most successful approach. You realise that the dispute cannot
be settled in the way of the natural sciences: Controlled experiments are largely
excluded and all the parameters involved keep changing all the time. You realise it
is time for some philosophical reflection.
As this story illustrates, if you widen the system boundary, more and more
scientific disciplines get involved. Questions and answers become more complex
and uncertain. People come into the picture, with their own social and cultural
characteristics. You as investigator enter the scene, with your skills, limitations and
biases. In this chapter, the focus is on epistemological issues and ways to handle
complexity and uncertainty, as these play an important role in sustainability science.
8.2 Models in the Natural Sciences
8.2.1 The Scientific Method

The underlying paradigm in science is that there are regularities in the world, which
are manifestations of dispositions and tendencies and capacities in the particulars
or ‘things’ of the material world. The regularities result from complex interactions
between these particulars, which are causal in the sense that they result from actions
(Chalmers 1999).1 To find out about the regularities, there is the scientific method.
1 Causality is not a prerequisite: The laws of classical thermodynamics, for instance, cannot be inter-
preted as causal laws (§7.3).
Its essence is to generate empirical laws from observations and measurements in con-
trolled experiments. It is an inductive process of inference, guided by pre-scientific
intuitions and notions. It yields descriptive, phenomenological statements or ‘sci-
entific facts’ and is at the core of the empiricist approach. Empirical reductionism
or scientific materialism is the radical interpretation that this is the only valid way
to acquire knowledge and it is associated with the Modernism worldview (§6.3). It
is rooted in a conviction: ‘ . . . all tangible phenomena, from the birth of stars to the
workings of social institutions, are based on material processes that are ultimately
reducible, however long and tortuous the sequences, to the laws of physics’ (Wilson
1998).
Upon reflection, it becomes clear that the statements or facts cannot be derived
solely from the senses: An appropriate conceptual framework and knowledge about
how to apply it are needed. The most convincing proof is that what once were
considered ‘scientific facts’ turned out to be invalid or subject to revision within a
novel conceptual frame. Facts not only precede theory, theory also precedes facts.
Before or with the experiments, a formal system is developed and used to generate
linguistic statements and scientific theories. This is a deductive process during which
the link with observations and measurements is less direct and the level of abstraction
is high. It is the core of the rationalist approach. Theory is derived from facts by logical
reasoning. Its essence is resounded by Simon (1969): ‘The central task of natural
science is to make the wonderful commonplace: to show that complexity, correctly
viewed, is only a mask for simplicity; to find pattern hidden in apparent chaos’.
Mathematics is a stronghold of the rationalist position because it is hard to imagine
empirical progress without the concepts and methods of applied mathematics.
There are several problems with a radical empiricist approach. Given that one
cannot do infinite numbers of observations, what are the rules to derive more gen-
eral statements than the one that strictly follows from a few observations? In other
words, when are logical deductions valid and what is, therefore, a justifiable gener-
alisation and validity domain? Most scientists would give the pragmatic answer that
a statement is true if it has a high probability to be true in the light of the evidence.
Another problem is that in contemporary science, much knowledge is not accessible
to direct observation; for example, think of electrons or genes.
A brief history of the ideal gas theory can illustrate the scientific process and the
role of models therein. An amount of gas in a cylinder is in classical thermodynamics
described on the basis of the following macroscopic observables: mass (m), temper-
ature (T), pressure (P) and volume (V). To quantify these observables, a reference
system is defined, which is used as the measuring device, for instance, a thermometer
for T and a manometer for P2 .
If you fill your bicycle tire with air, you will notice that it takes less effort in the
beginning than later on. You rediscover Boyle’s law (1662): The product of pressure
and volume remains constant for a given amount of gas if the temperature does
not change, so more gas in a fixed volume implies higher pressure. It took over 140
years before this observation was expanded into what became known as the Boyle-
Gay-Lussac law (1808): PV∼T. Yet, this model was not more than an observed
proportionality. With the corpuscular theory according to which gases consist of
2 The Système International (SI) of measures and units for the sciences is based on the definition of
such physical reference systems. See physics.nist.gov/cuu/Units/units.html.
Box 8.1. Definitions of models. There are many definitions and views of models.
‘Models and simulations of many kinds are tools for dealing with reality . . . Even
thousands of years ago, buildings, boats, and machines were first tested as small
models before being constructed on a large scale. Children’s games have always
been simulations of the world of grown-ups – using models of people, anim-
als, objects, and vehicles. The model worlds of mythology, legends, and reli-
gions . . . have guided the behavior of generations in all cultures . . . Models range
from miniaturised realistic representations of the original to technical drawings
to functional diagrams. They may consist of stories, fables, and analogies or be
expressed in mathematical formulae or computer programs’ (Bossel 1994:1–2).
One type of model already encountered in Chapter 2 are mental models.
Mental models are ‘deeply ingrained assumptions, generalisation, or even pic-
tures or images that influence how we understand the world and how we take
action . . . The discipline of working with mental models starts with turning the
mirror inwards; learning to unearth our internal pictures of the world, to bring
them to the surface and hold them rigorously to scrutiny’ (Senge 1990:8–9).
More formal distinctions between models are:
r deterministic vs. probabilistic or stochastic;
r analytical vs. simulation;
r continuous vs. discrete, depending on whether differential-integral calculus is
used or not.
Refer to the Suggested Reading for more in-depth treatment.
small particles (‘atoms’), a wide range of observations were brought together in the
formula PV = nRT with n the number of moles and R the gas constant.
Since then, the observed data have been refined and so have the mathematical
equations. For instance, Van der Waals proposed replacing V by V-b in order to
correct for the non-zero volume of the gas particles. There are now a number of
more advanced equations that describe the observations even better and across a
larger domain of variables. But such a more refined formal description of empirical
observations does not really add to our understanding. A next step was to formulate
the system in terms of statistical distributions of the properties of the gas particles
(speed, momentum).
The spiralling process of controlled experiments and induced hypotheses in
combination with deductive formalisms has led to many hypotheses, which became
accepted and mainstream – only to be later falsified and dismissed or modified when
new observations were made. Often, the earlier hypothesis is not refuted but shown
to have a limited validity domain. An example is Newtonian mechanics as a spe-
cial case in relativistic mechanics. Or the statistical thermodynamics approach that,
later combined with quantum mechanics, made it possible to relax the assumption
in classical thermodynamics of equilibrium and homogeneity in the intensive vari-
ables (§7.3). A new operational model emerged, using micro-level concepts and
incorporating rather than invalidating the previous model. The reverse also hap-
pens: A hypothesis is dismissed or ridiculed for lack of empirical support, only to be
re-established later on when its intuitive strength is confirmed (§5.3).
Figure 8.1. Modelling as an encoding of

a natural system (after Rosen 1985).
8.2.2 Models and the Modelling Process

In a scientific context, the word model refers to any formal representation of obser-
vations and measurements, ranging from empirical laws to universal theories. Often,
however, it is defined rather broadly and loosely. In everyday usage, a model refers
to some copy or image. It may be a physical replica, usually in miniature, or an
example for imitation. It can also be a description used to help visualise something
that cannot be directly observed. Modelling is a way of looking at and ordering the
world around us. In science, models are usually defined rather formally, for instance:
a model is a representation of a part of reality, based on a system of postulates,

data, and inferences presented as a mathematical description of an entity or state of
affairs.
Modelling is an activity with an objective: to get (better) understanding of some

part of reality and to directly or indirectly control or manage a system and to solve
a problem. This ‘part of reality’ is created by explicitly and consciously drawing a
boundary around what one thinks should be investigated and hence modelled: the
system boundary (§2.3).
Modelling can be viewed as a coding process (Figure 8.1; Rosen 1985). It creates
a relation between a ‘natural’ and a ‘formal’ system. The natural ‘real-world’ system
is the object system or target system and is identified by means of observables, that
is, qualities and quantities that can be observed and measured in a more or less
controlled way. ‘A Natural System is a collection of observables connected by mind
created relations’ (Rosen 1985). The formal system is the model system, which can
be any set of mental elements (objects, symbols and so on) and relations and rules
that permit the generation of statements – as in language or logic. The rules are used
to generate new statements, which are decoded in order to explain and predict the
behaviour of the natural system. This is why they are called rules of inference. In
this sense, a model is an approximation of a part of reality. Although this view of
modelling is inspired by the natural sciences, the social and psychological ‘interior’
reality can also be part of such a ‘natural’ system and investigated with participative
and introspective methods (§10.4).
The process of constructing a model in interaction with the external world is
sketched in Figure 8.2. The outside world (natural system) is perceived in a physiolo-
gical recording through the sensory apparatus – the five senses and their technical
extensions like microscope and telescope. This forms percepts in the human mind
that are converted into observations. In combination with previous experiences
Figure 8.2. The modelling process in science.
(ideas, awareness or emotions), a conceptual model is constructed. This may dir-

ectly lead to a prediction and resulting action (experiment). In a scientific process,
the conceptual model is formalised and translated into a mathematical model. This
permits more precise prediction and more targeted new experiments.
In principle, the modeller has total freedom in choosing the observables in the
natural system. In practise, however, this freedom is constrained by the available
measuring systems, which are part and parcel of scientific and technical advances. In
fact, the simultaneous development of experiments, measuring devices and formal
mathematical systems is at the root of the scientific revolution of the last couple
of centuries. The measuring devices – thermometers, manometers and so on – can
be considered as reference systems against which any system can be calibrated.
The domain of empirical observations and measurements has expanded enormously
with the development of microscopes, telescopes, echoscopes and all sorts of elec-
tromagnetic and physico-chemical measuring devices. Mathematical techniques have
become one of the hallmarks of scientific and technical progress. As with the measur-
ing devices in empirical analyses but more subtly, their use induces a methodological
bias. In the words attributed to Mark Twain, ‘If the only tool one has is a hammer,
one treats everything like a nail’. In other words, in order to be analytically solvable,
there is a tendency to simplify the system under study.
Sometimes one and the same natural system can be approximated by two dif-
ferent formal systems (Rosen 1985). This happens when some essential, though
more abstract, features of two natural or formal systems are comparable.3 In the
evolution from empirical laws to universal theories, scientists construct ever more
encompassing models. As long as there is no ‘unified theory’, two different descrip-
tions may be accepted as complementary, as in the case of the wave and the particle
3 In the mathematical idealisation, the two systems are said to be isomorphic.
a) b)
Figure 8.3. Representations of isomorphic relationships between ‘natural’ object and ‘formal’
model system: complementarity (a: left) and analogue or metaphor (b: right) (after Rosen
1985).
description of light (Figure 8.3a). Conversely, two different natural systems may
apparently be so much alike that they can be described with a single model. They
are analogies or, if the isomorphism is rather loose, metaphors (Figure 8.3b).
Analogies and metaphors play an important role in the construction and com-
munication of scientific knowledge. Their use is a powerful heuristic to understand
complex systems in terms of (a model of) a simpler and better understood system. It
allows a connection between empirical observations on the one hand and the logic
of the analog formal model on the other. The ‘borrowing’ of a formal system from
physics and chemistry to describe observations in biology or economic science did
advance science significantly. Famous examples are Hartley’s hypothesis that the
human heart works like a mechanical pump and Laplace’s comparison of the plan-
etary system with a mechanical clock. With the advent of the computer, electrical
networks became the metaphor for the human brain and even for society. Boulding
(1978) used the metaphor of the Cowboy Economy as an image of unlimited eco-
nomic expansion. The image of Spaceship Earth was a symbol of enlightened engin-
eering, often contrasted with the image of Mother Earth.4 The differences between
favourite metaphors reflect divergent interpretations of available knowledge and
divergent value orientations (Table 6.2). Sometimes analogies or metaphors are
merely evocative, such as the human psyche as a steam engine or economic processes
as analogues of chemical and mechanical equilibria. However, their use reflects the
search for universal principles governing the phenomenal world and helps to gen-
erate shared models in, for instance, biology and evolutionary economics and in
immunology, linguistics and institutional economics (Frenken 2006; Janssen 2002).
8.3 Strong and Weak Knowledge

Classical natural science owes its success to the reduction of its domain of investig-
ation to controlled experiments and to the combination of experiments and math-
ematics. The artificial constructs (‘machines’), which nowadays constitute the tech-
nosphere and in numbers exceeds the human population, are the outcome of these
4 Other examples are Awakening Earth to convey an evolution to higher collective consiousness
and Living Machines for technologies that make use of organic processes for manufacturing and
breakdown.
Box 8.2. The mathematisation of science. An important if not quintessen-

tial element of the scientific revolution has been the mathematisation of sci-
ence. It evoked a tension with the experimental tradition, reflecting the diver-
gence between deduction and induction. The French sociologist Bourdieu
argues that ‘la mathématisation est à l’origine de plusieurs phénomènes con-
vergents qui tous tendent à renforcer l’autonomie du monde scientifique et
en particulier de la physique . . . La mathématisation produit d’abord un effet
d’exclusion: avec Newton la mathématisation de la physique tend peu à peu
à instaurer un très forte coupure sociale entre les professionnels et les ama-
teurs . . . Deuxième conséquence de la mathématisation [est] la transforma-
tion de l’idée de l’explication. C’est en calculant que le physicien explique le
monde . . . Ceci entraı̂ne un troisième effet de la mathématisation, ce qu’on peut
appeler la désubstantialisation: la science moderne substitue les relations fonc-
tionnelles, les structures, aux substances aristotéliciennes et c’est la logique de la
manipulation des symboles qui guide les mains du physicien vers des conclusions
nécessaires. L’usage de formulations mathématiques abstraites affaiblit la tend-
ance à concevoir la matière en termes substantiels et conduit à mettre l’accent
sur les aspects relationnels’ (Bourdieu 2001:96–98).
scientific endeavours. The underlying search for more universal laws has led from
coarse-grained, phenomenological descriptions of the world to fine-grained, funda-
mental ones (Table 2.3).
Scientific knowledge comes in grades: Not all knowledge has the same status.
Can one devise a taxonomy of knowledge? The biologist Pantin made, in his book
The Relations between the Sciences (1967), a distinction between the ‘restricted’ and
the ‘unrestricted’ sciences, the latter being those where controlled experiments are
hard or impossible – as in geology, ecology and archaeology. Along this line, we
here hypothesise that a scientific theory develops by strengthening three elements
(Groenewold 1981):
r logical operations (l);
r codified experiences (e); and
r hypotheses (h) which relate the experiences.
The logical operations constitute a more or less formalised system of concepts and
rules. It can be a language syntax, differential-integral calculus or transition rules in a
cellular automata model (§10.5). They are the rules of the formal system (Figure 8.1)
and the language for conversion from conceptual to mathematical model (Figure 8.2).
It is the deductive part of science. For instance, a differential equation to describe
the exponential growth of a deer population is a formalisation of a particular set
of observations, in particular on animal reproduction. The more it is formalised,
the sharper the concepts and rules can be – but also the more the object system
is simplified. The codified experiences refer to human experiences reframed in an
experimental setting. They are the experimental set-up to interact with the natural
system (Figure 8.1) and are the outcome of the conversion from percepts to obser-
vations (Figure 8.2). This is the inductive part of science. For instance, enjoying wild
Box 8.3. On strengthening knowledge. A recent example of how scientific know-

ledge is strengthened and how reference systems matter is from climate change
science. From a natural science perspective, the climate change debate is full
of uncertainties. This is widely acknowledged as one can read in, for instance,
The Economist (August 13, 2005:65): ‘The [climate] system itself is incredibly
complex. There is only one such system, so comparative studies are impossible.
And controlled experiments are equally impossible . . . So there will always be
uncertainty and therefore room for dissent’.
For years, there has been an apparent discrepancy between what is happening
to temperatures at the Earth’s surface and what is happening in the troposphere.
Whereas recorded surface temperatures are rising, observations on troposphere
temperatures from weather balloons and satellites showed constant tropospheric
temperatures since the 1970s. Are the models wrong, are the data wrong – or is
there no warming at all, as climate skeptics say? Recent research suggests that the
data are wrong. Weather balloons often make measurements while being exposed
to sun’s rays and get overheated. This was known and corrected with a correction
factor. However, the correction factor has not been corrected for the design
changes that have been made to reduce the problem of over-heating. A second
error may have to do with the satellite measurements that are used to construct
tropospheric temperatures. The constancy found since the 1970s may be caused
by the change in the satellite’s orbital period, which decays because of friction
with the outer reaches of the atmosphere. If these corrections are introduced,
the trend in tropospheric temperatures is warming and better in agreement with
climate model simulations.
deer in a natural park is not codified but a well-documented count of wild deer in a
certain area and period is. The hypotheses connect and expand the experiences via
logical operations. The observation that the number of offspring of deer fluctuates
with the number of wolves can induce the hypothesis that the two populations are
interacting. The formalised hypotheses yield a scientific model.
A scientific theory becomes mature by gradually eliminating unnecessary hypo-
theses and sharpening the logical operations and the codifications of experience.
Much knowledge in physics and chemistry represents what we henceforth call strong
knowledge.5 It is grown out of hypotheses that have been falsified and replaced by
hypotheses better in agreement with observations. It is formulated in mathematical
models that are thoroughly empirically validated within their domain of observa-
tion and control. Yet, there is often still much scope for improvement. An example
is meteorology, where the combination of satellite data and simulation modelling
has led to an enormous improvement in weather forecasting, which has still not yet
ended.
A theory is strengthened by eliminating weak elements. Sometimes those are
obvious, as in false logic, misapplied statistics or a priori judgments with a claim
5 Often the words hard and soft are used to denote what is meant here with strong and weak. The
strong-weak terminology is preferred, because hard and soft are better used to refer to the degree
to which an aspect of reality can (or cannot) be influence (by men).
on absolute and universal validity. Sometimes their weakness is only obvious if one
demands empirical evidence, as in statements about non-observable entities with no
consequences for experience or in theological dogma’s or revelations. Sometimes,
statements reflect an intuition or speculation, which the Zeitgeist is not yet ready for
or not willing to consider for cultural or political reasons.
The success of the natural sciences and the associated Modernism worldview
suggest that the method of the natural sciences will penetrate all other fields of
knowledge and make them in due course strong and mature (§6.4). In that sense,
calling knowledge in the life and social sciences weak(er) sounds derogatory: It
suggests that scientific knowledge about life and society will become strong(er) with
further domination of the natural sciences. The issue, however, is complexity and
there are two reasons why such domination will not happen soon. First, the method
does not work for the complex systems in the real world. For a large complex system
like the earth atmosphere or parts of the biosphere, it is impossible to conduct
controlled experiments. At best, one can mimic the circumstances and do small-
scale field and laboratory experiments. With economic and social systems, controlled
experiments are hardly or not possible and the very experiment can alter the way
the system functions.6 Scientific statements about such complex systems are often
highly restricted and probabilistic, as there are so many unknown system variables
and interdependencies. Constraining assumptions that are helpful in physics and
chemistry, such as equilibrium and linearity, are in these domains inadequate and
provide pseudo-knowledge because there is no controllable reality in which they can
be tested.
A second reason why the monopolising tendency of the natural sciences has its
limits is that it rejects other sources of knowledge than via the scientific method.
Strong scientific knowledge is active (interaction) and public (shared), but it does
not imply that passive and private knowledge is weak in the sense of false or without
relevant evidence. Weak statements are not ‘truthless’ or useless. They can have
strong subjective meaning and value. Intuitive, speculative or revelatory insights
can also be at the birth of scientific theories. Here the tension between worldviews
is felt: to what extent should knowledge be based on empirical evidence (vertical
dimension) and to what extent should it be public and universal (horizontal dimen-
sion) (Figure 6.5)? This is an important question in the search for sustainability: how
to reconcile the existence of public, scientific knowledge with a universal claim of
validity about the material world with the reality of millions of individuals who have
personal knowledge with claims of validity about their exterior and interior world?
Therefore, statements about complex systems are usually probabilistic and non-
or multicausal. One approach to strengthen them is offered by applying Bayesian
conditional probabilities (Chalmers 1999). Without going into detail, the essence
is that a hypothesis is becoming stronger if the prior probability of being correct
increases with new evidence. The issue then is how to assign prior probabilities
to hypotheses. Most people do not believe this can be done in an objective way.
Prior probabilities represent the beliefs in hypotheses that scientists, as a matter of
fact, state or practise. The Bayesian approach opens the way to a more constructivist,
6 In physics, the interference between observer and observed causes an impossibility to measure
certain properties with arbitrary precision. It is known as the Heisenberg uncertainty principle.
postmodern view of truth. A few examples of statements and assertions can illustrate
the relevance and the problems of this approach.
Statement 1: ‘A wind farm with a wind speed of 6 m/s produces a power of 2 W
per m2 of land area’ (MacKay 2009).
This statement can be tested by setting up an experiment in a windtunnel (the
codified experience). The strength is less in the numbers, which are probably not
exact, than in the underlying universal physical principles. Probably no one would
dispute this statement, but most people would frown if you add ‘yesterday’ or ‘par-
ticularly in the United Kingdom’. An experiment will provide new evidence, which
will strengthen the statement – unless the prior probability assigned to it was already
very high because the physics is known and trusted. If the new evidence does not fall
narrowly within the relevant domain, the ambiguity in the word wind farm would
show up.
Statement 2: ‘A wide range of direct and indirect measurements confirm that the
atmospheric mixing ratio of CO2 has increased globally . . . from a range of 275 to
285 ppm in the pre-industrial era (AD 1000–1750) to 379 ppm in 2005’ (IPCC 2007
WGI:2.3.1).
This carefully framed statement is at the core of the hypothesis that humans
influence the climate. It is based on a series of direct observations and indirect recon-
structions. If one accepts the physics of gaseous diffusion and the methods used in
historical reconstruction, it is a strong scientific statement. The indicated uncertainty
takes into account that direct measurement of pre-industrial concentrations are not
possible.
Statement 3: ‘Work on simplified ecosystems in which the diversity of a single
trophic level . . . is manipulated shows that taxonomic and functional diversity can
enhance ecosystem processes such as primary productivity and nutrient retention’
(Ruiter et al. 2005).
A statement like this may have great importance in judging biodiversity – is
it strong knowledge? The word ecosystem is first used in an experimental frame
(‘manipulated’) and then in a broader context (‘ecosystem processes’). The formula-
tion suggests the need for new evidence from additional experiments to strengthen
the assertion. The notion of diversity, even if strictly defined in the experiment, may
easily be interpreted in a wider context of (bio)diversity and intensive agriculture,
although such generalisations are rather weak. By implication, it becomes a matter
of trust whether a statement like this is used for policy purposes.
Statement 4: ‘Poor [animal] nutrition is one of the major production constraints
in smallholder systems, particularly in Africa’ (Thornton 2010).
The prior probability assigned to this assertion is a difficult matter. For a local
practitioner, it may be an obvious everyday experience – but not all smallholders
will have a shortage of fodder and for them it is difficult to assess. Besides, what
is the precise meaning of ‘one of the major’? For outside experts with a broader
knowledge of smallholder systems (like the author), every smallholder faced with
nutrition problems for his animals will strengthen the statement. However, water
experts may see lots of evidence that water shortage is a production constraints.
With a statement like this, it matters already for its legitimacy who made it. If the
author has a solid reputation, most people will assign a high prior probability to the
statement but any experience to the contrary can weaken it.
Statement 5: ‘ . . . the individual is mainly concerned not with his absolute level
of success, but rather with the difference between his success and a benchmark that
changes over time . . . using economic tools, we argue that [it] can be evolutionarily
advantageous in the sense of improving the individual’s ability to propagate his genes’
(Rayo and Becker 2007).
The first statement is founded on ‘a large body of research’. It speaks of ‘an
individual’ and is, therefore, too general to be falsified. It is unclear how an absolute
level of success and an associated benchmark can be measured. It may be assumed
that such details can be found in the background literature. The association with
evolution is in the form of integrating information in an individual’s happiness
function. It contains sophisticated mathematics and no empirical content. In their
conclusion, the authors are aware of the speculative nature of their model, but do
not question the scientific validity of their approach. An exercise like the one in
this paper may offer insight in human behaviour, if the necessary qualifications are
added. Otherwise, it is more rationalisation than insight.
Statement 6: ‘ . . . the overabundance of young people with advanced education
preceded the political crises of the age of revolutions in Western Europe, in late
Tokugawa Japan and in modern Iran and the Soviet-Union’ (Turchin 2008).
This statement presents a crude correlation between two observations: The age
distribution of populations and political crises in history. It suggests a causal con-
nection. Many people will feel justified to assign a prior probability to the implied
hypothesis – but possibly for quite different reasons because their personal history
matters for the evaluaton of statements like this one. New events like the recent polit-
ical crises in African and Middle Eastern countries may strengthen the hypothesis
for some people, but others remain convinced that rising food prices or ingrained
corruption are the major determinants of crises. Whatever the verdict, observing cor-
relations like these are one of many steps in understanding complex social-ecological
systems.
In evaluating scientific assertions, it is also useful to occasionally shift the focus
towards the person of the scientist and his psychological traits and sociological
configurations. In his essay Science de la science et réflexivité (2002), the sociolo-
gist Bourdieu distinguishes three premises about the ‘scientific enterprise’ that are
good to remember in judging scientific statements. The first premise is that scient-
ists are driven by a reward system. What matters is recognition. Citation indices
and networks are important. It raises also the difficult question how scientists make
choices and deal with conflict. The second one is that science has largely an internal
autonomous dynamic, but this is not smooth. Instead, there are periods in which ‘nor-
mal science’ is suddenly confronted with a revolution. Such a sudden change is called
a paradigm shift by Kuhn in his book, The Structure of Scientific Revolutions (1962).
The internal driving forces are still unclear. A third premise is that science is largely
a ‘contextual game’ in the sense of Wittgenstein’s Philosophische Untersuchingen
(1953). The truth is not ultimate but constantly negotiated in a social domain
of stakeholders. It is helpful to keep these social and psychological mechanisms
8.4 Complexity 227
Box 8.4. The role of disciplinary background: deforestation. A meta-analysis of

the proximate and underlying causes of tropical deforestation has been published
by Geist and Lambin (2001). From a detailed literature search, they identified
three clusters of proximate causes of deforestation, that is, human activities that
directly affect the environment. They also listed five clusters of underlying causes
of deforestation: economic, demographic, technological, policy-institutional and
socio-cultural factors. For the underlying causes, they explored whether the dis-
ciplinary background of the scientific authors had an impact upon drivers and
causes perceived and reported. It is concluded that there is a significant correla-
tion for political scientists and ecologists between their disciplinary background
and the main cause identified, and that research teams that combine natural and
social science views show negligible bias. One may subsume that such correla-
tions are also found in many of the questions about the evolution of complex
social-natural systems. It appears that the present generation of spatial models
contributes to a more thorough and integrated understanding of deforestation
processes, which synthesises the various disciplinary filters.
in scientific practise in mind when dealing with (the ‘production’ of) knowledge in
sustainability science. They incorporate the relativism of Postmodernism.
A painful illustration is the occasional fate of practical knowledge, which in
certain contexts is referred to as indigenous or ‘vernacular’ knowledge. Scientific
knowledge, even if its claims and authority are legitimate, may well be wrong, irrel-
evant or both. There are some stories about how mainstream scientific insights were
forcefully applied in situations for which they turned out to be invalid (Earle 1988).
A corollary is that indigenous knowledge acquired by practise and containing great
accumulated experience and wisdom is often, as it should be, an ingredient of the sci-
entific endeavour. The Indian activist Vandana Shiva expresses this forcefully: ‘My
involvement with the Chipko movement of women protecting their forests . . . had
taught us that the powerless are not powerless due to ignorance but due to the
appropriation of their resources by the powerful . . . literacy is not a prerequisite
for knowledge . . . ordinary tribals, peasants and women have tremendous ecological
knowledge based on their experience. They are biodiversity experts, seed experts,
soil experts, water experts. The blindness of dominant systems to their knowledge
and expertise is not proof of the ignorance of the poor and powerless. It is in fact
proof of the ignorance of the rich and powerful’ (Vandana Shiva 2009).
8.4 Complexity
Several times in this chapter we used the word complex. Did you ever think about
what you mean when you say that a situation, a systems, a person is complex? If
you buy a new mobile phone, you may find it easy and simple to use, but difficult to
understand how it works. The scientist who co-designed it may consider its function-
ing as rather simple. The same situation is often encountered in daily life. You visit
a car mechanic, a medical doctor, a lawyer, and you expect them to handle issues
Box 8.5. Energy and complexity. A perspective that connects energy and com-
plexity is offered by Tainter (2000). ‘Human societies often seem to become
progressively more complex – this is, comprised of more parts, more kinds of
parts, and greater integration of parts. . . . Every increase of complexity has a
cost . . . The cost of supporting complexity is the energy, labour, time, or money
needed to create, maintain, and replace a system that grows to have more and
more parts and transactions, to support specialists, to regulate behaviour so that
the parts of a system all work harmoniously, and to produce and control informa-
tion . . . No society can become more complex without increasing its consumption
of high-quality energy, human labour, time, or money’ (Tainter 2000:6–7; §7.1).
which are (too) complex for you. You rely on an expert who is a professional. It is
tempting to say then that complexity is in the eye of the beholder. In other words,
saying that something is complex gives it a contextual characteristic.
In the last decades, the notion of complexity has emerged from a combination
of mathematics, computer software and applications in various fields of enquiry
denoted as complexity science or complex system science (see Appendix 8.1 for a
brief historical overview and the Suggested Reading). A new language is evolving
around concepts like dissipative structures, emergence and emergent properties,
non-equilibrium systems, self-organisation, self-similarity, non-linearity, bifurcation,
resilience, chaos, sensitivity for initial conditions, decentralised control, distributed
feedbacks and others. They give new and deeper content to concepts that have
become central in 20th-century science such as information, evolution, computation,
order and life (Mitchell 2009).
An oscillating mass at the end of a spring is a classic example of a simple system.
The standard approach works well: It can be subjected to controlled experiments
and it is not needed to introduce external elements to know the movement of the
spring. But, a few coupled springs can already give rise to apparently chaotic beha-
viour (§7.3). Conversely, the complexity of a coastline or of an ice flake turns out
to be apparent because a simple equation can generate similar patterns due to the
property of self-similarity (fractals). However, such deterministic complexity is not
the common connotation of complexity. Mathematicians have proposed the notion
of algorithmic complexity, which links complexity to information content. This is not
very interesting for our discussion either. Physicists have discovered complexity in
simple phenomena such as phase transitions, percolation phenomena and pattern
formation in networks and chemical reactions.7 They emphasise sensitivity for ini-
tial conditions and interdependence as key features of complex systems (Solé and
Goodwin 2000; §7.3). But there is not yet an agreed-upon way to measure complexity
quantitatively.
A less formal approach is to study systems experienced as complex, such as
the immune system, an insect colony or a stock market, and try to list their key
7 Nice examples are the Ising model and the percolator model, which can be inspected with quite a
few other ones in the Model Library of the NetLogo software package.
8.4 Complexity 229
characteristics (Mitchell 2009). The list of key attributes of complex systems usually
comprises elements, relationships, an environment and an internal structure – as do
all systems – plus learning and memory, emergent properties (‘more than the sum of
the parts’), self-organising and dissipative change and evolution (Manson 2001). At
an abstract level, complex systems do have some intriguing properties in common.
This gives rise to one possible definition (Mitchell 2009)8 :
A complex system is a system in which large networks of components with no central

control and simple rules of operation give rise to complex collective behaviour,
sophisticated information processing, and adaptation via learning or evolution.
This rather abstract and somewhat circular definition suggests that complexity can be
defined objectively. This may be incorrect. Perhaps, as suggested before, complexity
can only be defined in relation to the observer.
Two characteristics of complex systems are distinction, that is, variety (or hetero-
geneity), and connection, that is, constraints on parts as the result of interdependence
between parts. These properties are not objective; they depend on what is distin-
guished by and how they are related to the observer. Dimensions of complexity
such as identity and connection depend on what is and can be distinguished by the
observer.9 This view is expressed in most definitions proposed by social scientists,
for instance, Pavard and Dugdale (www.irit.fr):
A complex system is a system for which it is difficult, if not impossible to restrict its
description to a limited number of parameters or characterising variables without
losing [sight of] its essential global functional properties.
It is not this book’s intention to dig deeper into the definition of and the debate
about complex systems. Systems which apparently exhibit behaviour that is not
easily understood, handled or (re)constructed by (most) observers, are complex
systems. They are positioned somewhere between frozen order and chaotic anarchy
where the information content is highest.10 And, paradoxically, their micro-level
complexity sometimes disappears at the phenomenological macrolevel thanks to
their capacity for self-organisation. Whatever the precise definition, sustainability
science is mostly about complex social-ecological systems.
In this book, I use the notion of aggregate complexity in the following way: A
system A is more complex than a system B, if A has:
r more interaction with the environment in terms of the exchange of energy and
matter and thus information;
8 Already in 1995, over 30 definitions of complexity were given (Horgan (1995). For detailed dis-
cussions, see, for instance, the Journal of Artificial Societies and Social Simulation (jasss.soc.
surrey.ac.uk).
9 In anthropology the suffices ‘emic’ and ‘etic’ are introduced to differentiate between descriptions of
a system as seen from one of its members (emic) as against a description from an outsider vantage
point (etic).
10 It is still too early for scientifically sound generalisations. For instance, complexity of a coupled
system may be highest at some medium level of connectivity, where fluctuations do not yet average
out but elements cannot be considered isolated either.
high diversity
attitude:
much interaction
constructivism
weak
science
Figure 8.4. Aggregate complexity as di-
versity of and interaction between sys-
ity
ex
tem elements and as degree of subjective
pl
material immaterial
m
cognition. This illustrative sketch shows
co
objective e subjective
also the distinction between the natural
at
external internal
eg
and the social sciences in the context of

gr
ag
increasing aggregate complexity of the

system under consideration.
strong
science
attitude:
positivism
low diversity
no interaction
r a larger degree of diversity and connectedness of the system elements, both in

number and nature; and
r a larger ability of the system elements to make representations or models of past
and future actions (memory, learning, anticipation and so on).
Following Manson (2001), the discussion focuses on such aggregate complexity. Most
systems investigated in controlled laboratory experiments in physics and chemistry,
and the derived appliances in the technosphere, are of low complexity in this sense.
If experimentation is difficult or impossible or if observer (subject) and observed
(object) interact, the system is of high(er) complexity. Human beings are at the
highest level of complexity in terms of individual diversity, genetic and cultural
codes and interconnections – at least, that is the common presumption. Knowledge
about such systems tends to be weak(er).
These three features of complexity can be visualised with the scheme in Figure
8.4. It sketches two dimensions. The first one refers to complexity in terms of the
number, diversity and heterogeneity of the system’s elements under consideration
and of their interactions – the first two of the three features previously mentioned.
It is the vertical axis in Figure 8.4. Along this dimension, the difficulty to per-
form controlled experiments is one characteristic of complex systems and associated
uncertainties. The second dimension reflects the degree of ‘interiorisation’ in the
sense of an internal, subjective world – the last of the three features previously
mentioned. It is the horizontal axis in Figure 8.4 and represents the spectrum from
external/objective to internal/subjective. This axis is less common in discussions on
complexity. Yet, it is here that the rift between the natural and the social sciences is
felt most intensely (Hollis 2007; Döpfer 2005). It is an essential part of the artificial
intelligence debate: If human agents are included in models, then which ‘interior’
are they given? Chapter 10 reflects on the self-image of man in more detail.
The vertical dimension in Figure 8.4 can be interpreted as describing reality per
se (ontological) and the horizontal axis as the relation of the knower to reality (epi-
stemological). An increase in aggregate complexity of a system can be understood as
8.4 Complexity 231
an evolution to the upper right in Figure 8.4. It can be interpreted in terms of world-
views (§6.3). In the lower left, knowledge is seen through the lense of natural science
and is associated with positivism. The upper right is the domain of subjective and
transcendental knowledge of cosmos, world and life. Constructivism is the dominant
attitude.
With the study of real-world complexity, the role of uncertainty in science
has become bigger and controversial: ‘ . . . [the politicization of uncertainty leads
to] scholastic disputations with the ferocity of sectarian politics. The scientific
inputs . . . have the paradoxical property of promising objectivity and certainty by
their form, but producing only greater contention by their substance’ (Funtowicz
and Ravetz 1990). Some sustainability related issues are at the core of these dis-
putes: human-induced climate change and the risks of nuclear power and GMOs, to
mention a couple. The notion of ‘post-normal science’ has been proposed to deal
with the new situation:
Post-Normal Science has been developed to deal with complex science related issues.
In these, typically facts are uncertain, values in dispute, stakes high, and decisions
urgent, and science is applied to them in conditions that are anything but ‘normal’.
Managers and planners, whether individual or collective (firms, governments), rely

on value-laden interpretations of the situation in the face of irreducible uncertainty
and complexity. Faced with a complex reality, they often fall back on rather simple
‘rule-of-thumb’ recipes rather than on model outcomes. After all, management is
often about complexity reduction, for instance, by controlling or simplifying the sys-
tem and/or its environment. Therefore, in post-normal science, it is argued that for
models to be relevant they need a more comprehensive assessment and participa-
tion of stakeholders. This demands a critical attitude with regard to the status of
knowledge and an alertness for information quality and uncertainty. In post-normal
science, ‘we move from the dream of conclusive scientific demonstration to the
ideal of dialogue for reconciling real antagonisms’ (Ravetz 2006). It implies ‘softer’
forms of interference and control, for instance, in ecosystem management (Walters
1989).
In science, the empirical-reductionist mindset and paradigm are still powerful
and necessary, but they can no longer be our only guide. Despite their success and a
status not unlike that of rituals in pre-scientific cultures, they have to make room for
new ways of observing and experimenting with the use of new and powerful tools.
The availability of cheap and powerful computers, advanced software packages
and communication devices in connection with Internet and social media is already
changing the landscape of scientific knowledge. Table 8.1 contains a list of natural
science oriented methods and tools, with applications relevant for sustainability
science.11 Their use promises a better scientific understanding of (un)sustainable
development processes, because they incorporate features like non-linearity and
feedbacks and engaging models users as participants and stakeholders.
11 See Gilbert and Troitzsch (1999) and Feinstein and Thomas (2002), amongst many other books, on
(quantitative) methods in social science. Table 8.1 does not contain the ‘methods’ in the upper right
corner in Figure 6.5, which are usually considered as non-scientific: intuition, introspection, empathy
and participation, and contemplation and meditation.
Table 8.1. Methods and tools and application fields in complexity science
Method Fields of application
Integral-differential equations analytical models in engineering, environmental and economic

sciences and of population and evolutionary (eco)dynamics
Statistical and factor analysis econometric models; social science relationships and ‘stylised
facts’
Optimisation and control theory, engineering sciences; optimal resource depletion; least-cost
linear/dynamic programming abatement strategies
Linear matrix techniques economic input-output (I-O) theory; food webs
Systems science, systems dynamics, simulation models in resource and environmental economics and
cybernetics management
Catastrophe theory ecosystem dynamics; social [r]evolution
Network (or graph) theory food webs; economic, social and information networks
Game theory social dilemmas; common pool resource (CPR) management
Evolutionary game theory analytical models of species invasion; models in behavioural
economy
Cellular Automata (CA) simulation models of land-use land-cover change (LUCC) and of
urban dynamics
Complex Adaptive Systems (CAS) simulation of social-natural [co-]evolution; optimal strategy search
and Genetic Algorithms (GA) in complex adaptive systems
Multi-Agent Simulation (MAS) or simulation models in evolutionary economics and resource use
Agent-Based Modelling (ABM) dynamics; modelling spread of innovations, diseases,
information (in networks)
Participative: simulation gaming and simulation model of resource use management in interaction with
policy exercises participants; interactive simulation in social dilemma situations
Scenario analysis connecting qualitative storytelling and quantitative modelling for
resource use management
8.5 Metamodels and Organising Concepts

The acquisition of scientific knowledge – strong, public and with a claim to univer-
sality – generates expert knowledge and expert models. Expert models have been
and still are hugely successful in understanding and manipulating the world around
us, but they also demand much personal effort during construction and mainten-
ance. An individual person can only be an expert in one or a few domains, although
the options to acquire and share knowledge are increasing rapidly with the advent
of ICT and Internet. Therefore, most of our knowledge and skills in pre-analytic
everyday life come through observations and interactions with other people. We
memorise and learn to conceptualise them into more abstract schemes, rules and
theories. This ‘everyday’ learning and communicating process is, for most of us, in
constant interaction with the findings of science via the media, colleagues and so
on. We have the privilege that we do not need to discover and reinvent everything
ourselves.
An expert can transform a reference or ‘mother’ expert model into a simplified
version: a metamodel. According to Janssen et al. (2005):
A metamodel is a reduced-form or minimal model in the sense of a simplified and

coarse-grained (in space and time) version of an expert model or a relevant correla-
tion which is probabilistic in nature.
Box 8.6. Mental models, pants and cotton. The Dutch author Bertus Aafjes
tells a beautiful story in his travel account, Morgen bloeien de abrikozen (1954).
He and his wife were invited to visit an English missionary, who lived in a town in
southern Egypt and ran a school for the daughters of the small Christian minority.
Aafjes’ wife was wearing pants, which was most extraordinary in that place and
time – and within hours she was known all over the village as ‘omne pantalone’,
‘the mother of the pants’. The town lived from the cotton harvest in the province,
each and every thing was experienced in terms of cotton. Unfortunately, on the
morning after their arrival, the newspapers announced a further fall of the cotton
price and worse was to come when it appeared that the cotton harvest in Texas
was good beyond expectation.
This is dramatic news for the small community that depends on cotton sales,
and the Muslim cotton traders accused Aafjes’ wife of having the ‘evil eye’ and
threatening their sustenance. The missionary was not willing to give in to this
superstition and neither was Aafjes’ wife. Wherever she appeared – in pants –
people disappeared, as no one risked to be in contact with the evil eye. Cotton
prices kept falling and the situation became precarious. Then, one morning,
everything changed. Aafjes’ wife made her usual promenade and, much to her
surprise, she was invited by every shopkeeper and received all kinds of presents:
‘please be our honourable guest, mother of the pants . . . ’ She returned with lots
of presents and full of surprise. The missionary went out to find the cause of
this sudden change. As it turned out, an announcement had been made at the
townhall that a tornado had brought irreparable damage onto the cotton fields
in Texas. Cotton prices climbed and climbed, and Aafjes’ wife became the most
venerable person in town. Her admirers grew every day in numbers, as did the
presents and invitations.
In some situations, metamodels are more adequate in communication and more

useful for practical applications and policy purposes. Some people prefer the term
archetypical or elementary model. Inside are the explanatory mechanism of the expert
model in reduced form. In their simplest form, metamodels are like a manual or rule
of thumb.
Sometimes, no reference or ‘mother’ expert model exists. Or there are several
expert models around, each with their own expert(s) and content. In such a situation,
no mature expert model may be available in the near future and one has to be satisfied
with probabilistic statements and controversial schools and followers. New evidence
from experiments has to be awaited before the relevant knowledge becomes stronger
by refuting one or more of the hypotheses. In this situation, the available observations
and measurements are often presented as correlations, which appear to capture
salient features but for which no scientifically sound explanation can be given. The
history of science has many examples. It is hoped that the correlation becomes a
law – the scientists’ road to fame. The scientist looking for more evidence will speak
of hypotheses, the economist speaks of stylised facts, and the sceptical enquirer calls
them myths or speculations, Clearly, metamodels can reflect knowledge of quite
different status (strong versus weak).
Metamodels have a life cycle. After a period of excitement and controversy,

either new evidence leads to qualified reformulation and acceptance, or it vanishes.
A reformulation can, for instance, confine the validity of the proclaimed relationship
to only a few countries or periods. It may also be that the issue was ill-phrased in
the first place or that it remains controversial and has to await novel methods or
concepts before any advance is being made. For a while, there can be competing
models around and situations in which each individual has his or her own ‘correct’
model. Uncertainty is exploited to serve particular values and interests and can lead
to the ‘anything goes’ adage of Postmodernism, examples of which are given in the
thematic chapters.
Metamodels should be judged not so much for their scientific truth but for their
usefulness in a given situation. They should be ‘good enough’ – for instance – in
communicating – as basis for acting under uncertainty or as an invitation to do new
experiments or collect new data. It also depends on the user. A scientist judges a
situation in probabilistic terms and refines his model in scholarly dispute. A manager
prefers a more pragmatic interpretation for which a metamodel can provide the best –
and sometimes only – defensible strategy.
There are many relevant metamodels in sustainability science. You have already
encountered several ones, for instance, Zahavi’s law (§2.2). The huge amount of
country data on a variety of issues give rise to coarse-grained country-level hypo-
theses and theories, which have the status of metamodels.12 The chapters on demo-
graphy, agriculture and economy introduce several metamodels in the proper con-
text:
r Population-economy: population growth rates tend to decline with higher
income levels (§10.2)13 ;
r The Engel curve: people spend a lower fraction of their income on food at higher
income levels (§11.2);
r Economic growth and income: GDP-growth is low at very low and very high
income levels, and highest somewhere in-between (§14.1);
r Sectoral shifts: the activity pattern in economies tends to shift from agricultural
to industrial and subsequently to service sector activities – so-called (economic)
structural change (§4.1 and §14.1); and
r Dematerialisation: the use of minerals and energy per unit of GDP tends to rise
and then fall, called the Intensity-of-Use (IU) hypothesis (§14.2).
Income is often the major independent variable in these correlaton-style

metamodels.
The existence and use of metamodels is often a sign of complexity, uncertainty
and controversy. The following example illustrates the point. Since the environment
debate started in the 1960s, opponents of environmental regulation and taxes have
argued that both interfere with economic growth – but how? In the industrialised
countries, the argument was that economic growth is needed to pay for environmental
12 An example are the empirical analysis of country GDP-growth rates versus other variables in Barro
and Sala-i-Martin (2005; Chapters 13 and 14).
13 As to income, this is usually set equal to the ratio of gross domestic product (GDP) and population
of a country, in current or constant US$/cap/yr or in ppp-corrected I$/cap/yr.
Individual Social
development system
Government
system
Human system
Economic Infrastructure
system system
Support system
Environment &
resource system
Natural system
Figure 8.5a. The linkages between resources, economy and society (Bossel 1998).
measures. In less industrialised regions, the argument went differently: We first have
to become rich before we start worrying (and paying) for environmental degradation.
What is the truth about environmental expenditures and economic success? After
years of research and hundreds of publications, there is no conclusive answer at
the general level (Florax and De Groot 2005). The presumed negative correlation
between the strength of environmental policy and intercountry trade does not occur.
Some analyses use input-output tables and regression analysis, others use abstract
theories such as neoclassical trade theory or gravity models. The heterogeneity in the
different analyses is too large to provide a meaningful metamodel. Perhaps, more
data and more variables – think of innovation dynamics – will make it stronger over
time.
A less formal and more abstract way to deal with ill-structured problems in sus-
tainability science is the use of organising concepts – but people also use the words
conceptual model, frame or template. It is a way to frame the issues and express
and communicate ideas and values. Sometimes, organising concepts are close to an
analogue or metaphor and their adequacy and usefulness are valued differently in
different worldviews. The previously introduced regimes and syndromes (§4.5) and
the scheme of aggregate complexity (§8.5) are both examples.
Organising schemes and concepts can be used to analyse change in a struc-
tured way and at different spatial and time scales. Figure 8.5 presents two such
schemes. The first one is ecosystems as life-support systems (Figure 8.5a) and rep-
resents the relationship between ‘nature’ and ‘culture’ and between ecology and
economy (§10.1). Physical nature is seen as the material substratum – or life-support
system – upon which the use of food, water, energy and materials and, ultimately,
all human life is based. At the top are the social and cultural developments of
individuals and society. The institutional arrangements (government) and the eco-
nomic and infrastructure systems connect top and bottom at the intermediate level.
It is a common scheme in the analysis of global change phenomena and the twin
Figure 8.5b. The interactions between population and economy on the one hand and biogeo-
chemical cycles on the other (Rotmans and de Vries 1997).
issues of development and sustainability.14 Another scheme considers population

and economic activities as the drivers behind the need for and supply and use of
food, energy and water, which in turn impacts natural processes (Figure 8.5b). Such
a Pressure-State-Impact-Response (PSIR) scheme is one of the representations of
the ‘nature-culture’ nexus (§10.1). Another example of an organising scheme are
the ecocycles, which indicate how (eco)systems may evolve according to a sequel of
states (§9.7).
An important organising concept in sustainability science is transition. In
Chapter 4, it has already been discussed in the context of the agrarian and
(post)industrial regimes. Here, it is introduced in order to structure complex phe-
nomena at a high level of abstraction:
A transition is an evolutionary process in which social, economic, institutional and
technological structures develop in mutual interaction and change drastically in the
long run.
A formal representation of a transition is the logistic growth process dealt with

in Chapters 2 and 9. Amongst the first transitions identified was the demographic
transition followed by the epidemiological transition. Other transitions are the health
transition, the food or nutrition transition, the land use and forest transitions and the
energy transition and transport transition.
A generic process that is part of all these transitions are technological transitions.
These have been defined as processes of major and long-term changes in techno-
logies to fulfill societal functions such as transportation, housing, communication
and food provision. They not only involve technological changes but also changes in
user practises, regulation, industrial networks, infrastructure and symbolic meaning.
14 The scheme is reminiscent of Marx’ view of a material base or substructure and an ideological
superstructure. It is also reflected in the ‘value pyramid’ proposed by Daly, which arranges knowledge
and actions from ultimate means at the bottom to ultimate ends at the top (Meadows 1998). Both
are expressions of the axis materialist versus mental/spiritual (Table 6.1).
Increasing structuration
of activities in local practices
Socio-technical
landscape
(exogenous
context) Landscape developments
put pressure on existing regime,
which opens up, New regime
creating windows influences
of opportunity for novelties landscape
Markets, user
preferences
Socio-
Industry
technical Science
regime
Policy
Culture
Technology
Socio-technical regime is ‘dynamically stable’. New configuration breaks through, taking
On different dimensions there are ongoing processes advantage of ‘windows of opportunity’.
Adjustments occur in socio-technical regime.
Elements become aligned,

External influences on niches
and stabilise in a dominant design.
(via expectations and networks)
Internal momentum increases.
Niche-
innovations
Small networks of actors support novelties on the basis of expectations and visions.
Learning processes take place on multiple dimensions (co-construction).
Efforts to link different elements in a seamless web.
Time
Figure 8.6. Multi-level perspective on transitions (Geels and Schot 2007, adapted from Geels,
2002:1263).
In economic science, such changes have been studied in the context of capital stock
and factor substitution dynamics, for instance, the gradual change in steelmaking
from the Bessemer to the basic oxygen and later the electric arc process. One of the
first models of technological transitions was the logistic substitution model (Grübler
1999; §14.3). It is based on the simple predator-prey model and shows that there
is a remarkable robustness in such long-term substitution of processes. A broader
conceptual framework is being developed and applied at the interface of sociology,
innovation science and evolutionary economics. It distinguishes a micro-, meso- and
macro-level (Geels 2002; Geels and Schot 2007). At the micro-level, there are pro-
tective niches in which novel technologies (‘configurations that work’) can emerge.
The meso-level of socio-technical regimes is where technological trajectories are con-
solidated and incremental innovations occur. The macro-level is the slowly evolving
landscape of societal change (Figure 8.6).
Box 8.7. Hierarchy and consciousness. Possibly because it is in the nature of

humans to think in terms of hierarchies, many forms of hierarchical order have
been postulated in human history. There are many authors who have suggested
a hierarchy from a worldly, humanist orientation. Besides Marx’s distinction
between infrastructure and superstructure, one can think of Maslow’s spectrum
of needs (§6.1). Jantsch (1980) sees hierarchy in the unfolding of ‘The Self-
organizing Universe’. In the sense of the material-immaterial dimension in Figure
6.5, this insight appears to be part of perennial wisdom (§6.4). Schumacher (1977)
introduced levels of being, with reference to Thomas of Aquino, in his book
A Guide for the Perplexed. Teilhard de Chardin (1959) advanced complexity
and interiorisation as a third dimension alongside the infinitely large and the
infinitesimally small, and the noösphere as part of the process of human evolution.
Aurobindo (1955/1998), Elgin (1993) and others also see an evolution towards
an ever higher consciousness. The Eastern chakra doctrine expresses a similar
insight.
One might postulate an ‘ascending order of complexity’ and associate the levels
along the complexity dimension or axis with an ascending order of intentionality
and consciousness. Whether this is read as an unfolding or push from below as in
the materialist-empiricist orientation, or as a teleological drift upwards as in the
metaphysical spiritual orientations, is as yet an open question for each individual
human – as well as for human society at large.
A series of historical change processes is investigated within this framework, such

as the gradual displacement of sailing ships by steamships and of the horse-drawn
carriage to automobiles. The previously mentioned transitions are all characterised
by coevolution of social and technological forces. An illustration is the gradual shift
from surface water use to piped water and personal hygiene, where engineers and
users (or technology and culture) are both intrinsic to the change process (Geels
2005).
The transition concept is used to guide policies. Transition management emphas-
ises the opportunities to influence technological innovations in a desired direction.
Not unexpectedly, the applications still struggle with some pressing underlying ques-
tions such as: What is the end-point? Is it worldview-dependent? Can we steer
towards it, and if so, how? But the concept is useful in its attempt to transcend
partial and formal (meta)models and frame the broader, integrated perspective. It
also frames a prime aspiration in sustainability science: a sustainability transition
in which society moves towards high-efficiency, low-emission technological systems
and associated economic institutions and social lifestyles (Meadows et al. 1991).
8.6 Science in the Age of Complexity

It is one thing to have reflections on and make models of complex systems, but
quite another to manage such systems. Because they are complex, there are lots
of uncertainties about how they function and respond to interventions. Because
there are uncertainties, there are also controversies and every intervention will be
8.6 Science in the Age of Complexity 239
Figure 8.7. The role of (scientific) knowledge in situations of low versus high consensus on
values and on knowledge, and the corresponding management regimes (de Vries 2006). It
illustrates also a way to connect the science and policy domains.
rationalised within the frame of a worldview. Therefore, we must explicitly con-

sider the role of values and knowledge in decision making with regard to sustain-
able development. This is also the key message of the aforementioned post-normal
science.
A useful metamodel is shown in Figure 8.7 (de Vries 2006; Hisschemöller et al.
2009). The problem under consideration is evaluated in terms of (lack of) consensus
on knowledge and in terms of (lack of) consensus on values. The stronger the
knowledge, the more frequent there is a knowledge consensus. The more dominant
the worldview, the more frequent there is a consensus on values.15 If there is a large
consensus on knowledge and on values (lower left corner), science can be invoked
as a problem solver. It is the domain of applied science. Many environmental models
have served this purpose in the past – think of air pollutant dispersion or process
models for water cleaning. In management science, this is called the rational zone.
The buzzwords are planning, budgets, defined outcomes, and goal-seeking control.
It naturally fits the Modernist worldview. The usual way to deal with uncertainty is
statistics.
If strong knowledge is not (yet) available, but agreement on values prevail,
there is room for alternative models and scientists become advocates in the decision-
making arena (upper left corner). Some elements of the debate on human-induced
climate change are illustrative, for instance, the mechanisms of cloud formation in
climate change or the technical potential of renewable energy sources (Hulme 2009).
The expectation is that, with more research, such controversies will be solved and
15 This is of course a simplification. Often, the power relationships dictate a degree of agreement that
is as much the reflection of compromise and tactics as of genuine value consensus.
action can be taken. However, as long as the situation persists, vocal individual
scientists can play an influential role, and in business this is the corner for inspiring
motivators and visionary leaders.
If there is consensus on knowledge, but not on values, the situation is one of medi-
ation, offering rule systems and evaluation methods (lower right corner). ‘Science’
offers practical assistance to find ‘rational’ ways of managing complexity and uncer-
tainty. Professional consultancy offers value explication and decision support with
tools such as resource accounting, cost-benefit analysis and risk assessment. Also,
participatory methods such as simulation gaming and policy exercises can serve this
purpose (§10.7). These methods primarily explain, clarify and communicate in order
to teach ‘how to play the game’. An example are the various ways to bring about
a reduction in greenhouse gas emissions: Should the focus be on a well-designed
emission trading system or should some kind of equity-based allocation scheme be
negotiated (den Elzen et al. 2008)? Mission statements and partisan positions are
prominent tools to negotiate that which is known into action. This zone is associated
with politics: ‘We know damned well what the problem is, but how can we best
protect our interests?’
The fourth and last position is when there is lack of consensus in both knowledge
and values. The primary role of science here is problem recognition. It appears that
major global change issues such as the risks and consequences of a sudden collapse
of the West Antarctic Ice Sheet (WAIS) or dramatic decline in biodiversity are in
this quadrant. In the eyes of the planner/manager, this is the dramatic chaos zone
(upper right corner). In this corner of controversy on knowledge as well as on values,
the arguments, claims, interests and values on which policy options are based cost
much energy to organise. Here, too, novel interactive and participatory approaches
are useful.
A similar scheme has been introduced by Pielke (2007). It uses two axes: One is
the view of science associated with knowledge con/dissensus, and the other is the view
of democracy akin to value con/dissensus. This leads to four roles for the scientist:
the pure scientist (upper right), the issue advocate (lower right), the science arbiter
(upper left) and the honest broker of policy options (lower left). The importance
of the various schemes is more their usefulness in alerting us to the different role
of science than to their scientific ‘correctness’. Media play an important role in the
dynamics of the discourse.
8.7 Summary Points

Given the interdisciplinary and complex character of sustainability problems, one
must reflect on the nature of (scientific) knowledge. This chapter introduces some
epistemological considerations and basic ideas about models, uncertainty and com-
plexity. Key points to remember are:
r the scientific method is a combination of inductive empiricism and deductive

rationalism – the one cannot be without the other;
r a model maps an object or item in the ‘real world’ into a formal system with
formal (mathematical) rules;
r analogues and metaphors are complementary maps or representations;

r knowledge can be qualified on a scale from strong to weak, on the basis of an
assessment of logical operations, codified experiences and hypotheses;
r complexity can be defined in various ways, but aggregate complexity is primarily
an increase in the diversity of and interactions between system elements;
r metamodels are a way to represent and communicate knowledge about complex
systems (in expert models or as black box correlations); and
r organising concepts, such as transition, play a useful role in structuring the
investigation of complex systems.
One of the impulses behind science is the desire to gain reliable knowledge about
the world so that we can control it.
– Solé and Goodwin, 2000:1
For the Chinese the bamboo expresses the will to survive, the spirit to endure under
adverse circumstances. Great trees in their strength resist the winds and are broken;
but the pliant, yielding bamboo, twirled and tossed about madly in the storm, bends
and bows unresting, and survives.
– Diana Kan, Chinese Painting, 1979:45
Fire can burn but cannot move.

Wind can move but cannot
burn. Do men know it’s like
that with knowing and doing?
Till fire joins wind it cannot take a step.
– Devara Dasimayya, in: Speaking of Siva (Vs. 127).
Ed. A.K. Ramanujan. Penguin Classics 1973
. . . each of us knows exactly one mind from the inside, and no two of us know the
same mind from the inside. No other kind of thing is known about in that way.
– Dennett, Kinds of Minds: The Origins of Consciousness, 2001:2
SUGGESTED READING
A helpful guide about the status of scientific knowledge and the philosophy of science from a
natural science perspective.
Chalmers, R. What Is this Thing Called Science?, 3rd. ed. Queensland University Press/Open
An interesting collection of controversial views of U.S. authors, clarifying divergent worldviews.
Easton, T. Taking Sides – Clashing Views on Environmental Issues, 11th ed. Contemporary
Learning Series. New York: McGraw-Hill, 2006.
This paper gives sixteen reasons, other than prediction, to build models.
Epstein, J. Why Model? (2008). jasss.soc.surrey.ac.uk/11/4/12.html.
A thorough introduction into the philosophy of science from a social science perspective.
Hollis, M. The Philosophy of Science – An introduction. New York: Cambridge University
Press 2007.
A systematic introduction into what complexity is on the basis of a set of concepts and theories
in 20th-century science.
Mitchell, M. Complexity – A Guided Tour. New York: Oxford University Press, 2009.
Contributions and applications in the field of sustainable development, notably ecology and
anthropology.
Norberg, J., and G. Cummings, eds. Complexity Theory for a Sustainable Future. New York:
Columbia University Press, 2008.
A not uncontroversial but concise introduction in the rationale for a post-normal science.
Ravetz, J. The No-Nonsense Guide to Science. Oxford: New Internationalist,
2006.
An inspiring exploration of complexity in a variety of models and systems.
Solé, R., and B. Goodwin. Signs of Life – How Complexity Pervades Biology. New York:
Basic Books, 2000.
An introduction into environmental modelling with a stepwise introduction of complex system
approaches.
Wainwright, J., and M. Mulligan, eds. Environmental Modelling – Finding Simplicity in Com-
plexity. London: John Wiley & Sons, Ltd., 2004.
USEFUL WEBSITES
r www.bayesian.org/bayesexp/bayesexp.html is the website of the International Society of
Bayesian Statistics and applications.
r www.nusap.net/ is the NUSAP website on uncertainty in post-normal science.
r math.rice.edu/∼lanius/frac/ offers a website with fractal geometry examples; there are
many more on this and other complex system science theories and applications.
r www.inclusivescience.org/ presents ideas and principles regarding an ‘inclusive science’
in a/the New Age.
r www.image.nl/fair is the website of the FAIR emission allocation model.
Appendix 8.1 A Brief History of Complex Systems Science

Complex systems science (CSS) emerged in the 1970s and 1980s. In the wake of
research by early 20th-century mathematicians like Poincaré and Julia, rigorous
experimentation and mathematical analysis in combination with rapidly advancing
computing power resulted in several novel approaches to the study of real-world
complexity. In the 1960s and 1970s, Thom and Zeeman developed catastrophe the-
ory, showing how the qualitative nature of system behaviour depends on the equa-
tion parameters and introducing nonlinearities in a variety of models with broad
applicability. In the 1970s, Mandelbrot discovered fractal geometry while studying
complexity and chaos, of stock markets, amongst other phenomena, and increasing
computing power permitted to draw fascinating graphs. In 1976, May published his
discovery that simple discrete mathematical models could generate chaotic beha-
viour.
In 1980, Prigogine published the book From Being to Becoming, followed by the
book Order Out of Chaos, written with his colleague Stengers and based on earlier
work on irreversible thermodynamics. In this and his 1987 book, Die Erforschung
des Komplexen, with Nicolis, he announced the end of the Cartesian-Newtonian
paradigm. In retrospect, it was only one of the 20th-century events that signalled the
end of ‘modern’ science and the advent of ‘post-modern’ science.
Since the 1990s, the pathbreaking research has occurred in the life and social
sciences. Kauffman proposed ideas about how life could have evolved from self-
organising chemical reactions in his book, At Home in the Universe (1995). An
Appendix 8.1 A Brief History of Complex Systems Science 243
early application of game theory to animal behaviour introduced a completely new

understanding of complex dynamics in the book, Evolution and Theory of Games
(Maynard-Smith 1982), which has further been developed into the subdiscipline of
Evolutionary Dynamics (Nowak 2006). Watts and colleagues attracted attention with
the book Small Worlds (1999), which explores the science behind the phenomenon
that any two people, selected randomly from almost anywhere, are ‘connected’ via
a chain of only a few, six or seven, intermediate acquaintances. Graph theory got a
further boost when computing power permitted network analyses of large networks
such as citations, the World Wide Web and metabolic reaction schemes.
In the 1980s and 1990s, complexity science has become popular by some events
and books. Several books popularised the ideas of chaos and complexity and the
scientists around it, notably at the Santa Fe Institute of Complexity (www.santafe
.edu). To mention a few, Gleick’s Chaos (1988) made the new science the scene of
enthralling and personal theatre; Waldrop popularised notions such as the edge of
chaos in his book Complexity (1992); Gladwell wrote his book The Tipping Point
(2000) ‘on how little things can make a big difference’ after being inspired by the
nonlinearities and shocks of infectious disease epidemics.
9 Land and Nature
9.1 Introduction
At the dawn of the environmental movement in the 1960s and 1970s, there was
increasing concern about the damage done by humans to nature – and nature pro-
tection and conservation were the stated goals. Ecology and the emerging environ-
mental sciences were at the forefront. With the advent of the idea of sustainable
development in the 1980s, more emphasis was put on the legitimate aspirations
of many people to (material) well-being, that is, development. There was also an
increasing realisation that natural systems are always changing and evolving, and that
not much ‘undisturbed nature’ had been left after millennia of human evolution.
Nevertheless, ecology is still considered by many the core of sustainability sci-
ence. The word is derived from the Greek oικoσ , house, and λoγ oσ , reason or idea,
and it is, in a broad sense, the art and science of seeing things as a whole. As such,
it has its formal scientific offshoots in theoretical and systems ecology, and its more
social and transcendent expressions in social and human ecology. Its core idea has
also shaped new bridging disciplines like landscape ecology and environmental and
ecological economics. Ecology supports a rich interpretation of sustainability and
a broad view of the environment-development nexus. It should not be confounded
with environmental science, which has branched out in more practical and applied
forms across various disciplines (chemistry, economics and others). Ecology is cru-
cial for understanding the theory and practise of sustainable development, and more
than one chapter should be devoted to it. This is not possible so I refer to some
textbooks in the Suggested Reading.
Another core science in sustainability science is geography or, more broadly,
the earth sciences. Life has evolved in the biosphere in interaction with geological
change processes (§5.2). The resulting landscapes have been classified in terms of
soils, vegetation and other characteristic variables. The dynamic changes in land-
scapes and the general principles and mechanisms behind them are the topic of
land change science (Lambin and Geist 2006). Large parts of the natural world can
apparently be understood on the basis of a rather small set of physical-chemical prin-
ciples and exploited and (re)designed with help of the engineering (mineral, hydro-
logical and civic) sciences. Amplified by the exponential growth in population and
244
9.2 Earth: Soil Climate Vegetation Maps 245
productivity, hardly any landscape on earth has remained untouched by human activ-
ity. This chapter investigates the natural and anthropogenic changes in landscapes
and ecosystems.
9.2 Earth: Soil Climate Vegetation Maps

The interplay between geological features and forces and life as it manifests itself
in ecosystems is a necessary part of understanding (un)sustainable development. In
this section, we briefly introduce a few classifications and change processes in the
geosciences. Soils are the outcome of aeons of geological change processes and the
processes of life, including human life. The different soil types reflect differences
in chemical composition, topsoil layer thickness, degree of weathering and so on.
In combination with the incoming solar energy and its derivatives (such as winds,
clouds, precipitation and vegetation), soils are an integral part of landscapes and
climate. Figure 9.1 shows the FAO world soil map. Figure 9.2 shows the maps for
two important climate parameters: average annual surface temperature and average
annual precipitation. One should be aware that averages are constructed from a –
sometimes sparse – network of observation points. They conceal the large differences
in space and time, and local heterogeneities and monthly or even daily or hourly
fluctuations are often of great importance from a sustainable development perspect-
ive. Increasingly, with the development of geographical information systems (GIS)
and associated software, global and regional maps and time-series are available on
the Internet on geology, soils and climate. The interested reader can look up the
suggested websites.
The three maps in Figures 9.1 and 9.2 are a representation of the abiotic envir-
onment on which life can flourish in all its forms. It is the basis for the land cover
change (LCC) classification. Oldest and still widely used is the life zone classifica-
tion system proposed by Holdridge in 1947. It distinguishes thirty-three classes on
the basis of two dimensions: annual precipitation rate (rain in mm/yr) and potential
evapotranspiration (ratio of water loss in evaporation/transpiration and precipita-
tion) (Figure 9.3). If precipitation is high but the potential evapotranspiration is low,
you are in the humid tropical forests (lower right corner). At low latitudes and high
temperatures, these ecosystems fix large amounts of water in soil and vegetation. If
temperatures are high but there is not much rainfall, the evaporation rate is high
and you are in a (semi-)arid desert (lower left corner). The little water there is in the
system is not easily contained. Little rainfall and low temperatures are characteristic
of the polar deserts (upper corner). The various classes coincide with two features
mentioned before: geographical location (latitude) and elevation (altitude), both
are shown in Figure 9.3. A bioclimatic classification like this is useful to investigate
potential vegetation, that is, the vegetation that one would expect without human
impact. Of course, there will always be local variety due to the natural gradients in
altitude, temperature, soils and so on.
The Holdridge classification is used in several global change models to per-
form integrated assessment of atmosphere-biosphere interactions. For instance, the
IMAGE-model maps land cover and use changes in combination with greenhouse
gas emissions and climate change (Figure 9.3; Bouwman et al. 2006). Assigning to
246
FAO Soil map of the World
Acrisols Rendzinas Lithosols Greyzems Arenosols Rankers Yermosols

Cambrisols Ferrasols Fluvisols Nitosols Regosols Vertisols Solonchaks
Chernozems Gleysols Kastanozems Histosols Solonetz Planosols SALT
Podzoluvisols Phaeozems Luvisols Podzols Andosols Xerosols WATER
Figure 9.1. Soil map of Earth (source of map: FAO). (See color plate.)
Global average annual temperature, 1990
degrees Celsius
–27 - –20 –9.9 - –5 5.01 - 10 20.1 - 25
–19.9 - –15 –4.99 - 0 10.1 - 15 25.1 - 30.9
–14.9 - –10 0.001 - 5 15.1 - 20
Figure 9.2a. Climate map of Earth: annual average temperature and rainfall (source of maps: PBL). (See color plate.)
247
248
Global average precipitation patterns 1990
mm/yr
< 20 51 - 100 201 - 500 1,001 - 1,500 2,001 - 4,000
21 - 50 101 - 200 501 - 1,000 1,501 - 2,000 > 4,000
Figure 9.2b (continued) (See color plate.)
0.1
.5
25
62
io
an
rat
0.2
nu
5
12
5
n
al
tio
pre
ira
0
0.5
25
latitudinal altitudinal
cip
sp
regions belts
ita
tra
ti
50
on
po
1
polar desert desert desert alvar
va
(mm
00
1.5 °C
le
10
2
dry moist wet rain
tia
biotemperature
subpolar tundra tundra tundra tundra
alpine
ten
00
3 °C
20
po
4 dry moist wet rain

boreal desert
scrub forest forest forest subalpine
00
6 °C
40
8
desert moist wet rain

cool temperate desert
scrub
steppe
forest forest forest
montane
00
12 °C
16
80
warm temperate desert
thorn
dry moist wet rain
lower montane
desert steppe /
scrub forest forest forest forest
subtropical woodland premontane
0
00
24 °C
desert thorn dry moist wet rain
32
very dry
16
tropical desert
scrub woodland forest forest forest forest forest
critical
super- perarid arid semi- sub- humid per- super-
temperature
arid arid humid humid humid line
humidity provinces
Figure 9.3. Life zone classification on the basis of annual precipitation rate and potential evapotranspiration ratio (source: Peter
Haiasz, creatice commons).
249
250 Land and Nature
the crude LCC-classes average characteristic properties such as carbon uptake, it

allows an assessment of the role of the biosphere in climate processes. Figure 9.4 is
a more empirical and refined way to construct such a map from the growing number
of satellite data. It shows a potential vegetation map constructed from satellite and
census data and after eliminating human influences (Ramankutty and Foley 1999).
Scientists attempt to simulate the mechanisms of land and soil change processes
in order to understand and predict their dynamic evolution in global change integ-
rated assessment models (GC-IAMs). The models are improved continuously by
adding more detail such as plant types, and by increasing resolution, for example,
from monthly to hourly or even shorter timesteps and to smaller spatial units. This
permits a more detailed investigation of the complex interactions between vegeta-
tion, climate and climate change–inducing greenhouse gas concentrations. This topic
could fill a whole book in itself. In this chapter the discussion is confined to a few
aspects of undisturbed ‘nature’: what kind of processes occur and how do human
interventions affect ecosystems and their ‘services’? But, first, a few narratives from
the real world.
Box 9.1. Universal and global processes. Natural science principles have led to
an abstract, formal understanding of real-world phenomena at the micro-level, in
the form of a limited number of generic processes, which are widely taught and
used in environmental science, engineering and management. Such processes can
occur in many places on Earth, but they are local/regional in scale and appearance.
They are called universal to distinguish them from processes that involve the earth
system as a whole. The latter obey the same laws but happen on a much larger
scale, which is why they are called global. The distinction is not sharp, but scale
does matter.
Soil erosion due to wind and water, evapotranspiration in vegetation and the
dispersion of combustion products are ubiquitous and thus universal. In the 1960s,
it was discovered that the emission of sulphur and nitrogen oxides in countries
like the United Kingdom and The Netherlands acidified the lakes in Sweden. It
was a regional-scale phenomenon. Similarly, the dust storms in China and Africa
have reached regional-scale proportions and now also affect California and Latin
America. One of their causes may be the clearing of tropical forests. These pro-
cesses are (or are becoming) global (change) processes. Widespread but local
emission of chlorofluorocarbon (CFC) compounds penetrated the stratosphere
and broke down ozone molecules, causing the global phenomenon of the ‘ozone
hole’. When emitted from fossil fuel combustion and land use activities, green-
house gases like carbon dioxide spread quickly through the atmosphere and thus
cause a global increase in concentration, which in turn affects the radiation bal-
ance of the Earth – creating a global phenomenon. In turn, the resulting changes
in temperature and precipitation affect local conditions. The existence of univer-
sal change processes as part of global change phenomena is a characteristic setting
for many sustainability science issues. This micro-macro aspect is also referred to
as ‘nested dynamics’.
Ice Boreal forest Temp.deciduous forest Hot desert Tropical woodlands
Tundra Cool confier forest Warm mixed forest Scrublands Tropical forest
Wooded tundra Temp.mixed forest Grasslands/steppe Savanna
Figure 9.4. Potential vegetation cover based on Holdridge life zone classification (12 classes) (source of map: PBL/www.mnp.nl\image).
(See color plate.)
251
252
Trop. Evergr. Forest/Woodland Boreal Evergr. Forest/Woodland Dense Shrubland

Trop. Decid. Forest/Woodland Boreal Decid. Forest/Woodland Open Shrubland
Temp. Broadl. Evergr. Forest/Woodland Evergr./Decid. Mixed Forest/Woodland Tundra
Temp. Ndleaf. Evergr. Forest/Woodland Savanna Hot Desert
Temp. Decid. Forest/Woodland Grassland/Steppe Polar desert/Rock/Ice
Figure 9.5. Land cover map (15 classes) (source of map: Ramankutty and Foley 1999). (See color plate.)
9.3 Stories 253
9.3 Stories
9.3.1 Forest Fires1

One of the causes of forest destruction is fire. It is a natural phenomenon, and it
is instrumental in regulating ecosystems and rejuvenates their biodiversity. Natural
fires are mostly triggered by lightning, other fires are caused by agricultural practises
and other human activities. What are the impacts of fire? In Mediterranean Europe,
the annual area burnt and the average fire size has roughly been constant since
1990 and the number of fires is declining since 2000. However, they can still have
devastating impacts.
A group of French researchers has investigated representative Mediterranean
forest areas in the Provence (oak and pine trees). They compared the vegetation,
soil and biodiversity in areas where no fire had occurred for fifty years or more with
the same variables in areas where up to five fires had taken place in rather short
periods. It turns out that after some fifty undisturbed years, the forest has built up
some resilience against disturbances. The structure of the original forest has only
come back after 150 to 200 years. During those first fifty years of recovery, multiple
fire events can permanently degrade the potential of the ecosystem. The risk of
permanent damage is increased by the more frequent droughts as a consequence
of, most likely, climate change. There is a threshold effect: ‘A forest can recover
from a succession of three fires in 50 years (one every 25 years on average), but
the fourth one is critical . . . The combined effects of recurrent fires and droughts are
devastating’, according to the study coordinator Vennetier. What to do? Protecting
with priority the oldest forests (150–200 years), which are rare and contain species
that might get extinct, is one option. Better protection against fires is another one.
If unmanaged, the fires will transform the landscape into one of bushes (maquis and
garrigue) with considerable loss of flora and fauna.
It is difficult to get strong knowledge about the relationship between size and
frequency of forest fires and climate change. In June 2010, a ‘blocking anticyclone’
caused a heat wave of unprecedented intensity and duration in the southern part of
Russia and in eastern Ukraine. It is still unclear whether there is relation with climate
change. Summer temperature in large cities such as Moscow and Kazan had reached
39◦ C. In August, temperatures started to drop and the month was not the hottest
ever, but very dry. Human activity and natural processes, possibly self-ignition in the
extremely dry and hot conditions, caused huge fires in the forests and the peat-bogs.
The data are controversial, because of different definitions, measuring methods and
possibly distortion of information. Official data speak of an affected area of 1 to
1.5 million hectare (Mha), whereas satellite data from scientific organisations give
estimates of 5 to 12 Mha. Hundreds of thousands of professionals and volunteers were
involved in combating the fires and rescuing people. Liquidation of state protection
of forest and peat bogs, termination of work to prevent fires and abolishment of
monitoring and suppression of fires at early stages became the reason of catastrophic
1 Le Hir in Le Monde 4 août 2009. See also effis.jrc.ec.europa.eu/. The text on forest fires in Russia is
written by Sergey Minosyants, Ph.D. student at Mendeleyev University of Chemical Technology of
Russia and based on, The Conclusion of the Public commission on investigation of the reasons and
consequences of natural fires in Russia in 2010 (September 30, 2010).
254 Land and Nature
Figure 9.6. Nature reserves in North French-Guiana. The Trésor Reserve borders the twenty-
five times bigger Kaw-Roura (moerasbos: swamp forest; hellingbos: slope forest; natte
savanne: wet savanna; plateaubos: plateau forest). (See color plate.)
fires of forest and peat bogs. Preliminary estimates of the damage indicate that the
economic costs might exceed U.S. $375 billion or in the order of 25 percent of annual
GDP. Besides, an estimated 30–100 Mt CO2 were emitted and hospitalisation and
mortality rates increased significantly during the heat wave and the fires.
9.3.2 Biodiversity in South America: The Trésor Project2

The Trésor Foundation was founded in 1995 by enthusiastic and committed individu-
als at the University of Utrecht, the Dutch, French and U.S. branches of the World
Wildlife Fund (WWF) and the Dutch business community. The Foundation’s mission
is: to conserve biodiversity and develop sustainable land use in a way that is equally
beneficial to all the parties directly and indirectly involved, by means of cooperation,
consultation and local participation. The Foundation bought 2,600 hectare (ha) of
rainforest from the French-Guianese diocese in the northeast of French-Guiana and
later expanded with another 1,400 ha (Figure 9.6). The university made an assess-
ment of the area’s carbon storage capacity, which motivated companies to participate
with an investment to compensate part of their carbon emissions and as a way of
branding.
A ‘commercial’ nature reserve was born with three objectives. The first one is
protection of the biological diversity and the existing ecosystems in and nearby the
park. The second one is stimulation and support of scientific research in tropical
systems. The third one is providing botanic and environmental education by starting
or contributing to educational programmes for schools and visitors of the Trésor
Reserve. The park management is done by a voluntary local board (Association
Trésor) with three full-time employees at their service. Because the park is being
granted an official national park status (Réserve Naturelle Regionale) as a token of
social appreciation, their salaries are now paid for by the regional government.
Some original fundraising methods were created to finance the park and the
activities. A rainforest adoption system was created that makes it possible for private
2 This story is written by Joeri Zwerts and Vijko Lukkien at Utrecht University. See also www.iucn
.nl/about us/members 1/trsor foundation/.
9.3 Stories 255
contributors to adopt square meters of the reserve for a certain amount of money and
a certificate in return. In this way, donating money becomes a more personal affair,
even though the certificate gives no further rights to its owner, and the management
of the reserve can remain simple. Compensating for greenhouse gas emissions is
another way of fundraising. Several university departments compensate their CO2
emissions from flight trips with donations to the Trésor Foundation. For the future,
management seeks to facilitate corporate research for bioprospecting, as relevant
knowledge of the area can significantly reduce a company’s investment costs prior to
the research. In return, an agreement will be made to finance further development
of the area.
The project has large positive effects for the people in the area and as a result
receives much local support. The area gains much (inter)national interest and a local
ecotourism industry is developing. Investments create revenues and employment for
the local population. One example is the production of durable wooden signposts
for alongside the walking trails: The company that was asked to make them now also
provides them to other reserves across the continent. It illustrates the positive link
between local economic growth and nature preservation when park management
involves the local population. The neighbouring, and twenty-five times larger, Kaw-
Roura Reserve now also obtained official protected status, as a result of the example
set by Trésor.
The project illustrates what can be achieved by a number of committed and
knowledgeable individuals and by collaboration between local communities, envir-
onmentalists, governments, science and business. The central message is that in order
to accomplish your objectives, intensive, innovative and adaptable management is a
must. Poverty alleviation and education play a central role and respect for the local
situation is key because no two forests and no two cultures of the people that inhabit
them are alike. For the Foundation itself, after fifteen years, the future perspectives
are upscaling the reserve and disseminating knowledge to serve the general cause of
nature conservation.
9.3.3 Mining in Papua New Guinea3

Papua New Guinea is one of the world’s largest islands. Called the ‘Last Great
Place’, it is home to hundreds of unique species of animals and plants as well as to
an upward of 820 languages. The Porgera gold mine is situated in the highlands. It
produced around 18 tons of gold per year and more than U.S. $1 billion of profits in
2006, according to Barrick Gold, a Canadian corporation that assumed a majority
share of the mine. Barrick operates twenty-six mines worldwide and boasts of having
the industry’s largest reserves.
Porgera, New Guinea’s biggest gold mine, accounts for 72 percent of the coun-
try’s export earnings. It utilises the most advanced extraction technologies and heli-
copters fly people and gold in and out on a daily basis. Its extraction process creates
cyanide-laced wastewater that the company discharges directly into the local river
system. One millionth of a gram of cyanide in a litre of water is enough to kill the
fish. Because Porgera is literally at the top of the country, the streams into which
3 Part of this information is from the site http://www.corpwatch.org/article.php?id=14381.
256 Land and Nature
the toxins are dumped flow into many other tributaries before they reach the sea.
After fourteen years, the mine waste has slowly torn the hills from under the local
inhabitants and turned the small valley below into a choked river of dirt creeping
toward the Gulf of Papua 200 miles away. The large rainforests and its inhabitants
are also under threat: Half of the forests will disappear by 2020, if present defor-
estation rates continue. The causes of deforestation are exploitation for timber and
clearing by small farmers.
Since the 1970s, in the Indonesian part of New Guinea, the province Irian Jaya,
the American mining company Freeport has been exploiting some of the world’s
largest reserves of copper and gold. Here, too, huge profits are made while eco-
systems are destroyed. Massive deforestation for infrastructure took place. Almost
all processed material is released and fills up several square kilometres at the mouth
of the nearby river. These mines are part of what the United Nations Industrial
Development Organizations (UNIDO) calls a ‘gold rush in the Third World’ that
began in the 1980s and spread to Tanzania, Surinam and many other countries.
Mining for gold is one of the world’s most grotesque industries, consuming vast
resources and producing mountains of waste to produce a small amount of soft,
pliable metal with few practical uses. To make one gold wedding band, at least
twenty tons of earth must be excavated.
9.4 Land Cover Change and Degradation

Two key ingredients of ecosystems are soils and vegetation. Soils are the outcome
of very complex processes over long periods of time. They are at the interface
of the lithosphere, biosphere, hydrosphere and atmosphere and are under natural
conditions in dynamic equilibrium with climate and vegetation. Their classification
is based on a series of constituent characteristics, such as parent material, chemical
composition and physical features, topography and regional climate and vegetation
(Figure 9.1). Vegetation is the cover of plants and trees on the Earth’s surface.
It ranges from a small number of species in simple arrangements to very diverse
and complex ecosystems, depending on the prevailing conditions. Vegetation is
characterised by different spatial features and changes over different time scales. The
gradients in climate variables such as temperature and precipitation and in elevation
are the major determinants of spatial heterogeneity. Together, soils and vegetation
constitute the larger part of the biospheric stocks and flows of mass and energy.
An essential factor in soil-vegetation systems is water. Indeed, the role of water
in all its forms is so diverse and pervasive that it is hard to give a concise description.
The system of water stocks and flows is called the hydrological cycle (Figure 9.7). By
far, the largest amount of water on Earth is stored in the oceans. Other important
stocks are ice and groundwater. The major flows are evaporation and subsequent
transport and precipitation. With temporary storage in snow, soils and vegetation
(the equivalent of delays in a system dynamics context), water flows back to the
oceans via the several hundreds of river basins. These flows are large in absolute
amounts and in a human context – up to hundreds of cubic km per year (km3 /yr) – but
tiny compared to the three aforementioned stocks (<2 percent). I refer the reader to
the Suggested Reading and Useful Websites for more detailed information. Human
use of water is dealt with in §12.3.
Figure 9.7. The hydrological cycle (source of data: www.cgd.ucar.edu). Units are thousand
(103 ) km3 .
A major sustainability concern is the degradation of landscapes and, in partic-

ular, soils, with important consequences for ecosystem functioning. Human inter-
ference with soils, vegetation and water has occurred for millennia. With the ongo-
ing population growth and the advent of industrial agriculture, interventions have
increased in extent and intensity. Soils are said to degrade. But what is land degrad-
ation and to what extent is it human-induced? Also, how will it evolve under the
increasing pressures of human activity and climate change?
In many landscapes erosion by the forces of water, wind and gravitation is at
work as a perennial force behind changes in land cover. There has always been
erosion, as it is one of the geomorphological forces. It is a fundamental and complex
natural process. Physically speaking, erosion is the detachment of particles of soil and
surface sediments and rocks. The ultimate driving force of erosion is tectonic action,
which lifts the Earth’s surface from which material can then be eroded. It is the basic
formation mechanism of sediment and is, therefore, at the origin of much cultivable
land and fertile deltas. A second influence comes from climate, in particular, rainfall
intensity and rainfall pattern in interaction with vegetation. The third factor – which
is usually the focus in sustainable development contexts – is erosion induced by
human activities such as burning forest, ploughing and animal grazing. These are
in some areas like the Mediterranean and Asian river deltas age-old forces shaping
the landscape, with overgrazing and fire amongst the most influential processes. It is
difficult, however, to separate human-induced erosion from natural erosion.
Erosion and its impacts tend to be greatest in the so-called drylands (hyper-arid,
arid, semi-arid and dry subhumid), as these are most vulnerable with their poorly
developed soils and little vegetation. These drylands make up some 25 to 40 percent
of the earth surface and are habitat to one-fifth of the world population. They are
prone to desertification, which narrowly defined is a process of increasing aridity
and gradual dominance of abiotic over biotic processes (Okin 2002). Desertification
258 Land and Nature
has three dimensions: meteorological, ecological and human (Reynolds and Stafford
Smith 2002). The seasonality and frequency of (extreme) rain events are an important
factor in fragile drylands, and there are still large uncertainties in the causes of
variability in precipitation over the years. The natural vegetation in drylands is a
mixture of grasslands, shrublands and savannas, and the relatively unprotected soils
are poor in organic matter. Their use as rainfed or irrigated cropland and, much more
important, as rangeland can easily degrade soils by tillage and grazing. The human or
socio-economic dimension, therefore, often plays a large role, but in different forms
and at different scales. Debt and market failure are direct and immediate causes of
degradation, for instance, in the form of local overgrazing. Tourism and trade are
indirect driving forces.
One of the processes leading to desertification is salinisation. Semi-arid and arid
regions tend to be naturally salty, because evapotranspiration exceeds precipitation
and there is not enough water to leach out the soluble soil material. Salts from rivers
and sea intrusion and atmospheric deposition are not carried away to rivers and
seas and accumulates in the soil. This is exacerbated in closed drainage basins.
Human activity, notably water diversion for irrigation and dam construction, can
intensify salt accumulation. Improper management of irrigated soils may accelerate
it. Human-induced salinisation happened already in the early Mesopotamian civil-
isations (§3.3). An estimated 10 percent of the world’s irrigated area for crops is
severely degraded from salinisation.
One of the many definitions of land degradation is given in the UN Convention
to Combat Desertification and Land Degradation (UNCCD), often referred to as
the ‘desertification convention’ (Okin 2002):
[land degradation] is the reduction or loss of the biological and economic productivity
and complexity of terrestrial ecosystems, including soils, vegetation, other biota, and
the ecological, biogeochemical and hydrological processes that operate therein, in
arid and semi-arid lands, resulting from various factors including climatic variations
and human activities.
There is still little reliable information on land degradation globally. A map of

human-induced land degradation was made in the late 1980s by Oldeman et al. (1990)
at the International Soil Reference and Information Centre (ISRIC) as part of the
Global Assessment of Soil Degradation (GLASOD) project. It was based on surveys
amongst national experts on type (water erosion, wind erosion, chemical deteriora-
tion, physical deterioration and so on), degree (light, moderate, strong or extreme),
relative extent and causative factors of degradation. The latter comprised defor-
estation, overgrazing, agricultural activities (improper agricultural management),
overexploitation of vegetation (cutting for fuelwood and so on) and industrial activ-
ities (pollution). The GLASOD-analysis indicated that almost 2 Gha or 15 percent
of the land surface was degraded, of which four-fifths was due to erosion.4 There
are only a few other data sources on land degradation. One study estimates the
area affected by water and soil erosion at 1,64 Gha or 12 percent of the land area
(Leemans and Kleidon 2002). The Millennium Ecosystem Assessment (MEA 2005)
gives similar qualitative and quantitative estimates in its investigation of ecosystem
4 Land area is indicated in million hectares (Mha) or billion hectares (Gha). Sometimes km2 are used,
1 km2 = 100 ha.
services. In 2000, a FAO-based assessment of human-induced degradation found

severe degradation from deforestation, agriculture and overgrazing in countries
such as Iceland, Cameroon, Ethiopia, Nigeria and Turkey (20 to 30 percent) and
very severe in countries like Costa Rica (68 percent) (Lambin and Geist 2006).
In 2008, ISRIC published an updated ‘proxy global assessment of land degrada-
tion’ (Bai et al. 2008). In the new assessment, degradation is (re)defined as a long-term
decline in ecosystem function and productivity. The so-called normalized difference
vegetation index (NDVI) is taken as a proxy-indicator of ecosystem function and pro-
ductivity, which is based on satellite observations of net primary production (NPP)
corrected for changes in rain-use efficiency (ratio of NPP to precipitation). Trends in
the NDVI map over time are an indication of land degradation or improvement, but
can unfortunately not tell anything about the kind of change happening. The results
show that 24 percent of the global land area had suffered from a declining rain-
use efficiency adjusted NDVI during the period 1981–2003 (Figure 9.8). Severely
affected areas are found in Africa south of the equator, Southeast Asia, Australia,
the South American pampas and the high-latitude forests in Canada and Siberia.
An estimated 24 percent of the world population, or more than 1.5 billion people,
are affected. If overlaid with land use/cover maps, 19 percent of degradation in this
twenty-two-year time period took place on croplands and 43 percent on forested
lands. More rigorous analysis is needed to understand the relationships with, for
instance, population density and poverty.
The results differ substantially from the earlier GLASOD estimate, because the
latter did not distinguish between historical and current rates of land degradation.5
Historically, degraded lands have mostly become stable landscapes with low pro-
ductivity. In the new estimate of changes between 1981 and 2003, they hardly con-
tribute to the rate of degradation. In other words, the earlier legacy of humanity is
not recorded in the new data. It also explains why notoriously degraded areas in the
Nile Delta and the Iraqi marshes, amongst others, do not show up and why current
degradation is not primarily a dryland issue. The primary productivity-based NDVI
approach has some advantage over the earlier GLASOD approach, which is based
on expert opinion. It is globally consistent, repeatable, quantitative and measures
actual degradation. There are, however, problems of interpretation because other
effects cannot easily be separated from the data. A mix of both approaches, with a
focus on abandoned farmlands, may be the best strategy for the future.
Can we now answer the question whether land degradation is caused by humans?
Undoubtedly, erosion processes can be strongly modified – and usually accelerated –
by human activities such as forest clearing, agriculture (ploughing, irrigation), pas-
toralism (grazing), dam and road construction, surface mining and urbanisation.
Erosion of arable land is a serious issue. In some places, erosion rates are significantly
above the natural ‘undisturbed’ rates, which can cause accelerated river sediment-
ation and hill denudation. The rather scarce field experiments confirm that intense
rainfall storms and large-scale land clearing (bulldozing) are the largest contribut-
ors. According to the aforementioned GLASOD inventory, the partly overlapping
5 The earlier GLASOD assessment was ‘a map based on perceptions, not a measure of land degrad-
ation; its qualitative judgments . . . have proven inconsistent and hardly reproducible, relationships
between land degradation and policy-pertinent criteria were unverified . . . although, to be fair, its
authors were the first to point out the limitations’ (Bai et al. 2008:223).
260
80º 80º
60º 60º
40º 40º
20º 20º
0º 0º
–20º –20º
–40º –40º
–60º –60º
–80º –80º
Slope of linear regression of sum NDVI

Mollweide Projection
Central Meridian: 0.00
.0 – 04
.0 – 3
– 02
ve 01 1
ta 0
d
–0 3 – 0.0
N –0 0.0
te
ge –
–0 4 – –0.
2 0.
–
.0 <
ot .
–0
Figure 9.8. Global negative trend in rain-use efficiency-adjusted normalised difference vegetation index, 1981–2003 (source of map: Bai et al.
2008). (See color plate.)
Box 9.2. USLE, erosion and worldviews. Scientific knowledge of universal

processes is instrumental in understanding the processes that concern us from a
sustainability perspective. It is instructive to do simple erosion modelling exercises
with software packages such as NetLogo 4.1 or PCRaster, in order to get a feel for
process dynamics. Some of the suggested websites at the end of the chapter have
interactive models that use basic equations parameterised for specific situations
(Wainwright and Mulligen (2004)). Erosion models often use the Universal Soil
Loss Equation (USLE), because it is a simple and widely used way to assess
erosion processes.
Ecological historians like Grove and Rackham are critical of the use of the
USLE-equation: ‘The “Universal Soil Loss Equation” . . . is not really an equa-
tion, but an empirical relationship between observed rates of erosion and various
parameters of environment and land use . . . Collection of data began in the 1920s
and led to a number of regional soil loss “equations”, which were combined in
the USLE. The data that generated that Equation were gathered east of the
Rocky Mountains . . . the term “universal” means universal within the United
States . . . Empirical models will work only within the range of conditions over
which the original data were collected . . . The Mediterranean lies outside the
range of climates, soils, crops and cropping practices for which the USLE was
constructed. North American soils were generally not cultivated until the 19th
century, whereas Mediterranean soils have been cultivated and eroded for cen-
turies . . . The USLE fails to predict rates of sheet and rill erosion in the Mediter-
ranean. It is even worse at predicting total erosion, which it was never meant to
do . . . ’. (Grove and Rackham (2001).
Environmental modellers often have a different and more practical view:
‘ . . . erosion prediction is still dominated by the use of one model – the USLE. It
provides the user with a simple model which can be executed using nothing more
sophisticated than a pocket calculator, yet in model comparison studies predicts
annual erosion as well as any model. This may not reflect any superiority in terms
of the model, but simply the fact that a five-parameter empirical model can do as
well as a multi-parameter model in capturing the extremely variable response of
a hillside system’ (Quinton, in Wainwright and Mulligan (2004) 192).
activities of overgrazing (35 percent), agriculture (28 percent) and deforestation

(30 percent) are the major causes. Such results have to be interpreted with cau-
tion, however, because of inconsistencies, limited reliability and reproducibility and
divergent class definitions and interpretations. Besides, there is a difficult to decipher
the legacy of the past. Some experts believe that published data and maps grossly
overestimate the extent of land degradation and that awaiting better observations
in combination with clearer definitions is needed.
Indeed, controversy about the causes and the consequences of desertification
has increased in the last decades (van der Leeuw 1998; Grove and Rackham
2001; Reynolds and Stafford Smith 2002, Bai et al. 2008). Some areas have been
impacted by humans for millennia and humans and their animal herds are as much
part of the resulting social-ecological system as other species. In a long-term and
262 Land and Nature
larger-scale perspective, events are also more ambiguous in their effects. Eroded
hills are the remnants of sedimentation processes that created fertile valleys. Dam-
ming and diverting large rivers such as the Nile, Yellow and Colorado significantly
reduces the sediment flows and thus block the nutrient inflow in the estuaries (MEA
2005). Tectonic uplift and climate change processes are occurring slowly and over
long periods, and human populations may too easily and incorrectly be blamed for
erosion of (top)soil, particularly in fragile areas such as the Mediterranean. Land-
slides and the various forces of water are sometimes neglected or misinterpreted.
The role of trees may be overestimated, the role of other vegetation underestimated.
‘Ruined landscape theory is [thus] deeply pessimistic. If it is true, soils are uncon-
servable except by growing trees. All cultivation – even orchards – is ultimately
self-destructive; farmers can survive only by going out of business . . . ’ (Grove and
Rackham 2001).
So what about sustainability? The degradation processes occur at rather large
time-scales and are partially or wholly irreversible. For instance, organic matter
decomposition happens over five to twenty years, accelerated erosion over ten to
fifty years, and accelerated mineral weathering over centuries (Leemans and Kleidon
2002). This is often outside the time span considered in human resource management
and, therefore, often pernicious – remember the shift in risk spectrum as part of
societal collapse (§3.5). Land overexploitation and degradation should, therefore, be
viewed as part of the long-term evolution of landscapes and sustainable management
of soil vegetation systems as context-dependent. But the basic sustainability principle
remains: the (top)soil removal rate should not exceed the rate at which topsoil
accumulates from natural processes.
9.5 Ecosystem Dynamics: Population Ecology
9.5.1 Population Ecology: Logistic Growth

As you may recall from §2.3, exponential growth can never go on indefinitely.
At some point, growth constraining/stabilising forces start to work. This can be
introduced in a model by making the growth rate a function of the available resource
(Edelstein-Keshet 1988; Case 2000). It can be done in several ways, but, historically,
the first and most simple method stands out. In the 19th century, Verhulst had
observed processes of population growth of organisms under limiting environmental
constraints and proposed to describe these with an equation of the form:
dX/dt = αX (1 − X/K) (a > 0) (9.1a)
with X the population size (§2.4).6 This logistic growth equation can be rewritten in
the form of the population growth rate:
1 dX

X
=α 1− (9.1b)
X dt K
This shows that the growth rate of the population declines with population size.
The shape of the population over time is an S-curve. The reference behaviour
is initially fast growth, then a smooth transition towards the attractor or equilibrium
6 See Appendix 2.1 for mathematical details.
a)
600 50
Variable X Carrying capacity K
Inflow Outflow
500
40
Stock and flow values
400
30
300
20
200
10
100
0 0
0 50 100 150 200 250 300 350
Time
Figure 9.9a. Characteristic logistic growth path of the state variable Population (1), approach-
ing the carrying capacity value (2) of 400 (population) units after about 200 timesteps ((α =
0.03, t = 0.25, Runge-Kutta4). The lower curves indicate the in- and outflux (3,4), with the
influx initially larger than the outflux.
value K (Figures 2.10 and 9.9a). It is easily seen that for X << 1/2K the equation
becomes the one for exponential growth (dX/dt ∼ αX) and that for X >> 1/2K, it
approaches towards an attractor of value K (dX/dt ∼ α(X-K)). Setting dX/dt = 0
shows that there is one equilibrium state at X = K (saturation) and one at X = 0
(extinction). At X = K, the system is in dynamic equilibrium: outflow equals inflow.
K is associated with the carrying capacity because it is the maximum value to which
the population size X (0 < X < K) can grow in the specific environment.
The logistic growth equation or ‘model’ is a combination of a positive and a
negative feedback loop. It is used widely to describe growth processes, which are
confronted with external limits to growth. These limits constrain further growth and
can be food scarcity, density-related diseases, overcrowding and so on. A well-known
empirical case is bacterial growth in a resource-limited environment. In ecology, a
broader interpretation is that it describes the succession dynamics from a simple
pioneer to a mature climax ecosystem. In system dynamics, it is known as the Limits-
to-Growth archetype (§2.4).
In our simulated model world, birth and death are deterministic processes: At a
given population size, they are exactly a particular fraction of the population size. In
the real world, such processes are often discrete events – a mother gives birth to one
or two children but not to 1.5 child, and there is no law regulating its frequency. Only
the assumption of large numbers justifies the approximation. The discrete character
of (birth and death) events can be addressed by formulating the logistic equation in
discrete form.7 It gives some surprising insights (May 1976).
7 See §2.3. Applying simulation software to explore the model behaviour over time is always done
with a discrete timestep. It is important to test results for the sensitivity for timestep and integration
method. Seppelt and Richter (2005) investigated an ecosystem model constructed with different
modelling tools such as STELLA R
and MatLab. They found it hard to generate the same outcome
with different software tools and warn for the differences from uncertainty in integration methods.
264 Land and Nature
b)
700 50
Variable X Carrying capacity K
600 Inflow Outflow
40
Stock and flow values
500
30
400
300
20
200
10
100
0 0
0 50 100 150 200 250 300 350
Time
Figure 9.9b. Logistic growth in an oscillatory dampening mode (α = 0.03, t = 0.25, Runge-
Kutta4, delay period = 20 t).
Animals, and often people too, probably do not anticipate that the environment
has a maximum amount of food it can provide. The forces behind the (exponential)
growth process do not ‘see’ the environmental constraint posed by the carrying capa-
city, because the relevant information is not available and there cannot be an imme-
diate feedback in time and space. Such a delay as consequence of weak/incomplete
signals and absence of anticipation cause ‘overshoot’: The population grows above
the level that can naturally be sustained. The process can be simulated by adjust-
ing the equation in such a way that the outflow (death rate, the negative term in
equation 9.1a) responds to a change in the population size with a delay of one or
more timesteps. The result of a twenty-period delay is shown in Figure 9.9b. Instead
of a smooth approach towards the carrying capacity, the system gets in an oscillatory
‘overshoot-and-collapse’ mode.
In real-world processes, events are not only discrete but also stochastic. Birth
and death rates, as well as all kinds of phenomena in the environment, show up as
statistical variations in experimental data on population sizes. The stochasticity – that
is, randomness or noise – originates in processes that are not considered explicitly in
the model but influence the dynamics, such as fluctuating precipitation, an extremely
hot summer, a sudden disease or invasion, a social revolt, novel inventions and so
on (Turchin 2003). For instance, rain variability is amongst the factors that make
erosion a complex process and disease epidemics can cause erratic death rates in
populations.
A more general growth equation can be expressed as dX/dt = g(X)·X. The
function g(X) is called the intrinsic growth rate. Its logistic growth form g(X) =
α(1–X/K) can be extended to, in first instance:
dX/dt = g(X ) · X = (α0 + α1 X + α2 X 2 ) · X (9.1c)
The growth rate is now a parabolic function of the population size and has for
α 1 > 0 and α 2 < 0 a maximum at some population size Xmaxgr . The birth rate is
now proportional to population size at very low population levels, unlike in the
logistic growth (equation 9.1b). A biological explanation of this so-called Allee
effect, that has empirically been observed, is that mating is more difficult at low
population density.8 Many other growth functions have been proposed for a variety
of ecological and biological processes (Case 2000).
As nice as the notion of carrying capacity may be from a mathematical point
of view, its real-world interpretation is not without problems. Having studied the
millennial history of social-ecological systems in the Mediterranean, Grove and
Rackham (2001) conclude that carrying capacity is not a precise measurement: ‘Is
[carrying capacity] to be based on an average year, or the worst year in ten? Or a
hundred? Are the animals attended and moved around, or can they go where they
will? Even in a stable environment the ecological carrying capacity . . . is typically at
least 50% more than the economic carrying capacity.’ A more dynamic perspective
on carrying capacity, in niche construction theory, is discussed in an evolutionary
context in Chapter 10. The next section discusses an obvious refinement: interacting
populations of more than one species.
9.5.2 Population Models: Prey-Predator Dynamics9

