462759

ON THE DUAL USES OF SCIENCE AND ETHICS
PRINCIPLES, PRACTICES, AND PROSPECTS
ON THE DUAL USES OF SCIENCE AND ETHICS

PRINCIPLES, PRACTICES, AND PROSPECTS
edited by
Brian Rappert and Michael J. Selgelid
Practical Ethics and Public Policy Monograph 4

Series Editor: Michael J. Selgelid
Published by ANU E Press

The Australian National University
Canberra ACT 0200, Australia
Email: [email protected]
This title is also available online at http://epress.anu.edu.au
National Library of Australia Cataloguing-in-Publication entry

Title:

On the dual uses of science and ethics : principles, practices, and prospects /
editors, Brian Rappert, Michael J. Selgelid.
ISBN:
9781925021332 (paperback) 9781925021349 (ebook)
Subjects: Science and civilization--Moral and ethical aspects.

Science--Moral and ethical aspects.

Science and the humanities.
Other Authors/Contributors:

Rappert, Brian, editor.
Selgelid, Michael J., editor.
Dewey Number: 501

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system
or transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise,
without the prior permission of the publisher.
Cover design and layout by ANU E Press
Printed by Griffin Press
This edition 2013 ANU E Press
Contents
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
Introduction
1. Ethics and Dual-Use Research. . . . . . . . . . . . . . . . . . . . . . . . . 3
Michael J. Selgelid
Part I: Dual Use in Context

2. Nanotechnology and Dual-Use Dilemmas . . . . . . . . . . . . . . . . 13
Jim Whitman
3. What Does Neuroethics Have to Say about the Problem of

Dual Use?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Valentina Bartolucci and Malcolm Dando
4. Synthetic Biology as a Field of Dual-Use Bioethical Concern . . . 45

Alexander Kelle
5. Crops Agents, Phytopathology and Ethical Review . . . . . . . . . 65

Simon Whitby
6. The Super TB Experiment: Evolution and Resolution of an

Experiment with Dual-Use Concerns. . . . . . . . . . . . . . . . . . . 87
Nancy Connell
Part II: Ethical Frameworks and Principles

7. Moral Development and Ethical Decision-Making. . . . . . . . . . . 97
Judi Sture
8. Responsible Dual Use. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

John Forge
9. An Expected-Value Approach to the Dual-Use Problem. . . . . . 133

Thomas Douglas
10. The Doctrine of Double Effect and the Ethics of Dual Use. . . 153
Suzanne Uniacke
11. Unknowns in Dual-Use Dilemmas . . . . . . . . . . . . . . . . . . . 165

Michael Smithson
12. Moral Responsibility, Collective-Action Problems and the

Dual-Use Dilemma in Science and Technology. . . . . . . . . . . 185
Seumas Miller
13. Biosecurity and the Just-War Tradition. . . . . . . . . . . . . . . . 207

Koos van der Bruggen
14. The Precautionary Principle and the Dual-Use Dilemma. . . . . 223

Steve Clarke
Part III: Ethical Practices

15. Scientific Control Over Dual-Use Research: Prospects
for Self-Regulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
David B. Resnik
16. Contrasting Dual-Use Issues in Biology and Nuclear

Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
Nicholas G. Evans
17. Considering Contextuality in Dual-Use Discussions:

Is There a Problem?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
Louise Bezuidenhout
Part IV: Ethical Futures

18. Exploring the Role of Life Scientists in Combating the
Misuse of Incapacitating Chemical and Toxin Agents . . . . . . 293
Michael Crowley
19. WHO Project on Responsible Life-Sciences Research

for Global Health Security: Why Ethics and How to
Strengthen It?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
Emmanuelle Tuerlings and Andreas Reis
Conclusion
20. Ethics as . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Brian Rappert
Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
Contributors
Valentina Bartolucci
Valentina Bartolucci is a lecturer at the University of Pisa, Italy, and at the
Marist college campus of Florence, Italy. Valentina is also Visiting Fellow at the
University of Derby (UK). She maintains a close connection with the School for
Social and International Studies at the University of Bradford, UK. Valentina
has a degree in peace studies from the University of Pisa, an MA in conflict
management and human rights from the Sant Anna School of Advanced
Studies, and a PhD from the University of Bradford. Dr Bartolucci has previously
worked as a junior technical expert for the Italian Ministry of Foreign Affairs
in Morocco, and as a consultant for various international organisations (in
Italy, Burkina-Faso and India). She is currently offering consultancies on issues
linked to terrorism, organised crime, migration, and American and Middle
Eastern foreign policies for various national and international organisations as
well as government departments around Europe. In recognition of her impact
internationally, she was recently made a Member of the Aspen Institute (group
I protagonisti italiani nel mondo). In 2011, she was nominated for the Whos
Who Academic Excellence. Apart from her doctoral thesis on the governmental
discourse on terrorism, Dr Bartolucci has written numerous articles in the fields
of politics, critical discourse analysis, security and surveillance. Her research
mainly addresses the foreign and security policies of the United States and other
states, terrorism and counterterrorism, discourses on threat and emergency, and
strategic communication as anti-terrorism. She has recently been awarded with
a Fulbright Research Scholar Bursary for 201314 to be spent at Arizona State
University working on public diplomacy as anti-terrorism.
Louise Bezuidenhout
Louise Bezuidenhout completed her PhD in cardiothoracic surgery at the
University of Cape Town, South Africa, and has worked as a postdoctoral
scientist in the department of cardiovascular sciences at the University of
Edinburgh, UK. In 2008 she was the recipient of an EU scholarship for the
Erasmus Mundus Masters of Bioethics program run by the Catholic University
of Leuven, Radboud University Nijmegen and the University of Padova. These
dual interests in science and bioethics led her to enrol for a PhD in sociology at
the University of Exeter under Professor Brian Rappert, which she completed in
2013. Her research focused on the need for contextual sensitivity in life-science
vii
On the Dual Uses of Science and Ethics
ethics debates, particularly examining how scientists in developing countries

interact with topical ethical issues such as dual use. She is currently a research
fellow at the University of Witwatersrand in South Africa.

Koos van der Bruggen (born 1951) studied political science (international
relations) at Radboud University Nijmegen (the Netherlands). He wrote his
PhD thesis on criteria for the ethical judgment of nuclear deterrence at the
Vrije Universiteit (Free University), Amsterdam. Dr van der Bruggen has been
working at the Rathenau Institute, the Dutch organisation for parliamentary
technology assessment, for more than 12 years. He was involved in research and
debates on social and ethical aspects of technological developments (biomedical
technology, genetics and military technology). More recently, van der Bruggen
has been working at the Royal Netherlands Academy of Arts and Sciences
(developing a code of conduct for scientists regarding issues of biosecurity) and
Delft University of Technology (research on biosecurity and dual use). He was
also the secretary of the Committee of Inquiry into the Dutch involvement in
the Iraq War of 2003 (Davids Committee, 2009). At the moment he is working
as an independent researcher on peace and security issues. Dr van der Bruggen
has published (in Dutch and English) on just-war issues, on issues of ethics and
technology and on biological weapons.
Steve Clarke
Steve Clarke is a Senior Research Fellow in the Centre for Applied Philosophy
and Public Ethics at Charles Sturt University in Australia. He is also a James
Martin Research Fellow in the Institute for Science and Ethics in the Oxford
Martin School and Faculty of Philosophy at the University of Oxford, and a
Research Fellow in the Uehiro Centre for Practical Ethics, also at Oxford. He
holds a PhD in philosophy from Monash University and has previously held
appointments at the University of Melbourne, the University of Cape Town and
La Trobe University. Steve is a broad-ranging philosopher who has published
more than 60 papers in refereed journals and scholarly collections. His papers
have appeared in such journals as the American Philosophical Quarterly,
the British Journal for the Philosophy of Science, the Journal of Medicine and
Philosophy, Religious Studies and Synthese. He is the author of two books
(Metaphysics and the Disunity of Scientific Knowledge, Ashgate, 1998; and The
Justification of Religious Violence, Wiley-Blackwell, 2014) and the co-editor of
three books. The latest of these is Religion, Intolerance and Conflict: A Scientific
and Conceptual Investigation, edited by Steve Clarke, Russell Powell and Julian
Savulescu (Oxford University Press, 2013).
viii
Contributors
Nancy D. Connell
Nancy D. Connell is Professor in the Division of Infectious Disease in the
Department of Medicine at Rutgers New Jersey Medical School (RNJMS) and
the Rutgers Biomedical Health Sciences. Dr Connell, a Harvard University
PhD in microbiology, has made her major research focus the antibacterial drug
discovery in respiratory pathogens such as M. tuberculosis and B. anthracis. She
is Director of the Biosafety Level Three (BSL3) Facility of RNJMSs Center for
the Study of Emerging and Re-emerging Pathogens and chairs the universitys
Institutional Biosafety Committee. Dr Connell has been continuously funded
by the National Institutes of Health (NIH) and other agencies since 1993 and
serves on numerous NIH study sections and review panels. She has served on a
number of committees of the National Academy of Sciencesfor example, the
Committee on Advances in Technology and the Prevention of their Application
to Next Generation Biowarfare Agents (2004), Trends in Science and Technology
Relevant to the Biological Weapons Convention; an international workshop
(2010), and the Committee to Review the Scientific Approaches used in the FBIs
Investigation of the 2001 Bacillus anthracis mailings (2011). Current work is
with the National Academies of Sciences Education Institute on Responsible
Science, held in Jordan (2012) and Malaysia (2013).
Michael Crowley
Michael Crowley is the Project Coordinator of the Bradford Non-Lethal Weapons
Research Programme (BNLWRP) and is also a senior research associate for the
Omega Research Foundation. He has worked for 20 years on arms control,
security and human rights issues, including as executive director of the
Verification Research, Training and Information Centre (VERTIC). He also acted
as chairman of the Bio-Weapons Prevention Project. Prior to this he worked at
the Omega Research Foundation exploring options for effective restriction of the
development, trade and use of security equipment and technology employed in
torture and ill treatment. He has also managed the Arms Trade Treaty project
at the Arias Foundation in Costa Rica and worked as senior arms trade analyst
at BASIC. He has also held several research and policy positions with Amnesty
International, both in the UK section and at the International Secretariat. He
holds a BSc in genetics from Liverpool University, and an MRes and PhD from
Bradford University, where his doctoral thesis explored the regulation of riotcontrol agents, incapacitating agents and related means of delivery.
ix
Malcolm Dando
Malcolm Dando is Professor of International Security at the University of
Bradford. A biologist by training, his main research interest is in the preservation
of the prohibitions embodied in the Chemical Weapons Convention and the
Biological Weapons Convention at a time of rapid scientific and technological
change in the life sciences. His recent publications include Deadly Cultures:
Biological Weapons Since 1945 (Harvard University Press, 2006), which he edited
with Mark Wheelis and Lajos Rozsa.
Thomas Douglas
Thomas Douglas is a Uehiro Research Fellow in the Oxford Uehiro Centre for
Practical Ethics, Faculty of Philosophy, University of Oxford, and a Junior
Golding Fellow at Brasenose College, Oxford. He trained in medicine (MBChB,
Otago) and philosophy (DPhil, Oxford) and works in applied and normative ethics.
His current research addresses the ethics of producing dangerous knowledge
and of using medical technologies for non-medical purposes, particularly crime
prevention. He has previously worked on organ donation policy, reproductive
decision-making, slippery-slope arguments and compensatory justice.
Dr Nicholas G. Evans
Dr Nicholas G. Evans is an Adjunct Research Assistant at the Centre for
Applied Philosophy and Public Ethics, Charles Sturt University, Canberra. His
dissertation focused on the ethics of censoring dual-use research in the life
sciences through an examination of the history of nuclear science. Nicholas is
the author of a number of articles on military ethics, the philosophy of science,
and ethics, and has taught philosophy, military ethics and physics at universities
around Australia. His research interests include emerging military technologies,
public health ethics, concepts of responsibility and autonomy, and friendship.
John Forge
John Forge is an Honorary Associate in the Unit for History and Philosophy of
Science, Sydney University, having previously worked at the Universities of
New South Wales, Wollongong, Griffith and Macquarie. His research ranged
from the philosophy of physical science, especially explanation, in the early
years, to science and ethics later on. Forge has authored numerous articles and
six books, the latest being The Responsible Scientist (Pittsburgh, 2008), winner
of the David Harold Tribe Prize for Philosophy and the Eureka Prize for Research
Ethics; Designed to Kill: The Case against Weapons Research (Springer, 2012);
and On the Morality of Weapons Research (Ashgate, 2014).
Contributors
Alexander Kelle
Alexander Kelle is a political scientist by training and a Senior Policy Officer
in the Office of Strategy and Policy of the Organisation for the Prohibition of
Chemical Weapons in the Hague, Netherlands. He contributed to this volume
while senior lecturer in politics and international relations at the Department of
Politics, Languages and International Studies at the University of Bath, UK. Before
moving to Bath, he held positions at Queens University Belfast, the University
of Bradford, Stanford University, Goethe University Frankfurt and the Peace
Research Institute Frankfurt. His past research has addressed international
security cooperationwith a view to chemical and biological weaponsdualuse governance of synthetic biology, and the foreign and security policies of
Western liberal democracies.
Seumas Miller
Seumas Miller is a Professorial Research Fellow at the Centre for Applied
Philosophy and Public Ethics (CAPPE) (an Australian Research Council Special
Research Centre) at Charles Sturt University (Canberra) and the 3TU Centre
for Ethics and Technology at Delft University of Technology (the Hague) (joint
position). He is the foundation director of CAPPE (200007). He has authored
or co-authored more than 150 academic articles and 15 books, including (with
M. Selgelid) Ethical and Philosophical Consideration of the Dual Use Dilemma
in the Biological Sciences (Springer, 2008), Terrorism and Counter-Terrorism:
Ethics and Liberal Democracy (Blackwell, 2009), The Moral Foundations of Social
Institutions: A Philosophical Study (Cambridge University Press, 2010) and
(with I. Gordon) Investigative Ethics: Ethics for Police Detectives and Criminal
Investigators (Blackwell, 2013).
Brian Rappert
Brian Rappert is a Professor of Science, Technology and Public Affairs and Head
of the Department of Sociology and Philosophy at the University of Exeter.
His long-term interest has been the examination of how choices can be and
are made about the adoption and regulation of security-related technologies
particularly in conditions of uncertainty and disagreement. Recent books by
Rappert include Controlling the Weapons of War (Routledge, 2006), Biotechnology,
Security and the Search for Limits (Palgrave, 2007), Technology & Security (ed.,
Palgrave, 2007), Biosecurity (co-ed., Palgrave, 2009) and Experimental Secrets
(UPA, 2009). More recently he has been interested in the social, ethical and
epistemological issues associated with researching and writing about secrets, as
in his book Experimental Secrets (2009) and How to Look Good in a War (2012).
xi
Andreas Alois Reis

Andreas Alois Reis (MD, MSc) is Technical Officer in the Department of Ethics
and Social Determinants of Health at the World Health Organisation (WHO)
headquarters in Geneva. After medical education in internal medicine in
Germany, France and Chile, he pursued studies in health economics. His main
area of work is public health ethics, with a focus on distributive justice and
equitable access to health services, and ethical aspects of infectious diseases
such as HIV, pandemic influenza and tuberculosis. He has lectured and organised
training for WHO in more than 50 countries, and is currently serving on the
editorial board of Public Health Ethics.
David B. Resnik
David B. Resnik (JD, PhD) is a bioethicist and Institutional Review Board (IRB)
Chair at the National Institute of Environmental Health Science, National
Institutes of Health. He has published eight books and 200 articles on ethical,
legal, social and philosophical issues in science, medicine and technology, and
is Associate Editor of the journal Accountability in Research.
Michael J. Selgelid
Professor Michael J. Selgelid is Director of the Centre for Human Bioethics at
Monash University in Melbourne, Australia. He was previously Director of the
World Health Organisation Collaborating Centre for Bioethics at The Australian
National University. His research focus is public health ethicswith emphasis
on ethical issues associated with biotechnology and infectious disease. He is
editor of a book series in public health ethics analysis for Springer, Co-Editor of
Monash Bioethics Review, and Associate Editor of the Journal of Medical Ethics.
He co-authored (with Seumas Miller) Ethical and Philosophical Consideration of
the Dual-Use Dilemma in the Biological Sciences (Springer, 2008).
Michael Smithson
Michael Smithson is a Professor in the Research School of Psychology at The
Australian National University in Canberra. He is the author of six books, coeditor of two, and his other publications include more than 140 refereed journal
articles and book chapters. His primary research interests are in judgment and
decision-making under uncertainty, statistical methods for the social sciences,
and applications of fuzzy-set theory to the social sciences.
xii
Contributors
Judi Sture
Judi Sture is the Head of the Graduate School at the University of Bradford,
UK, where she leads two doctoral research training programs. She lectures in
research ethics and research methodology and is closely involved in devising
and developing postgraduate and ethics policy and practice at the university
and beyond. As a member of the Wellcome Trust Dual-Use Bioethics Group and
associate member of the Bradford Disarmament Research Centre, she is engaged
with colleagues from a number of UK and overseas universities in developing a
bioethics approach to counter-biosecurity threats in the life sciences. Judi holds
a BSc(Hons) in archaeology (University of Bradford), in which she specialised
in the study of human skeletal remains, and a PhD (University of Durham) in
biological anthropology, focusing on environmental associations with human
birth defects. Her research in biological anthropology continues, including
further work on developmental defects. She is currently engaged in analysis of
skeletal remains held at the Museum of London and is working with the British
Association for Biological Anthropology and Osteoarchaeology on developing
ethical practice in the profession.
Emmanuelle Tuerlings
Dr Emmanuelle Tuerlings has extensive experience on dual-use research,
biosecurity and global health security. She carries out training and consultancy
with a particular interest in public health. She was a scientist to the Biorisk
Reduction for Dangerous Pathogens (BDP) program at WHO headquarters,
Geneva, from 2004 to 2011, where she was leading the project Responsible Life
Sciences Research for Global Health Security. Before joining WHO, she was
based at the Harvard Sussex Program, University of Sussex, UK. She also worked
with several international and non-governmental organisations on issues related
to dual-use biological technologies and their governance. She holds an MSc and
a doctorate in science and technology policy from the University of Sussex (UK).
Suzanne Uniacke
Suzanne Uniacke is Director of the Centre for Applied Philosophy and Public
Ethics, Charles Sturt University, Canberra. She has held positions in philosophy
departments in England and Australia and visiting research fellowships at St
Andrews, Harvard and Stirling universities. Professor Uniacke has published
extensively in ethics, applied philosophy and philosophy of law and was editor
of the Journal of Applied Philosophy from 2001 to 2013.
xiii
Simon Whitby
Simon Whitby works at the interface between life and associated science and
national security communities to address the threat of deliberate disease in
the context of rapidly advancing dual-use science and technology. Whitbys
work contributes to the discourse on dual-use biosecurity and bioethics and
thus on raising awareness at governmental, civil society, life science and
industry levels about the ethical, legal and social implications of life science
and associated science research. He has been actively engaged in building a
worldwide capability in dual-use bioethics to engage life and associated science
communities in awareness-raising programs about the importance of responsible
conduct of life-science research. Significantly, he has developed a novel and
innovative online distance-learning masters-level train-the-trainer program as
well as short courses in applied dual-use biosecurity/bioethics.
Jim Whitman
Jim Whitman is Professor of Global Governance in the Department of Peace
Studies, University of Bradford, and Director of Postgraduate Studies for the
School of Social and International Studies. His latest book, Governance Challenges
of Nanotechnology, will be published by Palgrave in 2014.
xiv
Acknowledgments
A Wellcome Trust award, Building A Sustainable Capacity in Dual-Use
Bioethics, supported a January 2010 workshop titled Promoting Dual Use
Ethics at The Australian National University that underpinned the development
of this volume. A grant by the Research Councils UK (ES/K011308/1) enabled
Brian Rappert and Malcolm Dando to undertake work for their chapters.
An award from the Economic and Social Research Council covered the production
costs associated with making this volume Open Access through ANU E Press.
xv
Abbreviations
AAAS
AMA
ANU
APS
ASM
BBSRC
BDP
BIO
BMA
BNLWRP
BSL
BSLIII
Bt
BTWC
BW
BWC
CAPPE
CBA
CBRN
CBW
CISSM
COB
CSAIL
CSIS
CWC
DARPA
DDE
DHHS
DHS
DIY
DOC
EGE
ELSI
ENSN
American Association for the Advancement of Science

American Medical Association
The Australian National University
American Phytopathological Society
American Society for Microbiology
Biotechnology and Bioscience Research Council
Biorisk Reduction for Dangerous Pathogens program
Biotechnology Industry Organization
British Medical Association
Bradford Non-Lethal Weapons Research Programme
biosafety level
Biosafety Level Three facility
Bacillus thuringiensis
Biological and Toxin Weapons Convention
biological weapons
Biological Weapons Convention
Centre for Applied Philosophy and Public Ethics
costbenefit analysis
chemical, biological, radiological and nuclear
chemical and biological weapons
Center for International and Security Studies at Maryland
cat out of the bag
Computer Science and Artificial Intelligence Laboratory
Centre for Strategic and International Studies
Chemical Weapons Convention
Defense Advanced Research Projects Agency
doctrine of double effect
Department of Health and Human Services (US)
Department of Homeland Security (US)
do-it-yourself
Department of Commerce (US)
European Group on Ethics
ethical, legal and social implications
European Neuroscience and Society Network
xvii
EU
EVC
EVP
FBI
FDA
fMRI
GE
HAC
HSP
IACUCs
IAEA
IAP
IASB
IBC
ICPS
ICRC
INN
IPR
IRBs
ISAAA
ISU
IUBMB
MAS
MIT
MOOTW
MRC
MTA
NATO
NCD
NEST
NET
NGO
NIEHS
NIH
NMHP
NNI
xviii
European Union
expected-value criterion
expected-value procedure
Federal Bureau of Investigation (US)
Food and Drug Administration (US)
functional magnetic resonance imaging
genetic engineering
holistic arms control
Harvard Sussex Project
institutional animal care and use committees
International Atomic Energy Agency
InterAcademy Panel
Industry Association Synthetic Biology
institutional biosafety committee
International Consortium for Polynucleotide Synthesis
International Committee of the Red Cross
International Neuroethics Network
intellectual property rights
institutional review boards
International Service for the Acquisition of Agri-biotech
Applications
Implementation Support Unit (UN)
International Union of Biochemistry and Molecular Biology
marker-assisted selection
Massachusetts Institute of Technology
military operations other than war
Medical Research Council (UK)
material transfer agreement
North Atlantic Treaty Organisation
Neuroscience Center at Dartmouth
New and Emerging Science and Technology
Neuroethics New Emerging Team
non-governmental organisation
National Institute of Environmental Health Sciences (US)
National Institutes of Health (US)
no means to harm principle
National Nanotechnology Initiative (US)
Abbreviations
NPG
NRA
NRC
NRC
NSABB
NSF
ONR
OPCW
PCSBI
PHAC
PI
PNAS
PP
PTSD
QALYs
QTL
RAC
R&D
RCR
rEVP
RNJMS
SAP
S&T
SEU
sPP
TB
TMS
UK
UN
UNESCO
UNGA
UNMOVIC
UNODA
UNSC
UNSCOM
Nature Publishing Group

National Research Act (US)
Nuclear Regulatory Commission
National Research Council
National Science Advisory Board for Biosecurity
National Science Foundation
Office of Naval Research (US)
Organisation for the Prevention of Chemical Weapons
Presidential Commission for the Study of Bioethical Issues (US)
Public Health Agency, Canada
principal investigator
Proceedings of the National Academy of Sciences
precautionary principle
post-traumatic stress disorder
quality-adjusted life-years
quantitative trait locus
Recombinant DNA Advisory Committee (US)
research and development
responsible conduct of research
restricted expected-value procedure
Rutgers New Jersey Medical School
science affects people
science and technology
subjective expected utility
strong versions of the precautionary principle
tuberculosis
transcranial magnetic stimulation
United Kingdom
United Nations
United Nations Educational, Scientific and Cultural
Organisation
UN General Assembly
United Nations Monitoring, Verification and Inspection
Commission
UN Office for Disarmament Affairs
United Nations Security Council
United Nations Special Commission
xix
US
USDA
VBM
VERTIC
WHA
WHO
WMA
WMD
xx
United States of America

US Department of Agriculture
valuable biological materials
Verification Research, Training and Information Centre
World Health Assembly
World Health Organisation
World Medical Association
weapons of mass destruction
Introduction
1. Ethics and Dual-Use Research

Michael J. Selgelid
[W]ho is most capable of treating friends well and enemies badly in matters
of disease and health?
A doctor.
Isnt the person most able to land a blow, whether in boxing or any other
kind of fight, also most able to guard against And the one who is most
able to guard against disease is also most able to produce it unnoticed?1
Nuclear physics and contemporary genetics

One of the most dramaticand ethically problematicepisodes in the history
of science involved the making and use of the first atomic weapons. When key
discoveries of things like atomic fission and the chain reaction were made during
the revolution in physics under way during the first half of the twentieth century,
the scientists involved realised that the new knowledge gained might be used
for both good purposes (for example, energy production) and bad purposes (for
example, weapons making).2 The first atomic bombs were developed soon after
the discovery of the chain reaction in particular. The atomic bombs dropped on
Hiroshima and Nagasaki during World War II caused hundreds of thousands of
deaths.3 Since that time vast resources (which might have been used for other
purposes) have been devoted to nuclear weapons production, and humanity has
lived under the threat of nuclear holocaust for decades.
An implication of the contemporary scientific revolution in genetics and
biotechnology is that the life sciences, at present, are in a situation analogous
to that of physics when the key discoveries mentioned above were made. The
very same biological discoveries that might be used to benefit humanity (for
example, via advances in medicine) can sometimes also be used to cause harm
(for example, via biological weapons). In some cases, the harms in question
could be catastrophic. This is thus a key moment in the history of biologyand
a crucial time for ethical decision-making and policymaking in the life sciences.
1 Plato 1992, Republic, G. M. A. Grube (trans.), C. D. C. Reeve (rev.), Hackett, Indianapolis, pp. 59.
2 Schweber, S. S. 2000, In the Shadow of the Bomb, Princeton University Press, Princeton, NJ.
3 Rhodes, R. 1986, The Making of the Atomic Bomb, Simon & Schuster, New York.
3
Research with potential to be used for both good and bad purposes is now
commonly referred to as dual-use research. While almost any knowledge and
technology can be used for both kinds of purposes, the expression dual-use
research of concern4 is used to refer to research that can be used for especially
harmful purposesthat is, where the consequences of malevolent use would
be potentially catastrophic. Of particular concern are advances in genetics that
might enable development of a new generation of biological weapons of mass
destruction. (In the remainder of this chapter, I will use the expression dualuse research as shorthand for the expression dual-use research of concern.)
Controversial cases
Such danger is illustrated by the following controversial studies that have been
published during the new millennium.
The mousepox experiment

Australian scientists used genetic-engineering techniques to insert an interleukin
(IL-4) gene into the mousepox virus. Their aim was to develop a strain of
mousepox that would make mice infertileand thus provide a potentially
powerful new means of pest control. They unexpectedly discovered, however,
that the altered virus killed both mice that were naturally resistant to and mice
that had been vaccinated against ordinary mousepox. They published these
findings, along with description of materials and methods, in the Journal of
Virology in 2001.5 A danger is that the same techniques might enable production
of vaccine-resistant smallpox. Smallpox is one of the most feared biological
weapons agentsand vaccine is our only defence against it.
Synthetic polio
Following the map of the polio genome (published on the Internet), American
scientists (via mail order) bought and strung together corresponding strands of
DNA. Addition of the synthesised genome to cell juice (a solution containing
cellular ingredients, but no live cells) led to production of live polio virus that
paralysed and killed mice. The scientists aims were, inter alia, to show that
4 National Science Advisory Board for Biosecurity (NSABB) 2007, Proposed Framework for the Oversight
of Dual Use Life Sciences Research: Strategies for Minimizing the Potential Misuse of Research Information,
National Science Advisory Board for Biosecurity, Bethesda, Md, <http://oba.od.nih.gov/biosecurity/pdf/
Framework%20for%20transmittal%200807_Sept07.pdf> (viewed 7 April 2013).
5 Jackson, R. J., Christensen, C. D., Beaton, S., Hall, D. F. and Ramshaw, I. A. 2001, Expression of mouse
interleukin-4 by a recombinant ectromelia virus overcomes genetic resistance to mousepox, Journal of
Virology, vol. 75, no. 3, pp. 120510.
4
such a feat would be technically possible and to demonstrate that viruses are
ultimately just chemicals.6 They published their findings, along with description
of materials and methods, in Science in 2002.7 A danger is that similar techniques
might be used to create biological weapons agents, such as smallpox, which
bioterrorists and/or state-sponsored biological weapons programs might not
otherwise be able to access easily.
Reconstruction of 1918 Spanish flu