The logistic growth equation describes the dynamics of a single population, such
as bacteria, algae, deer, lynxes – or humans. However, populations never live in
isolation, so what happens if more than one species is introduced in the model? This
has been a research area for more than a century in ecology and biology. It had
already been noticed that animal populations sometimes show cyclical behaviour.
For instance, there are records dating back to the 1840s kept by the Hudson Bay
Company in Canada and their trade in pelts of the snowshoe hare and its predator
the lynx reveals that the relative abundance of the two species undergoes dramatic
cycles. Also, other ‘ecohistories’ show the importance of species interactions, for
instance, the Kaibab and Pribiloff Island narratives.
In ecology, three fundamental types of species interaction are posited: antagon-
istic interactions between two competitors; consumer-resource interactions between
a predator and its prey; and indirect cooperative interactions or mutualism in chains
of species (Roos and Persson 2005). The following discussion focuses on the best
known one: predator-prey. For one predator population of Y and one prey popula-
tion of X individuals, you get the classical formulation given by Lotka and Volterra
in the 1920s as a set of two linear differential equations:
dX/dt = aX − bY X = X (a − bY )
dY/dt = −cY + dXY = Y (dX − c) (9.3)
The equations reflect two common-sense interpretations (Figure 9.10):
r because the prey is food for the predator (X in dXY), it is part of the carrying
capacity of the predator; and
r more predators increase the prey death rate (Y in –bYX).
8 There may also be a maximum growth rate for human populations, when crowding causes the net
population growth to decline due to emigration and wars or a declining desire for children (§8.4).
9 See Appendix 9.1 for some mathematical details.
266 Land and Nature
Figure 9.10. The interaction between a species and its environment and between two species,
represented in a causal loop diagram.
Note that the equations represent exponential growth with a net growth rate depend-
ent on the population of the other species (equation 9.1b). By setting dX/dt = 0 and
dY/dt = 0, one finds the attractors or equilibrium states, meaning the state in which
both populations do not change. One stable state is the trivial (0,0) and the other is
(c/d,a/b). The coefficients a, b, c and d make up the so-called ecological community
matrix.10
A typical pattern over time of a system described by the Lotka-Volterra equa-
tions is shown in Figure 9.11a. When there is prey abundance, the predator popula-
tion increases. The prey population starts declining due to predator abundance, after
which predator population will go down again. The populations will oscillate over
time, but these oscillations are rather ephemeral because the actual path is sensitive
to the initial populations, as is seen in the phase diagram with trajectories for three
different initial predator populations (Figure 9.11b). The system is neutrally stable: a
delicate balance between stability and instability. If prey is harvested, the oscillations
are dampenend or even disappear altogether. Invasion of prey causes the oscillations
to be amplified (Figure 9.11c,d). Invasion of predators also stabilises the system, but
harvesting them tends to destabilise it. In other words, population sizes easily crash
or explode under small perturbations. This is one reason why you should not take
the model too seriously. As with the logistic equation, it is a stylised description of
a phenomenon, but it does convey the key point that species interaction can cause
oscillations in population sizes.
In the course of the years, numerous multi-population models have been pro-
posed and analysed – from simple refinements of the Lotka-Volterra model to ones
10 The model has actually been borrowed from chemical kinetics and provides in turn a metaphor for
more complex processes, for example, in economic and social systems – remember the arms race
model.
a)
200
Predator population
Prey population
Number of individuals
150
100
50
0
0 50 100 150 200 250 300 350
Quarter
b)
200
Initial # predators = 30
150
Prey population
100
50
0
0 20 40 60 80 100
Predator population
Figure 9.11a,b. Simulated populations of prey (1) and predator (2) for the system description
of equation 7.5b (a = 0,004; b = 0,16; d = 0,1; k = 0,3; t = 0.25, Runge-Kutta4). The
populations oscillate permanently around the equilibrium values (3,4).
with many species and classes. One extension is to introduce a finite carrying capacity
for the prey by replacing the term aX in equation 9.3 by aX(1–X/K). This changes
the steady state from a neutral stability to a stable spiral. If prey is harvested, the
system remains in a stable oscillation within the attractor basin for not too large
disturbances. Further extensions are, for instance, to add vegetation as food for
the prey and to introduce competition between species for the same resource. Such
models have been examined in great detail (Case 2000). In some cases, they provide
robust predictions, as in the case of the arctic lemming (Gilg et al. 2003). It is one
of the simplest real-world ecosystems, with one prey (the lemming) and four pred-
ators. Four years of empirical data are nicely reproduced with a straightforward
predator-prey model. Interestingly, no finite carrying capacity for the prey has to be
introduced – stabilisation comes from predator interaction. Nevertheless, the models
are criticised by empirically oriented ecologists because the relationships are often
268 Land and Nature
c)
200
Predator population
Number of individuals Prey population
150
100
50
0
0 50 100 150 200 250 300 350
Quarter
d)
200
150
Prey population
100
50
0
0 20 40 60 80 100
Predator population
Figure 9.11c,d. Simulated populations of prey (1) and predator (2) for the system description
of eqn. 7.5b (a = 0,004; b = 0,16; d = 0,1; k = 0,3; t = 0.25, Runge-Kutta4). The populations
oscillate permanently around the equilibrium values (3,4) but switch to a new trajectory after
a permanent disturbance (harvesting 3 prey, that is, 3 percent) is introduced in quarter 180.
weakly linked to field observations. It turns out that oscillations in real-world ecolo-
gical communities are more exception than rule because of other interactions, such
as positive feedbacks, and because evolutionary processes are omitted in population
models.
In this respect, an interesting modelling exercise has been done by Fay and
Greeff (2006). They built, starting from empirical data on lion, wildebeest and zebra
populations in Kruger Park in the period 1968–1994, a stepwise more comprehensive
three-species population model. In each step, they try to reproduce statistical time-
series with the model. The first one is the simplest predator-prey set of equations
(equation 9.2). Next, mutualism between wildebeest and zebra is introduced; it turns
out that historical data are badly reproduced. It becomes hardly better if a carrying
capacity is introduced for the prey species on the assumption that even with predators
there may be overpopulation. The resulting spiral point attractor gives no good fit.
In a subsequent step, a limiting factor to predation is introduced to reflect that only
Box 9.3. The Kaibab narrative: management on ill-understood systems. Prior

to 1907, the deer herd on the Kaibab plateau, which consists of some 727,000 acres
and is on the north side of the Grand Canyon in Arizona, numbered about 4,000. In
1907, a bounty was placed on cougars, wolves and coyotes – all natural predators of
deer. Within fifteen to twenty years, the larger part of these predators (over 8,000)
was wiped out and a consequent and immediate irruption of the deer population
followed (Figure 9.12). By 1918, the deer population had increased more than
tenfold. The evident overbrowsing of the area brought the first of a series of
warnings by competent investigators, none of which produced a much-needed
quick change in either the bounty policy or the policy dealing with deer removal.
In the absence of predation by its natural predators (such as cougars, wolves and
coyotes) or by man as a hunter, the herd reached 100,000 in 1924; in the absence of
sufficient food, 60 percent of the herd died off in two successive winters. By then,
the girdling of so much of the vegetation through browsing precluded recovery of
the food reserve to such an extent that subsequent die-off and reduced natality
yielded a population about half that which could theoretically have previously
maintained (Roberts et al. (1983); Meadows (1986)).
In the article The Rise and Fall of a Reindeer Herd, Scheffer (1951) tells a
similar story about forty reindeer placed in 1911 on the two Pribiloff Islands,
St. Paul and St. George, in Alaska to provide the native residents with a sustained
source of fresh meat. There were no predators except humans. The herd on
St. George grew quickly to 222 animals, after which it declined to a stable level of
forty to fifty animals. The herd on the larger St. Paul increased to more than 2,000
animals by the late 1930s, after which it collapsed to only eight animals in 1950.
It turned out that the reindeer survived in winter by eating certain shrublike
lichen. When the herd size increased, the reindeer consumed the lichen faster
than their regeneration rate and it needed only one severe winter to exterminate
most animals on St. Paul.
hungry lions kill. This dampens the fluctuations in population sizes but still leaves
some historical trends unexplained. Finally, two more assumptions were put into
the formal model: wildebeest calving in a six-week spring period instead of spread
out over the year, and the ‘cropping of lions’ in the period 1976–1984. This gives
a satisfactory resemblance between the simulated and historical data, considering
the errors in observations. Only after including historical contingencies, such as a
human intervention, did they get it right. This illustrates the necessity to combine
quantitative, mathematical models with qualitative, historical case studies.
An objection against the general population models is the root assumption of
homogeneity in space and behaviour. The isomorphism with the kinetics of chemical
reactions suggests that ecological interactions are basically an encounter of two or
more individuals. However, the interaction mechanisms are usually much more com-
plex and are part of a whole food web with space- and scale-dependent patterns. The
assumption of identical individuals is also unrealistic, in view of the consequences
of different life histories of individuals and the variance in their traits. This is not
unimportant, because if the models of ecosystem dynamics suggest generality but
270 Land and Nature
a)
b)
1200
sheep wolves
1000
Population size
800
600
400
200
0
0 50 100 150 200 250
Time
Figure 9.12a,b. Spatial dynamics of a simple, CA-based wolf-sheep predation model, with
grass ‘off’. It is almost impossible to find a limit cycle path. The lower left graph shows the
population trajectories (NetLogo 4.1).
fail in particular situations, ecosystem preservation and restoration policies may be

inadequate or even catastrophically wrong. Again, there is the tension between the
universal claims of theory and the contingency of practise (§6.3).
A logical next improvement is, therefore, to include spatial aspects and behavi-
oural rules, in combination with disaggregation into age, gender and other classes.
Introducing behavioural variety has been pioneered by evolutionary game theorists
like Maynard Smith (1982) and elaborated by Nowak and colleagues (2006). It is
investigated with cellular automata (CA) algorithms and GIS-based software pack-
ages (§10.4; Engelen et al. 1993) and with agent-based models (§10.5; Janssen 2002).
Some aspects of real-world complexity are introduced in this way and point at the
possibly stabilising role of spatial heterogeneity and behavioural variety.
c)
d)
450
400 sheep wolves grass
350
Population size
300
250
200
150
100
50
0
0 100 200 300 400 500 600 700 800
Time
Figure 9.12c,d. As in Figure 9.12a, but with grass regenerating at a 30 percent/year rate (grass
‘on’). The lower left graph shows the population trajectories (NetLogo 4.1).
The discussion about the stabilising feedbacks in predator-prey systems is nicely

illustrated with a spatially explicit simulation model. The predator (wolf) and prey
(sheep) move in space according to elementary rules such as ‘wolf-eats-sheep if
they happen to move into the same spot’.11 Figure 9.12a,b shows the square area
populated by wolves chasing and eating sheep and the population sizes over time. It
is a spatial representation of the predator-prey model (equation 9.2). The simulation
ends for almost any parameterisation in an unrealistic situation: either both wolves
and sheep extinct or sheep population growing unchecked to infinity. Do you see
the last wolf and the many sheep that still do not know their fate? The reason for
11 The model can be downloaded from the NetLogo website as a demo model under the name Predator-
Prey.
272 Land and Nature
deer pop observations probable carr cap St. Paul St. George
120000 2500
Deer population (Kaibab)
Deer population (Alaska)

2000
1500
60000
1000
500
0 0
1907 1917 1927 1937 1947
Year
Figure 9.13. The deer population on the Kaibab plateau population and the reindeer pop-
ulation on two Alaskan islands. An interactive simulation model is available on the site
forio.com/simulate/e.pruyt/ex4–6-the-kaibab-case-2/model/explore/ to explore the dynamics.
this system behaviour is clear: The sheep are not confronted with resource scarcity.
If grass is introduced into the model and it is allowed to regenerate towards its
carrying capacity (equation 9.1), the system exhibits oscillations in all of its three
state variables (wolves, sheep and grass) (Figure 9.12c,d). It is now a stable oscillation,
in agreement with the previous analysis. Thus, an elegant but simple mathematical
description may be faulty because of its artificial system boundary; in this case, the
lack of food for the sheep.
9.6 Food Webs: The Stocks and Flows in Ecosystems
9.6.1 Food Webs and Their Representations

A first step to deal with ecological complexity is setting up a good classification
system. The Swedish botanist Linnaeus was given the name Father of Taxonomy
for his Systema Naturae (1735) and Species Plantarum (1753), a plant classification
system. It is still widely used in botany. In subsequent centuries, the static classi-
fication was extended with a dynamic representation of trophic systems and food
chains. In ecology, populations per se are understood on the basis of individuals
with characteristics and interactions with individuals of the same and other popula-
tions, as members of an ecological community. Plants and animals, constituting the
elements of the system, are linked to each other through eating and being eaten,
and, therefore, several forms of ecological succession processes can happen. Such
interactions – predator-prey as well as competition, mutualism and parasitism – give
the community a dynamic structure and cause emergent properties which make it
a ‘sum more than its parts’. They form a food chain, in which producer-consumer-
decomposer functions and relationships determine trophic levels. Currently, most
ecologists prefer to speak of food webs. If one also includes the physical environment,
the system is called an ecosystem. This is a core concept in ecology.
Figure 9.14. Estimated flows in mW/m2 through the food web of the North Sea around 1970
(source of data: Ulanowicz 1986:50, after Steele 1974).
For a given classification, the species populations and their interactions can be
sketched in a box-and-arrow scheme or graph.12 The populations in a food web
sustain energy and material flows of carbon (C), nitrogen (N) and phosphorus (P)
amongst others through the system. The basis is formed by biomass from sunlight
via photosynthesis. Every component of the web uses part of the energy flow to
sustain itself. How much energy from the sun can be used depends on the location,
soil, climate and so on. A common measure for biomass via photosynthesis is net
primary production (NPP), usually expressed in ton per hectare (t/ha).
The actual measurement of the mass and energy flows making up a food web
is far from simple. Odum and colleagues (1976) have formulated and applied prin-
ciples of energetics in theoretical/systems ecology for the description of food webs.
Many ecosystems have been conceptualised and empirically represented in terms of
energy and mass stocks and flows – with many results of food web analyses having
been published. Figure 9.14 gives an example of the food web in the North Sea
around 1970. The boxes indicate the species. The linkages represent species interac-
tions expressed as energy flows. The total system throughput, or the sum of all the
12 Equivalently, they can be represented in an input-output (I-O) matrix, with the species classes
arranged in a square matrix and the coefficients indicating linkage strength. The I-O formalism is
widely applied in economics since Leontief proposed it in the 1940s and is used also in ecological
phenomenology (Ulanowicz 1986; §14.3).
274 Land and Nature
system flows, is 2.6 W/m2 . There are several things to note from such a food web
analysis:
r the system is driven by the solar energy inflow, but only a tiny fraction of it
(1.2 W/m2 ) is used (solar radiation is in the order of an average 100 W/m2 );
r the larger part of this energy flow is already dissipated, like that used for meta-
bolism, at the lowest level of the primary producers; and
r human preference for large fish taps into a very small energy flow at the very
top.
A food web representation suggests also a clear difference between growth, which
is an increase in the total system throughput, and development, which is a rise in the
average mutual information of the network flow structure.
9.6.2 Stability and Resilience

Since the concept of food webs gained prominence in the 1970s, researchers have
investigated the relationship between complexity and stability in ecosystem models.
Connectivity and stability, it turns out, are not simply linked (May 1976; Pimm 1979).
More realistic descriptions than a set of coupled differential equations are needed.
Community complexity does apparently buffer against perturbations, when environ-
mental heterogeneity creates subsystems (compartments) that permit higher-level
species to do adaptive foraging and in that way stabilise the system. Introducing
spatial and temporal variation and heterogeneity amongst the individuals in pop-
ulation models, which has become possible with advances in computational speed
and methods, gives new insights in the stability, resilience and persistence of biotic
communities. The application of network theory is also generating new insights into
the structure and evolution of ecosystems (§10.4).
Perhaps you remember a chemistry experiment from high school: You dissolve
a salt in hot water, let the solution cool down, put a small crystal in the now over-
saturated solution – then all of a sudden crystals form. Or you may have read about
landslides in countries such as Brazil or the Philippines where it rains for a long time
and, suddenly, a large mass of earth starts to slide and whole villages are covered
under mud. These phenomena are nonlinear, which means that the effect of a change
is not proportional to the change but, at a certain point, superproportional or cata-
strophic. The system jumps from one equilibrium to another one; for example, from
an oversaturated solution to crystals or from a stable layer of soil to a huge mass
of earth down in the valley. Usually, positive or reinforcing feedback loops are
involved. The word tipping point has become popular to describe such a sudden
change. It’s not evolution, it’s revolution. It has to do with stability, resilience and
multiple equilibria.
What, then, is (eco)system stability? Already in the mid-1990s, there were no
less than 163 definitions of 70 different stability concepts in ecological literature
(Grimm and Wissel 1997). In general,
stability of a system is the condition that it a) stays essentially unchanged (constancy);

b) persists through time (persistence); and c) returns to the reference state (or
dynamic) after a temporary disturbance (reversibility).
The property of constancy is most simple but begs the question: What in the system
is constant? The other two properties refer to the ability of a system to absorb
disturbances in its environment. A system is in a stable state when it develops forces or
movements that restore the original condition upon being perturbed from a condition
of static or dynamic equilibrium (Giampietro 2004). An indicator of stability is the
degree to which a system is amplified upon perturbations or disturbances. A mass at
the end of a spring or a pendulum in a rest position are both typical examples of a
stable system. If you pull the spring or push the pendulum, it will tend to restore its
initial stable state, which is no movement (Figure 7.6). Similarly, a ball in a bowl will
tend to come at rest at the point where the potential and kinetic energy are smallest,
which is on the bottom (§7.4). A system described by the predator-prey system of
equation 9.2 is stable in certain parameter domains, in the sense that a disturbance
will cause it to switch to a new oscillation but not make it explode or die out. As long
as a system can be plausibly described by a set of differential equations, stability can
mathematically be investigated (Appendix 2.1).
Many real-world systems do not behave according to the stylised linear mod-
els that were thought to be ‘reasonable approximations’ of real-world phenomena.
Instead, many ecological as well as economic and social systems exhibit non-linear
behaviour outside a certain domain. For example, if extreme rainfall causes a land-
slide or more intense hunting causes populations to collapse, the system is unstable.
The apparently stable climate system may exhibit amplifying feedbacks, such as the
mechanism that global warming accelerates the release of methane from permafrost
areas, which in turn accelerates temperature rise. If a shortage of a commodity leads
to rapid and large price fluctuations or one month of rioting brings down a gov-
ernment, the underlying system is in an unstable state. Such systems may have an
attractor, which is itself an oscillation – a so-called limit cycle – or have multiple
attractors and sometimes strange attractors.
To account for more complex system properties, the concept of resilience has
been introduced (Holling 1973). Here, again, there are several definitions, reflecting
the author’s background. A widely used definition is:
resilience is the ability to preserve identity under external perturbations or disturb-

ances and to return to the original structure when the perturbation or disturbance is
over.
A measure of the resilience of a system is the full range of perturbations over which
the system can maintain itself. For instance, the maximum wind speed at which a
tree branch can still remain unbroken by bending. Or the human body, viewed as a
system trying to maintain body temperature around 37◦ C, has a rather large domain
of environmental situations in which it can function. This condition has been called
homeostasis.13 The notion of identity makes the definition of resilience somewhat
ambiguous and a well-defined system boundary is important. For example, clothing
and heating equipment increase the domain of circumstances under which the human
body is resilient provided they are included in a newly defined, larger system. Their
provision, however, may lead to less resilience in the even larger system due to new
13 The essence of the Gaia-theory is that System Earth is in homeostatis, such as it has its own internal
feedback, which keep essential state variables within bounds (Schneider and Boston 1991).
276 Land and Nature
dependencies, for instance, on wool and fossil fuels (§3.4). Think of the 200,000-
inhabitant city of Yakutsk in Siberia, where −25◦ C is considered mild and people
die if fuel supplies fail because of rivers being frozen longer than anticipated.
Historically, there has grown a distinction between engineering resilience and
ecological resilience, one characteristic for the technosphere and the other for the
ecosphere. ‘One focuses on maintaining efficiency of function (engineering resili-
ence); the other focuses on maintaining existence of function (ecological resilience).
Those contrasts . . . can become alternative paradigms whose devotees reflect tradi-
tions of a discipline or of an attitude more than of a reality of nature’ (Gunderson
and Holling 2002).
Engineering resilience focuses on efficiency, control, constancy, and predictabil-
ity – all attributes at the core of desires for fail-safe design and optimal performance.14
Such a desire is appropriate for systems with low uncertainty in a largely controlled
environment. It concentrates on stability near an equilibrium state and is measured
as resistance to disturbance and speed of return to equilibrium. For instance, the
capability of a strained entity to recover its original condition (such as size, shape or
structural characteristics) after a deformation caused by stress can be considered as
system resilience (Giampietro 2004).
Ecological resilience focuses on persistence, adaptiveness, variability, and unpre-
dictability – attributes embraced from an evolutionary or developmental perspective.
The emphasis is on the instabilities associated with multiple equilibria and far from
equilibrium states. Indeed, it is better to think of basins of attraction or, broader
even, regimes instead of (point) attractors. Resilience is now measured by the mag-
nitude of disturbance that can be absorbed before the system undergoes a structural
change, which is in a mathematical sense the shift from one attractor basin to another
(‘regime shift’). Consequently, ecologists consider the maximum amplitude of dis-
turbance that still allows the system to return to the attractor basin as an indicator.
This notion of resilience is also penetrating the social sciences, where it is associ-
ated with the capacity of institutions to adjust to perturbations (Scheffer 2009). The
more resilient an institution, the less vulnerable it is to changes or perturbations
from the outside world. This aspect has recently become prominent with regard to
financial institutions and political regimes in an increasingly globalising and volatile
world.
Engineering resilience is an inadequate concept for dynamic, evolving ‘living’
systems with high uncertainty, more than one attractor basin and a ‘noisy’ environ-
ment. Ecosystems are such systems, with a bewildering richness and complexity:
‘Indeed, when considered in detail, many fluctuations in the populations of many
animals and plants show erratic patterns that are likely to be the joint result of
fluctuating external conditions and interacting intrinsic cycles in the ecosystem’
(Scheffer et al. 2001). If drawn to the outer bounds of the attractor basin, there is an
increasing probability to transit to another attractor, depending on the magnitude
of the disturbance and on the underlying resilience of the current state. There is a
loss of ‘identity’. Nonlinear thresholds, random fluctuations in the environment and
14 In a technology context, the German word Fehlerfreundlichkeit expresses this feature to indicate
that machinery should be designed in such a way that (human) errors do not cause irreversible
damage.
complex couplings across space and in time make it hard to predict their evolution.
It is also possible that ‘latent’ energy is built up endogenously in the system, which
is suddenly released.15 Human interventions further complexify the situation.
From a management perspective, the question is also what is desirable – an
issue at the heart of many sustainability debates. Often, humans intend to avoid
dangerous catastrophic shifts or at least moderate their extent and speed in order to
make adaptation more feasible. For instance, if the frequency of extreme floods or
droughts because of climate change were to increase, this may in many ecosystems
become a trigger for further and accelerated change. Risk analysis based on such
models can bring some clarity in possible management options. But from a techno-
economic or social-political point of view, the intention is often to move a system
out of an undesirable basin of attraction – an issue taken up in Chapter 14 under
the heading of ‘lock-in’. In any event, a better understanding is crucial: ‘The central
theme of food web research is the understanding of structure, function, dynamics,
and complexity . . . this task will determine in part the management of our ecosystems
towards sustainability’ (de Ruiter et al. 2005).
9.7 Catastrophic Change in Ecosystems

Real-world ecological communities are networks of highly structured, often dense
interactions in three dimensions and with individual trait variance. Their dynamics
has, traditionally, been understood as ecological succession, a process thought to be
controlled by some form of exploitation. A stage of rapid colonisation of open or
disturbed areas is succeeded by a stage of conservation, during which the emphasis
is on slow accumulation and storage of energy and material. So-called r-species (or
r-strategists) are associated with the first, exploitative phase: They have extensive
dispersal ability and rapid growth in a ‘winner-takes-it-all’ competitive environment.
In the second, conservative phase the so-called K-species (or K-strategists) prevail,
with slower growth rates and more elaborate forms of competition.
A more comprehensive scheme of ecological change processes is proposed by
Holling, in the context of adaptive resource use and management (Holling 1986,
Gunderson and Holling 2002; Figure 9.15).16 The process of transformation from
an initially unstructured state of r-strategists into a structured and stratified climax
community of K-strategists is still the basis. But observations suggest that there is
more: The climax community eventually becomes so complex that its own stability
is undermined and collapse becomes an inevitable consequence, as in the aforemen-
tioned self-organised criticality and societal collapse mechanisms (§3.5). At such
a catastrophic moment, all the energy that is tied up in the niches and interde-
pendencies of the climax community is released. This energy may be used again by
15 Such events – for instance, the sudden outburst of catastrophic forest fires – have a power law size
distribution and are named self-organised criticality. The power law size distribution implies that
there are many events of small size but a few of (very) large size. Natural phenomena such as forest
fires and earthquakes exhibit such distributions.
16 The authors usually represent the changes in the form of a lemniscaat, in which the r-K change is
one towards more potential and more connectedness. I use the more familiar system dynamics rep-
resentation, which suggests a resemblance with the worldview framework and dynamics introduced
in §6.5.
278 Land and Nature
Figure 9.15. A scheme of possible changes in ecosystems, as proposed by Holling and co-
workers.
opportunistic, fast-breeding and highly competitive r-species, who in the process pre-
pare the road for the energy-conserving strategies that characterise the K-species.
There is as yet only limited evidence to validate the scheme. This is not surprising
because in most systems an even more complex reality operates at multiple scales
in space and time. The adaptive ‘ecocycle’ dynamics has been applied as an organ-
ising framework to deal with the complexity of ecosystems and, by analogy, of social
and economic systems (Holling 1986; Gunderson and Holling 2002).17 The scheme
has heuristic value, as an extended analogy of economic and social systems with
divergent cultural perspectives as agents of change.
It has been known for quite some time that ecosystems occasionally experience
sudden, discontinuous change as, for instance, in the collapse of a climax community.
What is behind such ‘catastrophic change’? Throughout the last decades, nonlinear
behaviour has been studied with the help of formal models in order to understand it
and use it for adaptive resource management. Catastrophic change can happen when
the underlying system has more than one equilibrium state or basin of attraction. It is
a sudden shift in system structure and behaviour, a regime shift. A simple nonlinear
equation that represents this phenomenon is a third order differential equation of
the form18 :
dX/dt = −X (X 2 − α) (9.4)
Developed in the 1960s and 1970s by Thom and Zeeman, this was at the basis of
catastrophe theory. The essential point is that a change in the slow parameter α
influences the behaviour of the fast changing state variable X.
17 Holling (1986), well aware of the parallel with Schumpeter’s (1942) theory of economic maturity,
collapse and renewal, calls this transition from climax community to compost ‘creative destruction’.
18 See Appendix 9.2 for some mathematical details.
a)
4
X-value
Attr 01
Attr 02
2 Variance
State variable X
0
0 50 100 150
-2
-4
Time
b)
4
X-value
Attr 01
Attr 02
2
Variance
State variable X
0
0 50 100 150
-2
-4
Time
Figure 9.16a,b. Bifurcation diagram. The dots indicate the state variable X according to equa-
tion 9.4, for a change of the slow variable α from −5 to +5 (t = 0,01, Runge-Kutta4). The
grey curves are the attractor curves. In the upper graph (a), the random fluctuation equals 2.
In the lower graph (b), it is 20. The thin lines are a measure of the variance.
A simple simulation can clarify the model. Suppose that a slow variable, rep-
resented by the value of α in equation 9.3, changes in the course of a period of 100
quarters (25 year) from –5 to +5. During this change, the system will shift from one
attractor to three (one unstable, two stable) attractors (Appendix 9.2). This is indic-
ated in Figure 9.16a with the grey curves. Each point on the curve is a steady-state
and the emergence of the fork at α = 0 is called a bifurcation. The state variable X is
the fast-changing variable and remains, in the absence of environmental fluctuations,
at the attractor X = 0. However, because this attractor is unstable for α > 0, the
smallest perturbation will make it switch to one of the stable attractors (grey curves).
If there are such fluctuations, X will for α < 0, simply fluctuate around the then
stable equilibrium at X = 0. As soon as the bifurcation emerges, at α = 0, it depends
on chance which of the two pathways is chosen (Figure 9.16a). If the fluctuation
280 Land and Nature
Gradual Threshold Hysteresis Irreversible
Stability
n
tio
tio
tio
tio
Landscape
ita
ita
ita
ita
cip
cip
cip
cip
low high low high low high low high
Pre
Pre
Pre
Pre
Vegetation Vegetation Vegetation Vegetation
high high high high
Vegetation
Vegetation
Vegetation
Vegetation
recovery
Equilibria
collapse
collapse
low low low low
Precipitation Precipitation Precipitation Precipitation

Figure 9.17. The differences between gradual ecological change (a) and three different types
of regime shifts (b–d) using precipitation–vegetation interactions as an example (Gordon et al.
2008).
is large relative to the separation between the two branches, the system may even
switch from one branch to the other (Figure 9.16b). It is hard in this situation to
derive firm statements about the state of the system, because in the neighbourhood
of the bifurcation point, the stability landscape is rather flat. One consequence is that
environmental noise can have a relatively large impact on X around this point. This is
seen in the value of the variance during the simulation. Recent investigations suggest
that careful analysis of the patterns of fluctuations over time may be an indicator of
a critical transition (Scheffer et al. 2009). Such ‘before-the-collapse’ signals can serve
as an early warning indicator.
It is instructive, again, to illustrate regime shift graphically with the ball-in-the-
landscape metaphor (Figure A2.1). The ball indicates the position of the system in
the ‘stability landscape’. Let the x-axis represent a state variable X and the y-axis
a response variable Y, then the question is how the two are related (Figure 9.17).
If their relationship is linear, a change in X will cause a proportionate change in
Y. The relation becomes nonlinear when a threshold develops. A next step is the
emergence of a ‘fold’: The system has one unstable and two stable attractors and
gives rise to a phenomenon known in physics as hysteresis. This happens for α >
0. The system can follow either of two trajectories, depending on the initial state
(Figure 7.7). Fluctuations (‘noise’) in the environment may cause the system to flip
from one attractor basin to the next. For large α, the system enters irreversibly the
one or the other attractor basin. As discussed, the width of the attractor basin is a
defining characteristic of resilience.
For three categories of agriculture water interactions, regime shifts can be
envisaged as is shown in the graphs in Figure 9.17 (Gordon et al. 2008). The first
group is agriculture aquatic systems, where changes in runoff quality and quantity
cause regime shifts in downstream aquatic systems. The second one is agriculture
soil systems, in which changes in infiltration and soil moisture result in terrestrial
regime shifts. The third one is the agriculture atmosphere system, in which changes
in evapotranspiration result in regime shifts in terrestrial ecosystems and the climatic
system. The graphs illustrating these situations are for precipitation as state variable
and vegetation as response variable. The system under stress is characterised by two
branches of equilibria (b) and sometimes an area in which these two overlap (c,d) and
sudden change can occur. Note that each curve indicates values of the two interacting
variables (vegetation and precipitation) at which the system is in equilibrium.
There are several real-world examples that suggest the existence of the kind
of dynamics previously described. A first classical example is the Spruce budworm
and forest, which is about the interactive dynamics between the spruce budworm,
its predators and the boreal forest (Holling 1986; Meadows 2008). For centuries,
the spruce budworm has been killing spruce and fir trees periodically in North
America. When the virgin pine forests became depleted and the lumber industry
became interested in the spruce and fir, the budworm became a ‘pest’ and northern
forests were sprayed with the insecticide known as DDT to control it. Later, other
insecticides had to be used and spraying costs kept rising, but success in killing the
budworm was meagre. It was only after Holling and co-workers looked at it as a
system that they started to understand why.
It turns out that the budworms are crucial in maintaining forest diversity. They
are part of a natural stable ecological cycle, during which their favourite food, the
balsam fir tree, increases until with the help of a few hot summers the budworm
population explodes. Even its natural predators are not able to control it, and the
outbreak only stops when the balsam fir population is drastically reduced. The bud-
worm dies off from the combined forces of balancing loops: starvation and predators.
In mathematical terms, there are two equilibrium branches with regular catastrophic
change. Interference by spraying insecticides kept the budworm population under
control, but it also kept the budworm food stock (balsam fir) at a high level and
killed off the natural predators. In consequence, the forest managers set up a situ-
ation of ‘persistent semi-outbreak’, as Holling called it. A similar cycle has been
observed with regard to budworms, tree foliage and predators (birds). Two equilib-
rium branches exist: one with low budworm populations and young, growing trees;
the other with high budworm populations and mature trees. The essential element is
a nonlinear relation in predator (bird) effectiveness: As foliage increases, the bud-
worms become less visible and their population is no longer controlled. A similar
‘near-outbreak’ situation can occur in forests where effective fire protection prevents
natural fires from happening.
A second and well-researched case is Eutrophication of shallow lakes (Scheffer
et al. 2001). It has become another archetype of ecosystems under stress of a disturb-
ance such as a pollutant. The storyline is that an influx of nutrients from inflowing
fertiliser and wastewater and industrial effluent cause a growth of phytoplankton.
This causes bottom plants to disappear as they get less or no light, which makes
the lake look greenish and turbid. Also, small animals living on the bottom veget-
ation die off, several fish species disappear and a monotonous community is what
remains. Bird populations visiting the lake drop by an order of magnitude. These
observations can be formalised in terms of a catastrophic change model (equation
9.4). Turbidity is the slow-changing variable, nutrient concentration the fast variable
and there are two branches of stable equilibria: One with low and one with high
turbidity-nutrient levels that partly overlap (Figure 9.17c). A clear lake with bottom
vegetation will become more turbid as nutrient concentration rises, until a certain
threshold is reached – the critical turbidity. At this point, the bottom vegetation
collapses and the system jumps to the upper branch without vegetation and high
turbidity. ‘Overall, the diversity of animal and plant communities of shallow lakes
282 Land and Nature
Figure 9.18. The figure shows how ecosystems may undergo a predictable sequence of emer-
ging self-organised patchiness as resource input decreases or increases. Thick solid lines
represent mean equilibrium densities of consumers functioning as ‘ecosystem engineers’. The
area between the dotted arrows is the area of two attractors where catastrophic shifts between
self-organised patchy and homogeneous states, and vice versa can occur. The squares indicate
the associated vegetation patterns, dark colours representing high density (Rietkerk et al.
2004).
in the turbid state is strikingly lower than that of lakes in the clear state’ (Scheffer
et al. 2001).
The model indicates an important point for management: To bring the system
back to the original state along the lower branch, one has to go back to much
lower nutrient concentrations than when the switch to the upper branch occurred. A
detailed analysis of the shallow lake area De Wieden in The Netherlands has shown
that understanding such complexity is useful indeed. The official norm of 0.05 mg
total-P/litre, inspired by the aim of nature protection, requires significant reduction
of phosphorus (P) emissions. Much effort and money was spent on reversing the
trend by reducing the nutrient influx, mostly without success. The catastrophic shift
model indicates that the goal of clear water can more effectively and at lower cost
be realised by biomanipulation (removal of the benthivorous fish) (Hein 2005). It
has also consequences for the more optimal use of ecosystems services (Mäler et al.
2003).
A third example of nonlinear regime shifts in response to external and associated
internal ecosystem variables is when there is a positive feedback between consumers
(such as plants) in combination with limiting resources (such as water or nutrients).
This is quite common: Fine-scale interactions lead to spatial resource concentration
and self-organised patchiness because of endogenous causes. In such situations – for
instance, in semi-arid regions – a regime shift can happen if grazing pressure exceeds
a critical threshold, with possibly catastrophic consequences (Rietkerk et al. 2004;
Figure 9.18). This phenomenon has been investigated for heterogeneous landscapes
using a stochastic cellular automata model and field data from three regions in the
Mediterranean (Kefi et al. 2007). It is found that the patch-size distribution of the
vegetation follows a power law and that this can be explained from local feedback
interactions amongst plants. When the model is used to simulate the effects of
increasing grazing pressure, it turns out that the deviations from power laws seen
in the field data also emerge in the model simulations and, importantly, that they
always and only occur close to a transition into a desert. The researchers suggest
that patch-size distributions may be a warning signal for the onset of desertification,
a spatial equivalent of the previously discussed time-signal of a critical transition
(Figure 9.16). Imagine that our complex social-economic-cultural systems also have
such thresholds beyond which sudden catastrophic change might occur – quite a
different metaphor from the one of smooth ongoing growth in material welfare
usually presented.
MDG Goal 7: Ensure environmental sustainability.

TARGETS
1. Integrate the principles of sustainable development into country policies and
programmes and reverse the loss of environmental resources.
2. Reduce biodiversity loss, achieving, by 2010, a significant reduction in the
rate of loss.
‘Global deforestation – mainly the conversion of tropical forests to agricultural
land – is slowing, but continues at a high rate in many countries . . . Though some
success in biodiversity conservation has been achieved, and the situation may
well have been worse without the 2010 target, the loss of biodiversity continues
unrelentingly . . . ’
– The Millennium Development Goals Report 2010, New York
9.8 Biodiversity and Ecosystem Services

Nature has become increasingly an input for the production of marketable goods
and services of the world economy. Sometimes this is a very tangible input, such as
food and timber from the land and fossil fuel and mineral ore. There are also less
tangible forms, such as regulating water flows or satisfying the longing for beauty.
Ideas about and action against the diverse ways in which nature is threatened by
the advance of civilisation have a long history but are nowadays mostly associated
with nature conservation and protection of biodiversity. It is evaluated scientifically
and at a global scale in a number of large-scale assessments. According to the UN
Convention on Biological Diversity (CBD):
biodiversity is . . . the variability among living organisms from all sources including,
inter alia, terrestrial, marine and other aquatic systems and the ecological complexes
of which they are part; this includes diversity within species, between species and of
ecosystems.
284 Land and Nature
As with other definitions, defining something is one thing and operationalising it

is another. Several indicators are used to represent the state of biodiversity and its
change (Alkemade et al. 2009; PBL 2010b):
r the biome extent: the biome area according to climatic and geographical poten-
tial minus agri/urban areas and forest plantations (Figure 9.4);
r the mean species abundance (MSA): the ratio of actual and potential species
composition and abundance, with 100 percent being undisturbed nature by
definition; and
r the status of threatened species and the genetic diversity of domesticated species:
this is a rather diverse set comprising, for instance, the Red List Index with a
summary of threats for species.
Other indicators of biodiversity are coverage of protected or wilderness areas

(Table 11.1) and food web–based trophic indices.
The MSA can be estimated on the basis of biomes and from trends in the driving
forces of biodiversity loss such as land use for crops and grazing, forestry, infrastruc-
ture and others (Dirzo and Raven 2003, Benı́tez-López et al. 2010). Figure 9.19 gives
an impression of the MSA in the regions of the world in the years 1970 and 2000, and
one possible future trajectory for the year 2050. In the outlook for 2050, biodiversity
loss as measured in the MSA is assumed to continue at the rate of loss during the
20th century, a plausible assumption unless explicit and effective policies are imple-
mented (PBL 2010b). Not surprisingly, in view of the still limited understanding of
food web dynamics, there is controversy amongst scientists about the consequences
of further loss in biodiversity. However, there are good reasons to think that bio-
diversity is important for the stability of ecosystems by offering complementarity
through diversity in species functions and insurance through diversity in responses
to external disturbances. The rate of biodiversity loss is already, together with
the nitrogen cycle disturbances and climate change, considered to exceed accept-
able levels from an integrated planetary boundary perspective (Rockström et al.
2009).
The CBD has now 193 signatory nations. The process shows, however, the famil-
iar features of global negotiations: There is general agreement about the urgent need
to protect biodiversity, but there are widely different views on issues such as access
to genetic resources for medical purposes. It is also unclear how initiatives are to be
funded. It is proposed, for instance, to phase out part of the global subsidies to mar-
ine fisheries – estimated at U.S. $15 to 35 billion per year (Nature, 18 October 2010) –
in order to restore and protect nature, similar to the idea of abolishing subsidies to
fossil fuels in order to combat climate change. However, powerful interest groups
will not easily give in to such measures.
As in debates on other sustainability issues, the question of why to care about
biodiversity has been raised. Economists have pointed out that there are trade-offs
to be considered if certain levels of biodiversity are to be sustained. The argument
that many citizens consider it a necessity is countered with the observation that there
is a discrepancy between what people say and what they do. Therefore, what is the
economic benefit of preserving biodiversity?
Mean Species Abundance
1970

2000

2050
Legend
100%
20%
0%
40%
60%
80%
Figure 9.19. Map of terrestrial mean species abundance (MSA) in 1970 and 2000, and a
projection for 2050, based on calculations with the Globio model (Alkemade et al. 2009, PBL
2010b). (See color plate.)
286 Land and Nature
Already in the 1990s, estimates were made of the economic value of the services
nature is providing ‘for free’ (Hueting 1980; Daly and Cobb 1989; Limburg et al.
2002). The Nature issue on Pricing the Planet (15 May 1997) came up with an estimate
surprisingly close to the gross world product (GWP). Two-thirds of it was related to
marine biomes. Since then, more in-depth analyses are made of what has since then
become known as ecosystem services. The notion of ecosystem services was preceded
by the idea of ecosystem functions, in the sense that it is the function that results in
the supply of a service. A general definition is that:
an ecosystem service is the capacity of an ecosystem to provide goods and services
that satisfy human needs.
A telling example of ecosystem services is the honeybee. The bee is by far the most
important animal pollinator and crucial for an estimated one-third of global food
production. The decline in bee colonies is only partly understood and may ‘cost’
dozens of billion of dollars (Ratnieck and Carreck 2010).
The most comprehensive and authoritative assessments of the world’s ecosys-
tems and their services is the Millennium Ecosystem Assessment (MA) published in
2005 after four years of researching, assessing and describing by hundreds of scient-
ists all over the world. It has chosen ecosystem services as the core concept, with the
leading question: How well can ecosystems (continue to) provide the services that
people depend on but so often take for granted? The ecosystem services identified
in the Millennium Ecosystem Assessment are listed in Table 9.1 (MEA 2005). These
services can be grouped in a few classes:
r supporting: nutrient cycling, soil formation, primary production and so on;
r provisioning: food, fresh water, wood and fiber, fuel, medicine, genetic material;
r regulating: climate regulation, carbon sequestration, erosion and sedimentation
control, pollination, biological nitrogen fixation, disease regulation, water puri-
fication, storm and flood protection; and
r cultural: aesthetic, spiritual, educational, recreational.
All four are intricately linked to aspects of human well-being, such as material quality
of life, security, health and social relations. The four categories are widely accep-
ted, although names may differ, such as production instead of provisioning services.
Yet, it may be argued that the supporting services should not be listed separately
because, first, given their extent and complexity, there is no solid basis for their inclu-
sion and, second, their inclusion in valuation may lead to double counting as their
value is already part of the other services (Hein 2005). Ecosystem services in all four
categories are under various forms of stress: ‘ . . . humans have changed ecosystems
faster and more extensively than in any period in human history . . . The changes
made to ecosystems have contributed to substantial gains in human well-being and
economic development, but these gains have been achieved at growing costs. These
costs include the degradation of many ecosystem services . . . This [ . . . ] could get sig-
nificantly worse during the next 50 years . . . Reversing [ . . . ] while meeting increasing
demands for their services is a challenge’ (MEA 2005).
Many ecosystem functions have been an everyday experience for people since
the dawn of humanity, in particular the provisioning and cultural services. But the
scale of human interference and its consequences has instigated two changes:
Table 9.1. Ecosystem services (based on MEA 2005)
Ecosystem service Subcategory
Provisioning services
Food Crops
Livestock
Capture fisheries
Aquaculture
Wild plant and animal food products
Fiber Timber
Cotton, hemp, silk
Wood fuel
Biotic Genetic resources
Biochemicals
Natural medicines
Pharmaceuticals
Fresh water
Regulating services
Atmospheric and climate regulation Composition, temperature, precipitation, etc.
Water- and wind-related regulation Air quality
Water
Erosion
Water purification
Water treatment (removal, breakdown, etc.)
Biotic Disease and pest regulation
Pollination
Refugia
Natural hazard regulation
Cultural services
Values Cultural diversity
Spiritual and religious values
Educational values
Aesthetic values
Inspiration, sense of place
Cultural heritage
Social and economic Recreation and ecotourism
Social relations
Supporting services
Soil formation
Photosynthesis and primary production
Nutrient and water cycling
1. ever more people become aware, in the context of the modern scientific world-
view, how intricate the ‘web of life’ is and how dependent humans are for many
of its functions; and
2. as part of the modern capitalist market economy, the idea has spread that only
by accounting for such functions in economic/monetary terms will induce the
necessary prudence.
As to the first point, many so-called indigenous people still have a huge reservoir
of pre-modern experiences and insights about the functioning of their environment.
288 Land and Nature
Box 9.4. Attitudes towards biodiversity. On request of EU Directorate-General

Environment, Eurobarometer has asked more than 27,000 European Union cit-
izens in spring 2010 how familiar they are with the term biodiversity and with the
concept of biodiversity loss (EC 2010). Two-thirds of these citizens were familiar
with the term biodiversity in the sense of knowing the meaning (38 percent) or
having heard of it (28 percent). Thus, more than one-third (34 percent) have
never heard of the term and, unsurprisingly, respondents with low-level educa-
tion, manual workers and non-working respondents are significantly overrepres-
ented in this group. There are significant differences between countries. As to
biodiversity loss, 43 percent says that it is about certain animals and plants dis-
appearing and 19 percent that certain animals and plants are endangered. There-
fore, it is primarily associated with species, although a minority (13 to 18 percent)
mentions habitat change, declining forest and loss of natural heritage. Most know-
ledgeable and best informed are the citizens of Germany and Austria.
The survey also posed questions about the threats to biodiversity and the
importance of protecting biodiversity and natural areas. More than a quarter of
the interviewees mention air and water pollution as the greatest threat to bio-
diversity. Next come intensive farming, deforestation and overfishing (19 per-
cent), climate change (13 percent) and the creation of more roads, houses or
industrial sites and land use changes (9 percent). The perception of the serious-
ness of biodiversity loss varies widely. Whereas in Finland only 9 percent consider
biodiversity loss a very serious domestic problem, it is about two-thirds in Italy,
Greece, Romania and Portugal. A large majority consider it a very serious global
problem. Over three-quarters of citizens thinks that biodiversity should be pro-
tected for moral reasons but also because it is essential for economic prosperity.
One-third of the respondents opt for stricter regulation, about one-fifth prefer
a focus on better information. The results also indicate that there is relatively
low familiarity with the Natura 2000 European Union-wide network of nature
protection areas.
This has often been replaced and lost. Yet, it can still be a valuable source of
knowledge (Berkes et al. 2003). Regarding the second point, the argument is that
market prices in a private property setting are the most effective signals to secure
sustainable development. By ascribing ‘services’ to ecosystems in monetary terms,
one makes the average citizen aware of these processes and their precious role for
many societal processes – or so it is thought. But such ‘monetisation’ of nature is still
controversial, if only because it is hard to define and value those services adequately.
9.9 Nature and Sustainable Development

Clearly, people have quite different values, feelings and experiences with what we
call nature.19 For poor farmers and herders in the world’s villages, it is a source
19 The site www.6billionothers.org/ gives a fascinating account of the diversity in people’s views of
nature – as well as of other aspects of life.
9.9 Nature and Sustainable Development 289
of food and beauty but also a place of danger filled with wild animals, diseases
and floods. For all of us, the forces of Nature are feared in earthquakes and tsuna-
mis. Some people express fear of nature and its uncontrollable forces; for others
it is a source of spirituality and a way to connect to life. Some people and scient-
ists fear that the relation between man and nature is out of balance; for others
the urban jungle is as livable as the natural jungle. Underneath are the local and
global changes that people experience in their material life: less trees, less water, less
wild nature, more urban environments, more agriculture and livestock, and more
roads. For most urban populations, nature has become a rather far and away exper-
ience. Children in towns and cities have less and less direct contact with animals
other than with pets, in zoos and in animations. As one young urbanite expresses
it: ‘I’m happy in the urban jungle.’ It brings new forms of economic activity: In
2007, the pet food market for dogs and cats amounted to more than U.S. $45
billion and was dominated by only four global corporations. Other people worry
and battle. According to Greenpeace, ‘Without healthy, thriving forests, planet
Earth cannot sustain life. As much as eighty per cent of the world’s forests have
been degraded or destroyed. Greenpeace is campaigning for zero deforestation by
2020 to protect what is left of these extraordinary ecosystems’ (www.greenpeace
.org).
According to cultural theory, one may expect a limited set of perspectives
on nature (§6.3). An investigation of Australian attitudes towards wildlife, for
instance, shows a distinct difference between aboriginals and Anglo-Australians
(Aslin and Bennett 2000). Whereas aboriginal society shared a substantial con-
sensus about what is acceptable behaviour towards animals, the present Anglo-
Australian population has a more fragmented view on and less personal con-
nection to or responsibilities for wildlife. An analysis of various global change
scenarios inspired by cultural theory reveals a difference in emphasis regarding
biodiversity loss: For some it is an environmental problem; for others primarily a
social-cultural problem and for a third group an ethical-cultural issue (Beumer and
Martens 2010).
It is not my intention to make any prediction about the future evolution of
ecosystems and, closely related, social-ecological systems. But it is possible to make
some worldview-based explorations. Table 9.2 illustrates the differences in world-
view with a couple of questions and answers. In the Modernist paradigm, an evolu-
tion of man is perceived towards ever greater control and beneficial use of nature.
In its utopia, the loss of the natural world is successfully compensated by a combin-
ation of strictly managed nature reserves, three-dimensional (3-D) documentaries
on the television or Internet and biotechnological imitations and improvements.
In the eyes of the opposite worldview, those people with a more local and spir-
itually inspired belief system, the Modernist paradigm is an attitude of hybris and
flatness. The deep ecology movement and its teaching of ecosophy is one of its
manifestations. One should look for strategies that sustain all the diverse qualities
of life associated with nature. They certainly require rational decisions and practical
policies, as well as a broad understanding and appreciation of diversity, a moral
conviction about man’s relation to Nature and a deep respect for Life in all its
appearances.
290 Land and Nature
Table 9.2. Worldviews and ecosystem and biodiversity futures
ECOSYSTEMS AND BIODIVERSITY
Statement 1: No single species of living being has more [of this particular] right to live and unfold than any
other species. (Naess/Deep ecology)
In theory, yes. In Incorrect. We have to A valuable insight, that In theory, yes. In
practise, the human eat. Besides, it is God may bring us the practise, it is difficult
species is most who created the respect for other in view of the large
successful, but we do world and put us on sentient beings that populations. We
have to use our power top. many indigenous certainly should
sensibly. people still have. reserve some place
for other species.
Statement 2: Recreational tourism and partnerships with farmers are the most promising options for Nature
conservation.
This is a good example Someone has to pay for It is marketed as Nature Provided it is well
of public-private all those natural conservation, but in regulated on scientific
partnership success. It parks, so these fact it may become grounds, this seems
can help landowners options are an the largest threat to one of the options to
and farmers to interesting source of nature. It is already in get funding.
preserve biodiversity. revenues. Win-win, I many low-income
would say. regions.
Statement 3: Pressures leading to biodiversity loss are, in some cases, intensifying the consequences of this
collective failure, if it is not quickly corrected, will be severe for us all. (GBC 2010)
It is unclear how serious I don’t think the I know this to be true in Correct. Most scientists
the consequences will consequences for me my own environment, agree that it is a loss
be. A genebank and are serious. As long as so I assume that it is of valuable ecosystem
agreements on I can watch beautiful true also globally. services – water,
intellectual property nature movies on Extinction of species medicinal plants and
rights can reduce the television . . . is anyway terrible. so on. Time for strong
risks. international action.
Statement 4: Without healthy, thriving forests, planet Earth cannot sustain life. (Greenpeace 2010)
Can this scientifically be This is an abstract and This resonates with my The importance of the
proven? Nature turns pretentious own feelings and role of forests cannot
out to be quite statement, typical for intuitions about Life. be overestimated. We
robust . . . Maybe, a radical should do whatever
with climate change, environmental NGO. possible to preserve
forests will conquer them.
us.
Statement 5: Trees are my connection to the spiritual world.
Trees are respected in Sorry, I am not sure Trees, the sky, the Right, it is an important
many cultures, but what you mean. I ocean – they all make ecosystem service and
they are also don’t understand me so peaceful inside. we should be grateful
commercially New Age stuff. They are love to me for it, but as a
useful . . . in the proper sense of religion, it is rather
the word. primitive.
9.10 Summary Points

Land and nature are the substrate of human life. Geography and ecology are two core
disciplines in sustainability science and their subjects of investigation are landscapes
and ecosystems. Points to remember from this chapter are:
r Soil and vegetation constitute the life support system. Science has developed
classifications of life zones that are associated with soils and vegetation and with
climate. An essential part is water. Its role is represented in the hydrological
cycle.
r Nature is under the pressures from accelerating changes in land cover, and the
consequences such as soil erosion and overexploitation of water, fish and forest.
It is difficult, however, to assess degradation and to separate the natural and the
human causes.
r Erosion is an illustrative and important universal process of change, and one of
the processes at the root of declining resource quality (soil).
r Population ecology is about the species in ecosystems: how they reproduce, eat,
mate, die. Simple population models, such as the logistic growth and predator-
prey models, represent an elementary understanding of real-world ecosystem
dynamics. They also provide useful metaphors, for instance, for overshoot-and-
collapse and competition processes.
r Food webs provide insights in interactions and connections, experimental as well
as theoretical. They are complex, have multiple attractors and can exhibit cata-
strophic change. Stability and resilience are important properties of ecosystems
and food webs, but there are as yet no general theorems. The precautionary
principle and adaptive management, that works along with the systemic forces
instead of overlooking or fighting them, seem a good guide here.
r Ecosystems provide many less visible, but important services (supporting, pro-
visioning, regulating or cultural). Loss of biodiversity may destroy or degrade
ecosystem services. Their value should be incorporated in development plans
and management strategies.
The best people are like water.

They benefit all things,
And do not compete with them.
They settle in low places,
One with nature, one with Tao.
— Lao Tzu, Tao Te Jing
Furcht, nämlich – das ist des Menschen Erb- und Grundgefühl; aus der Furcht erklärt
sich jegliches, Erbsünde und Erbtugend. Aus der Furcht wuchs auch meine Tugend,
die heist: Wissenschaft.
Die Furcht nämlich vor wildem Getier – die wurde dem Menschen am längsten
angezüchtet, einschliesslich das Tier, das er in sich selber birgt und fürchtet: –
Zarathustra heisst es “das innere Vieh”.
Solche lange alte Furcht, endlich fein geworden, geistlich, geistig – heute, dünkt
mich, heisst sie: Wissenschaft.
— Friedrich Nietzsche, Also sprach Zarathustra
292 Land and Nature
Translation:
Fear, you see – fear is a basic, inherited feeling of man; original sin and original
virtue can be explained from this fear. My virtue grew also out of this fear, that is
called: science. That is to say: the fear of wild animals – it was bred in man over
a long period of time, including the animal that he hides within himself and of
which he is afraid: – Zarathustra calls it “the inner cow”. Such a long and ancient
fear, finally become subtle, sacred, intellectual – now, it seems to me, it is called:
science.
You are the forest

you are all the great trees
in the forest
you are bird and beast
playing in and out
of all the trees
o lord white as jasmine
filling and filled b y all
why don’t you
show me your face?
From: Speaking of Siva (Vs. 75).
Ed. A.K. Ramanujan, Penguin Classics 1973
SUGGESTED READING
A dozen story-research chapters on social-ecological systems around the world.

Berkes, F., and C. Folke, eds. Linking Social and Ecological Systems – Management Practices
and Social Mechanisms for Building Resilience. Cambridge: Cambridge University Press,
1998.
A collection of chapters on the philosophy and background disciplines constituting the envir-
onmental sciences.
Boersema, J., and L. Reijnders, eds. Principles of Environmental Science. New York: Springer
Science and Business Media B.V., 2009.
A thorough introduction into mathematical models in ecology.
Case, T. An Illustrated Guide to Theoretical Ecology. Oxford: Oxford University Press,
2000.
A broad and solid textbook introduction on ecosystems and environmental systems.
Jarvis, P. Ecological Principles and Environmental Issues. Essex, United Kingdom: Prentice
Hall/Pearson Education, 2000.
A comprehensive and beautifully illustrated overview of biodiversity and the possible impacts
of climate change.
Lovejoy, T., and L. Hannah, eds. Climate Change and Biodiversity, 2005, New Haven/
London: Yale University Press.
Freely downloadable assessment of the state of ecosystems and their services as of 2001–2005.
Millennium Ecosystem Assessment. Ecosystems and Human Well-Being – Our Human Planet.
Synthesis and Summary for Decision Makers. Washington: Island Press, 2005.
Thorough introduction on critical transitions in complex dynamical systems, mostly from an
ecological background and with models presented in appendices.
Scheffer, M. Critical Transitions in Nature and Society. Princeton: Princeton University,
2009.
Appendix 9.1 Prey-Predator Models and Stability Analysis 293
An extensive introduction in models of environmental system and the various modelling methods
and techniques.
London: John Wiley & Sons, Ltd., 2004.
USEFUL WEBSITES
r earthtrends.wri.org/index.php# is a site operated by the World Resources Institute with
data and maps on ecosystems and biodiversity.
r www.isric.org/UK/About+Soils/Soil+data/Thematic+data/Soil+Geographic+Data/ is the
site of ISRIC on World Soil Information, with definitions, data and maps.
r soils.usda.gov/use/worldsoils/mapindex/ has more than a dozen world maps on topics
such as soils, climate and biomes.
r www.globalsoilmap.net/ is a site for maps and contains a legend for a worldwide project
on constructing soil maps.
r www.fao.org/ag/agl/agll/wrb/soilres.stm has information and data on natural, land and
water resources.
r modis-land.gsfc.nasa.gov/ is the site of the MODIS team at the NASA/Goddard Space
Flight Center on global change research and resource management.
r www.resalliance.org is the website of the Resilience Alliance, which provides papers and
models about research of socio-ecological systems.
r www.ecologyandsociety.org/ is the site of the Journal Ecology and Society, on integrated
science for resilience and sustainability.
r www.millenniumassessment.org/ is the website of the Millennium Ecosystem Assessment
(MEA 2005)
r www.cbd.int/ is the site of the Convention on Biological Diversity (CBD) and
gbo3.cbd.int/ is the site of the Global Biodiversity Outlook, the flagship publication
of the CBD.
r www.iucnredlist.org/ is the IUCN-site on the Red List index of threatened species.
r wwf.panda.org/is the site of WWF with, amongst other items, the Living Planet index.
Modelling and software

r www.globio.info/ is the site of the Globio model, a tool to assess human activities on
biodiversity and developed and operated by PBL.
r www-binf.bio.uu.nl/rdb/books/mpd.pdf is a coursebook on population modelling in bio-
logy/ecology and the GRIND software, written by Rob de Boer of Utrecht University.
r www.aw-bc.com/ide/idefiles/navigation/main.html is a site that permits exploration of dif-
ferential equations interactively, for instance, logistic growth, predator-prey and spruce-
budworm bifurcation models.
r www.itc.nl/ is the site of Geo-Information Science and Earth Observation of Twente
University in the Netherlands, with courses and research on geographical and resource
modelling.
Appendix 9.1 Prey-Predator Models and Stability Analysis

Appendix 2.1 gives the description of a system of first-order ordinary differential
equations. In vector notation:
−
→ −
→
d X i /dt = A · X (A9.1)
with Xi the state variables and A the Jacobian matrix. In an ecological context, the
matrix A is the community matrix. It describes the interactions between the species
294 Land and Nature
in an ecological community. Setting dXi /dt = 0 for i = 1,2, one finds the attractor
{X1 *,X2 *}. A stability analysis reveals how the system responds to perturbations of
the equilibrium. In a prey-predator system (equation 9.3), for X1 = prey (hare, deer,
daphnia . . . ) and X2 = predator (lynx, wolves, fish . . . ), a12 < 0 and a21 > 0. The
a11 and a22 are the species-specific net growth rates. Stability analysis shows that
under certain conditions the system may exhibit a neutral oscillation, well-known
from mechanical analogues, or an unstable spiral (Figure 7.5). Of course, this model
is an extreme simplification of real-world ecosystems.
An example of a more extended and rather general predator-prey model is
given by:

dN λN N P
=r·N· 1− −κ ·
dt DN + KN N+E

dP λP P hP
=s·P· 1− − (A9.2)
dt DP + KP N
with P the size of the predator population and N of the prey population. The para-
meters s and r are the respective natural growth rates of the predator and prey
populations. The equations look impressive, yet they can be understood rather eas-
ily. First, for a vanishing predator population (P → 0), the first equation describes a
logistic growth process of the prey. Its dynamics then depends on the regeneration of
its food base. The second term in the first equation denotes the additional mortality
of prey because of the presence of predators, with κ/(N + E) the catch effectiveness
of the predator. In other words, a12 = r κ N/(N + E) > 0. Now the role of the para-
meter E is seen: For E → ∞, this term and, therefore, catch effectiveness becomes
zero as the prey has infinite space. Third, if the prey population becomes very large,
the second equation turns into the logistic growth process of the predator. h/N is the
additional mortality of the predator because its food, the prey, has become scarce.
For κ = h = 0 all interactions vanish. Therefore, this extended model incorporates a
number of ‘sensible’ assumptions about a predator-prey system. It can be shown that
such systems can under certain conditions show rather complex but stable oscillatory
behaviour around an attractor, a so-called limit cycle.
Appendix 9.2 Catastrophic Change and Bifurcations

When a parameter change in a system of differential equations causes the stability of
the equilibrium to change, a (local) bifurcation can happen: The simplest equation
to explain bifurcations is of the form:
dX/dt = β − X 2 (A9.3)
√
If one solves for the attractor, such as for dX/dt = 0, there are two roots: X* = ± β.
In the phase plane of dX/dt as function of X, for β > 0, the two attractors are at the
intersections with the x-axis (Figure A9.1). The positive one is locally stable, and the
negative one is locally unstable. For β = 0, the two coincide. For β < 0, the roots are
complex numbers and there are oscillations. If a system has a fast variable X and a
slow variable β, then a gradual increase in β moves the system to either the branch
with the stable or the branch with the then unstable attractor.
Appendix 9.2 Catastrophic Change and Bifurcations 295
10
β=0 α=-5
change in state variable: dX/dt
α=0 α=5
5
-5
-10
-3 0 3
state variable X
Figure A9.1. Phase space diagram for a second order (equation A9.3) and third order (equa-
tion A9.4) differential equation. The dashed curve and the two black dots are the curve and
the attractors for equation A9.3. The grey curves are the curve for equation A9.4 for α-values
–5, 0 and 5. The three grey dots are the attractors for the curve of α = 5.
Similar bifurcation behaviour happens in a system described by equation 9.4:

dX/dt = −X (X 2 − α) (A9.4)
The attractors occur for dX/dt = 0. One root is X = 0, √ and it represents a globally
unstable attractor. The other two roots are for X* = ± α. Unlike in the √ previous
case, for α > 0, the two non-zero attractors are both locally stable (X = ± α). This is
seen in the phase plane for α = 5 (Figure A9.1). For α ≤ 0, there is one globally stable
attractor. A change in the parameter α leads to a structural change in behaviour.
If, again, X is the fast variable, for instance, the size of a population, and α is the
slow variable, for instance, sensitivity to climate change, the time-path of the state
variable X will for slowly rising α gradually move from one with a globally stable
attractor to one which has two locally stable attractors and, therefore, can undergo
a rather sudden switch from one equilibrium state to another.
10 Human Populations and Human Behaviour
10.1 Introduction: The Image of Man

The previous chapter looked at ‘nature’ through the lens of geography and ecology.
But almost everywhere, ‘nature’ has been influenced by ‘culture’. The word culture
is rooted in the Latin word colere, which means to tend or till the land but is also
related to cultus and cultivate in the sense of care, labour and worship. The relation
between nature and culture is an ancient philosophical question. How deeply are
you rooted in ‘nature’ and how much and what kind of freedom can be realised in
‘culture’? The narratives about past societies give provisional answers. This chapter
focuses more explicitly on the species Homo sapiens in nature.
One can see nature as fundamentally subservient to human interests, as it is
phrased in some biblical texts and entrenched in modernity, or as a cosmos in
which humans participate as it is experienced in most pre-modern religions and
modern forms of ecospirituality (Figure 8.5). Clearly, there is more than one answer
to the question of how we see ourselves in nature. Are people not much more
than instinct-driven animals? Are human beings the pinnacle of evolution, on the
verge of a revolutionary jump in cosmic consciousness? Are we, basically, rational
survival-oriented individuals with an ever larger array of tools to control our destiny?
Or are we all of this? There is a huge perennial philosophy at our disposal as
source of reflection and insight. Philosophers like Wilber (2001) build an integrated
psychological-spiritual view in the footsteps of ancient traditions, incorporating novel
scientific insights. Social scientists have developed grand theories from which some
major characteristics of men as social beings emerge (§5.2; Table 6.3). Biologists
and economists have their own particular image of man. The latter prefer to see
men as rational beings; the former emphasise the biological roots in an evolutionary
framework.
An important development in this respect is the advance in understanding com-
plex systems (§8.5). We have associated complexity with an increase in behavioural
diversity of and interactions between system elements and with an increasing ‘interi-
ority’. Which survival strategies can viruses and bacteria develop is a crucial question
in understanding the spread of diseases. How do sheep behave when grazing – one
aspect of overgrazing. What are the rules and strategies of humans in converting
296
10.1 Introduction: The Image of Man 297
Figure 10.1. (Human) agent representations within the horizontal dimension of internal-
external and the vertical dimension of collective-individual. Complexity science approaches
are implicitly or explicitly using such representations (based on Ferber 2007).
land, catching fish, producing ores and fuels? What is the role of interaction between
people in choosing or not choosing organic food or favouring the car over public
transport? And, the most difficult one, how do people self-organise in multiscale
governance arrangements and institutions?
Many of the issues raised in sustainability science are so difficult to tackle because
of the role of humans. Increasingly, such questions are addressed with the novel
methods of complex system science (CSS). There are two reasons to consider these
methods and their applications in sustainability science. First, I introduce quality of
life as a largely subjective yet integral part of the search for sustainable development
(§6.2–4). Therefore, I must reflect on the psychological and social dimensions of
individuals and societies: values, beliefs and perceptions, expectations and attitudes,
mental constructs and contemplative practises.1 Such a reflection points at meaning,
at transcendence. Second, the methods and techniques of CSS allow experiments in
silico, to test hypotheses about people’s mental and inner functionings and about
their interactions with others in complement to laboratory and field experiments.
This chapter focuses on the 4-Quadrant framework originating in the field of arti-
ficial intelligence (Figure 10.1). Two dimensions are distinguished: external versus
1 The magical rituals and mythical accounts of the members of civilisations – their Great Stories –
can also play an important role in understanding the deeper, cultural drives, even in apparently
rationalist, modern societies.
internal and individual versus collective (see also Table 6.5). The horizontal dimen-
sion is about whether a person is primarily seen and modelled as an individual in
(relative) isolation, or as part of a larger group with mutual interactions (tribe, net-
work or society). The vertical dimension represents the degree to which the focus
of description is on external observable behaviour or on internal mental processes
(memory, imagination or anticipation). Of course, these are not black-and-white
distinctions.2
Mainstream science, operating within the paradigm of empirical reductionism,
has the tendency to focus on the E-I domain (lower right corner) where laboratory
experiments with animals or people give ‘objective’ scientific results and ‘strong’
knowledge. It is dominant in biology and medical science, and it is an interpretation
and aspiration for most scientists in the economic and social sciences. But how do
emotions, values, intentions determine and change individual behaviour? For that,
one has to consider the interior state of a person – the I-I domain (upper right corner).
This is considered the domain of psychology as well as of the esoteric and spiritual.
And how do the connections between people come into the picture? Individual
animals and human beings are never isolated and this recognition brings people into
the domain of the other social sciences – the E-C domain (lower left corner). Because
human beings are both (self-)conscious individuals and members of collectives, such
structures become internalised in the form of collectively shared social codes and
ideas – the I-C domain (upper left corner). It is here that the intangible and complex
elements of cultures, religions and ethics reside. In the university, philosophy and
the humanities are closest to this quadrant. As to methods, network and agent-
based models study emergent phenomena and evolutionary dynamics (E-C corner),
while investigations of cognitive science and artificial life complement the ‘natural’
introspective techniques (I-I corner).
This brief philosophical and methodological detour provides also a bridge
between the natural and the social sciences. It structures this chapter. The first
sections describe some important issues in demography and, in particular, popu-
lation models. Next, the chapter briefly explores the evolutionary perspective in
biology and discusses the image of man as seen in mainstream economic science and
recent extensions. I briefly introduce the methods, techniques and tools of CSS and
illustrate with a few applications how it can help to better frame and examine core
issues in sustainability science, among them the discourse on the relation between
science and religion and on the role of (self-)awareness. Thus, sustainability science
can become part of a personal journey through the diverse manifestations of human
life, towards a [r]evolution of consciousness. This is what an integral approach aims
for, ultimately.
World Population
6,890,729,773 on February 4, 2011 8:00 a.m.
world-population.com/
2 The horizontal dimension is similar to the group dimension in cultural theory and the individual-
collective axis in the worldview framework; the vertical dimension reflects partly the material-
immaterial axis in the worldview framework (§6.3–4).
4%
USA (4,5%) Brazil (2,8%)
W. Africa (5,8%) W. Europe (6,1%)
3% Russia (2,5%) Middle East (3%)
Pop-growth 5-yr avg %/yr
India (16,8%) China (20,4%)

Japan (2%) World (100%)
2%
1%
0%
-1%
0 20000 40000 60000
Income in 2005 $ ppp/yr/cap

Figure 10.2. Relationship between population growth and income for the world (black dots)
and nine world regions in the period 1970–2005, and their model-based extrapolation onto
2050 (source of data: PBL/World Bank).
10.2 Demography: Human Population Dynamics
10.2.1 Modelling Population3

The human population in the world has grown from less than one million at the start
of the Holocene to somewhere around 50 million individuals 4,000 years ago and to
more than 6.5 million in 2005 (Figure 5.1). The rate of growth has been declining
since the 1980s, with a consistent fall in the population growth rate in nearly all
countries. Income, measured as gross domestic product (GDP) per capita, is a major
driving force through a variety of mechanisms (§4.2). At the regional level, the net
population change rate declines exponentially with income – one of the metamodels
or ‘stylised facts’ presented in Chapter 84 (Figure 10.2). The former USSR republics,
including Russia, and Central Europe experienced negative population growth rates
for the period after the fall of the Iron Curtain in 1989–1990. The growth rate of
the United States’ population declined, but at a significantly higher level than that
of the population in Europe and Japan throughout the period. Probably, the flow
of interregional migration is a factor here. There are significant exceptions, each
with their explanations on closer inspection. For instance, South Africa experienced
a period of large socio-political change and a rise in the death rate because of
HIV/AIDS. Saudi Arabia’s combination of high population and high income growth
reflect its traditional culture and its oil wealth and immigrant inflow.
Understanding population dynamics is important because population size is an
important determinant of the prospects for quality of life and sustainability in a direct
3 I want to acknowledge the contribution of Henk Hilderink, at PBL, to this section.

4 Figure 10.2 is based on the world data of income as GDP/cap/yr (Y) and of population growth
(PopGR) (Bakkes et al. 2008, OECD 2008). For most regions, there is a good fit with a power curve
PopGR = aY−b (Figure 10.3). Also, an exponential decline curve gives a good fit. For low-income
regions, there is a widespread, but the power curve remains a good fit. The high-income regions
follow a logarithmic decline of the form PopGR = a-b LN(Y).
way, such as per capita resource use, and via indirect ways, such as health costs and
immigration pressure. Some people even insist that the present size of the human
population, having reached 7 billion in 2011, prevents any form of sustainability –
but this is impossible to prove without an idea about their quality of life. The
organisation and rising costs of health services – in rich countries as part of aging
and advancing medical technology and in poor countries as part of the intense desire
for adequate health care – are one of the population-economy links that determine
future quality of life for many. Another important link are the provision and costs
of education. I refer the reader to the Suggest Readings for more specific discussion
on these important topics.
Throughout history, people have tried to forecast the size of the population. At
the end of the 17th century, one estimate even went up to the year 20,000. The early
long-range projections were based on mathematical extrapolation of total numbers
or growth rates. In system’s terminology, the growth of a population (stock) is the
result of the difference between the total number of births (inflow), deaths (outflow)
and the net migration flow. The crude birth rate (CBR) is defined as the number of
births per 1,000 persons in a given year. Generally, it declines as a result of a decline in
the fertility rate, which is defined as the expected number of children born per woman
during her lifetime. The crude death rate (CDR) is similarly defined as the number
of deaths per 1,000 persons in a year. The death or mortality rates can be specified
for different ages of the deceased individuals. The resulting graph is an asymmetric
U-shaped function: Mortality is high for the children up to five years, relatively
low between five and twenty-five years and then steadily increasing for higher ages.
There is a non-negligible distinction between the two sexes, for example, women
have lower mortality rates, especially at higher ages. A more elaborate description
considers population age structure and income trends and inequalities (Lutz 2001,
2003).
A country’s population also changes because of international migration, while
within a country the population density changes due to rural-urban migration. Usu-
ally, migration trends are simply extrapolated from historical data. Net migration
is the outcome of two processes with different drivers: emigration and immigra-
tion. Determinants of migration are economic opportunities, conflict situations and
extreme environmental events. They operate at local, regional and global levels.
Long-term environmental change such as more intense droughts or sea level rise,
often in combination with overpopulation and conflicts, are becoming a more fre-
quent cause of migration. The number of environmental refugees is estimated in the
tens of millions and increasing.
The profound historical changes in fertility and mortality patterns is generally
referred to as demographic transition (§4.2; Figure 10.3). It is the transition from an
equilibrium situation with high birth and death rates (stage I) to another equilibrium
situation with low birth and death rates (stage III). The period in between (stage II)
is one where mortality is falling, sometimes rapidly, but the birth rate is still high
because people respond to the declining mortality with a delay that causes a tem-
porary difference between death and birth rates and hence fast population growth.
The transition process is visible in the age distribution of the population. Figure 10.4
shows age and gender classes for the population in four countries calculated for the
year 2010. One can clearly see the different structure of the population pyramid.
Stage I Stage II Stage III

>6 children/woman <2,2 children/woman
LE < 45 yr LE > 70 yr
Births/deaths per 1000
CBR
CDR
1750 1850 1950

Figure 10.3. The three stages of the demographic transition: changes in crude birth rate (CBR)
and crude death rate (CDR) and in life expectancy (LE). The historical data on crude birth
and crude death rate are for Sweden (source of data: en.wikipedia.org/wiki/File:Demographic
change in Sweden 1735–2000.gif under Creative Commons).
West Africa had the highest growth rates and now, as a consequence, has the largest
fraction of young people. Because the young people will also get children, such
broad-base pyramid structure represents a large population momentum – the notion
of inertia discussed with respect to stocks (§2.3). Western Europe and, to a lesser
extent, India are past stage II and the age cohorts of the high-growth periods are
past childhood. Japan is a clear example of an aging population.
The demographic transition is not an isolated process but part of a broader trans-
formation observed in a growing number of countries since the 18th century that is
associated with ‘modernisation’. It leads to considerable variety in fertility decline
between and within populations, as they are in different stages of the transition.
Yet, the pattern of the transition is remarkably consistent all over the world and
historical data indicate that fertility and mortality rates change only gradually. The
transition can be disrupted in some of the poorest countries or regions of the world,
if a combination of natural and societal forces cause collapse and people do not have
the money or connections to import food and/or are unable or unwilling to move
elsewhere.5 In essence, population models are not different from the models in pop-
ulation ecology (§9.5) – with one important exception: They do not consider a (fixed)
carrying capacity. The reason is that human populations have been quite successful
so far to overcome external constraints through outmigration and ingenuity-driven
intensification.
5 There are also countries where rigorous population policies have had a significant influence on
population structure and dynamics, for instance, Rumania and China. Also, the consequences of
large wars are seen, for instance, in the population pyramid of Japan (Figure 10.4).
302
Age India Age West Africa

100 100
Male Male
Female Female
80 80
60 60
40 40
20 20
0 0
15 10 5 0 5 10 15 15 10 5 0 5 10 15
Population (mln persons) Population (mln persons)
Age age Japan

Western Europe
100 100
Male Male
Female Female
80 80
60 60
40 40
20 20
0 0
5 4 3 2 1 0 1 2 3 4 5 3 2 1 0 1 2 3
Population (mln persons) Population (mln persons)
Figure 10.4. Age structure of the population for four countries in different stages of the demographic transition, based on simulation
with Phoenix model (Hilderink et al. 2009). The horizontal bars indicate the number of people (male/female) in a particular age group
or cohort, in millions.
10.2.2 Driving Forces

Using historical data and trends for CBR and CDR, one can construct a simple stock-
flow model and make population projections. But demographers try to understand
in more detail the determinants of birth and death rate and more refined concepts
and age-specific data have been developed (Lutz 2001, Hilderink et al. 2009). A first
step is to distinguish separate classes (cohorts) in the population, notably age group
and gender (Figures 2.10b and 10.4). The age structure has implications for future
population growth via the number of fertile women in the population. A second
step is to investigate the biological, socio-economic and behavioural determinants
of fertility and morbidity/mortality rates in each age/gender group. Figure 10.5 is
a diagram that portrays the most important determinants and connections. There
is a clear correlation between income and fertility/mortality rates, directly through
socio-economic and cultural factors and indirectly via state provision of education
and health services – another metamodel or ‘stylised fact’.
Human fertility is a biological process in which social, economic, cultural and
environmental variables play a role. The number of births mainly depends on the
potential number of children women can have (total fecundity), the rate of marriage,
and the degree to which contraception and induced abortion are applied. Fecundity
is related to nutrition, amongst other factors. Model analysis usually focuses on the
average total fertility rate (TFR), which is the number of children a woman would,
on average, have given birth to at the end of her reproductive life span if current
age-specific fertility rates would prevail.
Proximate drivers are people’s desire to have less children as incomes rise
because child mortality drops and old-age provisions improve, and the increase in
opportunities for women to use contraceptives and to make a career – and a decision –
of their own. Other determinants are largely socio-cultural and vary amongst soci-
eties and over time, sometimes even as part of explicit population policy. Some
of these relationships can be estimated from correlations with a woman’s social-
economic status. Women’s levels of educational attainment turn out to have an
especially strong effect on fertility in particular, more even than men’s education
levels or other characteristics of households such as wealth. Education levels are,
not surprisingly, correlated with income.
In many countries, there is a clear preference for male children (son preference).
This gives an upward pressure on fertility levels because many couples continue to
have children until a son is born, resulting in a higher TFR than without gender
preference. At the global level, this effect is rather small, but it may be significant at
the country level. Data from six countries with strong son preference – Bangladesh,
Egypt, India, Nepal, Pakistan and Turkey – show that the number of women preg-
nant at the time of the survey would be lower by 9 percent to 21 percent in the
absence of son preference. The effect becomes more evident with a rise in contra-
ceptive prevalence and a decline in fertility level (Nag 1991). In Sub-Saharan Africa,
Southeast Asia, Latin America and the Caribbean, fertility levels are relatively unaf-
fected by couples’ desire to have a child of a specific sex (INFO Project 2003). With
increasing education and continuing modernisation, son preference seems to fade
away gradually.
304
Figure 10.5 Causal loop diagram (CLD) with determinants of population. The important determinants of the flows (birth and death rate) are indicated.
a) 1000
100
Deaths per 1000
10
1
India, 1951 (41.9)
India, 1980 (52.8)
0.1 Canada, 1951 (66.3)
Canada, 1990 (81.1)
Base Mortality (86.0)
0.01
Age cohort
Figure 10.6a. Base mortality rates for the five-year age cohorts for Canada and India in the
years 1951, 1980 and 1990. The lowest curve is the base mortality rate. The corresponding life
expectancy is indicated between brackets (Source of data: Hilderink et al. 2009).
The causes behind human mortality are more varied than for fertility. The com-
mon approach is to start with an age-specific base mortality rate, such as the lowest
possible. An estimation for different age classes is shown in Figure 10.6a. Mortality
from explicitly listed health risks (diseases, accidents, war and so on) are estimated
from exposure to these health risks and added to the base mortality. Exposure is
estimated from socio-economic and environmental variables such as malnutrition
and deficient water supply and sanitation. These tend to correlate with income, edu-
cation, climate and other indirect variables. An additional factor to be considered
is the level and effectiveness of health services, which tend to be correlated with
health expenditures and, again, with income. To calculate the total mortality, an
excess mortality is added from unexplained and non-health risk–related ones. The
life expectancy (LE) can be calculated from the age-specific mortality rates: It is the
number of years a certain age group may expect to live given the present mortality
experience of a population (Table 4.2).
From these correlations, one can postulate causative factors that allow an estim-
ation of mortality rates as a consequence of health risks and as a function of age
and gender. Simulating the changes in the determinants of mortality, particularly
income, it is possible to reconstruct the death rate and its causes over time. There
are still large differences amongst the world regions (Figure 10.6b-e). The trends
in the causes of death in India changed significantly since 1950, one of the clear
marks of modernisation. In Western Europe, the major causes are the ‘welfare’-
related obesity, smoking and blood pressure, and the changes are much smaller. For
both regions, an extrapolation is shown, based on an officially expected income
growth path. In India, one can expect a significant increase in welfare-related
diseases.
From the probabilities of certain risk-disease combinations, you can calculate
the number of years lived with a disease and of years lost by premature death.
b) India 2010
0.12
Smoking Obesity
0.1 Blood Pressure HIV/AIDS
Socio-economic Status Malaria
0.08
Outdoor Air Pollution Indoor Air Pollution
Fraction
0.06 Water Supply and Sanitation Undernutrition
0.04
0.02
Age cohort
c) WEU 2010
0.07
Smoking Obesity
0.06
Blood Pressure HIV/AIDS
0.05 Socio-economic Status Malaria
Fraction
0.04 Outdoor Air Pollution Indoor Air Pollution
Water Supply and Sanitation Undernutrition

0.03
0.02
0.01
Age cohort
Figure 10.6b,c. Attributable mortality by exposure for ten health risks in Western Europe
and India in 2010, based on simulation with the Phoenix model (Hilderink et al. 2009).
The sum of these two is the disability adjusted life years (DALY). The DALY has
become the measure of the burden of disease in a population. The main risk factors
from which the DALYs are estimated are poverty- and environment-related risks
in poor countries and welfare-related risks in rich countries (Figure 10.7). Poverty
(childhood and maternal undernutrition) and environment (notably lack of adequate
sanitation and safe water) dominate risks in poor countries. For instance, in Africa,
childhood and maternal undernutrition account for more than 40 percent of all
attributable DALYs. In rich countries, welfare-related risks prevail. In Europe,
addictive substances (such as tobacco and alcohol) and diet-related and physical
inactivity factors (such as blood pressure and obesity) are responsible for the larger
part of health losses. These estimates point at the areas where policy interventions
are most effective in terms of gains in healthy life years.
d) India 2010
Undernutrition Water Supply and Sanitation
Indoor Air Pollution Outdoor Air Pollution
Malaria Socio-economic Status
HIV/AIDS Blood Pressure
Smoking Obesity
9
7
Deaths per 1000
Age cohort
e) WEU 2010
Undernutrition Water Supply and Sanitation
Indoor Air Pollution Outdoor Air Pollution
Malaria Socio-economic Status
HIV/AIDS Blood Pressure
Smoking Obesity
9
7
Deaths per 1000
Age cohort
Figure 10.6d,e. The epidemiological or health transition: attributable mortality by exposure
(per 1000 persons) in Western Europe and India in the period 1970–2050, based on simulation
with the Phoenix model (Hilderink et al. 2009).
Childhood & maternal undernutrition Sexual and reproductive risks

Environmental risks Other diet-related & physical inactivity
Addictive substances Occupational risks
Other risks
100%
80%
Risk factor attribution
60%
40%
20%
0%
Region
Figure 10.7. Risk factors for attributable DALYs for six regions and the world. The percent-
ages indicate which part of the total disease burden (DALY) can be attributed to each of the
seven risk factors (Source: WHO 2002).
The gradual shift in risks with rising incomes is called the epidemiological or
health transition (Martens 2002). Currently, one of the clear manifestations of this
phenomenon is the rapid rise of obesity and, as a consequence, diabetes and heart
diseases amongst the middle classes in countries such as India and China. The main
reason is the change in lifestyle, from traditional to fast food. Environmental risks
are present in all regions but are far more important in poor countries and cause
about 15 percent of the DALY loss worldwide. The effect of sexual and reproductive
risks, mainly HIV/AIDS, is clearly visible in Africa but forms a growing threat in
other regions such as Asia.
10.3 Evolution: Our Biological Roots

Human development has for a long time been primarily a co-evolution with plants,
trees and animals. What insights on sustainable development are received from a
biological perspective? The central concept is evolution, primarily in a Darwinian
sense. The two essential principles of evolution theory are (Mitchell 2009):
r natural selection on randomly generated, small variations in individuals is the
major mechanism of evolutionary change and adaptation, and
r macroscopic phenomena are understood by microscopic processes of gene vari-
ation and natural selection.
Box 10.1. Population regulation amongst the !Kung San. The !Kung San people
live in small bands in the hostile environment of the Kalahari desert. Studies
by anthropologists give insights in the interaction between humans and their
environment. A simple model has been constructed by Read (1998). It assumes the
resource base is fixed regardless of the intensity of foraging and hunting, and that
population growth is determined by the ‘textbook’ logistic growth equation with
a population density-dependent fertility rate. A behavioural element is added to
the model: !Kung San women want as many children as possible, but they are
concerned for the well-being of their family and balance the time and energy
spent on children and on family support–related activities. How much time they
can spend on the family depends on the resource characteristics, and so does
their desired family size. The fertility rate depends on the resource via the energy
expenditure of a !Kung San woman. If the energy expenditure exceeds a threshold
value, the fertility rate drops to zero. This causes a crash-and-boom periodicity
in the population size for a low energy expenditure per child – a pattern that may
have characterised the Netsilik Eskimo with periodic starvation from unexpected
and substantial changes in their resource base. If the energy expenditure per
child is high and women therefore have to take into account the time and energy
requirements for caring for offspring, there is still boom-and-bust behaviour, but
of a much lower amplitude.
Ethnographers have found that !Kung San women did not want additional chil-
dren unless they could care for them properly. Therefore, birth spacing decreased
with lower resource procurement efforts. This means that a woman’s decision on
how to allocate her energy between the well-being of the family, such as searching
for food, and the desire for more children is further tilted towards the former. In a
subsequent extension of the model, the threshold value for the energy expendit-
ure is, therefore, assumed to increase with population size in order to reflect
resource scarcity. This gives an even more adequate tracking of historical pop-
ulation size and resource adequacy. Instead of overshoot-and-collapse, there is
stabilisation. What is the lesson? First, it matters whether females place high
priority on the well-being of the family or instead focus on their desire for more
children. Resource abundance is usually not a control option, but birth spacing
is! Second, culturally mediated individual behaviour introduces anthropological
insights with possible population projections that are very different from what
(over)simplistic models tell us.
The evolutionary perspective on human nature emphasises the ability to change and
adapt, to adopt new behaviours and to produce social norms and culture. Human
nature is seen as adaptive and innovative, more than as a limitation or something to be
subdued. This view, with its idea of progress and its competitive search for efficiency,
is an important part of Modernism. ‘The theory of evolution by cumulative natural
selection is the only theory we know of that is in principle capable of explaining the
existence of organized complexity’ (Mitchell 2009).
The evolutionary perspective can offer unexpected and new vistas. The advances
in the life sciences (biology, cognitive and neuroscience) shed new light on the
question how malleable and diverse ‘human nature’ is. Some biologists claim that
genetics are the key to understanding individual and social behaviour. This view is
part of a long-standing thread in the scientific discourse, with Dawkins’ books, The
Selfish Gene (1976) and The God Delusion (2006), amongst the recent expressions.
Its rigorous application to social organisation of not only ants and bees but also
of humans has led to the controversial branch of sociobiology as described first by
Wilson in his book Sociobiology – The New Synthesis (1975). It emphasises the role
of natural selection in social behaviours such as mating and hunting. It borders on
biological determinism and has been used ideologically to justify social inequity and
racism. This variant of social Darwinism is unacceptable to those who cherish a
less materialist-empiricist worldview. Similarly, some neuroscientists are also mak-
ing claims such as ‘we are our brain’, on the basis of brainscans and experiments.
Strong statements about physico-chemical processes and behaviour under simplified
controlled conditions are broadened to weak statements about the free will, respons-
ibility, the soul and other complex notions, with sometimes strong ideological impact.
It reflects the tendencies of the (post)modern worldview.
The arguments that evolutionary biology provides the central organising prin-
ciples for understanding the behaviour of humans as well as other animals concen-
trates on a few observations and interpretations. First, human behaviour with respect
to offspring and consumption has biological roots, as is seen from the similarity with
animal populations. Why do people want offspring and under which conditions do
they reduce the birth rate? Reproductive success is in many, and not only tradi-
tional, societies positively associated with great wealth and high position. Biological
research suggests biological roots of this phenomeon: ‘Costly exaggerated displays
enable high quality males to honestly advertise their quality to potential mates and
rivals because only high quality individuals can bear the costs of the display’ (Penn
2003). People acquire display items to demonstrate group membership and they
need items worn by high-status individuals to rise in social status. The economist
Veblen (1899) introduced the notion of ‘conspicuous consumption’ and suggested
why people pursue extravagant goods such as fashionable clothes, luxurious cars,
and massive homes: ‘Consumption of . . . excellent goods is evidence of wealth . . .
the failure to consume in due quantity and quality is a mark of inferiority and
demerit.’
Biology also sheds light on competitive versus cooperative behaviour amongst
humans. There is a fierce debate going on about the balance between selfishness
and altruism. ‘It is human nature to want more than what is necessary to survive
and reproduce – more resources, more social status, more mates – but it is also
human nature to want fairness and to shame individuals that behave selfishly’ (Penn
2003). The study of primate behaviour points at the existence of altruism in animal
populations, such as behaviour that is beneficial for the receiver and has a cost
for the giver irrespective of motive (de Waal 2007). The rationality of this can be
understood in a context of reciprocity and interdependence, including a sense of
empathy, retribution and justice. There is also in silico evidence that higher forms
of cooperation and organisation can arise out of individual self-interest (Nowak
and Sigmund 2004). But the more complex and consistent moral considerations
are probably uniquely human outcomes of social interactions. Even apart from
the ethical implications, these issues matter in sustainability science because most
Box 10.2. Rationality in economic science. Economists usually interpret ration-

ality as self-regarding materialism and not as consistency. The British economist
Kay argues in his book Culture and Prosperity (2001) that this occurs to keep
the analysis simple: ‘Economists insist on rationality because they do not like
the alternatives . . . [this is why] economists adopt a concept of rationality that
reduces to self-interest. It seems to offer an anchor in an ocean of otherwise
unpredictable human behaviour. The assumption of rationality gives economics
rigor that distinguishes it from other social sciences . . . self-regarding, material-
istic behaviour would be the norm because no other behaviour could persist in a
market economy’ (Kay 2001: 212).
History provides interesting lessons about behavioural motives and social con-
text. During the Cold War, Intelligence Services from the United States, CIA,
and the former Soviet Union, KGB, actively recruited spies in Western Europe.
They realised that persons qualifying as a spy would have different motives. The
FBI distinguished four possible motives, indicated by the acronym MICE: ‘M’ for
Money, ‘I’ for Ideology, ‘C’ for being prone to be compromised (financially or
sexually) and ‘E’ for Ego. This is a summary of why individuals may become free
riders and break with group solidarity and cooperation.
environmental issues are collective-action problems (§5.4). To solve such problems,

social pressure and coercion are supposedly effective because people care about their
reputation: shame works, and public shaming in combination with other incentives
may force cooperation (Axelrod 1997).
Evolutionary biology indicates that people have certain traits that result in mal-
adaptation to changing circumstances. Cognitive research shows that human beings
can solve extraordinarily complex problems and yet fail to do simpler tasks because
‘human cognition and behaviour are designed for ecological rationality and we rely
on simple heuristics rather than complex algorithms to solve problems faced by our
ancestors’ (Penn 2003). Perhaps, the computers of the information age will help to
overcome these shortcomings. A related question is why people prefer the short
term over the long term. Evolutionary game theory suggests that it is the outcome
of weighing expected present and future fitness pay-offs under uncertainty in the
quest for survival. In a sustainability science context, an interesting development on
the interface between life and social sciences is niche construction theory (Oddling-
Smee et al. 2003). Its main tenet is that organisms change their environment and,
in the process, modify the selection pressures of their own and other species. Such
a co-evolution of populations and their environments is backed up by field and
simulation experiments and makes the carrying capacity a dynamic element in pop-
ulation ecology. The idea that species ‘engineer’ and control their environment is
another bridge with humans, who can be seen as the ultimate niche constructors.
An empirical illustration are the historical population estimates for England 1545–
1975 and Japan 1100–1992, which clearly show jumps in the carrying capacity as a
consequence of human-environment-technology interactions (Meyer and Ausübel
1999).
10.4 Homo Economicus and Its Critics
10.4.1 Consumers and Producers

In most of the earth and life sciences, there is only a descriptive and no prescriptive
component. Intentions and goals may be assigned to plants and animals, but there is
no ‘should’ or ‘ought to’ involved. The rain just falls in a gravitational field. Plants and
trees occupy their place (‘niche’) within soil and climate constraints. The predator
kills the prey and is not judged for it. Optimality and efficiency in natural processes
are evolutionary effective but without normative content. But inevitably, with the
introduction of human beings, normative and ideological elements enter. This is the
essence of human freedom and responsibility. Therefore, observations on economic
behaviour are always somewhere between ‘what is’ and ‘what-should-be-done’, fact
and value, science and ethics or means and ends. As part of sustainability science, a
reflection on the image of man in economic science is a necessity. Again, this book
merely scratches the surface of the vast literature, so review the Suggested Readings.
In the 20th century, economic theory has become dominated by the neoclassical
economic school with a particular image of man. Humans – or agents using modelling
parlance – in economic decision processes are rational and choose with perfect
information and foresight in a perfectly competitive market. As consumers, they are
in isolation maximising their utility within a budget constraint. As producers, they
are constantly in competition on ‘free’ markets. This Homo economicus species is
rooted in the Modernist interpretation of science (rational) and evolutionary theory
(survival and competition). Sophisticated theories and models have been constructed
in (mathematical) decision analysis to extend and refine this image of man, but
without changing the essence. In Chapter 6, the discussion of worldviews made clear
the inherent reductionism in this standard image of man in (micro)economic science.
Consumer theory presumes a concave utility function u = u(c) with u per capita
utility and c per capita consumption. This theoretical construct represents a set of
ordered, fixed preferences of an individual person (§6.1). A lot of attention has been
given to theories and practises of strategies to maximise utility in a narrow (con-
sumers) and broad (nations) setting. In most macro-economic models, an aggregate
‘national’ utility function U is maximised across generations in order to deal with,
for instance, climate change (de Vries 2010). In formula:
t=T
max U (C) · e−rt dt (10.1)

t=1
with C the national consumption and T the time horizon over which the decision
matters. The term e−rt is the discount factor. It converts the annual income streams
into the discounted sumtotal in order to take into account that people value the
present more than the future (time preference).6 Such a formulation reflects a util-
itarian ethic and a belief in perfect markets. The dominant criticism focuses on the
presumed rationality and the lack of interaction with others and attempts are made
6 The sum total is called the present value of the income stream. The underlying behaviour is referred
to as time preference.
to introduce more realistic behaviour such as cooperation and altruism (Nowak et al.
1990, Brock and Durlauf 2001).
Already in the 1960s, in his book Models of Man, Simon (1957) criticised absence
of cognitive and social capabilities in economic models and searched for a genuine
theory of human motivation to understand individuals in organisations better than
prevailing market theories. He suggested that people are satisficing, that is: looking
for adequate and not for optimal strategies in everyday decision making. Besides, it
has been empirically confirmed that uncertainty and risk perception play an import-
ant role (Kahneman et al. 1999; van den Bergh et al. 2000) and that context and
availability of and differences in information influence people’s decisions (Gintis
2000, 2005). Status, experience and novelty in postmodern consumer culture are to
be considered in a larger system perspective (Hirsch 1977; Pine and Gilmore 1999).
Consumption of goods and services is much more than satisfying basic material needs
for food, shelter and so on. ‘Material things offer the ability to facilitate our particip-
ation in the life of society . . . our attachment to material things can sometimes be so
strong that we even feel a sense of bereavement and loss when they are taken from
us . . . Novelty plays an absolutely central role in all this’ (Jackson 2009). Consump-
tion is partly a substitute for religious consolation, a filling up of the ‘empty self’ and
combating a sense of meaninglessness (Handy 1998; Figure 6.2b). In recent times,
the resulting restless desire of the consumer is merging with the restless innovation
drive of the entrepreneur.
Traditional economics fails to consider the extent to which people are guided
by noneconomic motivations. The theory ignores animal spirits: ‘The economics of
the textbooks seeks to minimise as much as possible departures from pure economic
motivation and from rationality’ (Akerlof and Shiller 2009). A possible remedy
consists of theoretical fragments dealing with real-world phenomena such as con-
fidence, fairness, corruption and bad faith and money illusion. In recent years, sig-
nificant refinements are introduced in economic theory, and connections are made
with empirical findings in psychology and sociology. At the interface, new discip-
lines emerge such as evolutionary economics and behavioural and experimental
economics.
Production theory assumes profit-maximising entrepreneurs who enter markets
until the last entrant no longer sees an opportunity to gain a profit. It implies a
constant search for cost-minimisation. In its simplest form, it is written as:

min C = min pi Xi under the condition: P = P0 · X1α · · · Xmµ (10.2)
i=1..m
with C the production costs, Xi the production factor i = 1..m, pi the price of
production factor Xi and P the output. Primarily, the factors are in neoclassical
theory labour L and capital K. The condition represents the economic production
function, such as the possible combinations of inputs with which the output P can
be realised. Applying mathematics produces elegant formulas. But the assump-
tion that an entrepreneur or corporation is rational and that the market is without
oligopolism, trade barriers and institutional failure is often invalidated. The real-
world economy is never in equilibrium, innovation is an evolutionary process and
there are inherent tendencies toward concentration and centralisation away from the
Box 10.3. Competition in a globalising world. In Friedman’s The World is Flat

(2005), the following anecdote appears about the competitive world economy.
A Chinese manager posts an African proverb, translated into Mandarin, on his
factory floor:
Every morning in Africa, a gazelle wakes up

It knows it must run faster than the fastest lion or it will be killed.
Every morning a lion wakes up.
It knows it must outrun the slowest gazelle or it will starve to death.
It doesn’t matter whether you are a lion or a gazelle,
When the sun comes up, you better start running.
ideal market.7 In the 1980s, the first attempts were made for a new foundation, which
is rooted in evolutionary mechanisms (Nelson and Winter 1982). In the new branch
of evolutionary economics, the Homo economicus is replaced by a pragmatic and
adaptive organisation that is in constant search for bettering its goals in interac-
tion with other organisations and the environment. Behaviour is self-sustaining and
self-reinforcing and not necessarily (cost-)efficient. Its metaphor is selection and
adaptation in an evolutionary process, not utility maximisation or cost-minimisation
in a mechanical world (Döpfer 2005; Frenken 2006). A similar critique and alternat-
ive descriptions of economic production processes are developed in system dynamics
(Sterman 2000).
The simplifications in economic theory, many of them fashioned after classical
physics, are understandable from a technical and a psychological vantage point.
For early agricultural societies, it may be possible to give a description of human
behaviour that is largely correct and permits formal modelling. For a modern, (post-)
industrial society, one has to bring in more variety in motives and behaviour, and
formal mathematical analysis becomes difficult. An example is the penetration of
new products in the market, where the usual assumption of no returns to scale turns
out to be invalid in many situations (Appendix 10.1). I refer the interested reader to
the Suggested Reading, and I come back to it in Chapter 14.
10.4.2 Games, Dilemmas and Cooperation

An interesting and growing field is game theory.8 It analyses the strategic interactions
between individual agents: How do people interact? The most elementary device
in game theory is the pay-off matrix. It gives the costs and benefits for an agent
of two or more possible actions in an interaction with one or more other agents.
Examples are the hawk-dove game, where two animals upon encounter can either
7 The shortcomings of conventional economic theory and the historical explanation in terms of
the aspiration to imitate the natural sciences and of the limitations of available analytical tech-
niques are addressed in, for instance, Kirman (1993), Ormerod (1998) and Kay (2004). The ana-
logy with 19th-century equilibrium thermodynamics and chemistry is a notorious feature (Döpfer
2005).
8 Its beginning is usually associated with the book Theory of Games and Economic Behaviour (1944),
by Von Neumann and Morgenstern. With the book Evolution and the Theory of Games (1972) by
Maynard Smith, biology, ecology and ethology have also become areas of application.
Table 10.1. Application fields of game theory
Situation Kind of strategies/dynamics Discipline
Gene selection Evolutionary strategies Genetic biology

Animal behaviour Hawk-dove game Etology/ecology
Competition-cooperation in reproduction
Consumers Social limits to growth Economics
Solid waste disposal
Wars Arms race: Security dilemma Sociology
Crime Prisoner’s dilemma
Common Pool Resources (CPR) Monitoring and sanction regime
Relationships Duel: Game of chicken Psychology
Stay/divorce: Battle of the sexes
attack (hawk) or retreat (dove); the big and the small monkey, collecting coconuts
with or without cooperation; the two car drivers that compete on the issue of who
will first slow down in the face of an abyss. Table 10.1 lists some of these situations
and contexts of application.
One of the widely studied and better known games is the so-called prisoner’s
dilemma. This and similar dilemma situations (chicken game or ultimatum game)
can be represented with a pay-off matrix as shown in Table 10.2. Each of the persons
involved can either act in his or her own selfish interest (defect) or in the collective
interest (cooperate). Interest is indicated with a benefit (reward or gain) that can
be expressed in money but can also involve recognition, love and so on. In the
prisoner’s dilemma, the prisoners – who are not allowed to communicate – are told
that if both plead not guilty, they both get a short-term sentence because of lack of
evidence. However, if both plead guilty, they get a medium-term sentence. If one
of them pleads guilty and the other does not, he is set free and the other gets a
long-term sentence. If the pay-off is interpreted as the number of years in prison:
c = q > a = b > r = s > d = p. Clearly, it is best for both people to plead not guilty.
But in the absence of communication and trust, each considers the other person’s
Table 10.2a. Pay-off matrix of a dilemma in a two-person game
Agent B / Your adversary follows strategy:
Agent A / You follow 1. cooperate 2. defect

strategy: 1. cooperate (a,b) (p,q)
2. defect (c,d) (r,s)
Table 10.2b. Pay-off matrix in a social dilemma
Everyone else follows strategy:
You follow 1. cooperate 2. defect

strategy: 1. cooperate (a,b) (p,q)
Win-win solution You are the sucker
2. defect (c,d) (r,s)
You are the free rider Commons tragedy
reasoning and concludes that the most rational strategy is to plead guilty! In game
theory jargon, the Nash equilibrium strategy.
It matters, of course, whether you play the game once, twice or the rest of your
life – it is all the difference between a purchase in Timbuktu, doing an exam or
stepping into a marriage. There are many similar dilemma situations that can be rep-
resented by a pay-off matrix and agent strategies (Barash 2003). For instance, if you
play as agent A for the first time and distrust your adversary, you may defect and be
proven correct (r,s). If you wish to do business with the other person and build up a
good reputation, you may first cooperate, until the other person defects once or twice
and you feel abused (p,q). It may also be you who first cheats (c,d). To figure out
which strategies are optimal is the essence of game theory – but reflecting upon such
situations is millennia old. The ‘players’ are presumed to be competitive and antag-
onistic in the pursuit of their goals, but the resulting behaviour depends crucially on
the perceived costs and benefits of each strategy. If one of the prisoner’s feels that
prison is the best place to be for a couple of years, the whole game changes. Being
an experienced criminal also makes a difference. Game theory can clarify thinking
and puts complex social interactions in a new perspective – and may be fun, too.
The common view amongst economists is that rational agents are not likely
to cooperate in certain settings, even when such cooperation would be to their
mutual benefit. Based on considerations such as those in Olson’s book The Logic of
Collective Action (1965), it is asserted that rational, self-interested individuals will
not act to achieve their common or group interests unless the number of individuals
in a group is quite small or there is coercion or another way to make individuals
act in their common interest. Elementary game theory seems to confirm this, but
more advanced computer-based simulations show much richer possibilities. An early
and famous iterative game experiment was the tournament organised by Axelrod
in the 1980s around the question: What is the best strategy in a repeated infinite
prisoner’s dilemma (Axelrod 1984)? It turned out that the tit-for-tat (TFT) strategy –
always respond with the same strategy as your opponent – was almost always the
winning strategy in terms of overall score. This tournament set the stage for many
explorations into the complexities of cooperative behaviour (Axelrod 1997). The
resulting evolutionary game theory suggests that in iterative games cooperative
behaviour is also not rational, because the argument of the prisoners can be extended
backwards to the very first encounter – the Nash equilibrium strategy. But in silico
experiments show that patches of cooperative agents do survive and the conditions
under which this happens are intensely investigated (Nowak and Sigmund 2004;
Helbing and Yu 2009).
Often and more relevant in the present context are the situations in which a
single person ‘plays’ against a large group of others with similar ends and means.
These situations are called social dilemmas. They represent rather complex situations
in which the gain or advantage being sought is limited by the fact that others are also
reaching for the same goal. A social dilemma is intricately related to the context of
a decision in space and time. The dilemma is that each player is in a narrow context
best off if he acts according to his own individual interest, but in a larger context,
he is better off if he acts according to collective interest. ‘Once you start identifying
social dilemmas, it’s difficult not to see them everywhere, whether in public affairs
or private matters’ (Barash 2003).
Individual rationality will not induce people to follow strategies in their own
long-term self-interest without externally enforced rules. But the assumption of
individual selfishness or rational egoism in economic theory has limited validity
in civic society. Experimental economics has established that the rational egoist
assumption does indeed predict behaviour in auctions and in competitive market
situations quite well, but that it is less general than often thought and confined to
particular spots in societies, such as economic science students. Empirical fieldwork
and simulation experiments confirm the everyday experience that individuals in all
walks of life and all parts of the world voluntarily organise themselves for a collective
purpose. Evolutionary game theory is increasing the understanding of this presence
or absence of cooperation and the conditions for and dynamics of its emergence
(Wright 2000; Nowak and Sigmund 2004).
Theoretical and empirical analysis of resource-related social dilemmas and the
establishment of design rules for sustainable resource use are a great challenge
for sustainability science. Resource allocation issues are often social dilemma’s
about allocation in a public goods (PG) or common pool resource (CPR) setting
(§5.3). Think of the following examples in a physical, economic or social scarcity
frame:
r watering your lawn during a drought;
r littering in your neighbourhood;
r crowding in a natural park area;
r diverting your income to tax havens in order to evade taxes;
r membership of a trade union that fights for higher wages – also for you;
r protesting in a dictatorship with secret police;
r emitting carbon dioxide that causes global warming;
r admission to a class on sustainability science.
These are all situations that can be framed in terms of a pay-off matrix, with different
strategies and outcomes. As discussed in Chapter 12, the sustainable use of resources
such as land, water and fish are indeed dependent on our creativity and willingness
to to deal with these situations.
The key is how to induce cooperation and, relatedly, coordination. Historical
analyses indicate that there are more options than state- or market-based gov-
ernance regimes, and laboratory experiments begin to generate relevant insights
about human behaviour (Ostrom 2000; Ostrom et al. 2002, 2008):
r individuals contribute significantly of their own endowments for a public good
in a PG game; their contribution declines as the game progresses and/or the last
round is announced;
r face-to-face communication produces substantial increases in cooperation;
r individuals who believe that others will cooperate in social dilemmas are more
likely to cooperate themselves; learning to play the game also matters; and
r the availability of a sanctioning mechanism (punishment or reward) tends to
change individuals who are initially the least trusting into strong cooperators.
Human behaviour is rich and heterogeneous, with some people more willing to
initiate and sustain reciprocal and cooperative behaviour than others. More research
Table 10.3. Worldviews, population futures and ideas about human behaviour
Population and Human Behaviour
Statement 1: Population growth is still the greatest threat to sustainable development.

This is incorrect. Unfortunately, this may Every human being We have already
Indeed, the ingenuity be true. The poor deserves respect and overshot the Earth
and labour of people should quickly stop every soul on earth carrying capacity.
is the very source of reproducing, for their has potential and Birth control should
societal wealth and and our welfare. purpose. It is inner be a high priority and
progress. Provide development, not education of women
contraceptives. number of people is the best approach
that counts. to it.
Statement 2: Aging populations in the rich countries will become an ever larger economic burden because of
high-tech medical treatments.
It is indeed one of the This fear is justified. It is This is typically the It is indeed one of the
problems we are one more argument preoccupation of a problems we are
facing. The rich will against immigrants wealthy and facing. It is important
insist on costly who put pressure on materialist society. to prepare for a
high-tech medication, the necessary social Personal well-being is decent health care for
but the poor will security more than individual and availability of
benefit later, too. arrangements. health. medical services for
all.
Statement 3: Humans are selfish by nature.
This is the safest bet in Correct. Life is about Incorrect. Human This is true for the
everyday life, but it is survival. beings have the larger part of the
not necessarily bad unique capacity for human population.
for society! empathy and Precisely for that
altruism, as part of reason, government is
their spiritual needed.
development.
Statement 4: Most people let their actions, not their words, be determined by money.
Correct. It can be seen Of course. If someone It is true for many One only has to meet all
all over the world. As tells you he’s people – but what else the committed people
the saying goes, blood different, don’t can one expect in a in NGOs and
is thicker than water. believe him! materialistic society? governments to know
that this is not true . . .
is needed to establish firmer rules and construct a more adequate and integrative
theory than the prevailing one in economic science. There are probably no universal
solutions based on simple models, and we should look ‘beyond panaceas’ in the
search for sustainable management of social-ecological systems (Ostrom 2009).
We have explored in this section the image of man as seen in demography,
biology and economic science. In Table 10.3, some sustainability-related issues that
pertain to these disciplines are addressed from different perspectives. Clearly, the
ways in which these sciences see man are not unambiguous ‘facts’ but instead reflect
differences in worldview. Following the framework presented in Chapter 6, we invite
you to trace some of the differences and find your own position.
Box 10.4. Corporations. Agents are often thought of as individual people. But
institutions such as governments and businesses are the higher-level and more
powerful agents. The most important actors in the modern industrial economy
are the multinational corporations (MNCs). If corporate sales and country GDPs
are considered equivalent measures of economic importance, 51 out of the 100
largest economic systems were corporations in the year 2000 (www.ips-dc.org).
In resource-related sectors (fossil fuels, metals, chemicals and food processing),
a dozen MNCs create the larger portion of value added.
The corporation was invented in 18th-century Britain and ‘separated ownership
from management – one group of people, directors and managers, ran the firm,
while another group, shareholders, owned it . . . [Its] genius as a business form
was – and is – its capacity to combine the capital, and thus the economic power,
of unlimited numbers of people’ (Bakan (2004) 6–8). Early 20th century, the
Anglo-American corporation became a legal ‘person’ and has become a remark-
ably efficient wealth-creating machine. Their operation reflects the principles of
laissez-faire capitalism.
The legal status and power of corporations has dark sides. Bakan, a Canadian
Law professor, warns that social and environmental values are for corporations
not ends in themselves but strategic resources to enhance business perform-
ance. ‘Greed and moral indifference define the corporate world’s culture [in the
USA] . . . as pressure builds on CEOs to increase shareholder value, corpora-
tions do anything to be competitive . . . the managers may be kind and caring
people but are allowed, often compelled, by the corporation’s culture to disasso-
ciate themselves from their own values’ (Bakan (2004) 9). He characterises the
(American) corporation as psychopathic: It is irresponsible, manipulates, has a
lack of empathy, has asocial tendencies and is unable to feel remorse. Its common
features are: obsession with profits and share prices, greed, lack of concern for
others, and a penchant for breaking rules. It is ‘an externalising machine’, with
predatory instincts and a built-in compulsion to externalise its costs. Although
this may be a value-laden caricature, the unethical and amoral behaviour on the
part of big multinational corporations is at the roots of many of the world’s social
and environmental ills.
Corporate social responsibility (CSR) is for this reason one of the contentious
sustainability related issues. Within the business community, CSR is considered
a sign of good entrepreneurship and part of the solution for sustainable develop-
ment. Opponents see it largely as window dressing and argue that ‘the people’
have to reconceive the corporation as it was originally intended: a public institu-
tion whose purpose it is to serve national interests and advance the public good.
This is a forteriori argued with respect to banks. Such a transformation will not
be easy, but within certain worldviews, it is an essential part of the sustainability
transition. One implication is that the state reclaims its responsibility in high-
income countries. In low-income countries with a weak state, corporate wealth
easily mingles with national and local elites at the expense of the majority of
citisens. Here, NGOs should oppose such practises.
10.5 Simulating Human Behaviour
10.5.1 Introduction
If the wish is to improve sustainable development models and introduce human
behaviour, which way should we go? There are two strands of scientific research that
seem to come together. On the one hand, there are the traditional, qualitative social
science theories, and on the other hand, there are the formal modelling approaches
of CSS. Increasingly, bridges are being built between the two. But for both, it is not
possible to study human behaviour without some ‘image of man’ and the tension
between a bottom-up and a top-down view of the world (§10.1). In the bottom-up
view, humans act individually in interaction with their direct neighbours – direct not
only in physical but also in affective, mental and social sense. In contrast, in the
top-down perspective, planning and order from above dominate individual human
behaviour. It is the tension between the ‘invisible hand’ of the market and the ‘Big
Brother’ eye of the state or corporation, represented by the horizontal axis in the
worldview scheme of Figure 6.5. Of course, the distinction is not sharp. This chapter
focuses on three approaches in CSS and gives some illustrative applications, to
illustrate the bottom-up simulation model approach. It can, at least in principle, give
an idea about when to oppose and when to give in to, or even use, the forces ‘from
below’ with intelligent regulation and institutions ‘from above’.
CSS methods are complementing, not replacing calculus. Classical mathematics
is extremely useful for a variety of analyses, but often the analytical models cannot be
solved, even for a skilled mathematician, unless the system is drastically simplified.
The models become quickly intractable if discrete heterogeneous objects, informa-
tional delays and feedbacks and many nonlinear interactions are to be included
(§2.4).9 Moreover, the interpretation becomes often too strenuous to be successfully
communicable. Most of the more interesting and useful models in sustainability
science, therefore, rely on simulation techniques and software. The advances in this
area are inextricably linked to the digital, discrete world of computers.
Let us take the notion of agent as starting point. It is the equivalent of element
in system dynamics and defined as:
an agent is a model entity which can represent animals, people or organizations;
can be reactive or proactive; may sensor the environment; communicate with other
agents; learn, remember, move and have emotions (Janssen 2002).
From a cognitive science and artificial intelligence point of view, the question is
how to construct an adequate representation of a (human) agent with perception,
beliefs and goals, which in combination with intentions and motivations lead to
(inter)action (Gilbert and Troitzsch 1999; Phan and Amblard 2007). One way is
to refine the internal representation of simulated agents, in the direction of the I-I
quadrant (Figure 10.1). Another direction is to improve the interactions between
simulated agents (their topology), in the direction of the I-E quadrant (Figure 10.1).
The two approaches are interrelated, as indicated in the scheme of Figure 10.8. There
are no sharp boundaries in the wealth of recent applications.
9 This is even more true if space is explicitly introduced in the form of partial differential-integral
equations.
Figure 10.8. Scheme of the ways in which agent-based modelling is applied: cognition,
behaviour and interaction in a dynamic environment.
10.5.2 Cellular Automata Models

Have you ever seen fireflies flashing on and off above the seawater? It turns out
that the behaviour of a firefly can be understood by the interactions with the neigh-
bouring fireflies. This is the idea behind the cellular automata (CA) approach. The
CA framework assumes a number N of ordered cells. The ordering can be in one
dimension (in sequence on a line or circle), in two dimensions (in a 2-D grid on a
surface) or in three dimensions (in a 3-D grid in a cube) or in other not necessarily
spatial configurations. Which cells are considered to be neighbours has to be defined
explicitly; for instance, a neighbourhood can contain four or eight cells around a cell
in a 2-D surface (Batty 2005). Usually, the cells are adjacent squares and represent
a part of an imaginary or real surface. Each cell works as a kind of minicomputer: It
is in a certain state S(t) at time t and changes into a subsequent state S(t + 1) at time
t + 1. The cell states are updated simultaneously according to some transformation
rule, or algorithm, that depends on the state of the cell and its neighbours. The rule
can be, for instance, to switch state if and only if all neighbouring cells are in a
different state.
There are many real-world phenomena that can adequately be described on the
basis of CA, for instance, the already mentioned fireflies, chemical oscillators such
as the famous Belousov-Zhabotinski reaction and computing devices. CA models
simulate behaviour of individual elements (bottom-up) without coordination by a
king, a government or a CPU in a computer (top-down), although some coordination
can be introduced with synchronous updating. Macroscopic order can emerge from
local transition rules.10
10 There are several software packages available to construct CA models. A user-friendly package
is NetLogo 4.1 that simulates agents on a grid. The predator-prey and the firefly model are both
examples. Another package, specifically for geographical applications, is PCRASTER. It has been
developed at Utrecht University and provides a high-level simulation environment for cartographic
and dynamic modelling. In more advanced software tools, it is also possible to relax the assumptions
and have, for instance, cells of different sizes and shapes and non-identical rules.
a) b)
Figure 10.9. Simulation of a segregation process amongst 2,000 people with a similar-wanted
threshold of 30 percent (a) and 70 percent (b). In the 30 percent situation, people are happy
at much higher levels of integration than in the 70 percent situation (Courtesy: Netlogo 4.1
Segregation model). (See color plate.)
One of the simplest, yet evocative CA examples is the Game of Life, constructed
by Conway in 1970. Cells can be ‘alive’ or ‘dead’, and their state in every subsequent
step depends on the state of neighbouring cells in processes of birth, survival, loneli-
ness and crowding. For instance, sudden catastrophic forest fire can be simulated and
understood in a simple CA-based model as self-organised criticality (§9.6). Another
interesting application is a model of the mechanism of neighbourhood segregation
proposed by the economist Schelling in his book Micromotives and Macrobehavior
(1978). He argued that the segregation of people from different ethnic backgrounds
(a macro-phenomenon) could be understood from people’s individual behaviour (a
micro-foundation). This is an example of a threshold or tipping-point process. In
the Netlogo implementation, agents – known in the NetLogo jargon as turtles –
have preference for a certain percentage of people in their direct neighbourhood
being ‘similar’ (ethnicity, profession or whatever is seen as defining identity). At any
moment, an agent looks around and ‘sees’ how many agents in the neighbourhood
are similar and decides on the basis of this information whether to move to another
location or not. If the number is above a certain threshold (percent similar-wanted),
he or she feels happy and will not move. Because agents are initially randomly
distributed, a number of people will for a given average threshold still have less
similar people as neighbours than they would like and they decide to move. For low
thresholds, after a few rounds, a stable pattern emerges with everyone being satisfied.
If the average threshold goes up, more and more people start to feel unhappy and to
move around. After some dozens of rounds, a stable and rather segregated pattern
consolidates the situation (Figure 10.9). For even higher thresholds (>80 percent),
the system cannot find a stable configuration and keeps oscillating indefinitely. The
model gives a vivid demonstration of the formation of spatially segregated patterns
from a single simple rule. The approach has been used to explain and predict loca-
tions of ethnic conflict and to suggest mechanisms to promote peace in regions with
ethnic or cultural segregation tendencies (Lim et al. 2007).
Urban and economic geography and the environmental sciences are probably the
disciplines where CA models are most widely created and used (White and Engelen
1997). CA models in urban and rural geography use GIS-data, which characterise
the grid cells (elevation, temperature and so on). They represent a potential field
with attractors or repellers that direct the activities through, for instance, the rela-
tive attractiveness of an area for rice growing, tourism or office space.11 Elaborate
transformation rules on neighbourhood, diffusion and noise have been investigated
to simulate the growth of cities (Batty 2005). Also, in the natural and engineering
sciences, CA models are becoming available and applied, for instance, on water
erosion and pollutant dispersal (Wainwright and Mulligan 2004). Detailed simula-
tions of land use dynamics on the basis of a variety of transformation rules are used
to understand processes such as deforestation and overgrazing (Verburg et al. 2003;
Bouwman et al. 2006). A model-based exploration of land degradation risks from
tourism in Crete is amongst the early CA models to explore sustainable development
issues (Clark et al. 1995). CA models are used in many other settings, for instance,
to understand land price developments and to investigate the impacts and adapta-
tion options in a situation of rising sea levels. Indeed, it is hard to imagine how the
inherently local and intricate dynamics of sustainability-related issues can be under-
stood without spatially explicit CA models. In Chapters 11–13, some applications
are presented in more detail.
A more theoretical application of CA models is the simulation of the evolution
of agent behaviour on a grid in a repeated prisoner’s dilemma (Lindgren 1991). Each
agent has at any timestep an encounter with one of its neighbours. When agents meet
each other, they can apply one out of four strategies in their encounter (always defect
DD, always cooperate CC, reply tit-for-tat DC and reply anti–tit-for-tat CD, with
the action on the left the response to the action on the right) and each strategy has an
outcome that is determined by the pay-off matrix A (Figure 10.10a). Agents mem-
orise the outcome of encounters and adjust their strategy according to the replicator
equation, which is a core equation in evolutionary dynamics (Appendix 10.2). The
evolution of simple strategies in a population can be simulated and visualised on a
grid and over time.
In a repeated prisoner’s dilemma with 1,000 players interacting with everyone
else and the occurrence of mistakes, an evolutionary path unfolds in the course of 600
generations, with different strategies emerging and disappearing (Figure 10.10b).
There are, for instance, long periods in which tit-for-tat (DC) is dominant, but
cooperative strategies (CC) are sometimes significant for considerable periods of
time. The introduction of mistakes makes the strategies more sophisticated but also
more cooperative. If mutation, duplication, innovations or a stochastic environ-
ments are introduced, even more complex patterns emerge (Eriksson and Lindgren
2002).
In silico simulations such as these alert to the richness of real-world interactions
and warn against general statements about human nature on the basis of simplistic
11 Geographical information systems (GIS) are data representations in metric space. Note that spatial
information on economic-geographic systems is in discrete form, which makes the CA approach
more amenable to simulate change in space than the analytical equivalent of partial differential
equations.
(a) Action for player B

Pay-off matrix:
Cooperate Defect
Action for Cooperate (3,3) (0,5)

player A Defect (5,0) (1,1)
(b) 1000
DC DC
# of individuals
DD
CD DD
CC
DC
CC
CD
CC
0
generations 600
Initial population, 250 each of memory 1 strategies:

CC – Always Cooperate
DD – Always Defect
DC – Tit-for-tat
CD – Anti–Tit-for-tat
Figure 10.10. (a) The pay-off matrix for cooperation or defection (b) a trajectory of sub-
populations in a repeated prisoner’s dilemma with mistakes (Lindgren 1991).
theoretical models. These models are also simplifications, but they complement
the outcomes of even simpler game-theoretic analyses. They instill amongst some
advocates a postmodern enthusiasm: ‘In an evolutionary system . . . there is no single
winner, no optimal, no best strategy. Rather, anyone who is alive at a particular point
in time, is in effect a winner, because everyone else is dead. To be alive at all, an
agent must have a strategy with something going for it, some way of making a living,
defending against competitors, and dealing with the vagaries of its environment’
(Beinhocker 2005). These models are like mirrors with which we uncover our own
image. There is still a long way to go, but there is the promise that, one day, they
will be of help in designing and implementing effective strategies and policies for
sustainable development.
Box 10.5. Small worlds. In 1967, the psychologist Milgram sent out a number of
packets to people who had agreed to participate in an experiment, namely that
the packets had to be sent to someone in Massachusetts, but the participants were
only allowed to send their packets to someone they knew by first name. It turned
out that only a median of five intermediaries were needed to get the package
from the participants to the person in Massachusetts. This has been called the
‘small-world phenomenon’: Each of us is only six steps away from every other
person on earth. Network theory has shown that in a network where every vertex
is connected to its two neighbours, only a few additional random connections
bring forth the ‘small-world phenomenon’ (Watts (1999)). A few shortcuts do
shrink the world dramatically. In other words, a few species or persons can make
a huge difference in overall connectivity and hence function and performance of
a system.
10.5.3 Interaction: Networks

The properties of systems with many individual elements can be understood from
the interactions between the elements and from their individual behaviour. Network
theory or mathematical graph theory is a generic way to look at the behaviour and
structure that emerge from interactions. It started centuries ago but has only recently
moved into the analysis of large networks with the availability of large datasets and
the enormous computing power of today. The emphasis in this brief introduction
is on relevant examples from various disciplines. It is not the intention to go into
mathematical details. For some basic definitions, I refer to Appendix 10.3 and the
Suggested Reading.
A network consists of a number of vertices (or nodes) and edges (or links)
connecting nodes. In first instance, the analysis of a network is a description of
the vertices and edges and an examination of the properties of the network. A
fundamental property is the degree of a vertex. It is the number of edges to or from
that vertex. A plot of the fraction of vertices of a certain degree as a function of the
degree is called the degree distribution. Some networks have a degree distribution,
which can be approximated with a binomial distribution. These are often random
networks because their structure is what one would get if the edges are randomly
distributed amongst the vertices. The degrees of nodes in random networks do not
differ strongly: All nodes have roughly the same degree. In contrast, many real-
world networks have a degree distribution that can be approximated with a power
law. Such networks have many weakly connected and a few strongly connected
vertices (‘hubs’).
The availability of large computing power makes it possible not only to analyse
a particular network at a given moment in time but also to simulate the evolution
of networks over time. It is one of the fascinating applications in modern network
theory. The construction of a network in silico is done by setting up a number
of vertices and connecting them by adding edges according to certain rules.12 The
construction rule for a random network is very simple: Add edges between randomly
chosen pairs of vertices in a network of fixed size and see what happens. With this
rule, the probability p that any two edges are connected increases and one expects
a binomial degree distribution. It turns out, upon adding edges and thus increasing
p, the relative size of the largest connected (or giant) component (ngiant comnponent /n)
jumps discontinuously from small (∼0) to large (∼1). The network has no particular
structure.
A more interesting rule to construct a network is to connect vertices with already
many edges with a higher probability than those with few edges, such as the most
connected vertices that are preferred in the growth process. Therefore, the mech-
anism is called preferential attachment. It generates in a growing network scale-free
structure. Figure 10.11a shows the dendritic configuration typical of such networks.
The degree distribution is approximated with a power-law function. In system terms,
it is a reinforcing feedback. In system archetype terms: Success breeds success, the
rich get richer.
12 The first attempts to examine the growth of networks were by Erdös, Rapoport and others in the
1950s.
a)
b)
Figure 10.11a,b. Example networks. Network (a) is an extremely efficient network, (b) an
extremely resilient one. Examples are all networks with 100 vertices and 250 edges (Brede
and de Vries 2009a).
There is an interesting link between efficiency and resilience that can be

examined with network analysis (Brede and de Vries 2009a). Efficiency of, for
instance, transport and communication networks can be related to the average
shortest pathlength. Efficient networks are star-shaped and consist of a periphery
made up of a highly connected core and outward branching nodes. Resilient net-
works are associated with the stability around a presumed existing stationary state
and have the property of redundancy (§9.6).13 One extreme is an efficient network
with a single ‘super-hub’ (Figure 10.11a). The other extreme is a network with long
loops (Figure 10.11b). Networks that realise a trade-off between both extremes
exhibit core-periphery structures, where the average degree of core nodes decreases
but core size increases as the weight is gradually shifted from a strong requirement
for efficiency and limited resilience towards a smaller requirement for efficiency
and a strong demand for resilience. Both efficiency and resilience are important
requirements for network design and a balance between the two may be a principle
of system robustness and resilience.
There is a rapidly increasing number of analyses of real-world networks and in
silico produced networks. Network analyses are done in different fields of science,
such as ecology (food web stability), transport and other infrastructure (congestion or
emergency), epidemiology (spread of disease) and the social and economic sciences
(information, power or money) (Appendix 10.3). Most investigations of networks
focus on the statistical properties of real-world networks. The rapid advance of
Internet and social media and the instabilities of financial systems accelerate the
interest in networks.
The networks in the different domains exhibit large structural differences. Many
metabolic networks appear to be scale-free networks, with a few highly connec-
ted nodes (Jeong et al. 2000). There are a few hubs that dominate the overall
connectivity and are, therefore, the system’s most sensitive elements. ‘Living
13 Mathematically, the smaller the largest eigenvalue of the adjacency matrix of the network, the more
stable and thus resilient it is.
1000
S. Europe
Population (thousands)
slope=–0.76
100
N. Europe
slope=–0.83
10
1 10 100 1000
Rank
Figure 10.12. Rank-size distribution for European cities in 1750 (Blum and Dudley 2001).
The lower curve is for northern European cities, and the upper curve for southern European
cities.
networks’ that grow by accretion do often have the dendritic shape, with telling
examples in biology and geography. But the structure of 16 high-quality food webs
from a variety of ecosystems in the United States differ from what one would expect
in random networks and in scale-free networks in terms of characteristic path length,
clustering coefficient and degree distribution (Dunne et al. 2002). Instead, the six-
teen food webs show a variety of functional forms and their size plays a role in the
network degree distribution. Investigations such as these enhance understanding of
structural complexity and help to design better ecosystem management strategies
(Ruiter et al. 2005).
There are also interesting applications in geography. The structure of river
catchments can be reproduced in in silico experiments with rules based on erosion
dynamics (Rodriguez-Iturbe and Rinaldo 1997; Buchanan 2002). These rules are
also characteristic of the percolator process in physics. In urban geography, the
frequency versus size histogram of cities and towns exhibit a scale-free structure:
a few very large cities, many towns and many more small villages.14 The urban
population distribution in northern and southern Europe around 1750 are examples
of such a rank-size distribution (Blum and Dudley 2001; Figure 10.12). They can
be produced by adding edges randomly and multiplicatively, that is, proportional to
existing density (Anderson 2005; Batty 2005). Is the mechanism really understood?
Early theories focused on the rich get richer preferential attachment mechanism that
is also at work in scale-free networks. It is not difficult to imagine how it might work:
Once a village becomes a town, it starts to attract more people than neighbouring
villages because of all kinds of advantages (trade, scale and so on), and this further
reinforces its growth. But this is probably not the whole truth, and a more accurate
representation of cities as entities in an evolving network is needed.
Increasingly, the social sciences use network analysis to explore structure of
financial, economic and social systems. An analysis of interbank payments between
14 It is called the rank-size distribution of city sizes and its power law representation is also called Zipf’s
law.
Box 10.6. Hierarchy in networks. Hierarchical networks with a modular struc-

ture are found in urban systems and organisations (Pumain (2006)). Intrinsic
hierarchy is found in a variety of communication networks, whereas networks
with a geographic organisation, such as power grid networks and Internet router
nodes, do not apparently have such a topology. In archaeology and anthropo-
logy, the observation of hierarchy in (past) human groups has been questioned.
There are reasons to assume that the ways in which human groups have organised
themselves are much richer, but the ones with a high degree of centralisation and
concentration may have left the most clear traces.
The sociologist Simon offers a rationale for the prevalence of hierarchical
structures: ‘Complex systems will evolve from simple systems much more rapidly
if there are stable intermediate forms than if there are not. The resulting complex
forms in the former case will be hierarchic. We have only to turn the argument
around to explain the observed predominance of hierarchies amongst the complex
systems nature presents to us. Amongst possible complex forms, hierarchies are
the ones that have the time to evolve’ (Simon (1969) 90–91/98–99).
commercial banks pointed at a network with both a low average path length and
low connectivity (Soramäki et al. 2007). There is a tightly connected core of banks
to which most other banks connect. The degree distribution is scale-free over a
substantial range. A recent analysis points at the high connectivity amongst the
financial institutions of the world (Schweitzer et al. 2009). Mutual shareholdings
and closed loops are a sign that the financial sector is strongly interdependent. It
can make the network unstable, obstruct market competition and pose systemic
risks.
Another social science application is interpersonal networks of friendship/
influence. Over thirty years old, there is the example of the analysis of networks
of influence regarding nuclear power policy in The Netherlands (de Vries et al.
1977). The vertices are the members of governing boards, directorates and advisory
councils of all the institutions (companies, ministries and councils) involved in the
nuclear policy process, or the stakeholders. Whenever a person has positions in two
or more of these institutions, he is assumed to be a channel for the exchange of
information and the exertion of influence. These ‘double-functions’ make up the
network edges. The analysis revealed that only a handful of people (<20) carried
a rather dense network of relationships that connected construction firms, (public)
electricity generating companies, research institutions and ministries. Such organisa-
tional networks around large-scale ventures can promote effective decision making,
but they also tend to exclude or ignore alternative viewpoints. Social network theory
has been greatly refined in the last decades, but the applications in sustainability
related areas are still rare.
Are network theory and applications relevant for sustainability science? Every
complex system that consists of many interacting units can be described and under-
stood as a network. A striking number of complex systems fit into a few, broad
classes. If rather rigid laws determine the development pattern of such systems, it
constrains the options for growth and planning. For instance, systems with scale-free
architectures are quite susceptible to targeted interventions – both positive and neg-
ative ones (Figure 10.11). The removal of one or a few ‘hub’ species from a food
web with a scale-free–like structure can cause its collapse. Targeted vaccination in
a population suffering from an epidemic outbreak can contain the spread of the
disease. As a corollary, scale-free network systems are remarkably robust against
random errors because most of the vertices can be removed without noticeable
impact on its structure. Therefore, the network structure may explain the robustness
of some ecosystems against random forms of interference (pollution or invasion)
and the vulnerability of centralised systems to (terrorist) attack. Understanding
such generic mechanisms and ‘laws’ of system structure and evolution can teach us
about a system’s boundaries and permissible trajectories and about its development
potential.
10.5.4 Multi-Agent Simulation: Behavioural Variety

When the emphasis is more on the representation of agent behaviour (such as
memory, choice criteria, and so on) than agent linkages, one tends to speak of
agent-based modelling (ABM) or multi-agent simulation (MAS).15 With the advent
of object-oriented simulation languages and platforms, it is increasingly feasible to
incorporate agents in simulation models. They are ‘intelligent agents’ in the sense
that they are autonomous entities in a virtual world, with a model of their envir-
onment, rules and possibly a model of themselves. Their ‘intelligence’ ranges from
response behaviour to information signals to anticipating the intentions of other
agents. In these simulated worlds, memory and adaptation are introduced by acting
on the basis of past experiences and discovering new strategies or rules. Agent-based
models have an inherent tendency to focus on heterogeneity and interdisciplinarity.
The self-organisation exhibited in ABM suggests that order emerges ‘from below’,
such as from rather simple rules at the individual level (Kauffman 1995; Holland
1996; Beinhocker 2005). According to some practitioners, it supports the market
ideology – an interesting link between worldview and scientific method.
Do you remember the story of Easter Island (§3.2)? A human population can
build a society and destroy it. The Lord of the Flies (1954) by Golding is another
famous story about an island. In this book, primitive rituals underneath a layer of
civilisation take over in a group of boys on an isolated island after an imaginary
plane crash. The story suggests that destruction does not have to come from cutting
all trees or polluting all water – it may come from a loss of ‘civilised’ behaviour.
In Chapter 3, other evidence of collapse because of human behaviour is mentioned
(§3.4). But how does such a process really work? ABM may help to understand such
processes with in silico experiments.
The first people to make and use ABM were probably the makers of computer
games (SimAnt, Civilization, SimCity and so on). A first, more rigorous model of
how individual agents can evolve is Sugarscape. It was presented in 1996 by Epstein
15 Also in differential equations and system dynamics models, one can interpret model behaviour
in terms of agents. For instance, the interaction between a prey and a predator (equation 9.5) is a
condensed representation of random encounters in a homogeneous environment, but the simulation
of behaviour is too restricted to be acceptable for social scientists.
and Axtell in their book with the telling title, Growing Artificial Societies: Social
Science from the Bottom Up. Again, the story happens on an island – the ‘ideal gas’
of the social scientist.
Imagine an island of 2,500 km2 divided in 50 × 50 squares (cells) of 1 km2 . The
cells have only one characteristic: the amount of sugar, which differs across cells. At
the start of the imaginary simulated world, there are agents with a limited behavioural
repertoire: look around in neighbouring cells for sugar (inspection), move and eat
sugar (metabolism). They differ in two respects: the inspection capability (how many
cells can be inspected) and metabolic rate (sugar needs for survival). The sugar is
unevenly distributed across the island, as with real resources. What will happen in
this simulated world? The agents – with different capabilities and positions – start
off and look for sugar. When they move around and hit more sugar than they need,
they accumulate it, which extends their life. If they cannot satisfy their needs, they
starve. Thus, agent A and agent B in Figure 10.13 develop in different directions
because of their different initial positions, inspection capability and metabolic rate.
Whereas agent A can collect hardly enough sugar to survive, agent B accumulates
with a bit of luck enough sugar to live on for a long time. Notice that there is no
direct interaction, unlike in the evolutionary game presented earlier.
What makes this simple model interesting is that, when run with many agents
and many times, it is possible to construct the statistics on the system in the form
of probability distribution functions of key variables. It turns out that the evolution
of this sugar economy shows some remarkable similarities with real-world observa-
tions. Although agents start from a random distribution (location and metabolism),
the simulation experiments generate a shift from an egalitarian situation with an
uniformly distributed wealth (accumulated sugar) to a highly skewed distribution in
which only a few agents own most of the wealth and a large group is near starvation.
This feature is consistent with the Pareto distribution found for income. It is an
emergent property in the sense that it is not the consequence of any one cause in
particular, but instead the outcome of the mix of initial distributions, rules of the
game and luck. Macroscopic order emerging from microscopic behaviour.
Much behaviour is mediated by information exchange and a number of ABMs
have been made to explore the penetration of innovations. An example is the simula-
tion of the spread of environmentally friendly innovations in a population of farmers
(Weisbuch and Boudjema 1999). Individual agents are given partial knowledge of
their environment and are endowed with different preferences and motivations and
an internal representation including the ability to learn and adapt. Encounters with
other agents, structured according to an imposed social network, can ‘infect’ a poten-
tial adopter with the innovation. With the assumption of full rationality of economic
theory, an exact prediction of the fraction of farmers Feq that will adopt the envir-
onmentally friendly option is possible and can have any value between 0 and 1.
But with social networks and agent heterogeneity, a form of bounded rationality is
introduced. Herd behaviour occurs and prediction is difficult. There is a tendency
for Feq to be near 1 or 0, not unlike what happens in the Polya model (Appendix
10.1). Another area of ABM-applications is the spread of diseases such as AIDS and
mad cow disease, inspired by epidemiological models (Dunham 2005). In Chapters
11–13, I present some examples.
Figure 10.13. Illustration of agent behaviour in Sugarscape. Each piece of vegetation accounts
for one units of resource (sugar).
Random events and rule-guided behaviour are standard features in ABM. Order
is seen to come from below and not as a result of a deterministic process – more like
a flight of starlings than an externally forced pendulum. The formulation in terms of
discrete events and objects makes it easier to link them to everyday experiences and
to cognitive maps. A possible drawback is that an ABM is more difficult to control
and validate. A fancy interface graphics may hide the internal complexity. Their
relevance in sustainability science is that it permits an experimental investigation of
the interactions between humans and their environment (Jager and Mosler 2007).
Computer simulations are the ‘laboratory’ in which social science hypotheses about
human behaviour can be tested and intervention policies can be explored. Given
the large and increasing complexity of sustainability issues, this may prove to be a
decisive factor in the design and implementation of adequate policies.
Box 10.7. Do ABMs have added value? Modelling human behaviour, including
motives and habits, leads to a more in-depth understanding of the forces behind
(un)sustainable development and resource exploitation and provides more
meaningful levers for policy experiments. But is it really giving new insights?
Janssen (2002) has explored the advantages of adding spatial heterogeneity to
models of range-land management in Australia. Experimenting with various types
of sheep behaviour using a relatively simple model – including fire and water-
points – it is found that non-uniform grazing due to different rules regarding
biomass density, waterpoint locations, and other sheep, may cause system beha-
viour to differ significantly from uniform grazing assumed in non-spatial models.
This could influence notions of ecosystem resilience and indicate the need for
more prudent exploitation strategies. A cautious conclusion is that ‘understand-
ing the implications of spatial processes is vital in the large open paddocks of
range-lands; however, it is likely that these processes are also important, perhaps
in a more subtle way, in other environments’ (Janssen 2002:123).
10.6 Summary Points

If designing policies for a sustainable future, one must ask the questions: Who am
I and who is the other? Or broader: What is the prevailing image of man? A brief
exploration into biology, demography and economics gives some indicative answers.
Here are some general insights:
r The image of man in science is largely focused on the individual and the outside,
as part of Modernism. The methods of complex system science (CSS) can com-
plement the traditional social science approaches in building stronger knowledge
about interactions between and cognitive aspects of individuals.
r Population dynamics (demography) and phenomena such as the demographic
transition are important for an understanding of (un)sustainable development.
The biological, notably the evolutionary, perspective on human behaviour has
deeply influenced the prevailing self-image in industrial societies.
r The image of man in economic science, known as Homo economicus, is biased
towards rationality and competition. The variety of human needs and behav-
iours is explored explicitly in new areas such as behavioural and experimental
economics and evolutionary game theory.
r Their relevance for sustainability is, amongst others, the study of strategic
behaviour in a public goods (PG) or common pool resource (CPR) setting.
It gives more realistic ideas about competition versus cooperation and the indi-
vidual vs. the collective and associated worldviews.
Novel methods to introduce heterogeneity in behaviour are cellular automata

(CA), network dynamics and agent-based models (ABM). They offer the prospect
to better understand human behaviour in complex social-economic systems, notably
at the micro-level.
Death is very likely the single best invention of Life. It is Life’s change agent. It
clears out the old to make way for the new.
– Steve Jobs, former CEO Apple
I suspect that the fate of all complex adaptive systems in the biosphere – from single
cells to economies – is to evolve to a natural state between order and chaos, a grand
compromise between structure and surprise . . . [from the complex system point of
view] we cannot know the true consequences of our best actions. All we players can
do is be locally wise, not globally wise.
– Kaufmann 1994:15/29
To the eternal triple question which has always remained unanswered, Who are we?
Where do we come from? Where are we going? I reply: As far as I, personally, am
concerned, I am me; I come from just down the road; and I am now going home.
– Pierre Dirac, quoted in Zeldin, The French
Il y a dans tout homme, à toute heure, deux postulations simultanées, l’une vers
Dieu, l’autre vers Satan. L’invocation à Dieu, ou spiritualité, est un désir de monter
en rade; celle de Satan, ou animalité, est une joie de descendre.
– Baudelaire
Translation:
There are in every man, at all times, two simultaneous tendencies, one toward God,
the other toward Satan. The invocation of God, or spirituality, is a desire to follow
the road upward; to invoke Satan, or animality, is taking delight in falling down.
SUGGESTED READING
An entertaining yet thorough introduction to game theory and its applications.

Barash, D. The Survival Game: How Game Theory Explains the Biology of Cooperation and
Competition. New York: Times Books, 2003.
A thorough introduction in the theory and applications of cellular automata, with emphasis on
geographical applications.
Batty, M. Cities and Complexity – Understanding Cities with Cellular Automata, Agent-Based
Models, and Fractals. Cambridge, MA: MIT Press, 2005.
An overview of environmental science with contributions from a variety of disciplines.
Boersema, J., and L. Reijnders. Principles of Environmental Sciences. New York: Springer
A popular, well-written account of network theory and its applications.
Buchanan, M. Nexus/Small Worlds and the Groundbreaking Science of Networks. New York:
Norton Company, 2002.
One of the early introductions on agent-based simulation models.
Ferber, J. Multi-Agent Systems – An Introduction to Distributed Artificial Intelligence. London:
Addison-Wesley, 2000.
A lucid and thorough introduction into evolutionary game theory and its applications.
Nowak, M. Evolutionary Dynamics – Exploring the Equations of Life. Cambridge, MA:
Harvard University Press, 2006.
Introduction into complexity science with applications in various fields.
Perez, P., and D. Batten, eds. Complex Science for a Complex World – Exploring Human
Ecosystems with Agents. Canberra: ANU Press, 2003.
Series of papers on causes and consequences, such as aging, urbanisation, life stages, kinship
networks and so on.
Demeny, P., and G. McNicoll. Population and Development Review 37. Special Volume on
Demographic Transition, 2011.
A practical guide on the technique of agent-based modelling.
Railsback, S., and V. Grimm. Agent-Based and Individual-Based Modeling. Princeton, NJ:
Princeton University Press, 2011.
Advanced introduction into nonlinear dynamic models and their applications.
Strogatz, S. Nonlinear Dynamics and Chaos – With Applications to Physics, Biology, Chem-
istry, and Engineering. Boston: Addison-Wesley Publishing Company, 1994.
USEFUL WEBSITES
r www.nicheconstruction.com/ is a site that gives a broad and well-referenced overview of
niche construction theory.
r www.unfpa.org/swp/swpmain.htm provides info on the State of World Population reports
by UNFPA.
r sedac.ciesin.columbia.edu/gpw/index.jsp is a site operated by CIESIN at Columbia Uni-
versity with many data and maps on population density, urban-rural and so on.
r www.pbl.nl/en/themasites/phoenix/index.html is the website of the demographic model
PHOENIX of the Netherlands Environmental Assessment Agency (PBL).
r www.ifs.du.edu/ is a site showing an interactive exploration of population-economic-
environmental futures of the world’s countries. The International Futures (IFs) model
can also be downloaded.
r www.who.int/en/ is the site of the World Health Organization (WHO) with the World
Health Report and data on health.
r www1.fee.uva.nl/cendef/ is the site of the Center for Nonlinear Dynamics in Economics
and Finance at the University of Amsterdam.
r www.bitstorm.org/gameoflife/ is a clear illustration of the Game of Life. Many more
versions, including 3-D ones, can be found on the Internet.
r jasss.soc.surrey.ac.uk/12/1/6/appendixB/EpsteinAxtell1996.html is a description of the
Sugarscape model.
r www.youtube.com/watch?v=rN8DzlgMt3M&feature=relatedshows a predator-prey
simulation in 3-D, with bottom-up rules creating macroscopic patterns.
Wikipedia has some excellent entries on complex system methods and applications.
SOFTWARE
r ccl.northwestern.edu/netlogo/ is the website of the Netlogo agent-based simulation plat-
form. Netlogo 4.1 can be downloaded for free.
r www.pcraster.nl is a GIS-MAS package developed at Utrecht University, with diverse
applications in geography and geophysics.
r www.pajek.org is the website of the network construction software package, freely down-
loadable.
r www.openabm.org/ is the OpenABM site with tutorials, a model library and discussion
forum on computational modelling
r www.railsback-grimm-abm-book.com/index.html is an introduction to agent-based mod-
elling procedures using NetLogo (Railsback and Grimm 2010).
Appendix 10.1 Models of Economic Decision Making

A simple model to allocate fractional shares amongst competing options is the multi-
nomial logit (MNL) model, a member of the logit/probit models. Suppose a company
or a consumer must make a decision between two alternative options – such as pur-
chasing equipment or a product of brand A or brand B. First, define a broadly
defined ‘cost’ of each option: price, reputation, environmental impacts, fuel use and
Appendix 10.1 Models of Economic Decision Making 335
recyclability. The higher the cost, the less attractive the option. There is a large
group of potential buyers and the number of times that option A is chosen in a cer-
tain period or number of decisions is called its market share µA . The MNL-formula
states that the market share of option i converges to:
c−λ
A 1
µA = or µA = (A10.1)
2 1 + (cB /cA )−λ
c−λ

i
i=1
with µi the market share of option i. The competing options are assumed to be fully
substitutable and the parameter λ is a measure of the substitution elasticity. A small
λ-value implies that cost differences hardly affect the choice (inelastic substitution).
Taxes or subsidies to affect choice, therefore, do not work. For a high λ-value, it is
the opposite.
The model is consistent with economic production theory, but for some products,
it is incorrect. Notorious examples are where the apparently most attractive option
lost are the QWERTY keyboard and the VHS video system. The mechanism is that,
once a product or process has reached a certain market share, economies of scale
and scope, access and inertia of infrastructure and other factors operate to give it
an even larger market share. Reinforcing feedbacks create one winner takes it all.
Economists call it positive or increasing returns to scale.
It can be simulated with a different model: the Polya generator (Arthur 1994).
The choice process is seen as a series of random choice events that are not inde-
pendent. A large number of producer or consumers choose in a sequence of n events
either option A or option B and option A has been chosen nA times. Let the prob-
ability that during the next event option A is chosen be equal to P(A) = F(nA ), such
as the probability depends on how often option A has been chosen before. If you
normalise F(nA ) into G(µA ) with µA = nA /n, it depends on the function G what
will happen. If G declines with increasing µA , there is one stable attractor and no
returns to scale. This is the usual assumption in economics. But if G is proportional
to P(A), there can be multiple attractors and positive or negative returns to scale
and the actual development becomes sensitive to what is chosen during the first
few events. Processes become path-dependent and inventions can get ‘locked-in’,
or become dominant even in the face of clear disadvantages. The objective may be
to get the system out of the attractor basin, instead of containing it within (§9.6).
History matters!16
An example of real-world processes with positive returns to scale are industrial
location: The probability that a new firm is established in a region increases
with the number of firms already there. It is known as first mover advantage and
operates in many infrastructure systems. An example is the choice of standards
in ICT-infrastructure/networks: If the task to be performed requires connections
with other users, each next user has an advantage in choosing the one with most
users connected. Other examples are alternating current (AC) versus direct current
(DC) electricity, gasoline versus biofuels, and private transport by car versus public
transport. Reputation is an example that illustrates the importance of information:
16 A vivid illustration of its consequences in economic decision making is given by Kirman (1993) with
his ant model.
Table A10.1. Descriptions of networks in different disciplines
Network theory Vertex Edge
Physics Site Bond

Computer science Node Link
Ecology/economy Point, site, place Link, connection, bridge
Sociology Actor Tie
The more an option earns a good reputation (quality, reliability and so on), the more
often it is chosen. This Polya generator model leads to a better understanding of how
technologies penetrate markets. It is important for the rapid, continued and wide-
spread introduction of more resource-efficient ways of producing and consuming.
Appendix 10.2 Replicator Dynamics

Let the pay-off for an interaction of strategy i with strategy j be given by {aij } then one
can define the fitness of a strategy by the product of its frequency and its pay-off. For
instance, if a strategy CC has a certain pay-off aCC and its frequency in a particular
round is xCC , then the expected pay-off in the agent population is:
n

fCC = xCC aCC (A10.2)
j=1
If pay-off is equated with fitness, the replicator equation determines the rate of
change in the frequency of the strategy:
dxCC
= xCC ( fCC − ϕ) (A10.3)
dt
with ϕ the average fitness across all agents and strategies (Nowak 2006). In this
way, behavioural strategies can invade the system. I refer the interested reader to
the Suggested Reading for more details.
Appendix 10.3 Network or Graph Theory

A network is defined as a set of system elements. The elements are called vertices
and the links are called edges, but there are different names in different fields
(Table A10.1). The number of vertices, n, is called the size of the network. The
edges represent some kind of exchange (goods, influence, progeny and so on) and
may be assigned directions, weights or sizes, as, for instance, in transport, trade or
communication channels. The vertices may have a variety of properties, such as
position in space, membership of a class (plants, animals, peoples, cities and so on)
or characteristics of such membership (weight, age, metabolic rate, communication
frequency, number of inhabitants and so on). If the vertices are used to formulate a
square matrix and the value of edges is indicated in the cells, the matrix is called an
adjacency matrix.
The degree of a vertex is the number of edges incident on, or connected to that
vertex. The shape of the degree distribution is an important network property. It can
Appendix 10.3 Network or Graph Theory 337
Table A10.2. Overview of systems that have been studied as networksa
Mean If
vertex- power-law
No. of No. of Mean vertex Clustering distribution:
Network Type vertices n edges m degree z distance l coefficient C exponent α
Metabolic Undirected 765 3,686 9.64 2.56 0.09 2.2

Marine food Directed 135 598 4.43 2.05 0.16 n.a.
web
Neural Directed 307 2,359 7.68 3.97 0.18 n.a.
Train routes Undirected 587 19,603 66.79 2.16 n.a.
Power grid Undirected 4,941 6,594 2.67 18.99 0.10 n.a.
Internet Undirected 10,697 31,992 5.98 3.31 0.035 2.5
Citation Directed 783,339 6,716,198 8.57 3.0
Sexual contacts Undirected 2,810 3.2
Physics Undirected 52,909 245,300 9.27 6.19 0.45 n.a.
coauthorship
Company Undirected 7,673 55,392 14.44 4.60 0.59 n.a.
directors
a Source: Newman 2003.
be approximated with an analytic function Pk , which should be interpreted as the

probability that a randomly chosen vertex has k connections. Strictly speaking, one
should distinguish a histogram, that is based on empirical data, from its mathematical
abstraction, that is called a probability distribution. When the distribution follows a
power law: Pk = k−α , one gets a straight line for a plot of Pk as function of k on a
double logarithmic or log-log scale. They are called power-law networks. They are
also called scale-free networks because they remain unchanged under rescaling of the
independent variable k with a multiplication factor m, as is seen upon introduction
of κ = mk (then pk′ ∼ κ −α = (mk)−α = m′ k−α = m′ pk ).
Another interesting large-scale property of networks is the mean degree z, which
is defined as the average number of edges connected to a vertex. It equals the ratio
of the number of edges and the maximum possible number of edges. The transitivity
or clustering coefficient C indicates the extent to which ‘the friend of your friend
turns out to be also my friend’. The community structure of a network is a measure
of the presence of subgroups of vertices with a high density of edges within them and
a lower density of edges between them. The resilience of a network is an indication
of a network’s response to the removal of vertices. If vertices are removed, paths
between connected vertices will become longer, and at some point, the vertices will
become disconnected.
Newman (2003) collected data on a variety of networks and characterises four
loosely connected categories (Table A10.2):
1. biochemical and bio/ecological networks: metabolic pathways, protein interac-

tion networks, genetic regulatory networks, food webs, neural networks;
2. technical and infrastructure networks: river and canal networks, road and railway
networks and airline routes, electric power grids and telephone and Internet
communication networks. Many of these networks are related to space- and
geography-governed infrastructure and designed for distribution;
3. information (or knowledge) networks: a classic example is the network of cita-

tions, more recent ones are the World Wide Web and networks of people’s
preferences for objects used in targeted advertising; and
4. social networks: intermarriages, sexual contacts, mail contacts, friendship rela-
tions, business and collaboration relationships, and influence networks of exec-
utives and politicians. The vertices are (groups of) people that are in some
forms linked to each other. They overlap with communication and information
networks.
11 Agro-Food Systems
11.1 Introduction: The Human Habitat

The topic of a previous chapter is ‘pristine’ nature, but pristine nature is becoming
scarce. Human populations have been very successful as a species and now have
impact upon one-third or more of the biosphere. Humans have always interfered
with the natural environment: Agriculture and pastoralism are sustained forms of
manipulation of plants and animals. During the process of agrarianisation, they
have learned to exploit soils, fish and forests for the provision of food, fiber and
fuel. The larger part of the world population is still engaged in agriculture, and
food production is still the most direct and largest interface between humans
and nature. In monetary terms, agricultural products make up only a few percent
of gross world product (GWP), partly because subsistence agriculture is outside
conventional economic statistics. If food processing, transporting and retailing are
considered, the economic role is much larger and in the order of 8 percent. But in
physical terms, agriculture and the larger food system are of enormous importance.
This chapter explores sustainability issues in present-day agricultural populations
and systems. Agriculture is understood here in the broad sense of agroecology and
the agro-food systems in the world, from local production to global trading and
processing. It connects the Latin words colere and ager, the latter meaning field.
There are many ways in which human populations interfere with the ‘natural’
world of biogeochemical cycles and ecosystems. Although one can discern some
universal tendencies in human-environment interactions, farming and animal hus-
bandry are intrinsically local and the natural cycles still dictate the human activities
in space and time. Land is locality par excellence. It is only logical, then, to focus
first on the human habitat. The relevant branch of science is geography, along with
ecology a core discipline in sustainability science.
11.1.1 Land and People

The natural energy flows are an essential part of the biosphere dynamics and the
mean net primary production (NPP) is a good measure of it. The NPP reflects the
growth of vegetation, which depends on location, soil quality, elevation, precipita-
tion, temperature and other factors. There is as of yet no internationally accepted
339
Table 11.1. Characteristic data for different land use land cover categories around the year 2000. Ten to
20 percent of the forested and mountainous regions are under some kind of protectiona
Population density
Area %Frac Mean Rural Urban
(mln Land NPP persons/ persons/ Income
Unit km2 ) (%) kgC/m2 km2 km2 ($/yr/cap)
Water Marine 349.3 (−) 0.15 0 0 0

Coastal 17.2 7.4 0.52 70 1,105 8,960
Inland water 10.3 4.4 0.36 26 817 7,300
Forest/woodland Forest: (sub)tropical 23.3 10.0 0.95 14 565 6,854
Forest: temperate 6.2 2.6 0.45 7 320 17,109
Forest: boreal 12.4 5.3 0.29 0.1 114 13,142
Dryland Dryland: hyperarid 9.6 4.1 0.01 1 1,061 5,930
Dryland: arid 15.3 6.5 0.12 3 568 4,680
Dryland: semiarid 22.3 9.5 0.34 10 643 5,580
Dryland: dry subhumid 12.7 5.4 0.49 25 711 4,270
Island Outside island states 2.4 1.0 0.72 82 1,220 12,162
Island states 4.7 2.0 0.45 14 918 11,148
Mountain Mountain 300–1,000 m 13 5.6 0.47 3 58 7,815
Mountain 1,000–2,500 m 11.3 4.8 0.45 3 69 5,080
Mountain 2,500–4,500 m 9.6 4.1 0.28 2 90 4,144
Moutain >4,500 m 1.8 0.8 0.06 0 104 3,663
Polar Polar 23 9.8 0.06 0.06 161 15,401
Cultivated Cultivated: pasture 0.1 0.0 0.64 10 419 15,790
Cultivated: cropland 8.3 3.5 0.49 118 1,014 4,430
Cultivated: mixed (crop 26.9 11.5 0.6 22 575 11,060
and other)
Urban Urban 3.6 1.5 0.47 0 681 12,057
a Millennium Ecosystem Assessment 2005.
land use/cover classification system, but several United Nations–led initiatives are
underway.1 Quantitative assessments have, therefore, still uncertainties and ambi-
guities. Marine ecosystems (oceans with >50 metres (m) water depth) are dominant
with more than two-thirds of the Earth’s surface. Much of the terrestrial surface is
covered by drylands: hyper-arid deserts (6.5 percent) and (semi-)arid and dry sub-
humid regions (34.3 percent) (Reynolds and Stafford Smith 2002). The remainder is
forest/woodland (28.4 percent), mountain areas (24.3 percent) and cultivated land
(23.9 percent). Due to definitions, overlaps occur and the sum does exceed 100
percent.
Data on land cover, NPP, population and income densities from the Millennium
Ecosystem Assessment (MEA 2005) are listed in Table 11.1. Not surprisingly, the
highest rural population densities tend to be at medium values of the mean NPP
levels. There is more to eat than at the low NPP values of the tundras, and there is, or
at least was, less competition with other species than in the tropical forests with their
1 Land cover refers to the physical and biological cover over the surface of land. Land use is defined by
natural scientists in terms human activities and by social scientists and land managers more broadly,
including the social and economic circumstances. Land cover can be observed in the field or by
remote sensing. Land use is more difficult to establish and needs an integrated approach.
high NPP-values. Probably the most thorough dataset of croplands and pastures so far
has been made by Ramankutty et al. (2008). They report 15 million square kilometre
(km2 ) or 1.5 billion hectare (Gha2 ) of cropland (90 percent confidence interval
12.2–17.1) and 28 million km2 or 2.8 Gha of pasture (90 percent confidence interval
23.6–30). This is 12 percent and 22 percent respectively of the total ice-free land
surface. Using a broader definition of agriculture, namely those areas where at least
30 percent of the land is used for cropland or highly managed pasture, agroecosystems
cover 28 percent of total land area excluding Greenland and Antarctica (WRI 2000–
2001).
Income (GDP/cap) as a measure of activity correlates positively with NPP, with
the exception of the desert and polar regions where the high income is from fossil
fuels or other resources, and the boreal forests and (sub)tropical forest/woodland
areas where climate and ecology suit other species better than humans. The economic
activity per unit area on the basis of population density data and economic data at
lower than national (state or province) level can be estimated (Nordhaus 2005).
The resulting gross cell product (GCP, or GDP per unit of cell area), on a 1◦ × 1◦
spatial grid, shows that income tends to increase with distance from the equator,
while GCP has a clear maximum in the temperate zones (5◦ C to 20◦ C temperature
zone).
It is also possible to map the distribution of human individuals with respect to
four geographical and climate parameters: altitude, nearness to permanent rivers
and sea coasts, temperature and precipitation (Small and Cohen 2004). The overlay
maps and their statistical representations clearly show the importance of climate,
vegetation and geography as determinants of human habitat (Figure 11.1). The data
indicate that human beings have been quite successful in adapting over a wide range
of temperature and precipitation, with the preferred niche largely between 10◦ C and
20◦ C and between 500 to 2,000 millimetres per year (mm/yr). It suggests a reinforc-
ing loop of success breeding success ‘The talent for ingenious adaptation . . . has
always provoked a need of yet more ingenuity . . . the very success of an ecological
adaptation creates a need to develop some means of keeping population growth
under control’ (Reader 1988). A second observation is that human populations tend
to live overwhelmingly (>70 percent) in areas less than 100 kilometres (km) from
large rivers, less than 1,000 km from a sea coast and below 1,000 m elevation. They
overlap with the large river plains in the world, and it points at a solid biogeographical
foundation for human evolution and development.
11.1.2 Anthromes
In §9.2, maps were presented of potential ‘natural’ vegetation. But the activities of
human populations are influencing directly and indirectly the larger part of the bio-
sphere, so it is more relevant and interesting to construct actual ‘humans-in-nature’
land use/cover maps – maps of humanscapes.3 Ellis and Ramankutty (2008) and
2 Land area is indicated in million hectares (Mha) or billion hectares (Gha). Sometimes km2 are used,
1 km2 = 100 ha.
3 Romanova and colleagues of Lomonosov State University in Moscow have reconstructed such
maps for the pre-Roman and Roman era in Europe and other regions and times – see de Vries and
Goudsblom (2002).
Temperature Variability Precipitation Variability

600 600
500 500
# People (x 1 million)
# People (x 1 million)
400 400
300 300
200 200
100 100
0 0
–10 0 10 20 30 0 10 20 30 40 50 0 1000 2000 3000 4000 0 2000 4000 6000
15 15
Land Area (million km )
Land Area (million km2)

2
12 12
9 9
6 6
3 3
0 0
–10 0 10 20 30 0 10 20 30 40 50 0 1000 2000 3000 4000 0 2000 4000 6000
200 200
2
2
150 150
Σ People / Σ km
Σ People / Σ km
100 100
50 50
0 0
–10 0 10 20 30 0 10 20 30 40 50 0 1000 2000 3000 4000 0 2000 4000 6000
Annual Temperature (ºC) Annual Range (ºC) Annual Precipitation (mmlyr) Annual Range (mmlyr)
Figure 11.1. Population density, calculated as the population in land of a certain elevation/
distance divided by its corresponding area (thick curves) (Small and Cohen 2004). The thinner
curves are a rescaling; the peak in this curve at 2,300 m is the Mexico Plateau.
colleagues (Ellis et al. 2010) have constructed maps of what they call anthropogenic
biomes or anthromes. It is based on the definition that:
a biome is a unit of species composition and ecological processes, and a function of

climate and climate-derived parameters.
The authors propose a total of twenty-one categories: dense settlements (two),

villages (six), croplands (four), rangelands (three), seminatural (four – mostly wood-
lands) and wildlands (two).4 The anthrome map is shown in Figure 11.2.
The anthrome map clearly shows the large differences with the potential vegeta-
tion map (Figure 9.5). The wildlands and, to a lesser extent, the seminatural wood-
lands and rangelands cover almost 50 percent of the world terrestrial surface. Human
settlements and land use processes increasingly dominate the other 50 percent. The
maps for the year 2000 confirm in more quantitative detail other statistical estimates
(Table 11.1), notably that:
r more than 80 percent of all people live in densely populated urban and village
biomes;
4 The categories have been slightly updated since the first anthrome map because of new insights –
for instance, one cropland and one wildland category are put under the new category seminatural
in view of rather ambiguous demarcations.
Figure 11.2. Map of anthropogenic biomes or anthromes (Ellis et al. 2010). (See color plate.)
r agricultural villages are the most widespread form of densely populated biomes:
one in four people lives in them; and
r most of the terrestrial biosphere has been altered by human residence and
agriculture, to the extent that now less than a quarter of Earth’s ice-free land is
‘wild’, with only 20 percent of it forested and more than 36 percent of it barren.
The anthrome maps have also been constructed for the years 1700, 1800 and 1900
(Ellis et al. 2010). The reconstruction and other data sources indicate that the era of
massive expansion into new lands (extensification) is over. Total cultivated area in
this classification has hardly changed since 1980 and amounts to about 700 million
hectare (Mha). There are still hundreds of millions of ha of cultivable land, but
their conversion – also for non-agricultural purposes – would have serious social-
economic and ecological impacts because many of these lands are in the tropical
forests of Latin America, Africa and East Asia.
The anthrome maps give an integrated view of both ecosystems and the human
activities within these systems, and they make Social-Ecological Systems (SES) the
natural unit of investigation. An SES is defined as:
an integrated system of ecosystems and human society with reciprocal feedback

and interdependence. The concept emphasizes the ‘humans-in-nature’ perspective
(www.resalliance.org).
Subsequent chapters often speak of SES because they offer the best way to identify
and appreciate local variety and contingency and to explore sustainable development
pathways that are based upon it.
4000
India China Other Asia
N Africa & M East C&S America Rest Of World
3000
Mln people
2000
1000
0
1950 1960 1970 1980 1990 2000
Year
Figure 11.3. Agricultural population in the world, 1990–2006, in million persons (source of
data: faostat.fao.org).
11.2 Agricultural Systems
11.2.1 Agro-Food Systems in the World

Many people in industrialised high-income countries tend to associate the natural
environment with rural areas and natural parks. Where they live, intensive agricul-
ture is concentrated in a few highly productive regions and dominates the food sys-
tem. Abandoned rural areas are re-inhabited by rich urbanites. People enjoy forests
and lakes in natural parks for recreation and travel as tourists and/or adventurers
in the remaining wild nature abroad. But for nearly 3 billion people, the natural
environment is an everyday experience, for sustenance, enjoyment and survival. In
the 1960s, about 80 percent of the world population was engaged in agriculture and
in 2005 still more than one-third of the world’s people are classified as agricultural
population: persons depending for their livelihood on agriculture, hunting, fishing
or forestry5 (Figure 11.3). They live in the agricultural villages and towns of the
world, much as their ancestors did for millennia. Essentially, they have a biomass
economy and their knowledge of local ‘ecosystem services’ is their most valuable
and necessary asset for their livelihood.
Figure 11.4 shows in a simplified way routes and endpoints of human popula-
tions in their food procurement. Several evolutionary pathways have led towards the
systematic and intensive interference with natural processes in modern agriculture.
Categories and transition processes are not sharply defined and estimated popula-
tions and areas are rather uncertain, but the scheme in Figure 11.4 can guide you,
with a few exceptions, through the various agroecosystems in the present-day world.
Very small numbers of humans are still living from hunting and gathering and
survive in the periphery of the anthroposphere – a couple of hundreds of thousands of
5 According to the FAO definition and data. The estimate comprises all persons actively engaged in
agriculture and their non-working dependants. Also, an estimated 800 million people are actively
engaged in urban agriculture (WRI 2000–2001).
Figure 11.4. Various routes from hunter-gatherers to forms of agriculture (de Vries and
Goudsblom 2002).
indigenous inhabitants of tropical forests and of the Arctic. Millions of herders live a
(semi-)nomadic life and practise pastoralism in landscapes with short-lived grasses as
the dominant vegetation: treeless temperate and tropical grasslands, savannas with
woody overstory, and arid shrubby grasslands. The area under permanent pasture is
huge but unevenly spread across the globe. While more than 80 percent in Oceania,
Sub-Saharan Africa, South America and Southeast Asia is considered pastureland,
the fraction is less than 15 percent in South and Southeast Asia. In the 1970s, an
estimated 10 to 20 million people lived more or less as nomadic herders, mostly in the
belt from the Sahara to Mongolia, in an area of about 2.5 Gha or 20 percent of the
total ice-free land surface (Grigg 1974). Their number has increased – one estimate
suggests 20 million to 25 million herders in the Sahel region alone (Niamir-Fuller and
Turner 1999).6 A recent estimate considers pastoral areas the home to more than
200 million people who are distributed over a grassland area of 5.25 Gha (∼0,04
person/ha). They own some 35 million cattle, two-thirds in the African continent,
and 9.5 million sheep and goats (Disperati et al. 2009). The numbers are imprecise
because of different definitions and classifications and lack of data. They also change
over time because grasslands spontaneously transit to forest and, reversely, forest is
cleared for pasture.
Nomadic herders are vulnerable for external disturbances such as a severe
drought. Therefore, and also to control and tax them, governments and aid agencies
encourage them to switch or diversify to subsistence farming. The success of such
a switch is unclear because mobility is their best strategy to cope with disturbances
6 The estimate refers to households in which 50 percent of gross revenue comes from livestock or
livestock-related activities.
(Pedersen and Benjaminsen 2008). Horticulture and agroforestry can be comple-

mentary sources of subsistence for these populations and thus enhance food security.
An outgrow of pastoral nomadism is extensive commercial grazing or ranching,
practised in large-scale, low-productivity farms in parts of the Americas, Australia
and New Zealand for the production of beef and wool. Where land was or became
scarce, it was followed by intensification and introduction of novel breeding and
other techniques. Cattle, pigs and poultry stocks are grown and processed at high
densities in large manufacturing plants (‘bio-industry’). It allows the optimisation
procedures and practises of engineering science and the large-scale output and sales
of animal-based products in present-day urban supermarkets. Evidently, the view of
nature of those involved has undergone a transformation, too.
Shifting cultivation, the most prominent and highly varied form of early farming,
still exists in parts of Southeast Asia, Latin America and Africa. Around 1970, an
estimated 50 million people were living from shifting cultivation, cropping 10 to 18
Mha in a total area of 100 to 110 Mha (∼0,5 person/ha) (Grigg 1974). More recent
estimates of the number of people involved indicate a range from 40 to 500 million
people, and the area may be a significant fraction of the 850 Mha of secondary
forest (Mertz et al. 2008). Depending on available nitrogen and duration of fallow
period, the system has yields in the order of 1 ton/ha (feeding ∼0,5 person/ha)
with low labour input and hardly any capital input. It is a useful system in tropical
forest regions. Yet, it is declining under the pressure of higher population density,
globalisation and rising food demand, which forces a transformation to permanent
agriculture and cash crop plantations.
The intermediate, modern and mechanised forms of agriculture can be divided
into two main groups: wet-rice culture and mixed farming.7 The wet-rice system
requires intensive land use with high labour inputs. It has several advantages, notably
high energetic yield and minor soil degradation (Lansing 2006).8 It has developed
mainly in Southeast Asia in the large river basins because much water is needed.
Natural floods are the most important irrigation form. Livestock is unimportant in
wet-rice regions. Because up to half of the rice harvested used to be consumed on
the farm, it is often considered subsistence farming. The so-called Green Revolution
has given a boost to rice yields to values of 3 to 4 ton/ha (feeding >2 person/ha),
with the use of hybrid varieties as well as with a need for abundant irrigation and
fertiliser. It also led to a rapid increase of production for the market, although the
fraction of rice traded internationally is still only about 7 percent of production.
The agricultural system, which has developed in Europe and spread to North
America and other temperate areas of European settlement, is known as the mixed
farming system. Three trends since the early Middle Ages determined present-day
mixed farming in Europe: reduction of the fallow period, the extinction of common
rights (enclosure) and increasing importance of livestock (Grigg 1974, 1987). Other
elements in its evolution were the introduction of the heavy plough and the three-
field system, the cultivation of roots and grasses as fodder, more and more intense
7 Mediterranean agriculture and tropical crop plantations are two other, more or less, separate sys-
tems.
8 Possibly, the success of rice cultivation allowed populations to grow to such densities that there was
no place for animals and that vegetarianism became the dominant food practice. It is an example of
a connection between cultural-religious practices and agricultural circumstances.
Box 11.1. Science for agriculture: Liebig’s Law. One of the early ‘laws’ about
agricultural systems is Liebig’s Law of the Minimum. It is a principle stating that
growth is controlled by the scarcest resource and not by the total of resources
available. The principle is founded in the observation that increasing the nutrient
flows does not lead to more plant growth – growth only increases if a nutrient is
applied, which is most scarce of all the nutrients needed by the plant. This nutrient
is called the limiting factor. This insight was essential for the development and
application of fertilisers. In chemical kinetics, the limiting factor is known as the
rate determining step.
Liebig’s Law has been extended to biological populations and natural
resources. In a logistic growth process, it is the scarcest resource in the envir-
onment that starts to slow down growth (§9.2). Scarcity often has a spatial and
temporal dimension. For example, the growth of a biological population may not
be limited by the total amount of resources available throughout the year, but
by the minimum amount of resources available to that population at the time of
year of greatest scarcity. The principle is also used in the formulation of economic
production functions and market substitution processes.
use of inputs in farming and a subsequent shift to more high-value products such
as livestock and vegetables. When real incomes started to rise in the 19th century,
a further shift from cereals to high-value products took place. With Europe’s mild
winters, fertile soils, long days in summer and reliable rainfall, European wheat
farmers nowadays attain amongst the highest yields in the world (6 to 8 ton/ha).
In the process, agriculture has become more and more dependent on science
and manufacturing for inputs such as industrial fertiliser, tractors, transport and
refrigeration equipment and fuel and electricity to run them. It also has to comply
with ever stricter hygiene, environmental and of late animal welfare standards, partly
as a result of the complexity and associated vulnerability of the system as a whole.
These developments were interwoven with trends in globalisation – which in the
19th century was tantamount to European colonialism – and commercialisation.
Modern mixed farming as practised in Europe and North America is now a highly
commercialised form of near-industrial activity with high yields and high levels of
non-labour inputs.
11.2.2 Food: Needs and Consumption

Evidently, any exploration of sustainable development has to acknowledge the great
variety and diversity of agro-food systems in the world. Farmers tend to simplify and
control their environment, but the combination of inherent variety of the land and
growing connectivity of food producers make the global agro-food system very
complex. The first question is: What are the food needs in the world? Often, one
speaks of food demand but the nutritional food requirement for a healthy person
can be determined quite well so the link between needs and demand is relatively
straightforward. Food demand and, if available, food consumption is determined by
the number of people and their nutritional needs and habits. This can be measured
not only in physical units but also as the part of income spent on food. In a 1895
article, the Belgian economist Engel analysed the cost of living of worker families and
observed that food expenditures as a proportion of income decreased with increasing
household income. In Belgium, for instance, the share of food expenditures in total
household expenditures was around 60 to 65 percent between 1860 and 1920, then
declined slowly to 50 percent by 1950 and dropped precipitously to less than 18
percent by 1980 (Swinnen et al. 2001). Economists speak about Engel’s law, which is
confirmed by more recent trends in emerging economies.
Economists define in the analysis of food demand the notion of income elasticity
as the percentage increase in food expenditures per percent increase in income
(Appendix 11.1). If the elastitiy is less than one, the product for which this happens
is considered a necessity. Examples are potatoes and other starchy staple food items.
If the elasticity exceeds one, it is called a luxury good. Income elasticities as well
as similarly defined price-elasticities are estimated from statistical data. The results
confirm that the fraction of starchy staple food in total food consumption decreases
with rising income, but that consumption of protein-rich food items like fish and
meat – the luxury items for the poor – increases.9 It implies that demand for food
depends not only on average income, but also on income distribution. It also points
at the high vulnerability of poor people for rising food prices. Throughout history,
rising and volatile prices of staple food have hit the poor segments of inegalitarian
societies the hardest and caused riots and revolts. It will not be different in the 21st
century, as the social unrest in African countries in the first decade of the century
shows.
These trends are part of the food or nutrition transition (Kearney 2010). There
are other trends that further complicate the analysis. The food industry endeavours
to upgrade and process food, which keeps the food expenditures higher than it would
otherwise be. In growing economies, food preparation at home is replaced by instant
meals and outside snacks and dining, as part of modernity. These trends show up in
macro-economic data as an increasing fraction of value-added in the food processing
and hotel and restaurant sectors.
11.2.3 Food: Resource and Potential Supply

A second look at the system is the biophysical resource perspective: How much
food can be supplied? Soils provide the substrate for vegetation – and thus for
food, fodder and fiber. Critical environmental variables are the climate, notably
temperature and precipitation, and derivatives such as length of growing season and
growing degree days. Soil properties such as organic matter and moisture content
are other critical variables. Based on these variables, classification in agro-ecological
zones and potential vegetation are made (Figures 9.2 to 9.5). It is possible to use an
extended set of variables to determine the potential yield of specific crops (wheat,
rice, maize and so on) for specific locations. It sounds straightforward, but it is not.
9 For instance, in 1996–1997 the fraction of income spent on food in Argentina, Brazil and Mexico
was more than 50 percent amongst the poorest 10 percent of the population and less than 20 percent
for the richest 10 percent (Sabates et al. 2001). Incorporation of income distribution may cause a 5
to 10 percent difference in food demand estimates (Cirera and Masset 2010).
For instance, one has to define precisely for which crop and in which situation,
because genetic potential, landscape and other factors also play a role. In a couple
of regional and global models, potential yields are calculated across large areas.
Figure 11.5 shows potential yields for the three most important grains (wheat, rice
and maize) around the year 2000 on a 0.5◦ × 0.5◦ (about 50 × 50 km) resolution as
simulated with the IMAGE2.2 model (Bouwman et al. 2006).
Of course, these potential yields are not realised in practise because of natural
variations and suboptimal management. The difference between the potential and
the actual yield is referred to as the yield gap. Licker et al. (2010) assign crop yield
data on a 10′ × 5′ grid for eighteen crops and separate them in 100 different climate
zones. Using global maps of growing degree days and soil moisture availability, they
analyse the role of other factors such as management practises across regions with
similar climates. The potential yield in a climate zone is in their analysis defined as
the lowest level of the best 10 percent of the grid cells (90th percentile). Therefore,
it is not a biophysical potential, but today’s near-maximum achievable yield. There
is a significant spread in the potential yields thus defined, even within similar climate
zones. Comparison between this potential yield and the actual yields indicate room
for improvement. The results point at possible improvements of 60 percent for wheat,
nearly 40 percent for rice and 50 percent for maize with practises now adopted by
the most productive farmers.
Neumann et al. (2010) follow a somewhat different agro-economic method.
They specify an agro-engineering production frontier, which relates the yield to
growth-defining, growth-limiting and growth-reducing factors. The first group con-
sists of climate-related variables such as photosynthetically active radiation (PAR),
temperature, CO2 concentration and crop characteristics.10 Taking into account pre-
cipitation and soil fertility constraints, they determine the production frontier. The
role of the growth-defining (frontier) and growth-limiting/-reducing (inefficiency)
factors is explored by estimating a log-linear relationship between yield and a set
of variables that are considered proxies for these growth determinants. Aggregated
across regions, the results for 26 regions suggest similar yield gaps as in the Licker
et al. (2010) study, namely 43 percent, 47 percent and 60 percent for wheat, rice and
maize, respectively. Defining efficiency as the ratio between the observed yield and
the production frontier for that crop in that region, it is possible to identify the extent
and determinants of yield gaps (Figure 11.6). Therefore, it seems that there is ample
room for yield improvements, but the researchers are cautious with firm conclusions
because what really can reduce the yield gap requires a better understanding of the
region-specific socio-economic factors.
11.2.4 More Food: Can It Be Supplied Sustainably?

Most of the increase in world food production since 1980 came not from more crop-
land, but from yield improvements. The agricultural land area increased with less
than 10 percent since the 1960s. Globally averaged yields have jumped between 1980
and 2000 from 3.3 ton/ha to 5.7 ton/ha (Lambin et al. 2003). The average price of
10 PAR indicates the fraction of sunlight that can be used by organisms in photosynthesis. It is radiation
in the 400 to 900 nanometres realm.
350
Temperate cereals Rice
Maize Millet, Sorghum
Yield (km2 harvested area per grid cell)

0 501 - 1,000 1,510 - 2,000 > 2500
< 500 1,010 - 1,500 2,010 - 2,500
Figure 11.5. Potential yields for some important crops, based on simulation with the IMAGE-model (PBL). (See color plate.)
Figure 11.6. Estimate of the efficiency in wheat production (Neumann et al. 2010). Efficiency is defined as the ratio of the observed output to the
corresponding frontier output. (See color plate.)
351
352
353
ten major crops more than halved in this period. This successful intensification has
been possible thanks to a series of changes: irrigation; multiple cropping; applica-
tion of more and more specific fertilisers, herbicides and pesticides; improved land
management (optimal planting, pest controls and so on); and genetic advances (rais-
ing the seed/biomass ratio or harvest index and new breeding techniques). More
innovations in the agrosciences are expected and some people talk about a second
Green Revolution and a livestock revolution (Jaggard et al. 2010; Thornton 2010).
It is unclear to which extent this is a solid expectation or wishful thinking. In any
event, it is widely believed that the increase in food production necessary to relieve
hunger and feed the 2 to 3 billion additional people is only possible with further
intensification. As the calculated yield gaps presented in the previous section indic-
ate, there is still room for improvement with existent practises and technologies. But
will it aggravate existing problems of erosion, salinisation, pollution and other forms
of degradation? Will it make farming less sustainable?
Soils are the crucial resource in this respect. A straightforward definition of
sustainable soil management is to preserve its inherent fertility. The inherent fertility
is in essence a legacy from the past: Some areas in the world are endowed with very
fertile soils of metres thickness, whereas in other places there is only weathered-
down rock. The natural processes that created the legacy are still at work, but they
are mostly slow in comparison with the rate at which humans are now intervening.
And not only the rate but also the extent and variety of human interventions have
become a matter of great importance. A number of soil characteristics determine
the suitability for food production. Sustainable soil management thus implies from
a system dynamics point of view (§2.3) that:
r the rate of outflow of nutrients does not exceed the rate of inflow by natural and
human processes; and
r the rate of inflow of fertility-reducing substances (‘pollutants’) does not exceed
the rate at which they flow out or are broken down to non-damaging substances.
The scheme in Figure 11.7 shows the most important processes in soil formation and
losses and in soil fertility maintenance.
There was widespread decline in ecosystem functions and productivity in the
last decades of the 20th century. It prohibits certain forms of land use and affects
hundreds of millions of people (Figure 9.7). There is controversy about causes and
consequences of such land/soil degradation (§ 9.4). Many soils in the world have been
degraded by a mix of natural and man-induced processes, notably wind and water
erosion. In fragile drylands with large natural variations in rain, human activities
have accelerated the degradation and the inherent soil fertility has declined because
of loss of organic matter and salinisation. Such irreversible processes on a time-
scales of several decades are already impacting on food yield and output in several
regions in the world (Brown 2004). The areas of most concern are the densely
populated regions in the tropical zones, where cropping and foraging intensity are
already above sustainable levels and where demand for food and feed is expected to
increase further.
Ongoing intensification requires more inputs and creates new and sometimes
harmful side effects. An important intensification measure is irrigation. Water is
a crucial input, but water use is unsustainable in many places, as discussed in the
Figure 11.7. Scheme of the factors involved in (un)sustainable soil use.
next chapter. More efficient water use is a necessity. There is also use of more
chemical and machinery inputs and a continuous flow of genetic innovations with
only partly foreseeable impacts. One of the consequences is a massive and still
increasing interference with the natural nitrogen (N) and phosphorous (P) cycles.
Figure 11.8 is a map of the N- and P-balances for natural ecosystems and agriculture.
In the densely populated regions, the emission densities (in kg/km2 /yr) are very
high. Excess values contribute to eutrophication of water and soils with deleterious
concequences in the long run.
Further intensification thus aggravates some of the undesirable trends that have
come along with the successes:
r the widespread and, in many places, still increasing use of inputs pollute soils,
rivers, lakes and seas, notably eutrophication from N- and P-compounds;
r increasing health risks at several levels, from the individual farmer who uses
herbicides and pesticides to the system-large disruptions because of disease
outbreaks or the impacts of using genetically modified organisms (GMOs);
r larger pressure on (ground)water resources for irrigation, contributing to further
loss of biodiversity and ecosystem services; and
r in a larger system context, the agro-food system becomes increasingly dependent
on finite resources, notably oil and phosphorous.
One can, therefore, expect that to supply more food will bring new risks, with
sometimes dramatically different consequences in different locations.
Parts of the agro-food system, notably meat production, contributes to green-
house gas emissions and the subsequent climate change. Direct effects from rising
2000 2000
356
-2 -1 -2 -1
N in kg km yr P in kg km yr
<100 <50
100 - 200 50 - 100
200 - 400 100 - 200
400 - 800 200 - 400
800 - 1600 400 - 800
1600 - 3200 800 - 1600
3200 - 6400 1600 - 3200
6400 - 12800 3200 - 6400
> 12800 > 6400
GO scenario 2050 GO scenario 2050
AM scenario 2050 AM scenario 2050
Figure 11.8. Total balance for (left) N and (right) P for natural ecosystems and agriculture for 2000 and for 2050 according to
two scenarios (Bouwman et al. 2009). (See color plate.)
11.3 Stories from the Real World 357
atmospheric carbon dioxide (CO2 ) concentrations have in the theory and the labor-
atory a positive impact on yields through a higher rate of photosynthesis and a higher
water use efficiency. It can be beneficial for some regions. Generally speaking, how-
ever, the consequences of higher surface temperatures and changes in precipitation
patterns from climate change as a consequence of rising atmospheric concentrations
of CO2 and other greenhouse gases can be severe: extreme rainfall or drought, heat
stress during heat waves, and large-scale adaptations and relocations. And despite
massive field tests and model simulations, it is still uncertain how yields will respond
in reality to changes in temperature and precipitation (IPCC 2007; PBL 2010a).
Rising concentrations of surface ozone (O3 ) in the Northern Hemisphere could
annihilate any yield increase from CO2 fertilisation (Jaggard et al. 2010). Whatever
the scientific details, there is widespread concern about the possible consequences of
climate change for agriculture and much research is done to identify the vulnerable
populations and to strengthen the capabilities for adaptation (UNEP 2007).
How all these changes will impact food production and food security is very
difficult to say. The human response in the form of innovations and lifestyle changes
will be varied and make conditional forecasts even more uncertain. A natural next
step is to investigate in more depth the social-economic and cultural processes: the
facts, mechanisms and models of land use and land cover changes. But first, I present
some stories from the real world in order to ensure appreciation of the human factor
and illustrate both the locality and the universality of the quest for a sustainable food
provision.
11.3 Stories from the Real World
11.3.1 Nomads in Mongolia11

In April 1998, the U.S. Embassy in Beijing reported on the tension between the
nomadic herders in Inner Mongolia and nearby farming people, who are invading
in large numbers and cause environmental havoc. It shows how closely natural
and human factors are intertwined. ‘Two hundred thousand impoverished farmers,
mostly Hui minority people from the Ningxia Hui Autonomous Region, devastate
the grasslands of Inner Mongolia each summer. They go there to harvest facai grass.
Land per farmer in Ningxia fell by one-third during the 1980s. . . . [and] Mongol
minority herders are no match for gangs of 20 or more Hui nationality farmers,
sometimes armed, who often poison wells to drive the herders away. (Chinese)
desertification experts say facai harvesting is the chief cause of desertification in
Inner Mongolia. A 1997 expert report . . . recommends a ban on facai harvesting and
commerce combined with poverty alleviation programs for the 200.000 poor farmers
involved. Human activities, rather than long range climate change, appear to be the
main causes of desertification in Inner Mongolia. Just as in other parts of China,
the map of severe poverty largely coincides with that of rapidly growing minority
populations, desertification and environmental devastation’.
11 The quotes in this story are from www.usembassy-china.org.cn/sandt/desmngca.htm and www.

reuters.com/article/idUST3183820071008.
According to a 2007 report from China, the large grassland steppes of Inner
Mongolia are turning into sand deserts, ‘The wild grass reached up to my knees in
the past,’ according to a 40-year-old herdsman. ‘But there’s very little grass now.
It hasn’t rained here in six years and we have to buy fertilizers and feed for our
livestock. We never needed these before.’ Desert area in China is estimated to have
increased from 17.6 percent of total land area in 1994 to about 27.5 percent in 2006.
According to the report, many homes in Inner Mongolia and other western provinces
have been swallowed up by sand, dumping sand in springtime, not only on Beijing
but also sending dust particles as far away as Korea, Japan and even the United
States. It is causing respiratory problems, especially for children and the elderly.
‘Eye infections are getting more serious and common because of the sandstorms’,
according to the chief of the Xilinhot City Peoples’ Hospital in Inner Mongolia.
A ‘Green Great Wall’ of 700 km barrier of shrubs and trees has slowed down the
desertification, but has not stopped it completely.
11.3.2 Can and Should Rural France Be Saved?12

‘When you understand that the take-it-or-leave-it prices now being offered [by
wholesalers] mean that you’ll pay more to produce crops than you’ll get back in
proceeds, you’re left with the choice of either becoming a slave to this impossible
system or find a niche to begin other activities.’ This statement is from a French
farmer who took over her parents’ farm and diversified from crops to horseback
excursions. She expresses a general trend in French agriculture. Since 1960, the
number of farms in France decreased from 2 million to 0.66 million in 2010. They
employ less than 3 percent of the workforce. A further decline to 0.32 million by
2020 is expected. A fourfold increase in labour productivity in the last 40 years has
transformed many farms into mega-operations run with an industrial mindset. Prices
paid for crops have fallen 60 percent in this period – and still the gains in efficiency
are insufficient to get a decent income because the wholesalers who supply food
processing companies have gotten access to even cheaper alternatives, such as wheat
from Ukraine and strawberries from Morocco.
It is a downward spiral of competition and globalisation, which, in combination
with elimination and a ‘green’ re-orientation of European Union subsidies, forces
farmers to look for alternatives. One avenue is bio-agriculture in which quality, fla-
vour and local tradition is emphasised. Another option is agritourism, which benefits
from the infatuation of urban French citizens with the countryside. A third option,
widely used also in other parts of Europe, is to let the farmer play a key role in land-
scape and nature management in its various forms. It appears improbable, however,
that these trends will make major inroads in mainstream European agriculture.
Can the farmer be saved? In the summer of 2010, a group of farmers persuaded a
fastfood chain to give them permission to sell their fruit and vegetables in the parking
place. There was widespread media attention: ‘The raison d’ être of farmers is to feed
people and manage the land, not the market . . . we work with a loss and consumers
cannot pay our products . . . why do I get 17 cents for a kilo of apples and does the
12 The first two lines are based on an article by Crumley, How to Save Rural France? in TIME August
2, 2010.
consumer pay 1,70 to 3 Euro’? Should the farmer be saved? Consumers benefit
from lower prices and global competition forces farmers to be more competitive
or leave it to – often poor – farmers elsewhere. From an equity and environmental
perspective, that is not necessarily worse than subsidising European, or American,
farmers. Inevitably, worldviews and painful trade-offs enter the evaluation.
11.4 Land Use and Cover Change
11.4.1 Land Use Changes and Its Causes

Land use and the associated changes in land cover or land use and cover change
(LUCC) are an important cause and consequence of global environmental changes
(§4.5). This is particularly true for the low-income regions in the world where the
already large population density is expected to increase even more, and for the
regions that are industrialising rapidly under the opportunities offered by global
markets. How and why did and does land cover and land use change happen?
Despite advances in data availability and modelling, understanding the causal mech-
anisms behind LUCC is fragmented. The complexity of meteorology and ecology is
compounded with the human dimension.
Evidently, the quest for food, fodder and fiber is at the root of LUCC as well
as the – often related – construction of dams and roads and mining and other
activities have played and do play an important role. The major process is the
still ongoing transition from local and small-scale subsistence farming to large-scale
industrial farming for an increasingly global market. Ongoing deforestation is part
of this process (Figure 4.4). In the last three centuries, massive conversion of for-
ested/woodland area and savanna/steppe/grassland area into cropland (from 300 to
400 and from 1,500 to 1,800 Mha) and pastureland (from 400 to 500 and from 3,200
to 3,300 Mha) has taken place (Lambin et al. 2003; Figure 4.4). In the 18th century,
conversion accelerated in Europe, the northern part of the Indian subcontinent and
in China. This was followed in the 19th century by rapid cropland expansion in
the newly developed regions of North America and the former Soviet Union. The
greatest cropland expansion in the 20th century took place in South America and
Southeast Asia. Much of the remaining forest in these regions and in Africa is under
pressure of deforestation, despite impressive forest regrowth and afforestation rates
in some parts of the world. On the other hand, tens of millions of hectares have been
taken out of production in United States and Western Europe in recent decades as
a consequence of rationalisation in agriculture.
Following the concept of transitions applied in demography (§10.3), the notion of
a land use/cover or LUCC transition is used to describe the manifold and multicausal
processes involved. As explained before, transitions are not deterministic trajectories
but rather co-evolutionary and reversible change processes with a variety of social,
economic and ecological feedbacks. A recent inventory of changes during the last
few decades within this framework indicates that (Lambin and Geist 2006):
r different parts of the world are in different phases of LUCC transition – for
instance, cropland is decreasing in the temperate and increasing in the tropical
zones;
r large-scale deforestation takes place in the Amazon Basin, in Southeast Asia

(rapid cropland expansion) and in boreal Eurasia (logging and fire); and
r there are still large gaps, inconsistencies and uncertainties in the existing data-
bases on LUCC (extent and rate of change).
The mechanisms behind LUCC are complex and multicausal. Often, population
pressure and poverty dominate the debate but one should also consider people’s
responses to economic opportunities within the local, national and global institu-
tional frames. Proximate (or direct) causes have to be distinguished from underlying
(or indirect or root) causes. In a survey of a series of case studies, it was found that
five fundamental underlying causes drive LUCC (Lambin et al. 2003):
r resource scarcity leading to an increase in the pressure of production on
resources;
r changing opportunities created by markets;
r outside policy intervention;
r loss of adaptive capacity and increased vulnerability; and
r changes in social organisation, in resource access, and in attitudes.
The most important determinant of what actually happens is probably the behav-
ioural responses of people to the perceived opportunities and constraints of markets
and policies. Increasingly, this is influenced by forces that operate at regional and
global scale and amplify or attenuate local factors.
In the context of worldviews, one recognises in LUCC transitions the tension
between the local biogeography and the associated customs – the small world – on
the one hand, and modern science and national/global government and business
interests – the big world – on the other (§6.3). This multi-scale character is often a
challenge in the search for sustainability. For instance, Raquez and Lambin (2006)
conclude from meta-analysis of forty-six case studies on factors behind sustainable
land use that ‘in general, governments seem to be reluctant to take into account
the diversity of environmental contexts to which local communities have adapted to
properly manage their natural resources. They tend to dictate land use management
strategies and deal with environmental problems in a top down manner’.
11.4.2 Modelling Land Use and Cover Change

LUCC processes are often conceptualised in the form of a land use change or transi-
tion matrix in order to simulate changes amongst land categories over time (Geist
2006). The basis is the collection, storage and analysis of data in a geographical
information system (GIS). Land conversion processes are also simulated at a rather
aggregate level in ‘top-down’ global change integrated assessment models (GC-
IAMs). They are the tools par excellence to study how the local biophysical and
socioeconomic changes have wider consequences through complex feedback link-
ages. In some of these models, the emphasis is on the biophysical processes, as in the
Integrated Model to Assess the Global Environment (IMAGE) model (Bouwman
et al. 2006) and the Global AgroEcological Zones (GAEZ) model (Fischer et al.
2001). Other models focus on aspects of the regional agro-food systems and the
agro-economic dynamics and links (IFPRI 2008).
Box 11.2. Modelling ecosystem dynamics and services: rangeland overgrazing

in Senegal. One of the challenges in (resource-)economic models is to incor-
porate the use of ecosystem services in a broader, complex dynamical context.
An example is the model by Hein et al. (2005) on rangeland management in
northern Senegal. It is an area where livestock grazing is the main source of
income. Sustained, heavy grazing pressures have an impact on the vegetation
which depends to a large extent on the annual rainfall. The data are available to
test different models. First, a simple model was constructed to assess the optimal
grazing pressure in animals/ha. Basic assumptions are a logistic growth process
of livestock growth, no feedback from grazing upon the pasture forage produc-
tion and hence upon livestock carrying capacity, given annual rainfall and fixed
prices. In subsequent steps, this benchmark model was refined with stochastic
variation in rainfall, a negative feedback of grazing upon rain-use efficiency, and
price fluctuations in a situation of drought.
The inclusion of the loss of resilience for drought because of high grazing pres-
sure indicates that an optimal long-term stocking rate should be 10 to 20 percent
below the benchmark model value. The current stocking rates are in the order of
50 percent above the indicated optima. There are clear warnings that sudden eco-
system change might happen, with catastrophic results for the inhabitants (§9.7).
An important task is to implement ecosystem dynamics and services in simula-
tion models of long-term climate-induced changes in the (semi-)arid regions of
the world. Only then can these stories and these models prepare for such events
and their consequences. Increasingly, agent-based models (ABM) are developed
to investigate situations like these and suggest strategies for sustainable manage-
ment (§10.5).
However, the uniqueness of local situations make it necessary to analyse LUCC

processes in a ‘bottom-up’ way, not only to perform local assessments but also to
identify and validate universal mechanisms used in GC-IAMs. It requires a rather
explicit and accurate image of ‘agricultural man’ (Verburg et al. 2003). At the lowest
scale is the unit of production, usually the family farm with its family structure, labour
division, wages and values, against the background of environmental variables such
as microclimate, soil characteristics and seasonal fluctuations. The actual processes:
clearing, burning, sowing, harvesting, soil erosion and so on, take place here. One
scale level higher, the landscape sets the social-economic context: How is the com-
munity organised; how are property rights organised and secured; how is access to
technology; and so on. Its biophysical counterparts are the landscape features like
altitude, topography and soil and water basin characteristics. Irrigation schemes and
erosion and sedimentation processes take place at this level. The region provides the
social, economic and cultural as well as biophysical embedding of farmers and com-
munities. The relevant macro-processes are colonisation and migration, large-scale
infrastructural developments, world market food prices and trade opportunities and
the like. Usually, assessments of food production are also made at this level on the
basis of climate parameters (temperature and rainfall patterns) and vegetation- and
soil-based crop potential (§11.2).
1972 2002 2022
Park boundary N
Buffer zone
Agriculture
Open forest
Closed forest
0 1 2 3 4 Kilometers
Figure 11.9. Generalised land use in San Mariano municipality in the northern Philippines:
1972 and 2002 based on interpretation of respectively aerial photographs and SPOT images;
2022 simulated by the spatial landscape-level model (Verburg et al. 2006). (See color plate.)
An example of a simulation model focussing on the local and regional scale is

the CLUE (Conversion of Land Use and its Effects) modelling framework (Verburg
et al. 2006). It uses a Cellular Automata (CA) approach and simulates land use
transitions driven by human and biophysical factors. For each land type, partly
overlapping suitability classes are assigned to the cells in a grid. These suitabilities
may change over time as the result of feedback mechanisms such as land degradation.
Regionally available food from the cells is calculated by incorporating the effects of
past land use, pests or diseases, annual fluctuations with existing technology levels,
economic values and two years of food/money reserves. In a second step, the regional
land-use demand is calculated from population and variables such as management
level, urban expansion rate, yield level and land use strategy (both a function of
[inter]national technology level). People’s values and preferences are considered,
too, in particular concerning cattle and natural vegetation. In a third step, a land use
allocation model translates regional demand into actual land use changes. In this
way, the model can be used to explore agricultural transitions, taking into account
food security and technological adaptation. The CLUE-model has been applied in
regions in China, the Philippines and Central and South America, where traditional
farming methods are gradually replaced or complemented by cash crops and other
forms of commercial farming and where pressure on the forest and soils leads to land
degradation and loss of biodiversity (Verburg et al. 2003; Wassenaar et al. 2007). It
is also used in the EUruralis project on the future of rural Europe (Westhoek et al.
2006; Verburg et al. 2010).
One illustrative application of the CLUE model at the local scale focuses on a
major biodiversity hotspot in the world: the Philippines, and in particular on the San
Mariano municipality in the north (Verburg et al. 2006). It is an area of 48,500 ha with
twenty villages and bordering one of the largest protected areas in the Philippines,
the Northern Sierra Madre National Park. The still growing population largely are
or descend from migrants. Logging companies deforested a large part of the area
between 1960 and 1990. Since the logging moratorium in 1989–1992, employment
declined and people switched to agriculture. A landscape-level analysis zoomed in on
an area at the border of the National Park. Aerial photographs and remote sensing
are used to present the changes between 1972 and 2002 (Figure 11.9a,b). It shows
the impact of large-scale logging, with road construction and farmer immigration –
a process that happened in many places in the world in the past centuries at an
accelerating pace. Using the most recent land use maps (1990), the CLUE model
was applied to forecast land use changes in the Philippines for the period until 2022
with an imposed deforestation rate based on trend extrapolation and for different
policies and scenarios (Figure 11.9c). Given the need for more agricultural land
and the resulting deforestation pressure in the Philippines, deforestation is expected
to continue at a rate of one-third of the rate in the 1972–2002 period. To design
and implement adequate policies to prevent ongoing illegal logging and conversion
to agricultural land, the models need to address the different scale levels of local,
regional and global forces in an integrated fashion.
A rapidly growing number of model-based analyses like these are made nowa-
days. They can help to understand the seemingly relentless drive behind deforesta-
tion at the different scale levels and help to formulate and implement feasible pre-
servation strategies. The simulations of biophysical potential and socio-geographic
change form a useful background to explore the prospects for ‘sustainable food
for all’ and for the effort to eradicate extreme poverty and hunger – Millennium
Development (MDG) Goal 1.
MDG Goal 1: Eradicate extreme poverty and hunger.

TARGET
Halve, between 1990 and 2015, the proportion of people whose income is less than
$1 a day.
According to the United Nations Millennium Development Goals Report (2010),
‘The number of people who are undernourished has continued to grow, while
slow progress in reducing the prevalence of hunger stalled – or even reversed
itself – in some regions between 2000–2002 and 2005–2007. About one in four
children under the age of five are underweight, mainly because of lack of food
and quality food, inadequate water, sanitation and health services, and poor care
and feeding practices.’
11.5 Towards a Global Industrial Agro-Food System
11.5.1 Diversity in Transition

Although there are undesirable trends such as overexploitation and pollution of
soils and water, the general view is that modern farmers have overcome ancient
environmental constraints. Expansion into new agricultural lands has almost come
to a halt. Intensification has led to a steady rise in yields and a declining trend in
food prices, on an almost constant cultivated area. The quantity and diversity of food
products offered on the local and regional market has increased. There are different
judgments about the quality. There is a shift in diets towards high-value food, but
junk food, obesity and diabetes are also part of the story. Improved logistics give a
better match between demand and supply, which reduces the vulnerability of human
Table 11.2. Agro-food systems: key indicators for some countries/regions in the world (2009)
Unit USA Japan EUa Korea, Rep S. Arabia RussianFed
Population (2009) mln 307 128 499 49 25 142

Income (2009) PPP US$/cap 46.437 32.433 30.543 27.195 23.421 18.938
HDI (2010) 0.902 0.884 0.854± 0.921 0.752 0.719
0.080
Ratio food import $/$ 0.7 20.7 1.2 6.4 9.3 4
and exports
(2008)
Cereals harvested ha/cap 0.18 002 0.11 0.02 0.02 0.3
area (2009)
Cereals production kg/cap/yr 1335 90 820 150 95 675
(2009)
Fertiliser use (2008) kg/ha/yr 290 620 295 730 555 45
Capital stock (2003) 1995US 10.8 35.7 10.3 15.7 29.6 2.7
k$/cap
Unit Mexico Brazil China India Vietnam Nigeria
Population (2009) Mln 107 194 1332 1155 87 155

Income (2009) PPP US$/cap 14.341 10.429 6.838 3.275 2.956 2.149
HDI (2010) 0.75 0.699 0.663 0.519 0.572 0.423
Ratio food import $/$ 1.5 0.15 2.2 0.55 0.7 4
and exports
(2008)
Cereals harvested ha/cap 0.09 0.1 0.07 0.08 0.1 0.12
area (2009)
Cereals production kg/cap/yr 290 370 360 205 490 195
(2009)
Fertiliser use (2008) kg/ha/yr 105 485 585 240 210 25
Capital stock (2003) 1995US 7.7 9.0 7.4 2.9 4.3 1.5
k$/cap
aPopulation weighted average of France, Germany and United Kingdom.
Source: FAO; for HDI: en.wikipedia.org/wiki/List_of_countries_by_Human_Development_Index&num;Very_
high_human_development_.28developed_countries.29.
populations to famine. Are we witnessing an irreversible change towards a global

agro-food system that fulfills the promise of eradicating hunger, even in remote
pockets? It is difficult to say, partly because there are still such large differences
between and within the usual unit of measurement: the country (Table 11.2). The
countries that produce food can roughly be divided into three groups: subsistence
biomass-based, agro-industrial and transition.
One-third to one-half of the world population still lives in a biomass economy,
with natural energy flows for their sustenance. They are located on the periphery
of the world system, across all continents but especially in Asia, Africa and Latin
America (Figure 11.3). Their livelihood is from subsistence farming and herding.
They are often poor in both income and wealth and are faced with increasing land,
water and energy constraints under the pressure of population growth, resource
depletion and globalisation in all its appearances. Most have low levels of education
and health and lack access to capital. Many still adhere to a mythical worldview,
with gods and rituals playing a direct role in their affairs. Often, they live in a
society with a rigid social hierarchy, dominance of male values and enforcement of
dogmatic beliefs. For them, survival values are still dominant. Food is more than
just calories: Its production and consumption is rooted in ancient traditions and
has profound social and cultural meaning. Their knowledge of local ecosystems is a
source of resilience. Some typical characteristics of these social-ecological systems
(SES) are:
r regional biogeography is a key defining factor, that leads to inherent diversity,
locality and plurality of human cultures and institutions;
r part of their activities and transactions is not monetised; therefore, it is in-house,
subsistence farming/herding and in natura exchange;
r livelihood is sustained on the basis of local resources which are exploited as com-
mon pool resources (CPR) with a mix of competitive and cooperative arrange-
ments;
r social stratification is often strong and the resource-based surplus is exploited
by local landlords or national government and industrial elites (cash crop or
mining);
r people are vulnerable for natural and human-made perturbations because of
limited resource and adaptation options.
Sustainable development for them is about the risks and challenges posed by a
combination of overpopulation and local resource overexploitation. Trade and out-
migration bring relief and opportunity, as well as dependence and new risks.
On the other extreme are the farmers in the advanced industrial economies in
the centre of the world-system. They are a mere 3 to 4 percent of the world’s agricul-
tural population, but their products represented more than two-thirds of the market
value of world agricultural output in 1995–1997 (Figure 11.3; WRI 2000–2001).13
Their farming activities have gradually been industrialised to the extent that their
relationship with natural ecosystems has largely vanished. Their business is high-
input industrial agriculture, as part of the increasingly global agro-food chain of
food producers, processors, traders and retailers. Their farming is large-scale grow-
ing of staple crops such as wheat and maize, production of high value-added fruits and
vegetables, and large-scale animal husbandry for dairy and meat products. Competi-
tion forces them to operate at ever larger scale and to aim for ever higher yields with
the latest biotechnological techniques. They also face risks: those associated with fin-
ancing the large capital requirements and trading on global markets (Roberts 2008).
Increasing globalisation tends to erode their income security. Postindustrial concern
about environment, ecological restoration and animal welfare confront them with
new and additional constraints and risks.
In between these two extremes is a large and growing number of farming families
who try to adapt to the rapid regional and global changes they are faced with. They
are in the midst of the nutrition and land use transition. In some regions, the Green
Revolution offered the opportunity for large-scale wheat farming on irrigated lands,
13 WRI (2000) gives an estimate of 80 percent for 1995–1997. It has declined since then, but reliable
recent data have not been found. The large share is partly because part of the food production does
not enter into the formal market accounts.
Box 11.3. Entering the cash economy. ‘Chotu Ram knew that “development”
was not necessarily to the advantage of everyone . . . Chotu Ram was a Kumhar,
or potter, by caste, but nowadays the farmers were cash-conscious and unwilling
to spare any of their land for his caste to dig clay. What’s more, clay cups and pots
were today considered inferior – stainless steel, plastic and, among the wealthier,
china had taken over. When the service of potters had been as essential to [the
village] Thakurdwara as that of barbers or washermen, Chotu Ram’s family had
been reasonably well looked after. They had received grain and other gifts that at
least provided their basic requirements. Now Chotu Ram had become a modern
man, he’d entered the cash economy, but the wages he received barely allowed
him to buy anything beyond the grain his family ate, and they certainly did not
cover the emergencies that every family in the village had to face from time to
time’ (Tully The Heart of India (1995) 22–23).
as in northern India. Others switched to high value-added fruits and vegetables

for the local market, with rising incomes and demand, as in China. Cooperatives
have been founded to market the milk from the small herds and satisfy the growing
demand for dairy products, as in villages of rural India. Many try to survive on a
mixture of subsistence farming, cash crop production and migrant work in urban
centres, as in Kenya and Nigeria. Many also become labourers on large-scale plant-
ations owned by foreign investors and producing cash crops for the global market,
as in Indonesia and Malaysia (palm oil), Ecuador (bananas) and Vietnam (coffee).
Rising food prices and prospects for biofuel have stimulated the interest of speculat-
ive foreign investors to buy land from these people – million of hectares are bought
in notably African countries with weak legal and government institutions (‘land
grabbing’).14 This may intensify if governments of land- and water-scarce countries
face an insecure food supply and enter the global food market for imports. Sustain-
able development for these people means job and income opportunities in order to
share in the basic amenities of modern life. In the process, many will end up in the
(mega)cities.
11.5.2 The Global Agro-Food System

Trade can diversify the food supply and thus reduce the vulnerability to large-scale
famine – as does conquest (§3.3). Both are an important ingredient of food security
since the earliest civilisations. In economic terms, trade was driven largely by a
comparative cost advantage: Some places are more suitable than others for food
production. The appearance of novel products (potatoes, sugar, tropical spices and
fruits), the use of military force by the colonial powers and the advances in the
agrosciences were also part of the story. In the second half of the 20th century, world
14 With economic, not military means, some Asian countries are repeating the history of colonialism.
Up to 50 percent of agricultural land in Madagascar is said to be sold to foreigners from Korea and
also in other African countries farmers are apparently selling their land to foreigners from China,
India and Saudi Arabia.
a)
1000
Maize
Rice (milled equivalent)
800
Wheat
mln ton/yr
600
400
200
0
1961 1971 1981 1991 2001
Year
b)
300
Chicken meat
250 Pig meat
Cattle meat
200
mln ton/yr
150
100
50
0
1961 1971 1981 1991 2001
Year
Figure 11.10a,b. World output of three dominant cereals (maize, wheat and rice) and of the
most important sources of animal protein, 1960–2008 (source of data: faostat.fao.org).
food production started to increase rapidly and so did food trade. Annual output
of the dominant cereals – or grains, notably wheat, rice and maize – almost tripled
since 1960 (Figure 11.10a). Besides, there are other staple food crops (potato, millet
or sorghum) and vegetables and (sub)tropical plantation produce (sugar, rubber,
coconut, palm oil, cacao, coffee, tea and many other crops). Extensive grazing and
dairy farming – notably for milk, butter and cheese – became large-scale and annual
production of meat from cattle, pigs and chickens increased more than fourfold in
the last five decades (Figure 11.10b).
Food production is the upstream source part of the global food chains. It rep-
resents an economic output in the order of 5 percent of GWP.15 But the larger part
15 The estimate of 1,300 billion US$/yr is for the years 1995–1997 and based on WRI 2000. More recent
data are difficult to find.
of value-added is in the downstream part of the chains. Total monetary transactions

in growing, processing, trading and retailing the ever larger array of food products
amounted in 2007 to an estimated US$3,500 billion or more than 8 percent of GWP.
About 80 percent had to do with food and beverages manufacturing and grocery
retail (www.etcgroup.org). These activities become ever more concentrated in the
hands of a dozen globally operating companies, who account for more than one-
third of the total world transactions. Food processor Nestlé and supermarket chain
Walmart are the respective leaders. The remaining 20 percent consist of the manu-
facturing of seeds, agrochemicals, pharmaceuticals, fertiliser production and biotech
activities. These activities are even more concentrated in the hands of a few multi-
national corporations: The top ten producers have 65 to 90 percent of the market,
with Syngenta, Monsanto and Bayer as the largest ones in 2007 (www.etcgroup.org).
A relentless competitive struggle for market share and expansion is the driving
force behind this concentration and the production and control of scientific and
technological knowledge. It is one of the forces behind and outcomes of global
capitalism.
Globalisation is a mixed blessing. The growth of financial institutions and the
movement towards open trade increasingly forces farmers worldwide to compete
with large-scale intensive farming and low labour-cost areas. Globally operating
agro-food producers and traders exploit differences in labour costs. In combination
with storage strategies and speculation, the increasingly connected system exhibits
food price instabilities. The clearest sign of failure is hunger. There are still more
than one billion undernourished people in 2010. There is enough food for them,
and probably also for the 8 to 10 billion people expected to live on the planet by
2050. But those who need it cannot afford it or have no access. Relative abundance
and improved logistics have made food aid a common practise. But it has also
encouraged populist policies to appease the urban poor – the panem et circenses in
ancient Rome. It is even argued that food aid aggravates the very problems it is
supposed to solve, because they make urban populations dependent on food from
elsewhere and discourage local farmers to make the necessary investments (Moyo
2009). Forward projections for some countries read like a Malthusian doomsday:
Investments in local farming fall short, traditional employment declines, fertility
rates remain high because education of women stalls, rural areas get depopulated
and city slums swell to ever larger proportions. The activities of NGOs for human
rights, poverty relief and environmental standards gain momentum, but still only
slowly redirect some of the basic trends.
In view of the growth in human populations and the trend towards more meat,
food demand is expected to increase. On the supply side is still considerable room
for expansion (§11.2). There is the prospect of innovations at an unprecedented scale
and in various directions, from high-tech GMOs to organic farming, from reducing
food chain losses to dietary changes and from ever more industrialised farming to
community-based local schemes. But these bring new risks and challenges. Can a
further increase in fertiliser and water use and in pesticides and herbicides use be part
of sustainable development? Is dependence on fossil fuels and control of the food
chain by a small group of powerful corporations compatible with sustainability? The
long-term future of agro-food systems are hard to imagine let alone predict. But it is
a near-certainty that a decline in soil fertility, water availability and waste absorption
Figure 11.11. Causal loop diagram of feedback processes in agro-food systems (partly based
on Sustainability Institute 2003).
capacity and an ongoing economic, social and cultural globalisation will make the
world agro-food system more complex and more vulnerable.
11.5.3 Markets, Scale and Innovations As Driving Forces

Many different mechanisms and forces are at work in the global agro-food system.
Some generic insights can be formalised in a system dynamics model in order to get a
better grasp of the options for a sustainable path.16 What are the mechanisms behind
the trends? One can distinguish several loops (Figure 11.11). The first one is the pro-
duction loop. It operates in every capitalist market system: The farmer-entrepreneur
makes an investment in food producing capital and, with other inputs like labour
and land, the production capacity generates output and revenues/profits. Part of the
profits are reinvested in order to further increase producing capital. It is a reinforcing
growth loop as long as profits are sufficiently high, which depends amongst others
on demand. Otherwise, capital owners will look for other opportunities.
Second, a balancing price competition loop is operating. Providing there is open
access to the market, competition amongst existing and new producers drives down
the average price. This, in turn, tends to increase demand until the profit margin
has declined again to the extent that farmer-entrepreneurs and capital owners put
their effort and money somewhere else. Producers can respond to stagnating sales
and profits with an attempt to find new markets. A subsidiary loop is marketing/
upgrading in which demand is driven up by processing and advertising. A more
effective and important way to boost profits is to bring down the cost of production
by increasing the scale of production and by technical innovations. Thus, a third loop
16 This analysis is inspired by the report Commodity Systems Challenges by the Sustainability Institute
(2003). The starting point was the observation that output and prices of agricultural products between
1950 and 2000 in the U.S. food system all show an upward trend in output and a downward trend
in price. The search for generic patterns is also the essence of the syndrome/archetype approach
(§4.5). See also Roberts (2008).
of upscaling/innovation (‘rationalisation’) leads to increasing productivity of labour

and capital and declining product cost. The prospect of rising profits will provide an
incentive to invest in new capacity, which in turn reinforces production growth.
These three feedback processes have given a boost to productivity in food provi-
sion and greatly benefited populations in the United States and Europe. Government
regulation for the sake of food security and consumer protection played an important
role, although there were different views on and forms of agricultural policies. But
in many places elsewhere, weak property rights and no or weak rule of law permit
parasitical politicians and vested business interests to reap short-term profits and
impede productivity growth. Market incentives that could respond to the pressures
of rising fertiliser cost and large losses during transport are blocked. These are causes
of hunger.
There are two more loops in a globalising and postindustrial world. First, des-
pite scale and innovations, the product may still be more expensive than equiva-
lent imports. In a world committed to open trade, producers are confronted with
competitive imports. This import loop adds to the capacity to deliver and puts a
downward pressure on price and an upward pressure on demand. Protective meas-
ures for reasons of food security and employment were and are a natural response
of governments. Second, with rising income people become more demanding with
respect to the environmental and social aspects of food quality and production.
They also become more aware of environmental, health and social impacts of the
industrialising and globalising agro-food system. It manifests itself as a regulation
loop: More stringent health, safety, environmental and animal welfare standards
tend to increase costs. This trend towards postindustrial values contributes to what
is perceived as a good quality of life, but puts also new strains on the system.
The United States and European agro-business model has with the fall of com-
munism become the dominant vision for the future of agriculture. Many consumers
will perceive it as a virtuous cycle of ever more food available at ever lower prices.
From the perspective of a farmer, it can also be experienced as a vicious cycle of
income erosion and increasing dependence. In an environmental resource context,
it poses serious long-term dangers. The system grows in response to increasing con-
sumer demand and shifting preferences, driven by increase in scale and competitive
innovations (Figure 11.11). Against such a larger system background, a number of
mechanisms and interventions determine its path:
r food surplus and trade: the scale/innovation loop may be so successful that a
food surplus results. Well-known examples are the European Union ‘milk and
butter mountain’ in the 1990s. The income ‘race to the bottom’ of United States
and European farmers in their relentless search for the cheapest supplies is
one consequence. Food exports in poor countries, promoted by governments
and business against the interests of their own population, is one implication of
global trade;
r food shortages and food security: like any commodity system, the agro-food
system experiences fluctuations in outputs and prices. Rapidly rising demand and
globalisation intensify this phenomenon. It is reinforced by short-term responses
such as speculation, social unrest and government protectionism and aggravated
by expansion of biofuel plantations, possibly climate change–related harvest
Box 11.4. Global food. ‘Paprika is the latest recruit to a revolution in Peruvian
farming . . . the country has added almost 400 different export crops to its tra-
ditional staples of coffee, cotton and sugar . . . Farm exports totalled $1,3 billion
in 2004, up a third on the previous year.’ (The Economist, September 7, 2005).
Ukrainian soils are the most fertile in the world, and Ukrainian wheat exports
has risen tenfold between 1995 and 2010, but the farmers do not benefit. Instead,
the introduction of export permits brought great fortunes to a small elite of
entrepreneurs and bureaucrats. Because banks ask 20 percent interest to small
farmers, international capital funds invest in Ukranian agriculture. Rationalisa-
tion and upscaling increase yields (NRC January 4, 2011). The harvest was good in
southern Africa in 2008 – but people cannot afford to buy the food. They spend
more than half of their income on food. The FAO estimates that low-income
food-deficit countries will import 2 percent less cereals but pay 35 percent more.
In the world food system, events unfold as a series of connected causes and
consequences too complex for a simple single story. Rich countries buy Chinese
goods while building up huge debts. The money is used by a Chinese middle
class to eat more meat. Chinese corn and wheat production fall short of local
needs. Simultaneously, the U.S. government decides to reduce oil dependence
by subsidising ethanol from corn. Exports start to fall and cause a 50 percent rise
in the corn price in Mexico, where it is a staple food. Worldwide droughts and
floods cause harvests to be less than normal in a number of countries. Argentina
and Brazil respond with rapid and large-scale expansion of soybean production
for export, partly on deforested lands. Although less than 10 percent of food is
traded internationally, markets do work and the additional demand causes food
prices to rise. The Russian government responds with a ban on exports, causing
food riots in Egypt with political consequences. And so on . . .
failures and natural catastrophes and famines. In the background are the ‘slow’
dynamics of resource degradation and environmental change;
r externalising environmental and social consequences: the push to bring produc-
tion costs down causes serious environmental damage and social destruction,
that is passed on to poor populations and future generations. This process, too,
has been intensified by globalisation. Competition has not brought innovations
throughout the food chain but also caused havoc to many poor farmers and
their communities (income disparities, debts, starvation and loss of local skills
and resilience);
r environmental and social regulation: increasing awareness and demands of
consumers in rich countries force governments to introduce more stringent
standards and regulations, with diverse impacts throughout the food chain. Food
chain lobbies tend to resist, arguing that it makes them less competitive vis-à-vis
foreign producers who are not faced with such standards. Governments of low-
income, food exporting countries consider such measures a form of protection-
ism and cannot or do not comply. On this not-so-level playing field, an increasing
number of trade disputes take place with the World Trade Organization (WTO),
governments, corporations and a multitude of NGOs as participants.
Although this analysis pertains primarily to important, globally traded crops such as
wheat, rice, maize and soybean, similar forces are at work in other parts of the agro-
food system. High-value commodities such as cotton, palm oil and tea, which are
cultivated on large-scale plantations, and fresh fruit and vegetables are increasingly
embedded in global corporate and trading networks. The Big Food and Big Retail
business conglomerates and sectors and the allied government departments operate
in a sense as effective top predators.
11.6 Perspectives on Food and Agriculture

There are two long-term ‘objective’ constraints in the global agro-food system:
Humans need a minimum amount of food for physical survival, and the food-
providing resource base is finite in size and in capacity to supply and adapt. Within
these bounds, the directions and options for a sustainable future are shaped largely
from the status quo framework of forces and powers. But quite different pathways
are still possible and the trajectory is a continuous balancing act between utopian
expectations and dystopian tendencies. Some pathways are more desirable, feasible
or possible than others. Some are more sustainable than others in terms of adequacy,
affordability, security and safety for the majority of present and future people and
for other species on the planet.
The system has for some decades evolved according to a neoliberal ‘free’ market–
oriented worldview. In theory, this leads to improvements with the side effects
remaining manageable through standards, labels and so on, under the guidance of the
United Nations and enlightened and proactive corporations. In practise, it is in quite
some places a postmodern nightmare in which private gain, speculation and exploit-
ation dominate people’s lives and in which prospects for local small-scale farming
erode because of lack of investment and infrastructure. Are major disruptions inev-
itable when food prices go up and become more volatile? Is more technology the
solution, and, if so, what kind of technology? Will resource overexploitation outstrip
yield improvements? Can small-scale, ecological and community-oriented forms of
agro-forestry and animal husbandry contribute to food security and quality? Is a new
solidarity and ethic needed to bring an end to hunger? The last section of this chapter
examines some prospects for the future with explicit use of divergent worldviews.
Many models are around that make conditional forecasts of future food and land
developments. Some trends are unmistakable. The human population will grow and
so will its needs and desires for food. Intensification and upscaling are inevitable.
But how, how fast and with what consequences is unclear. Many of the seeds of
change are hidden in small and fragmented events. Look at the following items in
the newspapers:
r The ‘light-green’ customer is discovered by supermarkets. Their concerns are

personal health, low prices and supermarket convenience, and they are less
‘fundamentalist’ than the original organic food market client. But Walmart,
where one out of every ten shopping dollars in the United States is spent, has
announced to compete more intensely on prices. And some nutrition specialists
claim that fast-food-plus-additives is the most healthy meal, despite evidence
that cheap food is bad for health.
11.6 Perspectives on Food and Agriculture 373
r ‘Without us food would be even more expensive’, says a commodity trader in a

2008 interview. Author of How To Make Big Money Investing In Metals, Food
And Energy (1998), he has great expectations of the second-generation biofuels,
notably jatropha curcas. It grows in semi-arid regions, does not need much water
and its nuts have to be harvested manually and are used to make biofuel. ‘The
commercial world jumps on it . . . the financial world is still asleep’. But according
to UN specialists, food trade is increasingly disturbed by investment banks and
pension funds, with profits as their sole aim. Their activities should be regulated
in order to prevent food crises.
r NGOs say that genetic technologies are pushed by an unholy alliance of science,
corporations and governments. A company spokesperson says that ‘they are
ready to be applied, but the fear of the label ‘genetically manipulated’ has to be
overcome’.
r Using satellite photos, Dutch farmers can do ‘precision agriculture’ and signifi-
cantly reduce fertiliser and pesticide inputs. ‘Comfort-class’ livestock buildings
are highly mechanised and animal-friendly. Farmers earn additional income
from recreational and educational services and renewable energy production.
r Eighty percent of Indian farmers has not more than 2 ha farmland. Yields are
rather low, inputs such as electricity and water are subsidised and rural employ-
ment is high. What is needed are investments in infrastructure and mechanisa-
tion – but such a long-term policy does not deliver votes.
r Some 800 farmers in the Ferghana valley in Kirgizistan are producing biocotton
with support of development NGOs. Certification helps them to get a higher
price. A Bio Farmers Public Union helps with access to the international market.
The goal is to double production by 2012.
r Pesticides that are forbidden nearly everywhere in the world are still used
on banana plantations in Nicaragua. Numerous children have distorted limbs,
psychic disturbances and heart defects.
r Coastal seas can provide endless amounts of algae and weeds, sources of
protein rich food. The first farms in nutrient-rich waters are established. It
reduces eutrophication and may also become a source of biofuels, according to
researchers.
r ‘The world is at its end’, laments an old herder. ‘We knew how to deal with
drought, but something in the world has changed’. The latest of a series of
droughts is about to give the death blow to the Samburu nomads in northern
Kenya. Like so many nomads, they will end up in the city slums.
r The Chinese government invests more than €100 million for equipment to create
artificial rain in some of its driest regions, in response to severe water problems.
A recent operation led to 17 percent more rain, it is claimed.
Each fragment reveals a particular perception and interpretation of the situation.