American researchers used similar techniques (to those used with polio) to
reconstruct the 1918 Spanish flu virus. The virus they created was more deadly
to chickens than any flu virus that had previously been studied. The purpose
of this research was to yield knowledge that would facilitate protection against
possible future influenza pandemicsfor example, vaccine production. They
published their findings, along with description of materials and methods, in
Science in 2005.8 A danger is that malevolent actors (for example, bioterrorists)
might use the published information to create and unleash the virus in
questionwhich was responsible for one of the worst epidemics in human
history, killing 20 to 100 million people over the course of a year or two.
Transmissible H5N1
The most recent and, to date, most controversial dual-use life-science research
involved the study of H5N1 (avian) influenza transmissibility among ferrets,
which provide the best model for influenza in humans. While it is estimated that
H5N1 kills 60 per cent of humans infected, it is not (currently) transmissible
between humans. Researchers in the Netherlands and the United States thus
conducted experiments that aimed to determine whether H5N1 might develop
into a human-to-human transmissible strain. Genetic engineering of the virus
and passaging of the altered virus between ferrets led to creation of strains
that were airborne and easily transmissible among ferretsthus indicating that
natural evolution of a human-transmissible strain of H5N1 might be possible.
Much debate surrounded the question of whether or not this research should
be published in detail. On the one hand, it was argued that publishing these
studies was important because this would facilitate vaccine development and/
or surveillance of relevant changes to H5N1 occurring in nature. The hope with
regard to surveillance is that this would enable earlier detection of emerging
6 Selgelid, M. J. and Weir, L. 2010, Reflections on the synthetic production of poliovirus, Bulletin of the
Atomic Scientists, vol. 66, no. 3, pp. 19.
7 Cello, J., Paul, A. V. and Wimmer, E. 2002, Chemical synthesis of poliovirus cDNA: generation of infectious
virus in the absence of natural template, Science, vol. 297, pp. 101618.
8 Tumpey, T. M. et al. 2005, Characterization of the reconstructed 1918 Spanish influenza pandemic virus,
Science, vol. 310, no. 5745, pp. 7780.
5
pandemic strains and thus earlier implementation of protective measures. On

the other hand, others argued that a human-transmissible strain of H5N1 could
kill millions or perhaps even billions of people if produced and unleashed by
bioterrorists or other malevolent actors (or in the event of accidental release
from research laboratories). The US National Science Advisory Board for
Biosecurity (NSABB) in December 2011 recommended that the findings of these
studies be published, but that detailed description of materials and methods be
omitted from the published articles.9 After a highly publicised World Health
Organisation (WHO) meeting reached the opposite conclusion in February
2012recommending that the studies eventually be published in full10the
NSABB reversed its initial decision in March 2012.11 Revised (fully detailed)
versions of the papers in question were finally published in Science12 and
Nature13 in June 2012. Part of the reason for the NSABBs reversal of decision
apparently involved misunderstanding regarding whether or not the airborne
transmissible strains were deadly to ferrets. While much of the initial public
discussion indicated that the strains produced in the Netherlands were just as
deadly as ordinary H5N1, it was later revealed that airborne strains created
in the lab were not deadly to ferrets. Another cited reason for the NSABBs
reversed decision is that the revised manuscripts they eventually approved
provided additional data regarding ways in which the studies might have public
health benefitsfor example, via surveillance.
Levels of governance
The dual-use phenomenon requires ethical decision-making by various actors
at different levels of the science governance hierarchy. Individual scientists
(insofar as they are at liberty) must decide what research to conduct and/or
publish. Research institutions (insofar as they are at liberty) must decide how to
regulate potentially dangerous research within their confines; how to educate
9 US National Institutes of Health 2011, Press statement on the NSABB review of H5N1 research, <http://
www.nih.gov/news/health/dec2011/od-20.htm> (viewed 7 April 2013).
10 World Health Organisation (WHO) 2012, Public health, influenza experts agree H5N1 research critical,
but extend delay, <http://www.who.int/mediacentre/news/releases/2012/h5n1_research_20120217/en/>
(viewed 7 April 2013).
11 National Science Advisory Board for Biosecurity (NSABB) 2012, March 2930, 2012 meeting of the
National Science Advisory Board for Biosecurity to review revised manuscripts on transmissibility of A/H5N1
influenza virus, <http://oba.od.nih.gov/oba/biosecurity/PDF/NSABB_Statement_March_2012_Meeting.
pdf> (viewed 7 April 2013).
12 Herfst, S., Schrauwen, E. J. A., Linster, M., Chutinimitkul, S., de Wit, E., Munster, V. J., Sorrell, E. M.,
Bestebroer, T. B., Burke, D. F., Smith, D. J., Rimmelzwaan, G. F., Osterhaus, A. D. M. E. and Fouchier, R. A.
M. 2012, Airborne transmission of influenza A/H5N1 virus between ferrets, Science, vol. 22, pp. 153441.
13 Imai, M., Watanabe, T., Hatta, M., Das, S. C., Ozawa, M., Shinya, K., Zhong, G., Hanson, A., Katsura,
H., Watanabe, S., Li, C., Kawakami, E., Yamada, S., Kiso, M., Suzuki, Y., Maher, E. A., Neumann, G. and
Kawaoka, Y. 2012, Experimental adaptation of an influenza H5 HA confers respiratory droplet transmission
to a reassortant H5 HA/H1N1 virus in ferrets, Nature, vol. 486, pp. 4208.
6
researchers working there (regarding dual use and/or ethics); what laboratory
security measures to put into place, and so on.14 Professional societies must
make decisions about the development, promulgation and/or enforcement of
ethical codes of conduct for scientistsand/or decisions about relevant (ethical)
education of their members. Publishers must make decisions regarding processes
of review of papers posing dual-use dangersand they must ultimately decide
which papers to publish. National governments must decide what research
to fund, and the extent to (or manner in) which things like research review,
publication review and/or relevant education of scientists will be mandated
and they must make decisions about the extent to which controls should be
placed on access to potentially dangerous materials. International governance
bodies (such as the WHO), finally, must make relevant decisions concerning
global policyfor example, whether or not there should be international
guidelines regarding dual-use research, or oversight thereof, and/or what the
content of such guidelines should be.
Ethical dilemmas
In all cases the decisions will be difficult. On the one hand, responsible actors
will want to take actions that will promote the development and use of beneficial
science. On the other hand, they will want to take actions that will prevent
the malevolent use of science (which might sometimes require avoidance of
generation and/or publication of potentially dangerous information). An
implication of the dual-use phenomenon, however, is that it is inherently
difficult to achieve both goals at the same time in the cases where the very same
research that is likely to be beneficial might also be used to cause harm. In the
case of governmental decision-making, for example, a laissez-faire approach to
scientific governance might facilitate scientific advance and the benefits thereby
enabledbut it might also lead to especially dangerous research getting done
and/or published. A more restrictive approach, on the other hand, might
prevent generation and/or publication of dangerous informationbut it might
also stifle beneficial scientific advance at the same time. Hence the expression
dual-use dilemma.
In any case, it is important to recognise that key decisions posed by dual-use
research are inherently ethical in nature. The decisions faced by the various
actors enumerated above largely concern: 1) the responsibilities of the actors
in question (for example, to what extent would a scientist be responsible if her
research is used to cause harm?); 2) issues of how one should go about promoting
14 Research institutions, of course, already do such things to varying degreesand numerous relevant
measures (for example, regarding biosafety) are required by law. The point here, however, is that additional
new measures are required to address dual-use research in particular.
7
benefits while avoiding harms or reducing risks (for example, should a paper be
published if this has a good chance of promoting a significant amount of human
wellbeing but a small chance of causing disaster?); and/or 3) questions about
values and value conflict (for example, how should governments strike a balance
between the goal to promote scientific freedom/progress and the goal to promote
security?). Issues regarding responsibilities, harms, benefits and valuesand,
ultimately, what ought to be doneare exactly the kinds of things that ethics
is about.
In the meantime, however, much of the debate about dual-use research has
involved scientists and security experts rather than ethicists in particular.15
While much bioethical discussion has focused on research ethics and the
ethical implications of genetics/biotechnology, it is ironic that relatively little
bioethical discussion has, to date, focused on dual-use research in particular.
This is unfortunate because, given the potential for catastrophic consequences,
dual-use research is surely one of the most important ethical issues regarding
research and genetics/biotechnology.
This volume
The chapters in Part I of this volume, Dual Use in Context, map the terrain
of dual-use issues emerging in various areas of life-science researchthat is:
nanotechnology (Chapter 2, Jim Whitman); neuroscience (Chapter 3, Valentina
Bartolucci and Malcolm Dando); synthetic biology (Chapter 4, Alexander Kelle);
agriculture (Chapter 5, Simon Whitby); and tuberculosis (Chapter 6, Nancy
Connell).
The chapters in Part II, Ethical Frameworks and Principles, explore the
relevance of various existing philosophical frameworks and/or tools of ethical
analysis to the dual-use problem (in cases where their relevance to dual-use
problems has received relatively little previous exploration). In particular, they
examine the impact of environments and institutions on moral development
and ethical reasoning (Chapter 7, Judi Sture); the ethics of weapons research in
general (and the relevance thereof to dual-use research in particular) (Chapter
8, John Forge); application of rational decision theory to the dual-use problem
(Chapter 9, Thomas Douglas); the relevance of the doctrine of double effect to
the dual-use problem (Chapter 10, Suzanne Uniacke); implications of uncertainty
for dual-use decision-making (Chapter 11, Michael Smithson); collective-action
problems associated with dual-use research (Chapter 12, Seumas Miller); the
15 Selgelid, M. J. 2010, Ethics engagement of the dual use dilemma: progress and potential, in B. Rappert
(ed.), Education and Ethics in the Life Sciences: Strengthening the Prohibition of Biological Weapons, ANU E
Press, Canberra, pp. 2334.
8
relevance of just-war theory to the dual-use problem (Chapter 13, Koos van
der Bruggen); and the relevance of the precautionary principle to dual-use
decision-making and policymaking (Chapter 14, Steve Clarke).
Parts III and IV, Ethical Practices and Ethical Futures, consider existing,
developing and future practices and policies regarding ethics and dual-use lifescience researchthat is, the prospect of self-regulation by scientists (Chapter
15, David Resnik); lessons learnt from the history of nuclear physics (Chapter
16, Nicholas Evans); dual-use governance in developing countries (Chapter 17,
Louise Bezuidenhout); the responsibilities of individual scientists in the context
of incapacitating chemical and toxin agents (Chapter 18, Michael Crowley);
and the WHOs project on Responsible Life Sciences Research (Chapter 19,
Emmanuelle Tuerlings and Andreas Reis). The Conclusion (Chapter 20) offers a
wide-ranging and agenda-setting summary by Brian Rappert.
This volume is a product of a workshop (funded by the Wellcome Trust) on
Promoting Dual Use Ethics (organised by Michael Selgelid and Brian Rappert)
held at the Centre for Applied Philosophy and Public Ethics (CAPPE) at The
Australian National University in Canberra in January 2010.
Part I: Dual Use in Context
2. Nanotechnology and
Dual-Use Dilemmas
Jim Whitman
Nanotechnology development and the

accommodation of risk
The abundant dual-use potential of nanotechnology arises from the fact it is not
materials-specific in either the organic or the inorganic realms. At the nanometre
scale, familiar materials can display altered, unexpected and/or amplified
qualities including tensile strength, viscosity, conductivity and antimicrobial
properties. Our ability to manipulate matter at the molecular and even atomic
levels opens up a breathtaking range of possibilities in engineering, medicine,
materials science, energy and computing, to list but a few. At the nanometre
level (a nanometre is one-billionth of a metre, approximately the diameter of a
strand of DNA), disciplinary boundaries between biology and chemistry have
become indistinct; and novel techniques notwithstanding, nanobiotechnology
is biotechnology conducted at the nano level.1 Technological convergencethe
project to create a single engineering paradigm combining nanotechnology,
information technology, biotechnology and cognitive scienceis being actively
pursued in both the United States and the European Union.2 These developments
have already extended our understanding of what science and technology can
achieve.
Although the proponents of nanotechnology and technological convergence
exalt the possibilities for human betterment, a great many malfeasant objectives
and adaptations are not merely a logical possibility, but also a risk integral to
their development and dissemination.
1 Many issues in nanoscience and nanotechnology turn on the practical aspects of disciplinarity, for
proponents of nanotechnology, science educators and practising scientists themselves. See Wienroth, M. 2002,
Disciplinarity and research identity in nanoscale science and technologies, in J. S. Ach and C. Weidemann
(eds), Size Matters: Ethical, Legal and Social Aspects of Nanobiotechnology and Nano-Medicine, Lit Verlag,
Berlin, pp. 15777.
2 Roco, M. C. and Bainbridge, W. S. (eds) 2002, Converging technologies for improving human performance:
nanotechnology, biotechnology, information technology and cognitive science, National Science Foundation/
Department of Commerce-sponsored Report, <http://www.wtec.org/ConvergingTechnologies>; Nordmann,
A. 2004, Converging TechnologiesShaping the Future of European Societies, <http://www.ntnu.no/2020/pdf/
final_report_en.pdf>.
13
Nanotechnology: Pace, promise and peril

The promise of nanotechnology might well be overstated in some quarters, but
there can be no doubting the political importance that has accompanied and
furthered its establishment. It is now sufficiently well embedded in research and
industrial settings in more than 60 countries that the appearance of nanomaterials
in a range of mundane consumer products is unremarkable.3 Indeed, such is the
rate of research breakthroughs and the development of new nanotechnology
products and processes that our deliberative systems and established means
of regulating them are already being outpaced.4 The following, from a report
on nanomaterials by the Royal Commission on Environmental Pollution,
under its depiction of risk governance, can also be read with a view to our
capacities to apply dual-use bioethics across the full span of nanotechnology
and nanobiotechnology:
[Currently], regulators face a Sisyphean task. Innovation is, or soon
will be, driving new products onto the market at rates that are orders
of magnitude faster than they can currently hope to manage with the
resources at their disposal. We heard from one regulatory body that it
was not even considering how to address third and fourth generation
nanomaterials because they were fully occupied with those currently
at the commercial stage. The magnitude of the task combined with
constraints on resources tends to create an attitude of regulatory
fatalism One expert likened the challenge of risk governance in this
field to that of shouting a warning to the driver of an express train as it
thunders past.5
And, in the commissions view, what is true for regulatory deliberation also
holds for public consideration more generally:
There remains in our view a real question about whether the capacity
for deliberation and (perhaps even more so) for public engagement in
modern democracies is sufficient to sustain an approach that seeks to
interrogate scientific and technological developments. The more specific
the focus, the more numerous the cases will be, but as the focus becomes
more generalized (looking at nanotechnologies as a whole for example),
the range of possible applications and implications threatens to make
dialogue unmanageable. Informed and inclusive deliberation (however
3 See the inventory of nanotechnology-based consumer products at The Project on Emerging
Nanotechnologies, Woodrow Wilson Center, <http://www.nanotechproject.org/inventories/consumer/>.
4 Whitman, J. 2007, The challenge to deliberative systems of technological systems convergence,
Innovation: The European Journal of Social Sciences, vol. 20, no. 4, pp. 32942.
5 Royal Commission on Environmental Pollution 2008, Novel Materials in the Environment: The Case of
Nanotechnology, <http://www.official-documents.gov.uk/document/cm74/7468/7468.pdf>, p. 66.
14
2. Nanotechnology and Dual-Use Dilemmas
conducted) on a huge range of potential developments seems as distant

a prospect as the resolution of many of the technical uncertainties
identified elsewhere.6
The commissions concerns appear to be validated by the difficulties that
have met the many efforts to engage interested publics in serious ethical as
well as practical deliberations about the likely impacts and implications of
nanotechnology-driven change:
This commitment to upstream public engagement raises many
unresolved questions. At what stages in scientific research is it realistic
to raise issue of public accountability and social concern? How and on
whose terms should such issues be debated? Are dominant frameworks
of risk, ethics and regulation adequate? Can citizens exercise any
meaningful influence over the pace, direction and interactions between
technological and social change? How can engagement be reconciled
with the need to maintain the independence of science, and the economic
dynamism of its applications?7
Even as these morally weighty yet essentially abstract considerations are
being aired, however, the development of nanotechnology is forging ahead,
with practical and moral implications that can only be guessed at, as the royal
commission describes with respect to whether nanomaterials might pose serious
risks to human health and ecosystem integrity.
Determining the fate of novel materials is vital when assessing the
toxicological threat they pose. Nanomaterials are illustrative of the
challenge. Techniques for their routine measurement in environmental
samples are not widely available, nor are we currently able to determine
their persistence in the environment or their transformation into
other forms. Laboratory assessments of toxicity suggest that some
nanomaterials could give rise to biological damage. But to date, adverse
effects on populations or communities of organisms in situ have not been
investigated and potential effects on ecosystem structure and processes
have not been addressed. Our ignorance of these matters brings into
question the level of confidence that we can place in current regulatory
arrangements.8
6 Ibid., pp. 745.

7 Kearnes, M., Macnaghten, P. and Wilsdon, J. 2006, Governing at the Nanoscale: People, Policies and
Emerging Technologies, Demos, London.
8 Royal Commission on Environmental Pollution, op. cit., p. 6. See also Morris, J., de Martinis, D., Hansen, B.,
Sintes, J. R., Kearns, P. and Gonzalez, M. 2011, Science policy considerations for responsible nanotechnology
decisions, Nature Nanotechnology, vol. 6, pp. 737.
15
When set against the products and promise of nanotechnology, the very
considerable risks and unknowns of the diffusion of nanomaterials might
be expected to generate a great many dilemmas both at policymaking and at
laboratory levels. Since these are not much in evidence either in public debate or
in the momentum of nanoscientific advance, it is clear that the powerful interests
involved and the expected gains and competitive advantages carry more weight
than either ethical disquiet or practical concerns. Although this is hardly a novel
position (of 30 000 bulk chemicals in use in the European Union, only 3000 or
so have been formally assessed for health and environmental effects),9 the moral
significance and practical risks involved in nanotechnologies on present and
planned scales cannot be regarded much as one might an incremental addition to
an already chemicalised environment. After all, the transformative qualities of
nanotechnology have been much touted by its proponents, which accounts for
the acknowledgment of serious ethical and societal implications that have found
expression in the academic as well as policy and policy-related literatures.10
The acknowledged possibility of deleterious but unwanted outcomes of
nanotechnology, however, in the form of kinds and degrees of toxicitythat is,
a biochemical risk of a familiar sort, albeit perhaps on an exceptional scaleis
not an end to the matter. It is in the realm of the many uses and adaptations of
nanotechnology in practically every field of the physical and medical sciences and
engineering that the larger dangers resideand as part of them, the prospect of
an explosion of dual-use issues. These are routinely gathered under the generic
term societal implications, with ethical issues often given freestanding
consideration, by which it is unclear where military nanotechnology is (or
should be) sited. After all, the research and development of military uses and
adaptations of nanotechnology are by no means either merely prospective or
peripheral to the establishment of nanotechnology. Although many of these
are at least nominally defensive,11 active research and already articulated warfighting possibilities and other means of inflicting serious harm have not been
9 Royal Commission on Environmental Pollution, op. cit., p. 50.

10 See, for example, Roco, M. C. and Bainbridge, W. S. (eds) 2001, Societal Implications of Nanoscience
and Nanotechnology, Kluwer Academic Publishers, Dordrecht; Roco, M. C. and Bainbridge, W. S. (eds) 2007,
Nanotechnology: Societal Implications II: Individual Perspectives, Springer, Dordrecht; The coevolution of
human potential and converging technologies, Annals of the New York Academy of Sciences, vol. 1013 (May
2004); Progress in convergence: technologies for human wellbeing, Annals of the New York Academy of
Sciences, vol. 1093 (2006); Banse, G., Grunwald, A., Hronszky, I. and Nelson, G. (eds) 2007, Assessing Societal
Implications of Converging Technological Development, Edition Sigma, Berlin; Allhoff, F., Lin, P., Moor, J. and
Wekert, J. (eds) 2007, Nanoethics: The Ethical and Social Implications of Nanotechnology, Wiley-Interscience,
Hoboken, NJ.
11 Kosal, M. E. 2009, Nanotechnology for Chemical and Biological Defense, Springer, New York; see also the
MIT Institute for Soldier Nanotechnologies, <http://web.mit.edu/isn/>.
16
precluded.12 Of the 13 US federal agencies which shared $1.6 billion under the
US National Nanotechnology Initiative for 2009, the Department of Defense was
allocated 28 per cent.13
Military uses and adaptations of nano-enabled products and processes are as
yet not concentrated around dedicated means of offensive war-fighting or new
weapons systems. Instead, the foreseeable nanotechnology developments most
likely to be adopted for military purposes have a very wide range of applications:
smart materials, micro-electromechanical systems, nanocomputing, robotics,
micro-sensors and many more. Jurgen Altmann has noted that if the production
facilities for raw material, feedstock, energy, and final products as well as the
transport systems are themselves produced by molecular nanotechnology,
a very fast increase of the production and distribution of military goods is
possible, and one possible outcome of this is that molecular nanotechnology
production of nearly unlimited numbers of armaments at little cost would
contradict the very idea of quantitative arms control and would culminate in
a technological arms race beyond control.14 In any event, since the worldwide
growth of nanotechnology research and development has been propelled by
the expectation that it will usher in a new industrial revolution, an age of
transitions and other epoch-defining transformations, government sanction of
nanotechnology developments has always had a strongly competitive edge
made all the more sharp now that it has been linked to conceptions/aspects of
national security.15 Realist fears inform the development of security dilemmas;
and the dangers are already apparent: By taking cues from Moscows centralized
and opaque institutions, Washington risks misperceiving Russias intentions
and calculations as well as prematurely locking into strategic competition.
The result, if unaltered, could convert the promise of nanotechnology into a
new realm of commercial rivalry and arms racing.16
The perils of nanotechnology, however, are not limited to arms races of the
sort that can, at least in principle, be addressed by dedicated arms-control
initiatives.17 There has already been concern over the comprehensiveness of the
12 Altman, J. 2006, Military Nanotechnology, Routledge, London. See also Shipbaugh, C. 2006, Offensedefense aspects of nanotechnologies: a forecast of military applications, Nanotechnology, pp. 7417; Kharat,
D. K., Muthurajan, H. and Praveenkumar, B. 2006 Present and future military applications of nanaodevides,
Synthesis and Reactivity in Inorganic, Metal-Organic and Nano-Metal Chemistry, vol. 36, pp. 2315.
13 US National Nanotechnology Initiative, Funding, <http://www.nano.gov/html/about/funding.html> .
14 Altman, op. cit.
15 Vandermolen, LCDR T. D. 2006, Molecular nanotechnology and national security, Air & Space Power
Journal, pp. 96106.
16 Stulberg, A. N. 2009, Flying blind into a new military epoch: the nanotechnology revolution, emerging
security dilemmas, and Russias double-bind, Paper presented at the International Studies Association
Conference, February 2009.
17 Whitman, J. 2011, The arms control challenges of nanotechnology, Contemporary Security Policy, vol.
32, no. 1, pp. 99115.
17
Biological Weapons Convention (BWC) and the Chemical Weapons Convention

(CWC) relative to the malfeasant purposes to which nanomaterials and products
can be put to use:
Article 1 of the BWC prohibits the development, production, and
possession of microbial or other biological agents, or toxins whatever
their origin or method of production Assuming microbial or biological
agents should be considered living things, can a toxin be something
inorganic or artificially created? Since the phrase whatever their
origin or method of production relates to toxins because of the comma
placement, it seems to include so-called mechanical devices that could
result from mature nanotechnology. In this sense, one can argue that
nanotechnologyparticularly nanorobotscan be treated as a toxin if
it causes harm similar to already known toxins. However, because the
BWC seems to deal only with biological organisms or products thereof,
a strong argument can be made that the artificially assembled products
that nanotechnology would produce cannot possibly fit under the
BWC. Perhaps the only way nanotechnology can fall under the BWCs
prohibitions without a doubt is if it were used to artificially create
replicas of known biological weapons or toxins. Only then would it
clearly be covered since no difference would exist between the natural
product and the artificial version. However, nanotechnology as a field
is much broader than these narrow biological replica applications.18
Nanorobots are not first and foremost an arms-control issue, but a dual-use one,
with the main lines of current research in the field dedicated to bio, industrial
and medical nanorobots.19 The wide applicability and adaptability of nanoenabled products and processes mean that their dual-use potential is more often
than not clearly evident. For example, where a nano-enabled medical advance
also holds the promise of more effective delivery of a chemical or biological
agent for offensive purposes, one might expect this potential to find expression
as a dilemma of considerable weight and intractability. Measured against the
many considerable benefits that can arise from their many uses, however, are the
research and development of any dual-use nano-development likely to present
practising scientists, technologists or policymakers with a dilemma?
18 Pinson, R. D. 2004, Is nanotechnology prohibited by the Biological and Chemical Weapons Conventions?
Berkeley Journal of International Law, vol. 22, no. 2, p. 298.
19 See the Special Issue of The International Journal of Robotics Research, vol. 28, no. 4 (April 2009): Current
State of the Art and Future Challenges in Nanorobotics.
18
Ethics, applied ethics and dual-use dilemmas