Each fragment also expresses hopes and fears about the ‘good life’. Each can be
assigned a position in the worldview space of Figure 6.5. And they are partly incon-
sistent, contradictory and incompatible, as is the world.
Table 11.3 is an attempt to express the different worldviews in terms of what
a person desires and expects and the policies he or she thinks are most effective in
reaching a world in which he or she would like to live. In the Modernist worldview,
Table 11.3. Worldviews and agro-food futures
Agro-Food Systems
Statement 1: Agricultural trade liberalisation should be promoted in order to improve food security and boost
incomes.
Correct. It is through We should be very Local food production is The advantages of food
food trade that cautious, because it to be preferred – more trade liberalisation are
famines in many increases our independence, less at least partly undone
low-income regions dependence in the environmental cost of by speculation and
have been prevented. critical domain of transport. If trade, food price volatility.
More food security for food. then Fair Trade!
all!
Statement 2: Sustainable organic farming is not only beneficial to us, and to animals and crops, but rather,
much more to our environment (www.theorganicfoods.net/).
This may be correct, but I am only interested in Correct. In a larger Correct. However, it is
these claims should be the health claims, and system perspective, still a luxury option for
as rigorously validated those are still organic food brings most people. Globally,
as for non-organic controversial. Besides, many benefits. it offers no solution
food organic food is still too Unfortunately, many because it would need
expensive. of these are not too much land.
recognised on
commercial markets.
Statement 3: The environmental movement has done much harm with its opposition to genetic engineering.
This is probably correct. Correct. It has delayed Unfair. The risks of There are strict
The fear and the necessary genetic engineering regulations regarding
resistance about innovations to keep are still unknown, and GMOs and the global
GMOs seems food cheap. It is also we should be very food situation can
unwarranted. The responsible for careful. It is mostly probably not without
risks are smaller than unnecessary hunger. pushed by greedy GMOs. Opposition
many other risks in corporations. may no longer be a
industrial society. responsible strategy.
Statement 4: Further intensification of agriculture is only possible with stringent environmental measures to
limit eutrophication.
Correct. Technically, it is I don’t know. There is Intensive agriculture is Essential for sustainable
already possible to already too much in many ways agriculture. In
reduce the use of regulation for farmers problematic. It is low-income regions,
inputs. Prices and – is it really a absolutely necessary the impacts are
innovations will solve problem? to reduce emissions – rapidly becoming
the problem. No and organic farming is serious. International
regulation needed. one of the solutions. standards should be
introduced.
Statement 5: The dramatic rise in private investment into agriculture has led to neglect of small-scale farming
and ‘land grabbing’.
Unfair. Upscaling and If the poor want to catch Wrong direction. This is This is an unintended
private investments up with the rich, this what the ‘free’ market consequence of
are the most effective seems one of the few does: economic globalisation. In the
way to increase ways that work. colonialism (slavery long term, it harms
productivity and and land grabbing) by local food security and
create employment in rich countries. should be regulated.
poor countries.
the problem is primarily seen as lack of efficiency and ingenuity. The solutions are
to increase scale, eliminate trade barriers, reduce losses, stimulate innovations and
ensure market-based incentives. Governments should not interfere and subsidies
should be abolished. In a more postmodern reality, corporations and governments
try to steer the world amidst the dangers of hunger, social unrest and environmental
degradation, and to sustain the existing power structure – as befits hierarchist insti-
tutions (§6.2).
The opponents will attack them from various sides. Markets only favour the rich,
technology brings new problems, corporate responsibility is window-dressing. ‘We’
should introduce fair trade and green label certificates, eat more vegetarian, forbid
child labour, support local cooperatives and farming communities, establish a fund
for climate adaptation, and value again local food and traditions. And ‘we’ should
avoid paternalism and neo-colonialism: ‘Are African pastoralists the panda bear
and ice bear of global agriculture?’ One direction is voiced by Wangari Maathai’s
‘challenge for Africa’ Green Belt movement and the Chipko movement in India:
fight forest destruction and promote small-scale production for the local market,
establish community-based cooperatives, practise organic agriculture and create
local employment. Another direction is to focus on UN-based regulation, with global
targets for poverty reduction and greenhouse gas emissions and with strict standards
for child labour, fair trade and sustainability. Somehow a balance must be found.
11.7 Summary Points

Food is an essential need of people. It is produced in the agro-food systems of the
world, which are diverse, local and increasingly connected. In combination with
population growth and environmental constraints, its adequate provision is one of
the great challenges. Some important points to remember are:
r The basis of agro-food systems is the human habitat, that can be categorised in
anthromes. Social-ecological systems (SES) are the natural unit of investigation.
r The present-day agro-food systems are the outcome of a millennial evolution
and the human population lived and still lives in quite different anthromes and
stages.
r Since the 1950s, world food production is a success story. With little more land,
much more food is produced. This intensification has unintended and undesired
side effects, such as input dependence and environmentally harmful outputs.
There is room for another 40 percent to 60 percent increase in food production,
but it may further endanger long-term sustainability.
r The causes of land use/cover change (LUCC) are diverse and local. Important
drivers are the opportunities for the poor and the conversion for urban areas
and infrastructure.
r The global food system is driven by a few causal loops, notably cost competition,
upscaling and innovation. Governments and multinational corporations have
become the major actors. Trade liberalisation and consumer demands for social,
environmental and health regulations are important co-drivers of development.
Both are important but only partly understood ingredients of a transition to a
sustainable food system.
The Months
Januar: By this fire I warme my handes,
Februar: And with my spade I delfe my landes.
Marche: Here I sette my nthinge to springe,
Aprile: And here I heer the fowlès singe.
Maii: I am as light as birde in bow,
Junii: And I weede my corne well ynow.
Julii: With my sithe my mede I mowe,
Auguste: And here I shere my corne full lowe.
September: With my flail I erne my bred,
October: And here I sowe my whete so red.
November: At Marinesmasse I kille my swine,
December: And at Christesmasse I drinke red wine.
15th century poem. The Oxford Book of
Medieval English Verse, Eds. Sisam,
– Clarendon Press, Oxford
Ït’s the land that feeds our children, you cannot own the land it’s the land that owns
you.
– Dolores Keane
The counter-seasonality of production in the Southern Hemisphere for Northern

Hemisphere consumers has created what has been termed ‘permanent global sum-
mertime’.
– Dicken, Global Shift, 2009:376
SUGGESTED READING
A number of case studies and models about resource management in local communities.
and Social Mechanisms for Building Resilience. New York: Cambridge University Press,
1998.
Although from the 1980s, this book is still a good, systematic overview of elementary interactions
between man and the biosphere.
Goudie, A. The Human Impact – Man’s Role in Environmental Change. Cambridge: MIT
Press, 1981.
A comprehensive overview of LUCC-related topics.
Lambin, E., and H. Geist. Land-Use and Land-Cover Change – Local Processes and Global
Impacts. The IGBP Series. Berlin: Springer, 2006.
A thorough introduction in economic aspects of resource and environment issues, partly on
food.
Perman, R., Yue Ma, J. McGilvray and M. Common, eds. Natural Resource and Environmental
Economics. Harlow: Pearson Education Ltd., 2003.
Twelve fascinating anthropological essays about peoples all over the world in their relationships
with the Earth.
Reader, J. Man on Earth: A Celebration of Mankind. New York: Perennial Library, 1988.
An historical account of the meaning of food in society.
Tannahill, R. Food in History. New York: Three Rivers Press, 1973.
Appendix 11.1 Income and Price Elasticity 377
USEFUL WEBSITES
Wikipedia presents good descriptions of resource and environmental economics definitions

and concepts. There are also sites on specific models and applications:
r glossary.eea.europa.eu/ is a general site of the European Environmental Agency with

brief explanation of terms in the area of environmental economics and policy.
r themasites.pbl.nl/en/themasites/image/index.html is the site of the Integrated Model to
Assess the Global Environment (IMAGE) model. Also, the HYDE project on historical
land use, the PHOENIX-model on population and some other models and databases can
be accessed via this site.
r www.iiasa.ac.at/Research/LUC/GAEZ/ is the IIASA-operated site where the data on
global agro-ecological zones and on the Harmonized World Soil Database (HWSD) can
be accessed.
r www.fao.org/ of the UN FAO with many news items and downloadable papers and
statistical data on food and agriculture; a subsite is on sustainable development, in partic-
ular: The State of Food and Agriculture (SOFA), The State of the World Food Insecurity
2010 (SOFI) and The State of Agricultural Commodity Markets (SOCO).
r www.grida.no/ is a site with extensive data, graphs and maps on Environmental
Knowledge for Change, operated by UNEP/GRID Arendal.
r www.itc.nl/ is the site of the ITC, a research and education institute on GIS-based data
collection and processing.
r www.cluemodel.nl is the site of the Conversion of Land Use and its Effects (CLUE)
model.
r www.wri.org/project/ is the site of the World Resources Institute (WRI) with data and
reports on major global environment and development issues.
r www.maweb.org/en/index.aspx offers an extensive guide to the Millennium Ecosystem
Assessment published in 2005.
r www.ifpri.org/ is the site of the International Food Policy Research Institute (IFPRI)
with an overview of their reports and projects.
r www.sustainablefoodlab.org/ is an organisation and is the site of a global partnership for
innovative solutions towards a sustainable food system.
r www.eururalis.eu is the site of Eururalis project, an extensive scenario analysis of futures
and policies regarding the rural areas in Europe.
r www.avsf.org/ is the site of the Agronomes sans Frontières, an NGO for assistance to
poor farmers worldwide.
Appendix 11.1 Income and Price Elasticity

The income elasticity can be expressed in formula for food expenditures F and
income I:
F/F
ε= (A11.1)
I/I
is the change during the period of observation, usually one year. If the ratio of
food expenditure and income in year t equals Rt and the income growth rate is α,
then it can be shown that:
Rt+1 F + F 1 + εα
= = (A11.2)
Rt I + I 1+α
The part of income spent on food will decline if ε < 1. The price-elasticity π is defined
in similar fashion:
F/F
π =−
p/p
with p the price of the food commodity. Because it is assumed that amounts decrease
for increasing price, it is defined with a minus sign.
12 Renewable Resources: Water, Fish
and Forest
12.1 Introduction: Lakeland

There are other renewable resources than soil such as water, fish and forest produce.
Let us make a trip to the imaginary country Lakeland to illustrate the nexus between
population and resource exploitation. The country consists of two subsystems: a lake
or coastal area with fish and their food – say, daphnia, a kind of water flea – and the
economy of people, fishing boats and a reserve of mineral ore with high gold content
(Figure 12.1). The Lakelanders get their food from fishing in the local lake and make
their boats and nets from local materials. There are only two capital stocks: boats and
nets, and mining equipment. They change because of investment and depreciation,
with the latter assumed to be inversely proportional to the economic lifetime. The
demand for fish is growing over time because of population growth. This means
more boats and nets. Therefore, you can make three reasonable assumptions about
the fishermen behaviour and the ecosystem:
r fishermen invest in new boats and nets in order to satisfy expected growing
demand;
r there is an upper bound to the annual fish catch; and
r gold mining causes pollution, which negatively affects the daphnia.
Each season the fishermen go out harvesting fish, which is sold on the local market at
the end of the season. Assuming a population growth of 1 percent per year (%/yr), the
Lakeland fishing grounds are sustainably exploited over the chosen simulation period
of about twenty years (1,100 weeks; Figure 12.2a).1 As long as the population and
fish demand hardly change, the fleet size remains about constant. Such a sustainable
state has existed in many places and for long periods, with different institutional
arrangements. Note that the model is meant to be an illustration and has not been
implemented for a real-world situation.
The islanders can ‘develop’ and enter into the large ‘world’ economy by opening
up the gold mine for exploitation and by allowing foreign fishing vessels to enter
the lake. Often, the population does not really have a choice and is simply exposed
1 The Stella
R
version of the model is available upon request.
379
380 Renewable Resources: Water, Fish and Forest
Figure 12.1. The Lakeland model world: a human population that sustains itself by fishing
and exploiting a gold mine, and an ecosystem with a prey-predator dynamic and pollution of
water and sediment.
to such developments by foreigners, either as military or commercial force. Aldous

Huxley gives a compelling description in his book Island (1962) how such a process
can unfold.
Exploiting the gold deposit can become the first and most important, export-
oriented sector in the economy. It not only provides revenues to the Lakeland
community or government, but it also generates mining waste that flows into the
lake and pollutes the water. The pollutant concentration influences the reproduction
rate of the daphnia, which in turn affects the fish population that predates on the
daphnia. The mining waste slowly accumulates in the lake sediment, building up a
Box 12.1. Lakeland – a model world. Lakeland is an island with a large and
beautiful lake that has been populated by fish as long as the Lakeland people
can remember. The Lakelanders loved to eat the large fish which swam around
so abundantly. The human population was in equilibrium with the ecosystem – a
sustainable state as long as population growth remained small. However, one day,
gold was discovered on the island, and it underwent ‘development’ in the form
of a gold mine. The mining caused water pollution. Then another ‘development’
event took place: Foreign fishing companies got permission to catch and export
fish. After a few years, the fishermen noticed that it was becoming harder and
harder to catch fish. Fish on the local market was becoming more expensive, and,
to their distress, the Lakelanders discovered that the water quality in the lake was
declining. Not long after that, the gold resources also became depleted.
a) Fish population Local fleet size Fish catch (per week)

25000 500
Fleet size and fish catch

20000 400
Fish population
15000 300
10000 200
5000 100
0 0
0 200 400 600 800 1000
Week
Figure 12.2a. Path of the fish population, the fleet and the weekly catch during the fishing sea-
son (note: time axis is in weeks). The catch changes the fish-daphnia system only temporarily
and insignificantly (t = 1, Euler).
b)
Fish population Local fleet size Fish catch (per week)
25000 500

20000 400
Fish population
15000 300
10000 200
5000 100
0 0
0 200 400 600 800 1000
Week
Figure 12.2b. As in (a), and with pollution from gold mining at 1%/yr, affecting the daphnia
population and, subsequently, the fish population and fish catch (t = 1, Euler).
stock of pollutant. As a consequence, even if pollution stops the concentration in the

water declines only slowly because the pollutant in the sediment dissolves again.2
A possible outcome is shown in Figure 12.2b. The growing pollutant concen-
tration in the lake reduces the fishing productivity, defined as catch per unit effort.
Fishermen respond by building and sending out more boats. But the mining waste
flow continues to harm the ecosystem, and after about ten years, the fish population
collapses (Figure 12.2b). The Lakelanders have to import fish – for which they can
2 In environmental science, this phenomenon is known in its possibly catastrophic form as chemical
timebomb. See Carpenter and Brock (2006) for the analysis of such a system for phosphorous in a
lake-sediment system and the detection of an early warning indicator of regime shift (§9.5).
c) Fish population Total fleet size Fish catch (per week)
25000 500

20000 400
Fish population
15000 300
10000 200
5000 100
0 0
0 200 400 600 800 1000
Week
Figure 12.2c. As in (a), and with permission for a maximum of 500 foreign boats to fish for
export (t = 1, Euler).
use the revenues from gold mining, provided these are not spent on luxury items or
transferred to a foreign bank. The lake is no longer a food resource and the sediment
has become loaded with pollutant (Figure 12.3a). The phase diagram plot in Figure
12.3b shows how fish population (x-axis) and daphnia population (y-axis) move in
time. The gold resource gets depleted and the ecosystem degrades to a much lower
ecosystem biomass. The system has shifted to a new equilibrium with a significantly
lower quality of life for a human population.
The other option for development in Lakeland is to let a foreign fishing fleet
enter the fishing grounds. The foreign boats fish for export to the world market
a)
120000
Poll in water
Fish population
80000 Poll in sediment
40000
0
0 200 400 600 800 1000
Week
Figure 12.3a. The pollution level in water and sediment for the 1%/yr gold mining (t = 1,
Euler).
b)
50000
40000
Daphnia population
30000
20000
10000
0
0 5000 10000 15000 20000 25000
Fish population
Figure 12.3b. Phase space diagram for the simulation with gold exploitation (black curve;
Figure 12.2b) and for the simulation with gold exploitation and foreign fishing boats as
discussed in the text (grey curve) (t = 1, Euler).
where prices are assumed to be constant. Whether and how many foreign fishing
boats will come fishing depends on the profit ratio. They will go elsewhere if the net
income, equal to fish sales minus operational cost, is less than the capital cost of the
vessels. The foreign fleet is in fact a top predator, with the financial sector with its
return on capital criteria as the ultimate top predator. The typical pattern is that the
foreign fleet rapidly enters the area, increases in size towards the maximum catch
permitted and declines in response to declining fishing productivity from overfishing.
Assuming that the government issues permits from the third season onwards
for a total of 500 ships, a possible outcome is shown in Figure 12.2c. The story is
that foreign boats capture the opportunity of high returns on capital investment, but
start leaving the area once the high return on capital can no longer be sustained.
The total fleet (local plus foreign) then reaches a sustainable exploitation level. The
damage is not irreversible, but government revenues from fishing permits and local
fish availability both decline. The situation may get worse if the government gives in
to the temptation to ignore pollution and overfishing and, instead, decides to ward
off local opposition and subsidise local fishermen for as long as it has revenues to do
so. Such a policy actually accelerates the collapse and is part of what is discussed in
Chapter 13 as the resource curse (Ross 2001). The combination of gold mining and
foreign fleet fishing spells disaster (Figure 12.3b).
More sustainable pathways are possible. The government can, for instance,
enforce environmental regulation on the mining waste flow or spend part of the
revenues on pollution abatement. It can also be more stringent with permits for
foreign fishing companies and tax them more heavily, particularly if mining is also
permitted. From a policy point of view, the trade-off for the Lakeland community
is between a variety of assets, each with its own indicators: ecosystem health, gold
resource, local food supply and supply security and, last but not least, government
income from taxing gold mining and fishing activities.
Table 12.1. A classification of renewable (bio)resources
Resource Stock Inflow Outflow
Soils (Chapter 11) Soils area/thickness Soil formation Soil erosion (dust storms
Organic content (sedimentation, etc.)
vegetation, etc.)
Rivers (Flow resource) Water flow downstream Natural outflow in sea
Abstraction (dams)
Groundwater Water reservoir content Natural recharge Natural discharge
Abstraction (pumps)
Fisheries Fish population Natural birth Natural death
Harvest/catch
Fish ponds Fish population Organised growth Harvest
(aquaculture) (farming)
Forests Tree biomass Regrowth, afforestation Natural death
Cutting/deforestation
Forest produce Game; fruit, nut etc. Reproduction Natural death
biomass Produce (re)growth Hunting
Harvest/catch
Wild animals Animal population Natural birth Natural death
(game) (deer, swine and so on) Hunting
Sunlight, wind (Flow resource) Electromagnetic Same (collector, panel,
radiation turbine)
Moving air mass
12.2 Renewable Resources
12.2.1 Renewable Resource Use: An Archetypical Model

Resources that can be harvested are called renewable (bio)resources when their
regrowth rate as part of natural processes is in the same order of magnitude as the
harvest rate. The regenerative and absorption/degradation functions of the envir-
onment (‘sink’) can also be considered a renewable resource, because they provide
ecosystem services and have limited regrowth rates. If the regrowth or regeneration
rate is happening on much longer timescales, as with fossil fuel and mineral ore
formation, resources are categorised as non-renewable or finite resources, which is
the topic of next chapter.3 If harvesting relates to the tapping from a flow, as with
river water or wind and solar energy, it is best to speak of flow resources. Their prime
characteristic is that they are, for all practical purposes, infinite in time as well as
local and fluctuating. They are ‘free’ but their conversion into useful energy requires
capital investments as with other resources. Table 12.1 gives a list of bioresources.4
Soils are an essential renewable resource in the agro-food system. Other such
resources are water, the fish in seas, rivers and lakes, timber, wild animals and
other produce in forests. To a large extent, they were and still are exploited ‘in
3 Also, fossil groundwater reservoirs can be considered finite, as their formation rate is very slow.
4 What actually constitutes a resource is not as objective as one may think. A resource is a part
of the natural environment that provides human beings (the opportunity of) satisfying needs and
desires. As such, it reflects prevailing needs and desires as well as technical skills and organisational
capabilities (§6.1).
the wild’, as open access resources, because people are less easily excluded from
using them and they are less easily appropriated (§5.2). Their overexploitation and
mismanagement affect ever more people and managing them sustainably becomes
an urgent task. Science contributes with research in scientific disciplines such as
fisheries and forestry economics and human ecology, and numerous models have
been formulated and applied. I refer the reader to the Suggested Reading.
The archetypical or generic population-resource models simulates a hunter-
gatherer or horticultural/agricultural population of size P, that exploits a local renew-
able resource R. The population growth rate is a function of the number of children
per woman during her fertile period and on the life expectancy at birth. These in turn
depend on the resource, which in this case is the available food. The resource has a
finite capacity to regenerate itself from harvest or other disturbances. In theory, it
can be used infinitely if it is harvested below the natural regeneration rate. Questions
to explore are:
r how will resource size and quality constrain the harvest rate, for instance,
through lower animal density, soil erosion and the like;
r how will this in turn affect the population dynamics, for instance, through
increased mortality and subsequent starvation or violent deaths from mutual
competition; and
r how will the human population respond, for instance, by outward migration,
adjustments of the number of births, technical innovations or concentration in
certain places where chances for survival are highest.
These questions are also addressed in the discussion on past civilisations and their
collapse in Chapter 4 as well as in the syndrome/archetype analysis in Chapter 4. A
simple mathematical description is the following model (Appendix 12.1):
dP/dt = bP − P/L (12.1a)
dR/dt = αR(1 − R/K) − βP (12.1b)
where P is population (stock), b is the birth rate parameter and L is average

life expectancy. R signifies the resource size (stock) and dR/dt its change (rate),
α its regeneration rate parameter and K its carrying capacity in the area under
consideration.5 The harvest rate H is set proportional to the population, with the
parameter β the resource use per person. Let us assume that the population does not
grow (dP/dt = 0 → b = 1/L). Given the harvest rate H = βP, sustainable exploitation
happens for dR/dt = 0. This condition is fulfilled if:
αR(1 − R/K) = βP (12.2)
It is useful to plot the exploitation rate dR/dt as a function of the resource R in

the phase plane (Figure 12.4a). The maximum sustainable yield (MSY) happens for
Hmax = 1/4 αK at which the resource regenerates at its maximum rate (Appendix
12.1). Exploitation is sustainable as long as H ≤ Hmax , which is possible for all
5 1/L is used for the mortality rate, which implies an exponential decay of the population if no
children are born (§2.3). We assume that the resource is homogenously distributed and follows a
logistic growth equation (§9.2).
Figure 12.4a. Phase plane diagram of the resource depletion

according to equation 12.2a. There are two attractors, one
stable and one unstable. The arrows indicate the direction in
which the fish population (stock/state variable) will move.
combinations of P and β for which Pβ ≤ Hmax . The carrying capacity for people in
this situation is Pmax = 1/4 αK/β. For P > Pmax exploitation exceeds the maximum
rate at which the resource can regenerate and the system is exploited unsustainably.
This extremely simple model shows that every growing population will at some
point be confronted with exploitation of its resource beyond the sustainable level.
The situation is worse if the resource regeneration rate is diminishing as a con-
sequence of human interference, for instance, as with soil erosion or age-specific fish
removal. This is tantamount to a decrease of the carrying capacity K. It is also worse
in the presence of random fluctuations in environmental parameters (May 1977). If
the carrying capacity is exceeded for prolonged periods and resource productivity
starts falling, the first option is to move to another area with or discover new unex-
ploited resources (extensify). A second option is to reduce demand for the resource
by more sober and efficient use or use of substitutes, for instance, fish from ponds
(reduce/substitute). A third option is to make the resource more productive by apply-
ing innovations, for instance, feeding the fish with agricultural remains (intensify).
These options, if successful, are effectively increasing the carrying capacity K for the
population.
Figure 12.4b. As Figure 12.4a, but with non-logistic

dynamics.
Box 12.2. Archetypical human-environment models. The simple mathem-

atical population-resource model suggests four typical population trajectories
(Figure 12.5). Curve (1) upper left is an exponential population growth (0.6%/yr)
leading to a growth from 10 to more than 400 individuals within 600 timesteps.
Without an upper bound, it leads to infinity and hence has to stop (§2.3).
Curve (2) upper right shows what happens if there is a limit: a smooth approach
towards the carrying capacity of 400 individuals, which may, however, easily
become oscillatory (overshoot-and-collapse) (§9.3). Curve (3) lower left shows
what happens when the limit is lifted due to technical ingenuity or behavioural
changes. This is the optimists’ world. The limit may also be brought down by nat-
ural processes or mismanagement. Curve (4) lower right shows this pessimists’
pathway.
The use of the logistic growth model to forecast growth and stabilisation of
human populations has led to notoriously wrong results. Some claim that the
carrying capacity is already exceeded with the present nearly 7 billion people.
Others have calculated that, from a food supply point of view, the carrying capa-
city for the human world population is 30 billion individuals or more. There is
not much empirical basis for extrapolation, so the value of K has to be chosen
from other considerations. This is difficult because human societies are complex
adaptive systems with continuous interaction between expansion and constraint,
opportunity and threat. They coevolve with their environment, as explained in
niche construction theory (§10.3).
12.2.2 Model Extensions and Management Principles

The model previously outlined is far too simple to be used for management. A
first objection, from the side of biologists and ecologists, is that the logistic growth
equation for R may be a good approximation for certain non-migratory animals and
biomass stocks, but a more complex description is needed to represent real-world
dynamics (§9.5). A possible consequence for resource management is shown in the
phase plane plot of dR/dt against R (Figure 12.4). In the logistic model, the area
RA < R < RB is the ‘safe’ region at a harvest rate H (Figure 12.4a). However, if
the natural regeneration rate is smaller or even below a certain threshold value of
the stock – because of density-effects, as shown in Figure 12.4b – the ‘safe’ region
is significantly smaller (Allee effect). The system has now three attractors. The
reproduction rate can at (very) low population densities also be higher than in the
logistic growth model. For instance, there is a possibility that reproduction rates go
up at low density as a consequence of genotypic adaptation, such as evolutionary trait
adjustment via earlier maturation, and not of phenotypic adaptation, which results
in less crowding and less competition for food. The ‘safe’ region is larger. What is
sustainable depends on what is proven to be the ‘correct’ model. Introducing features
such as seasonality, age-structure of populations, energy metabolism, trophic levels,
spatial interactions and genetic adaptation make the models more complex and
introduce almost certainly more equilibrium states and the possibility of sudden
change (§9.5).
388
a) b)
800 800
Logistic growth of X Logistic growth of X
Carrying Capacity Carrying Capacity
Exponential growth of X Exponential growth of X
Size of X
Size of X
400 400
0 0
0 120 240 360 480 600 Final 0 120 240 360 480 600 Final
Timestep Timestep
c) d)
800 800
Logistic growth of X Logistic growth of X
Exponential growth of X Exponential growth of X
Carrying Capacity Carrying Capacity
Size of X
Size of X
400 400
0 0
0 120 240 360 480 600 Final 0 120 240 360 480 600 Final
Timestep Timestep
Figure 12.5. Population-resource dynamics: four possible interactions.
A second objection comes from economists. They wish to know the least-cost
extraction rate so as to replace the simple extraction rate H = βP with a more
realistic harvesting function. In an open-access situation, everyone has the right to
fish, hunt, let animals graze or use upstream river water. The usual assumption is that
the population will exploit the resource up to the point where one more exploiter
has (perceived) gains that no longer outweigh (perceived) efforts. The following
introduces some simplifying assumptions not unlike those in the Lakeland world:
r the harvest H is a function of the harvesting effort E, measured, for instance, as
the number of boats times the fishing period;6
r the harvest also depends on the size of the stock R. The more resource, the
higher the harvest for a given effort level E;
r both the resource and the firm are experiencing diminishing returns, that is,
the additional harvest H per unit of additional effort E and of resource R
respectively is declining.
In an open-access situation, in theory, entry and exit are free and without cost;
therefore, new exploiters keep entering the business until for the last entrant the
profit is zero. If this is introduced as a behavioural rule in a competitive open-access
situation, the steady-state effort level will rise for higher resource price and for lower
resource cost (Appendix 12.2). Thus, if consumers are willing to pay higher prices
and/or firms can bring down costs by innovating, the resource will more intensely be
exploited. This is precisely the mechanism behind the near-extinction of several fish
populations and some other resources in the world.
The exploiter of the resource is expected to be primarily interested in max-
imising profits along a particular resource extraction path. Already in the 1960s,
mathematical models have been proposed in order to formulate optimal resource
extraction paths (Dasgupta and Heal 1979). An optimal harvest rate maximises over
time the profit (price minus cost). In formula:
t=T
max [(p − c) H] · e−rt dt (12.3)

t=1
with T the planning or time horizon in years, p the price and c the cost in money
per unit of catch. The term e−rt is the discount factor (§10.4). The variable r can be
interpreted as the interest rate on capital. The mathematics to solve this maximisation
problem has been dealt with extensively since the first models appeared in the 1960s
(Clark 1990; Perman et al. 2003). The agent in these models can be an oligopolistic
firm or a government planning agency. With a non-zero discount rate, the resource
is always exploited until depletion or extinction. More advanced models have been
constructed, but the basic economic rationale remains the same and so do the long-
run outcomes. I refer the reader to the Suggested Reading for more details. And
remember: All models are false, but some are useful. In other words, one has to strike
6 Often, a simple function of the form H = εEa Rb with a, b < 1 is used. Such a function is what
economists call an economic or engineering production function. This form is known as the Cobb–
Douglas production function (§14.2). An even simpler assumptions is that H is proportional to
E (a = 1, b = 0).
the balance between adding more scientifically relevant and interesting detail and
ensuring that the model is still transparent and useful for practitioners.
Is it possible to manage renewable (bio)resources sustainably? They used to
be open access resources that everyone could use and from which people could
not easily be excluded. If such a resource becomes less productive when it is more
deeply exploited, it is not only non-excludable but also becomes rivalrous. It is called
a common pool resource (CPR) (§5.3). In a widely cited paper entitled The Tragedy
of the Commons (1968), the biologist Hardin suggested that there is an inherent
tendency amongst humans to overexploit such a shared, common (or collective)
resource: ‘Picture a pasture open to all. It is to be expected that each herdsman
will try to keep as many cattle as possible on the commons . . . As a rational being,
each herdsman seeks to maximize his gain [and] concludes that the only sensible
course for him to pursue is to add another animal to his herd. And another; and
another . . . But this is the conclusion reached by each and every rational herdsman
sharing a commons. Therein is the tragedy’ (Hardin 1968). This is what the previous
simple model showed: Open-access CPRs are exploited until the very last unit as long
as someone pays for it. It was concluded that only less people or less freedom, under
one private or public centralised owner/manager, can avoid tragedy. The situation
that economically rational behaviour of individuals causes an undesirable outcome
for the collective is known in social science as a social dilemma (§10.4).
There have been many responses to Hardin’s paper, and a whole branch of sci-
ence has evolved around CPR-management. Currently, a widely shared view is that
a tragedy of the commons is not inevitable. Both markets and centralised control can
cause degradation and collapse. But there are other management regimes, besides
the options to limit access by private ownership or government-issued permits and
quotas. In her book Governing the Commons (1990), Ostrom has shown that people
all over the world have developed institutions to harvest the CPRs on their lands.
Some of these have lasted over 1,000 years and have survived droughts, floods,
diseases and economic and political turmoil. These management regimes operated
often in highly variable and uncertain environments. The appropriators, that is,
the individuals who use the CPR, have ‘designed basic operational rules, created
organisations to undertake the operational management of their common property
regimes, and modified their rules over time in light of past experience according to
their own collective-choice and constitutional-choice rules’ (Ostrom 1990).7 They
solved the problem of commitment and mutual monitoring – amongst them the
avoidance of free-riders and effective sanctioning – without resorting to either cen-
tralised power exercised by external agents or competitive market institutions.
Based on historical analyses, fieldwork and outcomes of behavioural experi-
ments, a provisional list of design principles for sustainable management of and
institutions for CPR can be constructed (§10.4; Ostrom 2000; Ostrom et al. 2008):
r clear boundary rules: who is in and who is not;

r local rules-in-use should restrict amount, timing and technology of harvesting,
relate benefits to inputs, and take local conditions into account;
7 The term common property regimes (CPR) has been replaced, after careful consideration, by the
term common pool resources (CPR) management (Ostrom et al. 2002).
r individuals affected by the resource regime can participate in making and modi-
fying their rules;
r users select their own monitors, who are accountable to the users or are users
themselves and who keeps an eye on resource conditions as well as on user
behaviour;
r sanctions should be graduated, depending on the seriousness and context of the
offence.
Often, resource exploitation occurs under a regime that violates most, if not all
of, these rules and in which an urban business or government elite is the primary
external force. The real threat are then loss of community and the anonimity that
facilitates overexploitation and shifting-the-burden mechanisms. Proper mixes of
centralised, top-down regulatory and local, bottom-up experiences and rules appear
most promising and should be experimented with. But it is a complex matter and
there are no one-for-all solutions.
MDG Goal 7: Ensure environmental sustainability.

TARGETS
C. Halve, by 2015, the proportion of the population without sustainable access to
safe drinking water and basic sanitation.
“If current trends continue, the world will meet or even exceed the MDG drinking
water target by 2015. By that time, an estimated 86 per cent of the population
in developing regions will have gained access to improved sources of drinking
water . . . ”.
– The Millennium Development Goals Report 2010, New York
12.3 Water Resources
12.3.1 Water Availability and Use

Water is essential for human life, as drinking water, for crops and animals, and for
many industrial processes. The hydrological cycle plays a crucial role in biospheric
processes and in human affairs, and water is with carbon dioxide (CO2 ) the building
block for organic compounds via photosynthesis (Figure 9.7). Hydrologists make
a distinction between freshwater in rivers, lakes and aquifers called ‘blue water’
resources and moisture in the soil and vegetation called ‘green water’ resources.
Blue water availability is the total river discharge, which equals precipitation minus
evapotranspiration or the sum of surface runoff and groundwater. It depends strongly
on seasonality and interannual variability in the flow. Ultimately, the water recharges
aquifers and flows off to the oceans. The larger part of rainfall infiltrates into the
soil and gets stored as green water on its way to the oceans and the atmosphere.
Its availability depends on soil water holding capacity and other soil characteristics.
It is moisture or water stored in the soil and productively used for transpiration
by vegetation. It is important for irrigated and rainfed crop growing and animal
husbandry, much less for industrial and domestic use.
Water availability varies in amount and time of year with the local situation.
Water use is diverse: A city using nearby river water for drinking water, for cooling
of electric power condensers and for diluting urban waste flows is very different
from a village dependent on rainwater collection or a groundwater reservoir for
irrigating crops. What is sustainable water use? As with all bioresource exploitation,
the answer has to consider the availability at a given time and place and of a given
quality with respect to the demand. Demand is related to needs and functions, but,
as with food, it is usually equated to water use for the past and projected use for
the future. The resulting supply-demand match has usually grown out of a long
history. Water engineering and management for drinking and irrigation goes back
thousands of years in the Middle East and Asia. Populations in arid and semi-arid
regions have developed all kinds of ingenious solutions and schemes, aware of the
common pool aspect of water. Famous are the qanats in Iran, constructions of shafts
and tunnels, which can deliver water without pumping. In Spanish Andalucia, the
Moors are remembered for their intricate irrigation canals, constructed during the
10th and 11th century, and the sophisticated water allocation schemes some of which
still survive. But in the last half century, the traditional solutions are in many places
overwhelmed by growing demand and changing lifestyles.
With the accelerating growth in human numbers and activities, the use of the
freshwater stocks (lakes, soil/groundwater) and flows (rivers) has increased in extent
and intensity. Water is no longer an open access ‘free’ resource, for which one does
not have to pay. It has become a potential source of profit and water-related manufac-
turing sales (pumps, pipes, filtration, desalination equipment and so on) amounting
already to more than €300 billion per year. Because water has deeply social and
cultural connotations, like food, its commodification as part of modernisation is still
resisted in many places (Robert 1994).
One estimate of global water use is that food production around the year 2000
consumed 6,800 cubic kilometres per year (km3 /yr) worldwide, of which 1,800 km3 /yr
are from blue water sources and the remaining 5,000 km3 /yr are green water flows
in evapotranspiration processes in rainfed agriculture (Falkenmark and Rockström
2006). The majority of assessments have focussed on blue water use by human
activities. An estimate of total global water withdrawal rates for human purposes in
the 1990s is in the range of 2,500 km3 /yr, if fossil groundwater and non-local blue
water are included (Hanjra and Qureshi 2010). The total withdrawal rate is estimated
to have more than doubled since the 1960s (Wada et al. 2011; Figure 12.6). Over 40
percent is in China and India. There are large differences in per capita availability
and withdrawal rates (Table 12.2). Significant fractions of rural populations have,
at least according to official estimates, no access to safe water for drinking and
sanitation, which impacts life expectancy and other quality-of-life indicators (§10.3).
In order to meet water demand, people have interfered in many ways with the
natural stocks and flows, notably by the construction of dams, embankments and
dikes to contain and use river flows and by construction of open and groundwater
reservoirs for irrigation. Dams to store water are an ancient solution to control
the fluctuating supply and demand, later in combination with power generation.
Worldwide the number of dams increased from less than 1,000 in the year 1900 to
over 30,000 in 2000, almost half of them being constructed for irrigation purposes
(Haddeland 2006). Besides direct interference in the water cycle, the natural water
Table 12.2. Indicators about worldwide water withdrawal, supply and accessa
Total
freshwater Water Water Water Access to
withdrawal Idem, per use for use for use for water
rate person agriculture industry domestic (urban/rural)
supply sanitation
Unit: km3 /yr m3 /cap/yr % of total % population serves
Africa 213.2 260 83.1 4.3 12.6 85/47 85/45

Asia 3,900.8 1,060 84.9 7.2 7.9 93/74 78/31
Europe 728 998 29.3 48.5 22.2 99/88 99/74
N America 510.9 1,073 44.1 33.9 22 100/100 100/100
S America 375 1,099 84.8 6.4 8.8 93/62 87/48
Oceania 31.4 1,094 64.9 10.4 24.7 100/67 100/78
a Source: Wada et al. (2011).
stocks and flows have been affected as a consequence of changes in land cover and
land use. In retrospect, the issue is always the tension between human activities and
natural possibilities. In other words: scarcity.
The first attempts to assess water scarcity at the global level were made by
Falkenmark and Lindh (1976), who classified national per capita water use from no
stress (>1,700 m3 /cap/yr)8 to absolute scarcity (<500 m3 /cap/yr). Later analyses have
refined the notion by including local features down to the scale of water basins. Other
indicators of scarcity have been proposed, such as the ratio of annual withdrawals
and availability rates (criticality ratio) and the ratio of total water demand and river
discharge. The higher their value, the higher the water stress. Usually, a threshold
value of 0.4 is used, above which there is high water stress. Based on indicators like
these, it is estimated that some 20 percent of the world population is confronted with
water scarcity, defined as a situation in which the amount of water use is close to
or exceeding that of water availability. In these areas, water is only available a few
hours per day or has to be bought at high prices from tank wagon or ship supply.
The annual averages conceal the large local differences and the role of seasonal and
interannual fluctuations. An additional complication in water assessments is water
quality. With increasing population density and agricultural and industrial activity,
water pollution is in many places a serious concern with important consequences for
people’s health (§10.3).
In the last decade, assessments are made on a spatially explicit river basin or
even more detailed level (Alcamo et al. 2003). One recent assessment is based on
the global hydrological model PCR-GLOBWB, which is used to calculate the water
stress on a 0.5◦ × 0.5◦ spatial resolution (van Beek and Bierkens 2008; van Beek et al.
2011, Wada et al. 2011). The model considers soil and vegetation characteristics and
runoff and evaporation flows, amongst others. An average climate is constructed
from climatological data for the period 1958–2001. Monthly water use and water
availability are calculated and converted into a distributed water stress index WSI:
W SIblue/green = Wdemand,blue/green /Wavailable,blue/green mln m3 /month (12.4)
8 Common units in water use are litre per person per day and cubic metre per person per day
(m3 /cap/yr). The conversion factor is 1 m3 /cap/yr = 1,000/365 ∼ 2.74 litre/cap/day.
It is done separately for blue and green water because the latter is only available for
crops. Next, a dynamic water stress index, DWSI, is calculated that quantifies the
severity of water stress by incorporating information on the exceedance of a water-
stress threshold value and the duration and frequency of the periods of exceedance.9
The dynamic water stress map for blue and green water is shown in Figure 12.6a.
There are spots of severe water stress throughout the world, but it is worst in the
densely populated and partly arid parts of India and China, in a Sub-Saharan belt
and in (semi-)arid regions of North America.
An important blue water source are groundwater aquifers. Their exploitation has
rapidly increased since the 1960s, particularly for agricultural and urban use. It can
at low cost be extracted with easily applicable water pumps, has excellent quality and
can be used with low transport losses in irrigation. In urban areas, clean groundwater
provides a substitute for polluted surface water. The largest withdrawals are in India,
the United States, Pakistan, China and Iran, together an estimated 70 percent of
global withdrawals (Giordano 2009). From a sustainable management perspective,
the ratio between the actual abstraction rate and the renewable recharge rate is the
relevant indicator and should be less than one. This ratio is alarmingly high in (semi-)
arid countries such as Saudi Arabia (9.5), Libya (8) and Egypt (3.5) and is increasing
rapidly in densely populated countries like Pakistan (1.1), Iran (1), Bangladesh (0.5)
and India (0.45) (www.worldwater.org). Globally, only 6 percent of the recharge is
abstracted – but global aggregates are misleading. Using the system principle that
outflow should not exceed inflow, groundwater use is utterly unsustainable in quite
a few places on earth.
Wada et al. (2010) assessed the average yearly groundwater recharge flows on
a 0.5◦ × 0.5◦ spatial grid and downscaled country-based groundwater abstraction
rates to the same spatial resolution by using water demand as a proxy. The resulting
maps give an indication of the extent to which groundwater depletion, defined as
abstraction minus recharge rate, occurs. It is a hot-spot map that coincides quite well
with the information from local qualitative and quantitative accounts. According
to this analysis, in the order of 40 percent of groundwater abstraction contributes
to depletion (Figure 12.6b). There is, for instance, severe groundwater depletion
in northern India. Naturally, there is a relation with the extent and type of agri-
culture and pastoralism in view of the large water demand for crops and animals
(Ramankutty et al. 2008).10
What are the prospects for sufficient and sustainable water supply for human
populations?11 Blue water river sources are now drained to such an extent that some
major rivers – Colorado, Nile, Ganges, Indus and Yellow – are almost dried up
for parts of the year upon reaching the sea. Some of the large inland lakes in the
world are shrinking because the inflowing river flows have declined to a trickle – the
9 The complex nature of water stocks and flows make the definition and interpretation of indices such
as the WSI and the DWSI a difficult topic. We refer to Wada et al. (2011) for the details.
10 In the analysis, only water demand for livestock is considered and it turns out to be only 1–2 percent
of total water demand. It, therefore, has supposedly only a minor impact.
11 Some water-scarce countries are diminishing the problem by importing food and thus, indirectly,
water. It is called ‘virtual’ water and is calculated in the form of the water footprint. It is only a
solution for countries that have the means to pay for the food imports, such as the oil-rich Middle
East countries.
0 - 0.2 0.2 - 0.4 0.4 - 0.6 0.6 - 0.8 0.8 - 1
Figure 12.6a. Dynamic water stress index (DWSI) based on monthly water scarcity index ≥0.4 over the period 1958–2001 (Low: 0–0.2, Moderate:
0.2–0.4, High: 0.4–0.6, Very high: 0.6–0.8, Extremely high: 0.8–1.0) (Wada et al. 2011). (See color plate.)
395
396
(a) 0-2 2 - 20 20 - 100 100 - 300 300 - 1000 (b) No Data 0-2 2 - 20 20 - 100 100 - 300 300 - 1000
(c) No Data 0-2 2 - 20 20 - 100 100 - 300 300 - 1000
Figure 12.6b. Groundwater depletion in the regions of the United States, Europe, China and India and the Middle East for the year 2000 (mm ·
a−1; clockwise from top left). (Wada et al. 2010; reproduced/modified by permission of American Geophysical Union). (See color plate.)
Box 12.3. Groundwater resources. Groundwater stems from precipitation,

which infiltrates the land surface and then percolates through a zone of unsatur-
ated water towards deeper, water-saturated layers. The two are separated by the
water table. Soil moisture in the unsaturated zone is important in the transport,
conversion and storage of dissolved substances and is one of the determinants
of water quality. Use of agrochemicals, for instance, can cause water pollution
that takes sometimes up to a century to fall to original levels. If the rock pores
in the layer below, the water table is saturated with water and the water can flow
under natural or applied pressure; it is a groundwater reservoir or aquifer. There
are different kinds of aquifers and they are usually exploited by drilling wells.
If the water has been sealed for very long periods, the aquifer contains fossil
groundwater and is effectively a non-renewable resource.
Development and use of groundwater is sustainable when abstraction can be
maintained indefinitely without causing unacceptable environmental, economic
or social consequences and risks. Stated in system terms, the groundwater out-
flow rate (natural drainage and withdrawal by humans) should not exceed the
inflow rate (rainfall, rivers or glaciers). Often, only the groundwater level in vari-
ous places can be directly measured and water managers rely on ‘black box’
approaches that correlate water levels to rain patterns and withdrawal rates.
satellite photos of the Aral Sea and Lake Chad tell a compelling story. With sovereign
nation-states still the major political unit, the use of river water in river basins such
as the Tigris-Euphrates, the Mekong, the Nile and the Colorado is bound to cause
allocation disputes (§5.3).
Although water withdrawal rates are relatively small in comparison with the
total water stocks and flows in the biosphere, abstraction and diversions cause
changes in the local and regional landscape and ecosystems. In combination with
other changes such as deforestation, increasing population densities, urbanisation
and climate change, it can have serious consequences for human habitation. Not
surprisingly, water is a central concern in the assessment of climate change impacts
and adaptation measures (IPCC 2007). Many regional water systems are exploited
unsustainably. It compromises ‘the ability of future generations to meet their own
needs’, in the WCED-definition of sustainable development.
There is a variety of solutions that are available and are increasingly implemen-
ted (Stikker 2007; Gleick and associates 2011). First, there is a large potential to use
water more efficiently, with techniques like drip irrigation, water-saving toilets and
showers, and recycling. Rainwater collection can bring relief in semi-arid regions –
installation is already obligatory, for instance, in new housing in parts of India.
Another option is to expand supply by seawater desalination. It is still energy- and
capital-intensive, but innovative designs are being tested and it seems eminently
suitable for use of solar energy. It will not bring relief everywhere, because water
transport over long distances is rather expensive and impractical. The implement-
ation of these options will be accelerated when water is priced more correctly, but
regulation and allocation schemes must be part of the solution as well. With growing
demand, water as a commons will thus slowly disappear and become part of the
market system.
12.3.2 Water for Irrigation: A Case Study and a Model

Because the water situations in the world are so diverse, there are lots of anecdotes
and stories but only few commonalities. For further inspection, the following is an
analysis of groundwater use in Spain (Fernández and Selma 2004). Many landscapes
in the Mediterranean are characterised by traditional irrigation systems along river
valleys, from which they take surface water. Since the opening of European and
global markets for Spanish producers, large-scale intensive agriculture has been
expanding. In the newly irrigated lands in the Segura basin in southeastern Spain,
the area of irrigated lands has increased between 1960 and 2000 from 1,200 hectares
(ha) to over 17,000 ha. Groundwater levels have on average fallen 15 to 20 metres in
this period. Overabstraction has resulted in seawater intrusion and water salinisation
and in outflow of nitrates and phosphates.12 The outflow from natural springs in these
four decades is reduced with a factor of ten, causing a decline in biodiversity. Valu-
able habitats as well as future profitability are threatened. ‘In south-east Spain, most
of the aquifers are over-exploited . . . The whole process clearly fits the desertifica-
tion syndrome . . . and constitutes a paradigm of environmental and socio-economic
unsustainability . . . the present total water demand, mostly originating from agri-
culture, amounts to 228% of the internal renewable resources of the Segura basin’
(Fernández and Selma 2004). In short, it is unsustainable.
The decline in size and quality of the local groundwater reserves from over-
exploitation is being countered by importing water from neighbouring groundwa-
ter reservoirs, re-use of wastewater flows and seawater desalination plants. The
hydraulic infrastructure (dams or river diversion) is extended and innovations (bet-
ter crop varieties, computer-aided fertilisation and irrigation) are introduced, but
these responses and the associated policies are controversial. A system dynamics
model has been constructed and implemented to examine the (ground)water policies
for the area (Fernández and Selma 2004; Figure 12.7). It has five sectors: irrigated
lands, profitability, available area, water resources and pollution. The crucial state
variables are the area of irrigated lands (tree crops, open air and greenhouse horti-
cultural crops) and the local groundwater resources (thirty-nine small aquifers). The
most important development loop is the conversion of drylands and shrublands into
irrigated lands (Figure 12.7 – bold arrows). This was considered profitable when it
began in the 1960s. Water demand started growing and groundwater withdrawal
caused declining water levels and rising pumping costs. It tended to decrease profit-
ability, in turn slowing down new irrigation projects. Later, also the decrease in
suitable lands and the environmental impacts put a brake on further development.
The model has been calibrated with historical data and ‘tells’ the water history
of the area. It is applied to explore three scenarios for the future: one scenario of
trend extrapolation, one with no increase in external water resources, and one with
some increase and technological intensification. The researchers conclude that the
currently implemented plans may well aggravate the unsustainability of large-scale
irrigated agri/horticulture, with a significant drop in profitability over the next two
decades. It is the lack of a systemic approach of the underlying driving factors that
leads to misconceived policies. The construction, application and lessons of models
12 In the Segura basin, soil salinisation is not affecting crop yields and profitability, because 80 percent
of irrigated lands use artificial substrata which are replaced every two years. Like the nitrates and
phosphates, the salt ends up outside the system – an externalisation of potential costs.
Figure 12.7. Simplified causal loop diagram of the large-scale projects on new irrigated lands in southeastern Spain. State variables are grey
(based on Fernández and Selma 2004).
399
such as this one is a necessary step in development projects if the scarce water
resources are to be used sustainably.
12.4 Stories
12.4.1 The Canadian Fish Drama

Northeastern Canadian fisheries – cod, haddock and other species – have been in
serious crisis since the 1990s. It is ‘the classic case of the failure of conventional
science-based fisheries management: the collapse of the northern cod of Newfound-
land and Labrador’ (Finlayson and McCay 1998). For centuries, the extraordinary
abundance of cod has been exploited by European settlers. It was done in the
summer months – storms and ice made it impossible in the winter months – with
relatively simple and inexpensive techniques. Cod fishery was the focal point of the
small coastal communities, which tended to have egalitarian social relations.
In the 1960s, the situation changed with the arrival of foreign trawlers, which
itself was driven by hungry post-war European and Russian populations and new,
cheap techniques such as the devastatingly effective factory freezer trawler. Fishing
now also happened during winter months by fleets that stayed all year round at
sea and were directed by their corporate or state owners wherever catch rates were
highest: ‘By the mid-1960s their numbers were so great that the Newfoundland fishing
banks at night were described as a “city of light”’ (Finlayson and McCay 1998). The
total catch went up from 300 kilotons per year (kton/yr) in 1956 to 783 kton/yr in
1968. Distant water fishing increased most rapidly. During these years, there was
minor and ineffective regulation because of a ‘lowest common denominator’ policy
and restrained enforcement power. It led to massive overfishing and collapse – total
catch precipitously declined to 288 kton/yr in 1975. A pragmatic sustainable yield is
estimated to be 200 kton/yr.
Local communities declined rapidly affecting some 35,000 people and inducing
large recompensation sums, and massive government funds went into restructuring
the industry after the collapse. This tragedy of an international, open-access com-
mons unfolded until the Canadian government closed the northern cod fisheries in
1995 for probably several decades. In retrospect, the scientific assessments of the
resource turned out to be wrong. Knowledge of the local fishermen was neglected.
Political and ideological factors played a role: ‘Scientists knew the truth but were not
heard or not allowed to speak because those charged with making fisheries policy
had reasons to favour more generous assessments’ (Finlayson and McCay 1998).
Bureaucratic inflexibility, public relations, a managerial industry perspective and
awkward relations between scientists and policymakers all were other factors. Did
anyone learn lessons? Possibly not, because most attention has gone to the crises,
not to the underlying causes.
12.4.2 European Union Fisheries Policy in Senegal13

‘For us, it has no sense or benefit because the industrial fishing boats don’t leave us
any chance of survival. They fish right up to the coast without being stopped and
13 This story is based on UNDP Human Development Report 2005 Occasional Paper (hdr.undp.org).
the government doesn’t have the means to control their activities. If the government
would listen to us, we wouldn’t sign an agreement with people who catch everything,
even the small fish,’ says Mario Alberto Da Silva, a West African artisanal fisher-
man. The fisheries sector is an essential component in Senegal’s development: It
provides 75 percent of local protein needs and generates about 100,000 direct jobs
for Senegalese nationals, of which more than 90 percent are in small scale (artisanal)
fishing. Another 600,000 people or 15 percent of the working Senegalese popula-
tion are employed in related industries. The rising global demand for fish combined
with pressures on world supplies means that Senegal’s ‘blue gold’ is an increasingly
valuable resource.
The UN Convention on the Law of the Sea (UNCLOS), signed in 1982 and in
force since 1994, provides for an Exclusive Economic Zone (EEZ) of 200 nautical
miles. This places 95 percent of the world’s fish stocks and 35 percent of the oceans
under the jurisdiction of coastal nations. The open access previously enjoyed by the
long distance fishing fleets was lost, so the European Union (EU) had to negotiate
access through fisheries agreements or through private license and joint venture
arrangements. Already in the 1980s, the Senegalese government first reduced sup-
port for the fisheries sector and then introduced export-stimulating mechanisms as
part of an agreement with IMF-World Bank. Since then, it has concluded a num-
ber of fishing access agreements, principally with the EU, that have a significant
contribution to government revenues. Like many African governments, short-term
financial compensation from fisheries access agreements was favoured over a thriv-
ing informal domestic sector from which it is difficult to extract revenue. Fish exports
have grown and the revenues have been an important source for debt repayments.
However, the value of this sector is being eroded by multilateral trade liberalisa-
tion. The pressure on West African fish stocks increased six-fold between the 1960s
and the 1990s, mainly as a result of fishing by the EU, Russian and Asian fleets. The
small-scale fishing industry of Senegal is now in direct competition with the fishing
fleets of the EU to supply both the local and the EU market. The risk of supply short-
ages and price increases on local markets looms ahead as fishing efforts shift from
locally consumed to export-oriented species. The EU has been strongly criticised
for its role in the depletion. It ‘exports’ EU fishing fleets to already resource-scarce
regions and partly subsidises these vessels. It concludes intransparent agreements
with signatory states who have quite limited capability to monitor or control the
EU fleet activities. Local interests suffer from its lobbying power. The fisheries
policies are in conflict with EU development policies in West Africa. To address
these criticisms, the EU has developed a new approach to Fisheries Agreements.
The latest agreement with Senegal (2002–2006) made some improvements but is still
an old-style ‘cash for access’ agreement.
12.5 Fisheries and Forests
12.5.1 World Fisheries

Fisheries and fishery management are amongst the best researched renewable
resource topics, which does not prevent their near-depletion. Fish is an import-
ant source of high-quality food for millions of people and has been so for a long
time. Originally, most fish was harvested from inland surface waters (rivers, lakes,
floodplains or lagoons). In high-income regions, fishing ‘in the wild’ is only for recre-
ation, but elsewhere it is still of direct importance for people’s livelihood. Inland
fisheries still provide an estimated 10 million ton/yr and possibly more because
of underreporting and illegal catch. The catch is a valuable source of protein-rich
food and employment. Governments tend to stimulate a transition to aquaculture
in private ownership because ‘governments and resource developers see inland fish-
eries as an impediment to their desires to expropriate the wealth of the rivers –
the transfer of generalized wealth (nutritional security, livelihoods) from powerless
people into focused income streams that benefit powerful people’ (Welcomme et al.
2010). As with pastoralists, there is a tendency of the state or those representing it
to ‘tame the wild’ and exert control according to universal principles (Scott 1998). It
is a process in which a common pool resource gradually comes under the influence
of government institutions and market incentives, in order to satisfy a growing food
demand through intensification.14 But a transition is not without problems because
aquafarming requires inputs and physical and institutional infrastructure, produces
different species and demands a different lifestyle of the farmer.
By far the larger part of aquatic food is nowadays coming from marine and
coastal fisheries. World capture of fish increased from about 20 million ton per year
(Mt/yr) around 1950 to 144 Mt/yr in 2006, including the 10 Mt/yr from inland fisheries
(Figures 4.6b and 12.8a). Increasingly, fish is produced from the high seas and from
coastal and inland aquaculture. The latter has spectacularly risen from a few Mt/yr
in 1985 to 52 Mt/yr in 2006, with a dominant role for Chinese aquaculture. Besides,
there is an estimated 11–26 Mt/yr illegal, unreported and unregulated catch and
9.5 Mt/yr discarded catch. Global fisheries contribute only a few percent to gross
world product (GWP), but directly and indirectly (vessels, processing and so on)
they provide employment and income to an estimated 42 million people (Garcia and
Rosenberg 2010). The sector as a whole is said to support more than 500 million
livelihoods. According to the FAO, fish contributes almost 20 percent to average
per capita intake of animal protein for more than 1.5 billion people, many of them
in low-income food-deficit countries (WRI 2000–2001). In some countries, fisheries
contribute more than 7 percent to GDP and are essential for poor segments of the
population.
Fisheries exploitation happens in stages, from undeveloped to developing to
mature and finally to senescent which is a state of overexploitation, depletion or
recovery. The statistics for 169 national fishing areas indicate that 27 percent is over-
exploited or depleted with respect to the state in which maximum sustainable yield is
supported (Garcia and Rosenberg 2010). Nearly 60 percent of fish landings are esti-
mated to come from fully/overexploited stocks (Figure 12.8b). There is widespread
agreement amongst experts that marine fisheries are at or near the upper produc-
tion limit of an estimated 80–125 Mt/yr (Garcia and Grainger 2005). This level can
probably not be sustained. To increase aquatic food output, the rapid growth in fish
from aquafarming and aquaculture must continue (Figure 4.6b). But overall system
efficiency is low, because the fish has to be fed, usually with seafish. The use of
chemicals to fertilise the ponds and protect the fish against diseases has already led
14 Intensive aquafarming in ponds can deliver up to 1,000 times higher yields than in natural production
situations (Welcomme et al. 2010).
160
140
120
million tonnes
100
80
60
40
20
0
50 55 60 65 70 75 80 85 90 95 00 06
(a)
1950–2006
senescent
mature
169 areas
developing
(b)
Figure 12.8. Trends in global fish catch (Garcia and Rosenberg 2010). (See color plate.)
to health and pollution risks and epidemic contagion of wild fish (Ford and Myers
2008; Arquitt et al. 2005). Sustainable aquafarming will be one of the challenges for
the future.
Fisheries depletion is not only a matter of quantity but also of quality. Food
web analysis of fisheries suggests that fishing grounds are ‘fished down’ in a double
sense. First, average depth of fish catches has increased from 150 to 200 metres (m)
to about 250 m with the shift to deep sea fishing. Second, the mean trophic level has
decreased in most coastal areas and the North Atlantic ocean (MEA 2005). Analysis
of catch data for the western central and north Atlantic and the southeastern and
northwestern Pacific reveal that since 1970 the mean trophic level of fish catch has
been declining in all of these fishing areas. This fishing down of marine food webs
is the result of sequential depletion of a fishing ground by catching species lower
down the food web (intensive or ‘deeper’), not by moving to new fishing grounds
(extensive or ‘broader’) (Pauly and Palomares 2005). But the empirical evidence for
this is still contested and more research is needed.
The exploitation mechanism in fisheries is not different from other resource sys-
tems. Fishing power, defined as the product of number of ships and potential catch
per ship, has increased sixfold between 1970 and 2005 and, in combination with
Box 12.4. Depletion and technology: Long-term trends. In a recent analysis

of UK fisheries, Thurstan et al. (2010) have compiled data on bottom-living fish
species landings into the United Kingdon from 1889 to 2007. They also estimated
total fishing effort, taking the innovation waves from sail ships to steam trawlers
and then, from 1950 onwards, to motor trawlers into account. The data show a
rapid increase in landings between 1890 and 1915, a further increase until 1950
followed by an accelerating decline. Around the year 2000, landings decreased to
half the 1890 level.
After factoring out the innovations introduced to increase catching power, such
as better gear and nets, it is possible to construct a reliable, long-term indexed
productivity of fishing activity measured as catch per unit of effort. It turns out
that between 1890 and 1920 productivity in ton per unit of fishing power dropped
dramatically, but the increased landings masked the steep decline in fish stocks.
Productivity doubled again between 1920 and 1960, as fishing vessels exploited
new grounds in the Arctic and West Africa. After 1960, a precipitous decline in
productivity set in and reached unprecedentedly low levels by 1980 – a thirtyfold
decline since 1889 (Figure 12.9).
Thus, these long-term empirical data confirm theoretical analyses that pred-
atory fish biomass is currently at most 10 percent of pre-industrial levels, and
that productivity has gone down so much that ‘for every unit of fishing power
expended today, bottom trawlers land little more than one-seventeenth of the
catches in the late 19th century’ (Thurstan et al. 2010). It is a classical tragedy of
the commons story and urgent action is needed to eliminate overexploitation of
European fisheries and rebuild fish stocks.
increasing demand, led to a relentless search for new fishing grounds and to their
subsequent depletion. The tendency to overinvest in capacity to exploit renewable
resources is found time and again.15 One probable cause is the misperception of
feedbacks and delays in the system (Moxnes 1998). In a broader context, the system
is seen to be fatefully drawn into the capitalist logic of catching as much fish as fast as
possible and putting the money into a bank account or, better even, investing it in still
profitable fisheries or other resource extraction business elsewhere. Are scientific
models useful for sustainable management? Not that much, it seems. Already for
decades, science-based regulation with quota and other measures are introduced and
implemented, but the negotiations and enforcements are slow and tedious and meas-
ures are often too late. Yet, coordinated regulation along the previously mentioned
CPR management principles is the only road towards sustainable exploitation.
12.5.2 Fisheries Models: Strategies and Interactions

The past century has taught that scientific advise in fisheries management has not
led to sustainable harvesting. Apart from the difficulty to adequately measure fish
15 The game Fish Banks (Meadows 1996) convincingly makes the point that sustainable exploitation
of open-access fisheries need some kind of coordination.
LPUP trawlers England and Wales Landings per unit of fishing effort England and Wales
70 40
60 35
30
50
Landings/unit/yr
Landings/unit/yr
25
40
20
30
15
20
10
10 5
0 0
1885 1905 1925 1945 1965 1985 2005
Year
Figure 12.9. Productivity in UK fisheries 1889–2007. Landings of bottom-living fish per unit of
fishing power of large British trawlers. Closed circles show landings per unit of fishing power
(LPUP) into England and Wales, open circles show landings per unit of fishing effort of large
British trawlers (corrected for changes in fishing power) into England and Wales (Thurstan
et al. 2010).
stocks, there is the problem that most scientific models tend to simulate the ecosystem
and then tell what is rational and optimal to do. But humans are as much part of
the resource system as the fish, and what is rational and optimal for the individual
fishermen tends to be irrational and suboptimal for the system at large – the essence
of a social dilemma. For instance, in the Nova Scotia Groundfish Fisheries, only
for one of three fish species – haddock – has a relation between catch and effort
has been observed (equation A12.2). Why? It turns out that including the well-
known fact that the birth rate of haddock is extremely variable, fish populations can
suddenly start to fluctuate wildly instead of the smooth equilibrium behaviour in a
deterministic simulation (May 1977; Allen and McGlade 1987). Apparently, human
exploitation can amplify the natural fast fluctuations and, therefore, resource systems
with a natural background noise can probably never yield a constant and satisfactory
economic return.
One way to improve the models is to introduce and simulate explicitly the beha-
vioural strategies of resource exploiters. An early example of such an agent-based
model (ABM; §10.5) is the investigation of two possible fishing strategies in an
open-access, spatially distributed resource area (Allen and McGlade 1987). Let us
suppose that there are two strategies, one that disregards catch data from boats and
chooses randomly where to fish, and the other that decides where to fish on the
basis of differences in (expected) profitability and can be considered an optimising
‘ultra rationalist’ skipper. The latter are rational optimisers cartesians. The former
are opportunists stochasts. Using an evolutionary selection mechanism (Appendix
10.2), the risk-averse cartesians go fishing where they know it to be profitable and
then deplete the zone and move on, whereas the risk-taking stochasts exploit in more
or less random fashion also new territory. The two strategies represent a trade-off
between efficiency and innovation, hence their names. Normally, the stochasts out-
compete the cartesians because they map the resource more quickly and completely,
and find more fish although they use information less efficiently. But the cartesians
can develop novel strategies in response and complex patterns may evolve.
From a sustainable development point of view, the lesson is that in real-world
situations resource use strategies can be expected to fluctuate between random beha-
viour with long-term benefits (knowledge or innovations) and efficient behaviour
with short-term gains (profits or stability). The resulting exploitation cycles can be
interpreted in favour of stochasts: ‘The model shows the opposition between short
and long term decision making . . . and the importance of management principles
which would maintain diversity . . . If we are to avoid a future of ferocious and ever
growing competition, in a shrinking world with a single perspective and the common
values of a single culture, then we must encourage “stochasts”, and the diversity and
expansion which only they can bring’ (Allen and McGlade 1987). The difference
between stochasts and cartesians echos the difference between the individualist and
the hierarchist cultural perspective (§6.3) and shows that multiple interpretations of
depletion and, therefore, divergent recommendations for management can be valid.
Significant advances in resource modelling have been made since this seminal
paper (Garcia and Charles 2007; Little and McDonald 2007). One extension of the
behavioural fisheries model is to introduce different fishing zones with different
carrying capacities and growth rates and several fishing strategies (Brede and de
Vries 2009b). What happens if a crowd of agents or fleets fish around in a few
separate, open-access fishing grounds without centralised coordination? Suppose
that you are a fisherman, then you can behave as a stochast, but there are other
strategies. For instance, if many people go fishing in the high-yield area A, you
imitate and join. If everyone does the same, A will soon be overexploited and yields
will decline. Some people start moving to area B, and the same will happen there.
This kind of crowd behaviour is known as the minority game (MG).16
An ABM has been constructed with this minority game and three other strategies
that can be followed by individual fishermen in order to explore the role of different
strategies for the exploitation history (Appendix 12.3). A series of model simula-
tions has been performed to see under which conditions strategies survive in an
evolutionary competition. If the harvesting pressure increases, the average catch
per vessel declines (Figure 12.10a) and the number of vessels following a particu-
lar strategy changes (Figure 12.10b).17 Which strategy becomes dominant depends
on the harvesting pressure. If harvesting pressure is low and resources are abund-
ant, the best performing agents are the stochasts, such as the ones that follow an
uncoordinated chance-based harvesting schedule (RAND agents; Figure 12.10a).
When the harvesting pressure increases, the the more cooperative COINS-strategy
starts to dominate in the agent population and the MGS-strategy also becomes more
16 The minority game was originally proposed to simulate behaviour in stock markets.
17 Harvesting pressure is calculated as the maximum possible catch for the given number of vessels
divided by the recovery period. It increases for shorter recovery periods. It is an equivalent of
resource scarcity or abundance.
max
RAND COINL
avg. catch per vessel

MGL
Figure 12.10a. The average catch per vessel

declines with increasing harvesting pressure. MGS,
Long-term (L) strategies perform better than COINS
short-term (S) strategies (Brede and de Vries
2009b). TGS
0
very low low overharvested severely
overharvested
harvesting pressure H
(a)
effective. In severely overharvested situations, the team game strategy (TGS) gives
the best results. The agents are assumed to have a short time horizon (S). A refine-
ment of the model is to introduce agents who make a long-term (L) projection
on the basis of their knowledge of the resource. The long-term strategies perform
consistently better than the equivalent short-term strategies (Figure 12.10a). The
COINL-strategy is now superior as long as the resource is not overexploited, but
beyond a certain harvesting pressure, some of the distributed resources are occa-
sionally overharvested and the minority game (MGL) strategy rises to dominance
(Figure 12.10b).
The simulation results suggest that the exploitation strategy co-evolves with
the intensity of exploitation. They also point at the inevitability of coordinated
monitoring and regulation if resource use becomes more intense. A cooperative
and long-term oriented strategy such as COINL emerges as the most sustainable
60 60
COINS COINL
MGS MGL
50 TGS 50 TGL
RAND
40 40
vessels
vessels
30 30
20 20
10 10
0 0
1,000 1,030 1,060 1,090 1,000 1,030 1,060 1,090
harvesting pressure H harvesting pressure H
(b)
Figure 12.10b. The average number of vessels that follow one of the strategies, as a function
of the harvesting pressure. The coordinated long-term (COINL) strategy is the most attractive
one as long as the resource is not overexploited (Brede and de Vries 2009b).
strategy: it relieves the pressure on the resource, reduces fluctuations and diminishes
the risk of overharvesting. Therefore, it can reconcile group interests and individual
interests. An interesting question is whether a population discovers such a COIN
strategy in time to prevent the collapse into the selfish (MG) and poverty (TG)
strategies. This brings in the question of regulation and policy.
Government policies have long influenced fishermen’s strategies. A comprehens-
ive fishery management system dynamics model has been constructed to explore the
role of government interference in the United States (Dudley 2008). It has three sub-
models. The first one is the ecosystem model based on a somewhat refined logistic
harvest function (equation A12.2).18 The second one is the investment model. If
the actual productivity, defined as catch per unit of effort, exceeds the desired or
target value, new ships are ordered. Otherwise, ships will leave the area in search
for more profitable fishing grounds but, before leaving, fishermen will attempt to
increase catch efficiency in order to sustain productivity. The third model simulates
lobbying and regulatory policy that affects the incentives and strategies chosen by
the fisheries managers.
The most important finding is that, in an open-access non-managed situation, an
oscillation in stock and catch (‘overshoot and collapse’) occurs with a frequency of
three cycles per century for a reasonable model parameterisation (Figure 12.11). In
the low-level periods, a large part of the stock addition consists of newborn fish and
this makes the system vulnerable to ecological and climatic fluctuations. This result
resembles the outcome for a minority game (MG) strategy and reflects the empirical
findings for the United Kingdom.
Second, full implementation of the management goal to regulate the number of
vessels on the basis of productivity gives in the long term the highest fish biomass
stock and income per unit of fishing effort. This more rational strategy resembles
the COIN strategy and is also the more sustainable strategy. It does not exhibit
boom-and-bust oscillations.19 However, the catch value in $/yr is lower than for a
less well-managed, short-term–oriented strategy. A worrying result of the model
experiments is that stock levels tend to decline to much lower levels than neces-
sary, if lobbying and political influences disrupt full implementation of management
goals. Unfortunately, this is what may have happened, at least in the United States:
‘Interpretation of what’s going on in fisheries has become very complex, not in the
least due to management failures, and, in response, more and more differentiated
regulation as well as litigation and non-compliance’ (Dudley 2008).
12.5.3 Fisheries Models: Behavioural Variety

Some researchers have introduced more complex agent behaviour than in the pre-
vious models. An example is the consumat model (Jager et al. 2000). The décor is
Lakeland, the imaginary island in an earlier section. The Lakelanders are endowed
18 The reproductive rate α is assumed to be a function of the biomass R and biomass growth is separated
into growth of existing stock and of new fish (recruitment). Besides, the ecosystem carrying capacity
K will drop in proportion with the cumulative fishing effort so the ecosystem will recover, but slowly
and non-linearly.
19 The insertion of random fluctuations in the model causes boom-and-bust behaviour even in a fully
managed fishery, in line with previous findings.
Catch, Biomass, CPUE, & Fishing Units

20,000 t/Year
100,000 t
200 t/(Year*units)
400 units
Fishing gear
15,000 t/Year
75,000 t Catch
150 t/(Year*units)
300 units
Stock
10,000 t/Year
50,000 t biomass
100 t/(Year*units)
200 units cpue
5,000 t/Year
25,000 t
50 t/(Year*units)
100 units
0 t/Year
0 t
0 t/(Year*units)
0 units
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Time (Year)
Figure 12.11. Trajectory of simulated fish biomass stock, catch, fishing gear and productivity,
calculated as catch per unit effort (cpue), for an unmanaged fishery when the cpue is sufficiently
low to cause overfishing to happen (Dudley 2008).
with different preferences and behaviours and are called consumats. They have
a number of stylised characteristics that are based on findings in economics and
psychology. In summary form, these characteristics are:
r they are equipped with certain abilities for fishing and mining and they allocate
their time to leisure, fishing and mining according to their individual preferences;
r they spend the money earned from gold mining on fish imports and on non-fish
consumption ‘luxury’ goods – the opportunities;
r their behaviour as producer (ability) and consumer (opportunity) depends on
their level of satisfaction and level of uncertainty.
The last point is important: Each consumat only behaves economically rational if
he is not satisfied and certain. This behaviour is called deliberation (reasoned or
individual) and is characteristic of the homo economicus (§10.4). But agents can
also behave differently: imitation (automated and socially determined) if satisfied
and uncertain, social comparison (reasoned, socially determined) if unsatisfied and
uncertain, and repetition (automated, individual) if satisfied and certain (Figure
12.12).
A consumat updates every timestep his memory with information on his own
and other agents’ performance. When he decides about job and spending according
to the decision scheme in Figure 12.12, it results in certain levels of needs satisfaction
and uncertainty. The needs considered are proxies for leisure, identity, subsistence
Figure 12.12. The agent model for consumats (after Jager et al. 2000). The level of needs
satisfaction and of uncertainty are important contextual determinants of people’s behaviour.
and freedom.20 The satisfaction of the need for leisure is related to the share of
the time spent on leisure and for identity to the amount of money the consumat
owns in comparison to other consumats with similar abilities. Needs satisfaction for
subsistence depends on fish consumption; for freedom, it is a function of the available
money. What can we learn from this model about sustainable development?
Sixteen consumats can satisfy their personal needs by exploiting two natural
resources: fish and gold (Figure 12.1). Which strategy a consumat chooses depends
on present and past levels of need satisfaction and uncertainty. If not satisfied and
certain, he or she aspires for more consumption and is certain enough to show
reasoned behaviour. If fully satisfied and highly uncertain, he or she will resort to
automated socially determined behaviour. This is similar for the two other cognitive
processes. Development in Lakeland happens because agents switch strategy if their
performance falls below the expectation they build up in their memory. The con-
sumats are in a social dilemma situation: Each consumat is inclined to get as much
income at the lowest effort possible, but in doing so they may collectively destroy
the option to sustain income and quality of life for the long term.
Let us look at some illustrative outcomes of this model if fishing is the only
option. To keep the analysis transparant, two archetypical consumats are defined:
the Homo economicus (He) and the Homo psychologicus (Hp). The former favours
deliberation, because he operates with high levels of need satisfaction and uncer-
tainty reduction. The latter is quickly satisfied and an uncertain imitator. If fishing
is the only option, the He use deliberate rationality to increase their income and
as a result the fish stock is depleted before year forty (Figure 12.13a). The Hp, on
the other hand, are quickly satisfied and do not fish more than needed. Most of
them spend their time as happy, lazy, poor repetitors and imitators, until the initial
budget has been fully spent, in year eighteen, and many of them become unsatisfied
and switch to deliberate behaviour in order to restore income (Figure 12.13b). As
a consequence, the fish stock remains at a high level (Figure 12.13a,b). The two
20 These needs are supposed to play an important role and are selected out of a larger set of needs, as
described by Max-Neef (1991) (§6.2).
1
25,000 repetition
0.9 deliberation
0.8 imitation
20,000 social comparison
proportion of time
0.7
0.6
fish stock
15,000
0.5
10,000 0.4
Homo
0.3
economicus
5,000 0.2
Homo
0.1
psychologicus
0 0
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48
time time
(a) (b)
30
Homo psychologicus
Homo economicus
25
20
Financial bud
15
10
0
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48
time
(c)
Figure 12.13a-c. Illustrative pathways in a fishing-only no-gold-mining consumat world for
the Homo economicus (He) and Homo psychologicus (Hp) in the first 50 timesteps (averages
for 100 simulation runs) (Jager et al. 2000). (a) The fish stock, (b) the proportion of time spent
in one of the four strategies (c), the financial budget.
consumat-types have a very different financial budget profile and in the He-collapse
period there are large income differences amongst the agents (Figure 12.13c).21 If
uncertainty is completely removed, the He does not overharvest and the Hp spends
even less time fishing. This is in agreement with the finding that uncertainty tends to
cause resource overexpoitation (May 1977).
If the consumats can engage in gold mining as an additional source of income, a
transition happens from a fishing into a mining community for both He and Hp. Both
do pollute the lake, with a negative impact on the fish stock (Figure 12.13d). The He
are keen to grasp the new opportunity and the fish stock is less exploited. The Hp
switch at a slower rate to mining, suffer more from pollution impacts and are forced
into an accelerated transition to mining. One sees here two different development
patterns with the same outcome: A growth in income (from mining) takes place at
the expense of the environment (pollution or depletion).
Discovering and implementing labour-saving opportunities is what innovation
and economic development are about. So what happens if a deliberating consumat
discovers a new fishing opportunity and increases his productivity (fish caught per
hour)? Nothing happens in the Hp world, because he or she is easily satisfied (is
21 Income differences are explored in more depth by giving the agents different abilities.
0.000004 Homo psychologicus

Homo economicus
concentration (kg/m)
0.000003
Figure 12.13d. Illustrative pathway of the pollut-

0.000002 ant concentration in the lake in a fishing and gold-
mining consumat world for the Homo economicus
and Homo psychologicus in the first 50 timesteps
0.000001
(averages for 100 simulation runs) (Jager et al.
2000).
0
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48
time
(d)
she lazy?) and does not engage in deliberation (is he stupid?). In a world populated
by He, the innovation makes one individual more productive than others and it
rapidly spreads in this competitive and entrepreneurial world through imitation. As
a consequence, the fish stock is depleted faster than without the innovation. This is
precisely what happened when fishing boats became more effective. It illustrates the
ambiguity of technology: Innovations in resource exploitation are both cause and
solution of sustainability problems. Of course, an illustrative model like this has to be
backed up by empirical fieldwork and to be extended with more realistic behaviour
(Janssen 2002; Perez and Batten 2003; Walker et al. 2006).
12.5.4 World Forests

Reconstructed maps of Europe show the enormous decline of forested areas since
medieval times. The more recent large-scale deforestation in parts of Canada and
the United States, in Argentina and Costa Rica, in Indonesia and in other parts
of the world has also been reconstructed and shows an acceleration in the rate at
which human populations deforest the planet (Turner et al. 1990; Williams 2008).
But in North America and Western Europe, the forested area is again increasing
for decades, as part of what some call the forest transition (§11.4; Satake and Rudel
2007).
Forests provide already for millennia important products such as fuelwood for
cooking and heating, timber for construction and charcoal for industrial processes.
Currently, wood is an important commodity and traded worldwide. Of late, global
wood use appears to stabilise for various reasons (Figure 4.5). Besides, forest produce
is a very diverse source of food and materials for forest dwelling populations. Forest
exploitation has been one of the first situations where the notion of sustainable
resource use emerged (van Zon 2002). It was also one of the first areas in which
enlightened management techniques have been introduced.
The assessment of forest resources, both quantity and quality, is even more
difficult and fraught with definition issues and uncertainties than of water and fish
resources. The major reason is the huge diversity and the complex dynamics of
forests, notably mature tropical forests (Bossel and Krieger 1991). Global estimates
of forest cover suggest an ongoing decline in forested area worldwide. Amongst the
driving forces is the large-scale clearing of forest for the production of soybeans and
Box 12.5. Forests in Japan. Forests in Japan have been under pressure already
for millennia (Totman 1989). More than 2000 years ago, rice culture caused the
first dramatic modifications of woodlands and bronze and iron smelting started to
put pressure on the forest due to the demand for high-quality charcoal. Metallurgy
led to new, more powerful tools and the assault on the woodlands intensified. The
warrior castes needed timber for ships, stockades and large residences and for
coffins buried in huge mounds rivalling the Egyptian pyramids in size. Wood was
also used as fuelwood for pottery and charcoal for weapons. Farmers also needed
wood for fuel but also for fodder and, most importantly, for green fertiliser mater-
ial which sometimes relieved cutting pressure. There was also a large demand for
the construction of monasteries, shrines and temples and, owing to termites and
rot, most wooden buildings had to be rebuilt every 20 years.
The deforestation had all kinds of consequences such as wildfire, flooding and
erosion, often forcing people to move. Kings and emperors often had to move,
possibly because local wood supplies dwindled. There were occasional attempts
to control the use of woodland on the part of governments and monasteries.
Ruling warriors tightened control to assure themselves of resources for military
use. It was a history of outright exploitation without concern for preservation or
reforestation.
When political struggles subsided in the 17th century, population and con-
struction rose rapidly and the demand for timber soared. The demands of the
peasant families led to widespread but less intensive use than the more concen-
trated demands from the rulers in the cities. Logging expanded and intensified,
erosion denuded mountains and damaged lowlands, and Japanese rulers were
forced into a combination of regenerative forestry and imports from the tropical
rain forest in nearby regions, particularly in Malaysia and Indonesia. Like with
food in 19th-century Britain, the pressure on scarce land resources was relieved
by importing resources from abroad (§4.3). Regulation to restrain consumption
was introduced, plantation forestry emerged by the late 18th century and most
forested areas came under some sort of management. It is one of the reasons that
Japan now remains more forested than nearly any other country in the temperate
zone.
for plantations of palm trees (for palm oil) and eucalyptus (for paper). But one has
to be careful with interpretations and conclusions. For instance, there is a rather
vague distinction between old (primary) and young (secondary) forest.
For forests, too, models have been constructed to understand the basic processes
and to design harvesting strategies (Acevedo et al. 1995; Bossel 1996). Of course,
given the variety in forests, the models have different characteristics and aims. They
used to focus on optimal harvesting strategies in the context of forest economics
(Clark 1990). Important variables are the rotation period and, as with fisheries
models, the discount rate. Such a framework had its origin in the hierarchist and
rationalist approach of the Modernist worldview, where the forester tries to shape
the trees according to the desire for control and efficiency. German foresters and
mathematicians even specified a standardised tree (Normalbaum) in an utilitarian
attempt to narrow the view of the forest down to commercial wood (Scott 1998). The
underlying worldview of Modernism fits in with the practise of maximising monetary
benefits from resource exploitation (Figure 6.5).
An example of an economic optimisation model is a global timber market model
by Sohngen et al. (1999) (equation 12.3). The authors calculated the optimal man-
agement of the world’s forests over the period 1995 to 2135. After parameterising
the model for nine world regions, the main conclusion is that only inaccessibility of
forests can protect biodiversity in a globalising market-driven world. Although of
little practical relevance, the model points out that economic rationality in open-
access resource exploitation tends to preserve resources only if the harvesting costs
are infinitely high and exploiters are forced to look for substitutes. Such high cost
occur when a resource is taxed very heavily or inaccessible. (However, what is
inaccessible for modern technology?) It should be added that neither demand nor
substitutes are considered in this model.
The reductionist optimisation approach has less appeal nowadays, with the dis-
covery of the rich biodiversity of tropical forests and the recognition of the wealth
of forest ecosystem services. The existence and complexity of ecosystem services is
only slowly incorporated into management models and practise. A simple way to
account for (the loss of) environmental and other services is to include externalities
in the harvesting model. A combination of consumer initiatives such as ecolabels
and producer considerations such as relative benefits for landowners of forestry and
agriculture can guide exploitation into more sustainable directions (Nilsson 2005;
Satake and Rudel 2007). A more integrated approach is to include topsoils, water
and ecosystems services such as erosion control and the possible occurrence of irre-
versible change due to thresholds explicitly in the model (Mäler et al. 2003; Hein
and Van Ierland 2006). This gives a natural link to biodiversity. Last but not least,
it is important to engage local stakeholders and concerns into the models and the
decisions, because forest exploitation is quite local in its causes and consequences.
This point deserves some special attention and is the topic of the next section.
12.6 Interactive Modelling for Sustainable Livelihood

Many applications in environmental science aim at models being used as decision
support systems (DSS). It can be combined with techniques such as scenario writing
and backcasting, as is done in policy exercises (§8.6; Toth 1988). The first attempts
to construct simulation games in environmental education and policy date from the
1980s. The rapid development in computer technology and software applications
has given a boost to the interactive formulation and use of models in close interac-
tion with stakeholders. The latest development is to run games and models through
the Internet. Modellers and experts interact with participants and stakeholders in
order to formulate a shared problem perception, to understand and communicate
the dynamics of the relevant processes, and to design feasible, mutually beneficial
strategies. The objective is a combination of teaching skills, conveying informa-
tion, increasing motivation and interest, changing attitudes and inducing evaluation
(Greenblat 1988; Bousquet and Voinov 2010). As discussed in §10.4, interactive
modelling can bring together the worlds of science, policymaking and computer
games. There are many initiatives, projects and platforms in simulation gaming and
12.6 Interactive Modelling for Sustainable Livelihood 415
decision making on environmental and sustainability issues, so I refer the interested

reader to the Useful Websites.
A group with longstanding experience is the CORMAS team in France
(www.cormas.fr). The CORMAS platform allows construction of spatially explicit
multi-agent simulation (MAS) or ABMs. This is done in interaction with stakehold-
ers and in combination with other participatory methods like role-playing games.
It has led to their own brand of companion modelling called ComMod (Daré et al.
2009). The basic element in the ComMod approach is experiential learning through
collaboratively tackling a question or an issue, co-constructing a shared represent-
ation (mental map or conceptual model), implementing it in a computer model
and visualising the dynamics, and, finally, collaboratively designing scenarios and
intervention strategies.
It is a learning and communication process between scientists and stakeholders,
during which new forms of knowledge and knowledge gaps come to the fore. The
participants progress from individual to collective mental models and are learning
for action and empowerment, both important aspects in a CPR context (§8.5). There
have been many and varied projects and workshops with the ComMod approach, for
instance, on integrated and shared water management (Brazil); water availability,
migration and (shortage of) farm labour (Laos); participation in how to manage and
develop multiple use regions in wetlands (France); and learning about negotiation
mechanisms (Brazil). It is as yet unclear to what extent people do change behaviour,
once they are back in everyday reality. The experiments with ComMod have not
yet led to general principles of sustainable management, but many data and insights
have been accumulated about the role of context in what people believe and how
they behave (§6.4).
An illustrative application of the ComMod approach concerns a clarification and
search for solutions in a conflict between sheep farmers, foresters and environmen-
talists in the Causse (‘plateau’) Méjean near Montpellier (Etienne et al. 2003). The
problem is that Scots pines and Austrian black pines are encroaching into a natural
ecosystem of high biological diversity. This affects the development objectives of
the stakeholders (sheep production, timber production or nature conservation) in
different ways. An ABM has been developed, using a 200 m × 200 m square grid of
5.726 cells (4 ha each). Biological characteristics are introduced for the initialisation,
formulated in the form of ‘spatial entities’ such as management units making sense
according to some specific perception of the ecosystem, such as already invaded
ridges, farming and woodland areas. Then, three sets of actors are defined, each
with their own strategy vis-à-vis the spatial entities: thirty-seven sheep farmers,
two foresters (one dealing with afforestation, one with the exploitation of native
woodlands), and a national park ranger (the ‘conservationist’).
The model has been used in an interactive decision setup, with an approach
based on four concepts:
r The point of view: the specific way in which each stakeholder perceives resources
and identifies management entities – not unlike the worldviews presented in
§6.3;
r The viewpoint: a spatial representation of a point of view, such as a vegetation
map, a map of tree ages, a map of tree stands or a map of landscape units;
r Indicators: a set of markers selected by the stakeholders to monitor the dynamics

of the system; and
r Scenarios: prospective management rules to tackle the problem at hand (such
as pine encroachment of the Causse).
During these exercises, it became evident that two aspects were particularly effective
in structuring the problem and looking for solutions: sharing experiences and views
of the underlying processes and comparing contrasting scenarios. A major conclusion
was that ‘it helped the participants to understand that all their views were legitimate,
but also subjective and partial’. Another example of an interactive modelling tool is
the Geonamica Dynamic Landuse Planning software package (Appendix 12.4).
12.7 Perspectives on Water, Fish and Forest

Just as with agro-food systems, the interpretation of what is going on and what
should happen with the world’s renewable resources diverge. Listen to the stories
from NGOs:
r the Mau Forest, Kenya’s largest area for water withdrawal, is under threat.
One-third of the 0.4 Mha forest has been cleared in the last decade because of
politicians giving out favours and illegal occupation;
r megacities are badly in need of water management. Yet, in megacities such as
Mexico City and Mumbai, there is hardly any long-term strategy for supply,
reuse and conservation of water;
r the largest rubber plantation, with 8 million rubber trees, is in Liberia. It is
operated by the Japanese/U.S. tire manufacturer Firestone/Bridgestone. Labour
conditions are bad: $4/day payment, use of pesticides (www.stopfirestone.org).
The following are the voices from the pragmatic world of engineers and politicians:
r ‘As long as the farm breaks even on a cash-flow business I’m OK’, the farmer
in Arizona said. ‘But my retirement is not such that I could suck up the water
bill if it keeps rising, and we still have a loan on the land. Wouldn’t take long to
suck all the retirement into that just to keep it’.
r ‘If we don’t do something, the most expensive part of Jakarta will disappear
into the sea’, says a director at the Agency for Development and Planning.
‘We investigate a 30-km dike in the sea, but it is the most ambitious option’.
Jakarta sinks every year about 10 centimetres (cm), mostly due to groundwater
pumping. The nineteen-year La Niña will, in combination with sea level rise, hit
hard again in 2025.
r ‘The economy [in Liberia] has always been associated with forest exploitation. I
do not say that REDD (Reducing Emissions from Deforestation and Degrada-
tion) is not a good strategy, but when you have to survive from day to day, your
first thoughts are not about climate change’. Can a carbon tax help to preserve
the forest in Liberia?
But there is always opportunity and hope:
r In 2009, a large investment fund advertised: ‘Water will become scarce. So it
will give many opportunities all over the world. Invest with Robeco’. Water is
a ‘growth market’ and investors and speculators are becoming more interested.
There are many water funds, and some note returns on investment of more than
30 percent:
r The World Wildlife Fund (www.wwf.org) works with corporations such as Uni-
lever in several countries on the introduction of ecolabels in order to promote
sustainable fishing.
r ‘L’Europe est sur le point de fermer ses portes aux entreprises qui prospèrent
sur la destruction des forêts’, says a Greenpeace activist in Brussels. Between
1995 and 2009, the trade volume in primary tropical wood has fallen 40 percent
due to ecolabels as well as depletion and rising domestic use in Indonesia and
Malaysia.
r Drip irrigation is a water conservation option with a large potential. Besides
saving water, it reduces weed problems, soil erosion, electricity use and helps to
reduce overexploitation of groundwater. In poor regions, it may mean survival
for farmers.
r Today’s scientific advances in water desalination dramatically increase our abil-
ity to transform sea water into fresh water and quench the thirst of 1.2 billion
people facing shortages of water. In 1956, Israel’s Prime Minister Ben Gurian
said it: ‘Die Wüste mit entsalztem Meerwasser zu bewässern mag für viele
volkommen unrealistisch klingen, aber Israel sollte keine Angst vor scheinbar
unrealistischen Ideeen haben, die durch die Macht von Visionen, Wissenschaft
und Pioniergeist die vorgegebene Ordnung ändern’.
These statements make it clear that the problems and solutions regarding renew-
able resource use are diverse and are viewed differently. In a security-oriented world,
the greatest risk is that conflicts about resources intensify – and the solution may be
more armaments. Advocates of the Modernist worldview have high expectations of
new technologies and rational property and pricing arrangements. The idealists who
voice other worldviews frame the issue in terms of equal access, of frugal use and of
vulnerability. As in Chapters 7 and 9–11, I have attempted to link more specifically
certain renewable resource issues to the different worldviews, with Table 12.3 as
the result. It is an invitation to the reader to explore his or her own position. The
challenge is to listen to each other, respect the unity in diversity and find sustainable
resource management somewhere in the centre.
12.8 Summary Points

This chapter used an imaginary country to illustrate pathways of (un)sustainable
resource management. Many models of renewable resource use have been construc-
ted and applied, with modest success: overexploitation happened and still happens
in many situations. The key insights to remember from the models and stories are
the following:
r Renewable or (bio)resources have a finite regeneration rate and can, therefore,

only be harvested sustainably below a certain harvesting rate;
r The driving force behind (over)exploitation is the size and productivity of the
resource-exploiting capital stock (groundwater pumps, fishing boats, tree-cutting
machinery and so on). Open-access (bio)resources will be overexploited without
some form of regulation;
Table 12.3. Worldviews and renewable resource futures
Renewable Resources
Statement 1: As a result of overfishing, overall ecological unity of our oceans are under stress and at risk of
collapse (www.overfishing.org).
Admittedly, This is an alarmist The web of life is very Scientists have found
fisheries are statement. I don’t complex and intricate, evidence that
overexploited. understand and I don’t so if you destroy a part, overfishing is affecting
But what is the believe it. you may destroy the whole food webs, not
ecological unity whole. just single species.
of our oceans and
how can it
collapse? Besides,
Nature is more
robust than we
think.
Statement 2: The rapid cost reduction in seawater desalination make this technology an important ingredient
of the transition to a sustainable water future.
This is certainly true It sounds like this is how Maybe. These In semi-arid regions, it is
for many of the they use their oil centralised, often the only option
semi-arid regions revenues in the Middle capital-intensive and relieves the
in the world. The East. Not bad, water is options tend to pressure on
oil-rich countries their scarcest resource. discourage small-scale groundwater. It
can already afford options such as more should be in balance
it, and their use efficient use and with other options.
makes further rainwater collection.
innovations
possible.
Statement 3: Certificates such as FSC for tropical timber will save the tropical forests.
Indeed, the Possible but improbable: It is good that people Green certificates are
introduction of There is so much who buy such timber working well for
green certificates corruption in those can make a ‘green’ imports in Europe.
is a very effective countries. Anyway, as choice. Still, they may Still, illegal logging,
way to long as the green consider to use population pressure
sustainable certificate does not Europe’s own timber – and growing local
resource use. increase the price. with treatment it is demand pose severe
Consumers want equally durable. threats to the tropical
it, and industry forests.
does it.
Statement 4: The world community should implement and enforce criteria for the ecologically sustainable,
socially appropriate and economically viable production and use of biomass (IUCN 2010).
A typical Biomass use may be Worldwide biomass Biomass plantations can
NGO-statement: needed in our energy production to satisfy have negative impacts
A political system. Its use should the energy addiction of for indigenous people,
compromise and primarily be justified on the rich is not a local food supply and
impossible economic grounds. solution. Criteria such the environment.
promise . . . as these may at least Regulation is critically
avoid serous damage. needed.
r Historical experiences with and recent investigations of common pool resource

(CPR) management suggest clear sets of rules to avoid the government plan-
ning versus private property dichotomy and to overcome the ‘tragedy of the
commons’;
r Water is a key resource with diverse and local characteristics. Duration and
frequency of mismatch between water demand and (blue and green) water
availability are already significant in several regions and are expected to
increase. Groundwater resources are overexploited in many places. Increas-
ing water stress can have significant consequences for quality of life of many
people;
r Global fisheries exploitation is occurring above sustainable yields and the car-
rying capacity is eroded. The switch to large-scale aquaculture raises new con-
cerns. Scientific models must be improved and include human behaviour and
(the ambiguities of) technological change to be valid and useful. The same is
true for global forestries.
r Interactive simulation gaming and participatory modelling are useful, new tools
to engage larger groups of people, as stakeholders and citizens, into the theory
and practise of more sustainable resource use.
The highest form of goodness is like water

Water knows how to benefit all things without striving with them
It stays in places loathed by all men
Therefore, it comes near the Tao.
Nothing in the world is softer and weaker than water;
But, for attacking the hard and strong,
There is nothing like it!
For nothing can take its place.
That the weak overcomes the strong, and the soft overcomes the hard.
This is something known by all, but practices by none.
– Lao Tze, quoted in Allerd Stikker, Water: The Blood of the Earth
Then they saw the Cedar mountain, the Dwelling of the Gods,
The throne dais of Imini.
Across the face of the mountain the Cedar brought forth luxurious foliage,
Its shade was good.
– Gilgamesh and Enkidu arriving at the sacred grove – Epic of Gilgamesh
SUGGESTED READING
A textbook with several chapters on renewable resources.