Implicit in the idea of a dual-use dilemma is that the beneficent purposes and/
or uses of a technology are roughly equal to the potential harm that could be
brought about by pernicious applications. If harmful and destructive uses for
a developing scientific or technological advance can be foreseen but are not
expressed, or not expressed as a dilemma, the dual-use potential will not be
sufficient on its own to necessitate the kinds of deliberation (scientific, managerial,
legal, political) that will regulate, slow, modify or halt the development in
question. Subsequent concerns about risks and dangers then become post-facto
legislative and law-enforcement matters. Similarly, regulatory initiatives such as
export bans to dangerous regimes or to untrustworthy end users do not address
dilemmas, but arise from cost and risk/benefit analyses.
Within established scientific and technological arenas, real and potential dualuse dilemmas are not abstract: their visibility in every sense is conditioned
by some combination of political priorities; inter-business and international
competitive impulses; patterns of public and private financing; and by the
narrowness of specialist research that can contribute to, rather than exhibit,
dual-use potential. Acknowledging this does not oblige one to subscribe to
one or another form of technological determinism, but it takes into account
that the proclaimed benefits on the plus side of dual use can serve promotional
purposes (of which, beware nano-hype);20 and that they are rarely if ever wholly
disinterested. As discussed further below, these and related considerations
narrow the range of actors likely to perceive or to voice dual-use dilemmas,
as well as the circumstances when they will be able or likely to do so. At the
same time, although many kinds of moral significance and forms of ethical
predicament can follow in the wake of scientific and technological advances,
these do not all take the form of dilemmas; there are, for example, detailed,
ethical arguments both for and against various forms of human enhancement
through genetic engineering.21
As a general consideration, dual-use dilemmas leave open the question of the
affected and/or responsible agentwho or what experiences the dilemmaand
indeed, whether such dilemmas are framed in ethical rather than, or in addition
to, prudential considerations. For the purposes of this argument, it can be
assumed that any dilemma arising from the possible misapplication of biological
knowledge will entail practical dangers of considerable moral significance. We
20 Toumey, C. 2004, Nano hyperbole and lessons from earlier technologies, Nanotechnology Law and
Business, vol. 1, no. 4, pp. 397405; Berube, D. M. 2006, Nano-Hype: The Truth Behind the Nanotechology
Buzz, Prometheus Books, Amherst, Mass.
21 Harris, J. 2007, Enhancing Evolution: The Ethical Case for Making Better People, Princeton University
Press, Princeton, NJ; Sandel, M. J. 2007, The Case against Perfection: Ethics in An Age of Genetic Engineering,
Belknap Press, Harvard, Mass.
19
can read as much into the definition of dual-use research of concern proposed
by the US National Science Advisory Board for Biosecurity: research that, based
on current understanding, can be reasonably anticipated to provide knowledge,
products, or technologies that could be directly misapplied by others to pose
a threat to public health, agriculture, plants, animals, the environment, or
materiel.22 Moral unease, however, is never automatically congruent with
estimates of the practical significance of possible dangers; and it is in the nature
of a dilemma that countervailing considerations are of roughly equal weight. In
addition, countervailing considerations in life-science dual-use conundrums
that is, those that bolster the case for either the positive or the negative aspects
of the dual use to be determinedmost often entail structural considerations.
This is because of the range of powerful interests involved and because the
encompassing arenas are international and/or global.
The application of bioethics or ethical reasoning more generally is not a matter of
informed individuals unfailingly applying knowledge to circumstance; and this
applies to scientists no less than to any other moral agents.23 In his comprehensive
study of scientists whose research was furthered by collaboration with the
Nazi regime, John Cornwell depicts the striking similarity of pressures facing
scientists then and now: The greatest pressures on the integrity of scientists
are exerted at the interface between the professional practice of science and
the demands of award-giving patrons. The narrative of scientific collaboration
with the Nazi regime reveals the pressures of hubris, loyalty, competition and
dependence leading to compromise. In the final analysis the temptation was a
preparedness to do a deal with the Devil in order to continue doing science.24
Even when there is no obvious devil in the form of a psychopathological regime,
scientific research can but rarely be understood as pure in the sense that it
can wholly be abstracted from the potential for pernicious uses; and ethical
concerns are embedded in risk/benefit calculations that do not present in clear,
dichotomous form. So it is that particular cases of the dissemination of biological
knowledge through open, peer-reviewed publication can be seen as posing
grave risks of misuse;25 but at the same time, there is no enthusiasm for general
prohibitions on scientific exchanges, on which so much beneficent advance
depends, and for which there is no commanding legal reach.26 It follows that in
22 National Science Advisory Board for Biosecurity (NSABB) 2008, Frequently asked questions, <http://
www.biosecurityboard.gov/faq.asp#1>.
23 Whitman, J. 2010, When dual use issues are so abundant, why are dual use dilemmas so rare? Research
report for the Wellcome Trust project Building A Sustainable Capacity in Dual Use Bioethics, <http://www.
dual-usebioethics.net/>.
24 Cornwell, J. 2004, Hitlers Scientists: Science, War and the Devils Pact, Penguin, London, p. 462.
25 Indicative examples are listed in Institute of Medicine and National Research Council 2006, Globalization,
Biosecurity and the Future of the Life Sciences, The National Academies Press, Washington, DC, pp. 534.
26 Marchant, G. E. and Pope, L. L. 2009, The problems of forbidding science, Science and Engineering
Ethics, vol. 15, pp. 37594.
20
such instances, the moral and/or prudential reasoning of individual researchers,

publishers and concerned institutions will be informed and conditioned by larger
interests and concerns, many of which provide the basis on which particular
dual-use dilemmas are generatedin this case, the maintenance of an open
epistemic community that will support the range of biological sciences integral
to twenty-first-century life. The cases of nanotechnology and technological
convergence have been conceived and established not only in ways that
highlight the beneficent potential (subtitles of key works include technologies
for human wellbeing and for improving human performance), but that also
depict ethical concerns and societal implications more as challenges to be
dealt with than as outcomes to be avoided.
It is against this background that we must set the potential of bioethics to
identify, articulate and address dual-use issues in the life sciences. What needs
to be borne in mind is that, in common with other specialist forms of applied
ethics within the life sciences (neuro-ethics, nano-ethics, and so on), dual-use
bioethics operates within fields of endeavour that are engines for generating
dual-use issues; it does not bring ethical scrutiny to bear on its own operating
environments. In one sense, this is as it should be: subfields of ethics are
delimited and specialistthat is their strength, particularly as the issues they
engage increase in number and complexity. Yet the growth of specialist, applied
ethics in the life sciences also has the effect of signalling the ethical validity of
fields they are established to scrutinise, however many dual-use issues they seem
set to produce. As the number and kind of ethical quandaries arising from farreaching scientific and technological advances proliferate, attention naturally
focuses on actual or impending difficulties, rather than on the conditions that
produce them. Of course, the relationship between medicine and medical ethics
is largely unproblematic in this regard, the much larger compass of modern
biosciences and bioethics less clearly so, since enterprises such as genetic
engineering remain contested in themselves and not merely in respect of some
of their particulars or applications. Nanotechnology is still more vexed.
Yet at the same time, there is a great deal of developed-world government
anxiety that, ethical and societal implications notwithstanding, nothing should
seriously impede the entrenchment and furtherance of nanotechnology and
envisioned forms of technological convergence for which it is the foundation.27
Hence the assertion in the path-breaking US National Science Foundation/
Department of Commerce (NSF/DOC) report on converging technologies that
[p]rogress can become self-catalyzing if we press ahead aggressively; but
if we hesitate, the barriers to progress may crystalize and become harder to
27 Fisher, E. and Mahajan, R. L. 2006, Contradictory intent? US federal legislation on integrating societal
concerns into nanotechnology research and development, Science and Public Policy, pp. 516.
21
surmount.28 Similarly, the principal EU report on nano-led convergence set

out the plan for forging ahead despite acknowledging that each of the likely
characteristics of converging technology applications presents an opportunity
to solve societal problems, to benefit individuals, and to generate wealth. Each
of these also poses threats to culture and tradition, to human integrity and
autonomy, perhaps to political and economic stability.29
Still more remarkable is the following assumption contained in the NSF/DOC
report: The ability to control the genetics of humans, animals and agricultural
plants will greatly benefit human welfare; widespread consensus about ethical,
legal and moral issues will be built into the process.30
Such a starkly instrumentalist understanding of ethical deliberation was
characterised by Paul Ramsey more than 30 years ago:
We need to raise ethical questions with a serious and not a frivolous
conscience. The man of frivolous conscience announces that there are
ethical quandaries ahead that we must urgently consider before the
future catches up with us. By this he often means that we need to devise
a new ethics that will provide the rationalization for doing in the future
what men are bound to do because of new actions and interventions
science will have made possible. In contrast a man of serious conscience
means to say in raising urgent ethical questions that there may be some
things that men should never do. Good things that men do can be made
complete only by the things they refuse to do.31
The reassurance in the EU report that [e]nlightened exploitation of discoveries
in nanotechnology, biotechnology, information technology and cognitive
science will humanize technology rather than dehumanize society32 also has
the effect of making ethical deliberation appear capable of precluding dilemmas.
Moreover, it appears that the task of practitioners in the social sciences and
humanities is to act as guides across the space between anxious publics and
prepared arenas, if not fixed ends. The following (also from the EU document
cited above) on nano-facilitated technological convergence outlines their roles:
The broadly transformative potential of [technological convergence]
sets limits to [its] public acceptance. The pace of the diffusion of new
technologies is constrained by the pace [at which we accept] and, if
so, accommodate them. Here the social sciences and the humanities are
28 Roco and Bainbridge, 2002, op. cit., p. 3.
29 Nordmann, op. cit., p. 4.
30 Roco and Bainbridge, 2002, op. cit., p. 5 (emphasis added).
31 Ramsey, P. 1977, Fabricated Man: The Ethics of Genetic Control, Yale University Press, New Haven, Conn.,
pp. 1223.
32 Nordmann, op. cit., p. 99.
22
needed to inform and accompany [technology convergence] research

and to serve as intermediaries. They should create settings within which
science and technology researchers on the one hand and the various
publics on the other, can learn from each other.33
Such a position does not invalidate the many genuine and methodologically
scrupulous efforts that have been made to engage both working scientists and the
general public in dialogue about the risks and uncertainties of nanotechnology,34
whether or not one agrees that the metaphor moving the debate upstream
adequately captures the degree to which debate is constrained by political,
research, industrial and military developments already well advancedand the
degree to which advocates of nanotechnology foresee the compass and role of
ethical deliberation. As depicted in the NSF/DOC report, the nanotechnology
effort should have many stakeholders in education, healthcare, pharmaceuticals,
social science, the military, the economy and the business sector, to name a few.
No less than a comprehensive national effort will be required. In addition, M. C.
Roco, widely regarded as the founding father of the US National Nanotechnology
Initiative, has recounted the efforts made in the 1990s to ensure widespread,
interlocking social and institutional momentum behind the establishment and
acceptance of nanotechnology:
The US National Nanotechnology Initiative was conceived as an inclusive
process where various stakeholders would be involved. In 1999 we
envisioned a grand coalition of academic, industry, governments, states,
local organizations, and the public that would advance nanotechnology
Creating a chorus of approval for nanotechnology, from 1990 to
March 1999 was an important preliminary step.35
Two considerations arise from the history of how nanotechnology has been
introduced, institutionalised and consolidated. First is the difficulty any
individual is likely to face in apprehending, acknowledging and voicing
ethical concerns over dual use that run counter to a prevailing ethos of tightly
interlocked institutions and interests. One need not subscribe to Daniel S.
Greenbergs judgment that [a]n infinity of researchable topics renders science
insatiable for money and increasingly indiscriminate in ways to get it,36
which depicts a corroded ethos, but leaves open the possibility for the ethical
33 Ibid., p. 18.
34 For example, see Rogers-Hayden, T. 2007, Moving engagement upstream? Nanotechnologies and the
Royal Society and Royal Academy of Engineerings inquiry, Public Understanding of Science, vol. 16, no. 3, pp.
34564; Kearnes, M., Macnaghten P. and Wilsdon, J. 2006, Governing at the nanoscale: people, policies and
emerging technologies, Demos, London; and Nanologue.net, 2006, <http://www.nanologue.net/>.
35 Roco, M. C. 2006, National Nanotechnology InitiativePast, Present and Future, <http://www.nano.gov/
NNI_Past_Present_Future.pdf> (italics in original).
36 Greenberg, D. S. 2001, Science, Money and Politics: Political Triumph and Ethical Erosion, University of
Chicago Press, Chicago, p. 463.
23
deliberation and moral courage of individuals. As psychologist Philip Zimbardo

has observed, however, [m]ost of us have a tendency both to overestimate the
importance of dispositional qualities and to underestimate the importance of
situational qualities when trying to understand the causes of other peoples
behavior.37 Zimbardo distinguishes between dispositional, situational and
systemic causes of behaviour; and although his interest is in trying to account
for cruel, inhumane and violent behaviours, his professional experiences,
ranging from the Stanford Prison Experiment to Abu Ghraib trial testimony,
illustrate how difficult it is for individuals to maintain and enact an oppositional
ethical position against a powerful, prevailing expectation. If, as quoted above,
the proponents of nano-led technological convergence foresee that widespread
consensus about ethical, legal and moral issues will be built into the process,
the practical prospects for dual-use bioethics will rest on levels of moral courage
unsupported by the records of social psychology or history.
Second, it might reasonably be asked whether, if ethicists and bioethicists
are stakeholders in a nano-led technological convergence initiative, they can
also bring ethical scrutiny to the enterprise itself. There is a strong case for
nonconformity with the functional roles that some advocates of nanotechnology
foresee for the social sciences and humanities:
We object to the narrow apprehension that the function of the humanities
and the social sciences consists only of achieving public trust concerning
nanotechnology. We believe that philosophy and ethics have a critical
function regarding the implementation of new technologies, which for
instance encompasses asking fundamental questions such as, What
impact will this new technology have on humanity? What is a good life?
Will this new technology affect the realization of a good life? What kind
of society do we want? and How does this new technology relate to that
kind of society? (And the list could easily be extended.)38
This argument has purchase as a matter of professional orientation, but still
greater meaning within contexts that contain such as the following, which
concerns the prospect of a nanotechnology-enabled battlefield in which soldiers
could be accompanied by machines capable of making their own decisions
(again, from the NSF/DOC report):
[C]ognitive scientists can do research on how a cyborg system makes
decisions about what constitutes a legitimate target under varying
conditions, including [the] amount of information, how the information
is presented, processing time, and the quality of the connection to
37 Zimbardo, P. 2007, The Lucifer Effect: How Good People Turn Evil, Rider, London, p. 8.
38 Ebbsen, M., Andersen, S. and Besenbacher, F. 2006, Ethics in nanotechnology: starting from scratch?
Bulletin of Science, Technology and Society, vol. 26, no. 6, p. 453.
24
higher levels of command. Practical ethicists can then work with

cognitive scientists to determine where moral decisions, such as when
to kill, should reside in this chain of command Practical ethicists and
social scientists need to act as stand-ins for other global stakeholders in
debates over the future of military nanotechnology.39
None of the above diminishes the importance of awareness and ethical
engagement of dual-use potentials by practising life scientists and by applied
ethicists, but it does contextualise it. Where, one might ask, are the ethicists and
their awareness of the dual-use potential of nanotechnology? We surely know
enough in outline about the practical potential and hoped-for applications of
nanotechnology (and the considerable efforts being expended to secure them) to
recognise them as morally weighty and ethically challenging matters. In respect
of dual-use bioethics as it applies to nanotechnology, moral engagement with
the project might best be thought of as pre-emptive rather than speculative.
Contrary to this, Alfred Nordmann has argued:
Ethical reflection of science and technology typically reacts to issues
that present themselves in the form of classical dilemmas, actual and
current predicaments, or hypothetical cases. In the case of reproductive
technologies, for example, ethical discussion has proven its relevance by
being very close on the heels always of novel techniques. In contrast,
nanotechnologies develop a tool-box for technological development.
As such they prepare the ground for a technical convergence at the
nanoscale. By enabling such a convergence, nanotechnologies create a
methodological challenge in that ethical engagement with presenting
issues becomes displaced by a perceived need to proactively engage
emerging issues. Lay and professional ethicists are only beginning to
meet this challenge.40
To characterise nanotechnological developments as a toolbox, however, and
to assert that the ethical challenges presented at this stage are methodological
are to abstract them from their social, political and economic contexts,
which shape their purposes and drive their momentum. In short, this line of
argument insulates nanotechnological developments from ethical scrutiny. But
the political sanction and investments that have directed the nanotechnology
enterprise have brought about a proliferation of new dual-use issues and have
extended or hybridised familiar ones. The weight and competitive orientation
of the interests involved have also established it on foundations shot through
39 Roco and Bainbridge, 2002, op. cit., p. 370.

40 Nordmann, A. 2007, If and then: a critique of speculative nanoethics, Nanoethics, vol. 1, no. 1, p. 34.
25
with moral and practical risks both profound and pervasive. These are hardly
propitious operating conditions for the scope and efficacy of either nano-ethics
or dual-use bioethics as applied to nanotechnology.
Dual use and dual-use dilemmas in

nanotechnology
As we have seen, even at this early stage of development, the thoroughgoing
international commitment to nanotechnology research and practical applications
has brought about a condition in which new scientific breakthroughs, novel
materials and processes and technological adaptations of nanotechnology are
coming on stream at speeds and with complex interrelations that challenge
professional comprehension at the community level. These have already begun
to outpace our wider deliberative systemsincluding our means of ethical
scrutiny. To this we can add the studied reluctance of nanotechnologys many
advocates to engage ethical deliberation outside a framework in which ethical
considerations are but one of a set of issues in a scientific and technological
progression that must not lose its inertia. This has been the case from the start,
as the following two examples illustrate. The first, a report by the US National
Research Council in 2002, explained why there was then a paucity of social
science research on the societal implications of nanotechnology: There appear
to be a number of reasons for the lack of activity in social science research on
societal implications.
First and foremost, while a portion of NNI [National Nanotechnology
Initiative] support was allocated to various traditional disciplinary
directorates, no funding was allocated directly to the Directorate of
Social and Behavioral and Economic Sciences, the most capable and
logical directorate to lead these efforts. As a consequence, social science
work on societal implications could be funded in one of two ways:
(1) it could compete directly for funding with physical science and
engineering projects through a solicitation that was primarily targeted
at that audience or (2) it could be integrated within a nanoscience and
engineering center.41
More recently, the US National Nanotechnology Initiatives funding opportunities
for societal implications in fiscal year 2010 are typical (and, it should be added,
only slightly above 1 per cent of the total disbursement):
41 National Research Council (NRC) 2002, Small Wonders, Endless Frontiers: A Review of the National
Nanotechnology Initiative, Committee for the Review of the National Nanotechnology Initiative, Division on
Engineering and Physical Sciences, National Research Council, National Academies Press, Washington, DC, p. 34.
26
Research directed at identifying and quantifying the broad implications

of nanotechnology for society, including social, economic, workforce,
educational, ethical, and legal implications ($5.78 million). The
application of nanoscale technologies will stimulate far-reaching changes
in the design, production, and use of many goods and services. Factors
that stimulate scientific discovery at the nanoscale will be investigated,
effective approaches to ensure the safe and responsible development
of nanotechnology will be explored and developed, and the potential
for converging technologies to improve human performance will be
addressed. The Nanotechnology in Society Network will extend its
national and international network.42
A second difficulty is that the range of uses of nanotechnology is so extensive
and the promise so considerable that dual use will not accurately or fully
capture the likely relationship between promise and peril. This is particularly
compelling in the case of nano-facilitated medical advances. For example, in
an experiment that shows great promise for treating brain cancer in humans,
University of Washington researchers have been able to cross the bloodbrain
barrier in mice. Until now, no nanoparticle used for imaging has been able
to cross the bloodbrain barrier and specifically bind to brain-tumor cells.
With current techniques doctors inject dyes into the body and use drugs to
temporarily open the bloodbrain barrier, risking infection of the brain.43
But the possibilities of nano-facilitated drug-delivery systems44 extend greatly
beyond the precise imaging of brain tumours, particularly with respect to
malign applications of neurobiology.45
Third, the very adaptability of nanotechnology makes a determination of dual
use at early stages of development very difficult, or introduces the possibility
at a threshold that is below that required for a genuine dilemma. For example,
in the following, it is very unlikely that a scientist engaged in the development
of nanowires would regard the full span of their possible applications as posing
an ethical dilemma:
University of Arkansas researchers have created assemblies of nanowires
that show potential in applications such as armor, flame-retardant fabric,
bacteria filters, oil cracking, controlled drug release, decomposition of
42 National Nanotechnology Initiative (NNI), Funding Opportunities, at NSF in FY 2010, <http://www.nsf.
gov/crssprgm/nano/core10_nsf_wta_nni.doc>.
43 Nanoparticles cross bloodbrain barrier to enable brain tumor painting, University of Washington
News, 3 August 2009, <http://uwnews.org/article.asp?articleid=51245>
44 Suril, S. S., Fenniri, H. and Singh, B. 2007, Nanotechnology-based drug delivery systems, Journal of
Occupational Medicine and Toxicology, vol. 2, no. 16, <http://www.occup-med.com/content/pdf/1745-66732-16.pdf>.
45 Wheelis, M. and Dando, M. 2005, Neurobiology: a case study of the imminent militarization of biology,
International Review of the Red Cross, vol. 87, no. 859, pp. 116.
27
pollutants and chemical warfare agents. This two-dimensional paper

can be shaped into three-dimensional devices. It can be folded, bent and
cut, or used as a filter, yet it is chemically inert, remains robust and can
be heated up to 700 degrees Celsius.46
In all probability, nanotechnology dual usethat is, malign possibilities and
pernicious applications as readily achievable as beneficent oneswill continue
to feature in research and development programs ranging across the physical
and medical sciences, but will rarely appear as dual-use dilemmas. The way
in which the nanotechnology and technological convergence enterprises have
been structured indicates faith at the policymaking level that the worrying
developments or applications of nanotechnology can be dealt with on a case-bycase basis, which essentially relegates dual-use conundrums to considerations
of risk. There is nothing in the way that nationally supported nanotechnology
programs have been initiated and sustained that acknowledges ethical quandaries
of a structural kind; and similarly, nor is it the case, as the Fink Report of 2004
asserts, that [b]iotechnology represents a dual use dilemma.47 Genuine
dilemmas are incapacitatingand there is little to suggest that our individual
and collective ethical unease over the biotechnology and nanobiotechnology
prospects will inhibit these enterprises, despite the number and range of dualuse possibilities they will continue to generate.
In order for dual-use potentials to be recognised and publicised as matters of
social concern and public policy, dual-use bioethics in nanotechnology will
need to be sited as far as possible upstream, towards current political initiatives
and research and development commitments. But the malign possibilities are
too wide-ranging in kind and too numerous to be dealt with as and when a
genuine dual-use dilemma appears. Indeed, bioethics and nano-ethics will find
little argumentative purchase in highlighting dual use if they concentrate on
the apprehension of dilemmas. After all, when set against the already extant
dual-use potential of nanotechnology, the number of expressed dilemmas is
exceedingly rare.48 What are required are ethical studies that regard the nanoenterprise from outside or above, which will provide a context for, and
inform, applied ethical studies, such as bioethics and nano-ethics can provide.
46 US National Nanotechnology Initiative, 2006, <http://www.nano.gov/html/research/home_research.html>.

47 Committee on Research Standards and Practices to Prevent the Destructive Application of Biotechnology,
National Research Council of the National Academies 2004, Biotechnology Research in An Age of Terrorism
[Fink Report], The National Academies Press, Washington, DC, p. 15.
48 Whitman, 2010, op. cit.
28
3. What Does Neuroethics Have to

Say about the Problem of Dual Use?
Valentina Bartolucci and Malcolm Dando
Introduction
It is clear that in the past advances in neuroscience were used for hostile as well
as peaceful purposes. Lethal chemical nerve agents, after all, interfered with
the acetylcholine neurotransmitter system1 and, during the twentieth centurys
EastWest Cold War, both sides clearly also made efforts to develop nonlethal chemical agents for various purposes.2 The use of some form of fentanyl
derivative(s) by Russian security forces to break the 2002 Moscow theatre
siege shows that today at least one major state has deployed such a weapons
system. Many commentators fear that Russia would be far from alone in having
an interest in developing novel incapacitating capabilities if the advances in
neuroscience provide suitable opportunities.3
This chapter sketches the areas of interest and methods used by neuroethicists to
ask what they have had to say about the problem of dual use: the fact that advances
in benignly intended civil neuroscience could produce materials, knowledge
and technologies that might then be used for hostile purposes by others. Of
course, it should be understood from the start that this is no small problem,
as has been made abundantly clear, for example, in the Lemon-Relman report
of the US National Academies in 2006, which, in its second recommendation,
stated that it was necessary to [a]dopt a broadened awareness of threats beyond
the classical select agents and other pathogenic organisms and toxins, so as to
include, for example, approaches for disrupting host homeostatic and defence
systems and for creating synthetic organisms.4
1 Dando, M. R. 1996, A New Form of Warfare: The Rise of Non-Lethal Weapons, Brasseys, London.
2 Dando, M. R. and Furmanski, M. 2006, Midspectrum incapacitant programs, in M. L. Wheelis, L. Rzsa
and M. R. Dando (eds), Deadly Cultures: Biological Weapons Since 1945, Harvard University Press, Cambridge,
Mass., pp. 23651.
3 Pearson, A. M., Chevrier, M. and Wheelis, M. L. 2007, Incapacitating Biochemical Weapons: Promise or
Peril? Lexington Books, Lanham, Md.
4 Committee on Advances in Technology and the Prevention of their Application to Next Generation
Biowarfare Threats 2006, Globalization, Biosecurity and the Future of the Life Sciences, The National Academies
Press, Washington, DC.
29
Host homeostatic and defence systems obviously would include hormones of

the endocrine system, neurotransmitters of the nervous system and cytokines
of the immune systemunderstanding all of which is critical to our continuing
advances in neuroscience.
The rise of neuroethics

As Robert Blank argued before the turn of the century, whilst there may be
some issues unique to the societal impacts of neuroscience, the issues raised
by advances in this area are basically similar to those raised in other areas of
medical advances.5 He suggested that there are, in fact, three clear levels of
policy dimensions in regard to all of these areas of technology:
1. [D]ecisions must be made concerning the research and development
of the technologies
2. The second policy dimension relates to the individual use of
technologies once they are available
3. The third dimension of biomedical policy centers on the aggregate
societal consequences of widespread application of a technology.6
These roots within the general growth of bioethical concerns are generally
acknowledged by neuroethicists. As Raymond de Vries noted in the 2007 special
issue of EMBO Reports on The Biology of Behaviour: Scientific and Ethical
Implications:7 Most histories of neuroethics are varieties of the technology
story. Illes and Bird (2006) place the history of neuroethics squarely in the
standard account of bioethics that runs from the Nuremberg Code in 1947, to
the 1964 Declaration of Helsinki to the Belmont Report in 1979.
The technology story, de Vries explains, is one in which new technologies bring
ethical questions that are too difficult for ordinary practitioners to answer and
they therefore need expert guidance from ethicists. Being a sociologist, de Vries
does not think this is the only possible history, but it is one that will be here
accepted for the moment.
Despite the acknowledgment of these long roots within bioethics, it appears to
be widely accepted that, as a particular field of study, neuroethics originated
5 Blank, R. H. 1999, Brain Policy: How the New Neuroscience Will Change Our Lives and Politics, Georgetown
University Press, Washington, DC.
6 Ibid., pp. 11 and 12.
7 de Vries, R. 2007, Who will guard the guardians of neuroscience? Firing the neuroethical imagination,
EMBO Reports, vol. 8 (S1), pp. S659.
30
3. What Does Neuroethics Have to Say about the Problem of Dual Use?
after the turn of the century. Martha Farah,8 in a thoughtful attempt to delineate
the field, argued, for example, that [b]eginning in 2002, neuroscientists began
to address these issues in the scientific literature and the field gained a name
neuroethics. At the same time, key meetings brought together large numbers
of experts and specific neuroethics membership organisations began to be
founded.9
The comparatively recent delineation of neuroethics as a special field it is not
too surprising; however, given its acknowledged deep roots within bioethics,
as Parens and Johnston carefully pointed out, it is important to understand that
bioethicists have made major errors in the past. If neuroethicists forget past
mistakes they could end up doing the same in the near future.10 Parens and
Johnston point to three particular problems that could easily arise in the new
field of neuroethics:
1. [T]he problem of reinventing the bioethical wheel
2. [T]he problem of exaggerating what the science can teach us about
who we are
3. [T]he problem of exaggerating what bioethics research can deliver.11
Furthermore, given the recent delineation of the specific field of neuroethics, we
should not be surprised to find that there is as yet no comprehensive, widely
accepted view of the scope of the field. For example, one might argue that finding
out what happens in the brain when ethical decisions are being made should be
central to any conception of neuroethics, but that does not appear to be what
practitioners have decided to do. Rather, the field seems to deal more with what
Farah called the practical and the philosophical: the implications of advances
in neuroscience for practical social issues and the implications of advances in
neuroscience for our understanding of ourselves (which, of course, overlaps to
some extent with the question of how we make ethical decisions). The sections
of Farahs paper provide illustrations of these different aspects (Table 3.1).
8 Farah, M. J. 2005, Neuroethics: the practical and the philosophical, Trends in Cognitive Sciences, vol. 9,
no. 1, pp. 3440, see Figure 3.1: Milestones in the history of ethics in neuroscience.
9 See, for example, The History of Neuroethics section of the entry on Neuroethics in Wikipedia.
10 Parens, E. and Johnston, J. 2007, Does it make sense to speak of neuroethics? EMBO Reports, vol. 8
(S1), pp. S614.
11From Wikipedia, op. cit., p. 64.
31
Table 3.1 Neuroethics: Practical and Philosophical
Practical: Brain imagining and brain privacy

Among the neuroscience technologies that present new ethical challenges of
a practical nature is functional brain imaging.
For example, in neuromarketing brain imaging is used to measure limbic
system response to a product that may indicate consumers desire for it.
Philosophical: Science and the soul
Recent neuroimaging research has shown a characteristic pattern of brain
activation associated with states of religious transcendence, which is common
to Buddhist meditation and Christian prayer.
The idea that there is somehow more to a person than their physical
instantiation runs deep in the human psyche and is a central element in
virtually all the worlds religions.
Neuroscience has begun to challenge this view, by showing that not only
perception and motor control, but also character, consciousness and sense of
spirituality may all be features of the machine.
Source: de Vries, R. 2007, Who will guard the guardians of neuroscience? Firing the neuroethical
imagination, EMBO Reports, vol. 8 (S1), pp. S659.
Neil Levy, in his introduction to the new journal Neuroethics, accepted this
twofold division of the subject but stressed that, to the extent that neuroscience
shows that we are less than rational and autonomous in our decision-making,
that must impact on our understanding of the practical impact of the advances
in neuroscience.12
In an attempt to advance the field of neuroethics, Georg Northoff argued
that both of Farahs two subdivisions of neuroethics (which might be simply
characterised as what we can do and what we know) should be termed
empirical neuroethics, and we should accept that there is no sharp distinction
between them (Table 3.2). So, in his view: Empirical neuroethics deals with
the empirical and practical aspects of the linkage between neuroscientific and
ethical concepts.13
12 Levy, N. 2008, Introducing neuroethics, Neuroethics, vol. 1, pp. 18.