Common, M., and S. Stagl. Ecological Economics – An Introduction. Cambridge: Cambridge
A special issue with a series of articles on system dynamic models in theory and practise.
Cavana, R., and A. Ford. Environmental and resource systems, System Dynamics Review
(Special Issue) 20 (2004): 2.
Thorough, mathematical introduction into renewable resource economics.
Clark, C. Mathematical Bioeconomics – The Optimal Management of Renewable Resources.
New York: Wiley Interscience, 1990.
This book has several system dynamics models in Stella R

and with diagrams about fisheries
and irrigation, with emphasis on spatial and commons aspects.
Costanza, R., B. Low, E. Ostrom, and J. Wilson, eds. Institutions, Ecosystems, and Sustaina-
bility. Ecological Economics Series. London: CRC Press, 2001.
Folmer, H., H. Landis Gabel, and H. Opschoor. Principles of Environmental and Resource
Economics – A Guide for Students and Decision-Makers. Cheltenham, UK: Edward Elgar,
1995.
Economics. Boston: Pearson Education Ltd., (2003).
An introduction into resource system dynamics models in Stella
R
, with diagrams about fisheries,
forests and so on.
Ruth, M., and B. Hannon. Modeling Dynamic Economic Systems. New York: Springer,
1997.
USEFUL WEBSITES
r glossary.eea.europa.eu/ is a general site of the European Environmental Agency with

brief explanation of terms in the area of environmental economics and policy.
r www.worldwater.org/ is an informative site on water in a series of reports The World’s
Water published since 1998, operated by Gleick.
r www.waterfootprint.org is the site of the waterfootprint, with definitions and calculations.
r ec.europa.eu/fisheries/cfp/fisheries_sector_en.htm is a rather complete and official
account of fisheries in the EU.
r www.worldfishcenter.org is a site on fisheries. It started as a centre in 1977 in the Phil-
ippines. The website states as its mission ‘to reduce poverty and hunger by improving
fisheries and aquaculture’.
r maps.grida.no/go/graphicslib have maps of global fisheries and global forests on the
site.
r www.fao.org has maps, statistics and documents such as the State of the World’s
Forests.
r www.worldagroforestry.org/ is a site with info and links on world agroforestry events and
issues.
r www.sciencemag.org/site/feature/data/deutschman/forest model.htm is a site on forest
dynamics models.
Websites on interactive models and games
r ivem.eldoc.ub.rug.nl/ivempubs/Software/Stratagemmanual/ is a description of the inter-

active game Stratagem (Meadows 1990), about running a country and experiencing that
stabilisation of population growth is a sine qua non for long-term sustainability.
r earthednet.org/Support/materials/FishBanks/fishbanks1.htm is one of the sites about the
game Fish Banks Ltd. (Meadows 1996), in which up to six teams get the opportunity to
experience, and prevent, the tragedy of the commons.
r www.rug.nl/fmns-research/ivem/publications/software/index has an overview of resource
models, among these the interactive model for electric power planning PowerPlan (de
Vries and Benders 1989).
r CORMAS (cormas.cirad.fr/) is the site on the CORMAS interactive simulation platform,
with practical recommendations for companion modelling (Daré et al. 2009). It is also
published in English.
r www.iemss.org is a site of the International Environmental Modelling & Software Soci-
ety (IEMSS), with information on a.o. the GEONAMICA R
software environment for
integrated environmental spatial modelling.
Appendix 12.2 Resource Use in the Simple Model 421
Appendix 12.1 The Simple Population and Renewable Resource Model

A general description of a resource or environmental sink exploitation model is:
dR(t )
= F [R(t ); G; t] − H(R) (A12.1)
dt
with dR/dt the rate of change of the resource or environment stock R and F[R(t);G;t]
a function expressing its degradation and regeneration. It accounts for depletion but
also external influences G and autonomous trends. H is the exploitation/harvest
rate. The maximum sustainable yield (MSY) or use is defined as the maximum value
of H for which the resource does not vanish. This occurs for dR/dt = 0, which is
the case for F[R(t);G;t] = H(R). If the resource dynamics is assumed to follow
a logistic growth path as given, then F = αR(1-R/K) in equation A12.1 and the
maximum value of H is found for the maximum value of dR/dt. Calculus tells us that
this happens for d2 R/dt2 = α(1–2R/K) = 0. This condition is fulfilled for a resource
stock of R = 1/2K. Therefore, the MSY equals Hmax = 1/4 αK. For a harvest H, the value
of dR/dt equals the difference between the two curves in Figure 12.4a. For values
of R between RA and RB , the resource will increase (dR/dt>0). For R exceeding
RB , the resource will decrease towards the stable attractor B with R = RB . For R
less than RA , the resource will decrease and become extinct (R = 0), as it moves
away from the unstable attractor A with R = RA . At or below the sustainable yield
level, the resource can – if it were to behave according to this simple description –
be exploited indefinitely. If the world were only that simple, this would reflect a clear
measure of sustainable resource use.
If the resource is an environmental sink, which is used for its capacity to absorb
and break down pollutants, the same formalism is used. H is the disposal rate,
and the function F represents the ‘cleansing’ capacity of the sink. For instance,
emitted carbon dioxide (CO2 ) disappears from the atmosphere in about 100 years,
on average, and certain chemicals such as polychlorinated biphenyls (PCBs) take
millennia to be broken down if ever at all.
Appendix 12.2 Resource Use in the Simple Model

If the resource is exploited as an open-access resource by individuals, it is possible to
derive the tragedy of the commons logic. If the harvest H is assumed to be a function
of effort E and resource size R, equation A12.1 can be written as:
dR(t )
= αR(1 − R/K) − εE a Rb (A12.2)
dt
For simplicity, we assume that there are no decreasing returns for an increasing
harvest rate (a = b = 1). The parameter ε is a measure of catch effectiveness (catch
per unit effort). The system is in steady-state equilibrium for dR/dt = 0, which
implies:
αR(1 − R/K) = εER (A12.3)
Rewriting shows that this is the case either for R = 0 (no resource) or for E = (α/ε)
(1–R/K). In other words, for each harvesting effort level E, there is a corresponding
stock size R for which the system is in a steady-state. Graphically, this implies that
the system is somewhere on the dotted isocline sloping downwards in Figure A12.1.
Figure A12.1. The points for which dE/dt = 0 and dR/dt = 0

(isoclines) in an open-access fishery according to eqn. A12.2.
At the intersection, the system is in dynamic equilibrium and
the last entrant will have zero profit.
What determines the economically efficient effort level E? Let the market price
of fish, p, and the costs per unit effort, c, be constant. Then, the profit of a fishing
firm equals the revenues B = pH minus the costs C = cE. Applying the open-access
principle of zero profit for the last entrant, the condition B = pH = C = cE translates
into E = pH/c. Because we assume proportionality, H = εER. Substitution gives
R = c/(εp) as the steady-state situation where the last entrant has a zero profit.
Graphically, it is the dotted vertical isocline in Figure A12.1. It proves that open-
access exploitation tends towards a stationary state of effort level E* and a stock
size R*. This situation is at the crossing point of the two isoclines. This is for E =
(α/ε)(1-c/(εpK)). The higher the regeneration rate (α) and/or the price (p) and the
lower the cost (c), the higher the steady-state effort level will be.
If the system is somewhere off the isoclines drawn in the graph, it will move
towards the attractor. How it moves is indicated with the arrows and can be seen
from the conditions:
r If R is on the right-hand side of the vertical isocline, R > c/(εp), which implies
that revenues exceed costs, B > C, and new firms will enter such as dE/dt > 0;
r if E is on the right-hand side of the sloped isocline, E > (α/ε)(1-R/K), which
implies that the harvest exceeds the natural growth rate, H > αR(1-R/K), and
the resource will shrink as in dR/dt < 0.
The position of the attractor depends, in this simple analysis, on the ratio between
natural growth rate α and catch effectiveness ε. It can be interpreted as the relative
force of resource (fish, tree and so on) and predator (people). The lower it is, the
more vulnerable the resource is for overexploitation. Or its corollary: The more
effective people become (ε↑), the more the resource is at risk. It also depends on the
ratio c/(εp), which is for given ε equal to the cost-price ratio. The higher the cost (c↑)
and the lower the price (p↓), the lower the risk of overexploitation. In this model,
high prices protect the resource and cost-reducing innovation do the opposite.
Appendix 12.3 Modelling Different Harvesting Strategies

Let us consider in more detail an investigation of different harvesting strategies
(Brede and de Vries 2009). Let there be a distributed resource consisting of j = 1..P
fishing zones, each modelled similar to the one introduced previously (equation
A12.2). Each agent chooses where to fish in the next season. The choice is determined
Appendix 12.4 The Geonamica Software 423
by a probability table that gives the chance that agent i fishes in zone j in the next
season. The table mirrors the agent’s view of the resource base in terms of catch and
profit performance. After each fishing season it is updated, using the information
about his and possibly also other agents’ last and previous performances. How an
agent builds this probability table is determined by the strategy it pursues. Therefore,
we define four strategies:
r the agent bases resource selection only on individual profit (Minority Game
(MG));
r the agent’s choice is motivated by the performance of a team it belongs to and
shares its catch with (Team Game (TG));
r agents use the impact of their catch on the performance of the whole community
(COllective INtelligence (COIN)); and
r the agent selects the resource to be exploited randomly (RANDom (RAND))
similar to a stochast strategy.
MG is a very selfish, individualist strategy. The COIN strategy is more ‘cooperative’
because agents base decisions not on their own profit but on the impact of their
effort for the community outcome. If played by many agents in the group, TG is a
very cooperative strategy.
Appendix 12.4 The Geonamica Software

Started by physicists and geographers at the Université Libre in Brussels, the Geo-
namica (formerly RamCo) software is one of the tools to incorporate local spe-
cificities with global developments. It has been used to investigate a variety of
resource and environment issues, such as the impact of climate change on the
Caribbean Islands (Engelen et al. 1993), land use planning in Indonesia (de Kok
and De Wind 2002), future land use patterns in the Netherlands, management
of the Wadden Sea in the Netherlands and floods risks (Engelen 2004; Nijs et al.
2004). It is developed and applied at the Research Institute for Knowledge Systems
(RIKS; www.riks.nl/products/geonamica) and at the Vlaams Instituut for Technolo-
gie onderzoek (VITO; Natuursimulator). Its basic methodology is cellular automata
(CA). The strength of this and similar models is the integration inherent to the
methodology and the interactive user-friendly mode.
The framework distinguishes three layers. The lowest layer contains the data on
the biophysical variables, such as elevation, water bodies, soil features and rainfall
patterns. It represents local (potential) suitability for various activities, for instance,
accessibility for housing and shopping, good soils for wheat or rice, attractive beaches
for tourism and so on. The highest level is an aggregate simulation of the driving
forces, notably population growth and sectoral macro-economic activity growth. It
is used as a scenario-generator. The intermediate level is based on a CA model.
It simulates the relative attractiveness of different forms of land use, such as agri-
culture, tourism or industry, against the background of suitability (lower level) and
driving forces (upper level). Changes in the driving forces cause changes in land
use, such as a conversion from agricultural land into housing areas, depending on
relative attractiveness. For instance, a premium price for a seashore area reflects
its suitability and attractiveness for tourism, which, however, declines with distance
Figure A12.2. Application of the Geonamica software to explore sustainable development in

western Sulawesi (Indonesia). The right side shows the high-level dynamic simulation model;
the left side shows the underlying base maps. The CA mechanisms relate both.
from the sea. The attractiveness potentials can interactively be adjusted to reflect
local expertise. Tools like these can identify the challenges and opportunities for sus-
tainable development of local/regional resources. An illustrative application is the
analysis of the impact of agricultural expansion on mangrove forests on the island of
Sulawesi in Indonesia (Engelen et al. 1995; Figure A12.2). The lower layer contains
a map with the actual use/cover and suitability for various activities of grid cells. The
upper layer is a system dynamics model that simulates regional demographic and
economic developments. The middle layer is the CA model, that contains the rules
according to which land use/cover change in response to regional drivers, on the one
hand, and biogeographical characteristics, on the other.
13 Non-Renewable Resources:
The Industrial Economy
13.1 Introduction: The Industrial Regime

Until the 18th century agriculture dominated the human economy, with almost
everywhere an important role for landed aristocracy, urban military and merchant
elites and religious institutions. Around 1750, a new social-ecological regime began:
the industrial regime. The beginnings were small, and hardly noticeable to most
contemporaries. While human history over the past 10,000 years has been the history
of the agrarianisation of the world, the history over the past 250 years has been the
history of industrialisation. The metabolic profile of the agrarian socio-ecological
regime and the industrial one are quite different (Schandl et al. 2009; §4.3). In the
industrial regime, energy and material use per capita is three to five times higher
than in agrarian societies. Population densities tend to be three to ten times higher,
energy and material use densities even ten to thirty times higher than in agrarian
societies. The fraction of biomass in energy supply is a factor three to then lower.1
The industrial regime operates on finite stocks that are produced in nature at
rates close to zero on a human timescale, hence the name non-renewable resources.
They are often referred to as minerals, from the Latin minerale, which means some-
thing mined, although later it broadened to ‘substances neither animal nor veget-
able’. Most importantly and already known and used in antiquity are sand, salt,
glass, limestone and of course metals, and for a few centuries also fossil fuels.2 Min-
ing and processing minerals to make metal objects has a history of several millennia.
Ancient Egypt was renowned for its rich gold, copper, silver and tin mines. There
has probably been silver (Ag) mining in Greece since 3500 Before Present (BP):
In the 5th century, between ten and thirty thousand miners were at work in the
mines of Laureion. A reconstruction of the exploitation history of lead (Pb), from
measurements of lead concentrations in Swedish lake sediments, put the first mining
back to the period 4000–3500 BP. Concentrations rose towards a peak during the
1 In the industrialised countries, the distinction between agriculture and industry has largely vanished.
Agri- and horticulture, livestock, fisheries and forestry have become almost completely industrial-
ised.
2 Minerals are usually understood as a chemical element or compound, which is also a (mineral) ore
in the sense that it contains a valuable constituent, often one of the elements defined as metals.
Mineral is the broader class, but I do not make a strict distinction between mineral and metal.
425
426 Non-Renewable Resources: The Industrial Economy
a)
Construction minerals
60 Ores and industrial minerals
Fossil energy carriers
Biomass
40
Gt/yr
20
0
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
Year
b)
10
Construction minerals
Ores and industrial minerals
8
Fossil energy carriers
Biomass
6
Gt/cap/yr
0
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
Year
Figure 13.1. Estimate of the most important mass flows in the world during the industrial era,
in absolute amounts (upper) and in per capita (lower) amounts (Krausmann et al. 2009).
Roman period (2150–1550 BP), declined thereafter and started to rise again with
the industrial revolution. It mirrors the European history of lead use of about 300
tons per year (t/yr) in 2700 BP to an estimated 80,000 t/yr around 1950 CE. Mining
and processing of minerals for iron (Fe), copper (Cu) and other metals has also
increased exponentially. Since 1900, the mass flows associated with fossil fuel use,
ores and industrial minerals and construction minerals has increased more than ten-
fold, whereas the biomass fraction in it has shrunk to one-third (Krausmann et al.
2009; Figure 13.1). The worldwide use of five key materials: cement, steel, paper,
aluminum and thermoplastics, has increased four- to sixfold between 1960 and 2005
and the exponential growth trend continues (Allwood et al. 2011).
The industrialisation process not only changes the material basis but also the
economic and social-cultural aspects of life. Unlike the rather static agrarian soci-
eties, the industrial economy has a strong endogenous growth tendency stemming
from the accumulation of capital and the desire for high returns in combination
with pure (science) and applied (technology) knowledge, as is discussed in Chapter
14. Average income rose exponentially, and colonial states and later multinational
corporations became the most important institutions. The rational mind and secular
values came to prevail in a world of economic rationalisation, a utilitarian ethic and
a liberal ideology – which also bred radical communism and fascism (§6.4).
The industrial economy is an everyday experience for an estimated 30 to 40
percent of the present world population. Its centre of gravity is shifting from Europe,
the United States and later the Former Soviet Union (FSU) and Japan to emerging
economies such as China. Some of its typical characteristics are:
r a delinking between regional biogeography and economic activities, except for
natural resource exploitation (minerals and fuels);
r a transition from local and diffuse land-based energy sources (wood-water-wind)
to concentrated and globally traded fossil fuel–based energy carriers;
r association of economic growth with core industrial sectors and physical through-
put: steel, cement, fertiliser and other materials, and intermediate capital goods;
r societal dynamics driven by resource and capital ownership and the forces of a
technological-industrial complex run by urban elites of bureaucrats, managers
and technocrats;
r a tendency towards concentration and homogeneity of economic activities,
largely driven by economies of scale and scope and preferential attachment
mechanisms;
r continuing competition between three dominant institutional clusters: tradition-
based communities, nation-states, and private enterprise corporations.
The signs of a next, postindustrial regime are already visible, with a crucial role
for information and communication technology (ICT). But large populations in the
world are still in the first stages of industrialisation. Therefore, it is widely expected
that, in conjunction with increasing population and economic activity, non-renewable
resource use and the associated emission flows to the environment will continue to
grow in the 21st century.
13.2 Non-Renewable Resource Chains: Extraction
13.2.1 Biogeochemical Element Cycles

The stocks and flows of materials and energy for human use should be seen in the
larger context of System Earth. Figure 13.2a shows a conceptual model of the human
population and economic subsystem within the finite biosphere. It highlights the role
of high-quality solar energy inflow and the Earth and cosmic environment as a sink
for low-quality energy outflows. The sun’s role in creating chemically stored exergy
(fossil fuels and metal-enriched mineral ores) and ‘closing the loop’ in the form of
recycling materials are also shown. From a sustainability perspective, it is not so
Table 13.1. Estimates of average metal abundance in crust and seawater, of total mobilisation
flows and the anthropogenic fraction therein.a Also, indications of richest ore and specific
mineralisability are given.b The elements for which the distribution is drawn in Figure 13.4 are
shown in bold
Total
mobilisation Richest ore
Avg conc in ppm rate Anthropogenic grade in q spec minera-
crust Seawater (TMR) fraction in TMR crust lisability
Metal g/Mg (ppm) g/Mg (ppm) Tg/yr % of total wt %
Aluminum (Al) 77,440 0.002 309 26 49.2 0.041

Iron (Fe) 30,890 0,002 848 90 0.087
Magnesium (Mg) 13,510 1,290 1,944 3
Carbon (C) 3,240 28 118,450 9
Titanium (Ti) 3,117 – 11 54 24.7 0.127
Sulphur (S) 953 905 766 22
Phosphorous (P) 665 0,1 538 6
Chlor (Cl) 640 19,354 6,264 2
Manganese (Mn) 527 0.0002 40 22 43.7 0.201
Zirconium (Zr) 237 – 1.6 78
Nitrogen (N) 83 150 6,067 7
Cerium (Ce) 66 0.2 56 0.15
Vanadium (Va) 53 0.003 1.6 72
Zinc (Zn) 52 0.8 20 47 4.3 0.215
Chromium (Cr) 35 0.0003 15 99 0.287
Niobium (Nb) 26 – 0.1 59
Lithium (Li) 22 0.2 0.2 51
Nickel (Ni) 19 0.001 2.3 69 0.153
Lead (Pb) 17 0.00004 3.9 84 5 0.286
Copper (Cu) 14 0.001 16 85 4 0.2
Gallium (Ga) 14 – 0.05 54
Cobalt (Co) 12 0.00005 3.5 6
Tin (Sn) 2.5 0.00001 0.3 78 1 0.277
Tungsten (W) 1.4 0.0001 0.04 95 2 0.287
Molybdene (Mo) 1.4 0.01 0.8 24
Uranium (U) 2.5 0.003 0.05 90 0.18 0.2
Arsenic (As) 2 0.004 0.1 63
Antimony (Sb) 0.3 0.0002 0.1 90 0.375
Bismuth (Bi) 0.1 – 0.01 97
Silver (Ag) 0.055 0.00004 0.03 58 0.056 0.294
Mercury (Hg) 0.06 0.00003 0.07 95 0.6 0.378
Palladium (Pd) 0.004 0.001 99
Gold (Au) 0.003 0.000004 0.003 100 0.002 0.298
Platinum (Pt) 0.0004 – 0.001 100 0.0006 0.198
a Klee and Graedel 2004.
b de Vries 1989.
much the absolute amount of anthropogenic flows but the amount relative to the
natural flows that matter. How much do humans interfere and does the interference
entail risks for ecosystems and human systems and, if so, how can they be avoided
or mitigated?
The elements are chemically or otherwise mobilised in natural and human-
induced processes. One can distinguish sequestration reservoirs in which nature
stores material for long periods and mobilisation reservoirs in which material is
transferred for much shorter periods (Figure 13.2b). Natural mobilisation flows are
associated with crustal weathering, sea spray and primary production by vegetation,
a)
Figure 13.2a. Conceptual model of human activities within an Earth system framework.
whereas anthropogenic mobilisation flows are from mining, fossil fuel combustion
and biomass burning. Klee and Graedel (2004) have analysed the stocks and flows
of seventy-seven out of ninety-two elements with the aim ‘to reflect the average rate
at which the state of a material is transformed from passive (e.g., in rock or soil) to
potentially interactive (e.g., in industrial products or in vegetation), with a focus on
the pedosphere . . . and the near-surface ocean’.
Not surprisingly, the mobilisation rates of the rare ‘precious metals’ such as
platinum (Pt) and gold (Au) and somewhat less rare elements like tin (Sn) and
tungsten (W) are dominated by human-induced, not natural flows (Table 13.1). The
anthropogenic mobilisation of elements of medium abundance, like vanadium (V),
zinc (Zn), chromium (Cr), nickel (Ni), lead (Pb) and copper (Cu) is significant and
b)
Figure 13.2b. Conceptual model of the stocks and flows of elements on Earth.
more than 65 percent of the total mobilisation flow. Anthropogenic interference with
abundant elements like aluminium (Al), titanium (Ti), sulphur (S) and manganese
(Mn) are also significant (22 to 54 percent). Some other abundant elements like
magnesium (Mg), carbon (C), phosphorous (P) and nitrogen (N) are part of such
large natural mobilisation flows that even their massive anthropogenic mobilisation,
notably in agriculture and fossil fuel combustion, represents only a small fraction of
the total mobilisation rate. Iron (Fe) is the exception here: Human use is an extreme
intervention (90 percent) in the natural flows. One possible explanation of the large
differences is that nature tends to rely for nutrients on water soluble elements – as
in the case of Ca in CaCO3 – whereas the mineral ores and metals used in human
structures are predominantly water-insoluble. Data such as these give an idea of
the extent of human interventions: The ratio of anthropogenic material flows and
natural stocks and flows is one example of a high-level sustainability indicator.
13.2.2 Classification
Two resources are crucial in the industrial transition: non-renewable mineral and
fossil fuel resources and environmental sinks.3 Non-renewable resources are exhaust-
ible or finite stocks of a substance, which are depleted upon exploitation. Their nat-
ural rate of formation happens at a geological timescale of thousands or millions
of years. Examples are mineral and oil deposits. Unsustainable use is, in essence,
exhaustion. Environmental sinks are natural systems such as soils, river flows and
the atmosphere, which are used by humans for ‘waste’ disposal. They are finite in the
sense that the disposed material is not immediately broken down or immobilised.
In the meantime, the disposed material may affect the system’s functions, often in
complex and only partly understood ways. In these respects, environmental sinks
are similar to renewable resources like fish and forest (Table 12.1). If the inflow of
a particular substance exceeds the rate at which the substance is eliminated and the
system regenerates, the sink is used unsustainably. Note that an environmental sink
can at the same time be a renewable resource, as for instance in the case of water
reservoirs. The two forms of use are usually in conflict.
Mining is at the ‘front-end’ or ‘upstream’ part of the resource chain. Mineral
and fuel exploitation (‘mining industry’) make up less than 1 percent of gross world
product (GWP) and the share in employment is much less. But the physical impact is
disproportionately large. Mining activities account for an estimated 7 to 10 percent
of global energy use, emit about 13 percent of global sulphur dioxide emissions, not
to mention the numerous other emissions, and impact upon 5.3 million square kilo-
metres (km2 ) of forest (WRI 2003). Even more than in agro-food systems, activities
are concentrated in a few large corporations. It is said that Glencore, the world’s
largest commodities trading company in 2010, controls 60 percent of zinc, 50 per-
cent of copper, 30 percent of aluminum and 25 percent of coal through its mining
3 What is considered a resource reflects the needs, skills and values in a society. For subsistence
farmers, farmland and pastures are the relevant resources and bauxite is useless if one does not
know about aluminium production and use. Uranium only became a resource once the discovery and
control of fission technology had developed in the 20th century. The same holds for environmental
sinks. Rivers were only recognised for their cleansing potential when water quality deteriorated.
Figure 13.3. Resource classification. The x-axis represents the probability that a certain
amount of resource is in place, the y-axis is an indication of the cost at which it is recoverable.
and trading interests.4 Mining mineral ores and fossil fuels is a complex process of
exploration and exploitation in a dynamic environment of markets, prices, techno-
logies, regulations and politics and with large uncertainties and risks (Harris 1984;
Yergin 1991). Complete certainty about how much resource and at which cost can
be extracted is only known when it is depleted. Added to this are the risks of market
disruptions and political conflicts. Investors, therefore, often want – and usually get –
high returns on their capital investment.
Resource assessments focus on two resource characteristics: the probability of
existence and the cost of extraction. Figure 13.3 shows a widely used classification
scheme. On the one extreme are the identified reserves. These refer to deposits that
are ‘proven’, meaning they do exist with >90 percent probability, and ‘economic’,
meaning profitable extraction is possible. The other extreme are the deposits that are
not yet discovered but which may be inferred from data and models of other resource
provinces. The estimates of the extraction costs are, at best, educated guesses. If the
exploration history of a region unfolds, the probability that the inferred resources
actually exist increases and their status is upwardly appreciated into possible
(>10 percent) and later on probable (>50 percent) reserves.
13.2.3 Availability, Exploration and Extraction: Two Models

In the context of sustainable development, the emphasis is usually on depletion and
a key question is: How big is the resource or, in expert jargon, what is the geological
resource base? First, I look at mineral ores. Solid evidence on resource ore quantity
and quality is limited by its very nature to those regions that have been explored.
However, it is possible to infer more speculative occurrences with hypotheses about
4 The company, founded in 1974, has been and still is surrounded by stories of fraud, tax evasion and
violation of human rights and environmental legislation. It is also an important player in world grain
trade and suspected of speculation on the grain market.
0.15
Ce
Hg
Fraction of total occurrence
Pb
Bimodal
0.10
0.05
0.00
1 10 100 1000 10000
Metal ore grade ppm (wt% * 10000)

Figure 13.4. The frequency distribution of three elements in the Earth’s crust as calculated
with the CAG model. The curves depict the fraction of the estimated total amount at a given
concentration interval. The dashed curve shows the possibility of a bimodal distribution.
how ores have formed in geological time. In 1954, the geologist Ahrens proposed
a ‘fundamental law of geochemistry’: The concentration or grade of an element
is lognormally distributed in a specific igneous rock. This – not uncontroversial –
crustal abundance geostatistical (CAG) model makes it possible to estimate the
geological resource in the form of a grade distribution curve (Appendix 13.1). This is
important because the extraction cost, in money and energy units, is to a large extent
determined by the ore grade. The outcome of such an estimation is the lognormal
distribution shown in Figure 13.4 for the relatively abundant lead (Pb), the very rare
mercury (Hg) and the rare earth element cerium (Ce). You remember the notion
of exergy introduced in §7.3? The high-concentration lead deposits have a chemical
potential with respect to the average crustal abundance that gives them a non-zero
exergy content. In other words, natural processes, fed by nuclear energy on the Sun
and Earth, did the separating work we benefit from (Appendix 13.1; Table 7.2).
The CAG model can put resource scarcity in a long-term perspective. For
instance, a news item such as ‘China’s export restraints on rare earth elements
has inflamed trade ties’ is clearly to be interpreted in a short-term market and trade
perspective and not (yet) considered a long-term scarcity concern. But there are at
least two caveats. First, the analysis does not include elements in seawater. Although
the concentrations are low, the amounts are huge (Table 13.1).5 Second, the evidence
from mining metal ores up to now does not exclude the possibility that, for chemical
or physical reasons, the distribution differs from a binomial one. It can turn out to
be a bimodal one, for instance – the dashed curve in Figure 13.4. This would lead to
a very different estimate of the amounts of metal available at low concentrations.
5 Some elements in seawater get concentrated by natural processes, such as manganese (Mn) nodules
on the ocean floor. Sometimes their increase is substantial on the scale of decades, which make such
occurrences nearly a renewable resource. They are a prime example of common pool resources
(CPRs) (§5.4).
There are other ways to estimate the geological resource base and its life cycle.
A widely known one is the logistic growth life cycle model (Appendix 13.2). It was
first stated in 1956 by the American geologist King Hubbert and is based on the
exploration and exploitation history of a region and has been applied mostly for oil
and gas. Using past oil exploration and exploitation statistics for the United States
(48) (onshore and offshore of the lower 48 states without Alaska), King Hubbert
found that the discovery rate per $ spent on exploration declined and forecasted that
output in the United States (48) would peak around 1970. Since then, United States
(48) oil production has declined within 5 percent of this prediction. The discovery
and subsequent exploitation of huge oil fields in Alaska retarded the decline for
a few years. But since 1986, the decline has continued and the United States has
become ever more dependent on imports – from Canada and Mexico, amongst
others. Accumulated oil output in the United States can accurately be simulated
with a logistic trajectory, with minor deviations during the period of the oil crises
and high oil prices of the 1970s and 1980s (Figure 13.5a). There is a delay of only ten
years between accumulated production with and without reserve, which means the
reserves have always been equal to about ten years of production.
In retrospect, King Hubbert’s prediction looks surprisingly correct, and his
model seems a useful metamodel. However, the explanation of the peak is not
necessarily a geological constraint due to depletion. It may also reflect the strategy
of U.S. oil companies to look abroad for cheaper crude oil supplies and to import oil
and oil products. For instance, in the Cold War era, the U.S. government restricted
for military reasons oil imports to less than 10 percent of oil use, which led to a rel-
atively deep depletion of U.S. resources and a cheap Middle East oil glut in Europe
(Yergin 1991).
Table 13.2 contains data on fossil fuel reserves and production for a couple of
countries, and there are clear signs of the occurrence of a life cycle profile in reserve
and production. For instance, production since 2000 in four countries outside the
Middle East (Mexico, Indonesia, United Kingdom and Norway) is probably past
their peak. Their share in world oil production declined from 23 percent in 1995 to 15
percent in 2009, although this too may partly be a reflection of market circumstances
and strategic reasons. For these countries, a decline in production and reserves may
pose serious challenges for economic development and force them into a transition
to other sources.
In principle, the logistic model is also valid for natural gas. Empirical data
for reserves and production of natural gas in The Netherlands, however, deviate
significantly from a logistic growth path (Figure 13.5b). The profile is more plateau-
like and this is probably due to the fast upward appreciation of the gas reserve in
the giant Groningen field in the 1960s. For coal and mineral resources, the logistic
model has limited validity, because local exploitation depends to a large extent on
other factors than depletion.
Many experts and nonexperts speculate these days about the time of maximum
world oil production – the global ‘oil peak’. One reason is that almost two-thirds of
world conventional oil production is from so-called giant fields (>500 mln barrel of
oil equivalent (bbl) or about 3 exajoule (EJ)), most of these having been discovered
decades ago and already for years declining in output. King Hubbert’s model is
often applied to calculate the world peak oil date. However, there are reasons to be
Table 13.2. Resource characteristics for some countries/regions in the world, 2009a
2009 USA Japan EU-27 Korea, Rep S. Arabia Russian Fed
Population (mln) 307 128 499 49 25 142

Income (PPP U.S.$/cap, 46,437 32,433 30,543 27,195 23,421 18,938
2009)
Oil, proved reserve (EJ) 13.5 3.0 126.1 35.4
Oil reserve as % of world 2.1 0.5 19.8 5.6
proved reserve
Oil, production rate (EJ/yr) 3.4 0 1.0 0 4.6 4.8
RPR (yr) – Oil 10.8 8.2 74.6 20.3
Natural gas
Coal
Import/use ratio – Oil
Natural gas
Coal
Military expenditures 4.3 0.9 1.8 2.8 8.2 3.5
(% of GDP)
2009 Mexico Brazil China India Vietnam Nigeria
Population (mln) 107 194 1.332 1.155 87 155
Income (PPP U.S.$/cap, 14,341 10,429 6,838 3,275 2,956 2,149
2009)
Oil, proved reserve (EJ) 5.6 6.1 7.1 2.8 2.1 17.7
Oil reserve as % of world 0.9 1.0 1.1 0.4 0.3 2.8
proved reserve
Oil, production rate (EJ/yr) 1.4 1.0 1.8 0.4 0.2 1.0
RPR (yr) – Oil 10.8 17.4 10.7 21.1 35.7 49.5
Natural gas
Coal
Military expenditures 0.5 1.5 2.0 2.6 2.4 0.8
(% of GDP)
a Source: BP, en.wikipedia.org/wiki/List_of_countries_by_military_expenditures.
cautious. First, unlike the United States, several regions in the world are relatively
unexplored and the model gives no clue about the size of the ultimately recoverable
resources. For instance, under the ice layers, Greenland may have 10 to 30 percent
of the as yet undiscovered oil and gas resources.6 Second, in several regions there
are large reserves of unconventional oil and gas deposits, the so-called oil and gas
shales and tar sands. Canadian oil sands alone amount to the equivalent of at least 10
percent of total conventional oil reserves. Another reason to be cautious is the limited
reliability of the statistical data on exploration and production. Major stakeholders
have strategic reasons to manipulate the data: oil companies for reasons of stock
market value and price speculation, and governments of Saudi Arabi, Russia and
other countries in order to optimise their revenues. What can be said with near
certainty, however, is that the world community faces in the next half century a
transition away from oil and with gas as an important transient fuel.
6 A dramatic story is unfolding in Greenland and other parts of the Arctic, where oil companies,
governments and NGOs are preparing themselves for the future battle when large oil and gas deposits
are discovered. The first international disputes about territorial claims have started. Will global
warming facilitate the exploitation? Can oil spills be avoided? Are indigenous cultures threatened?
a) USA crude oil

Cum crude oil prodn Cum prodn + reserves
Simulated reserve crude oil prodn
Simulated prodn
1600 30
25
1200
20
EJ/yr
800 15
EJ
10
400
5
0 0
1940 1950 1960 1970 1980 1990 2000
Year
Figure 13.5a. Oil accumulated production, reserves and annual production rates for the
United States (source of data: TEPD 2010, www.bp.com).
b) Netherlands nat gas Cum nat gas prodn Cum prodn + reserve
Simulated reserve nat gas production
Simulated prodn
200 4
150 3
EJ/yr
100 2
EJ
50 1
0 0
1950 1960 1970 1980 1990 2000
Year
Figure 13.5b. Natural gas accumulated production, reserves and annual production rates for
The Netherlands and some other countries (source of data: BP and Ministry of Economic
Affairs).
c) USA Persian Gulf Other

USA imports EU imports
160
120
80
EJ/yr
40
0
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
Year
Figure 13.5c. World crude oil production 1960–2010 for the United States, the Persian Gulf
and other countries. The lines show the crude oil imports for the United States and the EU.
Data for 2010 are provisional (source of data: TEPD 2010, www.bp.com).
13.3 Elementary Resource Economics
13.3.1 Supply Cost Curves

Minerals and fuels exist as deposits: rock layers with varying ore grades, subsurface
layers of coal or subsurface porous layers filled with trapped oil or gas. Exploration
produces knowledge about the probability of a deposit to exist and, in combination
with production tests, about the cost at which it can be extracted. Mining corporations
assess the economic value of a deposit along at least two dimensions:
r technology and production cost: given the existing exploitation technology, at
which cost can the resource be produced? Costs depend on depth and com-
position of an ore deposit, thickness and depth of a coal layer and depth and
size distribution of oil and gas fields. The separate deposits or fields in a region
are usually ranked in order of increasing extraction cost, which results in the
so-called long-term supply cost curve (SCC)7 ;
r processing and marketing: the resource, once extracted, requires processing,
transport and upgrading. Cost and price of marketable products will, therefore,
depend on product quality and standards, distance to markets and safety and
environmental regulations.
The larger picture has to be constructed from the, mostly confidential, data on ore
and coal deposits and oil and gas fields by mining corporations. There is a difference,
though, with food: Ore and fuel exploitation is inherently more global because of
the concentration of ‘value’ (exergy) by natural processes. It explains the early
globalisation and concentration of non-renewable resource markets and business.
7 The resource deposit, technology or substitute at which cost no longer increases upon extraction is
called the ‘backstop’ resource. In economics, the increase of the extraction cost is associated with
differential (or Ricardian) rent, as introduced in the 19th century by Ricardo for land resources.
a)
375 Oil
Low reserve
Medium reserve
Production cost (US $/bbl)
300 High reserve
225
150
75
0
0 20000 40000 60000
Resource (EJ)
Figure 13.6a. Long-term cost supply curve estimates for world oil (1 US$/GJ ≈ 7 US$/bbl).
World oil use in 2005 amounted to over 0,16 ZJ (van Vuuren 2007/USGS).
In practise, resource availability and depletion is often evaluated in terms of

the reserve production ratio (RPR), which is the ratio between (proven) reserves
and (actual) annual production. It is measured in years and used by countries and
corporations. Table 13.2 gives some RPR estimates. However, the use of the RPR
is misleading as a scarcity indicator because reserves are a function of exploration,
which is often driven by price (expectations). If prices go up, exploration is intensified
and new discoveries and re-appraisal of previously uneconomic deposits add to the
reserves.8 A better way to evaluate the resource situation of a country or of the world
at large is the aforementioned SCC. In Figure 13.6, it is shown for conventional and
unconventional world oil and gas resources as of 2005. For any given resource
size estimate, the curve indicates the estimated marginal cost. The uncertainty is
indicated by showing upper and lower estimates. The right-hand parts in Figure
13.6 are highly uncertain and speculative, which is reflected in the larger spread in
the estimates. Besides geological uncertainties, there is the ignorance about as yet
unknown technologies and environmental constraints.9
What do these curves tell us? In 2005, world crude oil extraction amounted to
about 155 EJ or about 27 billion barrels (bbl) (Figure 13.5c). The SCC indicates that
crude oil reserves at <50 $/bbl would last at the 2005 extraction rate for another
seventy years.10 It may also be 100 (lower SCC) or 35 (upper SCC) years (band in
Figure 13.6a). The reserves at <100 $/bbl would last 70–190 years at the 2005 extrac-
tion rate. But, of course, at these costs several alternatives become very competitive.
Organisations like the International Energy Agency (IEA) expect in the 2011 World
8 Reappraisal is the process of re-evaluating the resource in existing and abandoned mines and oil/gas
fields, with the new insights and techniques. For instance, novel techniques to recover more oil out
of the carrier sediment or better seismic exploration techniques to more accurately assess the 3-D
reservoir characteristics.
9 The curves in Figure 13.4b can also be considered SCC inasfar as there is an inverse proportional
relationship between cost and metal concentration.
10 In other words, the RPR is 70 years at a cut-off extraction cost level of 50 $/bbl.
b) Gas
150
Production cost (US $/bbl)
100
50
Low reserve
Medium reserve
High reserve
0
0 25000 50000 75000 100000
Resource (EJ)
Figure 13.6b. Idem, for natural gas. World gas production/use in 2005 amounted to almost
0,11 ZJ in 2005 (van Vuuren 2007, www.usgs.gov).
Energy Outlook (WEO) that world oil production rises 15 percent between 2010
and 2030, with most expansion coming from unconventional deposits. Other agencies
are more pessimistic and expect a world oil peak to come before 2020 (TDBT 2004;
ZTB 2010). The uncertainties give room for controversy and speculation and dif-
ferent interests and worldviews influence the analyses and expectations.11 Whatever
may turn out to be correct, there will be rising tensions in the coming decades in
the form of volatile and high oil prices and intense strategic games and political
conflicts.
A similar situation exists for natural gas, but it is widely believed that the
resources are still larger, that there are interesting opportunities to convert gas
to liquids to curb oil shortages, and that there may be a backstop in the form of huge
amounts of unconventional gas (Figure 13.6b). Most experts in business and govern-
ment count on natural gas as a key transition fuel for the next decades because of
its relative abundance, the potential it offers for further efficiency gains and its rel-
atively low carbon content. There are downsides, too. The necessary infrastructure,
in the form of long-distance pipelines and liquid natural gas (LNG) plants and ships,
is capital-intensive, and the production of unconventional gas in the form of shale
gas – and similarly of shale oil and of oil from tar sands – is not only more costly but
is also expected to cause serious environmental risks from large use of water and
chemicals.
The dynamics of resource exploration and exploitation is summarised in Figure
13.7. Demand drives investment in capital stocks for exploitation. This relative short-
term mechanism is supplemented with three longer-term loops. The first one is deple-
tion: The unit cost tends to go up with accumulated production, which has a negative
influence on demand via the price elasticity. Simultaneously, companies invest part
of the revenues and profits into cost-decreasing R&D and innovations. This tends
11 For instance, a 2010 WikiLeaks rumour indicated that Saudi Arabian oil reserves are overestimated
with an amount in the order of 2,000 EJ.
Figure 13.7. The most important dynamic mechanisms that determine long-term trends in
resource supply, costs and prices. The items in italic are influences from other subsystems.
to stabilise or lower unit cost, if there is sufficient competition on (global) markets.

Another part is invested in exploration, when the reserve base is considered too low
or when newcomers enter the market. This permits a constant or increasing identi-
fied reserve. Fourth, the more stringent safety and environment regulations tend to
increase unit costs, although these are difficult to separate from the other mechan-
isms. The resulting price changes will probably be upward and have an influence on
substitutes such as biomass-based fuels or plastics. If the reserve base becomes less
than desired by companies or governments and there is also limited competition, one
can expect price fluctuations because of uncertainty and speculation. Geopolitical
and technological developments make the story still more complex.
13.3.2 Innovation: The Learning-By-Doing Mechanism

The SCC not only reflects the observation that costs of minerals and fossil fuels
increase because of depletion but also, in the last decades, because environmental
and social costs are increasingly internalised in the cost, for instance, via envir-
onmental and safety regulations.12 Therefore, one expects costs and prices of non-
renewable resources to rise. Why then have the production cost of most metals and
fossil fuels gone down in the course of the 20th century, despite an accelerating
extraction rate and more stringent regulation (Allwood et al. 2011)? One answer
is: exploration. Resources are not necessarily discovered in order of cost and newly
discovered fields may turn out to be cheaper to exploit than those already under
12 Indeed, it is argued that lax implementation in a competitive and greedy industry are the cause of
disasters like the BP oil spill in the Mexican Gulf in 2010.
Box 13.1. Carbon sources and sinks. As discussed later on in this chapter, one
of the environmental issues is the change in the radiation balance of the Earth
as a consequence of rising concentrations of so-called greenhouse gases, notably
carbon dioxide (CO2 ). Because CO2 is emitted upon burning fossil fuels, notably
coal because it has a two times higher emission per unit of energy than natural gas,
mankind faces a somewhat weird source-sink problem. On the one hand, people
desperately long for cheap carbon in order to sustain or enter the industrial era
with its cars, appliances and so on. On the other hand, every kilogram (kg) of
coal, oil or gas discovered is almost certain to cause an emission in the order of 2
to 4 kg of CO2 , which will remain in the atmosphere for on average 100 years.
An affordable and profitable energy supply system requires more carbon and
oil companies and governments frantically search for carbon all over the world
in order to increase the existing proven reserves of about 3,400 gigaton carbon
(GtC). Reducing the risk of large-scale damage from climate change, for instance,
according to the 2◦ C target of the EU, implies that the cumulated carbon emis-
sions in the 21st century should not exceed the range of 800 to 1,000 GtC. The
reconciliation of this source-supply quandary may become the ultimate challenge
to human aspirations.
exploitation. An equally important answer is: technology. Extraction costs tend to

fall over time because of innovations (‘technical progress’). Innovation-driven cost
reductions are often spectacular, as in offshore oil and gas production – although
the move towards offshore fields was itself the consequence of depletion of onshore
resources.
The cost-decreasing effects of innovations, both incremental and breakthrough,
have been generalised into the concept of the learning or experience curve. It is
an empirical fact that key performance indicators of industrial processes – such as
cost per unit of output – decline over time because of the incremental technical
and organisational innovations and increases in scale.13 In-depth analysis shows the
complexity and contingency of the underlying processes (Wene 2000; Junginger et al.
2010). Learning-by-doing, as it is called, is incorporated in many models according
to a simple power law:
−p
y x
= (13.2)
y0 x0
for a cost/performance variable y and an learning-by-experience variable x. The

suffix 0 indicates the time at which learning is supposed to have started. Usually,
the cost/performance variable y is specific investment or operational cost and the
variable x is the cumulated output over a certain period. The learning rate p can also
be expressed by the progress ratio ρ, which indicates the factor with which the cost
13 Scale in industrial production can be either the size of plants, as with power plants, or the number
of items being produced, as with solar photovoltaic (PV) panels. Both aspects contribute to cost
reductions because of standardisation, shared overhead costs and so on.
Figure 13.8. Learning-curve relationships for energy technologies in the EU (source: PBL,
Junginger et al. 2010).
measure y decreases on a doubling of x.14 Equation 13.2 has gotten the status of a
law, although it rather is a metamodel.
Data for seven electric power generation technologies are drawn in Figure 13.8
in a log-log plot. It approaches a straight line over a large domain of cumulated
installed capacity, as it should for a power law, but the slope differs for different
technologies and changes over time. Progress ratios are for these technologies in the
range of 0.65 to 0.95. The implications are significant. If the cost of solar-photovoltaic
(PV) panels, for instance, decreases with 10 percent (ρ = 0.9) for the first doubling
of output (in MWpeak), then you need one more doubling of the cumulated output
for another 10 percent decline in costs. If the first doubling took 1 year and output
remains constant, the next 10 percent cost reduction takes two years. But the next
10 percent cost reduction will take four years at constant output – and the next one
eight years! This explains why learning-by-doing is only fast in time if production
grows fast as well and why otherwise massive subsidy is needed to stimulate demand.
Thus, long-term cost trends are the effect of two mechanisms: depletion and
learning (Figure 13.7). Because the costs of most metals and fossil fuels have declined
during most of the 20th century, the learning effect appears to have dominated.
Recent trends suggest that for several minerals/metals and fossil fuels the two effects
increasingly offset each other and costs remain more or less constant for long periods.
But reliable cost data are scarce and most available data are about price. Ideally,
there is an equilibrium between demand and supply and the price is closely correlated
with (marginal) cost. But the world market for metals and fuels fluctuates between
a sellers’ market when demand exceeds (perceived) supply and prices tend to rise,
14 If the logarithm is taken on both sides, a plot on a log-log paper gives a straight line. It is also seen
then that ρ = 2−p (p = –log(ρ)/log(2) > 0).
and a buyers’ market when supply exceeds demand and prices fall. One reason is
the inevitable system delays due to long construction times of new installations and
mines (5 to 10 years) and fluctuations in demand in response to changing economic
activity (§2.3). For instance, the rapidly increasing demand for steel and oil in China
was a major cause of oil price volatility in the early 21st century. Such volatility
spreads throughout the economy, because prices of natural gas and oil are coupled
and fossil fuels are important inputs for most industrial products. Speculation makes
it worse.
In the longer term, most experts anticipate higher prices, not only because of
depletion but also in view of (geo)political tensions and environmental policies.
Political upheavals such as the fall of the Iron Curtain in 1989 and the Gulf Wars and
Middle East uprisings create uncertainty and influence business profits and economic
policy. Important from a sustainability perspective are the trends in environmental
indicators and policies. More stringent regulation to protect biodiversity, in com-
bination with human rights protection, affects metal prices. Strong and consistent
greenhouse gas emission taxes raise the fossil fuel prices directly, but also indirectly
(§14.3).The effects of these factors are mitigated or exacerbated by trade and it is not
surprising that mineral/metal and fossil fuel prices experience significant and hardly
predictable fluctuations.
13.3.3 Optimal Depletion: The Resource Curse and Resource Security

Resource depletion should be judged in the broader context of the transition towards
sustainable metal and energy use systems. Economists have examined the optimal
resource extraction rate from a macroeconomic point of view (Dasgupta and Heal
1979; Ströbele 1984). In a capitalist setting of private ownership, an optimal extrac-
tion path H(t) to deplete a finite amount of resource Rult is found for:
T
max e−rt [(p − c(CP)).P] dt (13.3)

0
under the condition that dCP/dt = –P and CP ≤ Rult , with CP cumulated production
and P production. T is the planning horizon, r is the discount rate. In words, the
optimal extraction path is when the discounted net profit e−rt (p–c)P is maximised
over the project duration (§10.4, §12.2). If costs are known as a function of depletion,
the optimal path can be calculated.
In an ideal world, it can be shown that an economically optimal resource extrac-
tion path requires an annual percentage change in the resource price equal to the
discount rate during the exploitation period.15 In formula: dp(t)/dt = r. This out-
come has become known as the Hotelling rule, after the economist Hotelling who
formulated it in the 1930s. Therefore, the price should rise exponentially unless
the extraction cost goes down or a cheaper alternative or a constant cost ‘back-
stop’ option shows up within the exploitation period. For low-risk projects and in
15 The price p is the net-price which can be considered to represent rent or royalty and is given by P =
p + MC with P the gross market price and MC the marginal cost of production, such as the cost to
produce the last unit of the annual flow (Perman et al. 2003).
Box 13.2. Theory and observations. In economic theory, the relation with
empirical observations is a difficult one. The Hotelling rule has never been unam-
biguously confirmed from empirical data. One reason is, as often in economic and
social sciences, that neither the discount rate nor the net profit can be measured –
they are both aggregates to be estimated with proxies from incomplete data. ‘The
Hotelling rule is an economic theory . . . a theory is not necessarily correct . . . [a
theory] may fail to “fit the facts” because it refers to an idealised model of reality
that does not take into account some elements of real-world complexity. How-
ever, failing to fit the facts does not make the theory false; the theory only applies
to the idealized world for which it was constructed . . . The history of attempts
to test the Hotelling principle is an excellent example of the problems faced by
economists . . . many of the variables used in our theories are unobservable or
latent variables. Shadow prices are one class of such latent variables. The best we
can do is to find proxy variables for them. But if the theory does not work, is that
because the theory was poor or because or proxy was not good?’ (Perman et al.
(2003) 527–529). This situation is often dealt with in the form of metamodels and
stylised facts, as is discussed in §8.6.
situations of abundant and cheap capital (small r), the price increase will be smaller
than for high-risk projects and capital scarcity (large r). The Hotelling rule points
at the existence of resource scarcity rent, which comes on top of the cost to extract
and produce it. Such rents are difficult to measure and sometimes huge, and their
appropriation is a continuous battle between national and international business
elites and governments.
Optimal resource exploitation models are also applied for society at large, with
the optimal path being the one that maximises intergenerational social welfare
(§14.1). It remains mostly academic but raises a couple of normative issues such as
the form of a societal welfare function and the value of the discount rate (Dasgupta
2008; Stern 2008). But the relation between economic performance, resources and
technology is far more complex than can be dealt with in these models. Neverthe-
less, optimal resource algorithms are widely used, for instance, in estimating the
consequences of climate change impacts and policies, but the methodology, assump-
tions and outcomes are controversial (IPCC 2007).
Real-world behaviour is often far from economically optimal and rational
(§10.4). Therefore, theoretical analyses should be complemented with stories from
the real world. The extraction of rich metal ores and fossil fuel deposits was and still
is as much a matter of military adventure and political power struggles as it is of eco-
nomically rational behaviour (Yergin 1991). The major mining companies are from
countries like Australia and Canada, but many of the richest ores and, therefore,
most profitable mining opportunities are currently in countries in the periphery of
the industrial centres. The mining activities often generate large revenues for the
host country, but they also disrupt local cultures and communities and cause envi-
ronmental havoc. In countries with authoritarian or weak governance structures, the
revenues and job opportunities tend to go for the larger part to the ruling elites – at
least, this is what journalists found out about the mines in Papua, Niger and other
places and what is broadcast about the bank accounts of the former dictator fam-
ilies of some Asian, African and Middle East countries. This and the bad working
conditions of the miners and world market price fluctuations cause regularly social
unrest. This too is part of unsustainable development.
One particular hypothesis has received much attention, namely that a state’s
reliance on either oil or mineral exports tend to make it less democratic.16 This is
called the resource curse (RC) hypothesis. An analysis of 113 countries in the period
1971–1997 suggests a link between the large oil wealth of a nation and its governance:
‘[there are] three causal mechanisms that link oil and authoritarianism: a rentier
effect, through which governments use low tax rates and high spending to dampen
pressures for democracy; a repression effect, by which governments build up their
internal security forces to ward off democratic pressures; and a modernisation effect,
in which the failure of the population to move into industrial and service sector jobs
renders them less likely to push for democracy’ (Ross 2001). Recent econometric
analyses for a large sample of countries for the period 1970–1995, confirm that natural
resource abundance, with mineral production share in GDP as a proxy, correlates
negatively with economic growth. It suggests a resource curse. However, the effect
reverses sign if other possible explanatory variables such as corruption, investment,
openness, terms of trade and schooling are included. In a long-term perspective,
there is evidence that abundant natural resources tend to crowd out other income-
supporting activities, notably savings (Papyrakis and Gerlagh 2004, 2006).
Not surprisingly, a peculiar relationship has evolved between natural resources,
in particular oil and gas, and arms. Huge amounts of money are flowing into the
Middle East region from oil and gas export revenues. This influx, itself unstable and a
source of dependency, is part of an arms race. In the 1990s, between 20 percent and 40
percent of annual oil export revenues in Saudi Arabia were spent on weapon imports,
mostly from the European Union (EU) and the United States. In 2009, military
expenditures of Middle East countries as % of GDP were between 3.4% for Syria and
8.2% for Saudi Arabia (Table 13.2). Buying and selling arms stems from the under-
standable need of protection, but it may be one of the great pseudo-satisfiers (§6.2).
Another issue that enters the equation is the concern about energy security
in industrialised and industrialising countries (Kruyt et al. 2009). The dependency
of a number of states on oil and gas imports has been growing for decades and
keeps increasing (Table 13.2; Figure 13.5c). The interests at stake are huge, even
apart from environmental concerns like climate change. For instance, the EU’s
dependence on energy imports is increasing again, because the indigenous resources
have rapidly been exploited after the 1972 and 1979 oil crises. In 2005, the EU
imported about 50 percent of its energy with 45 percent of oil from the Middle East
and 40 percent of natural gas from Russia. It is expected to rise to 70 percent by 2025
(OECD/IEA 2010). The monetary value of energy imports is still only 6 percent
of total imports, but the physical dependency can have serious economic and social
risks in the face of global depletion. A similar situation exists for the United States,
16 The high-income country version is the Dutch disease, according to which the easy revenues from
a large and cheap to exploit resource cause governments to overspend, make citizens complacent
and incur inflation. Of course, one may assume that the equivalent of windfall profits for the mining
corporations have similar, though less publicised effects.
13.4 Stories 445
where oil imports amounted to twice the indigenous production. Japan is even more
dependent on fossil fuel imports. China and India are probably soon to follow. The
energy transition to noncarbon energy sources such as hydropower, nuclear power
and renewable sources (wind and solar) reduces the tensions, but oil price volatility
slows the transition down because investors avoid the risks in the absence of clear
signals of high future oil prices. The same is true for the materials transition, where
efficiency and recycling initiatives suffer from uncertain and volatile primary metal
prices. In the longer term, the use of finite resources is unsustainable. So what are
they used for? That is the topic of the next section – but first some stories.
13.4 Stories
13.4.1 Oil and Power17

Natural resources such as metal ores, coal and oil have been throughout history at
the centre of power struggles and ideologies. ‘The rapid rise of Russian production,
the towering position of Standard Oil, the struggle for established and new markets
at a time of increasing supplies – all were factors in what became known as the
Oil Wars. . . . there was a continuing struggle involving four rivals – Standard, the
Rothschilds, the Nobels, and the other Russian producers. At one moment they
would be battling fiercely for markets, cutting prices, trying to undersell one another;
at the next they would be courting each other, trying to make an arrangement to
apportion the world’s markets among themselves; at still the next, they would be
exploring mergers and acquisitions. On many occasions they would be doing all
three at the same time, in an atmosphere of great suspicion and mistrust . . . ’ (Yergin
1991). This sounds familiar, but it describes the 1890s – not much has changed.
In March 2001, U.S. oil majors were the leading foreign investors in Kazakh-
stan’s hydrocarbon riches. Six months before September 11, 2001, U.S. Ambassador
to Kazakhstan Jones said the United States wanted ‘Kazakhstan to develop its energy
resources . . . and have better access to world (oil) markets. . . . a stable central Asia –
and Kazakhstan in particular – would allow stable exports of Caspian oil to inter-
national markets . . . the US government backed the export pipeline from Kazakhstan
to Russia’s Black Sea port of Novorossirsk . . . Washington also supported two other
planned pipelines, which will eventually link the Caspian region . . . on the Mediter-
ranean. But . . . Washington would continue to oppose plans to export Caspian oil to
neighbouring Iran, which it accuses of sponsoring international terrorism . . . the US
Congress was likely to extend sanctions against Iran after they expire in August.’
It had long been known that Afghanistan had significant deposits of gemstones,
copper and other minerals, but in the summer of 2010, U.S. officials said they had
discovered and documented major, previously unknown deposits, including copper,
iron, gold and industrial metals like lithium. A Pentagon team, working with geo-
logists and other experts, had shared its data with the Afghan government. It was
working with the Afghan Ministry of Mines to prepare information for potential
investors. A few weeks later, Afghan officials said they believed that the American
17 The quotes are from the site www.nytimes.com/2010/06/18/world/asia/18mines.html.
estimates of the value of the mineral deposits – nearly $1 trillion – were too con-
servative. They could be worth as much as $3 trillion. The Ministry of Mines also
announced that it would take the first steps towards opening the country’s reserves
to international investors. Two hundred investors from around the world have been
invited to offer suggestions for how to develop the iron ore deposits at the Hajigak
area of Bamian Province, according to the principal mining specialist for Afghanistan
for the World Bank.
13.4.2 The Promise of Gold: Tambogrande18

In the 1990s, a controversial mining project was proposed for Tambo Grande, a town
of roughly 18,000 people in northern Peru. The government of Peru conceded three
blocks of land totaling 87,000 hectares (ha) to the Canadian Manhattan Minerals
Corporation. One 10,000-hectare concession, right in the town of Tambo Grande,
is particularly attractive to Manhattan Minerals. It is a massive polymetallic deposit
known as TG-1 that has gold sitting closest to the surface. Manhattan Minerals can
mine the gold first and finance the underlying copper and zinc extraction with the
revenues generated early in the project’s life.
The Tambo Grande valley is a green valley on Peru’s desert coast, which came
into being in the 1950s after building a reservoir and irrigation system with World
Bank money. It is now producing 40 percent of Peru’s mangoes and limes. These
and other crops create about U.S. $2 billion in revenue annually and permanently
employ roughly 15,000 people, and more during the harvest. A majority of the 65,000
people living in the area make their living directly or indirectly from growing fruit.
If the project were to get the green light, it would deeply affect the life of the
people in the province of Piura, changing it from a region dominated by agriculture to
one dominated by mining. Roughly half the townspeople in Tambogrande will have
to relocate to make room for the one kilometre long open-mine pit next to the Piura
River. A conflict about the mine plans is growing since the Fujimori government
granted the concession without consulting the local population. Fruit growers in the
arid region say that toxic mine wastes from the project will contaminate the (recent
and World Bank-funded) irrigation system. Environmentalists wish to protect the
fragile dry tropical forest and local officials fear the influx of labourers. ‘In the San
Lorenzo valley the people are not rich but they aren’t paupers. They have families
and animals and roads and schools, and they can live’, says Ulisses Garcia, 35, leader
of Tropico Seco, a youth environment organisation. ‘If Tambogrande is exploited it
will rape the nucleus of this area. It is a demented project. If they carry it out we will
have to leave the area. Once the mine happens, who will want to purchase produce
from this area? It will ruin everything. It is crazy to do this mine from a social point
of view’.
Opposition against the project increased over the years, despite promises by
Manhattan Minerals to build new houses and water and sewer infrastructure. With
the Peru mining ministry holding a 25 percent stake in the mine project, the people
18 The sources of this story are www.oxfamamerica.org, www.foei.org and www.ichrdd.ca/site/. See
also icarusfilms.com/new2007/tam.html about the film Tambogrande: ‘A compelling account of how
the global can become painfully local.’
of Tambogrande do not feel protected by their own national government. This

combination of uncertainty and vulnerability has led to frequent demonstrations in
the town, one of which turned violent in February of 2001. Since then, both the
Mayor of Tambogrande and the Archbishop of Piura have spoken out against the
mine. The mayor circulated a petition against the mine and got over 70 percent of
the town to sign it. In 2008, the government decided to block the project, informing
the mining company that it had failed to meet financial criteria necessary for digging
the proposed open-pit gold mine. There are similar stories all over the world, many
with a less fortunate outcome.
13.5 Resource Chains: Material Use and Efficiency
13.5.1 Assessment Methods

The exploration and extraction process is the ‘upstream’ part of the resource chains.
It operates to satisfy the demand for and make possible the use of materials in
the ‘downstream’ part of the resource chains19 (Figure 13.2b). They make up the
metabolism of industrial societies and are the topic of industrial ecology. In this part
of the chain, the resource is transformed in myriad ways and becomes in-use stocks in
appliances, vehicles, houses, offices and so on. Ultimately, it ends up in landfills and
air, water and soils. Several methods exist in resource analysis, from primary resource
to final product use and disposal. Usually, the flows are expressed in physical terms.
For heterogeneous product groups (such as machinery), high-value products (such
as pharmaceuticals) and aggregate sectoral output such as services (such as health
and education, financial or consultancy), monetary units are more suitable. The level
of analysis can be the unit process (such as pumps, mixers, furnaces, evaporation
and cooling units) or a combination of processes (such as a factory). Because of
limited data availability and other reasons, many analyses are done at economic
(sub)sector level, with particular interest in technological change. Also, the entire
economy can be the unit of investigation, for instance, in international comparisons
for benchmarking with regard to efficiency and dependency issues.
There are many analyses of resource use chains for a variety of products and
processes, including agricultural ones. Most analyses are stationary state descriptions
for stocks (inventories) and flows (in- and outputs) in a given year. Increasingly,
dynamic analyses are made. I refer the reader to the Suggested Reading and Useful
Websites for details and results. The most widely used methods are:
r Substance Flow Analysis (SFA): the flows of individual elements and compounds
through society. A typical application is for toxic substances;
r Material Flow Analysis (MFA): the flows of bulk materials through society, which
are faced with depletion or cause environmental impacts during production,
consumption and waste management; and
19 Usually, the words resource use and resource demand are interchangeable. Quantitative historical
analysis equates use and demand by definition. Resource use is the observable quantity in the statis-
tical data bases. Resource demand is by implication equal to use for the past and a nonobservable
expectation and projection for the future. The latter is considered equal to projected use for all
practical purposes (such as investment decisions).
r Life Cycle Assessment (LCA): the resource and environmental impacts of a

product or a service, using MFAs and SFAs for the relevant compounds and a
methodology for weighing inputs, outputs and impacts.
While SFA and MFA usually study material and energy flows at product/process
level, the system boundaries in an LCA are drawn broader. An LCA will usually
include an impact analysis, which is an evaluation of impacts in various categories
such as resource scarcity, ecotoxicity and global warning (Baumann and Tilman
2004). Dedicated software packages for LCAs are available, such as the Chain Man-
agement by Life Cycle Assessment (CMLCA) package. In view of its importance,
energy analysis has turned into a separate discipline with specific principles and
methods (§7.4; Blok 2006).
System boundaries should be carefully specified in order to avoid confusion
about and misinterpretation of the outcomes. Common system boundaries are
cradle-to-factory gate and cradle-to-grave. The system cradle-to-factory gate cov-
ers all steps from the extraction of resources from the environment up to the final
product as it is leaving the factory gate. The system cradle-to-grave goes one step
further and includes the product use and its treatment in a waste management sys-
tem. If the flow to the grave is minimised or even disappears, because almost all
of it is reused within the same or other manufacturing processes, one speaks of a
cradle-to-cradle approach. There are increasing numbers of firms that actively pur-
sue cradle-to-cradle strategies. They hope to exploit first mover advantages and
gain a competitive advantage when resources become scarce, as part of a long-term
strategy. The emphasis shifts from delivering goods to the provision of services and
from a linear chain to a circular metabolism, but the long-term macroeconomic
effects are still difficult to assess.
A recent analysis of the global tin (Sn) cycle illustrates, as one example out
of many, the MFA (Izard and Műller 2010). Tin is already known and used for
more than two millennia, but it is quite rare. The known economic reserves are
only enough for twenty-two years of global extraction at the present rate and are
concentrated in only three countries. The authors have constructed a dynamic model
of the global tin cycle, which is calibrated on the available data for the period 1927–
2005. It simulates the usual stages of production, fabrication and manufacturing, use
and waste management. It is difficult to trace where the tin ends up in the final stages
because it is used in small quantities in many applications: cans and containers, parts
for vehicles, plumbing in construction and solder and tinning in electronics, which
are the most important ones. Tin has accumulated in different stocks throughout
the world economy (Figure 13.9). In total, 17 million ton (Mt) of tin have been
mined from an estimated 28 Mt of resource. Of this, 13 Mt or nearly 70 percent
has ended up in landfills. Much tin is used in short-lived products, notably cans and
electronics, which also end up in dissipated form in the landfills. This explains why
only 2.2 Mt is in products-in-use (plus another 1.9 Mt in association with steel) and
why the recycling rate is quite low. MFAs like this one show that it matters for key
aspects of the life cycle for which products the refined metal is used. The specific
features of resource chains make it necessary to investigate each metal metabolism
separately.
Lithosphere
Tailings & slag
30 In-use stocks
Steel
Landfill
20
Tg
10
Reserve Base
Reserves
1930 1940 1950 1960 1970 1980 1990 2000
Year
Figure 13.9. Cumulative global tin stocks 1927–2005 (1 Tg = 1 Mt) (Izard and Müller 2010).
13.5.2 Dematerialisation: The Intensity-of-Use Hypothesis

What is driving use of materials and fuels?20 Who are the users and for what purposes
and how efficiently do they use it? And, what is the role of material and fuel substitu-
tion and material reuse and recycling? Usually, income (GDP/cap) is considered the
driving force behind the use of materials and fuels. Econometric analyses focus on
income and price elasticities to better understand the determinants (§11.3; Common
and Stagl 2005). The following introduces two metamodels, which are derived from
country-based empirical analyses and are used to extrapolate future trends.
As early as the 1950s, it was observed for the United States that the inputs
of materials and semi-finished goods becomes relatively less important vis-à-vis
nonmaterial inputs such as labour, capital and services. The concept ‘intensity of use’
was introduced to draw attention to the phenomenon that over time (time-series)
and across nations (cross-country) the use of materials and fuels per unit of GDP
tends to rise and then fall as function of income (Malenbaum 1978). This bell-shaped
form of the intensity-of-use curves has subsequently been called the intensity-of-use
(IU) hypothesis and is associated with the notion of dematerialisation. It has two
postulates (Bernardini and Galli 1993):
r the IU of a given material (in t/€) tends to follow the same pattern for all
economies: at first increasing, then reaching a maximum and declining as a
function of income21 ;
20 This paragraph consistently speaks about materials (from minerals) and (fossil) fuels. Because fossil
fuel is still the dominant source of energy, resources can still be equated to material and fuel stocks
and flows. If renewable energy gains in importance, this is no longer a valid assumption.
21 The inverse of intensity of use is resource productivity (in €/ton). Some materials and most fuels are
widely traded, and there can be substantial differences between production and consumption in a
a) Oceania, USA Africa, Brazil, Middle East, Mexico
C-Europe, Russia, Ukraine China, India, Indon, Korea, SE Asia
Japan, W-Europe Curve fit
const 300 kg/cap/yr const 600 kg/cap/yr
0.2
Intensity of cement use (kg/1995 ppp$)
0.15
0.1
0.05
0
0 5000 10000 15000 20000 25000 30000
GDP/capita (1995 ppp$)
Figure 13.10a. The relationship between cement use intensity (t/$) and income (GDP/cap).
The dots indicate the cement use intensity (t/1995 $) for some regions in the period 1971–
2000. The solid line is the IU hypothesis curve; the dotted curves are two lines of equal cement
consumption per capita (t/cap) (source of data: PBL).
r the maximum IU tends to be lower for countries, which had a later eco-
nomic takeoff, such as Japan and South Korea and the presently emerging
economies.
The empirical data for cement use in the period 1971–2000 in world regions indicate
an overall trend in line with the IU hypothesis (Figure 13.10a). The data can be fitted
to a function of income.22 However, there are large differences between the regions.
United States–style countries have higher IU values than European countries and
Japan, and there are large differences amongst low-income countries; for instance, a
factor five between Korea and Brazil, countries with comparable incomes. A similar
proximate confirmation of the IU hypothesis is found for steel (Figure 13.10b) and
other metals, but for paper and plastics there is no leveling off yet. The inverted
U shape is also visible for total commercial fuel use (coal, oil or gas), but it breaks
down when traditional fuels (biomass) are included (Grübler 1999).
country. The data in this paragraph are all based on consumption or use. Note that the IU hypothesis
implies an income elasticity that declines with the increase in income.
22 The curve used to describe the intensity-of-use and also, in the next paragraph, the emission intensity
as function of income is used in the IMAGE energy model (Van Ruijven et al. 2009).
b) N.America-W.Europe-Japan Brazil, Mexico

Africa, Middle East C-Europe, Russia, Ukraine
India (18) East Asia
Oceania (24)
const 300 kg/cap/yr const 600 kg/cap/yr
0.1
Intensity of steel use (kg/1995 ppp$)
0.05
0
0 5000 10000 15000 20000 25000 30000
Figure 13.10b. Idem, for steel use intensity (source of data: PBL).
Several explanations have been suggested. The first one is the structural change
in economic activity (§4.3). In the early stages of industrialisation, there is a rapid
expansion of material- and energy-intensive infrastructure (roads, railways, dwell-
ings and so on). Economic growth in affluent postindustrial societies, on the other
hand, stems largely from technology-driven, less material-intensive service activities
(education, medical care, hygiene and safety, design and so on). A second explana-
tion is the role of prices and technology: Industrial sectors with high-material inputs
have a large incentive, particularly in a situation of high or volatile prices, to develop
and implement innovations that make resource use more efficient or provide attract-
ive substitutes such as plastics for steel. Conversely, the lack of substitutes is also part
of the explanation why the IU hypothesis does not (yet) hold for paper and plastics.
A third, related factor is that rising prices make recycling more attractive for the
producer while making the solid waste problem more tractable. Because innovat-
ing and recycling require a techno-economic and social-political infrastructure, the
high-income countries are leading in dematerialisation. However, globalisation in
its various manifestations allows the application of the latest technologies in low-
income countries, at least in principle and with the necessary basic infrastructure –
a phenomenon called ‘leapfrogging’. This explains why the IU curves are lower for
the latecomers in industrialisation.
Because so many forces are simultaneously at work, one cannot expect more than
a correlation between intensity-of-use and income. For instance, the intensification
Box 13.3. More than mass alone? Environmental NGOs as well as corporations
and bureaucracies often equate sustainability with the technical and behavioural
options to reduce a person’s use of materials and energy while simultaneously
embracing a culture of consumption. Their sites and brochures are flooded with
data on physical stocks and flows and how to reduce them. An example of such
an obsession with facts is the site energyfacts.bp.com/ze/recycling.aspx, which,
on the virtues of recycling, states that ‘recycling aluminium cans saves 95 per-
cent of the energy used to make the cans from virgin ore. . . . Recycling one glass
bottle can save enough energy to light a 100-watt bulb for four hours . . . Recycling
one tonne of aluminium saves the equivalent in energy of 2,350 gallons of gas-
oline. That’s equivalent to the electricity used by the typical home in a 10-year
period . . . Recycling one glass jar can save enough energy to run a television
for 3 hours. . . . Recycling one aluminium can saves enough energy to run a TV
for three hours – or the equivalent of a half a gallon of gasoline. We use over
80,000,000,000 aluminium soda cans every year. The impact of such data floods
on human behaviour are uinclear – but you should in any case avoid 100-watt
light bulbs.’
of trade flows makes it more difficult to find reliable statistics and less probable to
find country-based correlations. Also, the differences in variables other than income,
such as climate and culture, and the continuous change in material use because of
novel applications – think of computers and mobile phones, which were absent
before 1985 – cause country-specific deviations from the trend.
Often, the IU hypothesis is interpreted as a sign that economies will converge
to roughly equal ‘best practice’ levels of material intensity and that materials and
fuels pose no limits to economic growth. Such a generalisation is unwarranted. First,
the material and energy use per unit of economic activity declines, but in a growing
economy, the absolute use of materials and energy keeps increasing. Second, the
high-tech innovations that are part of dematerialisation require often new and rare
materials. A well-known example is the platinum used in catalytic converters in cars
and another one is the use of rare earth elements for electronic applications and
batteries. Finally, dematerialisation involves the use of substitutes that may have
scarcity problems of their own. The obvious case here is the substitution of metals by
oil- and gas-based plastics. Mineral and fuel scarcities change but do not end. Their
depletion remains an issue in sustainable development.
It seems correct to conclude that ‘while qualitatively correct, the dematerial-
isation model is quantitatively an oversimplification’ (Bernardini and Galli 1993).
But there is another unsustainable part of the industrial economy. The mining
and processing of materials and fuels creates also large flows of ‘waste’, burdening
environmental sinks. Upstream, huge amounts of solid waste are generated in min-
ing and processing processes. One-fifth (20 percent) of global energy- and process-
related CO2 emissions arise from the production and processing of only five materi-
als: steel, cement, plastic, paper and aluminium (Allwood et al. 2011). Downstream,
the discarded materials and burnt fuels accumulate in landfills and air, water and
soils. Let us examine another metamodel.
a) N America & Australia C & S America

Africa OECD Europe
SU / Russian Federation & Ukraine India & Other Asia
Japan, Korea & China Curve fit #high
Curve fit #low
250
emission intensity (kg SO2/yr/cap)
200
150
100
50
0
0 5000 10000 15000 20000 25000 30000 35000

Figure 13.11a. The relationship between SO2 emission intensity (t/cap/yr) and income
(GDP/cap) for some regions in the period 1971–2008. The dots indicate the empirical SO2
emission intensity (t/cap/1995$) and the solid line is a possible curve fit (source of data: PBL).
13.5.3 Richer and Cleaner?

In 1955, the economist Kuznets suggested that income inequality will initially
increase with rising income per capita and then decline – another bell-shaped or
inverted U curve. The hypothesis was that political mechanisms such as voting by
the poor for redistributive taxes and financial mechanisms such as credit markets
could explain it. But research at the World Bank did not find evidence for a Kuznets
curve in a 1960–1990 seventy-three country database. What the data do indicate is
that overall economic growth coincides with growth in the income of the poorest
segments of society. This is the ‘trickle-down percolator’ hypothesis, which states
that economic growth will automatically benefit the poorest segment of society.
In the early 1990s, a Kuznets-type relationship was hypothesised between the per
capita emission of pollutants (in t/cap/yr) and income. It is called the Environmental
Kuznets Curve (EKC) hypothesis. Early work for the World Bank Development
Report in 1992 popularised this idea to the extent that an influential economist
claimed that ‘there is clear evidence that, although economic growth usually leads
to environmental degradation in the early stages of the process, in the end the best –
and probably the only – way to attain a decent environment in most countries is
to become rich’ (Beckerman, in Stern 2004). Again, such a generalisation is not
justified.
What do the facts say? The empirical sulphur dioxide (SO2 ) emission intensity
as a function of income for the years 1971 to 2008 seems to confirm the EKC hypo-
thesis (Figure 13.11a). The early industrialised and recent industrialising regions
in the world exhibit an inverted U shape. Explanations are economies of scale in
b) C & S America S Africa

SU / Russian Federation & Ukraine M East
China & India & Other Asia OECD Europe & Japan
Curve fit N America & Australia
25
emission intensity (kg CO2/yr/cap)
20
15
10
0
0 5000 10000 15000 20000 25000 30000 35000

Figure 13.11b. Idem for CO2 (source of data: PBL).
abatement techniques, more strict regulation as people become more affluent and
changing sector structure and trade patterns. Per capita emission in newly industri-
alising countries are significantly lower than for the older ones, probably because
of the availability of cheap abatement technologies and less coal use. For CO2 , the
curve does slope downwards, but there is as yet no maximum (Figure 13.11b). One
explanation is that CO2 emissions are from essential and rather dispersed sources;
another one is that international competition initiates a race to the bottom in emis-
sion standards that obstructs or at least postpones the decline phase. For both SO2
and CO2 emissions, there is a large spread in the data and, again, a significant differ-
ence between North America and Australia on the one hand and the more densely
populated Europe and Japan on the other. Several researchers have investigated
the EKC hypothesis in an econometrically rigorous way and some find the methods
invalid and the evidence spurious (Wagner 2008). Overall, ‘the majority of studies
have found the EKC to be a fragile model suffering from severe econometric mis-
specification . . . it seems unlikely that the EKC is an adequate model of emissions or
concentrations’ (Stern 2004).
Evidently, the validity of the EKC hypothesis depends on the definition and type
of pollutants and on the specification of causes. Using urban pollutant concentration
data, for instance, leads to a different conclusion than using national SO2 and CO2
emission data. Once again: such a metamodel has some validity in specified places
and periods. Both the IU and the EKC hypothesis may be an aggregate feature of
advancing economies, but for a deeper understanding of mechanisms and policies
one has to broaden the scope towards the material and energy transition, with an
explicit consideration of use efficiency, substitution and reuse and recycling. For
most new chemicals released on the market, the intensity-of-use and per capita
emission probably increase with higher income. The brings us to the very end of
resource chains: the sinks. But again first some stories.
13.6 Stories 455
13.6 Stories
13.6.1 Pearl River Estuary, China23

There are numerous environmental histories, tracing human interference with the
natural environment throughout the chain from mining to accumulation over time
and space. Measuring the concentrations of certain compounds in soils and sedi-
ments yields an archeological print of industrial activity. The sediment quality in
coastal wetlands of the Pearl River Estuary in Guangdong Province in China gives
such a print – and reason for concern. The wetlands are used for land reclamation,
aquaculture and wildlife protection, but meanwhile they also serve as one of the
main ultimate sinks for large amounts of heavy metals discharged from the rapidly
developing Pearl River Delta. QuSheng Li et al. (2007) have investigated the con-
centrations of a series of heavy metals and their chemical speciation. The results
show that the sediments are significantly contaminated by Cd, Zn and Ni with con-
centrations in the range of 3–5, 240–346 and 25–122 milligram per kilogram (mg/kg),
respectively. The sediment quality is no longer meeting the requirements of the
current wetland utilisation strategies and notably Cd and Zn pose a potential health
risk. The sediments have effectively become a metal resource.
The pollution causes social tensions between farmers and city dwellers because
the latter are rich enough to pay for wastewater treatment plants but do not allow
the farmers downstream in the Delta to use the treated water. ‘If we [also] continue
to pollute the water at its source, all the efforts of cities downstream will be wasted’,
according to Liu Chen, director of the Protection Bureau of the Pearl River Water
Resources Commission. Water pollution may be particularly damaging during floods,
when it can contaminate soils and accumulated pollution can be released (‘chemical
timebomb’).
Many suggestions have been made for a cleanup of the water but none has
been implemented so far. In many areas in the world, such events unfold – it is
one of the well-known syndromes (§4.3). It also appears to be part of a transi-
tion: The high-income regions are increasingly cleaning up and applying stricter
emissions standards, whereas most low-income regions are still in the early, rather
high-polluting stage. The latter do benefit, however, from the awareness and tech-
nological development in the high-income regions.
13.6.2 Water As a Commodity: Ban on Bottled Water in Australian Town24

The market has discovered scarcity as something to be desired. In 2007, the Stock
Fund Utilities Department of the Dutch bank ING announced that ‘prospects remain
good. In the states of California and New York energy shortages threaten and it
drives the price up’. Similar advertisements are seen for water. Drinking water used
to be a service ‘freely’ offered by nature. With growing populations and hygiene
23 This story is largely based on the site www.asianews.it/news-en/Pearl-River-pollution-a-

serious-concern-3264.html.
24 This story is based on www.norlandintl.com/blog/2009/04/bottled-water-market-share-volume 28.
html and an article in Le Monde 22 juillet 2009. See also www.circleofblue.org/waternews/2010/
world/peter-gleick-bottled-and-sold-whats-really-in-our-bottled-water/.
standards and increasing food demand, water demand is rapidly becoming a scarce
commodity in many places – or at least, that is a widespread perception. A process of
‘commodification’ has started. Every scarcity offers an opportunity for value added:
Profit and employment, and its exploitation, therefore, tends to be supported by
business people and politicians.
In the first decade of the 21st century, an estimated 200 billion water bottles
are annually consumed and bottled water has become worldwide a U.S. $60 billion
industry. Total bottled water consumption amounted to some 35 billion litres, largely
in high-income regions. In 2007, a U.S. citizen drank, on average, 125 litres of bottled
water per year, up from 7 litres per year in 1976. Massive marketing efforts have
led to this change. Selling one litre of water at a price many times the production
cost (and many times the price of the equivalent from the public water system)
is undoubtedly a marketing success. The bottled water market is expanding at the
expense of soda drinks, as these are increasingly seen as a cause of obesity. ‘Drinking
water instead of three sugary drinks per week for a year will spare you seven pounds
of fat’, says one of the Nestlé ads.
The industry worries that increased activism on the alleged environmental
impact of bottled water can affect sales negatively. The billions of mostly plastic
bottles, also from soft drinks, are now causing waste problems on local to global
scales. This is pushed aside by the International Bottled Water Association (IBWA):
‘There’s little if any measurable evidence that activists have had an impact upon
bottled water sales. Bottled water is well-established and popular with consumers
who rely on its convenience, healthfulness and refreshing taste . . . [but] Consumers
must also be made aware of the bottled water industry’s outstanding record of envi-
ronmental stewardship, protection, and sustainability . . . Bottled water containers
are 100 percent recyclable.’ It is promised that the fully biodegradable bottle is next
in line.
But not everyone is getting in line. In Bungadoo, a 2,000-inhabitant town 120
kilometres (km) south of Sydney, the community council decided to put a ban on
plastic bottled water. The decision was taken in response to a project of a large
company to pump water in Bungadoo and then bottle and sell it. Permission for the
project has been refused, but the decision is reconsidered in the Land and Environ-
ment Court. ‘We are a small community in favour of environmental sustainability’,
says a local shopkeeper. ‘Bundy on tap’ has become the slogan, now that the com-
munity council has decided to install tapwater systems. Australians pay more than
2 A$/litre for bottled water, whereas tapwater is almost free. The industry fights the
idea with the arguments of unemployment and obesity.
13.6.3 Organotin Compounds As Antifouling Agents25

There is little doubt that the accumulation of marine fouling organisms on ves-
sels and man-made structures at sea increases drag and vessel fuel consumption,
with substantial consequences in terms of economics and emissions. Tributyltin
(TBT) compounds and the less widely used triphenyltins are extremely effective and
25 This story is based on Santillo et al. (2001), Tributyltin (TBT) antifoulants: a tale of ships, snails and
imposex. In: EEA, Late lessons from early warnings: the precautionary principle (2001): 1896–2000.
relatively economical as antifouling biocides. Use of TBT in marine antifouling

paints dates from the 1960s and accelerated greatly in the 1970s with the rapid take
up of organotin-based paints by the shipping industry and small boat owners. Some
decades later, the use of organotin antifoulants has now resulted in widespread and
sometimes severe environmental effects. Unusual in the field of ecotoxicology, the
evidence linking cause and effect is irrefutable. The phenomenon is by some con-
sidered as ‘the best example of endocrine disruption in invertebrates that is causally
linked’.
Two regional case histories were instrumental in identifying low-dose effects
of TBT and initiating development of the first regional controls: the collapse of
the shellfish industry in Arcachon Bay (off the Atlantic coast of France) and the
reporting of widespread imposex in dogwhelks, a kind of snail, from southern UK
coastal waters. Early predictions that TBT would degrade rapidly in surface waters
were partly falsified. Half-life estimates in the order of days for eutrophic surface
waters are contrasted with up to several years for residues in nutrient-poor waters and
marine sediments, especially anaerobic sediments. The persistence of TBT, and its
toxicity to sediment communities, raises the prospect of delayed recovery of damaged
ecosystems and represents an unpleasant legacy for the authorities responsible for
dredging operations. Organotin bioaccumulation was also underestimated, and the
specific role of biofilms in enhancing bioaccumulation and toxicity has only recently
been recognised. Accumulation in top predators was simply not envisaged.
13.7 The Sink Side: Environment and the Industrial Economy
13.7.1 Resource Chains: The Sink Side

Human activities have always interfered with ‘undisturbed nature’. It shows up
most conspicuously in the legacy of deforested lands and abandoned mining areas in
Europe and the United States. Copper and other metal mining pits scar many land-
scapes in South America, Africa and Asia. Oil spills did and do happen, polluting
the oceans and coastal waters and rivers. Less visible are the large emission flows of
carbon, sulphur, nitrogen, phosphorus and other compounds into air, water and soil
compartments. Previously unknown substances enter into the natural environment.
The following looks in more detail further downstream the resource chain.
Table 13.3 lists the number of substances/compounds that are emitted into the
biosphere as regular part of the industrial economy. At the global level, many emis-
sion rates follow the familiar trend of exponential growth, with doubling emission
rates every two or three decades, but in the high-income regions some emission
rates are going down since the 1990s (cf. EKC). Some substances have a long resid-
ence time, because they stay for long periods in the atmosphere and are only slowly
absorbed in soils, lakes and sediments and, ultimately, in oceans and rock. Others
may decompose within days, or rain out onto rivers and land and accumulate in
organisms. Natural cleaning mechanisms help to mitigate harmful impacts by break-
ing down or immobilising the substances. It is one of the ecosystem services (§9.7).
Are these substances pollutants? The word pollution is rooted in Latin-Greek words
for filth, dirt and mud. It originally meant to contaminate or desecrate and was only
in the 19th century for the first time used in regard to the natural environment.
Table 13.3. Some important anthropogenic emission in industrial societies and an indication of their
emission rate in the EUa
Emission rate
EU27 (kton/yr)
Compound (‘pollutant’) 2009 Main impacts
Acidifying compounds SO2 : 5015 Damage to animals and vegetation, and buildings
(SO2 , NOx , NH3 . . . ) NOx : 9374
NH3 : 3770
Eutrophying compounds N: 156b Decrease in water transparency (increased turbidity);
(N, P . . . ) P: 31b fishing harvest loss; algae growth; ecological impacts
such as on phytoplankton
Aerosols, particulate matter PM2.5 : 12093 Adverse impacts on human health, depending on
(PMx ) PM10 : 1971 particle size
NMVOC 7761 Smog (from VOC and NOx ) causes tropospheric
ozone (O3 ) concentration increase with damage for
humans, ecosystems, crops and materials
Heavy metals (Cd, Pb . . . ) Cd: 0,1 Adverse impacts on human health and ecosystems (Pb,
and persistent organic Pb: 2,05 Cd, Cu, Hg, dioxins and so on)
compounds (PCBs, TBT) PCB: 0,003
Greenhouse gases (CO2 , CO: 24073 Enhanced radiative forcing, causing rise in
CH4 , N2 O . . . ) temperature and sea level
Radioactive compounds Adverse impacts on human health and ecosystems
a Source: www.eea.eu.
b Discharges of nutrients from urban waste water treatment plants for eight EU countries.
Pollution is sometimes a too value-laden term. All human activity interferes

with the natural environment starting with breathing, which converts oxygen (O2 )
into CO2 . But the industrial economy has intensified some of the waste flows already
characteristic of agricultural economies, such as some metals. It has enormously
increased the flows of fossil fuel related elements (C, S and N) and it has created a
large series of new substances and compounds, many of which did not occur on Earth
before. The monitoring and investigation of substances and their fate and effects is
the essence of environmental science. How to evaluate this part of the industrial
economy in the context of sustainable development?
To understand the consequences of the anthropogenic emissions into air, water
and soils for the function and quality or health of ecosystems and human popula-
tions, one usually has to go down to the local scale. It requires model simulations
of the spread of substances in space and time, the chemical reactions involved and
the harmful interferences caused. In environmental science jargon, it is about emis-
sion, immission and exposure rates and about the probability of harmful impacts.
Within a pressure-state-impact-response (PSIR) framework (Figure 8.5), one can
then identify responses to adapt to and prevent or reduce the consequences. It makes
sense to distinguish between relatively short-term and local forms of pollution and
long-term global interferences. A crude classification of the physical outflows of
substances in human activities can be based on four dimensions:
r impact concentration: what is the ratio between the spatially distributed sub-
stance concentration and the local natural concentration at which undesirable
interference with natural and human life occurs;
r degradation rate: at which rate is the substance broken down or otherwise made
inert in comparison with the relevant natural processes;
r ‘strangeness’: to what extent is the substance alien to natural stocks and flows
and, therefore, a priori difficult to evaluate as there is no natural equivalent; and
r ‘waste’: at what net cost and effort can the substance be kept inside the industrial
system by reconfiguration, reuse or recycling?
Measured along these criteria, one can distinguish the extremes. Radioactive
plutonium (Pu) qualifies as an extreme, where it is known to cause cancer at very low
doses, occurs in nature at extremely low concentrations and is toxic for millions of
years. At the other extreme are discarded aluminium objects, as it is abundant in the
Earth’s crust and the human use rate is still rather small in comparison with natural
processes. In between are thousands of substances. This book does not intend to
give a complete overview, but instead focuses on some problems of importance in a
sustainability context. I refer the interested reader to the Suggested Reading.
13.7.2 Enduring Environmental Problems

In the 1960s, air and water pollution became dramatically visible in North America
and Europe, because in many places the natural removal mechanisms were no longer
sufficient to balance/buffer pollutant emissions. Amongst the first substance known
as industrial pollutants with long-range and long-term effects are the acidifying
compounds (sulphur dioxide SO2 , nitrogen oxides NOx and ammonia NH3 ). The
wind disperses these substances, that mainly result from burning fossil fuels, into the
air and cause harm, damaging the health of people and trees. They cross borders
and give rise to international transboundary long-term environmental problem. It
became all too obvious that the atmosphere is a common pool resource (CPR)
(§5.4).
Evidence in the 1980s suggested that British emissions were to blame for the
acidification of Swedish lakes. Similar long-range effects were identified in the United
States. This led to successful policies to reduce acidifying emissions. Model-based
negotiations were instrumental in the design and implementation of these policies
(Haas et al. 1993; Hettelingh et al. 2009). In the Netherlands and Europe, large
reductions have been achieved for SO2 , but reduction of NOx and NH3 turns out to
be more difficult. Scientists have established safe levels with respect to soils (‘critical
loads’) and these will still be exceeded by 2020 in parts of northwestern Europe.
Emerging economies such as China and India are rapidly facing the same problem,
as their use of fossil fuels increases and the health and environmental impacts become
evident.26
It has been said, in retrospect, that the risks and damage of acid rain in Europe
are less severe than initially feared. But what would have happened without these
policy-induced reductions is hard to say. A recent, careful analysis of 30-year acid
rain history in the Netherlands concludes that ‘in the past, no unnecessary measures
26 The maps on the sites www.geiacenter.org and www.eea.europa.eu/data-and-maps are good illus-
trations of past changes in emissions in Europe and in Asia.
have been taken to counter acidification. The seriousness of the acidification prob-
lem has been visibly reduced, although it has definitely not been solved yet’ (PBL
2010c). In retrospect, policies were an adequate application of the precautionary
principle. Although some impacts, for instance, on forests may have been exagger-
ated, the underlying science has been proven right. Also, the evidence is there that
the benefits of avoided damage are substantial. Acid rain did inflict damage to life
in tens of thousands of lakes and streams in Norway and Sweden, many previously
with salmonids in great numbers. Natural conditions come back only very slowly
despite the billions of Swedish kronor spent on liming these lakes for over twenty
years. At the same time, it appears that the costs of emission reductions were over-
estimated because the success of cost-reducing innovations was underestimated. It
has become a famous case study about the interactions between pricing, regulation
and innovation.
Finally, and important in a sustainability science context, the emissions of sul-
phur compounds and also, particularly, nitrogen compounds are shown to have other
than acid rain effects and not only at the regional, but also at the global scale. For
instance, exposure to nitrogen in the air reduces life expectancy, deposition of nitro-
gen reduces biodiversity and too high concentrations in drinking water increase the
risk of cancer according to the European Nitrogen Assessment.27 The emission of
NOx in the stratosphere by airplanes is affecting the Earth’s radiation balance. In
view of the expected impacts and the increase in fossil fuel combustion, stringent
international acidification policy measures are rightly an integral part of sustainable
development.
Another well-known pollution story is about chlorofluorcompounds (CFCs). In
the 1930s, it was discovered that CFCs are useful in a host of applications: pro-
pellants and foams, refrigerators, air conditioning apparatus and fire extinguishers
(Haas et al. 1993, Bryson 2003). In the 1970s, it was discovered that the concentration
of ozone molecules in the upper atmosphere (stratosphere) above Antarctica had
fallen dramatically (the ‘ozone hole’).28 A major consequence is increased UV radi-
ation on the Earth’s surface, which has harmful effects for human health and crop
growth – the two best known effects. By 1980, it was convincingly proven that CFCs
were the culprit. By 1990, the industry in the industrialised countries no longer res-
isted plans to stop CFC production. The Montreal Protocol is considered one of
the great successes in international environmental negotiations (Parson 1993). But
it is also a warning: History shows the long delays in understanding the problem
and reaching agreement and how long it takes before the consequences show up. In
this case, preindustrial ozone concentrations are not restored before the year 2060,
the incidence rate of skin cancer in places like Australia will continue to rise during
the 21st century and adaptation is the only option for the next generation. Both the
health risks and the fear of it are reducing the quality of life of humans on Earth.
Some effects, such as the genetic effects of increased UV radiation from ozone
depletion, are so difficult to assess and so beyond imagination that it is hard to
27 www.nine-esf.org/ENA.
28 Initially, it was overlooked because the computer had truncated the observation as an impossible
outlier.
agree on any kind of collective action. The result is an ongoing controversy that
reflects the divergence in worldviews. This is most obviously the case with the largest
environmental issue of our times: the risk of climate change due to emission of CO2
and other greenhouse gases (GHG) into the atmosphere (IPCC 2007; PBL 2009).
This so-called (enhanced) greenhouse effect – or its popular term ‘global warming’ –
is dominating many sustainability debates and environmental policy discourses.
It is now beyond doubt that there is a steady increase in the atmospheric con-
centrations of gases which are known to ‘trap heat’, such as CO2 , methane (CH4 )
and dinitrogen oxide (N2 O). Deforestation reduces the natural fixation rate of CO2
in photosynthesis and has a similar effect. The resulting change in radiative forcing
is expected to cause a rise in global average surface temperature. How much and
at which rate is still quite uncertain because of divergent estimates of the climate
sensitivity, amongst others. But there is broad agreement that stabilisation of the
temperature rise to 2◦ C since the pre-industrial era – which is the stated EU target –
can only be reached with a reasonable probability if world greenhouse gas emissions
are drastically reduced (Meinshausen 2006). Climate change will have a variety of
impacts on local temperature and precipitation patterns, but here too are still many
uncertainties. While the trends in global average near-surface temperature are in
agreement with scientific expectations, a possible change in frequency and intensity
of extreme events in the last decades is more difficult to establish and still controver-
sial. The severe floods and droughts in parts of Asia and Africa in recent times are
an indication of the suffering that changes in rainfall patterns can cause for millions
of people. In the long term, significant sea level rise is anticipated, with devastating
consequences for the large coastal megacities.
People can respond basically in two ways: adaptation to actual or imagined
impacts and mitigation in the form of emission reductions. Some adaptation will
happen anyway, because some further rise in temperature is already inevitable.
Both the impacts of climate change and the cost at which it can be mitigated, that
is, at which greenhouse gas emissions can be reduced, differ amongst countries. This
makes it a global social dilemma with many uncertainties and different perceptions
and interests (§5.2; de Vries 2010). The payoff matrix has a complicated and dynamic
structure that can give rise to a rich variety of strategies for the different actors
(Table 10.2b). It spells a continuation of the already difficult negotiations. It also
means a proliferation of interpretations and controversies.
In 1992, the United Nations Framework Convention on Climate Change
(UNFCCC) was signed. In 1997, the Kyoto Protocol was adopted, and in 2005 it
entered into force. Its major objective was to stabilise worldwide emission of green-
house gases, as a first step to much more drastic cuts later on. The argument was
and is that the costs made now prevent presumably larger and irreversible dam-
age later. The value of the discount rate becomes a crucial point in this reasoning
(§10.4; Dasgupta 2008). Another argument is that emission reduction has a series
of cobenefits, such as reduced negative health impacts from air pollution, and helps
realise other sustainable development goals (Metz et al. 2002). Since then, the cli-
mate science justifying such drastic and, in the public perception, costly measures
suffered from a loss of legitimacy. This was fuelled by active campaigning by fossil
fuel interests, but it could happen because scientists and NGOs underestimated the
Box 13.4. Reuse: the road towards a circular economy. Following an earlier
systems-oriented definition of sustainability, material use chains have to become
closed to the extent that finite sources are not depleted without generating substi-
tutes and inflows in the environment do not exceed the natural decay rate. This is
currently almost never the case. Metals and plastics in ‘waste’ flows are dispersed
across the planet and accumulate to often hazardous concentrations in parts of the
biosphere. For instance, polyethylene, the most widely used plastics, is produced
at a rate of 80 Mt/yr and is showing up in many places, including in large soup-like
accumulations in the oceans. Recycling rates are still small. The post-industrial
economy has to be a circular economy, in which many manufactured goods are
reused somewhere along the chain. From an exergy point of view, it is preferred
to prevent dissipation and introduce reuse as much as possible upfront, near the
mines and the factories, or as secondhand goods. This is not easily organised in
an economy oriented towards profits and novelty, so most reuse is done at the
end of the chain by recycling (glass, paper or plastics) and incineration.
In the Netherlands, households produced 9 million tons of ‘waste’ in 2010,
double the amount produced in 1970. The total levels off, but the fraction of
plastics has tripled in this period. About half of it is collected separately. A
number of towns are now experimenting with plastics recycling because cleaning
up litter is costly. Citizens are responding positively, as they did with glass and
paper recycling. The plastic is compressed in the trucks that empty the containers
and then transported to large installations for separation and reuse in construction
materials and other applications. It is only accepted if it satisfies certain quality
criteria. One of the innovations is to put sensors in the containers and only collect
waste from full containers. A recently opened installation in Rotterdam applies
high-tech separation technology to produce PET, PE and PP fractions that are
sold to automotive and other manufacturers. Waste collection and treatment
is still largely in the hands of the municipalities, but it is one area where public-
private partnership works. As in metal and other recycling activities, raw material
price fluctuations are one of the obstacles. A deposit-refund scheme, particularly
for the huge quantities of bottles, is perhaps a better solution, because it is more
upfront and captures a larger fraction of value-added. Glass bottles with deposit is
often still the most environmentally friendly option – and aluminum cans the least.
But the main maxim should be: reduce before reuse, expressing the challenge of
using materials more efficiently while containing the rebound effect.
intricacies of postmodern science (§8.6; Metz 2009). In 2009, at the Conference of

Parties (COP-15) in Copenhagen, the clash between the different views of and solu-
tions for the problems and the certainties and the uncertainties was so vigorous that
‘global warming’ has again received a more modest position in the manifold trends
and events that make up the sustainability challenge. With many countries not hav-
ing met the targets for the first Kyoto commitment period (2008–2012), a new global
treaty has to be negotiated unless the world community decides to wait for the results
of local bottom-up initiatives and measures. For a more in-depth discussion, I refer
the reader to the Suggest Readings and Useful Websites.
13.7.3 Persistent Chemicals

More difficult to assess but possibly equally serious in their long-term consequences
are the numerous emissions, leaks and disposals of toxic persistent substances. They
range from oil spills and mine waste to industrial emissions in water and air, from
dissemination of food additives and exotic metals to garbage in the streets and
plastic in the oceans. One of the best documents regarding pollution in the context
of sustainable development is a study for the European Environmental Agency
(EEA) that was published in 2001 called Late Lessons from Early Warning: The
Precautionary Principle 1896–2000 (EEA/Harremoes et al. 2001). It contains a series
of investigations into the use, neglect and possible misuse of precaution in dealing
with occupational, public and environmental hazards. The fourteen cases provide
interesting and instructive insights about how one may try to deal with risk, scientific
uncertainty and ignorance. Two dozen environmental scientists tried to answer three
questions for these case studies:
r When was the first credible scientific early warning?
r When and what were the main actions and inactions on risk reduction taken by
regulatory and other responsible bodies?
r What were the resulting costs and benefits of the actions or inactions?
A fourth question was: What can we learn from the answers? The study takes
explicitly a long-term perspective.
The issues are complex and the impacts only visible in the course of decades,
so it is difficult to get strong knowledge (§8.3). There are always unexpected and
unintended consequences and conflicts between economic and social interests. The
assessment of (avoided) costs and benefits is therefore difficult and controversial
and a 100-year perspective is needed for the evaluation of the consequences of the
persistent chemicals manufactured at ever larger rates. The theoretical framework
for the study was the precautionary principle:
The main element of the precautionary principle . . . [is] a general rule of public policy
action to be used in situations of potentially serious or irreversible threats to health
or the environment, where there is a need to act to reduce potential hazards before
there is strong proof of harm, taking into account the likely costs and benefits of
action and inaction.
Its application reflects differences in worldviews and, more specifically, in people’s

attitudes towards risks, technology and nature (§6.3). The principle can be viewed
as a necessary consequence of postnormal science.
There is often a gap of decades between the first ‘early warning’ and decisive
action. A vivid illustration is the observation by a UK factory inspector made in 1898
about the evil effects of asbestos and a UK government decision, one century later,
to ban asbestos on the evidence that past asbestos exposure was expected to lead to
250,000–400,000 asbestos cancers in Western Europe in the period 2000–2035. Also
in the other cases, the early warning was in retrospect justified. Ionising radiation was
discovered by the end of the 19th century and the perception of its effects has slowly
developed ever since. Protection standards are firmly established in legislation, but
the long-term impacts in the form of cancers are still uncertain because they are only
known from undesirable bombing and accidents. There are now over 400 nuclear
power plants, many of them in need of dismantling and replacement. The handling
of the associated stocks and flows of radioactive material will probably be one of
the exacting tasks in the first half of the 21st century. Since 1897, benzene is known
to be a powerful bone marrow poison. Increasingly used for tyre and explosives
manufacturing, benzene had caused by 1939 widespread poisoning of workers with
leukaemia as one of the consequences. After more convincing evidence of benzene
as a cause of leukaemia, the standards became much more stringent with the Benzene
Decision in 1980 in the United States in the context of the Occupational Safety and
Health Act (OSHA).
Another case study is about polychlorinated biphenyls (PCBs). First synthesised
in 1881, its damage to the skin was known by 1899. Mass production started in 1929,
and by 1965, it had become a major public issue because of its danger to humans and
animals. PCBs were never intentionally spread into the environment. They were used
to replace more flammable, unstable or bulky products, in particular, ingredients of
electrical apparatuses. The first crucial observation was the discovery in 1966 that the
Baltic Sea fauna had remarkably high concentrations, leading to the hypothesis that
these compounds were apparently persistent in living tissues with bioaccumulation as
a result. Experiments in the 1970s showed clear correlations between pathological
uterine changes in female seals and concentration of contaminants, particularly
PCBs. A link with another pollutant, DDT, was suspected. Other evidence came
from an incident with rice oil in Japan: PCBs were found to be the cause of the ‘rice
oil disease’ (yusho). Their persistency led to accumulation in river sediments, with
concentrations in the range of 12 to 24 mg/kg in Rotterdam Harbour. Tests in 1977
and 1988 indicated significant PCB accumulation in eels and other fish. The Dutch
population was found in the 1980s to have the highest PCB contamination levels in
the world.
In the 1970s and 1980s, new observations and experiments – sometimes unin-
tended, as in another rice oil poisoning accident in Taiwan – led to a refinement of
the scientific models. For instance, it was found that the number and position of the
chlorine atoms played an important role, that the PCBs change during bioaccumula-
tion and biodegradation, which makes them more dangerous, and that the dose and
timing of exposure, in particular in pre- and postnatal children, are important factors.
By now, most of the PCB chain is understood. The main producer, Monsanto, ini-
tially denied any health impacts, but mounting evidence led to a gradual phasing
out of PCB production between 1972 (Sweden) and 1995 (world). But there will
be a toxic legacy for decades to come: The 1.5 million tons produced between 1929
and 1988 (excluding USSR and China) have spread via the atmosphere to even the
remotest places and are still spreading from the equipment in which they were used.
The other case studies in the report are about fisheries, halocarbons and several
other compounds, antimicrobials and hormones and the mad cow disease. They
show that the dangers did not only affect workers but also the public at large,
as with a carcinogenic compound such as benzene that is a hazard for all those
exposed to gasoline. Their study also identified cases of a (near) false alarm. In
their recommendations, the authors of the study stress the importance of long-term
monitoring, evaluating alternative options, engaging also ‘lay’ and local knowledge
and ensuring regulatory independence, amongst others.
13.8 Perspectives on the Industrial Economy 465
The stories reveal a pattern in the political process. When explicit risk analysis
and evaluation became part of legislature and mandatory, corporations began to
hire consultants to downplay or put in doubt the scientific observations about tox-
icity and impacts. The conflictual nature of this feature of the industrial economy
became clear in corporate lobbying with antipublic health campaigns and policies
to circumvent environmental laws. This is possible and understandable in view of
the social dilemma character and the complexity and value-ladenness. How much
are the individual and the collective willing to pay to reduce the risk that – as yet
unknown – members of the population prematurely die of cancer? How much cer-
tainty is possible, to what extent can medical advances repair or mitigate the health
damage, how far in space and time should the effects be considered? These questions
will remain part of the quest for a sustainable future.
13.8 Perspectives on the Industrial Economy

The interpretations and expectations about non-renewable resources and envi-
ronmental degradation differ widely amongst the various stakeholders. It can be
illustrated with news items, that are a mixed bag of good and bad stories:
r Over 300 factories in Brazil produce ethanol from sugarcane on about 8 million
ha of land, resulting in over 22 million tons in 2008. Ethanol is a cheap substitute
for imported oil, does not compete with food and offers the prospect of further
expansion. It makes up one-fourth of gasoline use in Brazil – is it the solution to
fossil fuel depletion and climate change?
r ‘Congolese are learning to sleep with an empty stomach’, says a laid-off mine
worker. The Congo has abundant mineral resources. Since 2003, the mining
industry grew spectacularly because of the booming demand from Asia. But
with the financial crisis in 2008, investors left the country and a quarter of a
million people lost jobs and income. The mineral wealth is also a cause of
continued war and corruption.
r High oil prices make it economical to squeeze crude oil out of shale oil deposits,
and U.S. oil production is expected to increase in the coming decade. But the
new techniques, such as hydraulic fracturing that fuel the shale gas boom, cause
large greenhouse gas emissions and create other environmental risks. In the
search for the ‘black gold’, the North Pole becomes Oil Pole and Kazakhstan oil
becomes part of a global scramble.
r ‘Garbage searchers in Rio hard hit by falling prices of recyclable materials.’
Thousands of people search the megacities’ municipal waste dumps for recyc-
lable material, as if it were the stomach of a huge organism. ‘Their lives are hard
and vulnerable . . . they have no security nets to fall back on.’ They started to
collect frying oil in response to falling prices.
Controversial supply-side issues are: How much resource is geologically available?
How much can be extracted and at what cost? How large are the possibilities to
substitute scarce resources with more abundant ones? Less controversial, but still
uncertain are the availability of oil and gas and of rare minerals and the distribution
of the benefits from resource exploitation. Controversial too are the seriousness of
enduring environmental problems such as greenhouse gas–induced climate change
and the spread of persistent chemicals. Appreciation and interpretation of the need
and cost of effective waste and reuse/recycling and of environmental and social
impacts of mining also differ, sometimes widely.
Governments and corporations tend to see resource and environment issues as
primarily global. They propose rational solutions, in line with the modernist world-
view, that are based on some, sometimes hotly debated, mix of market incentives and
government regulation. For many citizens, it is mostly the local aspect that counts:
Am I affected by air pollution? Do I have to spent ever more money on gasoline for
commuting? Is my health affected by the chemicals in the water? Can I insure my
house against floods? The larger context is not or at best partly visible and full of
uncertainties. Plus, who is to blame? In a highly connected world, a fair distribution
of the costs to reduce risks and help the victims is no easy matter. Negotiations will
be tough and it is easy to create a sense of unfairness and scapegoats. As in previous
chapters, I present a table with statements and their evaluation from the four differ-
ent worldviews introduced in §6.4 (Table 13.4). The reader is again invited to give
his or her own evaluation.
13.9 Summary Points

The industrial regime brought profound changes to the world and a significant part
of the world population is still in the midst of it. Cheap energy based on fossil
fuels provided the engine, but it becomes less accessible now that ever more people
aspire for the industrial form of modernity. Features of the industrial economy to
remember in a sustainable development context are:
r the Earth’s crust contains huge amounts of elements that are exploited for
industrial purposes. There are reasons to assume that resources are large and
that the sink side – interference with natural stocks and flows – is a more serious
constraint than the (re)source side;
r basic elements of resource dynamics include the classification of resources
according to size and probability, the supply cost curve as a representation of
resource size as function of extraction cost, and the interplay between depletion
and learning;
r fossil fuels are essential for today’s advanced and emerging industrial eco-
nomies. The transition to less accessible deposits in combination with growing
demand will cause economic and geopolitical tensions. The role of combus-
tion products in changing the Earth’s climate is slowly becoming another stress
factor;
r resource use trends depend on a complex mix of factors. There are clear signs
of lower resource intensity (t/€) with rising incomes (IU hypothesis), but the
causes are only partly understood. The hypothesis that a society can outgrow its
environmental problems (EKC hypothesis) is at best partially valid;
r the industrial economy is causing large outflows of substances in the natural
environment. Because some substances have long residence times, are not or
only partly broken down and move across compartments and borders, they
pose serious challenges for sustainable development. Examples are the enduring
environmental problems of acidification, ozone depletion, the enhanced green-
house effect and persistent chemicals.
Table 13.4. Worldviews and non-renewable resource futures
Non-Renewable Resources
Statement 1: Before 2025, the world is faced with severe oil shortages and associated economic crises.
It may happen, but not Probable so we should Probable so let us Probable and possibly
because oil resources focus on energy prepare for it. Rising dramatic for the
are depleted but security from other oil prices will low-income regions
because of sources (nuclear, stimulate energy that struggle with oil
geopolitical tensions coal). efficiency and price–induced
and lack of frugality. inflation.
production
investments.
Statement 2: Ensuring and safeguarding access to rare earth and other strategic mineral resources should be a
strategic policy priority.
Correct. Strategic I don’t know. Business I don’t know. Society It is a UN responsibility
materials are essential and government can no longer to arrange
for modern industry, should take timely function without multilateral
so governments action – beware of the high-tech and this agreements on scarce
should make it a top Chinese. vulnerability is now resources. Limiting
priority to safeguard becoming manifest. state sovereignty is to
their access and Simplify! be considered.
supply.
Statement 3: Recycling of metals such as aluminium and copper can be left to the market.
In principle, yes. But It is more difficult for We should first stop to This is an important
citizens have to be new products like consume and waste so strategy.
informed and computers. It is a much! Reuse, buy International
secondary metal regional but difficult secondhand. And if coordination is
prices stabilised. The market. recycling is needed, it needed to ensure
government has a is a community task. quality standards for
role here. labour conditions and
product quality.
Statement 4: Attention should be given to cancer from air pollution, more than to cancer from smoking.
Most important is the There are too many Modern man poisons his Both forms of cancer
choice of the environmental interior and exterior should be prevented.
consumer: if he wants regulations. The links in the name of But air pollution is
a cleaner car, car between air pollution progress and money. more of a social
manufacturers will and lung cancer are Regulation and dilemma, so stricter
respond with lower still unclear. market will not be international
specific emissions. enough – our attitude standards for traffic
has to change. are to be applied.
Statement 5: The Earth is on its way to the next Ice Age – global warming is at most a temporary slowdown.
It is an illusion that we It is good that some An irresponsible Irresponsible. There is
can control the future scientists put climate attitude. Life is quite some evidence
evolution of the change in perspective, precious, and humans that human-induced
world. Let’s trust our if only to calm down do not have the moral climate change may
ingenuity and the alarmists. Most right to interfere in trigger catastrophic
adaptability! climate policies are this way. system change. This
costly and will have large
unnecessary. consequences for our
(grand)children.
Geologists have estimated that if a single thimbleful of water were poured into a
river, after only a few years the circulation on a global scale would be so complete
that a similar thimbleful of water taken from anywhere on Earth would contain some
molecules from the original thimbleful. Others have expressed much the same idea
regarding atmospheric circulation by estimating that every breath of air we inhale
contains molecules that were once breathed by Galileo, Aristotle, and the dinosaurs.
– Chaisson 2001:193
If we look around we can see evidence on all sides of the wonderful revolution
which oil has wrought upon our very existence. It has brought in the motorcar for
pleasure and commerce . . . it has contributed to the conquest of the air, and is even
revolutionising movement across the seven sees . . . it has directly provided us with
the fastest and most formidable capital warship afloat . . . It may be said in very truth
that the community is becoming daily more and more dependent upon petroleum;
it promises to survive as the one force which will drive the world.
– Talbot, All About Inventions and Discoveries. Cassell and Company Ltd.,
London, 1916
SUGGESTED READING
Collection of papers in industrial ecology with theory and practice.