13 Northoff, G. 2009, What is neuroethics: empirical and theoretical neuroethics, Current Opinion in
Psychiatry, vol. 22, pp. 5659.
32
Table 3.2 Aspects of Empirical Neuroethics
Technology
Brain imaging
Pharmacological
enhancement
BMI and N-P

enhancement
Questions
Old technology
Safety
Researchers
obligations
Validity
Safety
Validity
New questions
Intentional deception
Neuromarketing
Personal
characteristics
Attention
Memory
Mood
Equity
Military research on
cyborgs
Source: Northoff, G. 2009, What is neuroethics: empirical and theoretical neuroethics, Current Opinion in
Psychiatry, vol. 22, pp. 5659.
He writes, however, that [a]lthough there has been much discussion of various
issues in empirical neuroethics, the discussion of methodological and conceptual
issues and thus theoretical neuroethics has remained rather sparse so far.
Thus, Northoff argues for the need for theoretical neuroethics to focus on the
methodological and conceptual aspects of the linkage between neuroscientific
facts and ethical concepts and for a theoretical neuroscience that is able to give
proper weight to both norms and facts in a normfact circularity.
Towards the end of his paper, Northoff notes that his ideas may strike some
as a mere playground for theoreticiansa criticism that might be particularly
applied to stressing theoretical neuroscience in this paper on the severely
practical issue of dual use. We think, however, he has a general point to make on
neuroethics methodology that is significant: If neuroethics wants to establish
itself as a separate discipline that is different from its neighboring disciplines
like philosophy, ethics and neuroscience, it must develop a special methodology.
In this view, a discipline or field of study cannot claim to be distinct just
because it studies certain things; it must have developed its own distinctive
methodology as well. It would certainly be wrong to consider that this will be
a simple task given that there are still great philosophical differences about how
we should go about understanding our brains and behaviour.14
14 Evers, K. 2007, Towards a philosophy for neuroethics, EMBO Reports, vol. 8, pp. 84851.
33
Although the recent conceptualisation of neuroethics originated in the United

States and Europe, research and publications on neuroethics are increasingly
international. Lombera and Illes suggested that despite the broad scope of the
field, neuroethics has attempted to frame its efforts in terms of four pillars:
brain science and the self, brain science and social policy, ethics and the practice
of brain science, and brain science and public discourse.15
On this basis, they carried out a wide-ranging literature survey and concluded
that their results demonstrated a steady increase in global participation in
neuroethics from 1989 to 2005, characterized by an increase in numbers of
articles published specifically on neuroethics, journals publishing these articles,
and countries contributing to the literatureand they clearly expect this trend
to continue as neuroscience advances and the associated technologies spread
around the world.
Neuroscientists and neuroethicists have been commendably interested in the
problem of communication with the public about their work16 and about how
neuroscientists might be better enabled to carry out such tasks. Yet there are
also reasons to believe that neuroscientists do not readily take to ethical analyses
and the communication of their ethical positions,17 and that they are often not
trained to do so.18 This, of course, is cause for considerable concern because, as
the sociologist Raymond de Vries pointed out, there are other possible accounts
of the rise of bioethics besides the technology story. For example, de Vries cites
one account in which [u]nlike the conventional technology story, in which
bioethicists are cast as the guardians who oversee and regulate doctors and
scientists bioethicists are less-than-critical allies of medicine and medical
science.19
Raymond de Vries comments that neuroethicists would do well to understand
these different possibilities because if they fail to do so, they are less inclined
to appreciate the way in which funding sources, and the structure of industry
and academic research, shape bioethics and neuroethics. Against that brief
background context, the rest of this chapter is concerned with some examples
of what neuroethicists doparticularly in regard to the problem of dual use.
15 Lombera, S. and Illes, J. 2009, The international dimension of neuroethics, Developing World Bioethics,
vol. 9, no. 2, pp. 5764.
16 Illes, J. et al. 2005, International perspectives on engaging the public in neuroethics, National Review
of Neuroscience, vol. 6, no. 12, pp. 97782; Illes, J. et al. 2010, Neurotalk: improving the communication of
neuroscience, National Review of Neuroscience, vol. 11, no. 1, pp. 120.
17 Wolpe, P. R. 2006, Reasons scientists avoid thinking about ethics, Cell, vol. 125, no. 6, pp. 10235.
18 Sahakian, B. J. and Morein-Zemir, S. 2009, Neuroscientists need neuroethics teaching, Science, vol. 325,
p. 147.
19 Blank, op. cit.
34
Case studies
Whilst the intention here is to concentrate on the ethical issues involved in the
practical consequences of the advances in neuroscience, it should be made clear
that this is not because we underestimate the importance of the philosophical
issues raised by modern neuroscience. This point has been made regularly in
reviews of the emerging field: an inadequate model of the brain and mind is
likely to cause great misunderstandings and practical difficulties.20 For example,
if someone is in a persistent vegetative state but neuroimaging shows that he or
she is able to respond to some questions,21 can informed consent be achieved for
treatment? Clearly, the brain is not without response, but is there an ability to
decide on complex questions?
If such questions at the more philosophical end of the spectrum of the empirical
issues are set aside, it is clear that neuroethicists have considered questions
arising from three types of new technologies: neuroimaging; pharmacological
enhancement; and non-pharmacological enhancement (such as brainmachine
interfaces, transcranial magnetic stimulation, and direct current stimulation).
There is also now some consideration being given to how the combination of
these technologies with genomics22 and information technology23 may affect the
ethical questions.
Some of the ethical questions that arise with these new technologies are far from
new. In this regard neuroethics has carefully noted the issues relating to safety
of the technologies, the validity of results drawn from complex analyses derived
from neuro-imaging techniques, and how these images are understood by the
media and general public. Neuroethicists have also discussed the obligations
that researchers have for their findingsfor example, what to do about findings
that suggest that the person involved could have an increased likelihood of
illness in the future.
Despite acknowledging these issues that have long-running parallels in bioethics,
neuroethicists writing makes frequent reference to the fact that some of the
issues that arise from the application of these new technologies are novel. As
20 Glannon, W. 2009, Our brains are not us, Bioethics, vol. 23, no. 6, pp. 3219.
21 Owen, A. M. et al. 2006, Detecting awareness in the vegetative state, Science, vol. 313, p. 1402.
22 Tairyan, K. and Illes, J. 2009, Imaging genetics and the power of combined technologies: a perspective
from neuroethics, Neuroscience, vol. 164, pp. 715.
23 Amari, S.-I. 2002, Neuroinformatics: the integration of shared databases and tools towards integrative
neuroscience, Journal of Integrative Neuroscience, vol. 1, no. 2, pp. 11728.
35
Farah and Wolpe express it: The brain is the organ of mind and consciousness,
the locus of our sense of selfhood. Interventions in the brain therefore have
different ethical implications than interventions in other organs.24
With regard to neuroimaging, there are numerous discussions of the possibility,
and thus implications, of being able to detect when people are intentionally
carrying out a deception. There has been wide discussion of the implications of
being able to detect, through neuroimaging, peoples desire for certain products
and the consequences of the growth of neuromarketing. Concerns have also
been expressed about the dangers to privacy if such personal characteristics can
be elucidated by neuroimaging.
With respect to pharmacological enhancement, there are again issues of safety,
validity and communication of findings to a non-expert audience that are not
specific to the concerns of neuroscientists and neuroethicists. Again, however,
the neuroethicists writing clearly indicates that they believe the possibilities of
enhancing attention, memory and mood and the reverse possibility of damping
down memories (to help people suffering from post-traumatic stress disorder)
do raise new ethical issues. What happens, for example, to those who choose
not to be enhanced, or who cannot afford to be enhanced, in an enhancementridden society? And, again touching on the philosophical, what happens to our
sense of self and worth if we can have a better attention span, memory or mood
not by work to achieve such developments but by popping a pill?
A similar set of novel and not so novel issues arises in regard to non-pharmaceutical
enhancement, but one unusual point can be noted in the concerns expressed
about military funding of work on brainmind interfaces and of new breeds of
cyborgs. Nevertheless, it is clear that this worry about the military implications
of the advances in neuroscience and related technologies is very limited in the
neuroethics literature. The limited nature of the neuroethical debate was noted by
Joelle Abi-Rached in her discussion of the launch of the European Neuroscience
and Society Network. In her view, this has to change and the debate must
also include the controversies surrounding the potential application of various
technologies in neurosecurity and counter-terrorism.25
She correctly references Jonathan Morenos book Mind Wars: Brain Research and
National Defense26 when making her point. Even there, howeverin the only
extended discussion of the possible misapplications of various technologies
there is very little discussion of the problem of dual use. Indeed, although in
24 Farah, M. J. and Wolpe, P. R. 2004, Monitoring and manipulating the human brain: new neuroscience
technologies and their ethical implications, Hastings Center Reports, vol. 34, no. 3, pp. 3545.
25 Abi-Rached, J. M. 2008, The implications of the new brain sciences, EMBO Reports, vol. 9, no. 12, pp.
115862.
26 Moreno, J. D. 2006, Mind Wars: Brain Research and National Defense, Dana Press, Washington, DC.
36
a 2007 editorial in Science, Henry Greelys reflections on neuroethics included

the view that [o]ther studies may have military implications: Suppose brain
stimulation created an indefinitely awake and alert soldier or pilot? Will
neuroscience be a new source of dual-use technologies such as those we worry about
for biological or chemical warfare? (emphasis added).27
It is very difficult to find even a reference to, let alone a discussion of, the
problem of dual use in the neuroethics literature, even when it deals with
national security issues.28
For those trained in natural science, a further striking feature of the neuroethics
literature is the generality of the discussion around the ethical implications of
neuroscience.29 There is very good reason to believe that the ongoing advances
in civil neuroscience could produce materials, technologies and knowledge that
could later be applied, for example, to develop new forms of incapacitating
chemicals for use in the War on Terror. As well, they could lead to the
erosion and eventual demise of the prohibition on the use of the modern life
sciences for non-peaceful purposes that is embodied in the Biological and Toxin
Weapons Convention and the Chemical Weapons Convention.30 So what are
civil neuroscientists to do to help protect their work from such misuse? Does
it perhaps conform to the paradigm described by Lawrence Schmidt and Scott
Marratto:
The technological imperative has opened every aspect of human life to
relentless transformation. The assumption is that it is always ethically
acceptable to experiment to find out whether we can do something, and
if we can, we ought to. But the adoption of the technological imperative
has meant the liberation of means from ends.31
That, of course, is a huge question, but one way to approach it, perhaps, is to
look more closely at one of the technologies that neuroethicists have focused
upon: cognitive enhancement.
27 Greely, H. 2007, Editorial: neuroethics, Science, vol. 318, p. 533.

28 Conli, T. et al. 2007, Neuroethics and national security, The American Journal of Bioethics, vol. 7, no. 5,
pp. 313; Resnik, D. B. 2007, Neuroethics, national security and secrecy, The American Journal of Bioethics,
vol. 7, no. 5, pp. 1426.
29 See, for example, Illes, J. 2007, Empirical neuroethics, EMBO Reports, vol. 8, pp. S5760, Figures 2 and 3;
and Glannon, op. cit., Figure 2.
30 See <www.opbw.org> and <www.opcw.org>.
31 Schmidt, L. E. and Maratlo, S. 2008, The End of Ethics in a Technological Society, McGill-Queens University
Press, Montreal and Kingston.
37
Cognitive enhancement
In December 2008, Nature carried a commentary by Henry Greely and five
colleagues titled Towards responsible use of cognitive-enhancing drugs by the
healthy. The stated aim of this piece was to propose actions that will help
society accept the benefits of enhancement, given appropriate research and
evolved regulation.32
Arguing that these drugs are just another means devised by our innovative
species to improve itself, they suggested cognitive-enhancing drugs should
be viewed in the same general category as education, good health habits and
information technology. They listed what they viewed as standard arguments
against the use of these drugscheating, unnaturalness and drug abuseand
dismissed them. They did accept, however, that questions of safety, freedom
(from coercion to enhance) and fairness (towards those who could not afford
enhancement) would need to be addressed. They also suggested a program
of research, professional guidance, public understanding and regulation
development be undertaken so that responsible use was facilitated. In a direct
critique of this commentary, two Canadian neuroethicists struck a much more
cautious note, arguing that the question of safety was far from settled: it is
important to stress that sizeable gaps exist in our current understanding of
the effects, both positive and negative, of neuropharmaceuticals on healthy
individuals.33
And they pointed out that the simple tasks studied so far in laboratories do not
reflect the complexity and diversity of activities in learning and thinking. They
thus implied that there was a degree of overstatement amongst those favouring
cognitive enhancement in the same way that gene therapy had been oversold by
supporters. A more neutral term than cognitive enhancement, they suggested,
might be non-medical use of prescription drugs. These critics also had concerns
about the impact on health resources of interest in cognitive enhancementfor
example, about the safety research program suggested by advocates eating up
resources that were more urgently needed elsewhere, for example, in treating
people who are ill. In short, there does not yet appear to be a settled view
amongst neuroethicists about cognitive enhancement.34
Yet there is a considerable literature on this subject and certainly enough to ask
two questions relevant to dual use: 1) do neuroethicists dealing with cognitive
32 Greely, H. et al. 2008, Towards responsible use of cognitive-enhancing drugs by the healthy, Nature,
vol. 456, pp. 7025.
33 Racine, E. and Forlini, C. 2009, Expectations regarding cognitive enhancement create substantial
challenges, Journal of Medical Ethics, vol. 35, pp. 46970.
34 Harris, J. and Chatterjee, A. 2009, Head to head: is it acceptable for people to take methylphenidate to
enhance performance? BMJ, vol. 338, pp. 15323.
38
enhancement consider the issue of dual use, and 2) do the investigations of the
mechanism of enhancement by neuroscientists indicate any awareness of the
possibility that their work could be used for the very opposite manner by those
with malign intent?
Most discussions of cognitive enhancement are concerned with the use of drugs
such as: methylphenidate (used medically to help people suffering from attention
deficit hyperactivity disorder) to improve attention; modafinil (used medically
to help people suffering from sleep problems to improve alertness; and SSRIs
selective serotonin reuptake inhibitors (used medically to help people suffering
from problems of mood)to improve the mood of people who are unwell. A
particularly interesting subject in regard to this chapter is the use of the drug
propranolol, not to improve memory but to help people to forget emotionally
laden traumatic eventsmemories that recur in post-traumatic stress disorder
(PTSD).35
In such discussions it is certainly possible to find what appears to be approval
of military-funded research that could fundamentally change war-fighting and
force employment because of, for example, the elimination of the need for sleep
and the maintenance of a high level of cognitive performance.36 On the other
hand, it is rare indeed to find a clear recognition in the neuroethics literature of
the problem of dual use in regard to cognitive enhancement. As Kathinka Evers
pointed out:
It has been suggested that therapeutic forgetting is interesting for
military purposes, for example, to provide soldiers with propranolol
before a battle. A problem here is that if it helps them forget what they
have been subjected to it also helps them forget what they have done to
others.37
This comment is clearly a special case because Evers is one of the few bioethicists
who has addressed the general issue of dual-use bioethics.38
If we turn to the neuroscience, it is clear that safety issues involved in cognitive
enhancement are sometimes well understood. As Cakic recently pointed out:
For psychostimulants such as methylphenidate, the dangers are real and
35 Glannon, W. 2008, Psychopharmaceutical enhancement, Neuroethics, vol. 1, pp. 4554.

36 Sahakian, B. J. and Morein-Zamir, S. 2010, Neuroethical issues in cognitive enhancement, Journal of
Psychopharmacology [epub ahead of print, doi: 10.1177].
37 Evers, K. 2007, Perspectives on memory manipulation: using beta-blockers to cure post-traumatic stress
disorder, Cambridge Quarterly of Healthcare Ethics, vol. 16, pp. 13846.
38 Research Ethics and Bioethics, University of Uppsala, <www.crb.uu.se>.
39
relatively well known. Aside from its abuse potential, methylphenidate may
aggravate mental illness, produce sleep disturbances and is associated with
cerebrovascular complications.39
It is also clear that we know a great deal about how emotion-laden memories
are laid down in mammals and the way in which propranolol can be used to
interfere with the role of noradrenaline in the consolidation and reconsolidation
of such memories.40
Whilst accepting that some modest improvements in capabilities may be achieved,
recent detailed reviews of the science and ethics of cognitive enhancement have
emphasised caveats: first doses most effective in facilitating one behavior
could at the same time exert null or even detrimental effects on other cognitive
domains. Second, individuals with low memory span might benefit from
cognitive-enhancing drugs, whereas high span subjects are overdosed.41
And, finally: evidence suggests that a number of trade-offs could occur. For
example, increases of cognitive stability might come at the cost of a decreased
capacity to flexibly alter behaviour.
This particular review also discusses six ethical issues found in the literature:
safety; societal pressure; fairness and equality; enhancement versus therapy;
authenticity and personal identity; and happiness and human flourishing. In
the last of these it does refer to the possibility that the blunting of memory
could involve violation of a duty to remember and bear witness of crimes and
atrocities. Thus it touches on the question of dual use, but again there is no
indication that the general pointthat all of the work in cognitive enhancement
could be dual usehas been understood.
Another recent review that notes the problem of such misuse in the blunting of
memories without drawing the general conclusion about dual use does make the
crucial point that our knowledge of the brain remains limited: the fundamental
question is: are we technically ready and do we have sufficient basic knowledge
to develop such drugs without risking a deadly brain doping?42
A possible response is to say that cognitive enhancement has been much
overhyped and that major brain modifications are years away and therefore
39 Cakic, V. 2009, Smart drugs for cognitive enhancement: ethical and pragmatic considerations in the era
of cosmetic neurology, Journal of Medical Ethics, vol. 35, pp. 61115.
40 Tully, K. and Bolshakov, V. Y. 2010, Emotional enhancement of memory: how norepinephrine enables
synaptic plasticity, Molecular Brain, vol. 3, pp. 1524; Dando, M. R. 2007, Scientific outlook for the
development of incapacitants, in Pearson et al., op. cit., pp. 12348.
41 de Jongh, R. et al. 2008, Botox for the brain: enhancement of cognition, mood and pro-social behavior
and blunting of unwanted memories, Neuroscience and Biobehavioral Reviews, vol. 32, pp. 76076.
42 Lanni, C. et al. 2008, Cognitive enhancers between treating and doping the mind, Pharmacological
Research, vol. 57, pp. 196213.
40
we should not worry too much about the state of neuroethics and its lack of
coverage of the problem of dual use. That is indeed what one neuroscientist
stated recently in a review of a book of essays on neuroethics: the participants
in this discussion often claim that their speculative approach provides us with
the unique opportunity to discuss the ethical consequences of new technologies
before they are fully developed However, do we really need a debate on a
technology that will probably never materialise?43
In this view, we do not really need to be concerned about neuroethics because
the debate amongst neuroethicists is not of great importance. Many people took
amphetamines during and after World War II so the present uses of drugs for
cognitive enhancement is nothing new and, given the limited benefits that are
likely to be available for the foreseeable future, the present phase will surely
pass quietly away.
Back to dual use

The problem with the benign viewpoint voiced above (which does not take
into account malign misappropriation of neuroscience advancements) is that
whilst it may be appropriate for such civil uses of cognitive enhancement, it
surely cannot be said to apply to the problem of military dual use. Here, in
addition to a long history of misuse of the ongoing advances in neuroscience,
we clearly have evidence of state-level development and use of novel chemical
incapacitants. Furthermore, there is an obvious concern that continued interest
in such developments could lead to the erosion and, potentially, the destruction
of the prohibition of chemical and biological weapons embodied in the Chemical
Weapons Convention and the Biological and Toxin Weapons Convention.
It is not as if it is difficult to find evidence of real concern amongst experts about
this issue. For example, when the International Committee of the Red Cross
launched its appeal on Biotechnology, Weapons and Humanity, the eminent
neuroscientist Professor Tamas Bartfai worried about the dangers of the misuse
of advances in neuroscience. He noted that the acetylcholine system had been
the target of successive generations of lethal nerve gases and for drugs designed
to help people suffering from Alzheimers disease, and went on to point out a
number of clear-cut ways in which current developments in the neurosciences
might be subject to misuse.44 Similarly, Robert McCreight, a senior US civil
servant, gave a lecture on the implications of the advances in neuroscience for
43 Quednow, B. B. 2010, Ethics of neuroenhancement: a phantom debate, [Books Forum], BioSocieties, vol.
5, no. 1, pp. 1536.
44 Bartfai, T. and Sellstem, A., Neurobiologyweapons and humanity, Presentation at the launch of the
appeal on Biotechnology, Weapons and Humanity, International Committee of the Red Cross, Montreux.
41
national security and future strategic weapons in January 2007. As he noted:

Scientific research, concept development, examination of bioethical issues
related to enhanced mental health and considerable long term funding support
has been ongoing for several years in the broad area of neuroscience.45
Then he asked: Have we adequately analysed and discussed the dual-use
implications of neuroscience, particularly its various military applications, and
the extent to which operational safeguards and societal controls are needed
to manage or control its most destructive weapons outcomes or debilitating
systems?
McCreight did not think that there had been adequate analysis and discussion
and suggested that unless measures were taken we may face new categories of
weapons before 2010 held by several nations both friendly and hostile.
Certainly, some practising neuroscientists have raised concerns about the
hostile misuse of advances in neuroscience,46 and there are a few bioethicists
who have tried to contribute to a better understanding of the problem of dual
use.47 Given this relatively early stage in the development of neuroethics, it is
not clear whether neuroethicists will have anything special to add to the work
of bioethicists in general, but if they do aspire to help as architects of moral
space by fostering open and constructive dialogue, discussion and debate
about critical issues,48 in order to develop smarter regulation where that is
necessary49 then surely there is no better time than now to begin to develop
their view of the neuroethics of dual use.
Conclusion
The field of neuroscience has grown considerably in the past decade. Advances
in neuroscience already have raised important questions on a wide range of
policy issues, such as those affecting neurotoxins and the environment, mental
health, child development, cognitive enhancement, criminal behaviour, the
safety and efficacy of pharmaceuticals and medical devices, and the ethics and
regulation of emerging discoveries. To an ever-increasing understanding of the
45 McCreight, R. 2007, Protecting our national neuroscience infrastructure: implications for homeland
security, Presentation to National Security and the Future of Strategic Weapons, George Washington
University, Institute of Crisis, Disaster and Risk Management, <http://www.chds.us/?fs:file&mode=dl&drm=
..%2F..%2Fresources%2Fsummit%2F%2Fsummit07&f=McCreight-GeorgeWashUniv.ppt&altf=McCreightGeorgeWashUniv.ppt>.
46 Bell, C. 2010, Neurons for peace: take the pledge, brain scientists, New Scientist, vol. 2746 (8 February).
47 Atlas, R. M. 2009, Responsible conduct by life scientists in an age of terrorism, Science and Engineering
Ethics, vol. 15, pp. 293301.
48 Robert, J. S. 2009, Toward a better bioethics, Science and Engineering Ethics, vol. 15, pp. 28391.
49 Sutton, V. 2009, Smarter regulations, Science and Engineering Ethics, vol. 15, pp. 3039.
42
brain mechanisms associated with core human attributes and values should
also correspond an increasing interest in the possible dual-use implications of
such advancements, giving the various ways, not always benign, in which the
new knowledge could be used. Neuroethics addresses the various philosophical
issues around the relationship between brain and mind as well as practical issues
about the impact upon society of our ability to understand and manipulate
the brain. The problem of dual use, consisting in a malign appropriation of
knowledge initially designed for benign purposes, should be an important
focus of neuroethicists analyses. In particular, the long-term applications and
impacts of neuroscience are likely to be powerful and profound. As pointed out
by Marchant and Gulley: Military and intelligence agencies, with the most at
stake from such applications in terms of both benefits and risks, recognize the
potential of neuroscience to revolutionize intelligence gathering and warfare.50
Furthermore, as pointed out by Jonathan Moreno, the 11 September 2001
attacks have resulted in increased efforts to exploit all technical possibilities
for enhancing security. The Pentagons Defense Advanced Research Projects
Agency (DARPA) is supporting work at Lockheed Martin on remote brain prints
and the scientist in charge already claims to be able to tell if a person is thinking
of a certain number. In the words of Moreno, a striking aspect of much of this
and other national security work being done in the field of neuroscience is that
it is dual usepotentially applicable to medical therapy or other peaceful
purposes as well as combat, riot control, hostage situations, or other security
problems.51
Worryingly, from a preliminary review of the literature on neuroethics, it clearly
emerges that, while publications abound on issues such as lie detection, informed
consent for certain patients and around the implications of neuroimaging, the
problem of dual use is very marginally addressed.
In the majority of research centres and institutions, no mention is made of the
problem of dual use, and the overwhelming assumption is that neuroscience
works for the betterment of humanity worldwide (see Appendix A). If this is
what neuroethicists are working for, more attention should surely be devoted
to the way advances in neuroscience could be misused. Leading neuroethicists
also very rarely address the issue of dual use in neuroscience, with few notable
exceptions.52
50 Marchant, G. and Gulley, L. 2010, National security neuroscience and the reverse dual-use dilemma,
AJOB Neuroscience, vol. 1, no. 2, pp. 202.
51Ibid.
52 Moreno, J. 2005, Dual use and the moral taint problem, The American Journal of Bioethics, vol. 5, no.
2, pp. 523; Moreno, J. 2008, Using neuro-pharmacology to improve interrogation techniques, Bulletin of
the Atomic Scientists; Huang, J. Y. and Kosal, M. E. 2008, The security impact of the neuroscience, Bulletin
of the Atomic Scientists.
43
Nevertheless, if neuroethicists have paid almost no attention to the problem

of dual use, the issue should be of great concern to both neuroscientists and
neuroethicists, who should critically approach dual use in neuroscience in order
to save neuroscience from dreadful distortions of its intended purpose. In the
current environment, a similar remarkable omission, if not addressed urgently,
will be soon deplored.
44
4. Synthetic Biology as a Field of