Ayres, R., and L. Ayres, eds. A Handbook of Industrial Ecology. Edward Elger: Cheltenham,
UK, 2002.
An introduction to the concepts and methods of energy analysis.
Some chapters deal with industrialisation, energy and materials, and life cycle assessments.
Boersema, J., and L. Reijnders. Principles of Environmental Sciences. New York: Springer
One of the first, largely theoretical texts on resource economics.
Dasgupta, P., and G. Heal. Exhaustible Resources and Economic Theory. Cambridge: Cam-
bridge University Press, 1979.
Comprehensive treatment of learning curves for a series of energy supply and demand
technologies.
Junginger, M., W. van Sark and A. Faaij, eds. Technological Learning in the Energy Sector.
Cheltenham, UK: Edward Elgar, 2010.
An extensive treatment of natural resource economics in a largely neoclassical framework.
Economics. Harlow, UK: Pearson Education Ltd, 2003.
This paper is a good example of a strategic scarcity analysis of the world market for metals
(copper).
Rosenau-Tornow, D., P. Buchholz, A. Riemann and M. Wagner. Assessing the long-term
supply risks for mineral raw materials – a combined evaluation of past and future trends.
Resources Policy 34 (2009): 161–175.
An entertaining account of the history of oil in all its aspects
Yergin, D. The Prize – The Epic Quest for Oil, Money and Power. London: Simon & Schuster,
1991.
USEFUL WEBSITES
Resources
r www.globalpolicy.org/security/docs/minindx.htm is a site on the exploitation of natural
resources (diamonds, oil and gas, water, timber and minerals).
r webelements.com/ is an informative site about the elements in the periodical system.
Appendix 13.1 The Crustal Abundance Geostatistical (CAG) Model 469
r www.peakoil.com is the site on world (peak) oil production; there are many sites on oil
and gas (peak) production, for instance, www. aleklett.wordpress.com.
r www.iea.org is the site of the International Energy Agency in Paris, publisher of the
annual World Energy Outlook (WEO) and with extensive reporting on energy trends
and outlooks. See also www.bp.com for the Statistical Review of World Energy.
r minerals.usgs.gov/minerals/ is the Minerals Information site of the U.S. Geological Survey
(USGS), with well-organised data on on a commodity and country basis.
r cml.leiden.edu/software/software-cmlca.html is the site where the Chain Management by
Life Cycle Assessment (CMLCA) package can be downloaded free of charge.
r www.oxcarre.ox.ac.uk is the site of the Oxford Centre for the Analysis of Resource Rich
Economies, on the relationship between resources, economic welfare and governance.
Environment
r www.geiacenter.org/ is the site of the Global Emissions Inventory Initiative (GEIA) with
large datasets and maps on emissions worldwide.
r www.unep.org/ is the site of UNEP with data and documents on global environmental
issues, in particular the UNEP Yearbook on www.unep.org/yearbook/2011/.
r glossary.eea.europa.eu/ is the site of the European Environmental Agency (EEA) with
brief explanation of terms in the area of environmental economics and policy and down-
loads of reports including the report Late lessons from early warning can be downloaded.
r www.milieuennatuurcompendium.nl/ is the site with environmental data operated by the
Netherlands Environmental Assessment Agency (PBL) (in Dutch).
r themasites.pbl.nl/en/themasites/ is the site of the environmental models operated at PBL
and gains.iiasa.ac.at/ is the site of the GAINS model on air pollutants.
r www.ipcc.ch/ the site of the Intergovernmental Panel on Climate Change (IPCC), with
downloads of the five Special Reports and other IPCC publications.
r www.atmosfair.de/en/about-us/what-is-atmosfair/ is an example of a site that offers an
emission calculator for individual air travel and offers to compensate emissions.
Appendix 13.1 The Crustal Abundance Geostatistical (CAG) Model

The Crustal Abundance Geostatistical (CAG) model is constructed from three
assumptions (de Vries 1989):
r ore deposit formation processes (igneous activity, erosion-sedimentation,
solution-precipitation) can be viewed as sequences of discrete separation events;
r with each separation event, a mass of M units of material with average concen-
tration Gav is divided into two equal parts, one enriched, one depleted; and
r the separation events can be stylised in a series of steps, with each separation
step characterised by a certain average efficiency q (0 < q < 1).
The average concentration of element i, Gav,i , is called the crustal abundance. The
separation efficiency qi of element i is called the specific mineralisability. The value
of q tends to be higher for the oxidation- and corrosion-resistent ‘noble metals’ like
gold (Au) and platinum (Pt). Using the estimates of Gav,i and the data on known
deposit size and concentration, it is possible to estimate qi . Estimates of Gav,i and qi
are shown in Table 13.1.
For a specific mineralisability q, for each separation step, one half is enriched q%
and the other half is impoverished q%. This is graphically shown in Figure A13.1. In
formula:
T M = M · Gav = 0.5 · Gav (1 + qi ) + 0.5 · Gav (1 − qi ) (A13.1)
Figure A13.1. Natural enrichment of mineral ore: With each step, there is an enrichment in
one part and an impoverishment in the other part.
After b steps, one gets:

b
1
TM = M · Ckb · Gav (1 + qi )b−k · (1 − qi )k (A13.2)
2b
k=0
with Ckb the binomial coefficient and (1/2b )M the unit deposit size. The amount of
element i in concentration class k, Mk , equals:
k
1 b 1−q

Mk = b M · Ck · Gref (A13.3)
2 1+q
with Gref the grade of the richest occurring ore. If you plot the resource size as a
function of the ore grade, you get the lognormal distribution shown in Figure 13.4.
For instance, lead (Pb) has an estimated average concentration of 17 parts per
million (ppm) and an estimated q-value of 0.288 (Table 13.1). The high-concentration
deposits are the easiest to find and exploit. For lead, they contain in the order of 5
percent by weight (wt%) or 50,000 ppm Pb. The CAG model generates the probab-
ility distribution shown in Figure 13.4, which with a reasonable accuracy represents
those reserves that are known to have existed in the past and still exist nowadays. The
other two elements for which the distribution is shown are mercury (Hg) and cerium
(Ce). Mercury production is declining because of the discovery of toxic effects, but
it used to be produced almost exclusively from an exceptional 0.6 wt% (6,000 ppm)
deposit in Spain. Cerium is the most abundant of the rare earth or lanthanide ele-
ments, which are essential for the manufacturing of cell phones, computers, hybrid
cars and military and other equipment. Its average crustal abundance is estimated in
the order of 66 ppm. It is rather reactive and supposedly has a q-value of 0.15. The
CAG-model suggests that the richest ores – it is usually mined as a by-product – are
in the order of 0.1 wt% (1,000 ppm) and that rather large amounts are available at
slightly lower concentrations.
The enrichment process represents a form of exergy because it implies a change
in the chemical potential with respect to the earth environment. (§7.4). If you con-
sider the Earth’s environment as a sink with a certain composition (78.1 percent N2 ,
20.95 percent O2 , 59.1 percent SiO2 , 15.8 percent Al2 O3 , 6.6 percent FeO and so on),
then every system with these substances in a different concentration or composition
Appendix 13.2 The Logistic Growth Life Cycle Model 471
represents an amount of exergy. Why? Because if it were brought reversibly into

contact with the Earth’s biosphere, work could be extracted – even if it were done
irreversibly, although less than the maximum.
Appendix 13.2 The Logistic Growth Life Cycle Model

The logistic growth life cycle model is based on the following assumptions (King
Hubbert 1971):
r oil fields in an oil province contain a finite amount of recoverable oil, R , which
ult
is identified by exploratory drilling;
r the oil found per unit of effort will decrease – or the reverse: drilling effort per
unit of oil found will increase – and the per unit cost to discover the oil will
increase;
r oil once found will be extracted within a rather short period thereafter, as long
as there are no competitive substitutes.
Initially, there is an exponential growth in the amounts of oil discovered (discovery
rate) and, somewhat later, in the amounts of oil extracted (extraction rate). There is
a buildup of oil reserves, which is the technically and economically recoverable oil
that is discovered but not yet extracted. Within an oil province, the cost to discover
and to produce oil rises upon depletion. The oil from this province can no longer
compete with oil from other regions and the discovery and extraction rate start
falling.
The model proposed by King Hubbert to describe this process is the logistic
growth equation with the accumulated oil production CP as the state variable (equa-
tion 2.5):
i=t
CP
dCP/dt = αCP 1 − with CPt = Pi (i) (A13.5)
Rult
i=0
Pi is the annual extraction rate in year i and Rult is the ultimately recoverable oil
resource. The accumulated production CP follows a logistic path over time. The oil
reserves follow the same pattern but a couple of years earlier, and not as function of
accumulated oil produced but of oil discovered. The oil production rate dCP/dt = P
is a bell-shaped form with a peak halfway (Figure 2.10).
14 Towards a Sustainable Economy?
14.1 Introduction
14.1.1 An Archetypical Model

The industrialisation process, and the transition to a next regime, which is in full
swing but still not easily recognised and interpreted, consists of several changes in
succession (§8.6). I consider the postindustrial transition process a change from low
income and low resource use levels towards a new equilibrium at high- and stable
income and resource use levels. In stylised form, it can be summarised in seven
items:
1. a positive feedback process drives the growth of income (€/cap/yr), which leads
to (exponential) growth in activity chains;
2. the chains consist of exploration and extraction of resources, followed by pro-
cessing into manufactured goods, use and discarding;
3. fossil fuels are an essential high-quality energy resource, that enables in com-
bination with capital and knowledge accumulation a steady increase in labour
productivity;
4. the resource intensity of income (mass/€) first rises and then declines with income
(IU hypothesis);
5. the externalities1 along the chain: resource depletion, pollution impacts and
waste accumulation, are initially not perceived and not priced;
6. with a delay, externalities are incorporated in the resource price in response to
complaints of the more affluent citizens (EKC hypothesis);
7. the rising resource price stimulates a process of cost- and pollution-reducing
innovations and the development of substitutes and alternatives.
The affluent countries are in the later stages of the transition; in the low-income
countries, the first stages can be identified although the details depend on local
circumstances and there may be jumps (‘leapfrogging’). The possibilities to import
1 The word externality is used in economics to denote those costs or benefits that are not transmitted
through prices. With costs, it is a negative externality; with benefits, a positive one.
472
Figure 14.1. Causal loop diagram (CLD) of the resource chain process in an industrial
economy.
resources from abroad, the vulnerability of the natural environment and the world-
views and behaviour of people are important determinants of how the transition will
unfold. In this chapter, I focus on the techno-economic aspects.
An influence diagram of the archetypical transition process in an isolated econ-
omy is sketched in Figure 14.1. The core element is the physical resource chain,
with the source and the sink sides as discussed previously. Upstream are the non-
renewable resources that are extracted for industrial upgrading and processing and
then an input for manufacturing and marketing. When the products are sold on
the consumer market, it accumulates as resource-in-use. After the lifetime of the
product, its constituents end up via various routes in one of the sinks: landfills, air,
rivers, groundwater and soils. It accumulates and, for most compounds, is degraded
and eliminated. The route from resource-in-use can be prolonged in a round of
reuse or recycling and its extent and composition can be changed in waste treatment
plants.
A possible storyline is as follows. The non-renewable resource is an aggregate
of minerals/fuels and indicated with MF. Income growth causes an increase in MF
use, initially at an accelerating pace. The resource is still cheap, but the rapidly rising
extraction rate causes a cost increase: the depletion feedback loop. The MF market
price goes up and this induces more efficient MF use in the price feedback loop.
When income increases further, the MF intensity starts to decline (IU hypothesis).
With rising income, citizens start to protest against the pollution from resource use,
because they experience it as loss of quality of life. Government and industrial elites
are forced to invest in pollution abatement and in resource use efficiency: the exter-
nality feedback loop (EKC hypothesis). Additional costs are made in the economy
for waste treatment, material efficiency and recycling and for the associated physical
a) China Income 2010 EKS $/cap/yr Resource use
Japan Income 2010 EKS $/cap/yr Metal use (rescaled)
40000
ppp 2010 EKS$/yr & R-units/yr
30000
20000
10000
0
1950 1970 1990 2010 2030 2050
Year
Figure 14.2a. Simulation of the stylised postindustrial transition, with the historical income
growth in China between 1950 and 2010 and its extrapolation to 2060 to the level of Japan
in 2010. The graph shows income and resource use; the dots indicate normalised use of the
major metals (steel, aluminium, copper, lead and zinc) in China.
and institutional infrastructure (regulation, enforcement and so on). It shows up in

higher cost and price levels and reinforces the price and externality feedbacks. The
new, high-income equilibrium can be sustained as long as there are enough resources
available at the higher cost level or if a substitute is available.
A stylised postindustrial transition can be simulated with a simple model based
on the mechanisms sketched in Figure 14.1. For illustrative purposes, I use the histor-
ical income growth path of China between 1950 and 2010 as an exogenous input and
assume that it continues until 2060 at the rate that the Japanese people experienced
since 1965.2 The transition period, between 2010 and 2050, is one of steady increase
in affluence and one of social and environmental stress. A characteristic outcome is
shown in Figure 14.2a-b. Resource prices go up and the expenditures on the resource
as fraction of income increase significantly (from 5 percent to 21 percent in the sim-
ulation). A decoupling between income and resource use occurs after the year 2015,
induced by depletion and externality pricing effects. Nevertheless, by 2060, the indig-
enous resource is depleted by two-thirds and, despite environmental regulations,
there are serious negative environmental consequences. Mining and infrastructure
cause pollutant emissions – apart from large-scale displacement of rural popula-
tions. Waste disposal affects the food chain and have negative health impacts. What
the Chinese Deputy Minister of Environmental Protection in 2006 said comes true:
‘China has entered a phase of frequent environmental accidents. With a fast growing
economy, systematic risks of environmental accidents brought about by unscientific
placement and industrial structure will replace individual pollution to become the
No. 1 threat to our country’s environmental safety’. Infrastructure falls short of
2 In other words, in 2060, the average Chinese person will have the same income as the average
Japanese person in 2010. Interactions between income and population growth and differences in the
composition of production and consumption are not considered.
b) 1.00 4
res frac rem
other variables (normalized)

res frac rem (normalized)
DepletionMult
0.75 3
ExternalityMult
Resource price
0.50 2
0.25 1
0.00 0
1950 1970 1990 2010 2030 2050
Year
Figure 14.2b. The depletion and externality multipliers and the resource price for the simu-
lation in Figure 14.2a.
expectations and income inequalities increase, which gives rise to social tensions.
Also, there are long-term impacts, such as changes in climate and biodiversity, but
these tend to be ignored or denied in the face of more direct challenges and threats.
From a source perspective, this economic growth path is unsustainable because the
resource is depleted before the year 2100 despite a stabilised extraction rate. From a
sink perspective, it entails the risk of irreversible losses in quality of life. But such a
growth path is possible, as countries such as Taiwan and Korea have shown (TCEPD
1989).
The stylised transition sketched previously is a widely believed, official outlook
on the future. It presumes that countries can maintain economic growth rates and
thus, implicitly, can overcome a number of social, resource and environmental con-
straints. The real-world transition can differ from the simulated one for at least two
specific reasons. First, the IU and EKC hypotheses have limited validity. The IU
hypothesis presupposes the existence of a substitute, such as plastic from oil for
steel, that is itself part of another chain with similar dynamics. The EKC hypothesis
may be invalid or too late for indirect and long-term impacts, because citizens do
not experience the negative consequences themselves.3 A second reason is that no
country operates on its own. Historically, resource shortages have been countered by
military conquest, import of resources and outmigration of people. Nowadays, trade
is the preferred choice and mineral ore and fossil fuel are the largest bulk trade flows
in the world. Resource-scarce countries will try to exchange manufactured goods
and services for resources.4 Availability on the world market will come under stress
3 Negotiations about greenhouse gas emissions and loss of biodiversity and military adventures to
safeguard resource imports and waste exports testify that such a response is deeply ingrained.
4 Land-scarce countries did and do exactly this: importing food as soon as they are able to exchange it
for manufactured goods. The UK has done it since the 19th century, and some Asian countries just
started doing it. The phenomena of virtual water use and ‘land grabbing’ are in the same category
of responses.
R extraction rate R import rate
R in sinks (right scale) R net rem rate
20000 500000
400000
15000
R units/yr
300000
R units
10000
200000
5000
100000
0 0
2000 2020 2040 2060 2080 2100
Year
Figure 14.3. As in Figure 14.2a, but extending the simulation beyond 2060 with the assumption
of zero income growth after 2060 and the possibility to import resource.
and cause political tensions and fluctuating prices. A counterforce is that resource
extraction costs, at least temporarily, decrease because of innovations, but this will
partly be undone by oligopolistic price setting and safety and environmental regu-
lations (Figure 13.7). Real-world resource prices can also be expected to fluctuate
significantly, which in combination with mismatches between demand and supply
from system delays, feedbacks and noise effect the overall process. History abounds
with examples.
Nevertheless, our simple archetypical model highlights a few points. First,
imports are from a global perspective merely displacing the problem. This is seen
if we lengthen the time horizon to the year 2100 and open the country for resource
imports. Next, we assume that the country starts importing the resource at a price that
is initially higher than the price of the indigenous resource. Then, we also suppose
that after 2060 the economy does not grow anymore or, if it grows, at zero marginal
resource intensity, so resource use remains constant thereafter. With the substitution
elasticity assumed in this simulation, imports start around 2010 and make up more
than 70 percent of the total flow by 2060. In this longer time resource-trade perspec-
tive, the country has successfully postponed depletion of indigenous resources: Even
by 2150, still one-third of the initial resource remains (Figure 14.3). But this has been
possible at the expense of resource use elsewhere – an example of shifting the bur-
den (§2.4). On the sink side, the situation has aggravated because without a decline
in resource use, the stock of pollutants keeps rising towards a dynamic equilibrium
of high levels of accumulated pollution with local and global feedbacks. In this
long-term perspective, economic growth is dependent on imported, finite resources
and creates large and rising environmental impacts. It is, again though differently,
unsustainable (Ho and Vermeer 2006). Truly long-term solutions are more intense
resource efficiency and recycling and further development of substitutes. They are
needed, and without delay, for a path that can be considered a sustainability transi-
tion. In some countries, signs of this are visible. In others, only the resistance against
it has a voice. In many countries, short-term priorities still overwhelm all long-term
concerns.
14.1.2 Substitutability, Technology and Optimality

An industrial economy can make the transition to a sustainable state, provided the
average income stabilises and there are sufficient resources, substitutes and sinks.
During the transition, price-based market forces, government-coordinated regula-
tory policies and behavioural changes are all three needed. Even then, there will
be periods during which parts of the population suffer from negative environmental
and social impacts. Episodes of severe pollution on many locations appears to be
part of the overshoot dynamics, just like the stress of overpopulation during the
demographic transition, the increase of welfare diseases during the health transition
and loss of fertile land to degradation and urbanisation during the land transition.
The experiences in the United States and Europe show that part of the damage
is reversible and can be undone. Pollution-related diseases can be addressed with
medical expenses and technology; rivers and soils can be cleaned up or will slowly
recover if no longer under stress. But it takes time and effort. Some impacts are
irreversible – notably the expected change in climate and the loss of biodiversity.
The social impacts show up as a further increase in income and wealth inequality,
decline of community life and tensions in the provision of public goods. These can
be remedied by proper social security schemes and public planning, but here the
experience shows that it is difficult in times of high expectations, rapid globalisation
and large uncertainties.
Not surprisingly, the assumptions of resource substitutability and of techno-
logical change are the critical ones in macroeconomic analyses of growth and sustain-
ability. Per capita consumption can continue to grow until the investments needed
to compensate the declining resource quality become so large that the output avail-
able for consumption makes further net growth impossible. Beyond that point, per
capita utility inevitably declines. If environmental constraints are also considered,
the long-term feasibility of sustainable pathways is further narrowed down. Such
source and sink constraints, notably in mining efficiency, land yields and pollutant
elimination, are the causes of collapse in the Limits to Growth model (Meadows
et al. 1971). There are three solutions to such a ‘doomsday’ scenario. First, the
existence of a backstop resource, which is a substitute resource that is available in
nearly infinite amounts. A second solution is technology that makes it possible to
perform economic activities at vanishing resource and environmental intensity, as
is the case in extreme forms of the IU and EKC hypotheses. A third and officially
rarely discussed possibility is a change in lifestyle and values that breaks by a mix-
ture of necessity and creativity, the link between the use of finite sources and sinks
and the quality of life.5 For many people, this is what the sustainability transition
5 Such an extreme form of dematerialisation is known to economists as (resource-growth) decoupling

and features in the debate also as delinking, degrowth, décroissance and green growth. It might
happen in the form of the rapid expansion of cyberspace on the basis of ICT we are currently
witnessing. But this is in no way certain: Non-material space is also exploited in favour of more
consumption, in a mixture of commercial battles, directionless innovation and meaningless content
and similar to what happened in physical space with the European conquest of the Americas.
is about. It is implied in Stuart Mill’s famous statement: ‘I sincerely hope, for the
sake of posterity, that [the population] will be content to be stationary, long before
necessity compels them to it’. It would coincide with a less materialist worldview
(Figure 6.3).
Macroeconomists have explored the conditions for sustained economic growth
with finite resources primarily in the form of an optimal control problem (Dasgupta
and Heal 1979; Ströbele 1984). The idea is to maximise the discounted utility over the
time horizon considered under two constraints: First, the macroeconomic production
function discussed later, and second, the finiteness of one or more sources and sinks
(§12.3 and §13.3). In formula:
T
max e−rt [U (C)] dt (14.1)

0
with U = U(C) the social utility function and C consumption. T is the social plan-
ning horizon. It can be shown analytically that, under particular and simplifying
assumptions, macroeconomic output can be sustained forever on the condition that
the resource is completely substitutable (Perman et al. 2003).6 In economic science,
this assumption is known as weak sustainability. It implies that natural capital can
in the long term be substituted completely by man-made capital and the incorpor-
ated ingenuity. In theory, complete source substitution cannot be excluded. It would
be an economy with complete recycling and an energy supply based wholly on a
resource that can be produced in near infinite amounts at constant cost. Solar power
is probably the only option that qualifies as such a backstop resource. Complete sink
substitution is more difficult to imagine. If sustainability is defined as the situation in
which production- and consumption-related waste flows ending up in environmental
sinks do not exceed the rate at which they are absorbed and broken down, then weak
sustainability implies that ever larger waste flows must be treated and eliminated.
This itself will require large amounts of exergy and materials. One only has to con-
sider the energy balance of the Earth to realise that this will pose limits to growth at
some point in time and place.
One final point about these rather academic modeling exercises. In macro-
economic models, supply and demand are assumed to equilibrate in the time period
considered (usually one year). Therefore, responses to source depletion and sink
overexploitation are timely and smooth. Moreover, the response activities are
assumed to share in the innovation-driven productivity increase that is supposed
for the economy at large. Therefore, the calculated costs of adaptation are in most
macroeconomic model calculations small and declining and the simulated economic
growth trajectories hardly differ from the ones without resource constraints.7 For
instance, most integrated economy-climate models suggest that policies to stabilise
6 Many other formulations and models of the optimisation problem have been explored, for instance,
incorporating technical change and resource and environmental constraints, feedbacks and costs.
7 In these simulations, there is also hardly any or no consideration of the economic damage from
changes in, for instance, climate and biodiversity. The overall costs of such policies are, therefore,
unknown, because the costs of no policies in the form of non-avoided damages cannot be known.
Some economists argue that this is irrelevant, because any future costs vanish with the discount rates
of 3%/yr and more in such simulations.
atmospheric concentrations would postpone income growth with only a few years
(Azar and Schneider 2002; IPCC 2001, 2007). For these reasons, most participants in
the sustainability debate prefer strong sustainability, which implies limited substitut-
ability between natural and man-made capital. Long-term concerns about climate
change, biodiversity and other issues in relation to economic prosperity becomes
then a matter of (very) small risks of (very) large and irreversible damage – a world
far away from deterministic optimality (Meinshausen 2006).
In a broader context, one can reframe the situation in three questions. How much
decline in quality life can a social-ecological system sustain before it collapses (resi-
lience)? How productive and circular can the industrial system become (efficiency)?
How can a decent quality of life be secured at low and stationary levels of physical
throughput (sufficiency)? The answers to these questions are to be discovered ‘along
the road’ and it is unclear whether and when to speak of a sustainability transi-
tion. Any assessment is complicated by the fact that, in the new postindustrial
economy, new issues emerge. What is the role of the information or experience
economy? How can we sustainably manage the commons? Can there be a human
capitalism (Pine and Gilmore 1999; Barnes 2006; Tasaka 2009)? To find answers, I
continue with a brief discussion on what economists and economic theory have to
say about sustainable development.
14.2 Theories of Economic Growth
14.2.1 Classical Theories

The history of economic thought is associated with a number of mainly British
and American scholars. The history of economic science is in itself important and
revealing if one wants to understand present positions in the debate on sustainable
development. Unlike in the natural sciences, the subject of analysis is so complex
and dynamic that no economic theory consolidates into natural law. Economic the-
ory more or less starts with the late 18th-century scholar Adam Smith and his book
Inquiry into the Causes and Nature of the Wealth of Nations (1776). It explained
how increasing trade and market size leads to specialisation, which causes labour
productivity to increase. This in turn increases income per capita and, subsequently,
market size. This positive feedback loop only starts to operate when market size
exceeds a certain threshold, as it did in Smith’s time in response to novel techniques,
colonial conquests, falling transport costs and other factors. Smith also looked for
another mechanism than church and state that could tie together society, and he
found it in the famous invisible hand of the market mechanism. If the individual is
given the opportunity to pursue his own self-interest, in an environment of likely
motivated individuals, this leads to competition. This in turn produces the goods soci-
ety wishes to have and at prices that people are willing to pay. Thanks to competition,
selfish motives yield the unexpected outcome of social harmony.
A second line of thinking in classical economics is associated with the name
of Thomas Malthus. In his Essay on the Principle of Population (1798), Malthus
showed himself more sceptical about societal progress than Smith. He argued that
unconstrained population growth will inevitably outstrip food production because
land is a finite resource and food production has diminishing returns, which means
that twice as many labourers on the land produce less than twice as much food. The
consequence is declining food per person, followed by a rise in mortality and a stabil-
isation with overshoot (§9.5). In other words, natural processes tend to put limits to
growth. Malthus concluded that population control was necessary. By breaking the
link between population growth and output growth, he felt an escape was possible,
but only if it was engineered.
The evidence in support of Malthus’ theory is quite convincing for most of his-
tory and most countries – although not for Europe in the past 200 years because key
assumptions turned out to be less and less valid. First, Malthus failed to recognise
the importance of capital and technology, which allow farmers to escape the dimin-
ishing returns of land and to increase yields significantly. Second, better hygiene and
sanitation, medical advances and birth control options in combination with rising
income led to a reduction in birth as well as death rates in the European countries.
Two hundred years of sustained growth of economic output in Western countries
suggests that Malthus was wrong. However, 10,000 years of pre-industrial history as
well as the large group of countries that still seem locked in a no-growth poverty
trap show that rejection of ‘Malthusianism’ is premature. More inclusive models of
the various stages of and transitions in economic growth will have to explore the
dynamics in more detail (Galor and Weil 2000).
With Smith and Malthus, discussion focused on two early and extreme posi-
tions in the debate. One optimistic, suggesting that growth will accelerate forever
and more or less automatically, and one pessimistic, predicting that growth cannot
endure no matter what is done. The ‘perfect world of Adam Smith’, as Heilbroner
(1955) calls it, did, of course, only exist in small pockets in time and space. Much
of 19th- and 20th-century political economy was about the world’s imperfections:
the permanent threat of scarce resources for a growing population (Malthus and
Ricardo), the exploitation of labourers (Marx), the absence of coordinated planning
and redistribution (Saint-Simon and Stuart Mill), the elimination of competition by
large oligopolistic enterprises and the psychopathology of the capitalist class society
(Veblen) and the creative destruction of technological change (Schumpeter). Nev-
ertheless, Smith’s idea of the market as the coordinator has become, in more or less
idealised forms, a pillar of the Postmodernist worldview (§6.5).
The early political and ethical works of classical economic theory are gradually
narrowed to a body of highly abstract and rather esoteric concepts and hypotheses.
One explanation for this is the desire, in the context of the Modernist worldview, to
become an objective ‘natural’ science (§8.3). Modern economic (growth) theory has
a narrow perspective and is largely associated with (Anglo-American) capitalism.
Ethical issues such as a fair reward distribution between the ‘production factors’
capital(ists) and labour(ers) and, later, between the rich and the poor and between
present and future generations were formalised and rationalised or disappeared
altogether from mainstream economic literature. Many social scientists, including
many political and welfare economists, are critical of this development and its con-
sequences and of the validity of the underlying ‘image of man’ (§10.4). Nevertheless,
economic growth theory has generated insights that are relevant for the sustainable
development discourse and some of these are introduced in subsequent sections.
Again, this is a vast topic, so I refer the reader to the Suggested Reading for further
study. The focus is on models first: ‘Every discipline that I am familiar with draws
Figure 14.4. Flows in the economic system and its relations with the outside world. The black
arrows represent money flows. The dashed arrows represent their physical and informational
equivalents.
caricatures of the world in order to make sense of it. The modern economist does
this by building models, which are deliberately stripped down representations of the
phenomena out there’ (Dasgupta 2007).
14.2.2 Economic Growth Theory

The starting point for our investigation is the scheme in Figure 14.4. It is an extended
version of the standard flow diagram of economic textbooks. Transactions between
producers and consumers are at the centre of the (market) economy. Governments
and banks are two other important subsystems. In the core part of the industrial
economy, labour and capital are the production factors and the rewards (wages
and dividends) are used for consumer expenditures. Money is the numéraire in this
system. The monetary flows are attached and add value to the underlying physical
stocks and flows. Besides, there are large information flows to facilitate transactions
between producers and consumers and between entrepreneurs, labourers, govern-
ments and bankers. Most economic activities, regulations and data are defined at the
level of the ‘sovereign’ states, that are mutually linked through exchanges of goods
and services, information and people. Not all human activities are registered as part
of the formal economy: The monetary economy is continuously, and in manifold
ways, interacting with people living in subsistence and informal economic systems.
Both formal and informal economy are embedded in the larger life support system
(Figures 8.6 and 13.1). In a sustainability context, it is important to have this more
inclusive perspective on ‘the economy’.
It was long known that the use of tools and machines make human labourers
more productive. It was formalised by Solow (1957) in the neoclassical model of
economic growth, in which capital accumulation plays a central role. Capital K and
labour L are the production factors. Land and resources, including fossil fuels, were
initially left out because of their small and declining role. The increase in the capital-
labour ratio allows output per worker, that is, labour productivity, to increase.8
Because it implies that the same output can be produced with less labourers, the
process is referred to as capital-labour substitution because workers are replaced by
machinery. The capital stock K is built up from investments I. It wears out and is
taken out of production and depreciates at a presumedly fixed rate m.9 Thus:
K dK
≈ = K̇ = I − mK (14.2)
t dt
In simulation models, the equation is treated as a discrete equation, but in its deriva-
tions, it is usually expressed as a differential equation. The equation describes a
pseudo-dynamic process, during which the system jumps from one equilibrium state
to the next. If the population increases at rate n and the fraction of workers in the
population remains constant, the growth of the labour force is given by dL/dt =
nL. Using the macroeconomic bookkeeping identity that output Y equals the sum
of consumption C and investment I and normalising variables with respect to L,
equation 14.2 becomes:
k̇ = y − c − (m + n)k (14.3)
with y = Y/L output per worker and k = K/L capital per worker. This equation
states that the stock of capital per worker equals the per worker investments minus
the per worker capital depreciation and the dilution from population growth.
You can specify the model if you postulate a production function Y = Y(K,L) that
represents all the possible combinations of production factors {K,L} that produce a
certain output Y.10 Following the aggregate trends in both agricultural and industrial
economies, the production function is assumed to exhibit constant returns to scale
and diminishing returns.11 Using these properties and assuming that the production
function is normalised (Y0 = 1, K0 = 1, L0 = 1 at time t = 0), a suitable production
function is of the form:
Y = Kα L1−α (14.4a)
8 The production factors capital K and labour L can be expressed per unit output. It is called the
factor intensity. The inverse is called the factor productivity and tells how much output is produced
per unit of factor input. More specifically, the output per unit of labour is called labour productivity
and indicated with a small letter: y = Y/L.
9 The depreciation rate is often approximated with 1/L with L the economic lifetime of the capital
stock (Chapter 2). I suppress the index t (time) in this and subsequent equations. The assumption is
that the period t is long enough to let the system come into equilibrium – usually one year.
10 Normally, only the input combinations that give the highest possible output are considered. In that
sense, it represents a production frontier and the best available technology. Real-world performance
will be lower. According to economists, this justifies the assumption of perfect rationality in human
behaviour: The models do not claim perfectly rational behaviour of real-world persons but, rather,
explore the conditions and outcomes if such behaviour were to happen.
11 Constant returns to scale is the property that total output increases with a factor n if all inputs
increase with a factor n: Y(nK, nL) = nY(K,L). If it is less (more), there are decreasing (increasing)
returns to scale. This assumption is shown to be invalid in the service and information economy
(Appendix 10.1). A production function exhibits decreasing or diminishing returns if doubling of
one of the inputs gives less than a doubling in output. It is effectively equal to a stabilising feedback.
The opposite of a positive feedback is referred to as increasing returns.
historical (k,dk/dt) (Belgium) labour productivity y = kα
depr and pop rate (m+n)k kα-(m+n)k (=c*)
2.0
1.5
dk/dt=kα-(m+n)k
1.0
0.5
0.0
-0.5
0 1 2 3 4 5
k=K/L
Figure 14.5. The neo-classical economic growth model. Labour productivity y increases in
proportion to kα (upper grey curve). Output partly goes to capital depreciation (mk) and
to population growth (nk). The difference is available for per capita consumption c (dashed
black curve). The steady-state levels k* and c* are at the intersection of the two curves (open
dot). (alpha = 0.35; m = 25%/5 yr; n = 7%/5 yr). The black dots indicate estimated values for
Belgian agriculture 1880–1980 (source of data: Swinnen et al. 2001).
or normalised with respect to L:

α
Kα L1−α Kα

Y K
y= = = α = = kα (14.4b)
L L L L
This is the widely used formulation proposed in the 1920s by Cobb and Douglas
(CD). It states that a rising capital-labour ratio k increases labour productivity y but
with diminishing returns and with constant returns to scale. Inserting equation 14.4b
into equation 14.3 gives:
k̇ = kα − c − (m + n)k (14.5)
This is the basic equation of the neoclassical growth model (Barro and Sala-i-Martin
2004). For dk/dt = 0, there is a steady-state {c* ,k* } at which per worker consumption
level c* is given by:
c∗ = k∗α − (m + n)k∗ (14.6)
at the capital-labour ratio k*. Equation 14.5 is graphically shown in Figure 14.5 with
the curves for kα and (m + n)k as a function of the capital-labour ratio k. For c < c*
the economy will grow towards higher k-values.
Is there any empirical confirmation? By way of example, I inserted in Figure 14.5
the empirical data for agriculture in Belgium for the period 1880–1989 (Swinnen et al.
2001). Between 1880 and 1930, the capital-labour ratio k fluctuated between 0.46
and 0.54, and there was no substitution and no productivity increase. After World
War II, the k-value started to increase rapidly and these data can be reproduced
with a CD production function. It suggests a growth path towards a steady-state,
but in reality, the k-value dropped significantly in the 1980s. Even in the relatively
simple agricultural sector, other mechanisms than factor substitution were operating,
such as innovation waves, international trade, government interference on behalf of
consumers or farmers and the possibility of capital-land substitution and capital-
labour complementarity in poor countries (Haley 1991).
From a sustainable development perspective, the neoclassical growth model
suggests the best of possible worlds and a recipe for sustainable development. It
predicts that per capita income levels can exceed subsistence levels permanently
through capital accumulation, that is, rising k, in contrast to Malthus’ conviction. It
indicates that an economy has a natural tendency to reach a zero-growth steady-state
and that per capita income differences between countries have a natural tendency to
disappear.12 Unfortunately, even apart from resource and environment constraints,
the real world behaves differently. Capital accumulation and capital-labour substi-
tution undoubtedly played an important role in economic growth. But time-series
and cross-country analyses of economic growth in industrial economies during the
20th century show that they explain only half or less of the observed output growth
and that no income convergence amongst the countries of the world takes place
except in a few ‘convergence clubs’. These shortcomings in the theory have been
acknowledged since the 1980s (Barro and Sala-i-Martin 2004; Helpman 2004).
One reason of failure is the high aggregation level of description, interpreta-
tion and measurement of the production factors. Labour reflects a wide variety in
skill and age levels. In advanced economies, fossil fuel is almost completely substi-
tuting for physical labour and most labour is involved in control and information.
Economists have introduced the notion of human capital in order to incorporate
the differences and changes in skill levels with proxies, such as number of years
of schooling. It provides some additional explanation (Helpman 2004). Capital also
represents a variety of items and undergoes quantitative and qualitative change.
Table 14.1 shows data for the capital stocks in four countries as quantified in mon-
etary units by the national statistical offices. Per person capital stocks have roughly
linearly increased over the last decades in these countries.13 The data indicate that
in the order of two-thirds of the capital stock consists of dwellings, buildings and
infrastructure and that the structural composition in monetary terms changes only
slowly. But scale effects, factor returns and capital-labour substitution for dwellings,
buildings and infrastructure are different from those in sectors such as information
and communication technology (ICT). Indeed, this is one reason to use the simpler
production function: Y = AK, introduced by Leontief. Defining the savings rate σ =
I/Y and replacing kα – c with σ Ak in equation 14.4, the growth equation becomes:
dy
= (σ A − m − n)y (14.7)
dt
12 The growth rate is lower for higher capital-labour ratio’s, so poor countries with low capital per
worker levels can grow fast and eventually converge with the rich countries at high k-levels (con-
vergence).
13 The increase in k, that is, in capital per worker, is probably lower in these economies because the
labour force grew faster than the population. This has partly been compensated by less working
hours (Chapter 4).
Table 14.1. Structure of capital stocks in Switzerland (2006), Finland (2006) and Australia
(1993–1994)a
Switzerland 2006 % Finland 2006 % Australia 1993–1994 %
Construction Building 55.8 Residential buildings 40.0 Private

Civil engineering 15.2 Non-residential 26.7 Dwellings 32.5
buildings
As part of construction: 27.6 Other structures 12.6 Construction 17.2
Machinery and equipment
Fabricated metal products and 14.4 Other machinery and 15 Equipment 15.5
machinery equipment
Office machinery and computers 1.2 Public
Electrical machinery and 2.3 Dwellings 1.9
apparatus n.e.c.
T.V., radio and communication 2.1 Construction 23.6
equipment and apparatus
Medical, precision and optical 3.8 Transport equipment 3.1 Equipment 4.8
instruments, watches and
clocks
Motor vehicles, trailers and 0.8
semi-trailers
Other transport equipment 2.9
Cultivated assets 0.2 Major improvements to 1.1
land and so on
Computer and related services 1.3 Software, knowledge, 1.5
entertainment, art
and so on
a Sources: www.rba.gov.au/statistics/op8 index.html,
www.bfs.admin.ch/bfs/portal/en/index/themen/04/02/04/key/Stock cap.html and

www.stat.fi/til/pka/tau en.html.
Assuming a closed economy in which the savings rate is constant and equal to the
investment rate, this model stipulates a continuous exponential growth in output as
well as in capital stock and consumption. Although this is the historical experience
in certain periods and places, it lacks explanatory power. Already a century ago,
economists turned to another determinant of growth: the knowledge incorporated
in capital and labour or, in short, technology.
14.2.3 The Role of Technology, Learning and Behaviour

In the 1950s, technological change was already identified as a major source of long-
term economic growth. But technology and innovation are elusive concepts that
are hard to measure or observe directly. Investment in physical capital and equip-
ment; in human capital and in research; and development (R&D) are all forms of
technology and causes of productivity growth. There is not yet a validated theory
of how each of them contributes and how they interact. Madsen et al. (2010) did
an extensive econometric analysis of the determinants of economic growth in the
UK in the period 1620–2006, including proxies for sectoral shares, coal availability,
government expenditures, urbanisation and trade barriers. They conclude that ‘the
empirical estimates so far give support to the hypothesis that productivity growth
Box 14.1. Is technology the ultimate force driving growth? This book has dis-
cussed technology and transitions several times – but do we understand the under-
lying dynamics? In a rather narrow, techno-economic context, these phenomena
are mostly substitution of one product or process by another in markets, as in the
CD production function. Several model formulations have been proposed for the
substitution process, one of the most widely known being the logistic substitution
model. It derives from the predator-prey and logistic growth models (Fisher and
Pry 1971).
If you change the state variable X from an absolute to a relative variable Y,
with Y = X/K, you can rewrite it into its normalised form: dY/dt = αY(1 − Y).
Y can be interpreted as the occupancy of a particular niche. It represents the
fraction f of the market occupied by a particular species or technology. It has been
observed that f follows a S-shaped growth curve over time for many technological
substitution processes (Grübler 1999). If the observed f-values are plotted as
log[ f/(1 − f )] over time, it is found to approximate a (segmented) straight line for
a variety of long-term historical techno-economic processes. The model can be
applied to more than one variable. Examples are the rise and decline of various
transport modes in France, the substitution of steel manufacturing processes and
the gradual switch from biomass-based to modern fuels and then electricity for
lighting in the United Kingdom (Figure 14.6a–c). Other data indicating such
transition dynamics are the substitution of energy carriers and the replacement
of steam engines by explosion engines.
in Britain, until the twentieth century, was predominantly a race between techno-
logical progress and population growth’. Income changes are well explained by the
changes in population growth rate and the changes in research intensity, the latter
being approximated by patent applications per worker. The rather high population
growth before 1900 was a drag on the technology-driven income growth. On the
other hand, the continuous research, patenting and implementation effort was a
driver of economic growth. But trade openness, financial (de)regulation, inflation
and unemployment rates, income distribution, political and civil rights, the size of
government and public infrastructure investments all influenced productivity growth
and continue to do so (Helpman 2004; Barro and Sala-i-Martin 2005).
The first and simplest way to include technology in the production function is
with an exogenous factor, for instance:
Y = AKα L1−α (14.8a)
The term A is called the total factor productivity (TFP) and can, not surprisingly,
explain a large part of productivity growth.14 ‘Technology’ becomes an all-explaining
deus ex machine, a manna from heaven. One way to estimate A is through the idea
14 For instance, in OECD countries, 16 to 47 percent of GDP growth in the period 1960–1995 is not
explained by contributions from capital and labour (Barrro and Sala-i-Martin 2005).
(a)
F Market share
1–F fraction F
102 0.99
101 0.90
0.75
Horse Railway
100 0.50
0.30
Waterway
10–1 0.10
Truck
Pipeline
Air
10–2 0.01
1800 1825 1850 1875 1900 1925 1950 1975 2000
(b)
F
1–F Fraction F
102 0.99
Puddel
101 0.90
Bessemer
Open-hearth 0.70
100 0.50
Oxygen
0.30
10–1 0.10
Electric
10–2 0.01
1850 1900 1950 2000
Year
Figure 14.6a,b,c. (a) Logistic substitution in transport modes in France since 1800, (b) in
global steel manufacturing process technologies since 1850 (c) and in lighting technology
(Source: Grübler 1999, www.iiasa.ac.at).
(c)
UK Lighting Services by Technology
(Fractional Shares in Billion Lumen-hrs, logit transformation)
100.00 99%
10.00 90%
T0 = 1830 T0 = 1929
f/(1–f)
biofuels (candles & whale oil)

1.00 Δt = 67 years Δt = 23 years 50%
fossil fuels (gas and kerosene)
electricity
0.10 10%
0.01 1%
1800 1850 1900 1950 2000
of learning-by-doing, that expresses A explicitly as a function g of accumulated

knowledge by workers (equation 13.2):
T
⎛ ⎞
Y = g⎝ Yτ dτ ⎠ f (K, L) (14.8b)
τ =0
Often, an S-shaped curve is assumed for g, with the argument that, within a certain
regime, additional output no longer generates any productivity increase when the
niche is filled (§13.3). Assuming a CD production function, the new formalism implies
again a path towards a steady-state. Another way to endogenise technology is to
emphasise the knowledge of workers and introduce human capital as a production
factor: H = Lh. The variable h is a measure of the quality of the worker that can
be estimated from education levels, amongst others (Barro and Sala-i-Martin 2004,
Weber et al. 2005).
At least three more mechanisms link technology to output. The first one is
the role of factor price–induced innovation. Production theory indicates that cost-
minimisation implies a capital-labour ratio that is proportional to the inverse of the
price ratio: (K/L)opt ∼ (pL /pK )opt (Samuelson 1947). Relative factor costs induce
technical changes that allow substitution of the cheap for the expensive factor. It
implies a de facto development and implementation of innovations away from the
most expensive input, as in labour-saving mechanisation. Second, R&D investments
generate innovations in a variety of ways, partly unplanned and unforeseen and
with long-term consequences. Economists emphasise that private rewards for the
innovator in the form of an entry barrier, for instance, through intellectual prop-
erty rights (patents), are a necessary condition. But technological breakthroughs
and lock-ins, organisational skills and other factors also play a role.15 Third, and
relatedly, education is a necessary ingredient of technology-induced productivity
growth. Economists have introduced it in the rational agent model by making edu-
cation a function of expected future earnings. In combination with demographic
changes, it can provide a better understanding of the long-term population-economy
nexus (§10.3; Galor and Weil 2000).
More in-depth insights into the mechanisms of economic growth come from
innovation theory and business dynamics (§10.4). Since the 1980s, some economists
have more radically incorporated the role of science and technology and of human
behaviour in economic theory and models. Building on earlier work by Veblen,
Schumpeter, Simon and others, Nelson and Winter laid the foundation of evolution-
ary economics in their book An Evolutionary Theory of Economic Change (1982).
The Homo economicus is replaced by a pragmatic and adaptive individual or organ-
isation, who is imperfectly informed, operates with bounded rationality and learns
from previous experiences.16 On the basis of reproduction, selection and mutation
processes, agents choose or imitate strategies that are more rewarding and succes-
sul vis-à-vis their objective. The evolutionary mechanism of survival under selection
pressure is thus central in these models (van den Bergh et al. 2007). It has the strength
of the biosciences but creates a ‘bottom-up’ bias that tends to neglect or deny the
‘top-down’ coordination and regulation mechanisms (§10.5).
Evolutionary economists attempt to simulate two major groups of actors in an
economy: producers and consumers.17 A small group of agents (producers and con-
sumers) ‘drive’ the system to more and novel products for consumption, exploiting
the desire for profit, status and novelty. Technological change is incorporated in the
investment decisions of the producers. If it is completely absent, the model economy
is in a steady-state with zero growth. When it is turned on, the model economy starts
to evolve in a permanent disequilibrium. Although the vocabulary of evolution-
ary economists is different, several model features (such as desired or anticipated
versus actual values) resemble elements of the system dynamics models – and, not
surprisingly, so do the conclusions. Two illustrative examples are summarised in
Appendix 14.2.
System dynamic models of economic growth address some of the shortcomings
in economic theory by introducing agents (mechanisms) that have local information
about profitable opportunities for change but proceed, in the absence of global
information about the (future) system, incrementally in a direction that improves the
profit or another target or performance indicator (Sterman 2000). Change happens
in incremental steps, in a gradient-following process. The engine of economic growth
15 It may well be that – hardly predictable – long waves of general-purpose technologies (GPT), such
as the steam engine, electricity and the computer, are more important for long-term TFP changes
than the processes of incremental innovations
16 In the process, the 19th-century metaphors of classical thermodynamics and mechanics are replaced
by another set of – also 19th-century – metaphors from Darwinian biology (Döpfer 2005).
17 What is called evolutionary economics has overlaps with innovation studies and behavioural and
experimental economics. Methodologically, it uses the novel method of ABMs (§10.5).
is conceived of a series of connected feedbacks, or increasing returns in economist

terminology. Important ones at the level of corporations are:
r unit production cost: there is continuous drive to lower unit cost through R&D
(§11.6). Traditionally, the ways to reduce costs are through economies of scale,
economies of scope and learning-by-doing. All three can work as positive feed-
back loops through which unit cost declines;
r unit development cost: in many modern knowledge-intensive industries, the
upfront development costs, are a large fraction of total cost and the actual
production cost are small or negligible (chips, software or music as examples).
Once underway, there is an enormous drive to create large sales to recover the
upfront cost;
r product awareness: firms will use advertising and sales efforts to promote their
products. In combination with word of mouth and media attention, this may
create a positive feedback towards ever larger sales and market share.
New product development, acquiring mono- or oligopolistic market power, mergers
and acquisitions, promoting workforce quality and loyalty, and access to cheap
capital by high profits and growth rates are other elements in the battle for growth of
firms and corporations. A particularly important mechanism is the positive feedback
from interaction synergies and network effects, with significant path dependencies
(§10.5).
Still, there is no conclusive answer to the question whether economic growth
tends towards a sustainable steady-state or not. Does it matter what is produced
and consumed? Perhaps, the transition to a new service- and information-driven
economic regime will relieve the tensions posed by finite resources, as examined in
§14.1. Does human behaviour matter? Mainstream economic science sees individuals
as rational agents, who immediately and intentionally adjust to whatever is econom-
ically optimal under new conditions (Appendix 14.1). Economic growth models can,
therefore, not make meaningful statements about a change towards postindustrial
values (§4.2). Do nonlinear feedbacks in the system influence growth patterns? The
analytical growth models discussed so far are pseudo-static in the sense that the
economic system is assumed to reach an internal price-volume equilibrium at any
moment considered (§7.2). ‘[Equilibrium thinking has] so permeated economics that
very few attempts have been made . . . to develop theories in which the existence of
cycles is an integral feature of the economy’ (Ormerod 1998). Real-world supply-
demand mismatches are probably endogenous and can significantly influence the
economic and financial system, but they are absent in economic growth models.
The economic system is far too complex to conceptualise in a single scheme,
even if there were agreement on the mechanisms. Nevertheless, following the system
dynamics approach, this book offers a causal loop diagram (CLD) of three subsys-
tems in the economy that are discussed in this and Chapter 10: consumer behaviour,
manufacturing, and capital and technology (Figure 14.7). The first one is the con-
sumer side, in the upper part of the CLD. It is a reinforcing feedback representation
of a complex process, that revolves around the individual needs and desires for posi-
tive experiences. In Figure 14.7, it is shown as the need or desire to drive a bicycle. In
agreement with psychological research, the level of happiness/contentment is influ-
enced by processes of comparison, competition, repetition and habituation (§6.2).
Figure 14.7. A simplified causal loop diagram representation of the economic system. It por-
traits three subsystems: the consumer side (above), the producer side (lower right) and the
capital and technology side (lower left).
This is indicated with two negative feedbacks. The fulfillment of needs and desires
is satisfied with help of private and public goods or satisfiers and capabilities. The
capabilities are to a large extent supplied in the form of public goods (infrastructure,
education or health), although there is no sharp distinction with privately supplied
(pseudo-)satisfiers. Tax is one of the main mechanisms to finance public goods.
The goods and services that provide the satisfiers and capabilities have to be
produced. This is represented with another feedback loop, in the lower right of the
CLD. In essence, demand leads to investments from which costs and prices follow.
The underlying mechanisms are complex market and institutional processes, but the
widely shared assumption is that lower prices will lead to higher sales in the longer
run in competitive markets. As we have seen, however, the dominant force is technol-
ogy. It is represented in the feedback loop in the lower left of the CLD. Savings and
profits provide capital for expansion and replacement investments, but also for R&D
and innovation. In combination with capital-labour substitution and economies of
scale/scope, it leads to higher labour productivity. This increases income and tends
to decrease unit costs. In this way, technology drives the larger, positive feedback
process of economic growth. It also satisfies the desire for new products that is a con-
sequence of the repetition/habituation process amongst consumers and is stimulated
by advertisements. Although the financial sector is an important or even crucial
component of the economic system from a sustainable development perspective,
Box 14.2. Are values driving growth? Is economic prosperity following our
values or the other way around? There is evidence of a link between the long
Kondratiev waves and political and social values in Western nations during the
last two centuries (Sterman 1986).18 ‘During periods of long-wave expansion,
material wants are satisfied, and social concerns turn to civil liberties, income
distribution, and social justice . . . As the expansion gives way to decline, con-
servatism grows, and political attention returns to material needs . . . During the
downturn, the accumulation of wealth becomes the overriding concern, at the
expense of civil rights, equity, and the environment . . . The variation of political
values is primarily the result of entrainment by the economic cycle’ (Sterman
1986). A deeper understanding of the interacting economic, technological and
social forces is one of the great challenges for sustainability science, more so now
that these interactions have spread globally and are faster than ever before.
I do not discuss it here because there is no coherent body of scientific knowledge

about it yet. Instead, I continue with three other contributions of economic science to
the sustainability discourse: physical inputs in the production function, the empirical
input-output formalism and the dynamics of resource efficiency.
14.3 Source and Sink Constraints in the Economy
14.3.1 Structural Economics: The Input-Output Formalism

A large part of the cost of economic goods and services consists of rewards for
the production factors labour L (wages or salaries) and capital K (dividends or
rent), together making up the value added (VA). Labour – in essence, organisation
and knowledge because most physical work has been substituted for by fuels and
electricity – is usually the largest cost component. Capital, in the form of buildings,
machinery and equipment, is an important second component.
Until the oil crises of the 1970s, energy was not considered in economic anal-
yses and neither were materials. This is understandable because coal and later oil
and gas became ever cheaper in the industrial economy and were considered an
operational input, not a potential constraint. Since the oil crises in the 1970s and
1980s, attempts are made to introduce energy E and materials M explicitly into the
production function, the so-called KLEM production function.19 The results are,
however, rather inconclusive because of the fact that energy can be both substitute
and complement to labour and capital and that technological change is treated at
too high an aggregation level.
18 Including feedbacks, notably the inherent oscillatory tendencies of firms and reinforcing mechanisms
among firms and labour and capital markets, is one of the ways in which long-term Kondratiev waves
are explained (Sterman 1986).
19 The costs of energy and materials as fraction of total factor costs are still small except for a few
sectors such as mining and petrochemicals.With more pressure on land for food and biomass-based
fuels, land may also return into the manufacturing production function.
From a natural science perspective, it has been argued since the 1930s that energy
inputs, if properly measured, are a key explanatory factor of economic growth. More
recently, the role of energy and, particularly, exergy in the economic production
has been examined in detail. It appears that the inclusion of useful work (‘exergy
services’) explicitly as an input in the production function yields an almost perfect
correlation with GDP growth for the period 1900–1975 (United States) and 1960–
1993 (United States, Japan and Germany) (Warr et al. 2002; Ayres and Warr 2005).20
The apparent trend break around 1975 in the United States and other advanced
economies is possibly explained by the oil price hike induced efforts to increase
energy productivity and by the rise of ICT. The analyses make it clear that useful
work is a sine qua non for the growth in economic output realised in the high-income
regions in the 20th century and being underway in emerging economies in the 21st
century.21 It strongly suggests that GDP growth in the low-income regions of the
world must concur with an increase in the use of exergy services, notably electricity.
National economic models distinguish economic sectors, each with their produc-
tion function. Each sector delivers part of its output directly to consumers, including
government and investments. This is called final demand. The remainder is delivered
to other sectors and are called intermediate deliveries. Together with the primary
inputs labour and capital, they make up the input-output (I-O) table (Appendix
14.3). From the I-O table, it is immediately seen how much input from an economic
sector j is used per unit of final demand output from sector i, in €/€. This is the direct
intensity of that input – and the input can be energy, steel or whatever specific entry
the table contains. If you choose labour or capital as the input, you can calculate
the direct labour or capital intensity. Most material- and energy-intensive sectors
have a rather low labour intensity and a rather high capital intensity, at least in the
high-income regions where wages are high. Not surprisingly, direct expenditures on
energy are higher when you buy for €100 products from the chemical sector than
when you spend it on childcare – twenty times more in the Netherlands in 2001. But
those same €100 generate six times more employment when spent on childcare as
compared to chemical products.
It is also possible to calculate the monetary flows associated indirectly with a
final demand output purchase – the indirect intensity of that input. For instance,
the energy used to deliver 1 kg of garbage bags has a direct energy intensity α ij
with i energy sector and j bag manufacturer, but there is also energy used for the
production of the plastic used in the bag (α jk with j plastic manufacturer and k
bag manufacturer) (Appendix 14.3). The sum of direct and indirect intensity yields
the total intensity. The differences in total intensities between sectors are smaller
than the direct intensities. Take the example of childcare: Part of your €100 spent
on childcare will soon be spent afterwards on gasoline by one of the employees
travelling by car as part of her service sector job. This adds to the total energy
intensity of those €100 of expenditures.
An illustrative outcome is given in Figure 14.8a,b. It is done on the basis of
a time-series of annual consumer expenditures (final demand) and I-O tables and
20 For commercial energy instead of exergy use the correlation with empirical data is much weaker.
21 Two other conclusions are that electric power is of more than proportionate importance because
of its high quality, and that there is a discrepancy between the elasticity of production and the cost
(a)
60
Petrol
Electricity
Direct energy requirement per capita (GJ)
50 Natural gas
Household fuel
Solid fuels
40
30
20
10
0
1948 1953 1958 1963 1968 1973 1978 1983 1988
Year
(b)
70
Transport and communication
Other consumption
Indirect energy requirement per capita (GJ)
60 Leisure
Hygiene and medical care
Clothing and footwear
Household effects
50 House
Food
40
30
20
10
0
1948 1953 1958 1963 1968 1973 1978 1983 1988
Year
Figure 14.8a,b. The direct and indirect energy use per capita per year for Dutch households
from 1948 to 1988. The indirect energy use is given for eight consumption categories (Vringer
and Blok 2000).
some simplifying assumptions. The rise in income from about 7,000 to 18,000 NLG
(Guilders) per person per year between 1948 and 1988 coincided with an increase of
shares of energy and labour (Kummel et al. 2002). The latter points at the necessity of higher energy
prices and lower labour costs/prices.
3.4 percent per year (%/yr) in total energy use per year per person. Direct energy
use started to rise rapidly after 1965, partly in response to the discovery of the large
Groningen gasfield (Figure 14.8a). In 1990, less than half (44 percent) of total energy
use was spent directly on energy in the form of fuel and electricity (Figure 14.8b).
The remainder was indirect energy use, with a significant but declining fraction of
it for food. I refer the reader to the Suggested Reading for more detailed methods
and results.
There are some pitfalls in the use of I-O tables. One of them is how to deal with
trade in open economies. An accurate treatment is to include the energy embodied
in imports and in exports on the basis of transport and I-O tables for the trade
partners. Although this is a rather tedious exercise, it can shed light on an element
of unsustainability in globalisation, namely that stricter environmental regulation
in high-income regions makes it, in combination with cheap labour in low-income
regions, profitable for corporations to move their (material- and energy-intensive)
sectors away from high-income countries towards countries with less strict regulation.
Theoretically, this is simply using a comparative cost advantage. Practically, it can
lead to a slowdown or even decline in environmental and other regulation.22 Analyses
based on I-O analyses with estimates of embodied energy in imports/exports indicate
that at least part of the dematerialisation (in MJ/€) in the United States, Europe
and Japan has occurred because of shifts in economic activity to Mexico, China and
other low-income regions.23
I-O tables can be of great help in linking physical flows to monetary flows. The
elements of the I-O matrix are converted to physical units with use of (average)
sectoral prices and double-checked with statistical data on physical flows in so-called
satellite accounts. These represent I-O tables in physical flows and are sometimes
called physical input-output tables (PIOT) (Ayres and Ayres 2002). A systematic
connection between monetary and physical I-O stocks and flows was first applied in
energy analysis. Later, environmental accounts were used to assess the the use of
materials and amount of emissions per unit of final demand (t/€) (Wilting et al. 2008;
Peters et al. 2011). In this way, the direct physical input (energy or material) and
output (emission) flows associated with consumer and government expenditures can
be calculated. A further extension is to include more detail on the primary factor
contributions, for instance, separating non-paid agricultural labour and distinguish-
ing income classes. Such a framework is called a social accounting matrix (SAM). For
low-income agrarian societies, this is a more adequate framework than the monetary
I-O table (Morrison and Thorbecke 1990).
An I-O matrix gives a static picture or snapshot of the structure of an economy.
One of the challenges is to connect economic growth models and I-O data in a
transparent and effective way. One way to do this is to isolate investment flows as a
separate final demand category. In this way, the empirical data on structure can be
22 The country with the lowest standards in environmental regulation – and other regulation, such as
labour conditions and nature protection – attracts industries in search of higher profits and force
other industries to lobby for less strict regulation in their country. This is the race-to-the-bottom
dynamic.
23 This is one of the aspects that complicate the negotiations about carbon emission reductions in the
context of international climate policy, because it invalidates a simple GJ/€ measure as indicator for
emission targets.
Box 14.3. Rebound effect. One interesting feedback phenomenon is the so-
called rebound effect (Polimeni et al. 2009). It is simply explained. If cost-effective
measures to reduce energy and material use in industrial production help to
lower production cost, they will also in the longer run reduce the product price,
if competition is working. This in turn probably leads to a higher demand for the
product and hence to production expansion and a higher overall material-energy
use. As a consequence, a part of the benefits of the measure (material-energy
saving) are annihilated by growth. It was famously stated by Stanley Jevons in his
book The Coal Question (1865): ‘It is wholly a confusion of ideas to suppose that
the economical use of fuel is equivalent to a diminished consumption. The very
contrary is the truth’.
A similar mechanism occurs in private households when money not spent on
fuel because of better insulation or lower room temperature is spent in other
ways, thereby leading to additional activity and hence additional use of energy.
Some have called it the Torremolinos effect, suggesting that the saved money is
used to fly with TransEasyRyan to a Spanish resort. Rebound effects – and other
economic feedback effects such as carbon leakage – are studied extensively and
their extent is still a matter of controversy.
combined with a standard economic growth model in order to simulate the system
as it goes from one equilibrated situation at time t to the next at time t+t. In
between, the technological coefficients are adjusted to represent trend- or expert-
based expectations about (future) technologies (Sassi et al. 2010). This is probably
one of the more solid ways to explore future demands on natural resources (Duchin
and Lang 1994).
14.3.2 Resource Efficiency and Pollution Abatement:

Economic Mechanisms
The dynamics behind the changes in material and energy intensity is an important
topic in industrial economics/ecology and also in sustainability science (§13.5). Most
goods and services are produced with a variety of inputs and consist also of many
different ‘products’. Engineers prefer to construct production functions on the basis
of physical flows (engineering production functions), but this is only possible for a few
basic processes (steel, cement and some others). Currently, complex manufacturing
processes are investigated in dedicated technology assessments. Monetary analyses
in production economics offer nevertheless some useful insights in the mechanisms
of resource efficiency, or resource saving as it is popularly called, and provide some
guidelines for a sustainable resource use policy.
Assume that a manufacturing plant produces P units/yr of a good with machi-
nery valued as a capital stock K. It uses E energy units/yr. Energy use can be
reduced by investments, and thus substitute capital for energy. At the level of a
manufacturing plant, office or house, there are numerous examples of additional
investments, for instance, more insulation, double glazing, larger heat exchangers
and so on, that lead to a reduction in energy required to produce the same
output.24 It is described with a production function, for which we choose again
the simple CD form:
P = P(K, E) = At P0 Kα E β t/yr, m3 /yr, . . . (14.9a)
At is the productivity-increasing technology factor. The total costs to produce P units

is given by:
C = pk K + pE E €/yr (14.9b)
with pi the price of factors i = K, E respectively. An equivalent formulation can be

used for materials M. The formulation implies that an energy flow can be substituted
with a capital stock. It is assumed that there are no positive or negative effects
on output from increase in scale, such as constant returns to scale (α + β = 1).
Empirically, the production function is a set of points that represent combinations
of inputs K and E in existing plants, long-term analyses or future plants.25 They
represent the production frontier.
What does this simplified model tell about energy efficiency? Suppose that a
firm is producing on the production frontier. Any change in the pE would induce a
switch to a technology using less E and more K as long as the savings exceed the
cost. In other words, as long as:
pE E ≥ pk K€/yr (14.9c)
with prices pK and pE the respective factor prices and under the assumption of cost
rationality.26 To understand this process, I express equation 14.9a-b in per unit of
product P for a given output level P0 . Assuming At = A0 , this yields:
k = A−1/α
0 e(α−1)/α (14.10a)
c = pK k + pE e (14.10b)
with c = C/P0 , k = K/P0 and e = E/P0 . Plotting the product cost c as a function of
the energy input e shows the two components (Figure 14.9). The linear part are the
energy cost and proportional to e (equation 14.10b). The downward sloping curve are
the capital cost expressed as function of e (equation 14.10a). The black curve are the
total costs at an energy price pE = 10. The grey curve are the total costs at an energy
price pE = 1. The arrow indicates that an increase in energy price pE should induce an
additional capital investment of about 50 percent that reduce energy use with about
60 percent. This ‘energy-saving’ operation makes the unit cost about ten units less
24 The empirical data have as yet not given unambiguous confirmation of capital-labour substitution at
the aggregate level of a sector or an economy. As said before, this is because capital and energy are
not only substitutes but also, and certainly in a longer time perspective, complements (mechanisation,
automation or robotisation) (Frondel and Schmidt 2006).
25 Note that the input of material or energy per unit of output is the material intensity (t/€) or energy
intensity (MJ/€).
26 It is assumed that the additional investment K is depreciated over an economic lifetime 1/m
(equation 14.1). This can be done in the form of an annuity, which includes depreciation and rent in
the cost of capital pk. The reduction in material and/or energy flows ME is on an annual basis.
k-cost e-cost
TotalCost at pE=5 TotalCost at pE=1
50
40
Cost
30
20
10
0
0 2 4 6 8 10
Energy input (normalized)

Figure 14.9. Substitution dynamics between capital and energy. The dashed curves indicate
the per unit cost as function of the energy input (A0 = 1, α = 0,5, pK = 10 and pE = 5). The
upper dashed curve are total costs. The linear curve are energy costs. If pE goes up, the total
cost and energy cost curves move upwards, and the minimum cost level will be at a lower
energy input.
costly than doing nothing (the grey area). The plant is then operated at higher fixed
capital cost but lower operational (energy) cost. It can be proven that the minimum
cost in the total cost curve is for Eopt = (pK /pE )Kopt .27 It represents the optimal
investment in energy efficiency improvement and corresponds in engineering terms
with, for example, a certain size of a heat exchanger surface, a certain thickness of
insulation material or a certain number of steps in a distillation column. Or, in your
house, with a certain thickness of insulation or triple-glazed window surface.
The example illustrates the importance of price responses in resource efficiency.
It also offers two other insights. First, it is the ratio of energy and capital price, not the
energy price per se, that induces efficiency gains. If the price of capital pK increases
too, for instance because the investor can get a higher return elsewhere, then the
reduction in energy use will be lower. Second, the approach presumes a continuous
set of techniques being available as energy-savings options. In reality, there will be
a discrete set of available techniques on the production frontier which are always
‘on the move’ in the K-E-plane because of (expectations about) innovations, energy
prices, environmental regulation, changing tax regimes, wage negotiations and so on.
Indeed, these options are usually only developed when the energy price (is expected
to) increase.
Manufacturers, as well as office managers and households, will often combine
replacement investments with expansion investments. This makes it more difficult
to find optimal estimates of E and K. In the practical evaluation of whether to
27 Of course, energy also substitutes for labour and environment, for instance in automation and
pollution abatement. This makes the analysis more complicated. The optimal factor allocation has
historically been applied to capital and labour and was one of the controversies in economic history
(Samuelson 1947).
switch from technology 1 → 2, often the simpler criterium of the payback time (PBT)
is applied:
K2 − K1
≤ PBT (14.11)
pE (E2 − E1 )1
This rule-of-thumb investment criterium says that a firm will only decide to switch
to a more energy-efficient technique if the ratio of additional investment and annu-
ally saved expenditures is less than PBT years. The lower PBT, the stricter is the
criterium. The higher pE , the more efficiency-investments are made.
This type of analysis gives a feel for the energy and material savings potential and
for the opportunities to reduce the physical flows in manufacturing. Also, it helps to
distinguish different efficiency potentials. The theoretical potential – which is in phys-
ical terms but not in monetary terms determined by the laws of thermodynamics –
cannot be realised. It is some vertical line to the left in Figure 14.9.28 The technical
potential at any given moment will rarely be realised, because it will run against
the rising cost barrier and becomes too expensive. The economic potential is given
by the optimal investment at the minimum overall cost. Note that the cost curve
in Figure 14.9 is the equivalent of the supply cost curve (SCC) in energy supply
(§13.3). Therefore, its empirical equivalent is called an efficiency supply cost curve
(ESCC). The ESCC may fall over time when incremental innovations in the process
of learning-by-doing bring down the production frontier and the economic potential
approaches the technical potential.29 Even more important are technological break-
throughs, such as fundamental innovations that lift the technical potential, thus
creating a new branch in production space. In some situations, the total cost curve
is almost flat, which suggests that energy efficiency is possible at no additional costs.
Such possibilities are called no-regret or win-win options. But the common situation
is that, after the ‘low-hanging fruit’ has been picked, more efficient technologies and
organisational measures are accompanied by higher cost.
From this brief exposition, it is seen that the potential to reduce the resource
intensity of a process is a moving target. It depends on the changes in factor prices
and on the progress made in technology and organisation. Moreover, other product
and process innovations and marketing dominate business decisions, because in
most firms the costs of energy and materials are only one, often minor, element
in a dynamic and competitive environment of capital and labour related decisions.
Moreover, a large fraction of office managers and households still have only limited
knowledge of energy/material costs and, if so, see them in relation to their income –
and conclude, for instance, that a tough negotiation about wages brings in more
money than investing in resource efficiency. Finally, there is the rebound effect.
These barriers to sustainable resource use have to be overcome by better information
and indicators and by gradual changes in perceptions and behaviour.
Upon widening the system boundary, the production space becomes more-
dimensional and the production frontier is no longer a well-defined set. Other,
complementary effects are identified that may stimulate but also hinder resource
28 Strictly speaking, one should, therefore, use a production function of the form Y ∼ Kα (E − E0 )1−α .
29 Innovation should be interpreted here in a broad sense: It refers to technical as well as organisational
and legal skills, practices and measures.
Box 14.4. The industrial economy growth paradigm. Although most people
are not familiar with economic growth models, the idea that economic growth
is good and necessary is widespread. In April 2010, an editorial comment of
a Dutch quality newspaper stated that ‘a healthy economic growth is the best
way to relieve the debt burden. Whatever other considerations, this should be
a short-term priority’. One often hears mentioning the unemployment rate as
an argument for economic growth. But in 2010 the debts accumulated from pro-
growth financial policies and rescuing of the banking system became the argument
for economic growth.
In February 2007, the same newspaper had as its headline that the Dutch
Central Bureau of Statistics (CBS) announced 2006 as an excellent year for
the economy ‘with a growth of 2,9% . . . The growth in 2006 is however smaller
than the 3,25% which the cabinet used as its starting point in its Miljoenen-
nota . . . Germany announced this morning 2,7% over 2006 and France 2%. The
Dutch economy may in the fourth quarter have been affected negatively by the
very mild autumn. This caused a lower demand for natural gas by consumers
and the energy sector therefore contributed less to the growth’. This article high-
lights at least three phenomena: the government (always) expects higher growth,
a country judges its growth performance in competition with others, and less
energy use is seen as bad for economic growth.
As a third example of the economic growth addiction: numerous are the news-
paper items that state that ‘economic growth is good for the environment’. But
growth-enhancing expenditures to combat pollution amongst the rich is very dif-
ferent from expanding economic output to satisfy basic needs of poor people.
All three examples are symptoms of the fundamental dilemma: with growth the
boundaries and feedbacks of a finite earth, without growth the risks of social
instability – a genuine quandary for governments. Human social and economic
systems are like ‘somebody who is struggling not to fall forward and can only
prevent this by walking forward’ (Jantsch 1980).
efficiency investments. For instance, the actual realisation of a switch from technol-
ogy 1 → 2 often turns out to incur additional ‘costs’ such as gathering inform-
ation, surveying the work, maintenance arrangements and so on. Such costs are
called transaction costs and they are often overlooked – making it less puzzling that
efficiency investments are not made or lagging. Other obstacles to realising the eco-
nomic potential are status-driven behaviour, incomplete information on prices and
available techniques, and diverging values, for instance, between cultures. It can also
work in the other direction, as with status, and if one includes the positive extern-
alities such as lower emissions and other co-benefits, there may exist even negative
cost options.
The preceding formalism can also be applied to pollution abatement. Invest-
ment decisions are made in order to replace part of operational costs for waste
management and emission charges by fixed capital costs. Those operational costs
now concern non-market priced environmental goods such as clean air or water, and
society must agree on some price, via taxes or emission trading, or introduce and
enforce regulation and standards in order to elicit the necessary investments (Pearce
1990; Perman et al. 2003).
14.4 Economic Growth and Sustainable Development
14.4.1 GDP and the Need for a Better Indicator

The usual measure for the standard of living is the gross domestic product (GDP)
per person (Appendix 14.4). In combination with the expectations and aspirations
of people, heightened in today’s globalising world of communication and adver-
tisements, there is an enormous momentum for ongoing economic growth of GDP.
Such growth is hoped to satisfy the large still unmet basic needs of the population
in low-income regions and provide jobs in order to distribute rising welfare and
prevent social unrest. It is also driven by the desires of an emerging middle class
for the luxuries of modern life. In the high-income countries, only a combination
of steady innovations, advertisements and high levels of working hours can bring
ongoing growth in GDP and income. Here, too, it is considered the recipe for jobs
because employment is the core mechanism of wealth redistribution and meaningful
social positioning. In both low- and high-income regions, growth is propelled by the
desire for returns on capital, a continuous battle for corporate profits and market
share, large advertising and sales efforts and the stimulants provided by the financial
system. Governments are also keen on economic growth because it is a prominent
way to satisfy the large needs for public goods (roads, schools, hospitals or army).
Besides, politicians depend on budgets for their status as a person and as a nation –
and occasionally for their personal wealth. At a psychological level, growth in income
satisfies the needs for personal success, social distinction and novelty.
Therefore, income growth is expected to continue, driven by the spontaneous
urge to activity which Keynes called animal spirits. From a sustainability perspec-
tive, one should ask whether continued economic growth is possible in view of the
constraints of a finite planet. Will it not start to interfere with its very aim of a better
quality of life? And, relatedly, is a growing income the best or only way to well-
being? I leave it to the reader to construct his or her own answer, with the previous
chapters as possible guidelines. But whatever the answer, the question needs to be
addressed as to whether income measured as GDP/cap is an adequate indicator of
well-being in a sustainable development framework. The answer is no, for various
reasons (Hueting 1980; Daly and Cobb 1989; Daly 1996; van den Berg 2009).
Surveys suggest that income is only at low levels clearly correlated with well-
being – and even there it depends on context (§6.1). The use of average income as the
standard measure of well-being makes it worse, because it neglects the role of income
differences and comparison. However, there is a more down-to-earth accounting
objection: In the calculation of GDP, there is no distinction made between costs and
benefits. This may cause a serious overestimation of GDP as a source of quality of
life. Many market-related transactions contribute to GDP but are not benefitting
anyone and are actually costs (negative externalities). Well-known examples are the
expenditures associated with traffic accidents, pollution abatement and natural dis-
asters. Advertisements are counted as benefit, although they largely reflect corporate
battlecries. The experience of nature is to an ever larger extent associated with travel
and accommodation expenditures. Financial transactions show up as contributions
to GDP, although they are at least partly an indicator of excess greed and risk. And
one can also wonder about the medical and psychiatric expenditures to combat the
side effects of life in a (post)industrial society.
A second flaw is that there is no distinction between stocks and flows. Resource
use is accounted for in GDP on the basis of the monetary value of the extraction
flow, without accounting for the decrease of the resource stock. It is like counting
only the money you withdraw from your account, without considering the debt you
are building up meanwhile.30 Similarly, the degradation of environmental sinks is
not counted in GDP, whereas it incurs future costs in order to maintain quality of
life that are counted in future GDP. Thus, many negative externalities are either
counted as positive contributions or not counted at all or are counted showing
up as ‘contributions’ to future GDP. This partly explains newspaper statements
like ‘Catastrophic floods stimulate economic growth’ or ‘Pollution boosts industrial
investments’.
Two alternative indicators explicitly address these shortcomings (§5.5).31 The
first one is the Index of Sustainable Economic Welfare (ISEW), which tries to account
for contributions to GDP that are actually costs (Daly and Cobb 1989). It has been
constructed for many countries. Using a set of conventions and assumptions, well-
being measured as ISEW stopped growing in most countries some decades ago.
The second indicator is the genuine savings rate, which focusses primarily on the
adequate inclusion of changes in natural and human capital. This one has also been
constructed for many countries. The results suggest that GDP growth is significantly
overestimated in countries that overexploit their oil and forest resources, whereas
it is underestimated in those that have invested in education. I refer to the Useful
Websites for more details. Interestingly, there are also reasons to suspect that GDP
and income underestimate our well-being, or at least welfare and utility. This stems
from the fact that GDP only counts activities that are associated with formally
registered monetary transactions. In pre-industrial Europe and still in parts of non-
industrialised countries, many goods and services are produced and consumed within
the confines of family and village. In many places, labour is still paid in natura in
exchange for protection and market transactions take place outside the monetary
system (§3.3).32 GDP underrepresents these economic activities. Some attempts
have been made to calculate their monetary equivalent, resulting in more weight for
the role of agriculture.
A second, overlapping category of activities that are not counted are house-
hold activities. Statistics and economic textbooks see households as places of
consumption and leisure. Including non-monetised transactions in households by
valuing working-hours at ongoing average wage rates would increase GDP with
the non-market activities (cooking meals, cleaning clothes, rearing children, do-
it-yourself repair and so on). For subsistence, for households in agrarian but
even more so in industrial societies, the value of the production in households
30 In this sense, the debt crisis in many countries if not different from the resource squandering in
others.
31 There are many additional indicators that are used to evaluate a country’s economic situation. For
instance, the Human Development index (HDI) and the Corruption Perceptions Index emphasise
social-political aspects, whereas the Competitiveness Index and a host of financial indicators reflect
supposedly the economic situation.
32 In large parts of rural India, for instance, more than 90 percent of human activities is estimated
to be in the informal economy and not recorded in monetary balances (Dasgupta and Singh
2005).
(gross household production or GHP) may be significant. In the United States, it has
been estimated to be in the order of one-third of GDP. In Australia, the time spent on
meals, laundry, childcare and shopping was in 1975–1976 equivalent to an estimated
60 percent of total GDP (Ironmonger and Sonius 1989).
The net result of these deficiencies in GDP are difficult to assess. On the one
hand, ever more activities are incorporated in the formal economy through increasing
employment of women, the desire for control and taxes by governments and most
politicians, and the pressures and conveniences of ICT services offered to citizens
by banks and governments.33 On the other hand, the transition to a postindustrial
service, information and experience economy generates a wealth of new activities –
think of Internet websites and financial products – that are not or only indirectly
included in the national statistics.
Yet another mismatch between income and well-being stems from the difficulty
to measure the contributions of a non-ending wave of innovations to individual and
collective well-being that are counted in GDP only with their monetary equivalent.
I now get a thousandfold faster computer for 100 hours of paid work than thirty
years ago for 100 hours of work. The opportunities and positive externalities of ICT
have increased beyond imagination and exceed by far the growth of income per se.
This is often not perceived because of habituation and real and imaginary inflation
(§6.2). It seems inevitable that the usefulness and adequacy of GDP and income as
an indicator of ‘the good life’ will further deteriorate.
The many facets and ambiguities of well-being and quality of life cannot be
expected to be covered with a single indicator. Nevertheless, it is necessary and
useful to look for indicators that can set targets for and use measurements of progress
towards sustainable development (§5.4). The challenge is to find a proper balance
between the universality and locality and between the material and the immaterial.
Universality of an indicator is desirable in order to reflect strong knowledge of the
world and to make comparisons possible, but locality is needed in order to respect
the large diversity in people and their quality of life situations and experiences. The
material aspects of quality of life have to be covered because aspects such as adequate
food and access to health services are essential and measurable ingredients of the
good life. But the immaterial aspects should not be forgotten, as access to education
and the right to justice are equally important though harder to measure determinants
of a good quality of life.
14.4.2 Beyond Models: Welcome in the Real World

Elementary models of economic growth, it seems, make up a consolidated body
of theory, closely connected to prevailing attitudes and practices, and there is as
yet no coherent alternative.34 Yet, there are numerous shortcomings and criticisms
of mainstream economic theory (§10.4). This is not surprising because the (world)
economic system is extremely complex and strong knowledge about it is not to
be expected. The inadequacy of economic theory is, however, partly because of an
33 It has been estimated that the entrance of women to the formal labour market was responsible for
up to one-third of GDP growth in the Netherlands in the 1990s.
34 See, however, the list of Useful Websites for various new directions.
overemphasis on mathematical formalism: ‘The human mind is built to think in terms

of narratives, of sequences of events with an internal logic and dynamic that appear
as a unified whole . . . It is generally considered unprofessional for economists to base
their analyses on stories. On the contrary, we are supposed to stick to the quantitative
facts and theory – a theory that is based on optimisation, especially optimisation of
economic variables’ (Akerlof and Shiller 2009). I, therefore, discuss in a more story-
like fashion two issues that are particularly relevant in a sustainable development
context and difficult, if not impossible, to incorporate in models: public goods and
governance. For more sophisticated treatises, I refer the reader to the Suggested
Reading on these topics.
Knowledge and infrastructure – including our cultural heritage – are in essence
public goods and provide public services. It is evident that accumulation and diffusion
of knowledge increases productivity. It is driven by educational institutions and
R&D investments and has important positive externalities in an industrial economy.
Infrastructure, for water, energy and transport and, more broadly, for education and
health services and social security is equally essential for economic prosperity.35 They
generate positive externalities or spillovers: the effects benefit more people than just
the ones who are directly involved. For reasons of planning, cost and fairness, they
use to be provided by governments and are considered a ‘natural monopoly’.36
It can be expected that public goods become more, not less, important in an era
of increasing population, rising expectations and growing pressures on the natural
life support system. Increasing densities of people and activities will naturally cause
the need for more regulation and organisation (Figure 12.10). Health, education and
transport services are constituents of the ‘good life’, in the sense of use and capability
to use. Their inherently public goods character demands collective arrangements and
planning, realisation and enforcement. Similarly, their rivalrous twin: common pool
resources (CPR), are crucial for quality of life, as previous accounts of groundwater,
fisheries and forests show. Identifying and governing the‘global commons’ such as
the atmosphere and oceans is one of the greatest challenge for sustainability.
The neoliberal trend of the last decades towards deregulation and privatisation
in the public domain is heading for the wrong direction. It has been legitimised
with the argument that competition leads to more efficiency and lower costs, but
the validity of this argument is one of the worldview-related controversies (§6.4).
Innovations and infrastructural developments cannot and should not be directed
by consumer markets only, because the needs of those who are less powerful and
assertive in the market place – amongst them non-humans and future humans – are
not met. Finding a balance between unregulated market-driven processes and state-
and community-based planning is a precondition for most sustainable development
initiatives. New social arrangements, entrepreneurial models and consumer lifestyles
35 In an analysis of physical indicators (paved roads, telephones, electric power) for a set of countries
over the period 1950–1992, there is ‘clear evidence that in the vast majority of cases infrastructure
does induce long run growth effects’, but with a great deal of variation in the results across individual
countries (Canning and Pedroni 2004).
36 Knowledge and infrastructure have both a public good/service character, but there are also differ-
ences. Knowledge per se is not rivalrous, whereas most infrastructure has congestion effects that
make their use often rivalrous (§5.4). See Table 5.1 with respect to various forms of management.
must be explored, representing a societal contract that bridges the extremes of full
state control and anarchic individualism.37
One of the big questions is whether the necessary new forms of governance
and institutions are possible within the current capitalist system and without a new
‘grand story’. In the 1990s, the system of neoliberal capitalism with democracy as
its political counterpart was by many seen as the ‘end of history’: Nothing better
was to be expected (Fukuyama 1992). Economic theory and practice took market
capitalism and its basic tenets as the starting point. History has proven its superiority –
or so the myth goes. Property rights, competition, the right to free enterprise and
free exchange of goods and services brought the rich countries today’s prosperity.
The reality is different, with its corporate oligopolies, alliances between business and
government and militarily supported control of resources. In the course of the 20th
century, the most destructive aspects of European and American capitalism were
mitigated by government regulations and programs.
With the fall of communism, the forces that restrained capitalism were gone. In
combination with the ICT/revolution, globalisation intensified competition, eroded
the tax basis and the legitimacy of governments and gave room for free-riding and
illegal and criminal practices (Castells 1996, Reich 2007). But the expansion of
the global economy with large, new players (Brazil-Russia-India-China or BRIC)
also brings a new world order with new rules. Whether and how democracy will
fare in this situation is uncertain. Will the ‘emerging economies’ (BRIC) give the
world new models of economic development and political governance? For the
moment, it seems that, in the 21st-century Internet world, NGOs in all their diversity
are probably the best if not only way to draw the divergent private interests in a
permanent dialogue on how to safeguard the overarching public interests.
These events and trends influence the prospects for sustainable development.
The following mentions five areas of concern and directions for solutions.
r The struggle for resources. Rising demand in combination with scarcity and envi-
ronmental change lead to price volatility and speculation, which in turn trigger
social unrest. It induces corruption and war in states with large resources but
weak governance and institutions. World price and trade agreements for specific
commodities are part of the solution. Transfer of technologies and introduction
of standards and labels for more efficient resource use and recycling are another
part.
r The income and wealth gap. A conventional view on economic inequality is that
wealth ‘trickles down’ from the richer to the poorer strata of society with a rise
in average income. ‘A rising tide lifts all boats.’ This view rationalises wealthy
people’s desire for more. But the measured effects of growth on the poorest
segments of society are controversial. Often, a more appropriate aphorism is
‘the winner takes it all’. The erosion of middle classes and taxpayer ethics is
a matter of concern in this respect. There is probably an ‘optimum’ income
inequality, when the incentive to take risks and create wealth is in balance with
the fairness needed for citizen’s compliance. Outside such a balance, social and
37 Recent examples of such explorations are Matsutani (2006), Barnes (2006) and Jackson (2009).
economic stability is at risk. Restoration of government legitimacy and other

forms of redistribution than through paid jobs and taxes are needed.38
r Trade and globalisation. More open economies are said to have higher economic
growth rates. But trade effects are mixed up with trends in transport cost, cap-
ital mobility and credit and debt formation. In theory, trade can encourage or
discourage the growth of income per capita. In practise, the already rich coun-
tries appear to benefit most.39 There is a need for whole-chain analyses in order
to understand and evaluate the effects of trade on local, regional and global
sustainability. Globalisation is perhaps already in the overshoot domain and
targeted forms of protectionism are needed.
r Corporate responsibility. In the absence of legitimate global governance, sus-
tainable management of the ‘global commons’ cannot succeed. The short-term
and quick and high returns still dominate most business. Design and imple-
mentation of more stringent rules for corporations, as laid out in, for instance,
Corporate Social Responsability (CSR) and Earth Charter, are needed and some
corporations already take the lead. The financial system in particular has to be
transformed towards more responsible behaviour.
There are a few principles for a sustainable economy that have to be strengthened
in the coming decades:
r internalise, encouraged by regulation and prices, an ethic that does not permit
shifting the burden onto people far away in space and time and onto other
species;
r as part of it, redirect consumerist lifestyles towards more immaterial lifestyles
that emphasise moderation, intelligence, cooperation and sharing;
r follow basic principles of ecology: there is no such thing as ‘waste’ in nature,
interdependence is the rule not the exception, (bio)diversity is the best insurance
for adaptiveness and spread of risk – and there is only one energy source: the
sun;
r redirect technical innovations towards a better balance between (rich) consumer
wants and (poor) human needs and stimulate a ‘green economy’ of reuse, recy-
cling and renewables;
r apply the subsidiarity principle: global governance where necessary and local
governance where possible, in order to give room to the greatly needed dynamic
and creative forces of civic communities and entrepreneurial capitalism;
Because the arguments are largely based on weak knowledge, there are uncertainties
and controversies in the interpretation and valuation of what is and of what ought
to be. Ethics and the image of man are inherent parts of the discourse. Worldviews
inevitably play a role in the interpretation and appreciation of (models of) economic
growth. As in previous chapters, I offer the reader at the end of this chapter four
38 In this respect, initiatives around basic income (www.basicincome.org/bien/), local exchange and
trade systems (LETS: www.gdrc.org/icm/lets-faq.html) and green business networks are interesting.
Households are also an important agent in redistribution of income and wealth via transfers, grants
and gifts. One estimate for the United States indicates these transfers are three times the government
and private charity flows.
39 There are, for instance, indications that protectionism was promoting economic growth before World
War I and negative thereafter (Helpman 2004).
different perspectives on economic growth in reply to four questions (Table 14.2).