Dual-Use Bioethical Concern
Alexander Kelle
Introduction
Over the past decade synthetic biology has emerged as one of the most dynamic
subfields of the post-genomic life sciences. According to a European high-level
expert group, synthetic biology comprises the synthesis of complex, biologically
based (or inspired) systems which display functions that do not exist in nature
[and] is a field with enormous scope and potential.1 Some of the areas where
this expert group argues that synthetic biology could have a major impact
include biomedicine, a sustainable chemical industry, environment and energy,
and biomaterials. If the emerging discipline of synthetic biology can deliver on
the promises of some of its leaders and become as pervasive as computing has
become in the past few decades, we might very well be witnessing a fundamental
shift similar to the one that happened to chemistry with the introduction of the
periodic table. If synthetic biologists live up to some of the more far-reaching
expectations, biology ultimately may become a mechanistic science.
On the one hand, synthetic biology developments show promise of leading
to beneficial applications in a number of areas, such as drug development,2
biodegradation3 and biofuels.4 At the same time, the dual-use character of this
new technoscience carries with it the possibility of synthesised biological parts,
modules and systems being malignly misused. This dual-use potential hasat
a rather abstract level and with a focus on one particular subfield of synthetic
biology, that is, DNA synthesisbeen recognised by practitioners in the field
as well as analysts. While this is a positive development, however, these mostly
technical governance measures that are addressing DNA synthesis capabilities
need to be broadened so as to cover all aspects of synthetic biology and to allow
for a comprehensive bioethical analysis of the fields dual-use implications. In
addition, part of the discourse on the broader societal implication of synthetic
biology can be traced back to the debates on ethical, legal and social implications
1 Synthetic Biology. Applying Engineering to Biology, Report of a New and Emerging Science and Technology
(NEST) High-Level Expert Group, European Commission, Brussels, 2005, p. 5.
2 See Neumann, H. and Neumann-Staubnitz, P. 2010, Synthetic biology approaches in drug discovery and
pharmaceutical biotechnology, Applied Microbiology and Biotechnology, vol. 87, pp. 7586.
3 See Kirby, J. R. 2010, Designer bacteria degrades toxin, Nature Chemical Biology, vol. 6, pp. 3989.
4 See Dellmonaco, C. et al. 2010, The path to next generation biofuels: successes and challenges in the era
of synthetic biology, Microbial Cell Factories, vol. 9.
45
(ELSI) of genetic engineering. Yet, given their limited nature, these debates do
not provide a solid foundation for a comprehensive discussion and assessment
of synthetic biologys dual-use potential.
This chapter will first outline the scope of synthetic biology as a new subfield in
the life sciences in which different science and engineering disciplines converge.
This will be followed by a brief discussion of synthetic biologys potential for
malign misuse as well as some of the proposals for governance of this new
technoscience. Thus far, the mostly technical character of these proposals
has resulted in a rather limited appreciation of the wider governance issues
related to the full breadth of approaches usually subsumed under the synthetic
biology label. The final section will discuss both academic and institutional
contributions to a bioethically informed discourse on the misuse potential of
synthetic biology.
Scope of synthetic biology as a subfield in the

life sciences
Attempts at defining synthetic biology
Not surprisingly for a discipline that is still in its formative stages, several
definitions exist for synthetic biology. One that has received the most attention
describes synthetic biology as the design and construction of new biological
parts, devices, and systems, and the re-design of existing, natural biological
systems for useful purposes.5 This definition clearly reflects the approach
to synthetic biology pioneered by scientists at the Massachusetts Institute
of Technology (MIT), and puts centre stage the idea to develop a registry of
standardised biological parts that can be assembled in devices and systems with
predefined functions. Although the MIT framing of the issue area has certainly
sparked the development of the whole field (for example, by organising the first
international Synthetic Biology Conference in Boston in 2004, by supporting
the Biobricks Foundation, and by organising the annual iGEM student
competitions), an exclusive focus on the parts-based approach to synthetic
biology tends to overlook other important subfields in synthetic biology (see
Table 4.1 for an overview). For ease of reference, four different substrands of
synthetic biology that have developed since the early years of the twenty-first
century are distinguished here6
5 See <http://syntheticbiology.org/Who_we_are.html> (viewed 6 November 2008).
6 This subdivision follows Schmidt, M. 2009, Do I understand what I can create? in M. Schmidt, A. Kelle,
A. Ganguli-Mitra and H. de Vriend (eds), Synthetic Biology. The Technoscience and Its Societal Consequences,
Springer, Dordrecht, pp. 81100.
46
4. Synthetic Biology as a Field of Dual-Use Bioethical Concern
engineering DNA-based biological circuits, by using standardised biological

parts
identifying the minimal genome
constructing protocellsin other words, living cells from base chemicals
creating orthogonal biological systems in the laboratory through chemical
synthetic biology.
Two enabling technologies also tend to be usually subsumed under the heading
of synthetic biology, although they have a more supportive role to the four
fields mentioned above. This applies first and foremost to the increasingly
more affordable large-scale DNA-synthesis capabilities that some companies are
providing,7 but also to the more generic bioinformatics capabilities utilised by
those attempting to identify a minimal genome.
Thus, synthetic biology at the very least aims at merging molecular biology
with engineering, by designing and producing new biological parts, devices
and systems. To achieve this goal, synthetic biology utilises high-throughput
commercial DNA-synthesis capabilities to provide the actual biological material
for the assembly of genetic circuits. In addition, synthetic biology is relying on
increasingly powerful information technology tools that allow for the modelling
of certain desired biological functions.
A report by the Royal Academy of Engineering in the United Kingdom has
detailed the design cycle that informs the engineering approach to synthetic
biology.8 Accordingly, the design cycle for biological systems starts with an
initial specification stage for bioparts, which is followed by a detailed design
step, drawing on the ability to undertake detailed computer modeling.9 In
the subsequent implementation stage, synthesised DNA is usually inserted into
E. coli, yeast or some other suitable chassis. In the final validation stage, the
functionality of the original specification is verified. Standardised bioparts that
have undergone such quality controls can then be utilised to build standard
devices, which, in turn, can be assembled into biological systems. As OMalley
has recently pointed out, however, research and development (R&D) in
synthetic biology have so far generally not been that straightforward. Instead,
strictly hypothesis-driven work is regularly complemented by [e]xploration,
iterativity and kludging.10 In other words, realisation of the engineeringdriven and standards-based plug and play ideal of many synthetic biologists
still awaits realisation and will for some time.
7 See, for example, the web pages of GENEART (<www.geneart.com>) or Blue Heron (<www.blueheronbio.
com>). Another prominent company in this area, Codon Devices, recently closed. See Hayden, E. C. and
Ledford, H. 2009, A synthetic biology reality check, Nature, vol. 458, p. 818.
8 The Royal Academy of Engineering 2009, Synthetic Biology: Scope, Applications and Implications, The
Royal Academy of Engineering, London, esp. pp. 1821.
9 Ibid., p. 19.
10 OMalley, M. 2011, Exploration, iterativity and kludging in synthetic biology, C. R. Chimie, vol. 14,
pp. 40612.
47
48
Finding the smallest possible

genome that can run a cell, to
be used as a chassis; reduced
complexity
Bioinformatics-based
engineering
Deletion of genes and/or

synthesis of entire genome and
transplanting the genome in a
cytoplasm
DNA-synthesis and
transplantation of Mycoplasma
genitalium
Designing genetic circuits, for

example, from standardised
biological parts, devices and
systems
Design and fabrication;

applying engineering principles
using standardised parts and
abstraction hierarchies
Design of genetic circuits on

the blackboard, inserting the
circuits in living cells
AND gate, OR gate;

genetic oscillator repressilator;
Artemisinin metabolism,
Bactoblood
Aims
Method
Containers such as micelles and

vesicles are filled with genetic
and metabolic components
Chemical production of
cellular containers, insertion of
metabolic components
Theoretical modelling and

experimental construction
To construct viable
approximations of cells; to
understand biology and the
origin of life
Protocells
DNA with different sets of

base pairs, nucleotides with
different sugar molecules
Searching for alternative

chemical systems with similar
biological functions
Changing structurally
conservative molecules such
as DNA
Using atypical biochemical

systems for biological
processes, creating a parallel
world
Chemical synthetic biology
Source: Schmidt, M. 2009, Do I understand what I can create? in M. Schmidt, A. Kelle, A. Ganguli-Mitra and H. de Vriend (eds), Synthetic Biology. The Technoscience and
Its Societal Consequences, Springer, Dordrecht, p. 84.
Examples
Techniques
Minimal genome
DNA-based bio-circuits
Table 4.1 Breakdown of Subfields of Synthetic Biology
According to a widely accepted historiography of the term synthetic biology,

it entered the scientific vocabulary in 1912 with the publication of the French
chemist Stephane Leducs monograph of the same title.11 As Campos outlines,
the desire among scientists to redesign life can be observed in a number of
different approaches throughout the twentieth century.12
Synthetic biology in its contemporary manifestation was initially promoted
under labels such as open source biology and intentional biology, which were
proposed for the renewed attempt by some to convert biology into a predictive
science by incorporating elements of the engineering design cycle. As Rob
Carlson, for example, stated in 2001: When we can successfully predict the
behavior of designed biological systems, then an intentional biology will exist.
With an explicit engineering component, intentional biology is the opposite of
the current, very nearly random applications of biology as technology.13 Besides
the Molecular Sciences Institute of the University of California at Berkeley
at which Rob Carlson was envisaging a distributed open-source biological
manufacturing system driving future industry, another early institutional hub
of contemporary synthetic biology was located at MIT, where in the Computer
Science and Artificial Intelligence Laboratory (CSAIL) Tom Knight had set up a
biology laboratory and started developing the Biobricks standard for biological
parts with support from a Defense Advanced Research Projects Agency/Office of
Naval Research (DARPA/ONR) contract on Computing with Synthetic Biology.14
The two main vehicles for developing the field of synthetic biology since then
have been the annual student iGEM competition as well as the four SBx.0
conferences. While the first two of these conferences were held in the United
States (Cambridge and Berkeley), SB3.0 took place in Zurich in the summer of
2007 and SB4.0 in Hong Kong in October 2008. Over the course of these SB
conferences it has become obvious that synthetic biology encompasses more
than only the parts-based approach emphasised here. While in this context
clearly some relabelling of more traditional biotechnology and molecular biology
approaches can be observed in order to participate in a newly established, cool
and potentially well-funded discipline, there clearly are different facets to
contemporary synthetic biology that go beyond the engineering of biological
parts, devices and systems.15
11 See, for example, de Lorenzo, V. and Danchin, A. 2008, Synthetic biology: discovering new worlds and
new words, EMBO Reports, vol. 9, pp. 8227.
12 Campos, L. 2009, That was the synthetic biology that was, in Schmidt et al., op. cit., pp. 521.
13 Carlson, R. 2001, Open source biology and its impact on industry, IEEE Spectrum, pp. 1517, as quoted
by Campos, op. cit., p. 17.
14 Knight, T. F. 2002, DARPA BioComp Plasmid Distribution 1.00 of Standard Biobrick Components, <http://
dspace.mit.edu/handle/1721.1/21167>.
15 For an even more detailed subdivision of the field than the one used here, see Lam, C. M. C., Godinho, M.
and dos Santos, V. 2009, An introduction to synthetic biology, in Schmidt et al., op. cit., pp. 2348.
49
Beneficial uses of synthetic biology

Optimistic assessments of synthetic biologys potential, such as the one contained
in the abovementioned report by an EU high-level expert group, expect it to
drive industry, research, education and employment in the life sciences in a way
that might rival the computer industrys development during the 1970s to the
1990s.16 The same group of experts envisages synthetic biology to have such a
dramatic effect by re-organizing biotechnological development in a way so that
research & development are likely to proceed in a much faster and much more
organized way.17 The report identifies six areas that could benefit from such a
streamlining of R&D processes: biomedicine, synthesis of biopharmaceuticals,
sustainable chemical industry, environment and energy, production of smart
materials and biomaterials, and counter-bioterrorism measures.18
Probably the most often quoted example for the imminent breakthrough
of a high-value synthetic biology application is related to the insertion of
an engineered metabolic pathway into a yeast strain to produce artemisinic
acid.19 This in turn can be converted into artemisinin, which forms the basis of
antimalarial drugs. The key goal of the attempt to synthesise this artemisinin
precursor is to reduce production costs for the therapy and thus increase its
availability in developing regions of the world. As Chang and Keasling have
pointed out in relation to this approach, the genes related to the biosynthetic
pathway for a target natural product are transplanted from the natural host into
a genetically tractable host system such as E. coli or S. cerevisiae.20 Similarly,
synthetic biologists have made some progress in utilising genetically modified
microbes to produce biofuels.21
16 Synthetic Biology, op. cit., p. 13.

17Ibid.
18 Ibid., pp. 1317.
19 Keasling, J. et al. 2006, Production of the antimalarial drug precursor artemisinic acid in engineered yeast,
Nature, vol. 440, pp. 9403, <http://www.nature.com/nature/journal/v440/n7086/abs/nature04640.html>.
20 Chang, M. C. Y. and Keasling, J. D. 2006, Production of isopronoid pharmaceuticals by engineered
microbes, Nature Chemical Biology, <doi:10.1038/nchembio836>.
21 See, for example, Clomburg, J. M. and Gonzales, R. 2010, Biofuel production in Escherichia coli: the
role of metabolic engineering and synthetic biology, Applied Microbiology and Biotechnology, vol. 86, pp.
41934; Keasling, J. D. and Chou, H. 2008, Metabolic engineering delivers next-generation biofuels, Nature
Biotechnology, vol. 26, pp. 2989.
50
Synthetic biologys potential for malign misuse

Potential misapplication(s)
The ability to understand, modify and ultimately create new life forms at the
molecular level clearly represents a scientific paradigm shift with a substantial
misuse potential. In general terms, synthetic biology can be misused to engineer
biological parts or modules that increase the efficiency/stability/usability of
known warfare agents or to synthesise new ones.
While currently there are still formidable challenges to overcomefor example,
in the creation of synthetic pathogens22there is a general acknowledgment
that these hurdles will be lowered by scientific and technological advances
over the next few years. In the words of a report jointly published by MIT,
the Centre for Strategic and International Studies (CSIS) and the J. Craig Venter
Institute: In the near future, however, the risk of nefarious use will rise because
of the increasing speed and capability of the technology and its widening
accessibility.23 While this statement refers to the risks related to synthetic
genomics, it is also applicable to the wider area of synthetic biology.
A brief look into two specific applications of major impacts expected from
synthetic biology as described in the abovementioned New and Emerging
Science and Technology (NEST) reportsmart drugs and vectors for therapy
should suffice to illustrate the quantum leap in biological warfare or bioterrorist
capabilities that may result from advances in synthetic biology. According to
the NEST report, a smart drug includes a diagnostic module that is capable
of directly sensing of molecular disease indicators it will only become
active in cells affected by disease.24 The misuse potential of such smart-drug
technology is obvious: if the sensing mechanism were programmed to detect
other, not disease-related indicators and/or the activated chemical compound
were to harm, not cure or would simply be administered in the wrong dosage,
considerable harm could be done with such a device. Similarly, it is conceivable
that newly designed or modified viral vectors that can deliver healthy genes
to the target tissue or that can recognize specific cells and target them for
22 See Epstein, G. L. 2008, The challenges of developing synthetic pathogens, Bulletin of the Atomic
Scientists, vol. 64, pp. 467.
23 Garfinkel, M. S., Endy, D., Epstein, G. L. and Friedmann, R. M. 2007, Synthetic Genomics: Options
for
Governance,
<http://www.jcvi.org/cms/fileadmin/site/research/projects/synthetic-genomics-report/
synthetic-genomics-report.pdf>, p. 12.
24 Synthetic Biology, op. cit., p. 14.
51
destruction25 could be easily diverted from their intended benign use to malign
applications that would, for example, aim at delivering pathogenic genes or
target not cancer, but nerve or other essential cells.
Recognition of dual-use concerns

In parallel with the focus of early activities in synthetic biology in general,
which have been concentrated in the United States, the origins of the discourse
on preventing the misuse of this new field of scientific inquiry also can be traced
back to the US scientific and policymaking communities. In 2004, for example,
George Church put forward A Synthetic Biohazard Non-Proliferation Proposal
to address some of the biosecurity concerns of synthetic biology. The underlying
rationale for this proposal was his identification of the misuse of synthetic
biology as a low-probability, high-consequence event.26 At around the same
time, the first big synthetic biology gathering, SB1.0 at the MIT, also included a
couple of presentations on safety, security and ethical issues related to synthetic
biology. Echoing Churchs assessment of the misuse of synthetic biology as a lowprobability, high-consequence event, George Poste of the Biodesign Institute at
Arizona State University also made reference to the move away from the bug
towards biological circuit disruptors as the object of concern.27 At the Second
International Conference on Synthetic Biology, the third day of the conference
was dedicated to four key societal issues associated with synthetic biology, one
of which was biosecurity and risk.28 Discussions during the corresponding
session were informed by the preparatory work undertaken by the Berkeley
SynBio Policy Group, which had produced a white paper on Community-Based
Options for Improving Safety and Security in Synthetic Biology.29
Also in 2006 the National Research Council tasked a Committee on Advances
in Technology and the Prevention of their Application to Next Generation
Biowarfare Threats to analyse the impact of the revolution in the life sciences
on the evolving biosecurity threat spectrum.30 In dispensing of its task, this
committee rejected a list-based approach, for which it felt that, because of
the pace of research discovery in the life sciences, the useful lifespan of any
25Ibid.
26 Church, G. 2004, A Synthetic Biohazard Nonproliferation Proposal, <http://arep.med.harvard.edu/SBP/
Church_Biohazard04c.htm>.
27 Poste, G. 2004, Synthetic biology: charting rational public policies for the oversight and regulation of
vanguard technologies, 11 June, <http://openwetware.org/images/3/3a/SB1.0_George.Poste.pdf>.
28 See <http://syntheticbiology.org/SB2.0/Biosecurity_and_Biosafety.html>.
29 See Maurer, S. M., Lucas, K. V. and Terrell, S. 2006, From Understanding to Action: Community-Based
Options for Improving Safety and Security in Synthetic Biology, Goldman School of Public Policy, University of
California, Berkeley, Draft 1.1, <http://gspp.berkeley.edu/iths/UC White Paper.pdf>.
30 National Research Council 2006, Globalization, Biosecurity, and the Future of the Life Sciences. Committee
on Advances in Technology and the Prevention of their Application to Next Generation Biowarfare Threats, The
National Academies Press, Washington, DC, <www.nap.edu/catalog.php?record_id=11567>.
52
such list would be measured in months, not years.31 To emphasise this point,
the report pointed out that [n]ew, unexpected discoveries and applications in
RNAi and synthetic biology arose even during the course of deliberations by
this Committee.32 Instead, the committee developed a classification scheme for
science and technology (S&T) advances with four different groups. These four
groups are
1. technologies that seek to acquire novel biological or molecular diversity
2. technologies that seek to generate novel but predetermined and specific
biological or molecular entities through directed design
3. technologies that seek to understand and manipulate biological systems in a
more comprehensive and effective manner
4. technologies that seek to enhance production, delivery and packaging of
biologically active materials.33
Synthetic biology is explicitly mentioned by the committee in relation to the
first two of these categories. A concise discussion of the future applications of
synthetic biology in the report acknowledges that DNA synthesis technology
could allow for the efficient, rapid synthesis of viral and other pathogen
genomeseither for the purposes of vaccine or therapeutic research and
development, or for malevolent purposes or with unintended consequences.34 It
is thus fair to conclude that the biosecurity community during the deliberations
of the Lemon-Relman Committee had clearly identified synthetic biology, albeit
with an emphasis on DNA synthesis and not the four subfields of synthetic
biology outlined above, as one of the technologies that will have a major impact
on the future biothreat spectrum.
This emphasis on DNA synthesis is also reflected in the approach of the National
Science Advisory Board on Biosecurity (NSABB).35 In order to conduct its work,
NSABB can set up working groups to address specific issues including one in
the field of synthetic biology. In the first phase of its work, the NSABB synthetic
biology working group has sought to address biosecurity implications of the
de novo synthesis of select agents.36 A report of the synthetic biology working
group on this issue was discussed during a NSABB meeting in October 2006
31 Ibid., p. 3.
32 Ibid., p. 103.
33 Ibid., p. 3.
34 Ibid., p. 109.
35 National Science Advisory Board for Biosecurity (NSABB) 2006, Addressing Biosecurity Concerns Related
to the Synthesis of Select Agents, National Science Advisory Board for Biosecurity, Washington, DC, <http://
oba.od.nih.gov/biosecurity/pdf/Final_NSABB_Report_on_Synthetic_Genomics.pdf>.
36 Select agents are those biological agents and toxins that can pose a severe threat to public, animal or plant
health, or to animal or plant products. For the current list of select agents, see <http://www.cdc.gov/od/sap/
docs/salist.pdf>.
53
and has subsequently been submitted to the US Government. Clearly, here as

well the focus of the threat assessment had not been on engineered bioparts and
modules, but on improved DNA-synthesis capabilities as they relate to select
agents. Thus, although it is reasonable to assume that since the end of 2006 US
policymakers have been aware of the potential biosecurity risks of synthetic
biology, discussion of these risks has up to now been limited to only a part of
the range of approaches subsumed under the synthetic biology label.
Evolving dual-use potential

Although the main focus of synthetic biologists is on the design and engineering
aspects of this new field of scientific inquiry, some also emphasise the fields
contribution to achieving a better understanding of life processes.37 Clearly,
this better understanding is sought in order to improve the human condition via
improved diagnostics, therapeutics and other beneficial applications; however,
as one study has pointed out, a better understanding of life processes in relation
to regulatory systems in the human body also opens new doors for potential
misuse of biologically active chemical compounds that can target the human
nervous, immune or endocrine systems with a higher degree of specificity.38 To
the extent that synthesised bioparts, devices and modules are utilised for such
misuse, synthetic biologys dual-use potential will represent an incremental
change in the misuse potential of the life sciences in general. In principle,
however, synthetic biology has a much more profound dual-use potential. If the
standardisation of parts and modules progresses to a point where these are truly
compatible and, in addition, can be inserted in a robust chassis for application/
dissemination, a threshold in dual-use potential will be crossed. Given the early
stage of development at which parts-based synthetic biology finds itself, and
the issues related to standardisation referred to above, and more generally the
fact that most of synthetic biology lies at the cutting edge of todays life-science
research, the progress made by synthetic biologists is most likely to be misused
first in an offensive state-level biological weapons (BW) program.
This will change, however, as the dissemination of the knowledge and hands-on
experience with synthetic bioparts and modules continue. The already mentioned
deskilling efforts by some leading synthetic biologists go hand in hand with the
dissemination of do-it-yourself synthetic biology kits and how to protocols. In
addition to these, the iGEM competition also supports this trend. Its broader
goals include: to enable the systematic engineering of biology; to promote the
open and transparent development of tools for engineering biology and to help
37 Synthetic Biology in Europe, TESSY Information leaflet, <www.tessy-europe.eu/public-docs/
SyntheticBiology_TESSY-Infomation-Leaflet.pdf>.
38 See Kelle, A., Nixdorff, K. and Dando, M. 2006, Controlling Biochemical Weapons: Adapting Multilateral
Arms Control for the 21st Century, Palgrave, Basingstoke, UK, esp. chs 46.
54
construct a society that can productively apply biological technology.39 Thus,

over time, the amount of specialist knowledge that is limited to (relatively) few
experts will become increasingly smaller, thereby increasing the likelihood
that sub-state actors and individuals will utilise synthetic bioparts and modules
for nefarious purposes.
Governance proposals for synthetic biology

From the formative days of contemporary synthetic biology, addressing the
question of governance mechanisms and their relative utility has been a feature
of the unfolding discourse on the disciplines societal implications. This early
engagement of the scientific community with the risks associated with synthetic
biology can in part be traced back to earlier debates about the ethical, legal and
social implications of genetic engineering.
As briefly mentioned above, one of the earliest proposals to address synthetic
biologys dual-use implications was put forward by George Church in his
Synthetic Biohazard Non-Proliferation Proposal. In it he advocated developing
a system for both instrument and reagent licensing, and to screen for select
agents, including a DNA agent clearinghouse.40
While Churchs proposal foresaw some form of government involvement or
oversight in both these systems, subsequent governance proposals placed
more emphasis on the self-governance efforts of the scientific community and
commercial DNA-synthesis providers.
Such proposals were discussed at the SB2.0 conference, which took place in
Berkeley, in May 2006, and during which a full day was devoted to discussion
of societal issues surrounding synthetic biology.41 The subsequently formulated
declaration of the conference contains four resolutions that aim at addressing
some of the dual-use implications of synthetic biologyin particular, DNA
synthesis that may give rise to safety or security concerns.42 The focus on DNA
synthesis is also reflected in two of the four resolutions contained in the final
declaration. These resolutions support the
development of improved software tools that can be used to check DNA
synthesis orders for DNA sequences encoding hazardous biological
systems
39<http://parts2.mit.edu/wiki/index.php/About_iGEM>.
40 The proposal is available at <http://arep.med.harvard.edu/SBP/Church_Biohazard04c.htm>.
41 Much of the debate was informed by the white paper produced by Maurer and colleagues mentioned
above.
42 See the revised public draft of the SB2.0 declaration at <https://dspace.mit.edu/handle/1721.1/18185>.
55
[A]doption of best-practice sequence checking technology, including

customer and order validation, by all commercial DNA synthesis
companies
[O]ngoing and future discussions within international science and
engineering research communities for the purpose of developing creative
solutions and frameworks that directly address challenges arising from
the ongoing advances in biological technology, in particular, challenges
to biological security and biological justice
[O]ngoing and future discussions with all stakeholders for the purpose
of developing and analyzing governance options such that the
development and application of biological technology remains
overwhelmingly constructive.43
In terms of practical next steps to be pursued, the declaration announces the
formation of an open working group in support of the improvement of existing
software tools for checking DNA sequences, as well as the completion of a
study to develop policy options that might be used to govern DNA synthesis
technology.44
This study, which was conducted jointly by the MIT, the J. Craig Venter
Institute and the CSIS, supported the trend to focus governance options in
relation to synthetic biology on DNA-synthesis technology. The reports
authors identify DNA synthesis itself, as conducted by gene-synthesis firms
and oligonucleotide manufacturers, and with the help of DNA synthesisers, as
the most effective intervention point for preventing the misuse of synthetic
genomics. The authors of the report concluded that for both gene foundries
and oligo manufacturers, a combination of screening orders by companies and
the certification of orders by a biosafety/biosecurity officer provide the greatest
benefits in terms of preventing incidents. The storage of order information by
firms was regarded as the most useful tool for responding after an incident had
occurred. Lastly, concerning equipment such as DNA synthesisers, the report
concluded that the licensing of both equipment and reagents was most likely to
enhance biosecurity by preventing misuse.45
Ideas developed in this report were subsequently taken up in proposals by the
(then existing) two industry associations in the area of synthetic biology. The
first of these groups, the International Consortium for Polynucleotide Synthesis
(ICPS), put forward a tiered DNA synthesis order screening process.46 This
43 Ibid., p. 3.
44Ibid.
45 Garfinkel, M. S., Endy, D., Epstein, G. L. and Friedman, R. M. 2007, Synthetic Genomics. Options for
Governance, <http://www.csis.org/media/csis/pubs/071017_synthetic_genomics_options.pdf>.
46 Bgl, H. et al. 2007, DNA synthesis and biological security, Nature Biotechnology, vol. 25, pp. 6279.
56
proposal would put DNA-synthesis companies and their industry association

at the centre of a governance structure that would, however, not be a selfcontained system of oversight, but rather rely on agreed-upon guidelines.
Such guidelines would be operationalised, inter alia, through lists of select
agents or sequences that would determine whether and how to process DNAsynthesis orders on the part of those companies that follow the guidelines.
Members of the Industry Association Synthetic Biology (IASB) have focused on
a number of interrelated issues that also revolve around the screening of DNA
orders by synthesis companies. These were formulated during a workshop that
was held in Munich in April 2008 on Technical Solutions for Biosecurity in
Synthetic Biology.47 Motivated by our responsibility for the scientific field to
which we provide services and products,48 workshop participants agreed on
the adoption of five distinct work packages
1. harmonisation of screening strategies for DNA-synthesis orders
2. creation of a central virulence-factor database
3. publication of an article on the status quo of synthetic biology
4. establishment of a technical biosecurity working group with members from
both organisations in order to discuss improvements and next steps for biosecurity measures
5. formulation of a code of conduct.49
With this last work package, the IASB started to expand its activities away from
the technically focused measures into the political arena of setting standards
and promoting best practices.
The focus of proposals put forward has been on the bottleneck of long-strand
DNA synthesis, which has evolved into one of the key enabling technologies
for parts and module-based synthetic biology. Some of these proposals place a
higher emphasis on government involvement than others, but all try to minimise
the negative impact governance might have on scientific progress and economic
benefits. By focusing on DNA synthesis these governance measures have in a
sense targeted the most developed part of the supply chain for the synthesis
of bioparts or modules, whose ancestors can be traced back to the advent of
recombinant DNA technology three decades ago, and which is relatively easily
targetable by traditional supply-side controls.
47 Industry Association Synthetic Biology n.d., Report on the Workshop Technical Solutions for Biosecurity
in Synthetic Biology, <http://www.ia-sb.eu>.
48 Ibid., p. 2.
49 Ibid., p. 16.
57
In contrast, the notion of making a library of bioparts and modules the object of
dual-use informed governance measures has yet to receive substantial attention.
It seems that in this context more thought has been devoted to issues of open
source versus intellectual property rights (IPR), with large biotechnology
companies increasingly discovering synthetic biology for commercial purposes.50
Unless a systematic discourse on dual-use governance structures for parts-based
and the other synthetic biology subfields identified above will commence soon,
it could thus be pre-empted by the IPR-driven attempts to formulate governance
solutions. This could also complicate the realisation of any bioethically informed
dual-use governance approach for synthetic biology that might be developed
out of the approaches and deliberations discussed in the next section.
Bioethical analyses to address the dual-use

aspects of synthetic biology
One of the earliest ethical commentaries on synthetic biology was published
in 1999 in Science51 and has been regularly quoted by practitioners in the field
and observers alike. It acknowledged that in order to ensure responsible use
of knowledge that could be applied to the construction of biological weapons,
we need to give serious thought to monitoring and regulation at the level of
national and international public policy.52
Yet, as Yearley has concluded a decade later, while Cho and her colleagues
highlight things that people may have ethical concerns about, the paper does
not set out or determine what the ethical analysis might conclude.53 More
fundamentally, Yearleys review of bioethics as a template to conduct an ethical
review of synthetic biology leads him to caution that this might actually be
counterproductive since the apparatus constructed to conduct the social and
ethical review will come to look like a mere legitimatory cloak for synthetic
biologys advance.54 Yet, Yearley remains somewhat elusive in specifying what
else might serve as a suitable foundation for an ethical review of synthetic
biology.
50 Oye, K. A. and Wellhausen, R. 2009, The intellectual commons and property in synthetic biology, in
Schmidt et al., op. cit., pp. 12140.
51 Cho, M. K., Magnus, D., Caplan, A. L., McGee, D. and the Ethics of Genomics Group 1999, Ethical
considerations in synthesizing a minimal genome, Science, vol. 286, pp. 2087, 208990.
52 Ibid., p. 2089.
53 Yearley, S. 2007, Review: the ethical landscape: identifying the right way to think about the ethical
and societal aspects of synthetic biology research and products, Journal of the Royal Society Interface,
<doi:10.1098/rsif.2009.0055.focus>, p. 3.
54 Ibid., p. 6.
58
Miller and Selgelid, in contrast, provide a much more detailed and in-depth
ethical and philosophical consideration of dual-use issues in the life sciences.55
Although not specifically targeted at synthetic biology, their analysis illuminates
many aspects of bioethical reasoning that are of relevance to synthetic biology
too. Based on a particularly morally problematic species of the dual-use
dilemma,56 in the form of a number of experiments of concern, they discuss
the permissibility of certain kinds of research, debate dissemination of dualuse research results, and analyse different ethically informed governance
models with which to tackle the dual-use issues presented by the biological
sciences. With regard to this last aspect, Miller and Selgelid discard both the
laissez-faire option of giving the individual scientist complete autonomy over
their research with dual-use potential and the rather draconian option of
complete governmental control. Instead they argue for either a mixed system of
institutional and governmental controls or a governance approach that would
rely on an independent authority being set up.57 Thus, while providing a
detailed discussion of ethical issues in relation to dual-use life-sciences research,
and a narrowing ofin their viewsuitable governance options, the analysis
remains in this latter dimension somewhat inconclusive.
In contrast with Miller and Selgelid, Ehnis discussion of the ethical
responsibilities of scientists engaged in dual-use research is more limited in
scope.58 He approaches the issue by discussing the basic conflict between the
freedom of science and the duty to avoid causing harm from two perspectives
that of moral skepticism and the ethics of responsibility by Hans Jonas.59 On
this basis, Ehni evaluates
four basic duties to define the prospective responsibility of scientists:
1) stopping research in some cases, 2) systematically exploring dangers
of dual use in some cases, 3) informing public authorities about possible
dangers resulting from research and the application of its results, and 4)
not publishing results and descriptions of research results and possible
dual-use applications.60
Given the nature of the dual-use issues at hand and the way science is organised,
Ehni concludes along the lines of Miller and Selgelid that [i]t is no solution to
55 Miller, S. and Selgelid, M. J. 2007, Ethical and philosophical considerations in the dual-use dilemma in
the biological sciences, Science and Engineering Ethics, vol. 13, pp. 52380.
56 Ibid., p. 531.
57 Ibid., p. 573.
58 Ehni, H.-J. 2008, Dual use and the ethical responsibility of scientists, Archivum Immunologiae Et
Therapiae Experimentalis, vol. 56, pp. 14752.
59 Ibid., p. 147.
60 Ibid., p. 151.
59
the dual-use problem to transfer total responsibility to individuals.61 Instead

he advocates a mixed authority for the governance of dual-use issues, the
details of which he also leaves unspecified.
While the recommendation of not assigning sole responsibility to individual
scientists is shared by Kuhlau et al., they point out that the moral duty to
prevent harm on the part of an individual researcher does exist and includes
both intentional and unintentional harms.62 Furthermore, such moral duty
linked to the professional role of the researcher carries with it a requirement for
an awareness of relevant regulation and potential dangers.63 This awareness in
turn entails a continuous process of reviewing ones work in a wider context.64
Based on this reasoning, they identify five criteria for the obligation to prevent
harm.
In order to take social responsibility and due care, life scientists should
strive to prevent harm that is: Within their professional responsibility
Within their professional capacity and ability Reasonably foreseeable
Proportionally greater than the benefits [and] Not more easily
achieved by other means.65
Kuhlau et al. subsequently apply these criteria to a number of proposed
obligations for life scientists in relation to dual-use issues. They conclude that
scientific responsibility does not involve preventing the act of misuse but
rather involves obligations concerned with preventing foreseeable and highly
probable harm (emphasis in original).66 Such reasonable obligations include,
in their view, the duties to consider negative research implications, to protect
sensitive material, technology and knowledge from unauthorised access, and to
report suspicious activities.67 While this again represents a useful clarification
of the obligation of life scientists including synthetic biologists in relation to
dual-use issues they face in their work, it leaves both the institutional and the
wider governance contexts unexplored within which these duties need to be
considered.
The European Group on Ethics (EGE) in Science and New Technologies to the
European Commission in its Opinion No. 25 on the Ethics of synthetic biology
usefully points out in this respect that [g]overnance is an overarching concept
61Ibid.
62 Kuhlau, F., Erikson, S., Evers, K. and Hglund, A. T. 2008, Taking due care: moral obligations in dual
use research, Bioethics, vol. 22, pp. 47787.
63 Ibid., p. 481.
64Ibid.
65 Ibid., p. 481 ff.
66 Ibid., p. 487.
67Ibid.
60
including legal, political and ethical considerations. Since synthetic biology may
result in major changes of traditional biology, governance needs to be reflected
on all these levels, finally entering the legal sphere.68
With respect to ethical consideration of synthetic biology, the EGE distinguishes
between conceptual and specific issues, and addresses both biosafety and
biosecurity under the latter heading.69 In its discussion of potential steps to
be taken, the EGE opinion states under the heading of Biosecurity, prevention
of bioterrorism and dual uses that [e]thical analysis must assess the balance
between security and transparency,70 and moves on to recommend that ethical
issues that arise because of the potential for dual use should be dealt with at the
educational level. Fostering individual and institutional responsibility through
ethics discussion on synthetic biology is a key issue.71
In addition, the EGE opinion contains three formal recommendations: 1)
linking dual-use bioethics to the Biological Weapons Convention (BWC) by
recommending that this international treaty should incorporate provisions on
the limitation or prohibition of research in synthetic biology; 2) requesting the
European Commission to define a comprehensive security and ethics framework
for synthetic biology; and 3) requesting the establishment of DNA-sequence
databases with supporting, legally based rules and procedures.72
A similarly wide-ranging attempt to chart the ethical issues surrounding
synthetic biology was undertaken by the US Presidential Commission for the
Study of Bioethical Issues (PCSBI), which in December 2010 produced its first
report, entitled New Directions: The Ethics of Synthetic Biology and Emerging
Technologies.73 Guided by five ethical principlesthat is, (1) public beneficence,
(2) responsible stewardship, (3) intellectual freedom and responsibility, (4)
democratic deliberation, and (5) justice and fairness74the report arrives at
18 recommendations, some of which are informed by the dual-use character of
synthetic biology or seek to address its implications. Of particular relevance
in this context are recommendations 12 and 13. Acknowledging the dynamic
character of the field and the resulting changes in dual-use issues of relevance,
the committee recommends periodic assessments of safety and security risks be
undertaken. It states:
68 European Group on Ethics 2009, Ethics of Synthetic Biology, Opinion No. 25, European Union, Brussels,
p. 36.
69 Ibid., pp. 424.
70 Ibid., p. 51.
71 Ibid., p. 52.
72Ibid.
73 Presidential Commission for the Study of Bioethical Issues (PCSBI) 2010, New Directions: The Ethics
of Synthetic Biology and Emerging Technologies, Presidential Commission for the Study of Bioethical Issues,
Washington, DC.
74 Ibid., p. 5.
61
Risks to security and safety can vary depending on the setting in which
research occurs. Activities in institutional settings, may, though certainly
do not always, pose lower risks than those in non-institutional settings.
At this time, the risks posed by synthetic biology activities in both
settings appear to be appropriately managed. As the field progresses,
however, the government should continue to assess specific security and
safety risks of synthetic biology research activities in both institutional
and non-institutional settings including, but not limited to, the do-ityourself community An initial review should be completed within
18 months and the results made public to the extent permitted by law.75
In case this review identifies significant unmanaged security or safety
concerns, recommendation 13 foresees changes to existing oversight and
control mechanisms with a view to making compliance with certain oversight
or reporting measures mandatory for all researchers regardless of funding
sources.76 This last point would lead to a significant tightening of existing
oversight mechanisms as it would expand their reach beyond publicly funded
life-science research and oblige commercial research activities to abide by the
same regulatory framework.
Conclusions
This chapter set out to first illustrate that synthetic biology is one of the most
dynamic new subfields of the life sciences. It offers the potential to live up to
the promise that the discipline behind the label of genetic engineering has long
aspired to: the engineering of biological parts, devices and systems, either to
modify existing or to create new ones. By applying the toolbox of engineering
disciplines and information technology to biology, a wide range of potential
applications becomes possible, ranging across scientific and engineering
disciplines. Some of the anticipated benefits of synthetic biology, such as the
development of low-cost drugs or the production of chemicals and energy by
engineered bacteria, are potentially very significant. There are, however, also
significant risks due to deliberate or accidental damage. In a way, synthetic
biology can be described as the prototypical emerging dual-use technoscience.
Although first attempts at formulating governance mechanisms can be identified,
these are for the most part focusing only on a subfield, or, as some would say,
enabling technology, of synthetic biologythat is, large-scale commercial
DNA synthesis. The conceptualisation of dual-use governance processes and
structures for the bioparts and modules-based approach within synthetic
75 Ibid., p. 13.
76 Ibid., p. 14.
62
biology is, by contrast, still in its infancy (as is the case for the other subfields
outlined above). Academic work on the characteristics of dual-use issues from
an (bio-)ethical perspective has increased in numbers over recent years, but
stops short of considering embedding their analysis and recommendations
into the wider institutional or political context that synthetic biologists find
themselves in. Similarly, the opinions and recommendations of advisory bodies
and committees briefly discussed are quite generic and deal with dual-use issues
among many other ethical questions raised by synthetic biology.
What are required are thus more detailed analyses of the dual-use implications
of the whole of synthetic biology and systematic dual-use bioethics awarenessraising efforts that reach all practising synthetic biologists and that are
supplemented by education and training efforts as well as the formulation of
codes and other governance tools that go well beyond the rather technically
orientated order screening by DNA-synthesis companies.
63
5. Crops Agents, Phytopathology

and Ethical Review
Simon Whitby
Introduction
This chapter explores the dual-use quality of scientific research and technological
development in the field of phytopathology (plant science). It offers a brief
survey of naturally occurring pathogens that have been developed for use in
weapons and considers areas of convergence and overlap between the hostile
use of disease organisms as a form of warfare and the peaceful deployment of
bio-control and plant inoculants. The relevance of bio-control agent and plant
inoculant production to the Biological Weapons Convention (BWC) is then
considered. Included in this chapter also is a snapshot of some significant
developments in civil plant science, alluding to the scale and speed of progress
in plant science and technology.
I argue that since they can be used for both peaceful and hostile purposes, plant
science and technology raise issues of dual-use biosecurity concern that are
thus worthy of ethical consideration. In this connection, this chapter argues
that ethical review processes could usefully be located alongside deliberative
processes that facilitate consideration of the legal and social implications of
plant-science research. Deliberation regarding its potential as dual-use research
of concern may therefore be best located within the context of a comprehensive
system for oversight of scientific research such as that recommended by the US
National Science Advisory Board for Biosecurity (NSABB).1 Therefore, a brief
survey of the contours of the latter is included in the concluding section of the
chapter, which focuses specifically on the role of the principal investigator (PI)
the most critical element in the oversight of dual-use life-sciences research2
and the requirement to seek to ensure through improved awareness and training
that PIs are sufficiently aware of dual-use research issues and concerns.
National Science Advisory Board for Biosecurity, Bethesda, Md.
2 Ibid., p. 11.
65
Deliberate disease for hostile purposes:

Biological warfare
As part of its offensive biological warfare program that originated in the early
1940s, the United States developed an extensive anticrop program that resulted
in the production, development, stockpiling and assimilation of anticrop
chemical and biological weapons agents. As is well documented,3 chemical
anticrop agents and defoliants were utilised extensively during the course of
offensive operations in Vietnam in the 1960s and 1970s.
A range of biological warfare anticrop agents was the focus of extensive
research and development during the course of the offensive US anticrop
program between the mid 1940s and 1969, and research on production, scaleup, storage, dissemination and effectiveness saw the standardisation of agents
for crop destruction with targets including both staple food crops and narcotics
production. Since the United States provides an example of a systematic
offensive program from research to assimilation, attention will turn briefly to
list the agents standardised for the conduct of anticrop warfare.
Agents in the US stockpile as documented4 in 1970, including for the destruction
of food crops, are listed in Table 5.1. An early attempt to identify agents for the
destruction of narcotics crops is also included.
Table 5.1 US Stockpiles in 1970 for Food Crop Destruction
Agents category
Type
Hosts
Category A
agents
Stem rust of wheat
Wheat, barberry and certain grasses
Category B
agents
Rice blast
Rice, possibly some other grasses
Stripe rust of wheat
Wheat, barley, various grasses
Category C
agents
Hoja blanca of rice
Rice, wheat, corn, barley, rye, sorghum and

various other grasses
Bacterial leaf blight of rice
Rice and various grasses
Downy mildew of poppy
Species of Papaver and Argemone
Source: Stockholm International Peace Research Institute (SIPRI) 1973, The Problem of Chemical and
Biological Wafare, vol II, Stockholm International Peace Research Institute, Stockholm.
3 Cecil, F. 1986, Herbicidal Warfare: The RANCH HAND Project in Vietnam, Praeger, New York. See also,
Karnow, S. 1997, Vietnam: A History, 2nd edn, Penguin, New York.
4 Stockholm International Peace Research Institute (SIPRI) 1973, The Problem of Chemical and Biological
Wafare, vol II, Stockholm International Peace Research Institute, Stockholm.
66
5. Crops Agents, Phytopathology and Ethical Review
Following President Richard Nixons unilateral renouncement of offensive

biological warfare in 1969 the stockpile scheduled for destruction5 in 1970
included 158 684 lb (71 t) of the causal agent of wheat rust, and 1865 lb (846
kg) of the causal agent of rice blast (as well as munitions for their deployment).
In a 1980s study based on publicly available secondary sources, Geissler noted
that prior to 1969at a time of an emerging international consensus towards
the agreement of a complete ban on biological warfarethe majority of military
work on pathogens had involved bacteria and fungi. The last included those
agents that were developed to attack some of the worlds most economically
and socially significant food and cash crops. In a program of research and
development that appeared to parallel the US biological weapons (BW) program,
Iraqs modest late-1980s attempts to conduct research and development into
crop warfare also focused on investigations into the effectiveness of fungal plant
pathogens.6 According to Geissler, though, by 1983, the majority of military
work on pathogens had switched from bacteria and fungi to instead focus on
investigations into the effectiveness of viruses.
The latter featured as agents of choice in what might be regarded as secondgeneration programs, not least as they appear to have featured in work in the
former Soviet Union.7 With the advent of genetic-engineering techniques,
Geissler noted a renewed military interest in biological warfare post 1981, with
Dr Kenneth Alibek8 (a former Soviet biological weapons scientist) subsequently
alluding to activities in the former Soviet Union in the early 1980s that included
investigations into the production of genetically engineered antibiotic strains of
a number of zoonotic, and antipersonnel agents, including anthrax and glanders.
According to Geissler, these events were characteristic of the emergence of
a third generation of scientific and technological applications in offensive
biological warfare. Although there is no publicly available information relating
to third-generation research and development in offensive anticrop military
programs, it would be unwise to rule out the possibility of improvements in
destructive effectiveness. Indeed, outside military programs, ample evidence
suggests that routine genetic manipulation is being deployed for peaceful
purposes in the burgeoning area of research, development and deployment of
bio-control agents and plant inoculants in agriculture.
5Ibid.
6 United Nations Special Commission (UNSCOM) 1995, Report to the Secretary-General, 11 October.
7 Tucker, J. 1999, Biological weapons in the former Soviet Union: an interview with Dr. Kenneth Alibek,
The Nonproliferation Review, (SpringSummer), p. 2.
8Ibid.
67
Peaceful bio-control that could serve biological

warfare
The focus of research in this area was in the development and deployment of
naturally occurring pathogens and insects in the protection of crops from disease
caused by pathogens and disease caused or transmitted by insect vectors. The
deployment of bio-control agents and plant inoculants marks this area out as
distinct since their use for peaceful purposes is not prohibited by the 1972
BWC. Nevertheless, during the course of talks to negotiate a legally binding
protocol to strengthen the effectiveness and improve the implementation of the
BWC in the 1990s and early 2000s, discussion focused on initiatives intended
to ensure that scientific and technological developments in this area were in
compliance with the objectives and scope of the convention.
In a statement9 by the South African delegation to the Fifth Review Conference,
on 19 November 2001, Peter Goosen, Department of Foreign Affairs, reminded
states parties that in all of their work related to the BWC the threat against
plants was usually considered to be of a lower priority than the threat against
humans. This had occurred, Goosen pointed out, in spite of the widespread
appreciation amongst states parties that major elements of biological weapons
programs since the 1920s had been directed against crops and that numerous
plant pathogens had been researched, developed and produced together with
weapons as part of offensive BW programs for the purpose of the widespread
dissemination of anticrop agents.10
No clear distinctions, however, separate pathogens in this area from those
deployed in offensive biological warfare programs. Bio-control agents are living
organisms, such as bacteria, fungi, insects, mites or weeds, or microorganisms
that are used in the control of microbes or other organisms. A large number of biocontrol agents are currently availablefor example, in the United States, where
they are marketed as bio-pesticides and include bacteria such as Agrobacterium,
the widely used Bacillus thuringiensis that produces a protein toxic to species
of insect pests belonging to the orders Lepidoptera (caterpillars), Diptera (flies)
and Coleoptera (beetles and weevils), Pseudomonas and Streptomyces. Further
bio-pesticides include fungi such as Ampelomyces, Candida, Coniothyrium
9 Goosen, P. 2001, Statement by Chief Director: Peace and Security, Department of Foreign Affairs,
Pretoria, South African Delegation to the Fifth Review Conference of the Convention on the Prohibition
of the Development, Production, and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on
their Destruction, Geneva, 19 November, <http://www.opbw.org/rev_cons/5rc/docs/statements/5RC-OSSAFRICA.pdf>.
10 For a systematic study of such state offensive anticrop biological warfare programs, see: Whitby, S. 2001,
Biological Warfare against Crops, Palgrave, London.
68
and Trichoderma.11 Plant inoculants are formulations containing living microorganisms, used in the treatment and propagation of seeds and plant propagation
matriel for enhancing growth and disease resistance in plants. They are also
used for the restoration of the microflora of soil. Indeed the technologies
associated with the dissemination of such agents appear to equate with those
used in the dissemination of biological warfare agents.
Prior to the First Review Conference of the BWC, the Preparatory Committee
requested that depositary governments prepare a background paper12 on new
scientific and technological developments relevant to the convention and
invited states parties to submit their views on new scientific and technological
developments relevant to the convention. Prepared by experts of the depositary
governments, the review13 focused on new scientific and technological
developments relevant to the convention and looked inter alia at the microbial
control of pests.14 Since significant environmental and human health implications
arose from the deployment of synthetic chemical pesticides that had seen
extensive use in Vietnam, this section of the report noted environmental and
human health concerns and questioned the efficacy of the use of agents against
plants that might develop resistance to their use. The review noted, however,
that there had been a remarkable increase in interest in this area. This was
summarised as follows:
Microbiological methods involve the large-scale production of certain
live micro-organisms or their extractable toxins, the formulation of a
liquid or powder product and dissemination of the product by vehicle
or aircraft-borne sprays (or in rodent control, the use of ground bait)
over crops or forests. With live microbial agents death of insect or rodent
occurs through infection; with microbial toxins death is produced by
toxic effects. In some basic respects the whole sequence resembles biological
warfare. [Emphasis added]
Table 5.2 illustrates the methods of production and dissemination of viral,
bacterial and fungal bio-control agents of relevance to the BWC.
11 McSpadden Gardener, B. B. and Fravel, D. R. 2002, Biological control of plant pathogens: research,
commercialisation, and application in the USA, Plant Health Progress, [Online], <doi:10.1094/PHP-20020510-01-RV>.
12 Report of the Preparatory Committee for the Review Conference of the Parties to the Convention on the
Prohibition of the Development, Production, and Stockpiling of Bacteriological (Biological) and Toxin Weapons
and on their Destruction, BWC/CONF.I/5, 6 February 1980, <http://www.opbw.org>.
13 Not all states parties submitted information to the Secretary-General of the United Nations that referred
either directly or indirectly to potential problems posed by the use of microbial agents against crops. For the
purpose of this discussion, it has been necessary to refer selectively to the official documentation.
14 Report of the Preparatory Committee, op. cit., Appendix E.
69
The United Kingdoms contribution15 to a subsequent BWC Review Conference

focused also on the microbial control of pests. It included an assessment of
increased interest in biological control and noted apparent changes to the
methods of production:16
Increases in use since 1980 have not been spectacular except possibly
in the nations of Eastern Europe, where Lepidoptera pests are a greater
agricultural problem than elsewhere. In such nations about 30 different
microbial preparations or formulations have been developed and some
are produced on often multi-tonne scales. GE [genetic engineering] is
being applied in many nations to the development of improved and novel
agents for pest control. Obviously large-scale industrial microbiology is
a key aspect of production Insect viruses for pest control continue
to be relatively expensive to produce but the possibility that viruses
could be more cheaply and effectively produced through GE, rather
than by bulk production in insects, is likely to result in widely-adopted
production methods in nations where susceptible pests are a problem.
Table 5.2 Methods of Production and Dissemination of Viral, Bacterial and
Fungal Bio-control Agents
Viruses
Nuclear polyhedrosis and granulosis viruses: produced on a large scale by

a few nations, using mass rearing of insect hosts. The viruses extracted at
a concentration of 12 x 106 infective units/ml are sprayed by aircraft. Viral
insecticides are more expensive to produce than bacterial insecticides but they
have the possible advantage of high target specificity.
Bacteria
Bacillus thuringiensis: produced by several nations on a multi-tonne basis

in deep-aerated vessels. The final product contains about 3 x 1010 bacterial
spores/g, and is stable for two to three years. Disseminated by aircraft spray as
liquid or powder aerosol, the bacterium is highly valued for controlling a wide
variety of insect pests.
Bacillus popilliae: another agent produced and used in much the same way as
Bacillus thuringiensis for controlling Japanese beetle larvae.
Pseudomanas seruginosa (and Pl. fluorescens) and Chromobacterium
prodigiosum: produced and used in a few countries for dissemination by aircraft
spray on reservoirs (at 10 x 106 organisms/sq cm water surface) in mosquito
larvae control. These agents are, however, facultative* pathogens for humans.
Fungi
Various species such as Trichoderma, Sporotrichum, Beauveria and

Cuelomemvces are produced on a multi-tonne basis by several nations. They are
disseminated by aircraft spray to infect insect pests and sometimes to attack
other fungal diseases of crops. Additionally, a number of other microbial agents
are currently being studied or evaluated in field trails.
* Facultative pathogens include those with mechanisms for infecting human body tissue.
Source: Report of the Preparatory Committee for the Review Conference of the Parties to the Convention on the
70
15 Background Document on New Scientific and Technological Developments Relevant to the Convention on the
Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and
on their Destruction, BWC/CONF.11/4, 6. Microbial Control of Pests, p. 8, <http://www.opbw.org>.
16 Ibid., para. 6.2, p. 8.
Regarding methods of dissemination for microbial methods of pest control, the

UK contribution17 noted the following:
Methods of dissemination of microbial pest control agents continue to
be the subject of increasing R&D and trials. Mobile jet-engined devices
are capable of disseminating agent aerosols, notably insect viruses, over
vast tracts of land. There has been continuing R&D on ultra-low volume
spraying systems, methods of studying spray deposition, formulations,
the problems of disseminating dusts and powders, micro-encapsulation
and other relevant topics.
A section on developments of relevance to the BWC in regard to the microbial
control of pests was also included in the UK contribution, which noted the
following (Table 5.3).
Table 5.3 Microbial Control of Pests
a.
GE-derived bacteria with high toxin yields.
b.
The production through GE of toxins in species beyond those bacteria that produce
them in nature.
c.
Development of formulations aimed at enhanced retention of microbial viability during

storage and in aerosol.
d.
Protection of aerosolised micro-organisms by the incorporation of protective UV-light

screening dyes.
e.
Improvements in the spray-drying and milling of micro-organisms and toxins.
f.
The formulation of synergistic combinations of live micro-organisms and toxic

anticoagulants, together with drug-delivery systems.
g.
Development of automated production lines for insects, used in the production of

some viruses.
h.
Vastly increased knowledge of aerobiological aspects of dissemination and the

elucidation of the factors that control viability and stability in dissemination, aerosol
and in respect of persistence.
i.
Computer-controlled continuous culture systems and improved purification systems.
Source: Report of the Preparatory Committee for the Review Conference of the Parties to the Convention on the
In its concluding remarks, the UK contribution noted a greater potential for

abuse across a range of civil capabilities than was present at the time of the
First BWC Review Conference in 1980. In particular, with regard to microbial
methods of pest control, it noted
the biotechnology explosion in the civil sectors of many nations and
the realization of the potentials of GE and industrial microbiology.
We have drawn attention again to developments in microbial methods
of pest control and to the increasing knowledge of nations in the
17 Ibid., para. 6.2, p. 8.
71
large-scale production and dissemination of micro-organisms and

microbial products. Such developments in the civil sector are relevant
to the BWC and could be abused to support offensive programmes.
The South African submission18 commented extensively and in considerable
detail in regard to developments relating to plant inoculants and biological
control relevant to the convention. In regard to plant inoculants, separate
detailed subsections of the South African submission gave an overview of: the
history of their use and development; their purpose; their mode of action; types
of inoculants; methods of inoculation; production methods; and the relevance
of plant inoculants for the BWC. And in regard to bio-control agents, separate
detailed subsections of the South African submission gave an overview of:
differing approaches to biological control; the complexity of factors affecting
their application; and biological agents against plantsthe last including the
controversial area of attacking drug crops.
According to the South African submission,19 plant inoculants are relevant to
the convention in terms of:
a. A growing industry and more sophisticated production facilities that
have the potential to be diverted to BW producing facilities, as in the
case of vaccine production facilities.
b. The genetic research and development that is conducted to improve
the micro-organisms that form the active ingredients of inoculants.
c. The development of liquid inoculants that will make their application
by spraying and aerosolisation a possibility.
In comparison, the control of plant pests, weeds and plants with biological
control agents is relevant to the BWC in terms of:
a. The less clear distinction between the peaceful use of biocontrol
agents and their use as BW due to the dual-use nature of these agents.
b. Undesired plants, exotic plants or even noxious plants in one country
may be natural, essential and in many cases utilised for commercial
purposes (crops) in other countries.
The failure of states parties at the Fifth Review Conference to produce a
final declaration20 meant that the usual extended understandings were not
produced during the BWC in 200102; South Africa urged co-depositaries and
18 Ibid., para. 6.2, p. 5.
19 Ibid., para. 6.2, p. 7.
20 Pearson, G. S. and Sims, N. A. 2005, Preparing for the BTWC Sixth Review Conference in 2006, Bradford
Review Conference Paper No. 10 (February), <http://www.brad.ac.uk/acad/sbtwc/briefing/RCP_10.pdf>.
72
states parties to give careful consideration to the issues of biological control at

the Sixth Review Conference of the Convention in 2006 and recommended that
CBM declarations be extended to include animal and plant pathogen research
and production facilities.
As noted by Kelle and colleagues21and echoing some of the sentiments
expressed by Geissler abovebiological warfare capabilities have been
determined by developments in the life sciences, but the genomics22 revolution
has changed dramatically the nature and scope of biological warfare.
Recognition of the challenge posed by rapid advances in science and technology
was, for example, included in the Final Declaration23 of the Second Review
Conference of the Biological and Toxin Weapons Convention in 1986. Thus, such
concern was embodied in additional understandings agreed on by states parties
that:
The Conference, conscious of apprehensions arising from relevant
scientific and technological developments, inter alia, in the fields
of microbiology, genetic engineering and biotechnology, and the
possibilities of their use for purposes inconsistent with the objectives of
the Convention, reaffirms that the undertaking given by States Parties in
Article 1 applies to all such developments.
Subsequent reviews by states parties of scientific and technological developments
of relevance to the BWC have been conducted (on a voluntary basis) by states
parties, and since 1980 submissions to the five-year reviews of the convention
have noted that significant advances of relevance to the convention have taken
place in the fields of biotechnology, genetic modification and genomics.
21 Kelle, A., Nixdorff, K. and Dando, M. 2006, Controlling Biochemical Weapons: Adapting Multilateral Arms
Control for the 21st Century, Palgrave, Basingstoke, UK, p. 35.
22 The sum total of genetic information of an individual, which is encoded in the structure of
deoxyribonucleic acid (DNA), is called a genome. The study of the genome is termed genomics. Recently, the
order of most of the chemical building blocks, or bases, which constitute the DNA of the genomes of human
beings (estimated to amount to three billion), several other animal species and a variety of human pathogens
and plants has been determined. Over the next few years this remarkable achievement will be completed and
augmented by research into functional genomics, which aims to characterise the many different genes that
constitute these genomes and their variability of action. Such research will also determine how these genes
are regulated and interact with each other and with the environment to control the complex biochemical
functions of living organisms, both in health and in disease.
23 Second Review Conference of the Parties to the Convention on the Prohibition of the Development,
Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction, Final
Document PART II Final Declaration, BWC/CONF.II/13/II, <www.opbw.org>.
73
Science and technology: Phytopathology