More than in any of the previous chapters, the perspectives and questions about
the economic system will be at the core of the search for a sustainable future for
mankind.
In conclusion, mainstream economic theory and models give neither in a
descriptive nor in a prescriptive way much insight about how a sustainable eco-
nomic system looks like. The economic system must be transformed, preferably
proactively for the better and not reactively for the worse. The germs of the trans-
formation are around us. There is an enormous amount of and variety in ideas,
practises and experiments for a transition to a sustainable economic system. I refer
to the list of Suggested Reading and Useful Websites for further study.
14.5 Summary Points

Many of the sustainability issues regarding provision of food, water, minerals and
fuels are in essence about human behaviour in the face of scarcity – the topic of eco-
nomic science. Nearly all development has an economic aspect in the sense of trad-
ing off alternatives under boundary conditions – and sustainable development is no
exception. Our brief exploration of macroeconomic growth models has given some
insights about mechanisms and processes driving economic activities and growth in
the long term. A simple archetypical model of the postindustrial transition shows
possible pathways to a sustainable economy. But the complexities of the (world)
economic system in relation to sustainability are such that a qualitative discussion
in a worldview framework will always be needed. Here are some of the points to
remember from this chapter:
r (neo)classical models of economic growth focus on the role of capital, labour and
technology (capital-labour ratio’s, substitution, income convergence, steady-
state). The models are founded on oversimplified ideas about the physical and
the social world and are poorly reproducing the empirical data on economic
growth;
r technology and innovation dynamics, human behaviour and subsystems such as
the financial system, government policies and the informal economy must be
incorporated in the models to make them useful for sustainable development
strategies and policies. Complex system science will advance such efforts;
r empirical input-output (I-O) tables and treatment of resource efficiency and
pollution abatement as capital-resource substitution are useful approaches in
assessing overall system effects and resource and environment aspects of eco-
nomic activities and growth;
r we do need a more adequate indicator than GDP in order to measure well-
being and quality of life and to formulate targets and evaluate progress towards
sustainable development. Some promising efforts are underway;
r experimenting with novel arrangements and models of economic activity to
redress perverse incentives for and addiction to material growth and to organ-
ise more fair (re)distributions of what the Earth can sustainably produce are
amongst the great challenges for the 21st century. A proper balance between
global governance and local creativity and community is a crucial ingredient.
Table 14.2. Worldviews and directions for a sustainable economy
A Sustainable Economy
Statement 1: Eco-socialism, that is: the effort to centralise environmental decision making within a state
apparatus, will prove to be the most unsustainable system of all.
History has shown what There is no need for Power corrupts and Typical right-wing
comes from centralised planning: existence of absolute power nonsense. As explained
state power. Business unpriced goods is the corrupts absolutely. A recently by Chinese
should join forces with course of the more sustainable officials, planning is
govenments and civil problem. world starts with each necessary to grapple
society, in order to be Corporations and of us individually. The with the large-scale
an innovative leader in individuals seek out next step is to organise problems facing the
the world economy unpriced goods – that action-from-below at world. Cooperation
and do well in the is neither surprising community level. and coordination are a
‘Green Race’. nor to be condemned. Forget about the state. must.
Statement 2: The runaway spending at the top has been a virus that has spawned a luxury fever that, to one
degree or another, has all of us in its grip (Luxury Fever, 1999).
It may have been This is a moral The extravagance of the Luxury spending of the
exaggerated, but you sentiment that is rich shows the decline rich sends off the
forget how important based in envy. To of Western society wrong signal in a world
the purchases of the appreciate luxury is into materialism, plagued by financial
rich are for innovation an art, and everyone egoism and hedonism. and environmental
and employment. likes to taste its The punishment is crises. Governments
privileges – but that’s inner emptiness and need to regulate the
impossible. social falling apart. extreme differences in
wealth and income.
Statement 3: The existing system of market capitalism creates unemployment and inequity, that breed
insecurity and instability. Sustainable development is out of reach without its transformation.
It is true that poverty and Typical left-wing Correct. We need to find History has shown the
inequity are critical nonsense. Thanks to novel ways, for failures of unregulated
challenges for global market capitalism, instance a basic market capitalism. The
stability and most of us have a income. It may be the strengths of capitalism
sustainability. But the good life. Insecurity only viable way of have to be combined
root problems are and instability come reconciling poverty with strict regulation of
subsidies, lack of from the lazy, the relief and full labour conditions,
market incentives and losers and the thieves employment. Also environmental
incompetence of – amongst them the local money and practises, speculation
governments. immigrants. exchange systems are and finance, trade a.o.
part of a real solution.
Statement 4: Innovations and open trade are the solution to development – and they are also the solution to
sustainability problems.
Correct. Economic Correct. But in an This is the mantra of The evidence for an
models and data increasingly crowded modern globalism and Environmental
clearly show the world, there may be capitalism. Often, the Kuzents curve is not
importance of not enough for opposite happens: convincing. It is
innovations and trade everyone. Besides, local economies, dangerous to count on
in promoting growth. one has to ensure cultures and GDP growth as the
Only then, the poor intellectual property communities are way out.
will be able and willing rights and a level destroyed and
to support playing field for trade. replaced by uniformity
sustainability policies. and consumerism.
The difficulty lies not with the new ideas, but in escaping the old ones.
– John Maynard Keynes
The important problem of steady state would not be production but distribution.
You can no longer avoid the problem of relative distribution by resorting to growth.
– Herman Daly, personal communication
In industrial capitalism, the production of economic goods along with the system of
allocating them has conditioned the type of satisfiers that predominate . . . . [it] leads
to an alienated society engaged in a senseless productivity race. Life [is] placed at
the service of artifacts, rather than artifacts at the service of life. The question of the
quality of life is overshadowed by our obsession to increase productivity.
– Max-Neef, Human Scale Development, 1991
‘ . . . the modern economist has been brought up to consider ‘labour’ or work as

little more than a necessary evil . . . Hence the ideal from the point of view of the
employer is to have output without employees, and the ideal from the point of view
of the employee is to have income without employment . . . The Buddhist point of
view takes the function of work to be at least threefold: to give a man a chance to
utilize and develop his faculties; to enable him to overcome his egocentredness by
joining with other people in a common task; and to bring forth the goods and services
needed for a becoming existence . . . the Buddhist sees the essence of civilisation not
in a multiplication of wants but in the purification of human character . . . [which] is
primarily formed by a man’s work . . . While the materialist is mainly interested in
goods, the Buddhist is mainly interested in liberation . . . [and] it is not wealth that
stands in the way of liberation but the attachment to wealth; not the enjoyment of
pleasurable things but the craving for them’.
– E.F. Schumacher, Small is Beautiful, 1973
Stretch a bow to the very full,

And you will wish you had stopped in time.
– Lao Tze, Tao Te Jing
SUGGESTED READING
This textbook gives a comprehensive overview of macroeconomic growth models with detailed
mathematical analyses. The words resources and environment are not in the index.
Barro, R., and X. Sala-i-Martin. Economic Growth, 2nd ed. Cambridge: MIT Press, 2004.
A textbook with elementary notions in micro- and macroeconomics.
A concise and humane introduction in the core concepts of an essential yet complex scientific
discipline.
Dasgupta, P. Economics – A Very Short Introduction. New York: Oxford University Press,
2007.
An elementary textbook with both disciplinary and more general topics.
Folmer, H., H. Gabel, and H. Opschoor, eds. Principles of Environmental and Resource
Economics. New Horizons in Environmental Economics. Cheltenham: Edward Elgar, 1995.
An in-depth investigation of the events, persons and forces of the industrial revolution.
Heilbroner, R., and W. Milberg. The Making of Economic Society, 12th ed. Boston: Pearson
Prentice-Hall, 2005.
A critical exposition of the explanations of economic growth that are given in macroeconomic
literature.
Helpman, E. The Mystery of Economic Growth. New York: McGraw-Hill, 2004.
A detailed historical description of the evolution of concepts and theories.
Pearce, D. (2002). An Intellectual History of Environmental Economics, Annual Rev. Energy
Environ. 27 2002: 57–81.
A rather radical but in-depth critique of dominant financial-economic theory and practice.
Keen, S. Debunking Economics. London/New York: Zed Books, 2012.
An extensive treatment of natural resource economics in a broad neoclassical framework.
mental Economics. Harlow, UK: Pearson Education Ltd., 2003.
Overview of economic theory regarding natural resource use, from a variety of angles.
van den Bergh, J., ed. Handbook of Environmental and Resource Economics. Cheltenham:
Edward Elgar, 2002.
USEFUL WEBSITES (SEE ALSO CHAPTER 5)
Theory and Data

r www.beijer.kva.se/research/research.html is a site devoted to ecological economics
research.
r www.unifr.ch/econophysics/ is a blog about the need and suggestions for alternative,
physical science–oriented economic theories and models (‘econophysics’).
r www.neweconomics.org/ is the site of the New Economics Foundation, with non-standard
news and critical views on poverty, environment, etc.
r ineteconomics.org/about of the Institute for New Economic Thinking, www.green
economycoalition.org/ and www.communityeconomies.org/Home are three of the many
sites that engage citizens in the search for a more sustainable economy.
r www.basicincome.org/ is the Basic income Earth Network, exploring the idea of basic
income as a solution to inequity and unemployment.
r www.itcilo.org is organised by the International Labour Organization (ILO) and has
information on employment and corporations. See also www.ips-dc.org/reports/top 200
the rise of corporate global power# on trends in global corporate power.
r www.gdrc.org/informal/ is the site of the Global Development Research Center (GDRC)
with papers on the informal sector.
r www.wiod.org/index.htm is the site with the recent World Input-Output Database.
Another widely used set of I-O data is available as part of the Global Trade Analysis
Project (GTAP) at www.gtap.org.
r ec.europa.eu/environment/enveco/index.htm is the site on Environment and Economics
of the European Commission with illustrative documents and links.
Indicators, Models and Games

r csls.ca/iwb.asp is the site of the Index of Economic Well-being (IEW).
r www.econmodel.com/classic is a site with a description and exercises of classic macro-
and microeconomic models (neoclassical growth model or cobweb supply-demand
mechanism)
r www.iiasa.ac.at/collections/IIASA Research/idocs/Research/TNT/WEB/Software/
LSM2/lsm2-index.html?sb=3 is about the logistic substitution model as developed and
applied at IIASA.
r www.rug.nl/ees/onderzoek/ivem/publicaties/software/eap is on the interactive model
EAP, which calculates on the basis of I-O analysis and the energy impact of your
expenditures.
r www.btplc.com/Societyandenvironment/Businessgame/ gives the opportunity to experi-
ence society and environment dilemmas in a business environment (30-minute game).
Appendix 14.2 Evolutionary Models of Producers and Consumers 511
Appendix 14.1 A Simple Behaviour Model of Saving

A counterpart of the production growth model is the standard model of savings
behaviour of households (Barro and Sala-i-Martin 2004). The assumption is that
a household decides the savings rate σ , that is: the fraction of income that is not
consumed, on the basis of an intertemporal maximisation of the discounted utility.
Assuming equilibrium on savings and labour markets, the savings S per household
obey the equation:
dS
= rS + wL − C (A14.1)
dt
because the change in saved assets equals the inflow of interest at rate r and income
w minus the outflow consumption c. The discounted utility can be shown to be at a
maximum when:
dc
= γ (r − ρ) · c (A14.2)
dt
with ρ the household’s time preference and γ a measure of the marginal utility, that
is ± how fast utility declines for rising consumption levels. This behavioural rule,
which is associated with the British mathematician Ramsey, says that households
will postpone consumption for future generations if they expect the interest rate
to exceed the time preference (r > ρ → dc/dt > 0). Otherwise, they will prefer to
consume now and have a declining consumption level for future generations. This
economic image of household behaviour expresses the importance of short-termism
versus caring for the future. In short, the demands of capital markets and banks
versus sustainability. In combination with a model of behaviour of firms, the savings
rate becomes an endogenous part of the economic growth process. I refer to the
Suggested Reading for more details.
Appendix 14.2 Evolutionary Models of Producers and Consumers

This appendix briefly introduces two evolutionary economics models for illustrative
purposes (de Vries 2010). Dosi et al. (2008) have constructed an evolutionary, multi-
agent model of an economy in line with the foundational work of Nelson and Winter
(1982). There are F firms and L workers/consumers. The firms belong to either
consumption-good firms F1 or machine-tool firms F2 . The consumption-good firms
(F1 ) plan investment decisions, such as orders for the machine-tool firms, on expected
demand, desired inventories and desired capacity utilisation. The key equations
describing investment behaviour of firm j are:
r the desired production level Q in period t depends on expected demand and
desired inventory;
r the desired capital stock K in period t is a function of the (desired) production
level and the (desired) level of capacity utilisation;
r given the labour productivity of machine tool producer and vintage, the unit
labour cost c is calculated as the ratio of wages and labour productivity.
The actual investment only takes place above a certain trigger. A crucial part is that
new capital stock has a higher labour productivity and thus lower product cost.
A firm decides to scrap old capital according to a simple payback criterium and
orders new capital from a subset of suppliers (F2 ). Which machine-tools are bought
depends on the price and productivity of this subset of suppliers. As a consequence
of expectations and imperfect information, firms will perform differently in the
consumption-good market. This is simulated on the basis of the replicator dynam-
ics (Appendix 10.2). The authors identify a dozen of macro- and micro-empirical
regularities about modern economies and distill from their model a number of vari-
ables/distributions for comparison with macro-statistics. The results do reflect most
of the ‘stylised facts’ and generate fluctuations in investment, consumption, employ-
ment and other variables in agreement with empirical counterparts.
Another evolutionary economics model focuses on the direction and pace of
innovations from the co-evolution between producers and consumers (Safarzyn-
ska and Van den Bergh 2010). On the producer side, it simulates the probability
of technological lock-ins, path dependency and the role of quality improvements
and marketing in the evolutionary ‘survival of the fittest’. On the consumer side, it
simulates network effects, such as the influence of other individuals (the social net-
work) on the decisions of a person. Such effects can represent important feedback
mechanisms, which are from an innovation. In the context of innovation policy for
sustainable development, these feedback processes can be positive as in the band-
wagon or herd effect and its opposite, the snob effect, or negative as in conformity
behaviour. The existence of multiple equilibria in models like this one offers the
prospect of understanding sudden large changes in economic systems (§9.6).
Appendix 14.3 Input-Output Tables

An input-output (I-O) table or matrix A contains intersectoral flows between eco-
nomic sectors. The coefficients Aij of the square matrix A indicate the monetary
transactions between sectors i and j in the formal economy during a certain period,
usually a calendar year. In network terminology, the economic sectors are the ver-
tices and the transactions of buying from other sectors (input) and delivering to
other sectors and to final demand (output) are the edges (§10.5). The level of detail
depends on the available statistics and on the purpose of the analysis. If I-O tables
have only a few sectors, much information remains concealed in the (large) diagonal
elements of intrasectoral transactions. Lack of detail is because of lack of data and/or
proprietary information. For some countries, there are matrices with hundreds of
sectors while for other countries or regions the disaggregation is limited to a few
sectors only.
The basic scheme is shown in Figure A14.1 and an example is given in Table
A14.1. The columns indicate the payments for inputs to a sector, including the
payments for the so-called primary production factors: wages for labour, dividends
and interest for capital, and rent for land and other resources. The rows are filled
with the deliveries in money, that is, ± sales, of a sector to other sectors and to
the final demand. The latter comprises deliveries to final demand, that is, ± to
consumers, the government and exports minus imports. The sum of the column
elements equals the sum of the row elements. GDP is defined as the total payments
to the factors of production and is equal to the value added (VA) in an economy,
defined as the summed rows of primary inputs capital and labour. Inputs of resources
Appendix 14.3 Input-Output Tables 513
Table A14.1. Example of an input-output (I-O) table: India in 1998. Coefficients larger than 0.01 are
printed in bold, larger than 0.1 also in greya
1 2 3 4 5 6 7 8 9 10 11 12
1. Food crops 0.08 0.00 0.00 0.02 0.02 0.00 0.00 0.00 0.08 0.00 0.00 0.01
2. Cash crops 0.00 0.03 0.00 0.02 0.02 0.00 0.00 0.00 0.17 0.06 0.00 0.01
3. Plantation crops 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.01 0.00 0.00
4. Other crops 0.00 0.00 0.00 0.01 0.20 0.00 0.00 0.00 0.12 0.00 0.01 0.01
5. Animal husbandry 0.04 0.04 0.04 0.04 0.00 0.00 0.00 0.00 0.07 0.01 0.00 0.00
6. Forestry and logging 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.02 0.00 0.00
7. Fishing 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.01 0.00 0.00 0.00
8. Primary minerals 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.01 0.07 0.00
9. Food processing 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.09 0.00 0.00 0.01
10. Other agroprocessing 0.00 0.00 0.00 0.00 0.00 0.01 0.03 0.01 0.03 0.18 0.03 0.01
11. Industrial manufact 0.14 0.11 0.10 0.08 0.01 0.06 0.05 0.15 0.11 0.24 0.37 0.10
12. Services 0.04 0.03 0.03 0.02 0.04 0.02 0.02 0.03 0.15 0.14 0.13 0.11
Primary inputs 0.70 0.79 0.83 0.81 0.67 0.91 0.87 0.80 0.13 0.32 0.38 0.74
a Source: de Vries et al. 2007.
and ecosystem services are not valued until the moment that people add value by
digging, transporting or any other form of processing. The example of 1998 India
(Table A14.1) reflects the great importance of agriculture with its many subdivisions
and the distortion because of the high aggregation level of industrial manufacturing
and services.
Goods & services outflows to: final demand (priv cons, govt cons, exp/imp, investments…)
Xi = ∑ Xij + Fi = ∑ α ij X j + Fi
j j
Sumtotal columns:
Goods & services inflows from: other sectors and factor inputs
Intermediate deliveries Final Demand F total output from class i

X11 X12 X13 F1 X1
=α11X1 =α12X2 =α13X3
X21 X22 X23 F2 X2
=
X31 X32 X33 F3 X3
Factor inputs
P1 P2 P3
Sumtotal columns: total input into class j
T1 T2 T3
Figure A14.1. Scheme of an input-output (I-O) table.
In the 1940s, Leontief introduced the assumption that the inputs are proportional
to the outputs. In other words, the production function is a simple linear relationship
between output Y and the limiting production factor K or L (Y = min[AK,BL] with
A and B constants). The basic identity in an I-O table is then:
n

Xi = Xi j + Fi (A14.3a)
j=1
in an economy with n sectors and a final demand Fi for goods and services in sector
i. It states that the output of any sector i equals the deliveries to all the other sectors
and the deliveries to final demand. The assumption of a linear production function
implies that the output Xi can be written as a linear function of the inputs Xij :
Xi j = ai j X j (A14.3b)
In words, each matrix element is divided by the sumtotal of the column elements.
The elements α ij = Xij /Xj are called the technical coefficients. For example, if sector
j is the medical service sector and sector i is the electric power production sector,
the ratio Xij /Xj = α ij indicates the fraction of every € spent on medical services that
is used to purchase electricity. Inserting equation A14.3b into equation A14.3a and
continue the substitution gives:
j=n j=n
k=n

Xi = ai j X j + Fi = ai j X jk + Fi + Fi = · · · (A14.3c)
j=1 j=1 k=1
In matrix notation, this becomes X = (I-A)−1 F, with I the identity matrix and A the
coefficient matrix. The matrix A is a concise representation of the economic structure
of a country. For more in-depth discussion, I refer to the Suggested Reading (Ayres
and Ayres 2002; Hoekstra and Van den Bergh 2002; Perman et al. 2003).
Appendix 14.4 Gross Domestic Product

Gross domestic product (GDP) and income (GDP/cap) are probably the most widely
used indicator to compare a country’s performance and its citizens’ welfare. Because
of its widespread availability, it is often used as a proxy driver. In fact, it does correlate
in one way or another with almost everything. What is it?
The GDP is the value of all the goods and services produced in one year in a
given economy as expressed in the monetary System of National Accounts (SNA).
In other words, the value of the monetary transactions per year as measured in the
official statistics. Divided by the number of people in a country, it is the (average)
income in monetary units per year per person (GDP/cap). If income transfers with
the rest of the world are included, it is gross national product (GNP). If the goods and
services for investment are substracted, one gets the net domestic product (NDP).
GDP is defined from the production point of view as the sum of all value added
(VA), which equals the payments for the primary production factors labour and cap-
ital. It equals gross domestic income (GDI), which is defined as the sum of all incomes
Appendix 14.4 Gross Domestic Product 515
(wages, salaries, entrepreneurial income and all rents). From the consumption point
of view, the GDP is the sum of consumption expenditures (including government
expenditures, gross capital formation and the transactions with abroad). If the sys-
tem is considered to be in accounting equilibrium within the given time period, the
domestic (or national) product being the sum of all expenditures equals the domestic
(or national) income consisting of all production factor rewards.
15 Outlook on Futures
15.1 Introduction
When individual human beings began to experience the first flashes of consciousness,
there must have been an emerging anxiety about death of the individual. Biology
may have dictated the individual to put the survival of the species above that of the
individual, but individual physical survival became the preoccupation that determ-
ined actions, emotions and ideas, albeit extended to the nearest members of family
and tribe. Whenever threats to survival were absent, the individual and his kin could
pursue other qualities of life such as improved shelter and clothing. It was here when
development of the individual, and with it society, started.
Survival and reproduction of the individual and his kin has for ages been the
main if not only sustainability concern for humans. Only a few individuals broadened
their horizons and interests to larger areas and longer periods – the kings and priests
in recorded history. But, as seen in Chapter 3, levels of quality of life above mere
subsistence could rarely be sustained for more than a dozen of generations and
for more than a few small elite populations. For the majority of people, individual
suffering from illness, strenuous labour, oppressive overlords or natural disasters
was never absent or far away. One response to these realities of life was military
valour and conquest – the way of the warrior. Another one was transcending the
individual self, in art, love, sacrifice, meditation and compassion – the way of artists,
philosophers and priests.
The benefits of working together for a goal that transcended individual survival
were clear from early times on (Wright 2000). Already thousands of years ago,
Krishna told Arjuna in the Bhagavad Gita: ‘The ignorant work for their own profit;
the wise work for the welfare of the world’ (Easwaran 1985). Concern for fellow
men and for society is found in many ancient works of philosophy and religion
(Armstrong 2006). They offer reflections on suffering and compassion, on good and
evil and on endurance beyond individual birth, reproduction and death. Amongst
those who held positions of power, the concerns usually extended to the tribe, the
nation or the empire and ranged from years to decades. In philosophy and religion,
the emphasis was on salvation in the hereafter or becoming (again) one with the
cosmos. These were the early forms of inclusive thinking, and therewith of concern
about sustaining quality of life for the collective and the long term.
516
15.2 Outlooks 517
In the 21st century, the world has become a crowded place, with huge and still
growing numbers of human beings and of activities and connections. The finiteness
of the planet is in sight. Inclusive thinking in space and time is an urgent necessity
and possible pathways for the world and la longue durée have to be explored. This
chapter probes deeper into what the data, models and insights in sustainability
science can tell us about the future. I introduce scenarios as combinations of stories
and models in order to be somewhat prepared for an uncertain and complex future.
I also pose a few questions that remain, even though the answers have come closer.
Can forecasts of key trends be made with any certainty? Is information a necessary
step for transformation? What is the right mix of ideas, values and actions for a
sustainability transition? And what does this imply for the sustainability science
research and education agenda? I do not aim at an overarching synthesis or final
answers because this book is itself an attempt at synthesis. I merely sketch a platform
that can guide further thought and action in this new branch of science.
15.2 Outlooks
15.2.1 Sustainable Futures: Urban, Rural, Global

Cities, or more broadly, urbanised regions, are increasingly the centerpieces of the
human world. How will a sustainable city look? Upon entering it, you become part
of a centrally coordinated world of constrained individual freedom. This is a con-
sequence of the high density of people and activities and the need for maximum
resource efficiency. Physical resource flows are optimised for their use in a circular
economy: urban gardens and ‘living machines’ recycle waste, water and most mater-
ials are reused, and the built environment is largely supplied from local, renewable
energy sources. Remaining inputs from outside the system are mostly from the land:
food, (bio)fuels, electric power. The cities accommodate the centres of knowledge
and communication, of health and education, and of the latest accomplishments in
resource-efficient building and mobility. They have turned into the largest ‘complex
adaptive system’ on Earth, continuously reinventing and rejuvenating themselves.
An apt metaphor of this manager-engineer world is the urban metro system: continu-
ous adaptations with the latest technology in an infrastructure that lasts for centuries
and is operated in seamless cooperation between city government and private enter-
prise. It is a world of intelligence and comfort, where the physical body has to follow
the prestructured lanes, but the mind and the spirit can freely wander – at least in
principle. People have accepted and internalised the trade-offs between physical and
privacy constraints on the one hand and creativity, diversity and sustainability on
the other.
The countryside, the rural regions, are different. Here, sustainable develop-
ment means use of local resources to provide for own use and in exchange with
the urban areas in the region or country. Farmers grow food and produce fuels
and electricity for rural and urban areas on the basis of nutrient (re)cycling and
local flow resources. Animal husbandry is limited and integrated in the natural
cycles. Degraded lands in arid and semi-arid regions are restored, using the theory
and the practice of ecosystem dynamics. Information and communication technol-
ogy (ICT) plays a key role in making communication and transport appropriate
for low population densities and for interaction with urban areas. It is here, also,
that your aging parents live and where you go for a holiday or a rest. The experi-
ence of nature – its silence, beauty, diversity – provides the appreciated comple-
ment to urban life. Governance is mostly a village affair: a combination of local
autonomy and bi-/multilateral agreements with urban areas in regional exchange
networks.
In this future, urban and rural, city and village live in mutual and beneficial
interaction, as was often but certainly not always the case in history. Some call it
rurbanism and consider it feasible and sustainable. Certain forms of coordination and
regulation are necessary, but quite different ways of life can be accommodated. Of
course, there are always activities that exceed regional boundaries and sovereignty.
A dense network of mining, processing and shipping raw materials and manufac-
tured products has evolved and business travel and tourism connect people to an
ever larger extent. They satisfy the need for basic inputs from and outlets for else-
where and provide adventure, exchange, knowledge or, in short, development. But
they have consequences that can make development unsustainable without more
coordination and regulation. Think of acidification, the ozone hole, biodiversity and
climate change, as well as of economic interdependence, food price volatility, income
gaps and financial crises. In practise, it means the imposition of regional and global
constraints on regional and local development with UN-led targets, standards and
allocation mechanisms. Globally operating corporations and national governments,
working together in supraregional and global institutions, have to take the lead and
act. Is this a utopian illusion? Not really, because the outlines can be seen in some
wealthy enclaves as well as in some poor places.
Instead of a rosy future, we can describe dystopian futures. Resources are
squandered, its revenues are spent on armaments and used to provide luxury items.
Corruption and crime are rampant. Institutions fail to deliver necessary health, edu-
cational, financial and legal services. Public transport and other infrastructure lags
far behind. ICT is used for repression and manipulation. The features of this future
can also be seen around the globe, in poor as well as in rich places. But there will
never be a single story for the whole world. The future world will, like the world
of today, consist of a variety of regions, some of which realise their own brand of
sustainable development and others become more or less isolated islands of decline
and misery. The bewildering richness and complexity of the world contains the seeds
of all these possible futures – but how is their potential identified and how are those
worldviews and forces that can bring about a more sustainable future identified?
One option is to do imaging and visioning as in the previous tale about sustain-
able cities and rural regions. We combine scientific models and data with stories and
fiction.1 And we ask ourselves which events, behaviours and structures can make a
particular future happen and which ones prevent it from happening or make it even
impossible. Is a story plausible and feasible? Are the models able to adequately rep-
resent the story elements? Are the data consistent and relevant? Besides plausibility
1 We usually get our stories from the media and from personal friends and colleagues. The community
level is often underrepresented and a site like The Story Garden (www.storygarden.ca/) tries to fill in
the gap. It is one of the novel ways in which civic society can communicate and build up a reservoir
of experiences and experiments.
15.2 Outlooks 519
and feasibility, there is desirability.2 If you desire a sustainable future, what is it you
desire? A good quality of life – but for whom, where and for how long (§1.2)? Are
personal and societal developments in values and consciousness to be considered?
And how is personal quality of life connected to systemic resilience and social and
ecological fairness? Perhaps you have your own answers to these questions and, at
the same time, accept the diversity in other people’s answers. Of course, your answers
are not determining history, or certainly less than those of the ones in power or those
of the majority of the people. But scenarios can make each of us more effective in
identifying and using the proper indicators and levers in the system that initiate
change in the desired direction.
15.2.2 The Scenario Approach

The scenario method combines the qualitative aspect of stories (or narratives) with
the quantitative of models. Practitioners have given various definitions, for instance:
a scenario is a combination of qualitative story-telling and quantitative model-

ling, with the purpose to construct and explore with stakeholders different possible
futures, each with their own logic.
Scenarios are supposedly a tool for (better) (strategic) decisionmaking. Common

elements of the scenario method are (de Vries 2006):
r emphasis on the construction of alternative futures in order to prepare for
divergent plausible futures;
r use of both qualitative (‘story-telling’, narrative) and quantitative (‘modelling’)
approaches and challenging existing mental models;
r a training in finding key trends, recognising prevalent myths and imagining
attitudes of key players.
It is important to know for whom scenarios are made and for which purpose. Cred-
ibility, legitimacy and creativity are core ingredients of process and product. The
identification of the driving forces: ‘what makes it going’, and of predetermined ele-
ments, in particular slow changing variables, are at the core of a scenario construction
process (Schwartz 1995). They provide the structure or logic of a scenario and give
an idea of critical uncertainties. The scenario method, it is claimed, opens people
to multiple perspectives on the world and offers a complement or even an altern-
ative to the conventional languages of business, government and science in dealing
with complex and ill-structured questions (Duke and Geurts 2004). Scenarios are a
stepping stone to strategy, which is the art of deliberately recognising major trends,
establishing one’s own course of action and translating this into practical plans. Part
of scenario construction is the process of visioning. ‘Visioning means imagining,
at first generally and then with increasing specificity, what you really want . . . not
what you have learnt to be willing to settle for. Visioning means taking off all the
constraints of assumed “feasibility”’ (Meadows et al. 1991). In this sense, scenarios
express the ethos of their times.
2 The possible and the hopefully desirable overlap. To make things more complicated, they also
interact.
Box 15.1. The future and scenarios about the future. The word future is from
Latin futurus meaning that what is going to be, yet to be. The French equivalent
avenir is clear: that which is to come, à venir. In German and Dutch, the words
Zukunft and toekomst express the same: that which is to come. Much has been
said about ‘that which is to come or to be’. A few aphorisms:
Prediction is very difficult, especially about the future.
– Niels Bohr
The trouble with our times is that the future is not what it used to be.
– Paul Valery
The function of science fiction is not always to predict the future, but sometimes
to prevent it.
– Frank Herbert
Our view of the future affects the present as surely as do our impressions of the
past.
– Willis Harman
There is reason for skepticism about predictions and forecasts. In a large-scale

experiment on future predictions, it was found that ‘experts who did particularly
badly were not comfortable with complexity and uncertainty . . . and also more
confident than others that their predictions were accurate . . . Experts who did
better than average . . . were comfortable seeing the world as complex and uncer-
tain – so comfortable that they tended to doubt the ability of anyone to predict
the future’ (Gardner 2010, 26–27).
Governing elites have always been interested in anticipating future events
and, often, priests were involved in order to let the gods participate in judging,
legitimising and rationalising. It provided strategy as well as legitimation against
competing groups with opposing views of what the present is and the future
could or should be. Thomas More offered his Utopia as a visionary critique. Karl
Marx offered a rationale for the demise of capitalism. Jules Verne expanded
technical possibilities far beyond the known options and religious leaders have
promised mixtures of catastrophe and salvation. Prevailing ideologies and their
utopian challengers, variously considered rebels, visionaries or prophets, are in
essence radicalised worldviews. Currently, there are hundreds of individuals and
organisations who offer their view of the future, which ranges from alerts and
warnings to technological paradise and fundamentalist doom.3 Some are prag-
matic, ‘business-as-usual’ trend extrapolations. Others are engaging, socially or
technically radical utopias (Achterhuis 1998).
The scenario method has risen to prominence because old centres of author-
ity were increasingly challenged in the 1960s and 1970s, and the control paradigm
had met its limitations. New and powerful actors appeared on the scene: activist
3 Warnings of overshoot in views or actions can often be identified from the titles: Limits to Growth
(Meadows et al. 1971), Social Limits to Growth (Hirsch 1977), Limits of Organization (Arrow 1974),
Limits to competition (Group of Lisbon 1995) and Limits to Certainty (Giarini and Stahel 1993) are
examples.
15.2 Outlooks 521
Figure 15.1. Framework for a sustainable development assessment (de Vries and Petersen
2009). See text for explanation.
scientists, innovative entrepreneurs and committed citizen groups. A new legitimacy

with participation of stakeholders was needed. The scenario method, with its explicit
consideration of uncertainties, multiple perspectives and stakeholders, addressed at
least partly the new requirements of a postmodern era. Not everyone embraces it.
Engineers may not subscribe to the qualitative elements and social scientists are sus-
picious of quantitative models. CEOs and politicians may also show signs of dislike,
preferring command and control over participation and pluralism.
15.2.3 Sustainable Development in a Scenario Frame

Sustainable development faces macro-problems that are high in aggregate complex-
ity and low on consensus in worldviews (Figures 8.5 and 8.6). Exploring future devel-
opments and constructing interesting scenarios for sustainable development must,
therefore, consider both ends and means, both subjective and objective experiences
and knowledge. Figure 15.1 sketches a framework for such a broad sustainability
assessment. The lower part represents resources and technologies, both important
determinants of societal change. It includes the ‘autonomous’ dynamics of processes
such as resource depletion, pollution and degradation of the environment, and tech-
nological developments. The knowledge about it is rather strong, and its refutation
erodes the very foundation of rational discourse. The upper part represents the
individual and collective sense and view of well-being that has historically also been
an important driver of change. In this domain, subjective experience matters more
than objective fact. The issues are complex, the solutions are controversial – and
knowledge about them is rather weak.
The upper and lower parts are connected through capabilities that arise from
what resources and technologies can offer, and functionings that correspond to the
behaviour that satisfies our aspirations for ‘the good life’. The middle part consti-
tutes the essence of a person’s worldview. Explicit acknowledgement of worldview
pluralism is the basis for legitimacy in a modern democracy and prevents us from
falling in the trap of reducing the problem to a single solution for an – illusory –
single constituency. Sustainability is lost when one worldview starts to dominate and
society moves towards extremes (§6.5).
Scenarios are positioned in the upper right of Figure 15.1. They combine the
social and cultural outlook with the resource and technology potential. Scenarios are,
in the present context, constructed in order to examine possibilities and strategies
for a (more) sustainability development. This can be done from different vantage
points. One approach is to position oneself in the centre of the worldview space
and to identify values and beliefs, hopes and aspirations (utopia) and fears and
concerns (dystopia) in each of the different worldviews. It includes an evaluation of
capabilities, resources and technologies in the form of financial, economic, social and
media power and control. A scenario then becomes a policy program that is robust
and resilient in the face of change, including sudden shifts in the political landscape
and surprise events in the natural system (MNP 2004). Such an integral and non-
ideological approach is what governments (should) do, in the public interest, in an
attempt to identify and implement robust policies.
A scenario can also be constructed to explore possible futures with respect to
the pros and cons for the stakeholders and their worldview. It serves the strategic
objectives of stakeholders, such as providing warning signals and leverage points.
Given the aim to strengthen one’s worldview, important questions are: which values
to promote, which ideas to communicate, which facts to mention and which risks
and opportunities to spell out? If the particular worldview has support from power-
ful groups, it can effectively be marketed as a desirable view of the future, with
other scenarios and associated worldviews as potential threats. In this way, scenarios
become politicised and get ideological content. There are many examples of such
partisan scenarios supported by government institutions, corporations or NGOs, for
instance, in the areas of energy and food.
A third option is to start from policy goals and targets that have already been
agreed upon, such as the Millennium Development Goals or the European Union’s
2◦ C target for climate change, and investigate various model-supported storylines
that might meet the goals and targets (PBL 2009). The worldview pluralism serves
as a heuristic that can create new solutions or identify ignored trade-offs. It also
permits an assessment of policy options in futures that are dominated by worldviews
different from the existing one. This can, in combination with backcasting, enhance
the likelihood of successful and robust policies.
15.3 Scenarios for a Sustainable World
15.3.1 Four Stories

Numerous scenarios have been made in the last decades. One of those is the Special
Report on Emission Scenarios (SRES), that was written on request of the Intergov-
ernmental Panel on Climate Change (IPCC) in order to provide possible greenhouse
gas emission trajectories (Nakicenovic et al. 2000). Two dimensions are proposed,
along which future developments might evolve: globalisation versus regionalisation
Box 15.2. Decision making in politics. A textbook on sustainable development

should also address public decision making and politics. Politics can be described
as the area in society where issues transcending mere individual needs and activ-
ities are debated and decided and where disagreement is legitimate. It has its
own rules and rites. Miklos Persanyi, former Environment Minister of Hungary,
shared with me and others some experiences about life in politics. A basic rule
for politicians is:
If there is no must to decide, it is a must not to decide.
When you are in the science-policy interface dealing with politicians, obey the
following rules:
r draw/keep politicians’ attention;
r speak their language: be stupidly simple/primitive, talk about money, talk about
benefits for constituency/business;
r don’t use words like future, climate . . . instead, talk about costs, floods;
r become a media star, marry his/her daughter, or go golfing.
It is tempting to reproach politicians for their often narrow-minded, myopic, self-

interested behaviour. But they probably represent what a postmodern, media-
hyped society deserves.
On a more serious note, the notion of bounded rationality was introduced by
the sociologist Simon to indicate that people usually act on the basis of a few,
rather simple rules. Cognitive sciences are bringing new insights into how human
process information. Individual experiences are filtered by several subsequent
layers (Morecroft and Sterman 1992):
r tradition, culture and the like;
r organisational and geographical structure;
r information, management and communication systems;
r operating goals, rewards and incentives; and finally
r people’s cognitive limitations.
Illustrative examples of the first three layers in a sustainable development context

are the role of tradition in the introduction of energy-efficient and solar-based
cooking stoves in rural India; the role of the state in explaining why centralised
nuclear power thrives in some cultures and is resisted in others; and the role of
(perverse) incentives in an oil company that rewards its executives on the basis
of the oil reserves at the end of the year. The rationality of decision making is
perhaps even farther away from the real world than we imagined in our criticism
of Homo economicus (§10.4).
and government versus market. For each of the four resulting quadrants, a storyline
or narrative is developed that describes a plausible – but not necessarily probable –
path of key variables such as population, economic activity and energy use. The
keywords of the SRES storylines are summarised in Table 15.1, together with the
scenario names (A1, B2, B1, A2). There is resemblance between the SRES scen-
ario dimensions and those in the worldview framework presented in Figure 6.3.4
4 Globalisation vs. regionalisation coincides with the horizontal axis of big world vs. small world.
Government vs. market represents the difference between a less vs. more outspoken orientation on
Table 15.1. Keywords for the four scenario quadrants and associated features, in the worldview
framework of Figure 6.3. Besides the names B1, A1, A2 and B2, many other names have been proposed
for the scenarios. One set is indicated in bold
B1 collective and immaterial B2 individual and immaterial

(Absolute Idealism) (Subjective Idealism)
Government/church authority offers universal truth Truth is personal
Tech and organisation in the service of the soul Low/appropriate tech and human scale
Hierarchist tendency in collectivist setting Egalitarian in individualist setting
Welfare economics Buddhist economics
Our Common Future Small is beautiful
A1 collective and material A2 individual and material
(Objective Materialism/Modernism) (Subjective Materialism/Postmodernism)
Science is universal truth Truth is what can be sensed
Society high-tech and well-managed Tech and organisation for material wellness
Hierarchist tendency in collectivist setting Individualist tendency
State/corporate economics Market economics
The End of History Clash of Civilisations
The scenario logic is derived from hypotheses or metamodels about how the world
works. For instance, it was in SRES hypothesised that more ‘free market’ forces of
privatisation and deregulation lead to higher economic growth; that higher income
causes a more rapid stabilisation of population growth; and that protectionism
hinders trade and, therefore, economic growth. This led to consistent population
and economic growth paths, that were introduced into the energy and land use mod-
els. The latter were then run with assumptions consistent with the storylines. The
outcomes of the thus generated sets of scenarios were presented for world regions
until 2100 in the form of energy use and greenhouse gas emission pathways.5
How does the world evolve according to these scenarios and stories? I give a
brief caricature description. In the modernist A1 world, material wealth and high-
tech business are essential for a good quality of life. Everyone with specialised skills
and a competitive and risk-taking attitude can share in it. Market-driven economic
efficiency, minimal government regulation and open trade are preconditions. The
benefits accrue to a small part of the population, but wealth and novel goods and
services will over time ‘trickle down’ to the poor. Post-war performance of the United
States is the most visible proof of the correctness of this ideology.6 The archetypical
A1 citizen is convinced that society will become stagnant and backward if these
values and qualities are lacking or obstructed. Indeed, he represents the successful
transition from traditional to secular/rational values, that has been observed in many
places during the 21st century (§4.3)
This worldview is not shared by people who value small-scale enterprises and
cooperatives, social and cultural traditions, community, personal growth, spirituality
material welfare along the vertical axis. For a discussion of these scenarios and the proposed axes,
names, interpretations and shortcomings, I refer to Nakicenovic et al. (2000), de Vries (2006), Riahi
et al. (2007) and de Vries and Petersen (2009).
5 The scenarios are caricatures in the sense that they presume a particular and rather extreme world-
view to prevail over a long period of time.
6 The emergence of China as an industrial nation with a command-and-control form of capitalism is
interpreted by some as a proof that democracy may not be needed or even be harmful, whereas
others see it as a temporary phase until postindustrial values will break through also in China.
and nature. They manifest their worldview by protecting the small and local world in
the form of cooperation, solidarity and forms of enclosure. In this B2 world, citizens
find their identity in cherishing what is local, be it their vernacular, the village church
or a nearby forest. The focus is on local needs and livelihoods, to which technology
and governance should be geared. Protectionism is seen as a survival option for their
socio-economic as well as cultural way of life. They adhere to a civic society ideal,
that often clashes with the logic of rationalist modernity and global capitalism.
The B2 citizens cherish autonomy, but they cannot exclude the larger world with
its lures and necessities. Much to their discomfort, they are confronted with alarm-
ing stories about the large, evil world out there: dwindling fish stocks, water and
air pollution, the threat of climate change, destruction of local jobs, immigrants and
refugees and so on. Blaming the others is not helpful. Many of them, may, therefore,
adhere to the B1 vision, of a world in which international cooperation and solidarity
are needed in order to solve the large-scale and long-term social and environmental
problems the world is facing. Persons adhering to the B1 vision support universal
values and aspirations regarding global solidarity and peace, human rights and sus-
tainable management of the global commons. Theirs is a world of global governance
and value-inspired science. The Millennium Development Goals (MDGs) are an
expression of this worldview.
But many people currently dismiss the B1 world as a failed ideal and a bur-
eaucratic dystopia, pointing at the political quarrels, fraud and mismanagement in
the UN organisation and in development aid. They prefer the ‘realism’ of the post-
modern A2 world. They are joined by those who are disappointed by the promises
and technocracy of the A1 world and by those who have anyway no interest in the
societal and immaterial aspects of life. Their future brings a world of individualism
and hedonism, in which citizens resort to a mix of consumerism, opportunism and
fatalism. It offers the pleasures of the material and body culture and opportunities
for local politicians and entrepreneurs, but injustice and lack of social coherence
take their toll in the form of fear, disorder and crime. Politically, it manifests itself in
protectionism, nationalism and clientelism. Bilateralism characterises trade, govern-
ments emphasise security and military strength. Prominent beliefs are that markets
and money run the show and that governments waste tax money and cannot be
trusted. In many places, there is a continuous crisis of legitimacy of political and
financial elites, unless or until a military elite takes over.
As this brief description shows, the scenario method can be used to tell different
stories about the future, with different values and beliefs dominating society. For
the 21st century, it can be surmised that the trend towards the postmodern A2 world
continues for some time although in parts of the world people rather experience A1
modernism. It may spell an era of intensified conflict between nations about scarce
resources, environmental damages, novel (arms) technologies and cultural identities.
When the need for collective action becomes sufficiently clear and urgent, there will
be attempts at coalitions or even a world government to deal more effectively with
the global problems that most seriously threaten the prospects for sustainable devel-
opment (B1; Schrijver 2010). Meanwhile and along the edges, small-scale groups
and initiatives that stress the spiritual, the community and the natural will grow in
strength (B2; Hedlund-de Witt 2011). Can models be of any use in this exploration
of world futures?
15.3.2 Growth within Limits . . .

In the last decades, the number of future scenarios have grown exponentially. For
key variables such as population, economic activity, food supply, energy use and
other system stock and flow variables, there are new forecasts every year. The scale is
usually regional and the time horizon between ten and forty years ahead. In previous
chapters, some possible future trends in demography, biodiversity, agriculture and
energy are given However, the complexities and uncertainties of our future world
are such that most scenarios are rather short-lived. They also increasingly reflect,
with or without explicit storylines, a particular worldview or ideology. Therefore,
this book refrains from giving more scenario outcomes. Most scenarios are made
by professionals in institutions such as the UNFPA for population, the World Bank
and IMF for economic activity, the FAO for food, the IEA and the WEC for energy
and so on. Some institutions make integrated scenario analyses in a policy setting,
for instance the World Development Reports by the World Bank and the Global
Environmental Outlooks (GEO) by UNEP. Corporations and NGOs also make
and sometimes publish scenarios, for instance, Shell’s strategic planning scenarios
or the green growth scenarios of the European Climate Forum. Institutions such as
IIASA and PBL have developed Integrated Assessment Models (IAM) in order to
incorporate the latest scientific insights in the analyses and scenarios. I refer to the
Useful Websites for more information.
One thread of model-based scenario explorations is the Limits to Growth report
(1972) and its subsequent appraisals and updates: Beyond the Limits (1991) and
The 30-Year Update (Meadows et al. 2004; Meadows 2006). It is based on a system
dynamics world model and can be considered the first IAM. The updated scenarios
suggest, that the world is still following a path towards overshoot that results from the
connected processes of resource depletion, soil degradation and pollution (Turner
2008). A transition to a sustainable society is still possible, but markets and techno-
logy alone will not make it happen. The underlying worldview manifests the anxiety
about the A world and its technological and market optimism, and the necessary
change towards a B world. The model outcomes are at a high level of aggregation
(world), which gives them an inherent bias towards universalism and a tendency to
overlook the real-world diversity and contingency and its surprises.
Inspired by the Limits to Growth report, a number of other long-term models
have been constructed to investigate the world system past and future. One example
is the TARGETS model, an IAM that simulates the most important subsystems
of the global system and their interactions (Rotmans and de Vries 1997; de Vries
2001). The perspectives of Cultural Theory were used to incorporate the uncer-
tainties and controversies (§6.3). By distinguishing between the dominant beliefs in
society on the one hand and the, as yet unknown, world system on the other, uto-
pias and dystopias were constructed. If the optimism of the modernist worldview is
justified and its supporters are in power (the individualist utopia), the increasing pres-
sure from exponentially growing population and gross domestic product (GDP) can
be managed for the next hundred years because natural ecosystems turn out to be
resilient, resources abundant and new technologies efficient. The future will then be
high-tech and highly managed. If the optimism about the robustness and resilience
of our life support system turns out to be unjustified, a more dystopian future can
unfold in which tremendous suffering and damage – notably from climate change – is
to be expected. The future then follows a Limits to Growth overshoot-and-collapse
path. This is prevented if a more egalitarian perspective gets the upper hand, lead-
ing to deep reductions in energy and material use and corresponding reductions in
environmental damage and destructive feedbacks.
A mainstream scenario is one in which the hierarchist perspective of official
government institutions prevails. It is a world of ‘muddling through’: A series of
incremental policy measures can avert major catastrophes, but large long-term envir-
onmental risks are not properly addressed. This is caused by waiting too long for
more and more reliable knowledge and getting bogged down in bureaucratic inef-
fectiveness and political deadlocks. The only consolation in the face of repeated
and spreading disasters is perhaps that the emergency actions are well organised.
The simulation experiments carried out with the TARGETS model represent one
approach to manage the large uncertainties in basic, quantifiable variables such as
the size of population and economic activity, life expectancy, food intake and water
and energy availability.
A recent scenario study for the world is the Growing within Limits report (PBL
2009). Two scenarios, Trend and Challenge, are investigated with a focus on green-
house gas induced climate change and land use/biodiversity. The Trend scenario is
the dominant ‘business-as-usual’ view of the future amongst government and cor-
porate officials. It is a mixture of expectation and desire, largely following an A1
storyline. Its outline is sketched in Figure 15.2.7 The graphs are indicative.
There is probably least uncertainty about population size. Most demograph-
ers expect a stabilising trend towards a population between 8 billion and 10 billion
individuals for the period until 2050. Underneath are the demographic and epidemi-
ological transition (Figure 10.6). The assumption is that there will be no massive
catastrophes that could make the population decline. The band width of 2 billion
people results from uncertainties in trends in income, food, education, availability
of medical services and biodiversity. Food supply is also projected with rather large
certainty, because presumedly it is well defined in relation to population and can be
met in all but the worst places and periods. Biofuel plantations will not greatly affect
the picture.
More uncertainty exists about the rate and nature of GDP and income growth.
It is a dominant variable because it drives most other variables – at least in the
models. The Trend scenario assumes more than a doubling of average income in
the world between 2010 and 2050, with an uncertainty range that suggests it might
be considerably less. Such an income growth path is an average across populations
and the kind of economic activities and their global distribution is largely based
on extrapolations and the assumption of income convergence. Consequently, the
forward projections of resource use and environmental load have similar or higher
uncertainty bands (Van Vuuren et al. 2008). For instance, the use of energy is
expected to increase with a factor 3 ± 0.5. For more details, I refer to the Suggested
Reading and Useful Websites.
7 Such scenarios used to be indicated with business-as-usual, but the feeling behind the term has
disappeared.
528
Population Income
Global World Product (GWP)

In 2050: 2 à 3 times value in 2010
Energy use
Agro-food system In 2100: 2,5 à 3,5 Resource systems
times value in
2010
pasture
Eco- and environment systems
cropland
Figure 15.2. Possible pathways of important world system variables in the 21st century (PBL 2009). The solid lines are according
to the Trend scenario; the coloured bands are the uncertainty margins from the literature. Solid arrows indicate fairly well-known,
causal relationships. Dashed arrows indicate less-known feedback relationships.
Box 15.3. Economic growth projections. Most official organisations still use
a rather straightforward neoclassical growth model to project future economic
output. An example of a long-term model-based projection of world economic
growth is the OECD report ECO/WKP (2009) by Duval and De la Maisonneuve.
Its projection of economic output Y is ‘based on a standard aggregate Cobb-
Douglas production function with physical capital, human capital, and labour as
production factors and labour-augmenting technological progress, and [assumes]
that the production function is invariant both across countries and over time’.
Output per capita Y/Pop can be decomposed as follows:
1−αα
Yt Kt Lt
= At ht
Popt Yt Popt
where Kt /Yt , At , ht , and Lt /Popt denote the capital/output ratio, TFP, human cap-
ital per worker and the employment rate (defined here as the ratio of employmed
and total population), respectively, and α is the capital share in aggregate output.
From the graphs in Figure 15.3, it is seen that this official forward projection indic-
ates, like most others, exponential growth and regional convergence in economic
activity measured as GDP and income (GDP/cap) until 2050 in the world.
15.3.3 . . . But Is It Sustainable?

Accommodating up to 9 billion people at twice the average income, as presumed
in most official scenarios, puts large strains of the life support system. What are
the implications for sustainability if it becomes reality? Combining the insights of
the preceding chapters with the A1 worldview yields the following, plausible story
of the Trend scenario. Development is guided by maximising productivity and effi-
ciency through competition, innovation and abolition of trade barriers in the search
for high returns on capital savings and decreasing consumer prices. Multinational
corporations are the prime actors in an ongoing process of globalisation, whereas
governments, including international organisations, focus on safeguarding a level
playing field for the corporations and on securing access to resources and contain
environmental side effects (Dicken 2009; Schrijver 2010). Much creativity, organisa-
tional skill and willpower is spent on overcoming resource scarcity and environmental
degradation – the knowledge and money are available. The high income growth goes
together with increasing energy use, but depletion and environmental impacts are
met by resource efficiency, recycling, renewable energy and pollution abatement
in response to rising world market prices. The postindustrial transition sketched in
§14.1 succeeds. There are still uncertainties, especially around potential bifurcations.
Will it be solar or nuclear or coal energy that wins the game? Will recycling go as
deep as cradle-to-cradle? Are GMOs boosting a second green revolution? Whatever
the answers, it will be a future of hard work, and in which we have to work hard to
cope with the consequences of our hard work.
In the Trend scenario, bottom-up market processes in combination with corpo-
rate efficiency and organisation are expected to deal incrementally with upcoming
problems and events unfold without a comprehensive and pro-active assessment to
530
Economic income growth projections Regions Economic growth projections World (normalized)
(normalized) (OECD ECO/WK(2009)4)
(OECD ECO/WK(2009)4) 600
1500
2005 PPPs US$ (2000=100)

2005 PPPs US$ (2000=100)
1200 USA
GDP/worker
EU27+EFTA
GDP/cap
900 China
GDP
India 300
World
600
300
0 0
2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050
year year
Figure 15.3. One projection of economic output until 2050 (OECD; /www.olis.oecd.org/olis/2009doc.nsf/LinkTo/NT00000AE2/$FILE/JT03260306
.PDF.
make the sustainability transition happen. There is a decline in long-term strategic

planning and R&D investments in both corporate business and governments. Its
flavour can be tasted in particular mixes of planning ideals and market mechanisms
such as in the April 1990 issue of Scientific American: Managing Planet Earth and in
the optimal control solutions to greenhouse gas emissions in Managing the Global
Commons (Nordhaus 1994).8 A precondition for its realisation is that the ruling
elites must be sufficiently convinced of the seriousness of scarcity and degradation
in order to act and sufficiently convincing to induce the investments and implement
the policies that are needed.
The Trend future is exposed to two major risks, one external and one internal.
The external one is that the life support system – nature – turns out to be more fragile
than expected (Tables 6.2 and 6.3). The planetary boundaries are real indeed. A
series of natural catastrophes disrupts large parts of the global system; fragmentation,
egoism and protectionism hamper coordinated responses; and in many places, people
start to rely increasingly on local resources and institutions. The worst crises are
initialy relieved with large-scale aid, but people get weary of it and it often comes
too little and too late. In the terminology of the narratives, the world moves towards
some mixture of fragmented, regionalised A2/B2 worlds, because the forces behind
a B1 world have failed in a predominantly A1 world.
The second risk is internal: The breakdown comes from the inside, because
income inequality destabilises the necessary social contracts and because govern-
ments fail to organise the public goods needed for income growth and quality of
life. In quite a few places, social revolts disrupt the growth engine, institutions fall
apart and disorder spreads in a densely connected world. It is an erosion process that
manifests itself in widespread distrust in government and retraction into a private
consumer world. It is a move towards an A2 world, because the promise of rising
prosperity for the masses is not fulfilled for the poor or no longer fulfilling for the
rich, and partly driven by anticipated shortages and scarcities. Reputable analysts
interpret the 2008–2011 financial crises as the onset of such a storyline. Both risks
are exacerbated because of pernicious feedback loops. For those who can see, the
signs are there. It is phrased in a penetrating way by Yeats in his post–World War I
poem The Second Coming:
Turning and turning in the widening gyre

The falcon cannot hear the falconer;
Things fall apart; the centre cannot hold;
Mere anarchy is loosed upon the world,
The blood-dimmed tide is loosed, and everywhere
The ceremony of innocence is drowned;
The best lack all conviction, while the worst
Are full of passionate intensity.
It also shows up in the various names of recently constructed scenarios: Hyperindi-

vidualism, Fortress World, Tribal Society, Battlefield (de Vries 2006). Some argue
8 Options to rescue the planet by large-scale geo-engineering reflect an even more tense marriage
between the aim of large-scale engineering to ‘save humanity’ and a rejection of planning in favour
of ‘the market’ and ‘bottom-up’ self-organisation.
that a ‘great disruption’ is inevitable and that it will bring the necessary awakening
and a subsequent massive switch to more durable ways of life (Gilding 2010). Hope-
fully, mankind will accomplish sustainability, fairness and dignity for itself without
the suffering and perishment of hundreds of millions of people.
The A1 world of the Trend scenario and its natural devolution into an A2 world
is the wrong direction for sustainable development. It carries substantial environ-
mental and social risks for the long term and confronts the world with some difficult
trade-offs. The quantification of the environmental impacts suggest severe climate
impacts long before the end of the century. Greenhouse gas emissions have to be
halved at least, and before 2060, in order to reach the 2◦ C temperature target. In
an alternative Challenge scenario, it is shown that this is technically possible with
drastic improvements in energy efficiency and a deep penetration of non-carbon
energy sources. Large-scale carbon sequestration and storage (CSS) is needed in
the transition period. It implies massive and fast development and transfer of new
technologies and adaptation of users and institutions at all scale levels. In storyline
terms, it is only feasible in a B1-like narrative. The Challenge scenario is one of the
hopeful prospects, but many have begun to doubt its feasibility.
A second and different narrative sees a reconnection to the immaterial and
communal aspects of life as the essence. The emphasis is on personal growth and
self-awareness, as a source of inner conviction and of behavioural change and in
natural association with sustainable livelihood precepts. People experiment with
sustainable communities in a variety of ways, with their own local resources and
their own values and understandings. If such enclaves become mainstream, the
B2 world with scenario names such as Local Stewardship, Voluntary Simplicity
and Let Hundred Flowers Blossom gets a chance. Some have envisioned it as the
transindustrial society (Harman 1976). Others deride it as utopian because it is
ineffective in the face of global threats and the absence of global government. But,
as Paul Tillich once remarked, ‘Mensch sein heist Utopien haben’.
As discussed in Chapter 6, a genuine sustainable development should be built
on the balanced manifestation of all four worldviews: the integral worldview. In
practise, this means that we have to reorient existing values and beliefs in each of the
four quadrants in such a way that they contribute in their characteristic way to the
transition towards sustainability. Table 15.2 lists suggestions for reorientation. It is a
tentative list, in an attempt to include all the local and global and material and imma-
terial initiatives that are taken every day all over the world. Some of these concern
local habits and practises, others take by their very nature place in a more global
arena of ideas and institutions. Yet others can only be successful in the small and
idealistic worlds of committed individuals, from where some in retrospect turn out to
be the seeds of large and strong trees. Some of them are most attractive to the young
and adventurous, others will appeal to the elderly and prudent, and many will be
rooted in the rationality and pragmatism of the sciences. Together, they provide the
impetus for humanity to explore its wisdom, intelligence, empathy and sensuosity –
the four elements fire, air, water and earth.
For each of us, as individuals and inhabitants of a finite planet, there is an
invitation to contribute in various directions. Practise inclusive thinking and med-
itate upon the wisdom and compassion of enlightened ancestors. Confide in and
15.4 An Agenda for Sustainability Science 533
Table 15.2. Elements of sustainable development strategies within the four worldviews
B1 collective and immaterial B2 individual and immaterial

(Absolute Idealism) (Subjective Idealism)
Engage religious leaders and ethics Personal growth/consciousness
Address values of transcendence and stewardship Experiment with autarchy, community
Principles and institutions for Global Commons Stimulate arts, nature education, slowfood
Strengthen initiatives for fair trade and aid Permaculture/organic farming for local markets
Nature: protect biodiversity hotspots Experiment with forms of basic income
Design and implement subsidiarity principles Stimulate reuse and secondhand markets
Revitalise science and society reflection Distributed ICT-supported energy and transport
Experiment with public-private arrangements Transparency and integrity in product information
Cyberspace as common/public space Local autonomy in taxes, social security
Ecological restoration of degraded lands Nature: green city design
Envision and expand sustainable infrastructure Stimulate consumer labels and certificates
Participatory discourse on innovations ‘Greening’ banks and enterprises
Stimulate and direct long-term R&D Strengthen small-medium scale enterprise/skills
Regulation of (global) financial and tax systems Merge sustainability, cost and comfort
Engage business in transition technologies From junk to health food
Corporate Social Responsibility Address security and identity concerns
A1 collective and material A2 individual and material
(Objective Materialism/Modernism) (Subjective Materialism/Postmodernism)
strengthen institutions that enforce global standards, regulations and laws to man-
age the commons. Contribute to the knowledge and techniques that help reduce
resource inefficiency and fulfill the needs of the poor. Celebrate the body and its
expressions, yet do not identify with it.
15.4 An Agenda for Sustainability Science

Sustainability science has a more narrow scope than sustainable development. Nev-
ertheless, the list in Table 15.2 is also indicative of the topics to be researched
and taught in sustainability science. Most of these are introduced, explicitly and
implicitly, in the previous chapters. This book ends with a brief discussion of a
research agenda, defining science as the aspiration to produce, individually and
collectively, knowledge of the highest possible contextual validity (§8.3). It has its
stronghold in the lower left of the quadrants in Table 15.1, but it has an open mind
towards the other quadrants.
The natural and engineering sciences offer undoubtedly enormous prospects
for contributions to a more sustainable future. It is their responsibility because they
generate not only solutions but also problems. At present, their advances are largely
driven by the search for profits and the desire for innovative consumer goods. Over
time, this can lead to technologies that promote sustainability, as historical develop-
ments testify. However, the direction of science and technology must more intensely
be steered in accordance with societal priorities. Here, sustainability science can give
context and directions. Besides, there are sometimes real breakthroughs that may
invalidate much of ongoing deliberations. What about genetic engineering – are we
ethically and psychologically ready for it? What if man learns to manipulate nuclear
fusion reactions – would democracy survive? What would be the consequence of
rolling out geo-engineering, of deep control over individual mind and consciousness
or of a technology to manipulate gravitational force? In these domains, sustainability
science can offer critical reflection and initiate participatory technology assessments
in the science and society tradition.
We live in physical space and the sciences of ecology and geography contribute
in manifold ways to our understanding and organisation of natural and economic,
social and cultural space. They are the material linchpin between our quality-of-life
aspirations on the one hand and the resources to fulfill them on the other – in other
words: between the subjective ends and the objective means. Our knowledge of the
Earth is rapidly increasing in the three dimensions of space, time and structure. The
tools of complex system science are disclosing new and fascinating panoramas that
already play a large role in sustainable development–oriented design and imple-
mentation processes. It provides the necessary system perspective on the efficient,
reliable and affordable supply and use of water, energy and transport that are at
the core of sustainable cities. It can guide a more sustainable use of renewable
resources, such as soils, fish, forests, and of nonrenewable resources, such as fossil
fuels, phosphates and scarce minerals.
We live in mental and social space. Our interaction with the physical world
and with other living beings is mediated through the senses, the emotions and the
mind and (self-)awareness. We have created and learned habits, customs, languages,
ideas, calculus that frame our actions as individual-in-collective. Efforts in the social
sciences: economics, sociology, psychology and others, aim at the construction of
an image of man that explains and, if possible, predicts in probabilistic terms the
behaviour of the members of the species Homo sapiens. Insights from biology and
demography are an essential background. These are areas where contributions to
sustainability science are dearly needed. It represents another link between ends and
means.
On the agenda is the refinement of the image of man in economic science – the
Homo economicus. We should intensify the experiments and simulations in behav-
ioural economics in order to understand the decision of individuals as consumers
and producers. What determines the balance between egoism and altruism, between
competition and cooperation in individuals? What is the position of economic beha-
viour in the broader setting of social rules, ethical considerations and inner con-
sciousness? Other important research areas are the social construction of needs in
relation to innovation dynamics, and the necessary and sufficient rules to manage
common pool resources sustainably. Complex system science offers new vistas here,
too. Novel tools and methods for involving stakeholders and expert knowledge, such
as policy exercises and simulation games, are a helpful and timely development. If
we find answers, we are better prepared for the inevitable changes that will come and
the creativity that is needed. And for the possibly largest challenge of all: to sustain
the balance between the freedom and dignity of the individual and the solidarity and
brotherhood that is a necessity in a world of more than 7 billion human beings that
long for the good life.
Sustainability science just started, it seems.
Since 1900, a great dream has faded. To many in the West, their civilisation appears
to have gone wrong. Much that is unique about it has seemed to turnout to be
weakness, or worse . . . There seems to be little left for the educated to believe in.
The intellectual hegemony of western science has also been buttressed by its
sheer size and scale . . . [and] explains why science is now a major religion, per-
haps the religion, of our civilisation. . . . Of course, there are suspicions of the new
priesthood.
– Roberts, The Triumph of the West, 1985
Decadence is the subordination of the whole to the parts.

– Oscar Wilde
Whoever in this world overcomes his selfish cravings, his sorrows fall away from him,
like drops of water from a lotus flower.
– Dhammapada
He who knows he has enough is rich.

– Tao Te Ching
In the lower knowledge doubt and scepticism have their temporary uses; in the
higher they are stumbling blocks: for there the whole secret is not the balancing of
truth and error, but a constantly progressing of revealed truth.
– Bhagavad Gita, Commentary in the Words of Sri Aurobindo
The candles are different, but the light is the same.

– Anonymous
SUGGESTED READING
A journalistic account of predictions of the future in history and present.

Gardner, D. Future Babble: Why Expert Predictions Are Next to Worthless, and You Can Do
Better. New York: Dutton, 2010.
One of the books that presents the scenario method during its ascendency in the 1990s.
Heijden, K. van der. Scenarios – The Art of Strategic Conversation. New York: Wiley & Sons,
1996.
This report describes the SRES methodology and outcomes and can be downloaded from
www.ipcc.ch.
Nakicenovic, N., and R. Swart (Eds.) The IPCC Special Report on Emissions Scenarios
(SRES). Cambridge: Cambridge University Press, 2000.
An evaluation and update of the Limits to Growth (1971) report on the basis of the trends
between 1970 and 2000.
Meadows, D. L., J. Randers, and D. H. Meadows. Limits to Growth – The 30-Year Update.
Vermont: Chelsea Green Publishing Company, 2004.
A model-based investigation of two scenarios in order to find out about the resource and
environment constraints in a world with population and economic growth. It can be down-
loaded at www.pbl.nl/en/publications/2009/Growing-within-limits.-A-report-to-the-Global-
Assembly-2009-of-the-Club-of-Rome.
PBL. Growing within Limits. Bilthoven: Netherlands Environmental Assessment Agency,
2009.
USEFUL WEBSITES
r www.gaia.org/gaia/ecovillage/ is a site with explanation and network of ecovillages.
r www.iiasa.ac.at/ and themasites.pbl.nl/en/themasites/image are the sites with descriptions
of the models at IIASA and PBL used for scenario construction.
r econ.worldbank.org/ is the site where one can find the World Development Report oper-
ated by the World Bank.
r www.agri-outlook.org/ is the site of the Agricultural Outlook operated by the FAO and
OECD.
r www.worldenergyoutlook.org/ is the site of the World Energy Outlook (WEO) operated
by the IEA.
r www.unep.org/geo/ is the site of the Global Environmental Outlook (GEO) operated by
the UNEP.
r www.shell.com/home/content/aboutshell/our strategy/shell global scenarios/ is the site
with scenarios made and used by Shell.
r www.futurescenarios.org is a site with futures scenarios about the oil peak.
Glossary
This dictionary contains a list of acronyms and abbreviations and a brief etymology
and definition of key words and concepts introduced in this book. Etymology and
definitions are based on international English dictionaries and websites such as
www.etymonline.com/ and www.merriam-webster.com/dictionary. Wikipedia is also
a source for further enquiries.
Acronyms
Acronym Description Website

ABM Agent-Based Models www.openabm.org
BP Before Present; British Petroleum www.bp.com
CA Cellular Automata cell-auto.com/
CAS Complex Adaptive Systems
CE Christian Era
CFC ChlorFluorCompounds
CNG Compressed Natural Gas
CSD Commission on Sustainable www.un.org
Development (UN-based)
CSS Complex System Science
FAO Food and Agriculture Organization www.fao.org
(UN-based)
GDP Gross Domestic Product
GEO Global Environmental Outlook www.unep.org
GIS Geographical Information Systems
GMO Genetically Modified Organism
GWP Gross World Product
IAM Integrated Assessment Model
IIASA International Institute for Applied www.iiasa.ac.at
System Analysis
ICT Information and Communication
Technology
537
538 Glossary

IEA International Energy Agency www.iea.org
(OECD)
IFs International Futures model www.ifs.du.edu
IGBP International Geosphere www.igbp.ch
Biosphere Programme
IHDP International Human Dimensions www.ihdp.org
Programme
IMAGE Integrated Model to Assess the www.mnp.nl/image
Global Environment themasites.pbl.nl/en/
IMF International Monetary Fund www.imf.org
IMR Infant Mortality Rate
IPCC Intergovernmental Panel on www.ipcc.ch
Climate Change
IUCN International Union for the www.iucn.org
Conservation of Nature
LCA Life Cycle Analysis
MA Millennium Ecosystem www.millenniumassessment.org
Assessment
MAS Multi-Agent Simulation models
MDG Millennium Development Goals www.un.org/millenniumgoals/
MNC MultiNational Corporation
Netlogo 4.1 Netlogo agent-based simulation ccl.northwestern.edu/netlogo
modelling software
NPP Net Primary Production
OECD Organization for Economic www.oecd.org
Cooperation and Development
PAGES Past Global Changes network www.pages.org
PCRASTER PCraster environmental modelling pcraster.geo.uu.nl/
language
Phoenix Phoenix Population modelling themasites.pbl.nl/tridion/en/
framework themasites/phoenix/index.html
PBL PlanBureau voor de Leefomgeving www.pbl.nl
(Netherlands Environmental
Assessment Agency, formerly
MNP/RIVM)
PPP Purchasing Power Parity
R&D Research and Development
SAM Social Accounting Matrix
SCC Supply Cost Curve
SD Sustainable Development
SDI(S) Sustainable Development
Indicator (System)
SES Social-Ecological Systems www.resalliance.org
Glossary 539

SNA System of National Accounts
Stella
R
Stella system dynamics modelling www.iseesystems.com
software
UN United Nations
UNCED United Nations Commission on (‘Brundtland Commission’)
Environment and Development
UNDP United Nations Development www.undp.org
Programme
UNEP United Nations Environment www.unep.org
Programme
UNFPA United Nations Population Fund www.unfpa.org/public/
Vensim
R
Vensim system dynamics modelling www.vensim.com
software
WCDR World Commission on Disaster
Reduction
WCED World Commission on
Environment and Development
WEC World Energy Council www.worldenergy.org
WHO World Health Organization www.who.int/en/
(UN-based)
WMO World Meteorlogical Organization www.wmo.ch
WRI World Resources Institute www.wri.org
WSSD World Summit on Sustainable
Development (Johannesburg 2002)
WTO World Trade Organization www.wto.org
WWC World Water Council www.worldwatercouncil.org
WWF World Wildlife Fund www.wwf.org.uk
Adaptation
Synonym/associate: adaptability, adaptive capacity, coping capacity, resilience,
vulnerability
Etymology: from Latin adaptare, from ad near, and aptare to fit, aptus meaning apt,
fit.
In an evolutionary context, the modification of an organism or its parts that makes
it more fit for existence under the conditions of its environment. In a short-term and
local situation, adaptation means making or undergoing an adjustment to environ-
mental conditions.
Agrarianisation
Synonym/associate: agricultural transition
Etymology: from Latin root agrarius, from ager field.
The gradual intensification of the relationship between groups of humans, their
environment and each other. During this process, humans applied increasing
540 Glossary
amounts of labour and inputs to the land and raised agricultural output per unit
area and per unit time by applying more inputs (labour, fertiliser) and new tech-
niques (multi-/intercropping, irrigation).
Algorithm
Synonym/associate: arithmetic rule
Etymology: from medieval Latin algorismus, derived from Arabic al-khuw Arizmi,
an Arabian mathematician (825 CE).
A procedure for solving a mathematical problem (for instance finding the greatest
common divisor) in a finite number of steps that frequently involves repetition of an
operation. The broader meaning is a step-by-step procedure for solving a problem
or accomplishing some end, especially by a computer.
Analogy
Synonym/associate: resemblance, similarity
Etymology: from Greek analogia proportion.
If two things are similar in certain respects, the word analogy is used to infer that
they may also be equal in other respects.
Analysis
Etymology: from Greek, from analuyein to break up, to loosen.
The separation of a whole into its component parts and/or the identification or
separation of ingredients of a substance – as, for instance, in chemical analysis. In a
more abstract sense, it is proof of a mathematical proposition by assuming the result
and deducing a valid statement by a series of reversible steps. Less formally, it is an
examination of a complex, its elements, and their relations.
Anthropocene
Etymology: From Greek antrhopos man, human being and kainos new, fresh,
recent.
This word was coined by Störmer and Crutzen in their paper The Anthropocene
(Global Change Newsletter 41(2000)17–18). It was meant to indicate that with the
increase of human population and activities, the Earth has entered a new era.
Anthroposphere → see Sphere
Archetype
Synonym/associate: prototype, perfect example, generic structure
Etymology: from Greek archein to rule and typov type.
The original pattern or model of which all things of the same type are representations
or copies. In the psychology of Jung, it means an inherited idea or mode of thought
Glossary 541
that is derived from the experience of the race and is present in the unconscious of
the individual. In systems thinking, it refers to patterns of structure recurring again
and again.
Attractor
Synonym/associate: prototype
Etymology: from the Latin verb attrahere to pull or draw to.
A point or state to which a system is drawn or attracted. In mathematics, it is the
state in which one or more of the state variables do not change. In simple systems,
the attractor is a point in phase space. If the attractor is a closed trajectory in phase
space, it is called a limit cycle. If the attractor has a complex self-similar structure,
one speaks of a strange attractor.
Bifurcation
Synonym/associate: branching
Etymology: from Latin bifurcatus, of the verb bifurcare, from bi two, and furca fork.
The process which causes a system to divide into two branches or parts or during
which a system divides itself into two branches or parts.
Binomial distribution
Etymology: from Latin bi and nomen having two names.
A particular probability function, namely the one that develops from a series of
independent yes/no experiments each of which has a probability of success (yes) p.
The probability of getting k successes after n trials equals:

n
P(k; n, p) = pk (1 − p)n−k
k
with the first term the binomial coefficient.
Biodiversity
Synonym/associate: biological diversity, genetic diversity, species diversity, ecosys-
tem diversity
Etymology: from Greek bios life and from Latin divertere to turn in opposite direc-
tions.
The variety of ecosystems and species of plants and animals that can be found in
nature. Biodiversity matters at the level of genes, species and ecosystems. Genetic
diversity is the sum total of genetic information, contained in the genes of individual
plants, animals and microorganisms. Species diversity refers to the variety of living
organisms. Ecosystem diversity is about the variety of habitats, biotic communities,
and ecological processes in the biosphere, as well as the diversity in habitat and
542 Glossary
ecological processes within ecosystems. Preservation of biodiversity usually refers to

maintaining individuals or groups of species; conservation to maintaining the ability
of species to evolve.
Biogeography
Etymology: International Scientific Vocabulary.
The study and interpretation of geographical distribution of organisms (animals and
plants), both living and extinct.
Biome
Etymology (etymonline.com): from Greek biov life.
A vegetation type that contains ecosystem communities with similar growth forms
and interactions in animal, microbial and soil components and in above-ground
plants and trees. There is usually a relationship with climate and other environ-
mental factors. Biomes are abstractions: They have no sharp boundaries and mask
lower-scale heterogeneity. The concept is less useful for aquatic and mountainous
systems
Biosphere → see Sphere
Carrying capacity
Synonym/associate: equilibrium state, steady-state, attractor
The level of the population size, which the resources of the environment can just
maintain (‘carry’) without a tendency to either increase or decrease. At this popula-
tion density, birth rate equals death rate as a consequence of intraspecific competi-
tion. Mathematically, it is the equilibrium level or attractor in the corresponding
system description.
Causality
Synonym/associate: reason, motive
Etymology: from Latin causa a reason, interest, judicial process.
The relation between a cause and its effect or between regularly correlated events or
phenomena. A cause is a reason for an action or condition or a person or thing that
brings about an effect or a result. The notion of causality has also been defined more
formally. Let us assume that a situation is described at two given space-time points
{X1 ,t1 } (called the initial state) and {X2 ,t2 } (called the final state). If there exists a
dynamical equation of change that can be used to derive {X2 ,t2 } from {X1 ,t1 }, one
can speak of causation or causal linkage.
Glossary 543
Cellular Automata (CA)

Synonym/associate: robot, machine
Etymology: from Latin cella small room and from Greek automatos self-moving.
The name of a mathematical technique to describe change of a system element as a
consequence of changes in neighbouring elements.
Coevolution → see Evolution
Collapse
Synonym/associate: catastrophe, overshoot-and-collapse, disintegration
Etymology: from Latin collapsus, of the verb collabi: com with, and labi to fall, slide.
To fall or shrink together abruptly and completely, to break down completely or to
suddenly lose force, significance, effectiveness or worth.
Common Pool Resources (CPR)

Synonym/associate: ‘the commons’
Natural or humanly created systems (resource, facility) that generate a finite flow
of benefits, are available to more than one person and are subject to degradation
as a result of overuse. They can be owned by a community or a government or
be privately owned. When no property rights are established and its use is not
regulated, the CPR is under an open-access regime. Examples are a grazing area, a
fishing ground, a waterstream, the atmosphere, a natural park or a road.
Competition
Synonym/associate: rivalry
Etymology: from the Latin competitio and the Latin verb competere strive together,
from com with, and petere to strive, seek, fall upon, rush at, attack.
The act or process in which two or more parties acting independently attempt to
win or gain, for instance, two or more organisms competing for some environmental
resource in short supply. More generally, it denotes a contest between rivals. In the
World Economic Forum’s Global Competitiveness Report, it is defined as ‘the fitness
of a country’s economic institutions and structures to produce growth, in view of the
overall structure of the global economy’.
Complexity
Synonym/associate: complex systems, complexity theory, complexity science
Etymology: from Latin complexus totality, and the Latin verb complecti to encircle
or embrace.
The quality or state of being complex. As a noun, a complex is a whole being made
up of interrelated parts. As an adjective, it refers to obviously related units of which
544 Glossary
the degree and nature of the relationship is imperfectly known. Nonlinearity and
collective behaviour are characteristic features of a complex system. In the context
of socio-natural systems, it means differentiation in social, political and/or economic
structure combined with organisation that integrates diverse structural parts into a
whole.
Complex Adaptive Systems (CAS)

Synonym/associate: agent-based modelling (ABM), rule-based modelling, multi-
agent simulation (MAS)
Systems with inherent uncertainty in their dynamics that tend to have multiple stable
states and that exhibit self-organisation. More generally, it refers to systems that show
complex adaptive behaviour in response to environmental change – ecosystems but
mostly systems with human actors.
Conspicuous consumption
Synonym/associate: striking, extravagant
Etymology: Latin conspicuus, from conspicere to get sight of, from com with, and
specere to look.
Behaviour of extravagant and ostentatious displays of resources that function as
a competitive strategy to demonstrate wealth and social status. The term was
coined by the economist Thorstein Veblen. Conspicuous means obvious to the
eye or mind, attracting attention or marked by a noticeable violation of good
taste.
Contingency
Synonym/associate: accidental, possible, unpredictable
Etymology: from Latin contingent-, contingens, present participle of contingere to
have contact with, befall, from com with, and tangere to touch.
The quality or state of being contingent means being likely but not certain to happen,
of being not logically necessary, of happening by chance or unforeseen causes or
of being subject to chance or unseen effects. Contingency in socio-natural system
evolution is the dependence of conditions on the operation of previous processes.
Correlation
Etymology: from Latin correlation: com with, and relatio a bringing back or restoring.
A relation existing between phenomena or things or between mathematical or
statistical variables which tend to vary, be associated, or occur together in a way
not expected on the basis of chance alone.
Glossary 545
Cosmology
Etymology: from Greek kosmov meaning (good) order or orderly arrangement, and
logov discourse.
Life on earth, in this world, as opposed to the afterlife. Later its use was extended
to the known world, and still later to the physical world in the broadest sense: the
universe.
Dialectic
Etymology: from Greek dialektike (techne) (art of) philosophical discussion or
discourse.
Originally, reasoning by dialogue as a method of intellectual investigation. In
philosophy, it is a process of resolving or merging contradictions in character and in
which a concept or development changes into its opposite while also being preserved
and fulfilled. It got its meaning from the works of Kant and Hegel.
Discounting
Synonym/associate: disregard, discount rate, time preference
Etymology: from Latin discomputare, from dis not or lack of, and computare to
count.
The verb to discount means, in everyday market speak, to make a deduction from
usually for cash or prompt payment or to sell or offer for sale at a reduced price.
It also means: to leave out of account, to disregard or minimise the importance, to
make allowance for bias or exaggeration or to view with doubt. It is in this sense that
it is used in taking into account future events in present (financial) calculations.
Dissipative structure
Synonym/associate: dissipation, dispersion, diffusion, dissolution
Etymology: from Latin dissipatus, from the verb dissipare, from dis not or lack of,
and supare to throw.
The verb to dissipate, in a transitive sense, means: to break up and drive off (as a
crowd), to cause to spread thin or scatter and gradually vanish, or in physics: to lose
(as heat or electricity) irrecoverably. It may have a negative connotation: to spend
or use up wastefully or foolishly. In an intransitive sense, it means: to break up and
scatter or vanish.
Ecology
Etymology (m-w.com): from Greek oikos house and logos reason or idea.
A branch of science concerned with the interrelationship of organisms and their
environments as well as what this science describes: the totality or pattern of relations
between organisms and their environment. In the last three decades, several derived
scientific branches have emerged: human ecology, industrial ecology, social ecology.
546 Glossary
Economy
Synonym/associate: economic science, economics
Etymology: from Latin oeconomia, from Greek oikonomos household manager, from
oikos house and nemein to manage.
An archaic meaning of economy is the management of household or private affairs
and especially expenses. From this meaning, it has gone into two directions:
r equating it with good (household) management: thrifty and efficient use of
material resources, frugality in expenditures, or even an instance or a means of
saving on resources as in ‘economies of scale’. It may also concern the effi-
cient and concise use of nonmaterial resources such as effort, language, or
motion.
r equating it with the whole of production and consumption: the arrangement or
mode of operation of something.
As with ecology, there are several scientific branches derived from ‘the science of
the economy’: ecological economics, evolutionary economics and others.
Ecosystem
Etymology: from Greek oikos house.
A complex of a community of organisms and its environment functioning as an
ecological unit. In the UN Convention on Biological Diversity, it is defined as ‘a
dynamic complex of plant, animal and micro-organism communities and non-living
environment interacting as a functional unit’. Another definition is: any unit that
includes all of the organisms in a given area interacting with the physical environment
so that a flow of energy leads to . . . exchange of materials between living and non-
living parts within the ecosystem.
Ecotone
Etymology: from Greek ec out and tonov tension.
A narrow and rather sharp defined transition zone between two ecological com-
munities; they are usually species-rich and may have come into existence either from
natural or anthropogenic processes.
Emergence
Synonym/associate: rise
Etymology: from Latin emergere, rise out or up, bring forth, bring to light, from ex
out, and mergere to plunge.
The verb to emerge refers to biological processes of various superficial outgrowths
of plant tissue or of the penetration of the soil surface by a newly germinated plant.
In complexity science, it has been used in relation to emergent properties: system
Glossary 547
properties at a certain level of observation and abstraction, that cannot be observed

at a lower level of individual elements and are the product of the interactions between
the elements (and “more than the sum of the parts”). As with complexity and self-
organisation, the observer is involved in its definition.
Emic and etic
Etymology: from Greek etikon pertaining to.

The ending emic means relating to or involving analysis of cultural phenomena from
the perspective of one who does participate in the culture being studied, whereas the
ending etic means relating to or involving analysis of cultural phenomena from the
perspective of one who does not participates in the culture being studied.
Emulation
Synonym/associate: imitation
Etymology: from Latin aemulatus, of the verb aemulari to rival.
To imitate, with the connotation of ambition or endeavour to equal or excel others.
This aspect is emphasised in the expression of competitive emulation – a process in
which individuals, families, companies or states compete.
Entropy
Synonym/associate: (dis)order, chaos
Etymology: from Greek en in, and the verb trepein to turn.
In a natural science context, a well-defined quantitative measure of the microscopic
disorder of a system. Entropy change is caused by heat transfer, mass flow and
irreversible processes degrading energy from higher to lower order. In a broader
and looser sense, entropy is the degree of disorder or uncertainty in a system and
refers to a process of degradation or running down or a trend to disorder.
Epistemology
Etymology: from Greek episthmh knowledge, from epistanai to understand, know,
from epi with, and histanaii to cause to stand.
The study or a theory of the nature and grounds of knowledge especially with
reference to its limits and validity. Philosophers always warn to distinguish between
what can be known (epistemology) and what is (ontology).
Equilibrium (state)
Synonym/associate: attractor, steady-state, carrying capacity
Etymology: from Latin aequilibrium, from aequi equal and libra weight, balance.
In a mechanical sense, a state of balance between opposing forces or actions that is
either static (as in a body acted on by forces whose resultant is zero) or dynamic (as
548 Glossary
in a reversible chemical reaction when the rates of reaction in both directions are
equal). It is often used outside the mechanical context, as in systems theory: a state of
adjustment between opposing or divergent influences or elements, or in psychology:
a state of intellectual or emotional balance.
Eutrophication
Etymology: from Greek eu good and trojh nourishment.
The process of enrichment of water by nutrients, often stimulating plant growth and
as a result depletion of oxygen in the water.
Evolution
Synonym/associate: co-evolution, development, unfolding, growth
Etymology: from Latin evolutio unrolling, from the verb evolvere to unroll.
A process of continuous change from a lower, simpler, or worse to a higher, more
complex, or better state. In biology, it is specifically about the historical development
of a biological group, as a race or species. The underlying evolution theory asserts that
the various types of animals and plants have their origin in other preexisting types
and that the distinguishable differences are because of modifications in successive
generations. In a cosmological sense, it is a process in which the whole universe
is a progression of interrelated phenomena. In a social science context, it means a
process of gradual and relatively peaceful social, political, and economic advance –
in contrast to a revolution with its suddenness and discontinuity.
Exergy → see Energy
Fractal
Synonym/associate: self-similar, self-symmetric
Etymology: from Latin fractus fragmented or broken.
A geometrical or physical structure having an irregular or fragmented shape at all
scales of measurement between a greatest and smallest scale. They exhibit self-
similarity. The mathematician Mandelbrot (1977) has proposed the name fractal
geometry; it has been applied to such diverse fields as the stock market, chemical
industry, meteorology and computer graphics.
Gradient
Etymology: from Latin gradientem, derived from gradi to walk.
A (steep) slope of a road or railroad. In mathematics, it refers to the curvature in a
surface in phase or state space, such as the first derivative ∂F/∂Xi of a state variable
Xi for F = F(X1 . . . Xn ).
Glossary 549
Growth
Synonym/association: development
Etymology: from Old English growan and akin to Old High German gruowan
to grow. It supplanted in medieval times the Old English weaxan (cf. German
wachsen).
To grow, or the process of growth, means to spring up and develop to maturity, to be
able to grow in some place or situation and to assume some relation through or as if
through a process of natural growth. It also describes the phenomenon of increase in
size by assimilation of material into the living organism or by accretion of material
in a nonbiological process (such as crystallisation).
Heterogeneity
Synonym/associate: diverse, different
Etymology: from Greek eterogenev, from eter other and genos kind.
An area, a population and so on are said to be heterogeneous if they consist of dissim-
ilar or diverse ingredients or constituents. The opposite is homogeneity: consisting
of the same ingredients or constituents.
Heuristic
Synonym/associate:
Etymology: from Greek euriskein to find, discover.
A device or procedure that involves or serves as an aid to learning, discovery, or
problem solving by experimental and especially trial-and-error methods. In a com-
puter context, it refers to exploratory problem-solving techniques. In a more general
sense, it is a set of principles used in making decisions when not all possibilities can
be fully explored and consisting of guided trial and error.
Hierarchy
Synonym/associate: order, ranking
Etymology: from Greek ierov priest and arch, from archein to begin, rule, and arch
beginning, rule, archov ruler.
The first meaning of hierarchy in the English dictionary is a division of angels: the
celestial hierarchy. These religious roots also show up in the second meaning: a ruling
body of clergy organised into orders or ranks each subordinate to the one above it
and, more general, church government by a hierarchy. Later on, it has become a
wider term: a body of persons in authority, the classification of a group of people
according to ability or to economic, social or professional standing, or a graded or
ranked series as in a hierarchy of values.
550 Glossary
Homogeneity → see Heterogeneity
Identity
Etymology: from Latin identitas sameness, and the word idem the same.
The property of remain the same, having essential or generic character in different
instances. Identity can thus denote that which constitutes the objective reality of a
thing or a person. In the latter, psychological sense, it is the distinguishing character
or personality of an individual (individuality) and can serve as a way to establish
relations between individuals.
Indicators
Etymology: from Latin indicare to point out, show, from in in, and dicare to proclaim
and dicere to speak, to say.
An indicator is meant to provide information about part(s) of a system which is
useful within the context of the receiver. The usefulness depends on the objective
of the user, on the presumed relation between indicator and system, and on the
(scientifically) acknowledged relation between indicator and system. The degree to
which control over the system is possible in the form of choice (amongst alternatives)
matters too.
Inference
Synonym/associate: conclusion, deduction, consequence
Etymology: from Latin infere, from Greek ein in and jerein to carry, bring.
The act of making a statement or judgment that is considered to be valid or true on
the basis of another statement or judgment that one believes to be valid or true. It
also denotes the step from statistical data to a generalisation.
In silico
Etymology: from the Latin word for the element silicium or silicon.
Something (done) in or on a computer or in computer simulations/with computer
software. Coined recently after in vivo and in vitro, it stems from the notion that
silicium is an important element in chips (and the core element of sand).
Institution
Synonym/associate: (self-)organisation, establishment
Etymology: from Latin from instatuere to set up, from in in and statuere to establish,
cause to stand.
The word institution refers to a significant practice, relationship or organisation in
a society or culture, as in the institution of marriage. In a political/organisational
context, it means an established organisation or corporation, especially of a public
character, as for instance a college or university.
Glossary 551
Institutions can be viewed as a set of rules actually used by a set of individuals to

organise repetitive activities that produce outcomes affecting those individuals and
potentially affecting others. They structure human interaction, consisting of formal
(such as law) and informal (such as behavioural norms) constraints and enforcement
characteristics. Institutions are social constructs with their own normative, cognitive
and regulative (goals, values, paradigms or instruments) dimensions.
Irreversibility → see Reversibility
Isomorphism
Synonym/associate: similarity
Etymology (m-w.com): from Greek isov same and morjov form.
In mathematics, a one-to-one correspondence between two sets of objects.
Limit cycle → see Attractor
Marginal Cost
Etymology: from Latin margo edge, boundary space.
In economics, it is the cost of one additional unit. Mathematically, it is the first
derivative of the total production cost with respect to the coresponding input.
Metabolism
Etymology: from Greek metabolh change, from metaballein to change, from meta
after, beyond, and ballein to throw.
Metabolism is the set of chemical changes in living cells by which energy is provided
for vital processes and activities and new material is assimilated. In biology, it is the
thousands of biochemical reactions that sustain the processes of life and in particular
of cells and form metabolic networks. By analogy, industrial metabolism refers in
human/industrial ecology to the transformation of material fluxes in the natural
environment into usable ‘goods’ and their subsequent excretion, discarding and/or
recycling. In the social sciences, Marx and Engels introduced metabolism for the
material exchange between man and nature on a fundamental anthropological level
and as a critique of the alienating capitalist mode of production. It is a translation of
the German word Stoffwechsel, introduced in the 19th century for the description of
material exchanges within the (human) body.
Metaphor
Synonym/associate: comparison, analogue, symbol
Etymology: from Latin metaphora, from Greek metajerein to transfer, from meta
after, beyond, and jerein to bear.
A figure of speech in which a word or phrase literally denoting one kind of object
or idea is used in place of another to suggest a likeness or analogy between them. A
552 Glossary
thing is given a name that belongs to something else, but not all of the characteristic
properties are carried over together with the name. The purpose of using a metaphor
is to elucidate a new and unfamiliar concept by taking support from a familiar one.
Metaphors are violations of literalness – and it is from the form of violation that they
draw their emotive and cognitive effects. A prerequisite for correct use is that the
users have the same perception and experience of the named ‘thing’.
Model
Synonym/association: archetype, copy, image, replica, analogue, metaphor
Etymology: from Latin modulus small measure, from modus way in which something
is done.
Any physical or symbolic representation of certain properties and interactions of a
system. A model is, in a way, a primitive theory. A theory is, in a way, a generalised
model. The domain of a theory or model is the state space in which it is valid and
usable. A domain is always limited.
Nature
Etymology: from Latin natura course of things, derived from natus born.
The original from Latin word natura it denotes the course of things, the universe. The
medieval connotation of the word Nature is: essential qualities, innate disposition,
and also the creative power in the material world. Used for humans, as in human
nature, it refers to the inherent, dominating power or impulse of a person.
Noösphere → see Sphere
Observation
Etymology: from Latin observare to attend to, to keep, to follow.
A series of organised sensory experiences. Events are recorded within a certain time-
space domain. Recording can be qualitative, using verbal descriptions or ordinal
ranking, or quantitative, using measurement units. From a cognitive science per-
spective, observation is the interface between perception and learning.
Ontology
Etymology: from Greek ontov being and logov word, reason.
A part of metaphysics concerned with the nature and relations of being or existence,
or a particular theory about it.
Glossary 553
Organisation
Synonym/associate: establishment, society
Etymology: from Latin organum instrument, organ.
An administrative and functional structure, as a business or a political party. It refers
in a more formal sense to a set of entities that constitute a dynamical system and
are organising or being organised with a collective purpose. Self-organisation refers
to a set of dynamical systems, whereby structures appear at the global level of a
system from interactions amongst its lower-level components. It coincides with a
move towards lower probability and higher order, and increasing heterogeneity and
differentiation. It involves structures with function or purpose and it requires a non-
trivial fitness function. If a system involves only one or a few elements and/or follows
rules that can be easily understood, one does not speak of self-organisation.
Panarchy
Etymology: from Greek pan everything, all and archein to rule, command.
A nested set of adaptive cycles at different scales, that exhibits cross-scale inter-
actions. The concept is used by Holling and colleagues to capture the hypothesis
that ecosystems evolve through adaptive cycles that are nested one within the other
across space and time scales. Although many ecosystem observations are in con-
formity with the panarchy hypothesis, its validity and usefulness in ecological and
other (economic, social) domains is still controversial.
Paradigm
Synonym/associate: example, pattern, archetype
Etymology: from Greek paradeigma, from para alongside, near, and deiknynai to
show.
Originally, the word paradigm was simply synonymous with pattern or model. Later,
it was used for an outstandingly clear or typical example or archetype. With the work
of Kuhn, it acquired the meaning of a philosophical and theoretical framework of
a scientific school or discipline within which theories, laws, and generalisations and
the experiments performed in support of them are formulated. In a broad sense,
the (dominant) paradigm is the basic way of perceiving, thinking, valuing and doing,
associated with a particular vision of reality.
Pastoralism
Etymology: from Latin pastorem shepherd, also spiritual guide (shepherd of souls).
Related to pasture, from Latin pastura feeding, grazing.
The practise of herding as the primary economic activity of a society.
554 Glossary
Pedigree
Etymology: probably from a forked sign as from a bird’s footprint, in old French
pied de gru foot of a crane.
Ancestral line, descent, genealogical chart or table.
Perennial (wisdom)
Etymology: from Latin per through annum year.
Lasting through the years, enduring, permanent.
Persistence
Synonym/association: continuity, resilience
Etymology: from the Latin verb persistere, existing for a long or longer than usual
time or continuously.
A good or substance is persistent if it is retained beyond the usual period, continues
without change in function or structure, is effective in the open for an appreciable
time usually through slow volatilising, degrades only slowly by the environment or
remains infective for a relatively long time in a vector after an initial period of
incubation.
Phase plane
A plane (or space) with all the dimensions needed to describe dynamical physical
behaviour. For a single state variable X, it contains the dimensions X and dX/dt.
For a system with two or more state variables, it is the n-dimensional phase space
{X1 ..Xn }.
Phenomenology
Synonym/associate: appearance
Etymology: from the Greek verb phainesthai to appear, and related to phainein to
show.
An observable fact or event, or an object or aspect known through the senses rather
than by thought or intuition. As a philosophical school, it refers to the study of the
development of human consciousness and self-awareness as a preface to or a part of
philosophy.
Photosynthesis
Etymology: from Greek joton light and synthesiv composition, something put
together.
The synthesis process of chemical compounds with the aid of radiant energy and
especially light; in particular, the formation of carbohydrates from carbon dioxide
and a source of hydrogen (as water) in the chlorophyll-containing tissues of plants
exposed to light.
Glossary 555
Power laws
A mathematical description applying when the size S of a certain phenomenon
(event, property) and its frequency of occurrence N(S) are a straight line if plotted
in a log-log graph, that is, N(S) = aS−α . They are common in nature. Examples are
the energy release of earthquakes E and their frequency F(E). Power laws are scale
invariant: if S′ = βS, then N(S′ ) = aS′−α = aβ −α S−α = a′ S−α .
Redundancy
Synonym/associate: superfluity, abundance, diffuseness

Etymology: from Latin redundare come back, contribute, literally: overflow, from re
again, and undare rise in waves.
The property of abundance in the sense of being superfluous, but nevertheless useful
in case of emergency or failure as a backup.
Resilience
Synonym/associate: buffer, stability, robustness
Etymology: from the Latin verb resilire to jump back, recoil, from re back, and salire
to leap.
The capacity of a system to absorb disturbance, undergo change and still retain
essentially the same function, structure, identity, and feedbacks. A measure of the
resilience of a system is the full range of perturbations over which the system can
maintain itself. Resilience is often viewed and used as a general term meaning either
stability or robustness and linked to sensitivity and vulnerability.
Reversibility
Etymology: from Latin revertere to turn back.
The property of being capable of being reversed (in time). Irreversible is: not capable
of being reversed (in time).
Robustness
Synonym/associate: sturdy, resilient
Etymology: Latin robustus oaken, strong, from robur oak, strength.
The property of having or exhibiting strength or vigorous health, being strongly
formed or constructed or being capable of performing without failure under a wide
range of conditions. Robustness is associated with having or showing firmness, such
as having a solid structure that resists stress, not subject to change or revision, not
easily moved or disturbed. If a system has a well-defined identity, robustness refers
to the property to either absorb external perturbations from sheer size/mass or
redundancy or to use controls on the basis of anticipation and power.
556 Glossary
Scenario
Synonym/associate: screenplay, scenario method/approach, myth, strategy, plan,
visioning
Etymology: from Latin scaenarium place for erecting stages, from scaena stage.
An outline or synopsis of a play, more spcifically, a plot outline used by actors of the
commedia dell’arte. In a broader sense, it is a sequence of events especially when
imagined. As a methodology, the scenario method has become widespread over the
last few decades in futures research and strategic planning.
Science → see Chapters 1, 8 and 10
Sedentary
Etymology: from the Latin verb sedēre to sit.
Non-migratory; permanently attached.
Self-organisation → see Organisation
Self-similarity
Synonym/associate: self-resemblance, self-symmetry, scale invariance, fractal
Etymology: from Latin similis like.
A self-similar pattern or object is one whose component parts resemble the whole
and that remains invariant under changes of scale, that is, they are scale-symmetric.
Sphere
Etymology: from Greek sjaira globe, ball. Biosphere from Greek biov life. Pedo-
sphere from Greek pdon soil. Ecosphere from Greek oikos house, habitation.
Anthroposphere from Greek antropos man, human being. Sociosphere from Latin
socius companion. Technosphere from Greek technh art, skill or craft. Noösphere
from Greek noov mind.
A ball or body of globular form. Later, it broadened to the range of something or a
space where certain processes exist or dominate. In sustainable development context
one distinguishes:
r The biosphere, introduced by in 1883 by the geologist Suess and later reconceived
and refined by the chemist Vernadsky in the 1920s. Building upon insights from
geology and agrochemistry, Vernadsky coined it to indicate the ‘la région unique
de l’écorce terrestre occupée par la vie’ (Vernadsky, La Biosphère (1929):19), and
considered its investigation a necessarily transdisciplinary enterprise. The word
ecosphere is used in similar ways.
r The pedosphere is the outermost layer of the Earth with its soils. It is where soil
formation processes take place.
Glossary 557
r The anthroposphere is the part of the Earth where humans have a significant
impact on energy and material fluxes. The sociosphere is sometimes used to
indicate a similar domain.
r The technosphere is used to distinguish the part which consists of human-made
capital stocks and matter, energy and information flows.
r The noösphere was a word crafted by the palaeologist Teilhard de Chardin to
indicate the collective of human minds.
Spirituality
Synonym/associate: incorporeal, transcendent (in English also: church, clergy)
Etymology: from Latin spiritus of breathing, of the spirit.
Relating to, consisting of, or affecting the spirit, relating to sacred matters, con-
cerned with religious values or relating to supernatural beings or phenomena. In
English, it also means ecclesiastical rather than lay or temporal. In the official
definition of the British Department of Education, spirituality is the valuing of the
non-material aspects of life, and intimations of an enduring reality. The French spir-
itualité denotes that which is from the esprit and is disconnected from any material
manifestation.
Stability
Synonym/associate: resilience, firmness
Etymology: from Latin stabilis, from stare to stand.
The quality, state, or degree of being stable, such as having the strength to stand
or endure. The adjective stable means firmly established; unvarying, permanent and
enduring; not changing or fluctuating. It is also used in a more active, psychological
sense: steady in purpose, firm in resolution, not subject to insecurity or emotional
illness, as in a stable personality.
Strange attractor → see Attractor
Structure
Synonym/associate: construction, arrangement
Etymology: from Latin structura a fitting together, adjustment, building, from the
verb struere to build up, heap.
Something that is constructed, arranged or organised in a definite pattern.
558 Glossary
Sustainability → see Chapter 1
Syndrome
Synonym/associate: pattern
Etymology: from Greek syndrome concurrence of symptoms, concourse, from syn
with and dromov running, course.
In medical terminology, a syndrome is a complex clinical picture. Outside of medi-
cine it refers to a combination of phenomena seen in association. The concept of
syndrome has been introduced by the WBGU (1994) to understand Global Change
as a co-evolution of dynamic partial patterns of unmistakable character or symptoms.
Symptoms, derived from Greek and meaning accident or misfortune, is a departure
from normal function or feeling and an indication of the presence of disease or
abnormality.
Taxonomy
Synonym/associate: classification
Etymology: from Greek taxiv arrangement and nemein to manage.
The study of the general principles of scientific classification; more specifically, the
orderly classification of plants and animals according to their presumed natural
relationships.
Technosphere → see Sphere
Template
Synonym/associate: mold, example
Etymology: probably from French templet, diminutive of temple part of a loom,
probably from Latin templum plank.
Literally, a short piece or block placed horizontally in a wall under a beam to
distribute its weight or pressure (as over a door). It also is a gauge, pattern, or mold
used as a guide to the form of a piece being made. In a figurative way, it refers to
something that establishes a pattern or serves as an example, for instance, a molecule
(such as DNA) that serves as a pattern for the generation of another macromolecule
(such as messenger RNA).
Theory
Synonym/associate: hypothesis, speculation, conjecture
Etymology: from Greek qeorein the analysis of a set of facts in their relation to one
another.
The general or abstract principles of a body of fact or a plausible or scientifically
acceptable general principle or body of principles offered to explain phenomena.
Less strictly, it is a belief, policy, or procedure proposed or followed as the basis of
action. It may also be an ideal or hypothetical set of facts, principles, or circumstances,
as used in the phrase in theory.
Glossary 559
Topology
Etymology: from Greek topov place and grajein to write.
A branch of mathematics concerned with those properties of geometric configura-
tions (as point sets), which are unaltered by elastic deformations (as a stretching or
a twisting).
Transaction cost
Etymology: from Latin transactionem agreement, accomplishment, from trans
through, and agere to drive.
Generally speaking, it refers to costs that are made for an exchange in the market or
for particpation in market exchanges. Examples are the costs to acquire information
and to police and enforce contracts.
Utilitarianism
Etymology: from Latin utilitas usefulness, serviceableness, profit.
A doctrine that states that the useful is the good and that conduct should ethically be
judged for its utility, that is, the usefulness of its consequences. In practice, it implies
actions aiming at the largest possible balance of pleasure over pain or, in society, at
the greatest happiness of the greatest number. Utilitarianism is associated with the
British philosopher Bentham.
Utopia
Etymology: from Greek ou not and topov place.

The word was created by Thomas More for the title of his book Utopia (1516) about
a perfect, imaginary island. It is used to indicate a perfect but infinitely remote and
impossible to reach land (‘nowhere’) or situation.
Vicious/virtuous circle
Etymology: from Latin vitiosus faulty, defective, corrupt; from Latin virtus) moral
strength, manliness, excellence, from vir man.
A vicious circle or cycle is a series of events that reinforces itself through a positive
feedback loop, with a negative result. The response to a problem causes a new
problem that aggravates the situation. A virtuous circle or cycle is the same but with
a favorable outcome. Vicious circles can be converted into virtuous ones and vice
versa.
Vulnerability
Synonym/associate: sensitivity, lack of resilience
Etymology: from Latin vulnus a wound, and vulnerare to wound.
A strict definition of vulnerability is able to be hurt or wounded. In a broader
sense, it can be viewed as the potential for negative outcomes or consequences.
560 Glossary
Three distinct clusters of definitions for vulnerability have been identified: risk of
exposure to hazards, a capability for social response, and an attribute of places (such
as vulnerability of coastlines to sea level rise). Other descriptions are the interface
between exposure to the physical threats to human well-being and the capacity
of people and communities to cope with those threats, and the degree to which a
system or community is susceptible to, or able to cope with, adverse effects of such
as climate change. The common element is the potential of a system to be damaged
or deteriorate.
References
Aalbers, T., ed. Waardenoriëntaties, wereldbeelden en maatschappelijke vraagstukken: Ver-

antwoording van het opinieonderzoek voor de Duurzaamheidsverkenning Kwaliteit en
Toekomst. Report 550031002. Bilthoven, the Netherlands: Netherlands Environmental
Assessment Agency (PBL), 2006. Available at www.pbl.nl/search/node/550031002.
Acevedo, M., F. Urban, and H. Shugart. Models of forest dynamics based on roles of tree
species, Ecological Modelling 87 (1996): 267–284.
Achterhuis, H. Utopia’s Heritage (De erfenis van de utopie). Amsterdam: Ambo, 1998.
Adams, J. Risk. London: UCL Press Ltd., 1995.
Akerlof, G., and R. Shiller. Animal Spirits – How Human Psychology Drives the Economy,
and Why It Matters for Global Capitalism. Princeton, NJ: Princeton University Press,
2009.
Alcamo, J., P. Döll, T. Henrichs, F. Kaspar, B. Lehner, T. Rösch, and S. Siebert. Global
estimation of water withdrawals and availability under current and “business as usual”
conditions, Hydrological Science, 48(2003):339–348.
Alkemade, R., M. van Oorschot, L. Miles, C. Nellemann, M. Bakkenes, and B. ten Brink.
GLOBIO3: A framework to investigate options for reducing global terrestrial biodiversity
loss, Ecosystems 12 (2009): 374–390.
Allen, P., and J. McGlade. Modelling complex human systems: A fisheries example, European
Journal of Operations Research 30 (1987): 147–167.
Allwood, J., M. Ashby, T. Gutowski, and E. Worrell. Material efficiency: A white paper,
Resources, Conservation and Recycling 55 (2011): 362–381.
Ananthu, T. Gandhi’s Hind Swaraj. New Delhi: Gandhi Peace Foundation, n.d.
Anderson, C., K. Frenken, and A. Hellervik. A complex network approach to urban growth,
Environment and Planning A 38 (2006): 1941–1964.
Armstrong, K. The Great Transformation: The World in the Time of Buddha, Socrates, Con-
fucius and Jeremiah. New York: Random House, 2006.
Armstrong, K. Twelve Steps to a Compassionate Life. New York: Knopf Doubleday Publish-
ing, 2010.
Arquitt, S., X. Honggang, and R. Johnstone. A system dynamics analysis of boom and bust in
the shrimp aquaculture industry, System Dynamics Review 21 (2005): 305–324.
Arrow, K. The Limits of Organization. New York: W.W. Norton & Company, 1974.
Arthur, W. Increasing Returns and Path Dependence in the Economy. Ann Arbor: The Uni-
versity of Michigan Press, 1994.
Aslin, H., and D. Bennett. Wildlife and world views: Australian attitudes to wildlife, Human
Dimensions of Wildlife 15(2000): 15–35.
Atkisson, A. The ISIS Agreement – How Sustainability Can Improve Organizational Perform-
ance and Transform the World. London: Earthscan, 2008.
Aurobindo. Le Cycle Humain. Pondichéry, India: Sri Aurobindo Ashram, 1972/1998.
561
562 References
Axelrod, R. The Evolution of Cooperation. New York: Basic Books, 1984.