In the field of phytopathology, it became apparent by 1996 that genomics would
play an increasingly important role in plant biotechnology, and 2006 marked
a decade of important scientific and technological developments. During the
course of this period plant science provided a glimpse into the huge agronomic
and social potential of plants.
In recognition of the fulfilment of objectives set by the Arabidopsis genome
initiative in 1996, a news article in a December 2000 edition of Nature,24 titled
A green chapter in the book of life: the sequencing of an entire plant genome
is now complete , and the genome analysis in the same edition, signalled
a landmark in plant science. The sequencing of the remaining three of five
chromosomes, and therefore of the complete gene of Arabidopsis thaliana, a
small flowering plant that is a member of the mustard family commonly known
as thale cress, represented a major scientific and technological breakthrough,
with Arabidopsis representing a model organism in plant biology. This
development opened up the possibility of investigating the genetic complexity
of more economically and socially significant plant life. Citing Bonny, the World
Health Organisation (WHO)25 highlights the further potential benefits of the
new science and technology thus: The potential uses of modern biotechnology
in agriculture include: increasing yields while reducing inputs of fertilizers,
herbicides and insecticides; conferring drought or salt tolerance on crop plants;
increasing shelf-life; reducing postharvest losses; increasing the nutrient
content of produce; and delivering vaccines.
The cheap availability, short life cycle, small physical size and the small size
of the Arabidopsis genome118.7 million base pairsrelative to other more
complex plants, meant that plant science was now able to undertake, under
controlled conditions, the identification of the genes responsible for a wide range
of physiological processes. In fundamental terms, finding the genes responsible
for a plants physiological response to, inter alia, light, soil and soil nutrients,
bacterial, fungal and viral plant pathogens and insect pests, now promised to
be much simpler.
Rapid progress in plant science was soon also made regarding the rice genome.
The 10-nation International Rice Genome Sequencing Project announced the
24 Walbot, V. 2000, A green chapter in the book of life, Nature, vol. 408, p. 794. See also The Arabidopsis
Genome Initiative 2000, Analysis of the genome sequence of the flowering plant Arabidopsis thaliana,
Nature, vol. 408 (December), pp. 796815.
25 World Health Organisation (WHO) 2005, Modern Food Biotechnology, Human Health and Development:
An Evidence-Based Study, World Health Organisation, Geneva, p. 37, <http://www.who.int/foodsafety/
publications/biotech/biotech_en.pdf>.
74
sequencing of the second complete genome,26 that of rice, Oryza sativa, in 2005.
The economic and social importance of rice is significant. Rice is the worlds most
important food crop, consumed by more than half of the worlds population,
and to meet projected demand over the next 20 years production will have to
rise by an estimated 30 per cent.
In connection with progress in plant science outlined above, a first generation
of genetically modified crop products that have emerged in the marketplace
over the past decade has expressed a limited number of characteristics. In
particular, since 1996, millions of acres have been used for the production of
genetically modified crops with, in the case of the United States, large-scale
production of modified varieties of corn, cotton and soya (soya beans) enhanced
by gene-transfer techniques that confer herbicide tolerance and insect resistant
in crops. According to the International Service for the Acquisition of AgriBiotech Applications report Global Status of Biotech Crops,27 as early as 2005 the
United States had approximately 49.8 million ha planted with crops that were
the product of genetic modification. The extent to which some of these crops
have been adopted in US agriculture and the short time span over which this has
taken place are perhaps indicative of the willingness with which agricultural
enterprises in the United States have embraced such technologies in spite of
concerns raised regarding human health and the environment. According to
the US Department of Agriculture (USDA),28 planting of herbicide-tolerant soya
beans expanded from 17 per cent of US soya bean acreage in 1997 to more than
85 per cent in 2005. Planting with herbicide-tolerant cotton expanded from 10
per cent of US acreage in 1997 to more than 60 per cent in 2005. Planting with
insect-resistant transgenic corn and cotton containing the Bacillus thuringiensis
(Bt) toxin gene also increased significantly over this period in the United States.
USDA figures29 for 1995 reveal the [a]doption of all GE cotton, taking into
account the acreage with either or both HT and Bt traits, reached 79 percent in
2005, versus 87 percent for soybeans. In contrast, adoption of all biotech corn
was 52 percent.
According to the International Service for the Acquisition of Agri-biotech
Applications (ISAAA),30 global biotech planting had exceeded 1 billion ha by
2010.
26 For an analysis of some of the salient features of the rice genome, see: International Rice Genome
Sequencing Project 2005, The map-based sequence of the rice genome, Nature, vol. 436 (August), p. 793.
27 Available at: <http://www.isaaa.org/>.
28 United States Department of Agriculture (USDA) n.d., Adoption of Genetically Engineered Crops in the
U.S.: Extent of Adoption, Economic Research Service, United States Department of Agriculture, Washington,
DC, <http://www.ers.usda.gov/Data/biotechcrops/adoption.htm>.
29Ibid.
30 International Service for the Acquisition of Agri-biotech Applications (ISAAA) 2010, Global Status of
Commercialized Biotech/GM Crops: 2010, ISAAA Brief, <http://www. ISAAA.org>.
75
In the past two decades, genome studies have facilitated manipulation of the
genetic characteristics of food crops. Crops can now be produced with built-in
defences against insect pathogens such as Bacillus thuringiensis. They can also
be manipulated to delay ripening, as in the case of the slow-ripening Flavr Savr
tomato, which was approved for sale in the United States in 1994. Infertility can
be conferred on plant seeds, as in the case of the controversial Terminator gene.
Plant science is, however, now beginning to focus on a second generation of
crops that have been genetically modified to express a broader and more complex
range of plant traits. The challenge now also extends to assigning functions to
genes, and in the case of Arabidopsis much work31 had been done by 2007.
According to a recent edition of Current Opinion in Plant Biology,32 advances
in understanding at the level of functional genomics will result, in the case of
Arabidopsis, in:
An understanding of the networks through which these genes interact
to control plant development, metabolism, reproduction and other
fundamental processes will accelerate the advent of a new generation of
improved crop products to benefit growers, processors and consumers.
Bringing together knowledge of the function of genes and gene networks,
and of their regulation within the contexts of cell, organ, organism, and
environment will be crucial for achieving the level of precision in crop
engineering that will be required to fuel the development of the nextgeneration products.
Considerable progress has been made in regard to the re-annotation33
including both the structure and the functions of genesof the gene sequences
of Arabidopsis, and a similar re-annotation of rice is also under way.
The revolution in genomics signals a transition in plant biology from a
descriptive to a predictive science. As Dixon34 points out: Genomics (originally
DNA and transcript based, but recently extended to integrate the proteome and
metabolome) has revolutionized the speed of gene discovery for important plant
traits.
31 Lan, H., Carson, R., Provart, N. J. and Bonner, A. J. 2007, Combining classifiers to predict gene function
in Arabidopsis thaliana using large-scale gene expression measurements, BioMedCentral, <http://www.
biomedcentral.com/1471-2105/8/358>.
32 Salmeron, J. and Herrera-Estrella, L. 2006, Plant biotechnology: fast-forward genomics for improved
crop production, Current Opinion in Plant Biology, vol. 9, pp. 1779.
33 Rensink, W. A. and Buell, R. C. 2005, Microarray expression profiling resources for plant genomics,
Trends in Plant Science, vol. 10, no. 12 (December).
34Ibid.
76
Indeed a brief review of scientific and technological developments emerging

since the turn of the twenty-first century offers a glimpse into the broad range
of activities relating to a more sophisticated understanding of the function of
plant genomes.
A feature article35 published by the American Phytopathological Society
(APS) in 2000 described four significant areas of research and possible future
approaches involving genetic-engineering techniques that could confer plant
resistance to pathogen invasion. Research focused on enhancing resistance
with plant genes sought to identify the genes involved in defences against,
and resistance to, plant pathogens (resistance is against pathogens, rather than
diseases) to facilitate the conferring of disease resistance. Use of this approach
increased the plants ability to defend itself against pathogen invasion, and
plant biologists using recombinant DNA biotechnology could now adopt a
number of new strategies. As Fermin-Munoz36 demonstrated, the insertion of a
specific transgene conferred resistance not normally present in a host plant. The
insertion of transgenes also could trigger a plants intrinsic defence mechanism
against both pathogen invasion and abiotic stress. Also described were similar
techniques for the insertion of proteinaceous and non-proteinaceous compounds
having antibacterial and antifungal properties. One such example has led to
the discovery of a protein called harpin (produced by members of the plant
pathogenic bacterial genus Erwinia, which is sometimes thought of as a toxin
and sometimes as a defence chemical) that could be used prior to pathogen
invasion to activate crop defences.
Another approach37 was developed to protect plants from pathogen invasion
through pathogen-derived resistance. This strategy involved engineering
genes into plants, and important recent work has included the insertion of
viral transgenes as an alternative to the use of harmful pesticides against insect
vectors. Pioneering research has shown that both protein-mediated and RNAmediated pathogen-derived resistance can be conferred using viral transgenes,
with some success being achieved using a number of plant viruses38 affecting
alfalfa, cucumber, tobacco, tomato and potato.
35 Fermin-Muoz, G. A., Meng, B., Ko, K., Mazumdar-Leighton, S., Gubba, A. and Carroll, J. E. 2000,
Biotechnology: a new era for plant pathology and plant protection, APSnet Feature, May, <http://www.
apsnet.org/publications/apsnetfeatures/Pages/Biotechnology.aspx>.
36 Fermin-Munoz, G. A. 2000, Enhancing a plants resistance with genes from the plant kingdom, APSnet
Feature, May, <http://www.apsnet.org/publications/apsnetfeatures/Pages/EnhancingPlantResistance.aspx>.
37 Meng, B. and Gubba, A. 2000, Genetic engineering: a novel and powerful tool to combat plant
virus diseases, APSnet Feature, May, <http://www.apsnet.org/publications/apsnetfeatures/Pages/
GeneticEngineering.aspx>.
38Ibid.
77
Another approach39 for conferring plant disease resistance involves antimicrobial

peptides and proteins that confer antimicrobial properties, thus strengthening
immunity and resistance to fungal and bacterial plant pathogens. This review
noted that proteins potentially useful in improving plant disease resistance
could be found outside the plant kingdom in insects, animals, humans and
fungi. Antibacterial peptides/proteins and enzymes have been shown to
inhibit pathogen invasion in a variety of plants including some of considerable
socioeconomic significance. Furthermore, immunity or resistance to pathogen
invasion has been attempted at the molecular level by in planta expression of an
antibody against a protein necessary for pathogenesis.
The past decade has seen continued progress in the development of plant
disease-resistance mechanisms, and a wide array of new tools is being developed
to produce plants expressing a broader range of such traits. Interesting areas of
development were described in 2005 in Trends in Plant Science.40
In China, Wang et al.41 focused on improving rice, applying molecular markerassisted breeding, functional genomics and genetic modification techniques to
the identification of gene function in elite rice cultivars having important socioagronomic traits such as enhanced pest and stress resistance, good grain quality,
and high and stable yield potential.
In the area of molecular marker-assisted breeding, which is the application of
molecular biotechnologies to breeding, tools include marker-assisted selection
(MAS), quantitative trait locus (QTL) analysis, and genetic transformation
techniques. While plant scientists have enjoyed success in improving cultivars
with important traits using genes with known desirable traits, the adoption of
molecular marker-assisted breeding techniques may facilitate the identification
of desirable plant traits that result from the expression of multiple genes.
Wang et al. also delved into functional genomics,42 demonstrating how microarrays, reverse genetics and map-based cloning are being used for identifying
important characteristics in rice genes, including [e]xpression pattern,
chromosomal position, perceived biological function, and behaviour of alleles
under phenotypic selection.
39 Ko, K. 2000, Using antimicrobial proteins to enhance plant resistance in biotechnology: a new era
for plant pathology and plant protection, APSnet Feature, May, <http://www.apsnet.org/publications/
apsnetfeatures/Pages/AntimicrobialProteins.aspx>.
40 Dixon, R. A. 2005, Plant biotechnology kicks off into the 21st century, Trends in Plant Science, vol. 10,
no. 12; Neal-Stewart, C., jr, 2005, Plant functional genomics: beyond the parts list, Trends in Plant Science,
vol. 10, no. 12 (December).
41 Wang, Y., Xue, Y. and Li, J. 2005, Towards molecular breeding and improvement of rice in China, Trends
in Plant Science, vol. 10, no. 12 (December), pp. 61014.
42Ibid.
78
Although scientists have had some success in the development of transgenic

crops with resistance to plant disease and plant pests, a number of hurdles, not
least those relating to public health and the environment, have (in China) thus
far prevented the widespread commercialisation and human consumption of
rice varieties with improved traits.
In their article Microarray expression in profiling resources for plant genomics,43
Rensink and Buell discuss various approaches to facilitate the identification of
gene function and understanding of, inter alia, basic physiology, developmental
processes and environmental stress responses, using information derived from
micro-array platforms. They note, in particular, the significance for plant
researchers of micro-array-derived bio-information and the importance of the
worldwide availability, via the Internet, of bioinformatics data sets.
Likewise, in the article Genomics-assisted breeding for crop improvements,44
Varshney et al. note the importance of information on molecular markers
(functional markers), and the relevance of the rapidly advancing area of
bioinformatics, which is providing a means for the integration and structured
interrogation of data sets that will facilitate cross-fertilisation of disciplines in
the evolution of future genomics-assisted breeding.
A 2006 review in Current Opinion in Plant Biology45 provided a further glimpse
into research in functional genomics. For example, Bohnert et al. describe a
palette of tools that facilitates a more detailed understanding of the spectrum
of plant responses to developmental and environmental stimuli, including
tolerance to drought, soil salinity and cold stresses. According to Salmeron and
Herrera-Estrella,46 tools such as metabolite and protein profiling, subcellular
imaging, transcript clustering, comparative biology and reverse genetics reveal
a range of valuable genes, alleles and promoters. Research by Valliyodan and
Nguyen in the same edition shows understanding of abiotic stress tolerance,
specific gene components and transcriptional and cis-acting regulatory
elements important in possible future engineering of specific plant traits has
progressed significantly.
A review of work by Fernie et al.47 describes progress in the development of
genomics-based techniques that use molecular markers in the identification
of desirable plant trait alleles. According to Salmeron and Herrera-Estrella:48
43 Rensink and Buell, op. cit.

44 Varshney, R. K., Grner, A. and Sorrells, M. E. 2005, Genomics-assisted breeding for crop improvement,
Trends in Plant Science, vol. 10, no. 12 (December), pp. 62130.
45 Salmeron and Herrer-Estrella, op. cit., pp. 1779.
46Ibid.
47Ibid.
48Ibid.
79
Genomics will make possible a level of surgical precision in breeding that

allows such traits to be efficiently extracted by using molecular markers to tag
specific desirable alleles.
Alleles of importance for plant traits such as those related to nutrition and
yield have already been identified in some solanaceous and other important
crops. Also reviewed by van Schie et al. in the same edition is research on plant
fragrance mechanisms, and the possibility that it could be manipulated at the
genome level.
A further article on functional genomics centres on emerging techniques for
gene silencing using geminiviral vectors. Salmeron and Herrera-Estrella49
outline some possible applications as follows:
1) [T]hey can be used for functional genomics in plants species for
which the production of transgenic lines is difficult or time-consuming;
2) they provide the ability to work with genes whose knockout mutants
are lethal; 3) inoculation with Gemini viral vectors is rather simple and
phenotypes can be analysed a few days after the host has been infected;
and 4) they can be adapted easily for high-throughput genomic studies.
Malign applications
The above developments offer a glimpse of the rapid progress in plant biology
over the past 10 years. The genomics revolution opens up a range of new
possibilities for improvements in the quality and quantity of crop yields and
an increasing number of techniques and applications will become available for
combating disease that is caused naturally and accidentally; however, the same
developments also open up a range of possibilities for malign applications.
Analysts have for many years expressed concern regarding the ways in which
naturally occurring plant pathogens might be deployed for malign purposes.
This might involve the simple introduction into a crop species, for example, of
a pathogen to which no natural immunity exists. Van der Planks well-known
observations of the seemingly explosive spread of some plant pathogens in the
absence of immunity remain particularly salient in spite of great progress that
has been made in phytopathology over the past 40 years.
Scenarios may also include the introduction into crops of pathogens that have
mutated naturally in the environmentwitness, in this connection, the nearfuture possibility of the re-emergence in regions such as the Middle East and
in countries such as India of a new and virulent strain of wheat rust, Ug99
49Ibid.
80
(Ug99 has been found in Egypt at least and perhaps other countries in the
Middle East), and its predicted associated immense potential for social and
human destruction as it has been described by Borlaug50 in the New Scientist.
According to Mackenzie,51 to combat this mutation effectively, the production
of enough Ug99-resistant seed to plant our wheat fields might take up to eight
years.
As in the case of bio-control agent production discussed above, there is also a
possibility that crop pathogens could be genetically modified deliberately. The
result could include increased toxicity or pathogenicity. Or, as suggested by
Kelle et al.,52 plants innate immune systems are vulnerable to manipulation,
possibly affecting the response to pathogen invasion. For example, a plants
response could be manipulated so as to trigger systemic, rather than localised,
hypersensitive reactions to pathogen invasion. Nixdorff explains the mechanisms
involved in systemic plant resistance mechanisms thus: The main systemic
signals include salicylic acid, jasmonate and ethylene, which are produced
in response to wounding and insect attack. H2O2 is the most important
response mechanism to pathogen invasion
involved in downstream signalling, leading to the activation of signalling
cascades in Arabidopsis as well as activation of genes controlling the
production of proteins involved in HR [hypersensitive reaction].
[Therefore] plants may be attacked through their innate immune
systems, for example, by targeting either the receptors of signalling
cascades, or by inhibiting or producing an over-reaction in a signalling
cascade with the use of inhibitors of key components in that cascade.
Kagan et al.53 present concerns about the introduction of noxious DNA
material in the form of, for example, a bio-regulator into a bio-control agent
such as Bacillus thuringiensis in quantities sufficiently large to contaminate a
food product. Indeed, further to this, Chofness et al.54 also note in relation to
transgenic plants that such plants
could be malevolently engineered to produce large quantities of
bioregulators or toxic proteins, which could either be purified from
plant cells or used directly as biological agents. As with legitimate
production, using transgenic plants as bioreactors would eliminate the
50 Mackenzie, D. 2007, Rusting defences in the battle for wheat, New Scientist, 3 April.
51 Mackenzie, D. 2007, Billions at risk from wheat super-blight, New Scientist, 3 April, pp. 67.
52 Kelle et al., op. cit., p. 76.
53 Kagan, E. 2006, Bioregulators as prototypic nontraditional threat agents, Clinics in Laboratory Medicine,
vol. 26, no. 2 (June), pp. 42143.
54 Choffnes, E. R., Lemon, S. M. and Relman, D. A. 2006, A brave new world in the life sciences, the breadth
of biological threats is much broader than commonly thought and will continue to expand, Bulletin of Atomic
Scientists, vol. 62, no. 5 (SeptemberOctober), pp. 2633.
81
need for mechanical equipment normally associated with the process.

The technology would be limited to producing protein-based agents.
But because transgenic plants would be largely indistinguishable from
non-transgenic crops, biopharming could potentially provide a covert
means for producing large amounts of product.
Conclusion
This survey of offensive biological warfare agents, bio-control agents and plant
inoculants of relevance to the BWC, and recent trends in phytopathology
and plant technology, has alluded to a broad range of scientific discovery
and technological innovation. This highlights a number of areas where
considerationincluding ethical deliberationregarding the dual-use nature
of scientific discovery and technological application might usefully be applied.
Indeed, where deliberation might take place, who might be involved, and
what mechanisms and systems might be put in place to facilitate deliberative
processes have been the subjects of considerable attention in the United States.
Since 2006, the NSABB has been tasked,55 inter alia, with advancing thinking
around proposals for the development of recommendations for scientific oversight
measures and for recommending how such measures might be developed so as
to minimise the risk of misuse of scientific information. One of the challenges
inherent in pursuing this mandate is how oversight mechanisms might be
created that are both efficient and effective but constructed and implemented in
such a way as to mitigate the stifling of life-science innovation. This initiative
was viewed as a preliminary step towards establishing a mechanism of oversight
through the development and implementation of a comprehensive system for
the responsible identification, review, conduct, and communication of dual use
research.56 The NSABB proposed that a system of oversight might include the
following seven key features:
The development of Federal Guidelines for oversight of dual use life
science research
Enhanced levels of awareness of dual use research of concern amongst
practicing life scientists
Enhanced, ongoing, mandatory education that raises awareness of dual
use research of concern and addresses the roles and responsibilities of
life scientists
55 See NSABB, op. cit., Appendix A, p. 9.

56 Ibid., p. 7.
82
With appropriately trained Principal Investigators, a system of local

evaluation and review of research of dual use potential.
Risk Assessment and Risk Management
Mechanisms for Periodic Local and Federal Evaluation
Local and Federal Mechanisms for ensuring Compliance.57
In regard to a system of local evaluation and review of research of dual use
potential, the NSABB recommended that the initial evaluation of research
for its potential as dual use research of concern should fall to the principal
investigator (PI). Indeed, as further noted by NSABB, appropriate expertise
could thus be brought to bear in assessing the dual-use potential of scientific
research, but significantly NSABB also noted a requirement in this area for
appropriate training.58
As part of its recommendations, NSABB identified a series of oversight roles
and responsibilities and elaborated upon how both would relate to researchers,
research institutions, institutional review entities, ongoing independent review
and the Federal Government. In regard to the roles and responsibilities of
researchers, NSABB noted a further potential deficit in the area of awareness
and training. As it argued, PIs are the ones who will be best placed to discharge
their responsibility to be able to assess the kinds of knowledge generated, its
potential for misuse and how such threats might be mitigated. According to
NSABB,59 however, PIs would need both to be cognizant of the concept of dual
use research of concern and aware of the risk that technologies or information
produced by life sciences research may be misused.60 Indeed, NSABB61 places
considerable emphasis on improving levels of awareness in that it argues that an
enhanced culture of awareness is essential to an effective system of oversight
and is a critical step in scientists taking responsibility for the dual use potential
of their work.
This chapter argues that ethical discussion and deliberation, alongside
consideration of the potential social and legal ramifications of scientific research,62
could be usefully embedded in a broader system of scientific oversight such as
that envisaged by NSABB, and could have an important deliberative function
in assessing experimentation of dual-use concern in the area of plant-science
research. Indeed, opportunities for ethical and other types of deliberation could
57 Ibid., p. 8.
58 Ibid., p. 9.
59 Ibid., p. 11.
60 Ibid., p. 11.
61 Ibid., p. 9.
62 Ibid. See also Committee on Research Standards and Practices to Prevent the Destructive Application
of Biotechnology, Development, Security, and Cooperation, Policy and Global Affairs 2004, Biotechnology
in An Age of Terrorism, National Research Council, The National Academies Press, Washington, DC,
Recommendation 1: Educating the Scientific Community, p. 4.
83
be included at each respective step (PIs, institutional and independent review

entities, governments, and so on) in the research oversight process as identified
by NSABB.
Whilst the publication of the framework proposed by NSABB represented a
step-change in thinking about how to address the oversight challenges posed
by dual-use research of concern, as argued by NSABB,63 deliberative processes
will need to be informed by increased levels of awareness and education about
dual-use research of concern, including all applicable policies as well as the
provision of guidance and tools that facilitate compliance with the policies.
Much work needs to be done by ethicists, however, in respect of how such
review processes will be informed by ethical considerations. Indeed, awareness
raising, education and training courses will need to be developed that facilitate
ethical deliberation around dual-use research of concern but also that facilitate
deliberation and assessment of both the legal and the social implications of such
research. Such material must also be oriented to provide life scientists with
guidance and information that facilitate compliance with relevant guidelines,
policies and legislation.
Unfortunately, few accredited university courses or non-accredited short
courses currently exist that seek to engage life scientists in deliberation about
the ethical, legal and social implications of the scientific research they conduct.
One notable exception is a combined hybrid biosafety (bio-risk management)
and dual-use biosecurity training course64 that is being designed and developed
in a collaboration between the University of Bradford, UK, and the Public
Health Agency, Canada (PHAC). Following the implementation of Canadas
Human Pathogens and Toxins Act,65 which sets out a compliance requirement
that biosafety officers are trained and have requisite qualifications, this
initiative seeks to develop training material (with a Canadian focus) that engages
life scientists in deliberation about life-science research of potential dual-use
concern by addressing a range of dual-use issues of relevance to laboratory
biosafety and beyond the laboratory door. It is argued here that this combined
biosafety/dual-use biosecurity approach possibly represents a model of best
practice for the subsequent development of awareness raising, education and
training. The last could be incorporated into the training of plant scientists in
order that they are cognisant of dual-use research issues of concern and better
able to assess the ethical, legal and social implications of their work.
63 NSABB, op. cit., p. 15.