Axelrod, R. The Complexity of Cooperation: Agent-Based Models of Competition and Col-
laboration. Princeton, NJ: Princeton University Press, 1997.
Ayres, R., and L. Ayres, eds. A Handbook of Industrial Ecology. Cheltenham, UK: Edward
Elgar, 2002.
Ayres, R., and B. Warr. Accounting for growth: the role of physical work, Structural Change
and Economic Dynamics 16 (2005):181–209.
Azar, C., and S. Schneider. Are the economic costs of stabilizing the atmosphere prohibitive?
Ecological Economics 42 (2002): 73.
Bai, Z., D. Dent, L. Olsson, and M. Schaepman. Proxy global assessment of land degradation,
Soil Use and Management 24 (2008): 223–234.
Bakan, J. The Corporation: The Pathological Pursuit of Profit and Power. New York: Free
Press, 2004.
Bakkes, J., et al. Background Report to the OECD Environmental Outlook to 2030. Over-
views, Details, and Methodology of Model-Based Analysis. Bilthoven, the Netherlands:
Netherlands Environmental Assessment Agency (PBL), 2008.
Barash, D. The Survival Game: How Game Theory Explains the Biology of Cooperation and
Competition. New York: Times Books, 2004.
Barnes, P. Capitalism 3.0 – A Guide to Reclaiming the Commons. San Francisco: Barrett-
Koehler Publishers, 2006.
Barney, G., ed. The Global 2000 Report to the President: Entering the 21st Century. London:
Penguin Books, 1980/1982.
Barro, R., and X. Sala-i-Martin. Economic Growth. (2nd ed.). Cambridge, MA: MIT Press,
2004.
Batty, M. Cities and Complexity – Understanding Cities with Cellular Automata, Agent-Based
Models, and Fractals. Cambridge, MA: MIT Press, 2005.
Basavanna, in: Speaking of Siva (Translated with an introduction by A.K. Ramanujan). Lon-
don: Penguin Books, 1973.
Batty, M. Rank clocks, Nature 444 (2006): 592–596.
Baumann, H., and A.-M. Tillman. The Hitch Hiker’s Guide to LCA. An Orientation in Life
Cycle Assessment Methodology and Application. Lund, Sweden: Studentlitteratur AB, 2004.
Baumol, W. Macroeconomics of unbalanced growth: The anatomy of urban crisis, American
Economic Review 57 (1967): 415–426
Beder, S. The Nature of Sustainable Development. Newham, Australia: Scribe Publication,
1993.
Beinhocker, E. The Origin of Wealth – Evolution, Complexity and the Radical Remaking of
Economics. London: Random House, 2005.
Benı́tez-López, A., R. Alkemade, and P. Verweij. The impacts of roads and other infrastruc-
ture on mammal and bird populations: A meta-analysis, Biological Conservation 143 (2010):
1307–1316.
and Social Mechanisms for Building Resilience. Cambridge: Cambridge University Press,
1998.
Berkes, F., J. Colding, and C. Folke. Navigating Social-Ecological Systems – Building Resili-
ence for Complexity and Change. Cambridge: Cambridge University Press, 2003.
Bernardini, O., and R. Galli. Dematerialization: Long-term trends in the intensity of use of
materials and energy, Futures 25 (1993): 431–448.
Beumer, C., and P. Martens. Noah’s ark or world wild web? Cultural perspectives in global
scenario studies and their function for biodiversity conservation in a changing world, Sus-
tainability 2 (2010): 3211–3238.
Blanton, R., S. Kowalewski, G. Feiman, and L. Finsten. Ancient Mesoamerica. Cambridge:
Cambridge University Press, 1993.
Blum, U., and L. Dudley. Religion and economic growth: Was Weber right? Journal of
Evolutionary Economics 11 (2001): 207–230.
References 563
Boersema, J. Hoe Groen is het Goede Leven (How Green is the Good Life). Inaugural
Address. Amsterdam: Free University, 2002.
Boersema, J., and L. Reijnders. Principles of Environmental Sciences. The Netherlands:
Springer Science +Business Media B.V., 2009.
Boorstin, D. The Discoverers – A History of Man’s Search to Know His World and Himself.
New York: Vintage Books, 1985.
Boserup, E. Population and Technological Change. Chicago: The University of Chicago Press,
1966.
Bossel, H., and H. Krieger. Simulation model of natural tropical forest dynamics, Ecological
Modelling 59 (1991): 37–71.
Bossel, H. Modeling and Simulation. Wellesley, MA: AK Peters and Wiesbaden: Vieweg,
1994.
Bossel, H. TREEDYN3 forest simulation model, Ecological Modelling 90 (1996): 187–227.
Bossel, H. Earth at a Crossroads – Paths to a Sustainable Future. Cambridge: Cambridge
Boucheaud, J.-P., and M. Mézard. Wealth condensation in a simple model of economy, Physica
A 282 (2000): 536–545.
Boulding, K. Ecodynamics – A New Theory of Societal Evolution. London: Sage Publications,
1978.
Bourdieu, P. Science de la science et réflexivité. Paris: Raisons d’agir Editions, 2001.
Bourguignon, F., and C. Morrison. Inequality among world citizens: 1820–1992, The American
Economic Review 92 (2002): 727–744.
Bousquet, F., and A. Voinov. Editorial Thematic Issue – Modelling with Stakeholders. Envir-
onmental Modelling & Software 25 (2010): 1267.
Bouwman, L., T. Kram, and K. Klein Goldewijk, eds. Integrated modelling of global envir-
onmental change – An overview of IMAGE 2.4. MNP Report 500110002/2006. Bilthoven,
2006. Available at www.rivm.nl/bibliotheek/rapporten/500110002.pdf.
Bouwman, F., A. Beusen, and G. Billen. Human alteration of the global nitrogen and phos-
phorus soil balances for the period 1970–2050, Global Biogeochemical Cycles 23 (2009):
GB0A04.
Brander, J., and M. Taylor. The simple economics of Easter Island: A Ricardo-Malthus model
of renewable resource use, The American Economic Review, 88 (1998): 119–138.
Braudel, F. Civilization & Capitalism 15th–18th Century. New York: Harper & Row, 1979.
Brede, M., and B. de Vries. Networks that optimize a trade-off between efficiency and dynam-
ical resilience, Physics Letters A 373 (2009a): 3910–3914.
Brede, M., and B. de Vries. Harvesting heterogeneous renewable resources: Uncoordinated,
selfish, team-, and community-oriented strategies, Ecological Modelling 25 (2009b): 117–
128.
Brock W., and S. Durlauf. Discrete choice with social interactions, Review of Economic Studies
68 (2001): 235–260.
Brown, L. Outgrowing the Earth – The Food Security Challenge in an Age of Falling Water
Tables and Rising Temperatures. New York: W.W. Norton & Company, 2004.
Bryson, B. A Short History of Nearly Everything. New York: Broadway Books, 2003.
Buchanan, M. Nexus/Small Worlds and the Groundbreaking Science of Networks. New York:
W.W. Norton & Company, 2002.
Buckingham/USGS. Aluminium stocks in use in automobiles in the United States.
United States Geological Survey (USGS), 2010. Available at pubs.usgs.gov/fs/2005/3145/
fs2005_3145.pdf.
Buringh, P. The land resource for agriculture, Philosophical Transactions of the Royal Society
of London B310 (1985): 151–159.
Capra, F. The Hidden Connections – Integrating the Biological, Cognitive, and Social Dimen-
sions of Life into a Science of Sustainability. New York: Doubleday, 2002.
Carpenter, S., and W. Brock. Rising variance: A leading indicator of ecological transition,
Ecology Letters 9 (2006): 311–318.
Case, T. An Illustrated Guide to Theoretical Ecology. Oxford: Oxford University Press, 2000.
564 References
Castells, M. The Rise of the Network Society. Oxford: Blackwell Publishers, 1996.
Castells, M. The Power of Identity. Oxford: Blackwell Publishers, 1997.
Cavana, R., and A. Ford. Environmental and resource systems, System Dynamics Review
(Special Issue) 20 (2004): 2.
Cavane, R., P. Davies, R. Robson, and K. Wilson. Drivers of quality in health services:
Different worldviews of clinicians and policy managers revealed, System Dynamics Review
15 (1999): 315–340.
Chaisson, E. Cosmic Evolution – The Rise of Complexity in Nature. Cambridge, MA: Harvard
Chakrabarti, D. K. The Archaeology of Ancient Indian Cities. Delhi: Oxford University Press,
1997.
Chalmers, R. What Is This Thing Called Science? (3rd ed.). Berkshire, UK: Open University
Press, 1999.
Chisholm, M. The increasing separation of production and consumption. In The Earth as
Transformed by Human Action – Global and Regional Changes in the Biosphere over the Past
300 Years, pp. 87–102, edited by B. Turner, W. Clark, R. Kates, J. Richards, J. Mathews, and
W. Meyer. Cambridge: Cambridge University Press, 1990.
Cirera, X., and E. Masset. Income distribution trends and future food demand, Philosophical
Transactions of the Royal Society B 365 (2010): 2821–2834.
Claassen, R. Het eeuwig tekort. Amsterdam: Ambo, 2004.
Clark, C. Mathematical Bioeconomics – The Optimal Management of Renewable Resources.
Wiley Interscience, New York, 1990.
Clark, N., F. Perez-Trejo, and P. Allen. Evolutionary Dynamics and Sustainable Development:
A Systems Approach. Cheltenham, UK: Edward Elgar, 1995.
Clark, W., and N. Dickson. Sustainability science: The emerging research program, Proceed-
ings of the National Academy of Sciences of the United States of America (PNAS) 100 (2003):
8059–8061.
Costanza, R., B. Low, E. Ostrom, and J. Wilson, eds. Institutions, Ecosystems, and Sustainab-
ility. Ecological Economics Series. London: CRC Press, 2001.
Costanza, R., L. Graumlich, and W. Steffen, eds. Sustainability or Collapse? An Integrated
History and Future of People on Earth (IHOPE). 96th Dahlem Workshop. Boston: MIT
Press, 2007.
Costanza, R., L. Graumlich, W. Steffen, C. Crumley, J. Dearing, K. Hibbard, et al. Sustainab-
ility or collapse: what can we learn from integrating the history of humans and the rest of
nature? Ambio 36, (2007): 522–527.
Crosby, A. Ecological Imperialism – The Biological Expansion of Europe, 900–1900. Cam-
bridge: Cambridge University Press, 1993.
Daly, H. Steady-State Economics: The Economics of Biophysical Equilibrium and Moral
Growth. San Francisco: W. H. Freeman, 1977.
Daly, H., and J. Cobb. For the Common Good – Redirecting the Economy toward Community,
the Environment, and a Sustainable Future. Boston: Beacon Press, 1989.
Daly, H. Beyond Growth. Boston: Beacon Press, 1996.
Daré, W., R. Ducrot, A. Botta, and M. Etienne. Repères méthodologiques pour la mise
en oeuvre d’une demarche de modélisation d’accompagnement. ComMod www.cardere.fr
(2009).
Dasgupta, P. S., and G. W. Heal. Exhaustible Resources and Economic Theory. Cambridge:
Dasgupta, P. Economics – A Very Short Introduction. Oxford: Oxford University Press, 2007.
Dasgupta, P. Discounting climate change, Journal of Risk Uncertainty 37 (2008): 141–169.
de Geus, A. Modelling to predict or to learn? In Modelling for learning (special issue), pp.
1–5, edited by J. Morecroft and J. Sterman. European Journal of Operational Research 59
(1992): 1.
de Geus, M. The End of Over-Consumption: Towards a Lifestyle of Moderation and Self-
Restraint. Utrecht, The Netherlands: International Books, 2003.
References 565
de Kok, J., and H. Wind. Designing rapid assessment models of water systems based on
internal consistency, Journal of Water Resources Planning and Management, 128 (2002):
240–247.
de Roos, A., and L. Persson. Unstructured population models: Do population-level assump-
tions yield general theory? In Dynamic Food Webs – Multispecies Assemblages, Ecosystem
Development and Environmental Change, pp. 31–62, edited by P. de Ruiter, V. Wolters,
and J. Moore. Theoretical Ecology Series. Amsterdam: Elsevier, 2005.
de Ruiter, P., V. Wolters, and J. Moore, eds. Dynamic Food Webs – Multispecies Assemblages,
Ecosystem Development and Environmental Change. Theoretical Ecology Series. Amster-
dam: Elsevier, 2005.
de Vries, B., K. Uitham, and G. J. Zijlstra. Kernenergie in Nederland, Acta Politica 2 (1977):
161–206.
de Vries, B. Perspectieven op duurzame ontwikkeling, Lucht en Omgeving 6 (1989): 168–172.
de Vries, B. Effects of resource assessments on optimal depletion estimates, Resources Policy
15 (1989): 253–268.
de Vries, B., and R. Benders. Electric power planning in a gaming context, Simulation and
Games 20 (1989): 227–244.
de Vries, B. SusClime – a simulation game on population and development in a resource- and
climate-constrained two-country world, Simulation & Games 29 (1998): 216–238.
de Vries, B. Perceptions and risks in the search for a sustainable world – a model-based
approach, International Journal of Sustainable Development 4 (2001): 434–453.
de Vries, B., and J. Goudsblom, eds. Mappae Mundi – Humans and Their Habitats in a Long-
Term Socio-ecological Perspective. Myths, Maps, and Models. Amsterdam: Amsterdam
de Vries, B., M. Thompson, and K. Wirtz. Understanding: Fragments of a unifying per-
spective. In Mappae Mundi – Humans and Their Habitats in a Long-Term Socio-ecological
Perspective. Myths, Maps, and Models, pp. 257–300, edited by B. de Vries and J. Goudsblom.
Amsterdam: Amsterdam University Press, 2002.
de Vries, B. Scenarios: Guidance for an uncertain and complex world? In Sustainability or
Collapse? An Integrated History and Future of People on Earth (IHOPE) – 96th Dahlem
Workshop, pp. 378–398, edited by R. Costanza, L. Graumlich, and W. Steffen. Boston: MIT
Press, 2007.
de Vries, B., A. Revi, G. Bhat, H. Hilderink, and P. Lucas. India 2050: Scenarios for an uncer-
tain future. MNP Report. 2007. 550033002/2007. Available at www.pbl.nl/en/publications/
2007/India2050 scenariosforanuncertainfuture
de Vries, B. Environmental modelling for a sustainable world. In Principles of Environmental
Sciences, pp. 345–373, edited by J. Boersema and L. Reijnders. Netherlands: Springer
Science +Business Media B.V., 2009.
de Vries, B., and A. Petersen. Conceptualizing sustainable development: An assessment meth-
odology connecting values, knowledge, worldviews and scenarios, Ecological Economics
68 (2009): 1006–1019.
de Vries, B. Interacting with complex systems: Models and games for a sustainable eco-
nomy. PBL Background Studies Report number 550033003. 2010. Available at www.
globalsystemdynamics.eu; www.pbl.nl.
de Waal, F. Primates and Philosophers – How Morality Evolved. Princeton, NJ: Princeton
Dearing, J. Climate-human-environment interactions: Resolving our past, Climate of the Past
2 (2006): 187–203.
Demeny, P., and G. McNicoll. Population and Development Review 37 Special Volume on
Demographic Transition (2011).
Dennett, D. Kinds of Minds – The Origins of Consciousness. London: Orion Books, 1997/2001.
den Elzen, M., N. Höhne, and S. Moltmann. The Triptych approach revisited: A staged
sectoral approach for climate mitigation, Energy Policy 36 (2008): 1107–1124.
Devara, Dasimayya, in: Speaking of Siva (Translated with an introduction by A.K. Ramanu-
jan). London: Penguin Books. London.
566 References
Diamond, J. Collapse – How Societies Choose to Fail or Succeed. New York: Penguin Books,
2006.
Dicken, P. Global Shift – Mapping the Changing Contours of the World Economy (5th ed.).
London: Sage Publications, 2009.
Dietz, S., and E. Neumayer. Genuine savings: A critical analysis of its policy-guiding value,
Int. J Environment and Sustainable Development 3 (2004): 276–292.
Dietz, T., A. Fitzgerald, and R. Shwom. Environmental values, Annual Review of Environment
and Resources 30 (2005): 335–372.
Dirzo, R., and P. Raven. Global state of biodiversity and loss, Annual Review of Environment
and Resources 28 (2003): 137–167.
Disperati, P., J. van de Steeg, P. van Breugel, A. Notenbaert, J. Owuor, and M. Herreror.
Environmental Security and Pastoralism. International Livestock Research institute (ILRI)
Report, Nairobi, 2009.
Döpfer, K., ed. The Evolutionary Foundations of Economics. Cambridge: Cambridge Univer-
sity Press, 2005.
Dosi, G., G. Fagiolo, and A. Roventini. The microfoundations of business cycles: An evolu-
tionary, multi-agent model, Journal of Evolutionary Economics 18 (2008): 413–432.
Doucet, P., and P. Sloep. Mathematical Modeling in the Life Sciences. New York: Ellis Hor-
wood, 1992.
Douglas, M., D. Gasper, S. Ney, and M. Thompson. Human needs and wants. In Human Choice
and Climate Change. Vol. 1: The Societal Framework, pp. 195–264, edited by S. Rayner and
E. Malone. Columbus, OH: Battelle Press, 1998.
Duchin, F., and G.-M. Lang. The Future of the Environment. Oxford: Oxford University Press,
1994.
Dudley, R. A basis for understanding fishery management dynamics, System Dynamics Review
24 (2008): 1–29.
Duke, D., and J. Geurts. Policy Games for Strategic Management: Pathways into the Unknown.
Amsterdam: Dutch University Press, 2004.
Dumont, L. Homo Hierarchicus: The Caste System and Its Implications. Chicago: University
of Chicago Press, 1984.
Dunham, J. An agent-based spatially explicit epidemiological model in MASON,
Journal of Artificial Societies and Social Simulation (JASSS) 9 (2005). Available at
jasss.soc.surrey.ac.uk/9/1/3.html.
Dunne, J., R. Williams, and N. Martinez. Food-web structure and network theory: The role of
connectance and size, Proceedings of the National Academy of Sciences of the United States
of America (PNAS) 99 (2002): 12917–12922.
Earle, C. The myth of the southern soil miner: Macrohistory, agricultural innovation, and
environmental change. In The Ends of the Earth – Perspectives on Modern Environmental
History. Studies in Environment and History, edited by D. Worster. Cambridge: Cambridge
Easterlin, R. Modernization and fertility. In Determinants of Fertility in Developing Countries
(Vol. 2), edited by R. Bulatao and R. Lee. London: Academic Press, 1983.
Easton, T. Taking Sides: Clashing Views on Environmental Issues. Contemporary Learning
Series (11th ed.). New York: McGraw-Hill, 2006.
Easwaran, E. The Bhagavad Gita. London: Arkana Penguin Books, 1985.
Edelstein-Keshet, L. Mathematical Models in Biology. New York: Random House, 1988.
European Environment Agency (EEA). The European Environment – Synthesis. Copenha-
gen: European Environment Agency, 2010. Available at eea.europa.eu.
EEA. Late Lessons from Early Warnings: The Precautionary Principle 1896–2000. Copenha-
gen, European Environmental Agency, 2001.
Elgin, D. Awakening Earth: Exploring the Co-evolution of Human Culture and Consciousness.
New York: Morrow, 1993.
Elias, N. The Civilizing Process. Sociogenetic and Psychogenetic Investigations. Oxford:
Blackwell, 1967/2000.
References 567
Ellis, E., and N. Ramankutty. Putting people in the map: Anthropogenic biomes of the world,
Frontiers in Ecology and the Environment 6 (2008): 439–447.
Ellis, E., K. Klein Goldewijk, S. Siebert, D. Lightman, and N. Ramankutty. Anthropogenic
transformation of the biomes, 1700 to 2000, Global Ecology and Biogeography 19 (2010):
589–606.
Elvin, M. Three thousand years of unsustainable growth: China’s environment from archaic
times to the present, East Asian History 6 (1993): 7–46.
Engelen, G., R. White, and I. Uljee. Exploratory modeling of socio-economic impacts of
climate change. In Climate Change in the Intra-Americas Sea, edited by G. Maul. London:
Edward Arnold, 1993.
Engelen, G., R. White, I. Uljee, and S. Wargnies. Numerical modeling of small island small
socio-economics to achieve sustainable development. In Small Islands: Marine Science and
Sustainable Development, Coastal and Estuarine Studies Volume 51, pp. 437–463, edited by
G. Maul. Washington, DC: American Geophysical Union, 1996.
Engelen, G., R. White, I. Uljee, and P. Drazan. Using cellular automata for integrated model-
ling of socio-environmental systems, Environmental Monitoring and Assessment 34 (1995):
203–214.
Engelen, G. Models in policy formulation and assessment: The WADBos decision-support
system. In Environmental Modelling: Finding Simplicity in Complexity, pp. 257–271, edited
by J Wainwright and M. Mulligan. Environmental Modelling – Finding Simplicity in Com-
plexity. London: John Wiley& Sons, Ltd., 2004.
Epstein, J., and R. Axtell, eds. Growing Artificial Societies. Washington, DC/Cambridge, MA:
Brookings Institution Press/MIT Press, 1995.
Eriksson, A., and K. Lindgren. Cooperation in an unpredictable environment. In Artificial
Life VIII, edited by R. Standish, M. Abbass, and H. Bedau. Boston: MIT Press, 2002.
Etienne, M., C. Le Page, and M. Cohen. A step-by-step approach to building land man-
agement scenarios based on multiple viewpoints on multi-agent system simulations,
Journal of Artificial Societies and Social Simulation (JASSS) 6 (2003). Available at
jasss.soc.surrey.ac.uk/6/2/2.html.
Etienne, M., M. Cohen, and C. Le Page. A step-by-step approach to build-up land manage-
ment scenarios based on multiple viewpoints on multi-agent system simulations. Journal of
Artificial Societies and Social Simulation (JASSS) 7 (2003):
Falkenmark, M. and Lindh, G. How can we cope with the water resources situation by the
year 2015? Ambio 3 (1976): 114–122.
Falkenmark, M., and J. Rockström. The new blue and green water paradigm, Journal of Water
Resources Planning and Management. (2006): 129–132.
Fay, T., and J. Greeff. Lion, wildebeest and zebra: A predator-prey model, Ecological Mod-
elling 196 (2006): 237–244.
Feinstein, C., and M. Thomas. Making History Count – A Primer in Quantitative Methods for
Historians. Cambridge: Cambridge University Press, 2002.
Ferber, J. Multi-Agent Systems – An Introduction to Distributed Artificial Intelligence. London:
Ferber, J. Multi-agent concepts and methodologies. In Agent-Based Modelling and Simulation
in the Social and Human Sciences, pp. 7–34, edited by D. Phan and F. Amblard. Oxford:
Bardwell Press, 2007.
Fernández, J., and M. Selma. The dynamics of water scarcity on irrigated landscapes: Mazarrón
and Aguilas in south-eastern Spain, System Dynamics Review 20 (2004): 117–137.
Feynman, J., and A. Ruzmaikin. Climate stability and the development of agricultural societ-
ies, Climatic Change 84 (2007): 295–311.
Finlayson, A., and B. McCay. Crossing the threshold of ecosystem resilience: The commercial
extinction of cod. In Linking Social and Ecological Systems – Management Practices and
Social Mechanisms for Building Resilience, pp. 311–337, edited by F. Berkes and C. Folke.
568 References
Fischer, G., M. Shah, H. van Velthuizen, and F. Nachtergaele. Global Agro-ecological

Assessment for Agriculture in the 21st Century. IIASA/FAO Report, 2001. Available
at www.iiasa.ac.at/Admin/PUB/Documents/XO-01-001.pdf.
Fischer-Kowalski, M., and H. Haberl, eds. Socioecological Transitions and Global Change:
Trajectories of Social Metabolism and Land Use. Cheltenham, UK: Edward Elgar, 2007.
Fisher, J., and R. Pry. A simple substitution model of technological change, Technological
Forecasting and Social Change 3 (1971): 75–88.
Flannery, K. V. The cultural evolution of civilizations, Annual Review of Ecology and Sys-
tematics 3 (1972): 399–426.
Florax, R., and H. de Groot. Meta-analyse als hulpmiddel bij beleidsinstrumentatie, ESB 90
(2005): D8–D10.
Folmer, H., H. Landis Gabel, and H. Opschoor. Principles of Environmental and Resource
Economics – A Guide for Students and Decision-Makers. Cheltenham, UK: Edward Elgar,
1995.
Ford, A. Modeling the Environment. Washington, DC: Island Press, 2009.
Ford, J., and R. Myers. A global assessment of salmon aquaculture impacts on wild
salmonids, PloS Biology (2008). Available at www.lenfestocean.org/publications/plbi-
06–02-07 Ford1REV.pdf.
Forrester, J. World Dynamics (2nd ed.). Cambridge, MA: Wright-Allen Press, 1973.
Frenken, K. Innovation, Evolution and Complexity Theory. Cheltenham, UK: Edward Elgar,
2006.
Frondel, M., and C. Schmidt. The empirical assessment of technology differences: Comparing
the comparable, The Review of Economics and Statistics 88 (2006): 186–192.
Fukuyama, F. The End of History and the Last Man. New York: Avon Books, Inc., 1992.
Funtowicz, S., and J. Ravetz. Uncertainty and Quality in Science for Policy. Dordrecht, the
Netherlands: Kluwer Academic Publishers, 1990.
Gales, B., A. Kander, P. Malanima, and M. Rubio. North versus South: Energy transition
and energy intensity in Europe over 200 years, European Review of Economic History 11
(2007): 219–253.
Galor, O., and D. Weil. Population, technology, and growth: From Malthusian stagnation
to the demographic transition and beyond, The American Economic Review, 90 (2000):
806–828.
Garcia, S., and R. Grainger. Gloom and doom? The future of marine capture fisheries,
Philosophical Transactions of the Royal Society B 360 (2005): 21–46.
Garcia, S., and A. Charles. Fishery systems and linkages: From clockworks to soft watches,
ICES Journal of Marine Science 64 (2007): 580–587.
Garcia, S., and A. Rosenberg. Food security and marine capture fisheries: Characteristics,
trends, drivers and future perspectives, Philosophical Transactions of the Royal Society B
365 (2010): 2869–2880.
Gardner, D. Future Babble: Why Expert Predictions Are Next to Worthless, and You Can Do
Better. New York: Dutton, 2010.
Geels, F. Technological transitions as evolutionary reconfiguration processes: A multi-level
perspective and a case-study, Research Policy 31 (2002): 1257–1274.
Geels, F. Co-evolution of technology and society: The transition in water supply and per-
sonal hygiene in the Netherlands (1850–1930) – A case study in multi-level perspective,
Technology in Society 27 (2005): 363–397.
Geels, F., and J. Schot. Typology of sociotechnical transition pathways, Research Policy 36
(2007): 399–417.
Geist, H., and E. Lambin. What drives tropical deforestation? IHDP-IGBP LUCC Report
Series No. 4. Available at www.pik-potsdam.de/members/cramer/teaching/0607/Geist
2001 LUCC Report.pdf.
Geist, H., ed. Our Earth’s Changing Land – An Encyclopedia of Land-Use and Land-Cover
Change. Westport, CT/London: Greenwood Press, 2006.
Giampietro, M. Multi-scale Integrated Analysis of Agroecosystems. Advances in Agroecology.
Boca Raton, FL: CRC Press, 2004.
References 569
Gilbert, N., and J. Doran, eds. Simulating Societies – The Computer Simulation of Social
Phenomena. London: UCL Press, 1994.
Gilbert, N., and K. Troitzsch. Simulation for the Social Scientist. Buckingham, UK: Open
Gilding, P. The Great Disruption: Why the Climate Crisis Will Bring On the End of Shopping
and the Birth of a New World. New York: Bloomsbury Press, 2011.
Gilg, O., I. Hanski, and B. Sittler. Cyclic dynamics in a simple vertebrate predator-prey
community, Science 302 (2003): 866–868.
Gintis, H. Beyond Homo economicus: Evidence from experimental economics, Ecological
Economics Special Issue 35 (2000): 311–322.
Gintis, H. Behavioral game theory and contemporary economic theory, Analyse & Kritik 27
(2005): 48–72.
Giordano, M. Global groundwater? Issues and solutions, Annual Review of Environment and
Resources 34 (2009): 153–178.
Gleick, P., and associates. The World’s Water: The Biennial Report on Freshwater. Washing-
ton, DC: Resources. Island Press, 2011.
Gong, M., and G. Wall. On exergy and sustainable development – Part 2: Indicators and
methods, Exergy, an International Journal 4 (2001): 217–233.
Gordon, H. The economic theory of a common-property resource: The fishery, The Journal
of Political Economy, 62 (1954): 124–142.
Gordon, L., G. Peterson, and E. Bennett. Agricultural modifications of hydrological flows
create ecological surprises, Trends in Ecology and Evolution 4 (2008): 211–219.
Goudie, A. The Human Impact – Man’s Role in Environmental Change. Cambridge, MA:
MIT Press, 1981.
Goudsblom, J. Fire and Civilization. London: Penguin, 1992.
Goudsblom, J., E. Jones, and S. Mennell. Course of Human History: Economic Growth, Social
Process and Civilization. Armonk, NY: M. E. Sharpe, 1996.
Goudsblom, J. The past 250 years: Industrialization and globalization. In Mappae Mundi –
Humans and Their Habitats in a Long-Term Socio-ecological Perspective. Myths, Maps, and
Models, pp. 353–378, edited by B. de Vries and J. Goudsblom. Amsterdam: Amsterdam
Gourou, P. Les pays tropicaux. Paris: Presses Universitaires de France, 1947.
Grazi, F., J. Van den Bergh, and P. Rietveld. Spatial welfare economics versus ecological
footprint: Modeling agglomeration, externalities and trade, Environmental Resource Eco-
nomics 38 (2007): 138–153.
Greenblat, C. Designing Games and Simulations: An Illustrated Handbook. London: Sage
Ltd., 1988.
Grigg, D. The Agricultural Systems of the World. Cambridge: Cambridge University Press,
1974.
Grigg, D. The industrial revolution and land transformation. In Land Transformation in
Agriculture, edited by M. Wolman and F. Fournier. New York: John Wiley, 1987.
Grimm, V., and H. Wissel. Babel, or the ecological stability discussions: An inventory and
analysis of terminology and a guide for avoiding confusion, Oecologia 109 (1997): 323–334.
Groenewold, H. Evolutie van kennis, waarden en macht. Technische Universiteit Twente,
1981.
Grove, A., and O. Rackham. The Nature of Mediterranean Europe – An Ecological History.
New Haven and London: Yale University Press, 2001.
Grübler, A. Technology and Global Change. Cambridge: Cambridge University Press, 1999.
Grübler, A., and N. Nakicenovic. Long Wave, Technology Diffusion, and Substitution. Lax-
enburg, Austria: IIASA, 1991.
Gunderson, L., and C. Holling. Panarchy – Understanding Transformations in Human and
Natural Systems. Washington, DC: Island Press, 2002.
Haas, P., R. Keohane, and M. Levy, eds. Institutions for the Earth. Cambridge, MA: MIT
Press, 1993.
Haberl, H. Human appropriation of net primary production as an environmental indicator:
Implications for sustainable development, Ambio 26 (1997): 143–146.
570 References
Haddeland, I. Anthropogenic Impacts on the Continental Water Cycle. Ph.D. Thesis. Oslo,
Norway: University of Oslo, 2006.
Haley, S. Capital accumulation and the growth of aggregate agricultural production, Agricul-
tural Economics 6 (1991): 129–157.
Hall, S., D. Held, and K. Thompson. Modernity: An Introduction to Modern Societies. Bodmin,
Cornwall: MPG Books, 1996.
Handy, C. The Hungry Spirit – Beyond Capitalism – A Quest for Purpose in the Modern World.
London: Arrow Books, 1998.
Hanjra, M., and M. Qureshi. Global water crisis and future food security in an era of climate
change, Food Policy 35 (2010): 365–377.
Hardin, G. The tragedy of the commons, Science 162 (1968): 1243–1248.
Harman, W. An Incomplete Guide to the Future. Stanford, CA: Stanford Alumni Association,
1976.
Harris, D. Mineral Resource Appraisal. Oxford: Clarendon Press, 1984.
Harte, J. Toward a synthesis of the Newtonian and Darwinian worldviews, Physics Today 10
(2002): 29–34.
Hassan, F. Environmental perception and human responses in history and prehistory. In
The Way the Wind Blows: Climate, History, and Human Action, pp. 121–140, edited by
R. McIntosh, J. Tainter, and S. McIntosh. New York: Columbia University Press,
2000.
Hedlund-de Witt, A. The rising culture and worldview of contemporary spirituality: A soci-
ological study of potentials and pitfalls for sustainable development, Ecological Economics
70 (2011): 1057–1065.
Heijden, K. van der. Scenarios – The Art of Strategic Conversation. New York: Wiley & Sons,
1996.
Heilbroner, R. The Worldly Philosophers. New York: Simon & Schuster, 1955.
Heilbroner, R., and W. Milberg. The Making of Economic Society (12th ed.). New York:
Pearson Prentice-Hall, 2005.
Hein, P., C. van Kopen, R. de Groot, and E. van Ierland. Spatial Scales, Stakeholders and the
Valuation of Ecosystem Services. Wageningen, the Netherlands: Wageningen University,
2005. Available at dx.doi.org/10.1016/j.ecolecon.2005.04.005.
Hein, L. Optimising the Management of Complex Dynamic Ecosystems. Ph.D. Thesis. Wagen-
ingen, the Netherlands: Wageningen University, 2005.
Hein, L., and E. van Ierland. Efficient and sustainable management of complex forest ecosys-
tems, Ecological Modeling 190 (2006): 351–336.
Helbing, D., and W. Yu. The outbreak of cooperation among success-driven individuals under
noisy conditions, Proceedings of the National Academy of Sciences of the United States of
America (PNAS) 106 (2009): 3680–3685.
Helpman, E. The Mystery of Economic Growth. Boston: Harvard University Press, 2004.
Hettelingh, J.-P., B. de Vries, and L. Hordijk. Integrated assessment. In Principles of Envir-
onmental Sciences, pp. 385–420, edited by J. Boersema and L. Reijnders. The Netherlands:
Hilderink, H. World Population in Transition – An Integrated Regional Modelling Frame-
work. Faculteit der Ruimtelijke Wetenschappen. Ph. D. Thesis. Groningen, the Netherlands:
Rijksuniversiteit Groningen, 2000.
Hilderink, H., P. Lucas, and M. Kok, eds. Beyond 2015: Long-term Development and the Mil-
lennium Development Goals. Report 55002500. Bilthoven, the Netherlands: Netherlands
Environmental Assessment Agency (PBL), 2009.
Hirsch, F. Social Limits to Growth. London: Routledge & Kegan Paul, 1977.
Hirsch, R. Mitigation of maximum world oil production: Shortage scenarios, Energy Policy
36 (2008): 881–889.
Hisschemöller, M., J. Eberg, A. Engels, and K. von Moltke. Environmental institutions and
learning: Perspectives from the policy sciences. In Principles of Environmental Sciences,
pp. 281–303, edited by J. Boersema and L. Reijnders. The Netherlands: Springer Science
+Business Media B.V., 2009.
References 571
Hitlin, S., and J. Piliavin. Values: reviving a dormant concept, Annual Review of Sociology 30
(2004): 359–393.
Ho, P., and E. Vermeer. China’s Limits to Growth – Greening State and Society. Oxford:
Blackwell Publishing, 2006.
Hodgson, G. Decomposition and growth: Biological metaphors in economics from the 1880s
to the 1980s. In The Evolutionary Foundations of Economics, pp. 105–148, edited by
K. Döpfer. Cambridge: Cambridge University Press, 2005.
Hoekstra, R., and J. van den Bergh. Structural decomposition analysis of physical flows in the
economy, Environmental and Resource Economics 23 (2002): 357–378.
Hofstede, G. Cultural dimensions in management and planning, Asia Pacific Journal of Man-
agement 1 (1984): 81–99.
Holland, J. Hidden Order – How Adaptation Builds Complexity. Reading, MA: Helix Books/
Holling, C. Resilience and stability of ecological systems, Annual Review of Ecological Systems
4 (1973): 1–23.
Holling, C. The resilience of terrestrial ecosystems: Local surprise and global change. In The
Resilience of Terrestrial Ecosystems: Local Surprise and Global Change, edited by W. Clark
and R. Munn. Cambridge: Cambridge University Press/IIASA, 1986.
Hollis, M. The Philosophy of Science – An Introduction. Cambridge: Cambridge University
Press, 2007.
Homer-Dixon, T. The Upside of Down: Catastrophe, Creativity, and the Renewal of Civiliza-
tion. Washington, DC: Island Press, 2006.
Hueting, R. New Scarcity and Economic Growth. Amsterdam: North-Holland Publishing
Company, 1980.
Hulme, M. Why We Disagree About Climate Change. Cambridge: Cambridge University Press,
2009.
IEA (2004). See www.iea.org
IEA (2010). See www.iea.org
IFPRI (2008). See www.ifpri.org
Inglehart, R., and C. Welzel. Modernization, Cultural Change and Democracy: The Human
Development Sequence. New York: Cambridge University Press, 2005.
INFO Project. 2003. See Project Publications at info.k4health.org/about.shtml.
IPCC. Climate Change 2001 (TAR). Cambridge: Cambridge University Press, 2001.
IPCC. Climate Change 2007 (AR4). Cambridge: Cambridge University Press, 2007.
Ironmonger, D., and E. Sonius. Household productive activities. In Households Work, edited
by D. Ironmonger. Sydney: Allen & Unwin, 1989.
IUCN-WWF (1991). Caring for the world: a strategy for sustainability. IUCN-UNEP-WWF
Report. See www.iucn.org
Izard, C., and D. Müller. Tracking the devil’s metal: Historical global and contemporary U.S.
tin cycles, Resources, Conservation and Recycling 54 (2010): 1436–1441.
Jackson, T. Prosperity without Growth – Economics for a Finite Planet. London: Earthscan,
2009.
Jager, W., M. Janssen, B. de Vries, J. de Greef, and C. Vlek. Behaviour in commons dilemmas:
Homo Economicus and Homo Psychologicus in an ecological-economic model, Ecological
Economics 35 (2000): 357–379.
Jager, W., and H.-J. Mosler. Simulating human behavior for understanding and managing
environmental resource use, Journal of Social Issues 63 (2007): 97–116.
Jaggard, K., A. Qi, and E. Ober. Possible changes to arable crop yields, Philosophical Trans-
actions of the Royal Society B 365 (2010): 2835–2851.
James, W. The Varieties of Religious Experience – A Study in Human Nature (The Gifford
Lectures Edinburgh 1901–02). Glasgow: Collins Fountain Books, 1902/1977.
Janssen, M., ed. Complexity and Ecosystem Management – The Theory and Practice of Multi-
agent Systems. Cheltenham, UK: Edward Elgar, 2002.
Janssen, M., T. Kohler, and M. Scheffer. Sunk-cost effects and vulnerability to collapse in
ancient societies, Current Anthropology 44 (2003): 722–728.
572 References
Janssen, P., P. Heuberger, and A. Tiktak. Metamodelleren bij het MNP-RIVM. MNP-Report,
Bilthoven, 2005.
Jantsch, E. The Self-organizing Universe. Oxford: Pergamon Press, 1980.
Jarvis, P. Ecological Principles and Environmental Issues. Essex, UK: Prentice Hall/Pearson
Education, 2000.
Jeong, H., B. Tombor, R. Albert, Z. Oltvai, and A. Barabási. The large-scale organization of
metabolic networks, Nature 407 (2000): 651–654.
Jones, E. The European Miracle (2nd ed.). Cambridge: Cambridge University Press,
1981/2003.
Jung, C.G. The Nature of the Psyche. Princeton, NJ: Princeton University Press, 1960.
Junginger, M., W. van Sark, and A. Faaij, eds. Technological Learning in the Energy Sector.
Cheltenham, UK: Edward Elgar, 2010.
Kahneman, D., E. Diener, and N. Schwarz, eds. Well-Being – The Foundations of Hedonic
Psychology. New York: Russell Sage Foundation, 1999.
Kates, R., W. Clark, R. Corell, J. Hall, C. Jaeger, I. Lowe, et al. Sustainability science, Science
292 (2001): 641–642. Available at dx.doi.org/10.1126/science.1059386.
Kauffman, S. At Home in the Universe – The Search for Laws of Complexity. London: Viking,
1995.
Kay, J. Culture and Prosperity: The Truth about Markets – Why Some Nations Are Rich but
Most Remain Poor. New York: HarperBusiness, 2004.
Kearney, J. Food consumption trends and drivers, Philosophical Transactions of the Royal
Society B 365 (2010): 2793–2807.
Kefi, S., M. Rietkerk, C. L. Alados, Y. Pueyo, V. P. Papanastasis, A. ElAich, and P. C.
de Ruiter. Spatial vegetation patterns and imminent desertification in Mediterranean arid
ecosystems, Nature 449 (2007): 213–218.
King Hubbert, M. The energy resources of the Earth, Scientific American 225 (1971): 60–70.
Kirman, A. Ants, rationality, and recruitment, The Quarterly Journal of Economics 108,
vol. 1 (1993): 137–156.
Klee, R., and T. Graedel. Elemental cycles: A status report on human or natural dominance,
Annual Review of Environment and Resources 29 (2004):69–107.
Knappett, C., T. Evans, and R. Rivers. Modelling maritime interaction in the Aegean Bronze
Age, Antiquity 82 (2008): 1009–1024.
Kok, M., and J. Jäger, eds. Vulnerability of People and the Environment – Challenges and
Opportunities. PBL/UNEP Background Studies, 2007. Available at www.pbl.nl/en.
Kok, M., M. Lüdeke, T. Sterzel, P. Lucas, C. Walter, P. Janssen, and I. de Soysa. Quantitative
Analysis of Patterns of Vulnerability to Global Environmental Change. PBL/PIK/NTNU
Background Studies, 2001. Available at www.pbl.nl/en.
Krausmann, F., S. Gingrich, N. Eisenmenger, K.-H. Erb, H. Haberl, and M. Fischer-Kowalski.
Growth in global materials use, GDP and population during the 20th century, Ecological
Economics 68 (2009): 2696–2705.
Kruyt, B., D, van Vuuren, B. de Vries, and H. Groenenberg. Indicators for energy security,
Energy Policy 37, vol. 6 (2009): 2166–2181.
Kuik, O., and H. Verbruggen, eds. In Search of Indicators of Sustainable Development.
Dordrecht, the Netherlands: Kluwer Academic Publishers, 1991.
Kümmel, R., J. Henn, and D. Lindenberger. Capital, labor, energy and creativity: Modeling
innovation diffusion, Structural Change and Economic Dynamics 13 (2002): 415–433.
Lambin, E., H. Geist, and E. Lepers. Dynamics of land-use and land-cover change in tropical
regions, Annual Review of Environment and Resources 28 (2003): 205–241.
Lambin, E., and H. Geist, eds. Land-Use and Land-Cover Change – Local Processes and
Global Impacts. The IGBP Series. Berlin: Springer, 2006.
Lansing, J. Perfect Order: Recognizing Complexity in Bali. Princeton, NJ: Princeton University
Press, 2006.
Lao-Tzu. Tao Te Ching (Introduced by Burton Watson). Indianapolis/Cambridge: Hackett
Publishing Company, 1993.
References 573
Layard, R. Happiness: Lessons from a New Science. London: Allen Lane/The Penguin Press,
2005.
Leemans, R., and A. Kleidon. In Global Desertification: Do Humans Cause Deserts? 88th
Dahlem Workshop, edited by J. Reynolds and D. Stafford Smith. Berlin: Dahlem University
Press, 2002.
Leiserowitz, A., R. Kates, and T. Parris. Sustainability values, attitudes, and behaviors: A
review of multinational and global trends, Annual Review of Environment and Resources
31 (2006): 413–444.
Licker, R., M. Johnston, J. Foley, C. Barford, C. Kucharik, C. Monfreda, and N. Ramankutty.
Mind the gap: How do climate and agricultural management explain the “yield gap” of
croplands around the world? Global Ecology and Biogeography 19 (2010): 769–782.
Lim, M., R. Metzler, and Y. Bar-Yam. Global pattern formation and ethnic/cultural violence,
Science 317 (2007):1540–1544.
Limburg, K., R. O’Neill, R. Costanza, and S. Farber. Complex systems and valuation. Spe-
cial Issue: The Dynamics and Value of Ecosystem Services: Integrating Economic and
Ecological Perspectives, Ecological Economics 41 (2002): 409–420.
Lindgren, K. Evolutionary phenomena in simple dynamics. In: Artificial Life II, SFI Studies
in the Sciences of Complexity, Vol. X, edited by C. Langton, C. Taylor, J. Farmer, and
J. Rasmussen. Redwood City, CA: Addison-Wesley, 1991.
Little, L., and A. McDonald. Simulations of agents in social networks harvesting a resource,
Ecological Modelling 204 (2007): 379–386.
Lovejoy, T., and L. Hannah, eds. Climate Change and Biodiversity. New Haven/London: Yale
University, 2005.
Lowenthal, D. Awareness of human impacts: Changing attitudes and emphases. In The Earth
as Transformed by Human Action – Global and Regional Changes in the Biosphere over
the Past 300 Years, pp. 121–136, edited by B. Turner, W. Clark, R. Kates, J. Richards,
J. Mathews, and W. Meyer. Cambridge: Cambridge University Press, 1990.
Lüdeke, M., G. Petschel-Held, and H. Schellnhuber. Syndromes of global change: The first
panoramic view, GAIA 13 (2004).
Lumley, S., and P. Armstrong. Some of the nineteenth century origins of the sustainability
concept, Environment, Development and Sustainability 6 (2004): 367–378.
Lutz, W., W. Sanderson, and S. Scherbov. The end of world population growth, Nature 412
(2001): 543–545.
Lutz, W., B. O’Neill, and S. Scherbov. Europe’s population at a turning point, Science 299
(2003): 1991–1992.
MacKay, D. Sustainable Energy – Without the Hot Air. UIT Cambridge, 2009. Available at
www.withouthotair.com.
Maddison, A. Monitoring the World Economy. Paris: OECD Development Centre Studies,
1995.
Maddison, A. The World Economy. Vol. 1: A Millennial Perspective; Vol. 2: Historical Statistics.
Paris: OECD Development Centre Studies, 2006.
Madsen, J., J. Ang, and R. Banerjee. Four centuries of British economic growth:
The roles of technology and population, Journal of Economic Growth 15 (2010):
263–290.
Malenbaum, W. World Demand for Raw Materials in 1985 and 2000. New York: McGraw-Hill
Inc., 1978.
Mäler, K.-G., A. Xepapadeas, and A. de Zeeuw. The economics of shallow lakes, Environ-
mental and Resource Economics 26 (2003): 603–624.
Manson, S. Simplifying complexity: A review of complexity theory, Geoforum 32 (2001):
405–414.
Martens, P. Health transitions in a globalising world: Towards more disease or sustained
health? Futures 34 (2002): 635–648.
Maslow, A. Motivation and Personality. New York: Harper, 1954.
Matsutani, A. Shrinking-Population Economics – Lessons from Japan. Tokyo: International
House of Japan, 2006.
574 References
Max-Neef, M. Human Scale Development – Conception, Application and Further Reflections.

London: The Apex Press, 1991.
May, R. Will a large, complex system be stable? Nature 238 (1972): 413–414.
May, R. Simple mathematical models with very complicated dynamics, Nature 261 (1976):
459–467.
May, R. Harvesting natural populations in a randomly fluctuating environment, Science 197
(1977): 463–465.
Maynard Smith, J. Evolution and the Theory of Games. Cambridge: Cambridge University
Press, 1982.
Mazoyer, M., and L. Roudart. Histoire des agricultures du monde. Paris: Éditions du Seuil,
1997.
McGovern, T. The economics of extinction in Norse Greenland. In Climate and History
– Studies in Past Climates and Their Potential Impact on Man, edited by T. Wigley, M.
Ingram, and G. Farmer. Cambridge: Cambridge University Press, 1981.
McNeill, J. Social, economic, and political forces in environmental change: Decadal scale
(1900–2000). In Sustainability or Collapse? An Integrated History and Future of People on
Earth (IHOPE) – 96th Dahlem Workshop, edited by R. Costanza, L. Graumlich, and W.
Steffen. Boston: MIT Press, 2007.
MEA (Millennium Ecosystem Assessment). Ecosystems and Human Well-Being – Our
Human Planet. Synthesis and Summary for Decision Makers. Washington, DC: Island Press,
2005.
Meadows, D. H., D. L. Meadows, J. Randers, and W. Behrens III. The Limits to Growth. New
York: Potomac Associates, 1971.
Meadows, D. H., D. L. Meadows, and J. Randers. Beyond the Limits – Confronting
Global Collapse, Envisioning a Sustainable Future. Post Mills, VT: Chelsea Green Ltd.,
1991.
Meadows, D. Fish Banks Ltd. Durham: University of New Hampshire, 1996.
Meadows, D. H. Indicators and Information Systems for Sustainable Development. Report
to the Balaton Group. Sustainability Institute, 1998.
Meadows, D. L., J. Randers, and D. H. Meadows. Limits to Growth – The 30-Year Update.
Post Mills, VT: Chelsea Green Publishing Company, 2004.
Meadows, D. L. Evaluating past forecasts: Reflections on one critique of The Limits to Growth.
In Sustainability or Collapse? An Integrated History and Future of People on Earth (IHOPE)
– 96th Dahlem Workshop, edited by R. Costanza, L. Graumlich, and W. Steffen. Boston:
MIT Press, 2007.
Meadows, D. H. Thinking in Systems. Post Mills, VT: Chelsea Green Publishing, 2008.
Meinshausen, M. What does a 2◦ C target mean for greenhouse gas concentrations? In Avoid-
ing Dangerous Climate Change, edited by J. Schellnhuber, W. Cramer, N. Nakicenovic, T.
Wigley, and G. Yohe. Cambridge: Cambridge University Press, 2006.
Mertz, O., R. Wadley, U. Nielsen, T. Bruun, C. Colfer, A. de Neergaard, et al. A fresh look
at shifting cultivation: Fallow length an uncertain indicator of productivity, Agricultural
Systems 96 (2008): 75–84.
Metz, B., M. Berk, M. den Elzen, B. de Vries, and D. van Vuuren. Towards an equitable
global climate change regime: Compatibility with Article 2 of the Climate Change
Convention and the link with sustainable development, Climate Policy 2 (2002):
211–230.
Metz, B. Controlling Climate Change. Cambridge: Cambridge University Press, 2009.
Meyer, P., and J. Ausubel. Carrying capacity: A model with logistically varying limits, Tech-
nological Forecasting and Social Change 61 (1999): 209–214.
Migheli, M. Supporting the free and competitive market in China and India: Differences and
evolution over time, Economic Systems 34 (2010): 73–90.
Miles, E., A. Underdal, S. Andresen, J. Wettestad, J. Skjaerseth, and E. Carlin. Environmental
Regime Effectiveness. Cambridge, MA: MIT Press, 2002.
Mitchell, M. Complexity – A Guided Tour. Oxford: Oxford University Press, 2009.
References 575
MNP. Duurzaamheidsverkenning (Sustainability Outlook). Bilthoven, the Netherlands: Neth-

erlands Environmental Assessment Agency (PBL), 2004.
Morecroft, J., and J. Sterman. Modelling for learning (special issue), European Journal of
Operations Research 59 (1992): 1.
Morrison, C., and E. Thorbecke. The concept of agricultural surplus, World Development 18
(1990): 1081–1095.
Moxnes, E. Not only the tragedy of the commons: Misperceptions of bioeconomics, Manage-
ment Science 44 (1998): 1234–1248.
Moyo, D. Dead Aid – Why Aid Is Not Working and How There Is Another Way for Africa.
New York: Farrar, Straus and Giroux, 2009.
Mulder, K. Sustainable Development for Engineers. Sheffield, UK: Greenleaf Publishing,
2006.
Nakicenovic, N., Alcamo, J., Davis, G., de Vries, B., et al. Special Report on Emissions
Scenarios (SRES) for the Intergovernmental Panel on Climate Change (IPCC). Cambridge:
Nelson, R., and S. Winter. An Evolutionary Theory of Economic Change. Cambridge, MA:
Neumann, K., P. Verburg, E. Stehfest, and C. Müller. The yield gap of global grain production:
A spatial analysis, Agricultural Systems 103 (2010): 316–326.
Newman, M. The structure and function of complex networks, SIAM Review 45 (2003): 167–
256.
Niamir-Fuller, M., and M. Turner, eds. Managing Mobility in African Rangelands: The Legit-
imization of Transhumance. London: Intermediate Technology Publications, 1999.
Nicolis, G., and I. Prigogine. Exploring Complexity – An Introduction. New York: Freeman
& Company, 1989.
Nieuwlaar, E., and W. de Ruiter. Toegepaste Thermodynamica en Energieconversies.
Cursuscode GEO2–2212. Utrecht, the Netherlands: Utrecht University, 2010.
Nijs, T. de, R. de Niet, and L. Crommentuijn. Constructing land-use maps of the Netherlands
in 2030, Journal of Environmental Management 72 (2004): 35–42.
Nilsson, S. Experiences of policy reforms of the forest sector in transition and other countries,
Forest Policy and Economics 7 (2005): 831–884.
Norberg, J., and G. Cummings, eds. Complexity Theory for a Sustainable Future. New York:
Columbia University Press, 2008.
Nordhaus, W. Geography and macroeconomics: New data and new findings, Proceedings
of the National Academy of Sciences of the United States of America (PNAS) 105 (2005):
3510–3517.
Nowak, M., and K. Sigmund. Evolutionary dynamics of biological games, Science 303 (2004):
793–799.
Nowak, M. Evolutionary Dynamics – Exploring the Equations of Life. Cambridge, MA:
Nussbaum, M., and A. Sen, eds. The Quality of Life. Oxford: Clarendon Press Oxford, 1993.
Nussbaum, M. Women and Human Development: The Capabilities Approach. Cambridge:
Odling-Smee, J., K. Laland, and M. Feldman. Niche construction – The neglected process in
evolution. Princeton, NJ: Princeton University Press, 2003.
Odum, H. Environment, Power and Society. New York: Wiley, 1976.
OECD, 2008. Available at www.oecd.org.
OECD Factbook, 2007. Available at www.oecd.org
OECD/IEA. World Energy Outlook 2010. IEA Report. Paris: OECD. See www.
worldenergyoutlook.org/.
Okin, G. Toward a unified view of biophysical land degradation processes in arid and semi-
arid lands. In Global Desertification: Do Humans Cause Deserts? 88th Dahlem Workshop,
pp. 95–109, edited by J. Reynolds and D. Stafford Smith. Berlin: Dahlem University Press,
2002.
576 References
Oldeman, L., R. Hakkeling, and G. Sombroek. World Map of the Status of Human-Induced
Soil Degradation (2nd ed.). Wageningen, the Netherlands: ISRIC, 1990.
Opschoor, H. The ecological footprint: Measuring rod or metaphor? Ecological Economics
32 (2000): 363–365.
Ormerod, P. The Butterfly Economy: A New General Theory of Social and Economic Behavior.
London: Basic Books, 1998.
Ostrom, E. Governing the Commons – The Evolution of Institutions for Collective Action.
Ostrom, E. Collective action and the evolution of social norms, Journal of Economic Per-
spectives 14 (2000): 137–158.
Ostrom, E., T. Dietz, N. Dolsak, P. Stern, S. Stonich, and E. Weber, eds. The Drama of
the Commons. Committee on the Human Dimensions of Global Change. Washington, DC:
National Academies Press, 2002.
Ostrom, E., M. Janssen, and J. Anderies. Going beyond panaceas – Special feature, Proceed-
ings of the National Academy of Sciences of the United States of America (PNAS) 104 (2007):
15176–15178.
Ostrom, E. A general framework for analyzing sustainability of social-ecological systems,
Science 325 (2009): 419–422.
Papyrakis, E., and R. Gerlagh. The resource curse hypothesis and its transmission channels,
Journal of Comparative Economics 32 (2004): 181–193.
Papyrakis, E., and R. Gerlagh. Resource windfalls, investment, and long-term income,
Parris, T., and R. Kates. Characterizing a sustainability transition: Goals, targets, trends, and
driving forces, Proceedings of the National Academy of Sciences of the United States of
America (PNAS) 100 (2003): 8068–8073.
Parris, T., and R. Kates. Characterizing and measuring sustainable development, Annual
Review of Energy and the Environment 28 (2003): 559–586.
Parson, E. Protecting the ozone layer. In Institutions for the Earth, pp. 27–73, edited by P. Haas,
R. Keohane, and M. Levy. Cambridge, MA: MIT Press, 1993.
Pauly, D., and M.-L. Palomares. Fishing down marine food web: It is far more pervasive than
we thought, Bulletin of Marine Science 72 (2005): 197–211.
PBL. Growing within Limits. Bilthoven, the Netherlands: Netherlands Environmental Assess-
ment Agency, 2009.
PBL. Assessing an IPCC Assessment: An Analysis of Statements on Projected Regional Impacts
in the 2007 Report. Bilthoven, the Netherlands: Netherlands Environmental Assessment
Agency, 2010a.
PBL. Rethinking Global Biodiversity Strategies. Bilthoven, the Netherlands: Netherlands
Environmental Assessment Agency, 2010b.
PBL. Zure regen – Een analyse van dertig jaar verzuringsproblematiek in Nederland. PBL
Beleidsstudies (Buijsman et al.). Bilthoven, the Netherlands: Netherlands Environmental
Assessment Agency, 2010c.
Pearce, D. Economics of Natural Resources and the Environment. Baltimore: John Hopkins
Pearce, D. An intellectual history of environmental economics, Annual Review of Energy and
the Environment 27 (2002): 57–81.
Pedersen, J., and T. Benjaminsen. One leg or two? Food security and pastoralism in the
Northern Sahel, Human Ecology 36 (2008): 43–57.
Penn, D. The evolutionary roots of our environmental problems: Toward a Darwinian ecology,
The Quarterly Review of Biology 78 (2003): 3.
Perez, P., and D. Batten, eds. Complex Science for a Complex World – Exploring Human
Ecosystems with Agents. Canberra, Australia: ANU Press, 2003.
mental Economics. Harlow, UK: Pearson Education Ltd., 2003.
References 577
Peters, G., R. Andrew, and J. Lennox. Constructing an environmentally-extended multi-

regional input-output table using the GTAP database, Economic Systems Research 23
(2011): 131–152.
Petschel-Held, G., A. Block, M. Cassel-Gintz, J. Kropp, M. Lüdeke, O. Moldenhauer, et al.
Syndromes of global change: A qualitative modelling approach to assist global environ-
mental management, Environmental Modelling and Assessment 4 (1999): 295–314.
Phan, D., and F. Amblard, eds. Agent-Based Modelling and Simulation in the Social and
Human Sciences. Oxford: Bardwell Press, 2007.
Pielke, R. Jr. The Honest Broker. Cambridge: Cambridge University Press, 2007.
Pimm, S. The structure of foodwebs, Theoretical Population Biology 16 (1979): 144–158.
Pine, J., and J. Gilmore. The Experience Economy. Boston: Harvard Business School Press,
1999.
Polimeni, J., K. Mayumi, M. Giampietro, and B. Alcott. The Jevons Paradox and the Myth of
Resource Efficiency Improvements. London: Earthscan, 2009.
Pollock, S. Ancient Mesopotamia. Cambridge: Cambridge University Press, 2001.
Ponting, C. A Green History of the World. London: Penguin Books, 1991.
Proctor, R., and L. Schiebinger. Agnotology: The Making and Unmaking of Ignorance. Palo
Alto, CA: Stanford University Press, 2008.
Pumain, D., ed. Hierarchy in Natural and Social Systems. Berlin: Springer, 2006.
Railsback, S., and V. Grimm. Agent-Based and Individual-Based Modeling: A Practical Intro-
duction. Princeton, NJ: Princeton University Press, 2011.
Ramankutty, N., and J. Foley. Estimating historical changes in global land cover: Croplands
from 1700 to 1992, Global Biogeochemical Cycles 13 (1999): 997–1028.
Ramankutty, N., A. Evan, C. Monfreda, and J. Foley. Farming the planet: I. Geographic
distribution of global agricultural lands in the year 2000, Global Biogeochemical Cycles 22
(2008): doi:10.1029/2007GB002952.
Raquez, P., and E. Lambin. Conditions for a sustainable land use: Case study evidence, Journal
of Land Use Science 1 (2006): 109–125.
Ratnieks, F., and N. Carreck. Clarity on honey bee collapse? Science 327 (2010): 152–153.
Ravetz, J. The No-Nonsense Guide to Science. Oxford: New Internationalist, 2006.
Rayner, S., and E. Malone, eds. Human Choice and Climate Change (4 vols.). Columbus, OH:
Battelle Press, 1998.
Rayo, L., and G. Becker. Evolutionary efficiency and happiness, Journal of Political Economy
115 (2007): 2.
Read, D. W. Kinship based demographic simulation of societal processes, Journal of Artificial
Societies and Social Simulation 1 (1998): 1.
Reader, J. Man on Earth. New York: Perennial Library, 1988.
Redman, C. L. Human Impact on Ancient Environments. Tucson: University of Arizona, 1999.
Reich, W. Supercapitalism. New York: Knopf Publishers, 2007.
Renfrew, C., and J. Cherry, eds. Peer Polity Interaction and Socio-political Change. Cambridge:
Revi, A., N. Dronin, and B. de Vries. Population and environment in Asia since 1600 AD.
In Mappae Mundi – Humans and Their Habitats in a Long-Term Socio-ecological Per-
spective. Myths, Maps, and Models, pp. 301–352, edited by B. de Vries and J. Goudsblom.
Amsterdam: Amsterdam University Press, 2002.
Reynolds, J., and D. Stafford Smith, eds. Global Desertification: Do Humans Cause Deserts?
88th Dahlem Workshop. Berlin: Dahlem University Press, 2002.
Riahi, K., A. Grübler, and N. Nakicenovic. Scenarios of long-term socio-economic and envir-
onmental development under climate stabilization, Technological Forecasting & Social
Change 74 (2007): 887–935.
Richardson, G. P. Feedback Thought in Social Science and Systems Theory. Philadelphia:
University of Pennsylvania Press, 1991.
Rietkerk, M., S. Dekker, P. de Ruiter, and J. van de Koppel. Self-organized patchiness and
catastrophic shifts in ecosystems, Science 305 (2004): 1926–1929.
578 References
RIVM. Zorgen voor Morgen. Nationale milieuverkenning 1985–2010. Bilthoven, the Nether-
lands: RIVM, 1988.
Robert, J. Water Is a Commons. Mexico City: Habitat International Coalition, 1994.
Roberts, N., D. Andersen, R. Deal, M. Garet, and W. Shaffer. Introduction to Computer Sim-
ulation – A System Dynamics Approach. New York: Addison-Wesley Publishing Company,
1983.
Roberts, P. The End of Food. London: Bloomsbury, 2008.
Robeyns, I., and R. van der Veen. Sustainable Quality of Life – Conceptual Analysis for a
Policy-Relevant Empirical Specification. Report 550031005. Bilthoven, the Netherlands:
Environmental Assessment Agency (PBL)/University of Amsterdam, 2007.
Robinson, J. Ordinary Differential Equations. Cambridge: Cambridge University Press,
2004.
Rockström, J., W. Steffen, K. Noone, A. Persson, F. Chapin, E. Lambin, et al. A safe operating
space for humanity, Nature 461 (2009): 472–475.
Rodriguez-Iturbe, I., and A. Rinaldo. Fractal River Basins. Cambridge: Cambridge University
Press, 1997.
Rokeach, M. The Nature of Human Values. New York: The Free Press, 1973.
Rosen, R. Anticipatory Systems. New York: Pergamon Press, 1985.
Rosenau-Tornow, D., P. Buchholz, A. Riemann, and M. Wagner. Assessing the long-term
supply risks for mineral raw materials – a combined evaluation of past and future trends,
Ross, M. Does oil hinder democracy? World Politics 53 (2001): 325–361.
Rotmans, J. and B. de Vries. Perspectives on Global Change – the TARGETS Approach.
Ruth, M., and B. Hannon. Modeling Dynamic Economic Systems. New York: Springer, 1997.
Sabates, R., B. Gould, and H. Villarreal. Household composition and food expenditures: A
cross-country comparison, Food Policy 26 (2001): 571–586.
Safarzynska, K., and J. van den Bergh. Demand-supply coevolution with multiple increasing
returns: Policy analysis for unlocking and system transitions, Technological Forecasting and
Social Change 77 (2010): 297–317.
Sahlins, M. Stone Age Economics. London: Tavistock Publications, 1972.
Samuelson, P. Foundations of Economic Analysis (8th ed.). New York: Atheneum, 1947/1976.
Sassi O., R. Crassous, J.-C. Hourcade, V. Gitz, H. Waisman, and C. Guivarch. Imaclim-R: A
modelling framework to simulate sustainable development pathways, International Journal
of Global Environmental Issues 10 (2010): 5–24.
Satake, A., and T. Rudel. Modeling the forest transition: Forest scarcity and ecosystem service
hypothesis, Ecological Applications 17 (2007): 2024–2036.
Schäfer, A., J. Heywood, H. Jacoby, and I. Waitz. Transportation in a Climate-Constrained
World. Cambridge, MA: MIT Press, 2010.
SCEP. Man’s Impact on the Global Environment. Cambridge, MA: MIT Press, 1970.
Schandl, H., and N. Schulz. Industrial ecology: The UK. In A Handbook of Industrial Ecology,
pp. 323–333, edited by R. Ayres and L. Ayres. Cheltenham, UK: Edward Elgar, 2002.
Schandl, H., and F. Krausmann. The great transformation: A socio-metabolic reading of
the industrialization of the United Kingdom. In Socioecological Transitions and Global
Change: Trajectories of Social Metabolism and Land Use, edited by M. Fischer-Kowalski
and H. Haberl. Cheltenham, UK: Edward Elgar, 2007.
Schandl, H., M. Fischer-Kowalski, C. Grunhubel, and F. Krausmann. Socio-metabolic
transitions in developing Asia, Technological Forecasting and Social Change 76 (2009):
267–281.
Scheffer, M., S. Carpenter, J. Foley, C. Folke, and B. Walker. Catastrophic shifts in ecosystems,
Nature 413 (2001): 591–596.
Scheffer, M. Critical Transitions in Nature and Society. Princeton, NJ: Princeton University
Press, 2009.
Scheffer, M., J. Bascompte, W. Brock, V. Brovkin, S. Carpenter, V. Dakos, et al. Early-warning
signals for critical transitions, Nature 461 (2009): 53–59.
References 579
Scheffer, V. The rise and fall of a reindeer herd, The Scientific Monthly 73 (1951):
356–362.
Schneider, S., and P. Boston, eds. Scientists on Gaia. Boston: MIT Press, 1991.
Schnellnhuber, G., A. Block, M. Cassel-Gintz, J. Kropp, G. Lammel, W. Lass, et al. Syndromes
of global change, GAIA 6 (1997): 19–34.
Schnellnhuber, G., P. Crutzen, W. Clark, and J. Hunt. Earth system analysis for sustainability,
Environment 47(2005): 10–27.
Schrijver, N. Development without Destruction – The UN and Global Resource Management.
Bloomington: Indiana University Press, 2010.
Schumacher, E. Small Is Beautiful. New York: Harper & Row, 1973.
Schumacher, E. A Guide for the Perplexed. London: Harper Perennial, 1977.
Schwartz, M., and M. Thompson. Divided We Stand – Redefining Politics, Technology and
Social Choice. Hemel Hempstead, UK: Harvester Wheatsheaf, 1989.
Schwartz, P. The Art of the Long View: Planning for the Future in an Uncertain World. New
York: Currency Doubleday, 1995.
Schwartz, S. Beyond individualism/collectivism: New dimensions of values. In Individualism
and Collectivism: Theory Application and Methods, edited by U. Kim, H. Triandis, C.
Kagitcibasi, S. Choi, and G. Yoon. Newbury Park, CA: Sage, 1994.
Schweitzer, F., G. Fagiolo, D. Sornette, F. Vega-Redondo, A. Vespignani, and D. R. White.
Economic networks: The new challenges, Science 325 (2009): 422–425.
Scott, J. Seeing like a State – How Certain Schemes to Improve the Human Condition Have
Failed. New Haven, CT: Yale University Press, 1998.
Senge, P. The Fifth Discipline – The Art & Practice of the Learning Organization. New York:
Doubleday Currency, 1990.
Seppelt, R., and O. Richter. It was an artefact not the result: A note on system dynamics
model development tools, Environmental Modeling & Software 20 (2005): 1543–1548.
Shah, I. Learning how to learn – Psychology and spirituality in the Sufi way. London: Penguin
Books, London, 1978/1985.
Shiva, Vandana. The practice of earth democracy, Development Dialogue, What Next? 52
(August 2009). Available at www.whatnext.org
Sieferle, R. Ruckblick auf die Natur – eine Geschichte des Menschen und seiner Umwelt.
Munich: Luchterhand, 1997.
Sietz, D., M. Lüdeke, and C. Walther. Categorisation of typical vulnerability patterns in global
drylands, Global Environmental Change 21 (2011): 431–440.
Simon, H. The Sciences of the Artificial. Cambridge, MA: MIT Press, 1969.
Small, C., and J. Cohen. Continental physiography, climate and the global distribution of
human population, Current Anthropology 45 (2004): 269–277.
Smil, V. Energy in World History. Boulder, CO: Westview Press, 1994.
Sohngen, B., R. Mendelsohn, and R. Sedjo. Forest management, conservation, and global
timber markets, American Journal of Agricultural Economics 81 (1999): 1–13.
Solé, R., and B. Goodwin. Signs of Life – How Complexity Pervades Biology. New York:
Basic Books, 2000.
Soramäki, K., M. Bech, J. Arnold, R. Glass, and W. Beyeler. The topology of interbank
payment flows, Statistical Mechanics and Its Applications – Physica A 379 (2007): 317–333.
Sorokin, P. Social and Cultural Dynamics. London: Transaction Publishers, 1957.
Steffen, W., and P. Tyson. Global Change and the Earth System: A Planet under Pressure. New
York: Springer, 2004.
Steg, L., and C. Vlek. Social science and environmental behaviour. In Principles of Envir-
onmental Sciences, pp. 97–142, edited by J. Boersema and L. Reijnders. The Netherlands:
Sterman, J. The economic long wave: theory and evidence, System Dynamics Review 2 (1986):
87–125.
Sterman, J. Business Dynamics – Systems Thinking and Modeling for a Complex World.
Boston: Irwin McGraw-Hill, 2000.
580 References
Sterman, J., and L. Sweeney. Cloudy skies: Assessing public understanding of global warming,
System Dynamics Review 18 (2002): 207–240.
Stern, D., M. Common, and E. Barbier. Economic growth and environmental degradation:
The environmental Kuznets curve and sustainable development, World Development 24
(1996): 1151–1160.
Stern, D. The rise and fall of the environmental Kuznets curve, World Development 32 (2004):
1419–1439.
Stern, N. The economics of climate change, American Economic Review 98 (2008): 1–37.
Stikker, A. Water – The Blood of the Earth. Exploring Sustainable Water Management for the
New Millennium. New York: www.cosimobooks.com, 2007.
Ströbele, W. Wirtschaftswachstum bei begrenzten Energieressourcen. Berlin: Duncker & Hum-
bolt, 1984.
Strogatz, S. Nonlinear Dynamics and Chaos – with Applications to Physics, Biology, Chemistry,
and Engineering. Reading, MA: Addison-Wesley, 1994.
Sustainability Institute. Commodity Systems Challenges. A Sustainability Institute Report,
April 2003. Available at www.sustainer.org.
Swinnen, J., A. Banerjee, and H. de Gorter. Economic development, institutional change, and
the political economy of agricultural protection – an econometric study of Belgium since
the 19th century, Agricultural Economics 26 (2001): 25–43.
Tainter, J. The Collapse of Complex Societies. Cambridge: Cambridge University Press, 1988.
Tainter, J. Problem solving: Complexity, history, sustainability, Population and Environment
22 (2000): 3–41.
TCEPD (Taiwan Council for Economic Planning and Development). Taiwan 2000 – Balan-
cing Economic Growth and Environmental Protection. Taipei: Taiwan National University,
1989.
Tannahill, R. Food in History. New York: Stein and Day, 1973.
Tasaka, H. Invisible Capitalism: Beyond Monetary Economy and the Birth of New Paradigm
Economies. New York: Jorge Pinto Books, 2009.
Taylor, M. S. Buffalo hunt: International trade and the virtual extinction of the North Amer-
ican bison, American Economic Review 101 (2011): 3162–3195.
TDBT (The Danish Board of Technology). Oil-based Technology and Economy Prospects
for the Future. 2004. Available at www.tekno.dk/pdf/projekter/p04 Oil-based Technology
and Economy.pdf.
Teilhard de Chardin, P. L’avenir de l’homme. Paris: Editions du Seuil, 1959.
Teives Henriques, S., and A. Kander. The modest environmental relief resulting from the
transition to a service economy, Ecological Economics 70 (2010): 271–282.
TEPD (2010). Transportation Energy Data Book 2010. Available at cta.ornl.gov/data/
download29.shtml.
Tesfatsion, L. Introduction to the special issue on agent-based computational economics,
Journal of Economic Dynamics & Control 25 (2001): 281–293.
Thirgood, J. Man and the Mediterranean Forest – A History of Resource Depletion. New York:
Academic Press, 1981.
Thompson, M., R. Ellis, and A. Wildawsky. Cultural Theory. Boulder, CO: Westview Press,
1990.
Thompson, M. The dynamics of cultural theory and their implications for the enterprise
culture. In Understanding the Enterprise Culture, edited by S. Heap and A. Ross. Edinburgh:
Edinburgh University Press, 1992.
Thompson, M. Security and solidarity: An anti-reductionist framework for thinking about
the relationship between us and the rest of nature, The Geographical Journal 163 (1997):
141–149.
Thornton, P. Livestock production: recent trends, future prospects, Philosophical Transac-
tions of the Royal Society B 365 (2010): 2853–2867.
Thurstan, R., S. Brockington, and C. Roberts. The effects of 118 years of industrial
fishing on UK bottom trawl fisheries, Nature Communications 1 (2010): doi:10.1038/
ncomms1013.
References 581
Toth, F. L. Policy exercises, Simulation & Games 19 (1988): 235–276.

Totman, C. The Green Archipelago: Forestry in Pre-industrial Japan. Berkeley: University of
California Press, 1989.
Toynbee, A. Mankind and Mother Earth: A Narrative History of the World. Oxford: Oxford
Turchin, P. Historical Dynamics: Why States Rise and Fall. Princeton, NJ: Princeton University
Press, 2003.
Turchin, P. Complex Population Dynamics. Princeton, NJ: Princeton University Press,
2008.
Turner, B., W. Clark, R. Kates, J. Richards, J. Mathews, and W. Meyer, eds. The Earth as
Transformed by Human Action – Global and Regional Changes in the Biosphere over the
Past 300 Years. Cambridge: Cambridge University Press, 1990.
Turner, G. A comparison of The Limits to Growth with 30 years of reality, Global Environ-
mental Change 18 (2008): 397–411.
Tylecote, A. The Long Wave in the World Economy – The Present Crisis in Historical Per-
spective. London: Routledge, 1992.
Ulanowicz, R. Growth and Development: Ecosystems Phenomenology. San Jose, CA: Excel
Press, 1986.
UN (United Nations). Indicators of Sustainable Development – Framework and Methodolo-
gies. New York: UN, 1996.
UNAIDS. 2004. Available at www.unaids.org/en.
UNEP. Global Environmental Outlook 2. 2000. Available at www.unep.org/geo/GEO2000/
english/Index.htm.
UNEP. Global Environmental Outlook: Environment for Development (GEO–4). Nairobi:
UN, 2007.
van Beek, L., and M. Bierkens. The Global Hydrological Model PCR-GLOBWB: Con-
ceptualization, Parameterization and Verification. Department of Physical Geography,
Utrecht University, Utrecht, the Netherlands, 2008. Available at vanbeek.geo.uu.nl/
suppinfo/vanbeekbierkens2009.pdf.
van Beek, L., Y. Wada, and M. Bierkens. Global monthly water stress: 1. Water bal-
ance and water availability, Water Resources Research 47 (2011): W07517, doi:10.1029/
2010WR009791.
van den Bergh, J., A. Ferrer-i-Carbonell, and G. Munda. Alternative models of individual
behaviour and implications for environmental policy, Ecological Economics 32 (2000):
43–61.
van den Bergh, J., ed. Handbook of Environmental and Resource Economics. Cheltenham,
UK: Edward Elgar, 2002.
van den Bergh, J., A. Faber, A. Idenburg, and F. Oosterhuis. Evolutionary Economics and
Environmental Policy: Survival of the Greenest. Cheltenham, UK: Edward Elgar, 2007.
van den Bergh, J. The GDP paradox, Journal of Economic Psychology 30 (2009): 117–135.
van der Leeuw, S., ed. The Archaeomedes Project: Understanding the Natural and Anthro-
pogenic Causes of Land Degradation and Desertification in the Mediterranean Basin.
Brussels, Science Research Development European Commission Report EUR 18181 EN.
Brussels, 1998.
van der Leeuw, S., and B. de Vries. The Roman Empire. In Mappae Mundi – Humans and
Their Habitats in a Long-Term Socio-ecological Perspective. Myths, Maps, and Models, pp.
209–256, edited by B. de Vries and J. Goudsblom. Amsterdam: Amsterdam University
Press, 2002.
van Egmond, N., and B. de Vries. Sustainability: the search for the integral worldview, Futures
43 (2011): 853–867.
van Praag, B., and A. Ferrer-i-Carbonell. Happiness Quantified. Oxford: Oxford University
Press, 2004.
van Praag, B., and P. Frijters. The measurement of welfare and well-being: The Leyden
approach. In Well-Being – The Foundations of Hedonic Psychology, edited by D. Kahne-
man, E. Diener, and N. Schwarz. New York: Russell Sage Foundation, 1999.
582 References
van Vuuren, D., B. Strengers, and B. de Vries. Long-term perspectives on world metal use –
a model-based approach, Resources Policy 25 (2000): 239–255.
van Vuuren, D., B. de Vries, A. Beusen, and P. Heuberger. Conditional probabilistic estim-
ates of 21st century greenhouse gas emissions based on the storylines of the IPCC-SRES
scenarios, Global Environmental Change 18 (2008): 635–654.
van Zon, R. Duurzame ontwikkeling in historisch perspectief. Vakreviews Duurzame
Ontwikkeling. RU Groningen/KUN, 2002.
Veenhoven R. Is happiness relative? Social Indicators Research 24 (1991): 1–34.
Vennix, J. Group model building: Tackling messy problems, System Dynamics Review 15
(1999): 379–401.
Verburg, P., A. Veldkamp, and L. Fresco. Simulation of changes in the spatial pattern of land
use in China, Applied Geography 19 (1999): 211–233.
Verburg, P., W. de Groot, and A. Veldkamp. Methodology for multi-scale land-use change
modelling: Concepts and challenges. In Global Environmental Change and Land Use, edited
by A. Dolman, A. Verhagen, and C. Rovers. Dordrecht, the Netherlands: Kluwer Academic
Publishers, 2003.
Verburg, P., K. Overmars, M. Huigen, W. de Groot, and A. Veldkamp. Analysis of the effects
of land use change on protected areas in the Philippines, Applied Geography 26 (2006):
153–173.
Verburg, P., D. van Berkel, A. van Doorn, M. van Eupen, and H. van den Heiligenberg.
Trajectories of land use change in Europe: A model-based exploration of rural futures,
Landscape Ecology 25 (2010): 217–232.
Vervoort, J., K. Kok, R. van Lammeren, and T. Veldkamp. Stepping into futures: Exploring
the potential of interactive media for participatory scenarios on social-ecological systems,
Futures 42 (2010): 604–616.
Verweij, M., and M. Thompson, eds. Clumsy Solutions for a Complex World: Governance,
Politics, and Plural Perceptions. New York: Palgrave Macmillan, 2006.
Vringer, K., and K. Blok. Long-term trends in direct and indirect household energy intensities:
A factor in dematerialisation? Energy Policy 28 (2000): 713–727.
Vringer, K., T. Aalbers, and K. Blok. Household energy requirement and value pattern,
Energy Policy 35 (2007): 553–566.
VROM/ABF. 2009. Statistical Data on Housing in the Netherlands. See www.cbs.nl.
Wackernagel, M., N. Schulz, D. Deumling, A. Callejas Linares, M. Jenkins, V. Kapos,
et al. Tracking the ecological overshoot of the human economy, Proceedings of the
National Academy of Sciences of the United States of America (PNAS) 99 (2002):
9266–9271.
Wada, Y., L. Van Beek, C. van Kempen, J. Reckman, S. Vasak, and M. Bierkens. Global
depletion of groundwater resources, Geophysical Research Letters 37 (2010): L20402.
Wada, Y., L. van Beek, D. Viviroli, H. Dürr, R. Weingartner, and M. Bierkens. Global monthly
water stress: 2. Water demand and severity of water stress, Water Resources Research 47
(2011): W07518, doi:10.1029/2010WR009792.
London: John Wiley& Sons, Ltd., 2004.
Walker, B., J. Anderies, A. Kinzig, and P. Ryan, eds. Exploring Resilience in Social-Ecological
Systems – Comparative Studies and Theory Development. Canberra, Australia: CSIRO
Publishing, 2006.
Wall, G., and M. Gong. On exergy and sustainable development – Part I: Conditions and
concepts, Exergy International Journal 1 (2001): 128–145.
Walters, C. Adaptive Management. New York: Macmillan, 1989.
Warde, P. Energy Consumption in England & Wales 1560–2000. Consiglio Nazionale delle
Ricerche, 2007.
Warr, A., H. Schandl, and R. Ayres. Long-term trends in resource exergy consumption and
useful work supplies in the UK, 1900 to 2000, Ecological Economics 68 (2002): 126–140.
References 583
Wassenaar, T., P. Gerber, M. Rosales, M. Ibrahim, P. Verburg, and H. Steinfeld. Predicting

land use changes in the neotropics: The geography of pasture expansion into forest, Global
Environmental Change 17 (2007): 86–104.
Watts, D. Small Worlds – The Dynamics of Networks between Order and Randomness. Prin-
ceton Studies in Complexity. Princeton, NJ: Princeton University Press, 1999.
WCED (World Commission on Environment and Development). Our Common Future.
Oxford: Oxford University Press, 1987.
Weber, M., B. Volker, and K. Hasselmann. A multi-actor dynamic integrated assessment
model (MADIAM) of induced technological change and sustainable economic growth,
Ecological Economics 54 (2005): 306–327.
Weisbuch, G., and G. Boudjema. Dynamical aspects in the adoption of agri-environmental
measures, Advances in Complex Systems 2 (1999): 11–36.
Weisdorf, J. From foraging to farming: Explaining the Neolithic revolution, Journal of Eco-
nomic Surveys 19 (2005): 561–586.
Welcomme, J., I. Cowx, D. Coates, C. Béné, S. Funge-Smith, A. Halls, and K. Lorenzen. Inland
capture fisheries, Philosophical Transactions of the Royal Society B 365 (2010): 2881–2896.
Wene, K.-O./IEA. Experience Curves for Energy Technology Policy. Paris: IEA, 2000.
Westhoek, H., M. van den Berg, and J. Bakkes. Scenario development to explore the future
of Europe’s rural areas, Agriculture, Ecosystems and Environment 114 (2006): 7–20.
White R., and G. Engelen. Cellular automata as the basis of integrated dynamic regional
modelling, Environment and Planning B 24 (1997): 235–246.
WHO. 2002. Available at www.who.org.
Wilber, K. The Marriage of Sense and Soul: Integrating Science and Religion. Dublin: Gateway,
2001.
Wilkinson, R. Poverty and Progress: An Ecological Model of Economic Development. London:
Methuen & Co., 1973.
Williams, M. A new look at global forest histories, Annual Review of Environment and
Resources 33 (2008): 345–367.
Wilson, E. Sociobiology – The New Synthesis. Cambridge, MA: Harvard University Press,
1975.
Wilson, E. Consilience: The Unity of Knowledge. New York: Knopf, 1998.
Wilting, H., A. Faber, and A. Idenburg. Investigating new technologies in a scenario context:
Description and application of an input-output method, Journal of Cleaner Production 16S1
(2008): S102–S112.
Wood, M. Legacy – A Search for the Origins of Civilization. London: BBC Worldwide Ltd.,
1999.
World Bank. Expanding the Measure of Wealth – Indicators of Environmentally Sustainable
Development. New York: World Bank, 1997.
WRI. World Resources Report, Chapter 2: Taking stock of ecosystems. 2000–2001.
WRI (2003). World Resource Report.
Wright, R. Nonzero –The Logic of Human Destiny. New York: Pantheon Books, 2000.
Yates, R. War, food shortages, and relief measures in early China. In Hunger in History –
Food Shortage, Poverty, and Deprivation, pp. 147–177, edited by L. Newman. London: Basil
Blackwell, 1990.
Yeats, W. The Second Coming: Collected Poems of W.B.Yeats. New York: Scribner Paperback
Poetry, 1939/1996.
Yergin, D. The Prize – The Epic Quest for Oil, Money and Power. London: Simon & Schuster,
1991.
ZTB (Zentrum fur Transformation der Bundeswehr). Peak Oil – Sicherheitspolitische
Implikationen knapper Ressources. 2010. Available at www.zentrum-transformation.
bundeswehr.de.
Zuo Dakang. The Huang-Huai-Hai Plain. In The Earth as Transformed by Human Action –
Global and Regional Changes in the Biosphere over the Past 300 Years, edited by B. Turner,
W. Clark, R. Kates, J. Richards, J. Mathews, and W. Meyer. Cambridge: Cambridge Uni-
versity Press, 1990.
Index
acidification, 459 in plant/animal, 105, 404

adaptation. See also resilience, vulnerability plant/animal, 98
in human behaviour, 309, 314, 387, 478 biome, 284, 342
to environmental change, 54, 80, 357, 460 biosphere, 5, 78, 105, 109, 128, 224, 244, 339, 397,
Afghanistan, 445 427, 457, 471
agent-based model (ABM). See model birth rate. See fertility
agrarianisation, 54, 77, 105, 339, 425 bounded rationality, 489, 523. See also behaviour
agricultural bureaucratisation. See centralisation of power
area, 202
intensification, 354, 363 calculus, 14, 24, 29, 50–53, 197, 265, 278, 294, 482
markets and innovations, 366–372 Canada, 99, 265, 400, 443
products, 203, 339, 346 Capability Approach, 149, 491, 504
system, 66, 344–347, 366–372 capitalism, 92, 94, 124, 505. See also economics,
agriculture. See agricultural, agro-food chain, institution
system, yield carbon, 100, 250, 254, 273, 391, 430, 440
algorithm, 228, 311, 321, 443 carbon dioxide (CO2) concentration, 118, 357
analogue, 26, 114, 183, 201, 219, 221, 235, 278, 294 carbon dioxide (CO2) emission, 17, 28, 100, 182,
analogy. See analogue 194, 421, 454, 461
analysis. See system dynamics carrying capacity, 134
anthrome, 341–343 of (eco)system, 33, 263, 272, 361, 406
anthropocene, 120 of human population, 72, 74, 125, 385
anthropology, 158, 328 catastrophe theory. See bifurcation
anthroposphere, 105, 120 catastrophic change. See regime
archetype, 40–43, 45, 263, 325 causal loop diagram (CLD), 24, 55, 165, 490
Arizona, 56, 269, 416 causality, 40
arms race, 42, 151, 160, 444 Cellular Automata (CA). See model
attractor. See calculus, equilibrium centralisation
Australia, 289, 332, 456, 460 geographical, 86, 282
of power, 77, 95, 97, 161, 313, 328, 368, 436
Bali, 157 China, 17, 64, 72–74, 86, 109, 165, 179, 308, 362,
behaviour. See economic, utility, value 392, 442, 455, 474
maximising, 147, 312, 389, 405, 414, 443, 504, civic society, 124, 317
511 civilisation
simulation of, 320–324 Greek, 65
social, 310, 311, 314, 318 hydraulic, 65
bifurcation, 63, 279, 294–295 Indus, 63
binomial distribution, 325, 432, 469 Mayan, 66
biodiversity, 101, 127, 133, 225, 240, 253, 254, classical mechanics, 118, 195–201
283–288, 362, 398, 414, 442, 460 climate change. See carbon
biogeography, 360 adaptation, 341, 357, 397
biomass. See economy, energy policy, 440, 461–462
as fuel, 21, 106, 180, 202, 207, 425 processes, 28, 223
585
Downloaded from Cambridge Books Online by IP 132.248.9.8 on Wed Jul 29 22:24:26 BST 2015.
586 Index
Club of Rome, 104, 124 ecology. See also ecosystem

coal, 103, 179–181, 182, 485, 496. See also fossil population ecology, 265, 272
fuel, energy the science of, 119, 134, 244
coevolution. See evolution economic. See pollution
collapse, 57, 76, 78, 162, 277, 329, 381, 400, 477. externalities, 132, 142, 500
See also overshoot-and-collapse production function, 313, 347, 349, 478,
colonisation, 67, 74, 80, 277 482–485, 486, 492–493, 496
Combined Heat and Power (CHP), 204 substitution, 335, 427, 452, 478, 482, 488
common pool resource (CPR), 130, 137, 317, 390, transaction cost, 500
402, 459, 504 economics
comparative advantage, 97, 366, 495 environmental, 132
competition, 41, 55, 58, 60, 75, 165, 314–318, 358, evolutionary, 237, 314, 489, 511
365, 401, 479, 504 externalities, 502
complex adaptive system, 387 neoclassical growth theory, 482–485
complex systems science (CSS), 119, 168, 228, economy
242–243, 297 biomass, 182, 194, 344, 363–366
complexity, 227–231 industrial, 109, 427, 457, 465, 481, 500
social, 60, 77, 157 informal, 109, 481
concentration ecosystem, 103, 225
of power. See centralisation services, 258, 283–288, 344, 361, 414, 457,
of substances. See also carbon, pollution 513
consumption succession in, 253, 263, 272, 277
conspicuous, 310 Egypt, 61–63, 233, 394, 425
of goods and services, 127, 152, 312, 477, 478, electricity, 106, 138, 180, 185, 195, 206–208, 335,
482, 502 441, 493. See also energy
contingency, 270, 343, 440 emission. See carbon, nitrogen, pollution, sulphur
contingent valuation method, 149 empirical reductionism, 122, 168, 217, 298
cooperation, 60, 75, 80, 93, 173, 310, 314–318 energy. See coal, oil and gas, resource
Corporate Social Responsability (CSR), 506 efficiency, 21, 35, 107, 195, 206, 497
corporation, 93–95, 255, 289, 318–319, 368, 430, nuclear, 192, 210, 432
436, 446, 490 renewable, 107, 179, 206–208, 441
correlation, 40, 232 system, 206–208
cosmology, 121, 169 units, 185, 447, 495
creative destruction, 480 Enlightenment, 93, 103, 121
Crete, 65 entropy, 187, 191, 493
cropland. See agricultural, land environmental
Cultural Theory, 158–163, 289 concerns, 117, 133
Environmental Kuznets Curve (EKC)
Darfur, 201 hypothesis, 453, 475
Darwin, 103, 119, 170, 308 movement, 124
death rate. See mortality epistemology, 168, 230
deductive, 217, 222, 240 equilibrium
deforestation. See forestry dynamic, 26, 105, 187, 263, 476
dematerialisation, 449–452, 490, 495 multiple. See bifurcation, regime
demographic transition. See transition thermodynamic, 188, 194
Denmark, 210 Europe, 84, 92, 96, 123, 172, 207, 305, 459
depletion. See fishery, fossil fuel, land, forestry eutrophication, 101, 281, 355
desertification, 73, 257, 283, 357, 398 evolution
difference/differential equation. See calculus of networks, 325
differentiation, 60, 66, 77, 168, 553 of strategies, 323, 512
discount rate, 389, 413, 442, 461, 478, 511. See also theory, 119, 308
economics excludability, 129
dissipative process, 183, 193, 228 exergy, 191–194, 204, 432, 462, 470, 493
diversity. See biodiversity, complexity, resilience experience curve. See innovation
social, 297 exponential decline, 29, 412
social/behavioural, 158, 168, 406 exponential growth, 28, 40, 84, 222, 262, 427, 457,
doubling time (DT), 30 485
drought, 57, 61, 64, 73, 253, 345, 357, 361, 461
famine, 61–63, 72–74, 92, 106
Easter Island, vii, 57–60, 77, 329 feedback. See positive/negative feedback loop
Index 587
fertility. See population, soil heuristic, 158, 183, 221, 278

fertiliser, 101, 347, 368 hierarchy
fireflies, 321 in society, 60, 66, 157, 160, 328, 365
firewood, 58, 69, 107, 201 of needs, 147, 151
fishery, 122, 379–383, 400–404. See also model of of values, 174
fisheries Hohokam people, 56
fixed point. See equilibrium Holling, 277
food Holocene, 81, 105, 120, 129, 299
consumption, 74, 348, 368, 402 homo economicus, 409, 490. See also behaviour,
resources, 348–349 economics
security, 111, 346, 348, 357, 366, 370. See also Human Development Index (HDI), 140
vulnerability human habitat, 339–341
Food and Agriculture Organization (FAO), 147, Human Scale Development (HSD) theory, 150,
245, 371 151
food web, 272–274, 284, 403. See also ecosystem Hungary, 202
in North Sea, 273 hunting-gathering, 54
forest hydrological cycle, 256, 391
control of fire, 54, 179, 253, 276, 322
exploitation, 103, 363, 413 Ibn Khaldun, 171
resource, 412 identity, 80, 151, 169, 173, 275, 322, 409
forestry. See also firewood, forest, model of in silico, 297, 316, 323
forestry, wood inclusive thinking, 46
deforestation, 227, 256, 359–360 income. See also gross domestic product (GDP)
fossil fuel, 37, 107, 142, 179–181, 210, 425, 427, 441, distribution, 88, 135, 348, 486
459, 473, 484 inequality, 89, 165, 453, 505
France, 104, 358, 415, 486 Index of Sustainable Economic Welfare (ISEW),
141, 502
Gaia hypothesis, 119 India, 17, 21, 27, 30, 86, 109, 157, 165, 301, 366,
game theory, 311, 314–317 375, 392, 394, 445, 513
gas. See oil and gas indicator, 136–138
genetically modified organisms (GMOs), 163, 172, of resilience, 275, 280
231, 368 of sustainability, 3, 138–142, 175
genetics, 119, 310 inductive, 217, 222, 240
geographical information system (GIS), 245, industrialisation, 84, 93–95, 106, 122, 171, 425, 451,
360 472
geography inference, 217, 219
the science of, 244, 323 infrastructure. See energy, system, transport
globalisation, 89, 90, 128, 160, 162, 171, 347, 358, innovation. See economics
368, 445, 495, 506 learning-by-doing, 439–441, 488, 499
graph theory. See network lock-in, 43, 277, 512
Greece, 42, 65, 77, 288, 425 input-output (I-O) economics, 512–514
greenhouse gas. See carbon, climate change institution, 23, 77, 93–95, 158, 169, 174, 276, 390,
Greenland, 80, 434 506. See also organisation
gross domestic product (GDP). See economics, integral-differential equations. See calculus
indicator intensity-of-use (IU) hypothesis, 475. See also
and environmental polution, 453 dematerialisation
and resource intensity, 107, 449 interactive modelling, 414
and structural change, 90, 95 irrigation, 57, 62, 258, 346, 392, 397, 446. See also
as indicator, 141, 299, 341, 503 agricultural
definition, 514–515 isomorphism, 221, 269
gross household production (GHP), 503
Japan, 88, 413, 445, 464
halving time (HT), 30
happiness, 25, 148, 169, 226 Kaibab plateau, 269
health. See pollution Kazakhstan, 445
of population, 147, 165 knowledge
services, 305 strong vs, weak, 221–224
heat pump, 205 Kondratiev, 180
herders. See pastoralism Kondratiev (waves), 88
heterogeneity, 229, 256, 270, 274, 330, 484 Kung San people, 309
588 Index
land. See also soil of (post)industrial transition, 474–477

classification (Holdridge), 245 of economic growth, 481–483
degradation, 156, 256–259, 323, 362 of fisheries, 408–412
land use and cover change (LUCC), 359–360 of forestry, 413
use, 339–341 of population and renewable resource, 421–422
League of Nations, 122 predator-prey, 237, 265–272, 294, 383, 422
Life Cycle Assessment (LCA), 448 Sugarscape, 329
life expectancy. See population, mortality Modernism, 117, 140, 169, 217, 224, 239, 309
Linnaeus, 119, 272 modernity. See Modernism
livestock. See agricultural, pastoralism Mongolia, 345, 357
logistic growth. See model multi-agent simulation (MAS). See model,
agent-based
Maddison, 88 Mycenaean people, 65
marginal cost, 33, 437
market ideology, 329, 372, 504. See also economics natural gas. See oil and gas
maximum sustainable yield (MSY), 385, 421 Nature, 244, 289, 296
Max-Neef, 150 conservation, 133, 255, 284. See also
mean species abundance (MSA), 284 biodiversity
Mediterranean, 65, 67, 69, 102, 253, 262, 265, human nature, 309–311
283 negative feedback loop (NFL), 29, 31, 37, 263,
Mesoamerica, 65 271, 369, 473
Mesopotamia, 60–61, 64 net primary production (NPP), 259, 273, 339
metabolism, 105, 138, 447 network
metal. See mineral, mining applications, 326–329
metaphor, 60, 121, 183, 221, 235, 280, 314 theory, 274, 325–326, 336–338
Milankovitch (cycles), 118 Nile river, 61
Millennium Development Goals (MDGs), 126, nitrogen, 17, 100, 250, 284, 355, 430, 460
140, 152, 363 noösphere, 109, 121, 238
Millennium Ecosystem Assessment (MEA), 258,
340 observation, 35, 47, 217, 223, 233, 312. See also
mineral empirical reductionism
global tin cycle, 448 oil and gas. See energy
resources, 20, 192, 379, 425, 431, 436, 473. See logistic (King Hubbert) model, 471
also model, crustal abundance geostatistical resources and depletion, 434
(CAG). supply cost curve (SCC), 436
supply cost curve (SCC), 432, 441 organisation. See institution
mining, 452 self-organisation, 329
and environment, 259, 427, 446, 455, 457 social, 59, 65, 66, 77, 157, 170, 328, 504
gold in New Guinea, 255–256 Ostrom, 317, 390
in simulation model, 379–383, 408–412 overshoot-and-collapse, 45, 59, 142, 264, 269, 309,
industry, 430, 443 387, 408, 477
of metals, 64, 425 ozone, 101, 357, 461
model, 387–421
agent-based, 270, 298, 329, 361, 406, 415 paradigm
cellular automata (CA), 222, 270, 283, 321, 362, economic growth, 125, 135, 500
423 scientific, 120, 160, 216
consumat, 408–412 pastoralism, 55, 68, 74, 345, 347, 357, 361, 394
Conversion of Land Use and its Effects Peru, 446
(CLUE), 362 phase plane. See calculus
crustal abundance geostatistical (CAG), 432, phenomenological, 40, 47, 188, 222
469 Philippines, 362
expert model, 232 phosphorus, 67, 100, 282, 355, 430
hydrological (PCR-GlobWB), 393 photosynthesis, 273
IMAGE, 245, 349 plate tectonics, 118
Lakeland, 379–383 pollution
logistic growth, 31, 41, 86, 236, 262, 294, 309, abatement, 496–499
361, 421, 433, 471, 488 air pollution, 17, 46, 103, 215, 461
logistic substitution, 237, 486 CFCs. See ozone
metamodel, 232–235, 433, 441, 443, 454 classification, 458
network, 512 health impacts, 17, 103, 202, 455, 460, 464
Index 589
noise pollution, 17 revolution

persistent chemicals, 457, 463–465 agricultural, 54
water pollution, 288, 379–383, 397, 398, green, 346, 354, 365
408–412, 455 industrial, 84, 103, 171
population rivalry, 129
agricultural, 339, 344, 385 Roman Empire, 66–72
fertility, 55, 86, 264, 300, 303–305, 309, 368, 385 Russia, 69, 74–75, 84, 92, 253, 444
migration, 54, 75, 85, 92, 300, 475
model in demography, 299–300 salinisation, 60, 190, 258, 354, 398
model in ecology, 272 satisfier, 151, 444, 491
mortality, 55, 74, 86, 105, 112, 140, 294, 300, scenario methodology, 519–521
303–305, 480 Sen and Nussbaum, 149
world, 85 Senegal, 361, 400
positive feedback loop (PFL), 28, 152, 179, 341, Senge, 40
369, 490 shifting cultivation. See agricultural, land
Postmodernism, 169, 170, 172, 234 simulation game, 240, 415
post-normal science, 231, 239 social dilemma, 314–318
power law distribution, 276, 283, 325, 327, 337 sociobiology, 310
precautionary principle, 463 soil
preference erosion, 57, 67, 256–259, 386
drift, 151 fertility, 56, 99, 112, 349, 354
revealed and stated, 148, 149 solar power. See energy
preferential atachment. See network Spain, 69, 79, 171, 398, 470
pressure-state-impact-response (PSIR) specialisation, 77, 90, 479
framework, 236, 458 stability. See also resilience
price mechanism, 33, 479 in ecosystems, 266, 274–275, 280, 284
production. See economic, economy in human systems, 93, 159, 326
public goods (PG), 137, 317 mathematical, 293–294. See also calculus
steady/stationary state. See calculus, equilibrium
quality of life, 7–9, 139, 146–148. See also steam engine, 72, 106, 122, 179, 190, 221,
happiness, value 486
Sterman, 45
rationality. See behaviour, bounded rationality, sulphur, 17, 100, 250, 460
economics sustainable development indicators (SDI), 137
rebound effect, 496. See also economic syndrome, 109–111, 235, 385, 455
recycling, 20, 448, 452, 462. See also mineral system, 22–23. See also agricultural
redundancy, 326 agricultural, 366–372
reference drift, 151 complex adaptive system (CAS), 162
regime inertia, 26, 33, 43, 135, 301, 335
agricultural, 81, 181 social-ecological system (SES), 5, 261, 343,
industrial, 181, 425 479
shift, 277–283 transport, 40, 326
social-ecological, 104–109, 425 system analysis. See system dynamics
regulation, 93, 122, 136, 234, 370, 400, 404, 407, system dynamics, 36, 76, 101, 198, 263, 354, 398,
439, 489. See also economics, institution 490
religion, 95, 157, 166, 173, 296, 298. See also system theory. See system dynamics
paradigm, worldview.
resilience. See stability taxonomy, 222, 272
ecological, 68, 253, 275–277 technology. See economics, innovation
engineering, 275–277 technosphere, 221, 230, 276
social, 57, 72, 77, 94, 326 thermodynamics, 184–189, 203, 213, 217, 242,
resource. See energy 499
backstop, 438, 477, 478–479 Thompson, 158
classification, 384, 430 time preference. See discount rate
efficiency, 496–499 trade, 479. See also capitalism, economics,
open-access. See behaviour, common pool globalisation
resource (CPR) traffic accidents, 17, 30, 501
reserve-production ratio (RPR), 437 tragedy of the commons, 155, 390, 404. See also
resource curse (RC) hypothesis, 444 social dilemma
reversible. See thermodynamics transdisciplinary, 14
590 Index
transition urbanisation, 54, 87, 106, 259, 397, 477, 485

(post)industrial, 472, 480 utility, 147, 149, 312, 477, 490, 511. See also
demographic, 86, 300 consumption, happiness
energy, 180, 208, 445
food/nutrition, 348 value, 152–153, 489. See also behaviour, World
forest/land, 412 Value Survey (WVS)
health/epidemiological, 308 vulnerability, 111–114, 329, 363
land, 477
resource, 454 water
sustainability, 236–238 groundwater, 392, 394, 397
transport resource, 392
automobile. See car system use, 98, 392–397, 456
public, 43, 150, 335 windpower. See energy
world car system, 15–20 wood, 57, 60, 97, 103, 125, 179–181, 202, 412, 413
World Health Organization (WHO), 17, 147
United Kingdom (UK), 86, 148, 179, 404, 485 World Trade Organization (WTO), 371
United Nations (UN) World Value Survey (WVS), 95, 153, 165
Convention on Biological Diversity (CBD), 283 worldview. See Cultural Theory, paradigm, value
Convention on the Law of the Sea (CLOS), 401 and belief system, 164–166
Convention to Combat Desertification and and values, 96
Land Degradation (CCD), 258 in history, 168–170
United States of America (USA), 15, 43, 108, 148,
408, 433 yield, 60. See also food resources, system
universalism, 120, 170 Yucatán, 65
Urban PM10 concentrations in 2000
" "
" " ""
" "
" " " " " " "
" """ """" " " " "" "
" " """" " " "" """""" " " " "
"
"
"
"
"" "" " " " " """ """""
"" "" " " """ """ """" " " "" " "
""" "" "" "" """ """""" "
"""" " """ " " "" """"
" " """"" "
"""" "" """
""" "" " """
""
""" "
"" " "" " " """""""""" " "
""""""
"" " " " " " "
"""" "
" " "
" " " """ """"""""" """"
"" """""
""
"""""""""
"" """""""" " " " """ " """"""" "" " "
" " """ " " " " "
" " " " """"""" """"
""
"
""
""
""
"""
" """ ""
""""""
"""
"
" ""
""" "
""""""
""""" """
"" "
"
"""
"""
""
""" """"
" ""
"" " """ "" " " """ "
" " " " "
" " " "" """ "" "" """ """
" "" "" """" " "
"" " " "" "" ""
"
"""""" """ """" """ ""
""""" """"
""""" """
"" """" ""
""" "" " "
" " " """""""" """ "
"" """ " " " " "
""""""""" " ""
" " "" " """ "" " "
" " " """""" " "" " " " " "" """"
" ""
"
"" "
"""""
" "
" """""" "
"""""
"
"""""""
"""" "" " "
" " "
"""""" " " """ " " "
" "
""
""""""""""" " "" "
""" "
"" "
""" """""" ""
"" """""""
" " " """"" "" " """""" "
"
""" " " " "" "
" "" """ " "" " "
"""
" " " " " """""" " " """""""
" " ""
"""" """ "" " """"""" """""" """""" " " " "
""" " " """" """ """"
"" "
""
""
"" " ""
"" "" """ " " " """ """ " "
" " " " """ """ " "" "" """ """ " "" """"""""""" " " "" " " ""
""
""
""
"
"
"
"""
"
"
" " ""
"
" " "" " " "" """
"" " "" """
""""""""" "" "" "" " " " """" "" "" "
" "" """" """" """" "" " "" " ""
""" "" "
""""
"
" ""
""
" "
"
""
"""""
"
" " " " " " """" """"" "" """ "" """" " " " "" " """ " " "
" "" " " "" """" """" """ ""
" "" " ""
""
" ""
"""""
""
""
"""
"
""
"" "
"
" ""
" " "" "" """" "
"" " "
" """" " " "
" "
"""""" " " """""
"
" "
"" "" """"
""""""" " ""
"""
""
""
"""
""
"""
"""
"
"""""""""""
""""
"
"" " " " "" " " """ " " ""
"""" " "
" "" " " "
""
""
"
"""
""
""
""""""" "" " "
" """" " " """ "
""" " ""
"
"""""""""""" ""
""
""""
" " "" " " " " """""""""
"" "
"
" """
"
""
"
"
"
"
"""
"
"
"""
"
"""" " "
" ""
""" "
"
"" " " "" "
"
"
""
"
" " " " " " "" """ "" " " " " " "" """
""" """
""
"
"""
"""" """
"
" """"
""" " " "" "
"" ""
" "" """ "" """" "" "" " " "" " " " " """""" "" " "
" "" "" """ " " "" " " """" "" ""
""""
" """
" """""" """""" " "" "" " """ "" "
" "" " """ " " " " " """ " """" """"""" """
"
"""" """"""""
"
""
"
"
""""
"" " "" ""
""""
""""" " """ """"""" " """"""
""
""""""""
"" " """ ""
"
"""
""
" """""""
" " "
" " """ ""
"
"""""
""
"""""""""""""" "" "" """ ""
""
""
"
""" "
"""""" "
""""
" " """ "" "
"" ""
" "
"
"" " " " "
"
" " " """"" "" " ""
"""
"""
"
" " "
" """
" " " " "
" "
" "" "" " "
"""""" " " "" " " "" "" "
""""""" ""
""""" "" """"" " """ "
""" " " """
" " " " " "" """ """ " ""
"
""
"" "
"""" " """ ""
" """ " " """ "
""" " " " "
"
"""""
""
""""" """
"
" "" ""
" "
" "
"" "" "" " """"
" "" " """" "" ""
" """ " " " "" " " ""
" " " " "" "
""
""""
"""
""""
" """ ""
""" """"
" " """" " "" " " "" " "" "
" "" "" ""
" "" " " " " """ """""" """" """ " "
" "" " " ""
"" " "" " " """ " ""
"""
""" "" " " """ """
" "
""" " """
"
" " " """
"" " " " " " ""
Urban PM10 concentrations (ug/m3) ""
"""
"" " " ""
" """ """ " ""
" " "
"
" "
""
"
"" " "
"" " " " " " " " "
"" " "" " " " " "" ""
" < 20 "" " " " " " """"" """"""
"" " " " "" """" " " " "" " "
" " """
" "" " ""
"" "" " "
"" "
" 21 - 30 " "
"
" " " " "
"
"" "
" "" " "
" """ " "
" """ " "" " "" "" " "
" " " """ " """
"" "" " "
" 31 - 50 " """ """"" """"" " "
" " " "" """
"
"""
"""""""
"" " " "
" "" "" """" "" "
" "" " "" "" ""
" " "
" "
" 51 - 70 " " " "" ""
" """ ""
"
"
"
" " " """ "
" " " "" """
""" " "" ""
" > 70 ""
"" " ""
" " " "" ""
" "
"" "
" "
"
"
Figure 2.3b. Air quality in terms of Particulate Matter (PM10) concentrations in the cities of the world. GUAM concentration and population data
are coupled to city locations (x,y). The values are aggregated to an average value for IMAGE-model 0.5o × 0.5o grid cells. The cells are converted
into square points. The darker the colour, the worse urban air quality (source of data: PBL).
a)
b)
Figure 4.2.a,b. Map with population densities 1700 and 2000 (Source: HYDE/PBL).
a)
Figure 4.5a,b. Map with land use in 1700 and 2000 (Source: HYDE/PBL). The upper (a) shows
the fraction of a gridcell used as cropland; the lower (b) shows the fraction used as pasture.
b)
Figure 4.5a,b (continued)
Figure 4.12. Distribution of the eight vulnerability profiles in drylands worldwide according
to the vulnerability profiles in Figure 4.11 (Kok et al. 2011).
Figure 9.1. Soil map of Earth (source of map: FAO).
Figure 9.2. Climate map of Earth: annual average temperature and rainfall (source of maps:
PBL).
Ice Boreal forest Temp.deciduous forest Hot desert Tropical woodlands
Tundra Cool confier forest Warm mixed forest Scrublands Tropical forest
Wooded tundra Temp.mixed forest Grasslands/steppe Savanna
Figure 9.4. Potential vegetation cover based on Holdridge life zone classification (12 classes)
(source of map: PBL/www.mnp.nl\image).
Figure 9.5. Land cover map (15 classes) (source of map: Ramankutty and Foley 1999).
Figure 9.8. Global negative trend in rain-use efficiency-adjusted normalised difference vegetation index, 1981–2003 (source of map: Bai et al.
2008).
1970

2000

2050
Legend
100%
20%
0%
40%
60%
80%
Figure 9.19. Map of terrestrial Mean Species Abundance (MSA) in 1970 and 2000, and a
projection for 2050, based on calculations with the Globio-model (Alkemade et al. 2009, PBL
2010b).
Figure 11.2. Map of anthropogenic biomes or anthromes (Ellis et al. 2010).
Temperate cereals Rice
Maize Millet, Sorghum
Yield (km2 harvested area per grid cell)

0 501 - 1,000 1,510 - 2,000 > 2500
< 500 1,010 - 1,500 2,010 - 2,500
Figure 11.5. Potential yields for some important crops, based on simulation with the IMAGE-model (PBL).
Figure 11.6. Estimate of the efficiency in wheat production (Neumann et al. 2010). Efficiency
is defined as the ratio of the observed output to the corresponding frontier output.
2000 2000
-2 -1 -2 -1
N in kg km yr P in kg km yr
<100 <50
100 - 200 50 - 100
200 - 400 100 - 200
400 - 800 200 - 400
800 - 1600 400 - 800
1600 - 3200 800 - 1600
3200 - 6400 1600 - 3200
6400 - 12800 3200 - 6400
> 12800 > 6400
GO scenario 2050 GO scenario 2050
AM scenario 2050 AM scenario 2050
Figure 11.8. Total balance for (left) N and (right) P for natural ecosystems and agriculture for 2000 (Bouwman et al. 2009).
1972 2002 2022
Figure 11.9. Generalised land use in San Mariano municipality in the northern Philippines: 1972 and 2002 based on interpretation of respectively aerial photographs
and SPOT images; 2022 simulated by the spatial landscape-level model (Verburg et al. 2006).
0 - 0.2 0.2 - 0.4 0.4 - 0.6 0.6 - 0.8 0.8 - 1
Figure 12.6a. Dynamic water stress index (DWSI) based on monthly water scarcity index ≥ 0.4 over the period 1958–2001 (Low: 0–0.2, Moderate:
0.2–0.4, High: 0.4–0.6, Very high: 0.6–0.8, Extremely high: 0.8–1.0) (Wada et al. 2011).
(a) (b)
(c)
Figure 12.6b. Groundwater depletion in the regions of the United States, Europe, China and India and the Middle East for the year 2000
(mm · a−1; clockwise from top left). (WadaDownloaded
et al. 2010; reproduced/modified by permission of American Geophysical Union).
from Cambridge Books Online by IP 132.248.9.8 on Thu Jul 09 23:28:02 BST 2015.

De Vries - 2015 - Sustainability Science

Uploaded by

Copyright:

Available Formats

De Vries - 2015 - Sustainability Science

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

De Vries - 2015 - Sustainability Science

Uploaded by

Copyright:

Available Formats

SUSTAINABILITY SCIENCE

“This textbook is one of the first truly all-encompassing introductions to sustainab-

“Bert de Vries’ Sustainability Science is particularly welcome as it breaks ground

C Bert J. M. de Vries 2013

This publication is in copyright. Subject to statutory exception

First published 2013

Printed in the United States of America

Library of Congress Cataloging in Publication Data

ISBN 978-1-107-00588-4 Hardback

Additional resources for this publication at www.cambridge.org/devries

Cambridge University Press has no responsibility for the persistence or accuracy of

2 The System Dynamics Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3 In Search of Sustainability: Past Civilisations . . . . . . . . . . . . . . . . . . . 54

3.3.2 Egypt: The Nile and Its Rhythms 61

4 The World in the Past 300 Years: The Great Acceleration . . . . . . . . . . 84

5 Sustainability: Concerns, Definitions, Indicators . . . . . . . . . . . . . . . . 117

6 Quality of Life: On Values, Knowledge and Worldviews . . . . . . . . . . 146

6.2 Quality of Life and Values 146

7 Energy Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

8 On Knowledge and Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

9 Land and Nature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244

9.3.1 Forest Fires 253

10 Human Populations and Human Behaviour . . . . . . . . . . . . . . . . . . . 296

11 Agro-Food Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

11.4 Land Use and Cover Change 359

12 Renewable Resources: Water, Fish and Forest . . . . . . . . . . . . . . . . . 379

13 Non-Renewable Resources: The Industrial Economy . . . . . . . . . . . . 425

13.4.1 Oil and Power 445

14 Towards a Sustainable Economy? . . . . . . . . . . . . . . . . . . . . . . . . . . 472

15 Outlook on Futures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 516

15.3 Scenarios for a Sustainable World 522

insofar as it is relevant for understanding sustainable development: rise and decline

Whatever was created, needs

Water is one part,

Box 1.1. Sustainability science. ‘A new field of sustainability science is emer-

intricate. Development is an evolution from simple to complex in terms of technical

1.2 Sustainability Science

economics and so on (Boersema and Reijnders 2009). Gradually, it became clear

Box 1.2. Exactness. Don’t expect a Unified Theory of SES or Social-Ecological

1.3 Sustainable Development Is About Quality of Life

Sustainable development is development that meets the needs of the present

Elsewhere and later

1.4 Guidelines for the Reader

behind resource exploitation and environmental degradation (Chapters 11–14)? Of

Suggestions for a research program from a Global Change perspective.

r www.igbp.net/ is the site of the International Geosphere-Biosphere Program (IGBP) and

Appendix 1.1 United Nations Decade of Education for Sustainable

Education as the foundation of sustainable development was reaffirmed at the

Table 1.1. Key action themes of the UN decade of

Quality education Cultural diversity

Crude oil & minerals Car parts manufacturing

Car sales, insurance,

Car use: forensic, Repair, maintenance

Figure 2.1. The car system and related subsystems in society.

2.2 The World Car System

of active interventions by the automobile industry. The disadvantages of the car

0 100 200 300 400 500 600 700 800 900

Population NO2 SO2 PM10

Europe Paris 9.32 43 8 21

East.5 Producing an estimated 14 percent (passenger cars) to 23 percent (all trans-