64 The dual-use biosecurity element of this initiative has evolved from an existing online distance-learning
course that focuses on applied dual-use biosecurity education that has been running at the University of
Bradford since September 2010.
65 Human Pathogens and Toxins Act, 2009, Canada, <http://lois-laws.justice.gc.ca/eng/acts/H-5.67/index.html>.
84
Finally, building on research that was previously set out in the influential
2004 Fink Report,66 NSABB also set out seven criteria for the identification of
endeavours and discoveries that might trigger discussion and review. The
following seven classes of experiments might provide a useful framework for
considering the types of endeavours or discoveries that, if proposed, might
trigger review by PIs including, where appropriate, review and discussion
by informed members of the scientific and medical community before they are
undertaken or, if carried out, before they are published in full detail.67
The experiments include those that:
1. Would demonstrate how to render a vaccine ineffective. This would
apply to both human and animal vaccines. Creation of a vaccine-resistant
smallpox virus would fall into this class of experiments.
2. Would confer resistance to therapeutically useful antibiotics or
antiviral agents. This would apply to therapeutic agents that are used
to control disease agents in humans, animals, or crops. Introduction of
ciprofloxacin resistance in Bacillus anthracis would fall in this class.
3. Would enhance the virulence of a pathogen or render a nonpathogen
virulent. This would apply to plant, animal, and human pathogens.
Introduction of cereolysin toxin gene into Bacillus anthracis would fall into
this class.
4. Would increase transmissibility of a pathogen. This would include
enhancing transmission within or between species. Altering vector
competence to enhance disease transmission would also fall into this class.
5. Would alter the host range of a pathogen. This would include making
nonzoonotics into zoonotics agents. Altering the tropism of viruses would
fit into this class.
6. Would enable the evasion of diagnostic/detection modalities. This could
include microencapsulation to avoid antibody-based detection and/or the
alternation of gene sequences to avoid detection by established molecular
methods.
7. Would enable the weaponization of a biological agent or toxin. [Emphasis
added]
Whilst a detailed analysis of NSABBs oversight system is beyond the purview
of this chapter, the above criteria could be used to trigger interventions
66 Committee on Research Standards and Practices to Prevent the Destructive Application of Biotechnology,
Development, Security, and Cooperation, Policy and Global Affairs, op. cit.
67 NSABB, op. cit., p. 18.
85
in plant-science research at the level of PI. A multilayered oversight system

could include a system of checks and balances so as to ensure that appropriate
review is applied, where appropriate to do so, at each stage of the research/
oversight process. Application of the criteria to developments in plant science
and technology (as discussed above) would seem to suggest that discussion and
review concerning the ethical, legal and social implications of such research
would be triggered in no less than five out of seven classes of experiments
identified in Fink and by NSABB (see emphases).
86
6. The Super TB Experiment:

Evolution and Resolution of an
Experiment with Dual-Use Concerns
Nancy Connell
Introduction
Tuberculosis (TB) is a devastating disease that leads to 10 million deaths per year
worldwide.1 Despite enormous advances in TB research in recent years, the surge
of drug-resistant strains and deadly synergy with the HIV virus have threatened
to destabilise gains made in its control. During my career as a microbial geneticist,
I have studied the physiology of Mycobacterium tuberculosis, the causative agent
of TB, and its interaction with the human macrophage, its primary host cell. The
major function of the macrophage is to engulf and destroy invading organisms,
but many pathogens, such as TB, have developed intricate and clever ways to
avoid or subvert these host defences.
This chapter discusses how a classical approach to asking a very basic question
about microbial pathogenesis can lead to a surprising result and a dual-use
dilemma. The elimination of a single gene from M. tuberculosis created a strain
with increased virulence when measured in a model system (in vitro) but not
during animal infection (in vivo).
Experiments of concern
The concept of dual-use researchthat certain kinds of experimental endeavours
might be used for malignant purposes despite their original beneficent intent
has entered the biomedical lexicon.2 My personal interest in the matter grew out
1<http://www.who.int/tb/publications/global_report/2010/en/index.html>.
2 National Research Council and the American Association for the Advancement of Science (NRC/AAAS)
2009, A Survey of Attitudes and Actions on Dual Use Research in the Life Sciences: A Collaborative Effort
of the National Research Council and the American Association for the Advancement of Science, Washington,
DC; American Association for the Advancement of Science (AAAS) 2009, Building the Biodefense Policy
Workforce, American Association for the Advancement of Science, Washington, DC; American Association
for the Advancement of Science (AAAS) 2008, Professional and Graduate-Level Programs on Dual Use Research
and Biosecurity for Scientists Working in the Biological Sciences, American Association for the Advancement
of Science, Washington, DC; Atlas, R. and Dando, M. 2006, The dual-use dilemma for the life sciences:
perspectives, conundrums, and global solutions, Biosecurity and Bioterrorism, vol. 4, no. 3, pp. 111.
87
of a longstanding interest in biological weapons and arms control beginning in

the 1980s. At the time, many scientists were contributing to the development
of a Verification Protocol for the Biological and Toxin Weapons Convention
(BTWC).3 Part of that commitment included the responsibility of scientists to
be aware of the societal impact of their work and to communicate that impact
to the public.
In response to concerns about dual-use research, a number of advocacy groups
started up at this time such as Science for the People and the Committee
for Responsible Genetics (now the Council for Responsible Genetics).
Simultaneously, scientists and policy analysts realised that the potential for
harm caused by naturally occurring organisms and toxins could be easily
surpassed by the creative development of recombinant agents expressing altered
or novel virulence characteristics. Commonly available genetic technologies in
laboratories improved, allowing for a deeper understanding and dissection of
the mechanisms of infection and virulence.
As the Verification Protocol fell by the wayside in the autumn of 2001, one month
before the anthrax mailings in the United States, interest in strengthening the
BTWC took on new directions. One of those was an interest in how knowledge
and techniques being generated as part of peaceful work in the life sciences could
facilitate the deliberate spread of diseasewhat is now known as dual use. A
list of the experiments considered to be of dual-use concern was published in
several reports shortly after the events of 2001. This included the landmark
Fink Report.4 These lists included several approaches to creating novel mutants
of existing pathogens, such as: 1) developing antibiotic-resistant strains, 2)
expanding host range, 3) increasing transmissibility, and 4) altering virulence.
To understand a biological system, one of the first approaches is to perturb the
system and observe the effects of that manipulation. In a genetic analysis, the
approach is to alter or even eliminate the function of a gene and observe the
effects of that alteration on the biological process. By integrating information
from a number of different alterations of a single gene, its role or function in a
system or pathway can be deduced. In the case of disease-causing organisms,
each of the experiments of concern, listed above, fits into this classic approach
to understanding biological function
antibiotic-resistant strains are selected and characterised to understand the
mechanism of action of an antibiotic, and the identification of the target will
allow development of more effective antibiotics directed to that target
3 Zilinskas, R. A. 1998, Verifying compliance to the biological and toxin weapons convention, Critical
Reviews in Microbiology, vol. 24, pp. 195218.
4 National Research Council 2004, Biotechnology Research in An Age of Terrorism, The National Academies
Press, Washington, DC.
88
6. The Super TB Experiment: Evolution and Resolution of an Experiment with Dual-Use Concerns
host range mutants are developed to understand the mechanism of host range
restriction and to develop methods of blocking hostpathogen interaction,
the first step of infection
transmissibility mutants can be created to understand the mechanisms that
control transmissibility, to develop methods of reducing transmissibility
altering virulence can be used to identify the genes encoding virulence
factors, and interfering with these factors that may limit infection and
pathogenesis of disease.
Therefore, any comprehensive genetic analysis of virulence will eventually be
confronted with the possibility of an experiment of concern. The question for
researchers is whether to continue to pursue the line of inquiry or to find an
alternative approach. I now want to consider these issues in relation to my own
work.
The experiment: Nutrient transport mutants of

Mycobacterium tuberculosis
Microbes use small molecules such as sugars and amino acids to survive, and
pathogens that live in their hosts cells have a multitude of ways to acquire
these nutrients from their environment. In the case of M. tuberculosis, their
environment is the inside of mammalian white blood cells, where they replicate
quite happily; indeed, the bacteriums method of pathogenesis is to reside
within the very cells designed to protect the host against them!5
Our experimental approach was to create mutants of M. tuberculosis that were
defective in uptake of a certain class of small nutrients called peptides. These
are short (26) chains of amino acids that are transported by efficient transport
systems called ABC transporters.6 We were able to construct mutants of M.
tuberculosis lacking the genes that encode one of these transporters.
Simultaneously, we also discovered that a small tri-peptide called glutathione
had the unusual and unexpected property of being toxic to M. tuberculosis.7
This meant that in the presence of this small peptide, M. tuberculosis could not
grow. The concentration of glutathione in mammalian cells varies considerably
but is usually quite high; under some conditions, the concentration is much
higher than the toxic levels that kill M. tuberculosis. We wondered how this
5 Pieters, J. 2008, Mycobacterium tuberculosis and the macrophage: maintaining a balance, Cell Host and
Microbe, vol. 3, pp. 399407.
6 Detmers, F. J. M., Lanfermeijer, F. C. and Poolman, B. 2001, Peptides and ATP binding cassette peptide
transporters, Research in Microbiology, vol. 152, pp. 24558.
7 Green, R. M., Seth, A. and Connell, N. D. 2000, A peptide permease mutant of Mycobacterium bovis BCG
resistant to the toxic peptides glutathione and S-nitrosoglutathione, Infection and Immunity, vol. 68, pp. 42936.
89
situation could have developed: that a bacterium that spends its entire life cycle
in mammalian cells does so in the presence of a small molecule, glutathione, that
is toxic! We characterised this toxicity, and showed that in mouse-derived cells,
when the glutathione levels were high, M. tuberculosis did not grow well, but
when the glutathione levels were low, the bacteria grew very efficiently.8
We were, however, now at a crossroads. First, we showed that glutathione is toxic
to M. tuberculosis, and we found glutathione at toxic levels in all mammalian
cells. Second, we had created a mutant of M. tuberculosis in the laboratory that
was lacking a peptide transporter and incapable of accumulating small peptides,
including glutathione. Therefore, this mutant should have been resistant to the
toxic effects of glutathione. We showed that this was indeed the case.
But if glutathione is toxic to M. tuberculosis and present in all mammalian cells
then it might play a natural role in limiting M. tuberculosis growth. This was our
hypothesis. The logical next step was to see whether our mutant strain survived
better than its normal parent strain during intracellular growth in white blood
cells in tissue culture. It did. In fact, the mutant strain grew to three to five times
higher levels than normal parent strains when inoculated and grown in cultured
macrophage cell lines. Our experimental approach following a logical series
of steps had led us to create a hyper-virulent mutant of M. tuberculosis. We
had generated a strain that survived in host cells better than its normal parent
strain, as a result of the deletion of a single gene encoding one component of
a basic nutrient transport system. Interestingly, the dual-use implications of
our work were brought to my attention only during a discussion of that issue
with members of the Controlling Dangerous Pathogens Project at the Center for
International and Security Studies at Maryland (Elisa Harris and colleagues).
Apparently, I had partitioned my thinking about these matters, and separated
the experiment itself from the possible ethical concerns associated with it,
despite my own decades-long involvement with the issue of biological weapons
and the responsibility of scientists in preventing a biological arms race.
Virulence factors
Defining virulence is not a simple matter. By a general and superficial definition,
virulence is the degree of pathogenicity of a biological agent, and pathogenicity
8 Venketaraman, V., Dayaram, Y. K., Amin, A. G., Ngo, R., Green, R. M., Talaue, M. T., Mann, J. and Connell,
N. D. 2003, Role of glutathione in macrophage control of mycobacteria, Infection and Immunity, vol. 71,
pp. 186471; Venketaraman, V., Dayaram,Y. K., Talaue, M. T. and Connell, N. D. 2005, Glutathione and
nitrosoglutathione in macrophage defense against Mycobacterium tuberculosis, Infection and Immunity, vol.
73, pp. 18869.
90
is the ability of an organism to cause disease. Both terms are relative and highly
dependent on context. Furthermore, virulence and pathogenicity are frequently
used interchangeably.
Myriad factors contribute to the virulence of an infectious agent. Virulence
factors that cause severe damage to the host are called toxins. Some toxins
act directly on the host cell such as listeriolysin, which punches holes in the
membranes of host cells and breaks them open.9 Many viruses reproduce to such
high numbers in cellsgrabbing bits of the host cell membranes as they go
that the cells just break apart.10 Other toxins act directly on host cells to alter
their function severely. For example, one component of anthrax toxin destroys
the proteins that orchestrate the immune response and cause swelling in the
tissues surrounding the site of infection, while another component inactivates
a small protein, called G-protein, which regulates the movement of ions across
membranes; its inactivation leads to severe oedema.11 Other toxins block protein
synthesis, so the cell can no longer carry out normal functions (diphtheria
toxin),12 or interfere with neurotransmitter activities, as with the clostridial
toxins, botulinum toxin and tetanus.13 In some, but not all, cases elimination
of the gene encoding the toxin will render the bacterium completely harmless
(avirulent); this is true with Clostridium botulinum, the causative agent of
botulism. Conversely, expression of the gene encoding the botulinum toxin in
an alternative bacterial host would lead to the creation of a very dangerous
agent. Ken Alibek, a former Soviet bioweapons researcher, claimed that the
Soviet offensive program devoted considerable effort to splicing the botulinum
toxin genes into other bacteria for delivery.14
But in the case of Yersinia pestis (plague),15 for example, and Burkholderia mallei
(glanders)16both select agentsthere are many small, secreted virulence
factors that work together to damage the host; elimination of one or two of the
genes encoding the complex array of effectors will not lead to inhibition of
virulence, rather, just defects in the ability to cause disease (hypo-virulent,
9 Schnupf, P. and Portnoy, D. A. 2007, Listeriolysin O: a phagosome-specific lysin, Microbes and Infection,
vol. 9, pp. 117687.
10 Kaminskyy, V. and Zhivotovsky, B. 2010, To kill or be killed: how viruses interact with the cell death
machinery, Journal of Internal Medicine, vol. 267, pp. 47382.
11 Moayeri, M. and Leppla S. H. 2009, Cellular and systemic effects of anthrax lethal toxin and edema
toxin, Molecular Aspects of Medicine, vol. 30, pp. 43955.
12 Iglewski, W. J. 1994, Cellular ADP-ribosylation of elongation factor 2, Molecular and Cellular
Biochemistry, vol. 138, pp. 1313.
13 Popoff, M. R. and Bouvet, P. 2009, Clostridial toxins, Future Microbiology, vol. 4, pp. 102164.
14 Alibek, K. and Handleman, S. 1999, Biohazard, Random House, New York.
15 Viboud, G. I. and Bliska, J. B. 2005, Yersinia outer proteins: role in modulation of host cell signaling
responses and pathogenesis, Annual Review of Microbiology, vol. 59, p. 689.
16 Sun, G. W. and Gan, Y. H. 2010, Unraveling type III secretion systems in the highly versatile Burkholderia
pseudomallei, Trends in Microbiology, vol. 18, pp. 5618.
91
or attenuated). By the same token, expression of these individual effector

molecules in other bacteria would not likely lead to creation of a more virulent
organism, without the context of the entire genome for full virulence.
The many faces of virulence?

The experimental models used to identify and characterise these virulence factors
include bacterial culture conditions, infection of mammalian cells in culture and
small animal/non-human primates.17 Each of these approaches has its limitations
and can affect the way in which overall virulence is expressed. Indeed, the
terms hypo-virulence, hyper-virulence, avirulence and antivirulence18 are all
common in the medical literature. Yet the use of each of these terms must be
considered in its context.
Returning to the experiments described above, we had identified a mutant
strain of M. tuberculosis that exhibited hyper-virulent characteristics only in
the context of the cell culture modelthat is, mammalian cells grown in petri
dishes in incubators. We were at another crossroads: should we go ahead and
perform an experiment in an animal infection model of TB? Our laboratory was
under some pressure from funding agencies to demonstrate that glutathione
resistance was relevant in the larger scale model of mouse infection, which is
more closely relevant to human disease than the tissue culture model we used
for our more preliminary experiments.
We decided that the experiment was worth performing. We felt that if we could
show the relevance of our observation to TB disease in the whole animal, we
could convince the reviewers of our grant proposals that manipulation of the
glutathione system might be exploited to limit TB disease. Fully mindful of the
safety risks, we consulted our Institutional Biosafety and Institutional Animal
Use and Care committees, and we performed a risk analysis that weighed the
risks and benefits of the experiment. We concluded that knowing the answer
to this question was important for the projects direction and that our biosafety
plans were sufficient to carry out the experiment safely. Among our discussion
points was the observation that there is a high degree of variability among
clinical strains of M. tuberculosis, and our new strain was well within the range
of virulence observed in cell culture models. We were not dealing with a strain
of unusually extreme virulence. So we designed and executed a mouse infection
experiment under biosafety level three (BSL3) containment conditions with
17 OToole, R. 2010, Experimental models to study human tuberculosis, Advances in Applied Microbiology,
vol. 71, pp. 7589.
18 Casadevall, A. and Pirofski, L. 2002, Anti-virulence genesfurther muddling the lexicon? Trends in
Microbiology, vol. 111, pp. 41314.
92
appropriate safety and security measures. We found that there was no difference
in the survival or replication rate of the mutant compared with the normal
M. tuberculosis strain (unpublished observations). Thus, the hyper-virulent
behaviour we observed in the tissue culture model was not carried through in
an experimental system closer to the human disease state.
Our increased awareness of the dual-use implications of altering virulence
significantly impacted on our deliberations concerning additional testing of our
hyper-virulent mutant of M. tuberculosis. On the one hand, we were under
considerable pressure from our grant reviewers to demonstrate hyper-virulence
in the mouse model. If we did not demonstrate hyper-virulence then they might
not consider funding out project. On the other hand, our deliberations in
the laboratory about the nature of virulence and pathogenicity increased our
appreciation of the complexity of infectious potential and the inherent dual-use
nature of our research. These are the kinds of careful discussions that we would
urge scientists to have at many stages of their work.
93
Part II: Ethical Frameworks and

Principles
7. Moral Development and Ethical

Decision-Making
Judi Sture
A variety of approaches aimed at mitigating the intermittent friction between
science and society and the risks of malign use of modern scientific advances has
been defined: ethics,1 the responsible conduct of science,2 self-governance by
scientists,3 and top-down initiatives from policymakers and other authorities.4
These approaches have allowed for a number of perspectives on the challenges
that biosecurity poses to society and biotechnology today, but none has
considered the pre-existing values of stakeholders as they interact with these
challenges.
I would suggest that scrutinising the cultural pressures, forces and processes
that influence the development of the values that individuals bring to their
scientific work and policy views will shed further light on possibilities for
engagement with dual use. The lack of attention to this so far is possibly due to
the apparent lack of recognition of the role played by what may be called moral
development in ethical decision-making processes. We need to ask whether
the private and the public value sets held by individuals clash; is it possible to
hold multiple value sets in our private and professional lives and still manage to
operate effectively in one or both spheres? These are some of the questions that
need attention in approaching the potential challenges involved in deciding
what is good, bad, right or wrong in the context of scientific work and its effects
both inside and beyond the laboratory door.
The processes by which we learn and adopt moral stances have been recognised
and studied for more than 100 years. An influential work on what he termed
folkways5 by W. G. Sumner, professor of political and social science at Yale
University in the late nineteenth and early twentieth centuries, described and
interpreted the development of both individual habits and group customs as
they develop into morality-mediated usage in specific groups. He believed that
moral values and their associated behaviours achieve the status of unwritten
1 Selgelid, M. J. 2010, Ethics engagement of the dual use dilemma: progress and potential, in B. Rappert
(ed.), Education and Ethics in the Life Sciences: Strengthening the Prohibition of Biological Weapons, ANU E
Press, Canberra, pp. 2334, <http://epress.anu.edu.au/education_ethics/pdf/ch01.pdf>.
2 National Research Council/Institute of Medicine 2006, Globalization, Biosecurity and the Future of the Life
Science, The National Academies Press, Washington, DC.
3 Royal Netherlands Academy of Arts and Sciences 2008, A Code of Conduct for Biosecurity: Report by the
Biosecurity Working Group, KNAW, Amsterdam.
4 For example, Friedman, D. 2010, Israel, in Rappert, op. cit., pp. 2334.
5 Sumner, W. G. 1906, Folkways: A Study of the Sociological Importance of Usages, Manners, Customs, Mores,
and Morals, Ginn & Co., Boston, <www.gutenberg.org>.
97
laws for the individual and the group because popular usages and traditions,
when they include a judgment that they are conducive to societal welfare
exert a coercion on the individual to conform to them, although they are not
coordinated by any authority.6 I would argue that this is a relatively accurate
description of the position of most scientists and science stakeholders today in
the early years of the twenty-first century.
It is important that recognition of the role of pre-existing cultural folkways
(national/ethnic, economic, religious, social and technological values and so on)
and within these, domestic (familial) folkways or value sets, is factored in to any
consideration of how individuals reach a state of moral and ethical maturity in
their private and professional lives. This is particularly important in science,
where the rapid speed of advance typically outstrips the development of
appropriate ethical responses needed to deal with it and in which it is typically
supposed that all practitioners share a common set of professional values and
perspectives by virtue of their identity as scientists.
This chapter looks at the issues of moral development and ethical decisionmaking in terms of the publicly stated group value sets and individual, privately
held value sets. Later in the chapter, I describe the case of a conflicted scientist
as he tries to make sense of a number of professional/personal value conflicts
that face him in his daily work life. This unfortunate individual faces a set of
daily challenges that create an uncomfortable dissonance in him; he needs to
find a way in which to move forward without disadvantaging himself. Such
ways forward cannot be identified unless we have a better understanding of
the origins of the dissonance. By focusing on biological, psychological and
anthropological research into the development of value systems, within the
context of cultural identity, it is possible to highlight possible mechanisms that
may usefully be explored in further research.
Terminology
Many authors use the term morals to refer to ethical practice, standards or
beliefs; indeed it is common to find that the words morals and ethics are used
interchangeably. The very fact that there is this lack of clarity and consensus
in this vocabulary indicates a lack of understanding of the potential differences
between private and public values held by both individuals and groups. Because
this chapter focuses on the potential variance between private and public value
sets, I have chosen to be specific in the way I use these terms. For the purposes
of this chapter, the term morals will be used to refer to the attitudes and values
6 Ibid., p. 4.
98
7. Moral Development and Ethical Decision-Making
learned in childhood, pre-professional and private life, and the term ethics
to refer to the attitudes and standards learned and required in the context of
professional life.
The word moral(s) tends to be heavily laden with culture-specific values and
especially religious meaning, both of which are highly influential in forming
our early values and views. These values may belong to a value set separate
to that held professionally (although religious values may also direct ethical
values). Furthermore, many people think of morals as the attitudes they
learned from their personal and family lives, with ethics being something
outside the domestic sphere: a morality for the private sphere that is personal
and subjective, and a morality for the public sphere that is impersonal and
objective.7
The English word moresmeaning values, norms and, occasionally, virtues
is derived from the Latin plural noun mores. Mores refers to the traditions and
customs practised by specific groups, which relate directly to the subject of this
chapter. These traditions and customs are derived from accepted behaviours
and practices in a group rather than from actual laws, and consist of a shared
understanding of the manners and conduct that are acceptable to and expected
by the group. In short, either one conforms to the traditions and customs of the
group or one cannot function effectively as a full member. Indeed, an individual
whose values clash with the predominant values of the group may find that s/
he is not welcome in it at all.
Folkways, according to Sumner, are habits of the individual and customs of
the society which arise from efforts to satisfy needs. These folkways can be
understood as informal mores and, while not generally enforced in any way, they
tend to be perpetuated by members of the group emulating the behaviours and
attitudes of older or more established members. Once folkways are established,
they become, essentially, a value set that is shared by members of the group
and is maintained by group consent rather than by formal governance. Clearly
this process has implications in the biosecurity context when we are seeking to
develop, in effect, a new norm of biosecurity to which all scientists and science
stakeholders can subscribe.
In this chapter, the term culture refers to the culture of origin of the individual
(as in the ethnic sense), as well as to nationality and to religious, economic,
technological and linguistic groupings to which the individual may belong. All
of these dimensions impact on one another, having differing degrees of influence
depending on the context in which the individual is acting at any one time. One
of the outputs of such cultural influence is a value set, or group of value sets.
7 Catchpoole, V. M. 2001, A sociobiological, psychosocial and sociocultural approach to ethics education, PhD,
Centre for the Study of Ethics, School of Humanities, Queensland University of Technology, Brisbane, p. 274.
99
These may also be defined as beliefs, attitudes and perspectives on the world.
Clearly, these in turn impact on actions and behaviour patterns. Clashes occur
when an individual is faced with some conflict between the influences of one
or more cultural value sets in a given situation. Which value set should have
precedence?
Ethics may be defined as (potentially) universal standards and commitments
that support justice, care, welfare and rights;8 among other definitions, they may
also be defined as a principled sensitivity to the rights of others.9 The irony
here is, of course, that most people tend to believe that their own moral (private
and/or family derived) standards are also defined in the same way. Variations in
these have a role in defining scientists and policymakers ethical standards and
norms and must be recognised as key variables in any debate about what needs
to be done to prevent the hostile use of the life sciences. Nevertheless, bearing
in mind that cultural variations exist, it is probably reasonable to agree that
any desired socially responsible ethical approach involves the development
of adequate commitments that inform actions10 and that this includes an
understanding of what is beyond self-interest, and how we may justify that to
reasonable people and put it into practice.11 Even so, these aspirations, while
striving to be comprehensive and culture neutral, still incorporate approaches
that may not be prioritised by all cultures and therein lies part of the difficulty
in seeking a common set of values or norms. Everyone thinks that their existing
moral outlook is commonsense, and that their professional ethical values are
also commonsense. It is in this balancing act between the variation in moral
and ethical values as defined by diverse cultural backgrounds that we need to
find a way forward that allows individuals and groups to engage equally and
openly in the context of biosecurity as it challenges the needs of science in the
way it meets the needs of society.
Private morals and public ethics: Cultural

aspects of moral development
Given that cultural origins vary significantly, while providing a fundamental
platform of values from which each individual views the world, it follows
that everyone has at least one value set that influences the ways in which they
see, understand (or believe they do) and treat other people. Most people have
multiple value sets, and apply them variously in the different contexts of their
livesfor example, we may adhere to one value set in the family or domestic
100
8 Laupa, M. and Turiel, E. 1995, Social domain theory, in W. M. Kurtines and J. L. Gewirtz (eds), Moral
Development: An Introduction, Allyn & Bacon, Boston, pp. 45573.
9 Gilbert, N. 2001, Researching Social Life, Sage, London, p. 45.
10 Catchpoole, op. cit., p. 7.
11 Singer, P. 1993, How Are We to Live? Ethics in An Age of Self-Interest, Text, Melbourne.
sphere, but another in the work context; we may employ yet more value sets in
our relations with people we do not like, or with people we need, such as our
bank manager or our doctor.
How we learn about morality: Mechanisms of

development
This section introduces two processes of moral learning, which may be applied
in understanding the development of ethical thinking in the context of
professional science.
The development of a moral or ethical perspective presupposes the biological
capacity to do so. Growing up, we process moral learning as part of our social
heritage through our ability to symbolically represent the moral to ourselves,
which involves a relational aspect between ourselves and others:
[C]ulture serves to complement the process [of learning morality] in
important ways. Rather, we are led to moral experience and insight. Real
morality cant be forced on people, nor can they be fooled into having
it, nor do they just act on their moral instincts. Real morality does not
simply bubble up from beneath, nor is it imposed from the outside. In
each one of us, it must be discovered anew. The discovery process may
require great mental and emotional effort and may bloom only in the
right climate, but human beings see morality, recognize it, regardless of
what it is that they want or need or love or hate or feel compelled to do.12
We learn what is good and what is bad primarily in relation to ourselves as
individuals, but also, through social development, what is good and bad for
the others. These others will initially be the close others of family and
friends. Later, this can develop also into a sense of shared experience, and care
or empathy with the distant other: the outsider or even the enemy, the one who
is not a close other. Perhaps this conscious decision to have empathy for the
enemy or the outsider, the different one, is a measure of our ethical maturity.
Edward Wilson13 suggested that genetics and evolutionary processes shape the
human mind at least as much as culture, that a predisposition to religious belief
is genetically driven in humans and that acting altruistically through religious
belief can confer survival advantages to those willing to change their views.
Wilson also described the dark side of our predisposition to tribal affiliations
xenophobiaand suggested there is a need to globalise the tribe.14 In terms of
the debate on dual use and biosecurity, it may be useful to consider the relative
12 Goodenough, U. and Deacon, T. W. 2004, From biology to consciousness to morality, Tradition &
Discovery: The Polanyi Society Periodical, vol. 30, part 3, pp. 620.
13 Wilson, E. O. 1978, On Human Nature, Harvard University Press, Cambridge, Mass.
14 Wilson, E. O. 1998, Consilience: The Unity of Knowledge, Alfred A. Knopf, New York, p. 269.
101
positions of those defending the practice of research from security-inspired

interventions and those urging a change in attitude to the nature of scientific
freedom. These may be considered as belonging, at least in the eyes of some, to
different tribes and thus be subject to intertribal conflict.
In order to consider any change of allegiance (in our context, in terms of
changing what s/he thinks is ethical), an individual or group must be able to
reflect on and evaluate a range of cultural factors that pertain to the situation,
and then make an informed decision as to whether change of allegiance is the
right choice. The change may incur a short or a long-term cost, but the ability
to consider the change and the costs is part of a moral capacity, leading to the
application of moral agency. Essentially, the individual is faced with the choice
of allegiance or reprioritisation of values.
Griffiths15 considered that emotions play a key part in our moral response system,
and therefore in our capacity for moral agency, while Izard and colleagues
suggested that emotions have an adaptive role in equipping individuals to act
appropriately in order to survive.16 We can already see this in action in our
prioritisation and negotiation of values between the private morals and public
ethics in any given situation.
Wilsons views have been criticised by the evolutionary biologists Stephen Jay
Gould and Richard Lewontin, who oppose ideas of genetic determinism, and
the anthropologist Marshall Sahlins, who holds to the view that cultural factors
are a powerful driver of peoples behaviours, distinct from genetic influences.
Other researchers, however, have also supported a biological mechanism in the
development of values. The evolutionary biologist Robert Trivers17 suggested
that reciprocal altruism (all parties helping each other in order to survive) is
an evolutionary predisposition that offers the best chance of survival of the
individual and the species. Moore,18 however, has pointed out that it is not
always easy to distinguish between altruism and selfish acts disguised as
altruism. Richard Dawkins19 suggests that some biological determinants are
non-altruistic and reflect a predisposition to look after oneself, rather than
considering possible self-sacrifice for the good of others. Williamson20 argues
that the human concepts of shame and guilt function in evolutionary terms
as restraints on social behaviour, and points out that human self-interest is an
15 Griffiths, P. E. 1998, What the Emotions Really Are: The Problem of Psychological Categories, University
of Chicago Press, Chicago.
16 Izard, C. E. 1984, Emotioncognition relationships and human development, in C. Izard, J. Kagan and R.
B. Zajonc (eds), Emotion, Cognition and Behavior, Cambridge University Press, Cambridge, pp. 1737.
17 Trivers, R. 1971, The evolution of reciprocal altruism, Quarterly Review of Biology, vol. 46, pp. 3557.
18 Moore, J. 1984, The evolution of reciprocal sharing, Ethology and Sociobiology, vol. 5, pp. 514.
19 Dawkins, R. 1976, The Selfish Gene, Oxford University Press, Oxford.
20 Williamson, D. 1998, Mixed feelings, The Australian Review of Books, June, pp. 1415, cited in
Catchpoole, op. cit., p. 60.
102
obvious factor in survival: those of our ancestors who looked after their own
interests survived to reproduce themselves, and those who did not, failed to do
so.
Work by Krebs and Janicki21 indicated that the evolutionary development
of norms, or memes as Dawkins may refer to them (which may be defined as
culturally transmitted units of data such as ideas, habits and so on), has resulted
in four identifiable ways in which people operate ethically
1. they try to get others to invoke the norms that they themselves hold
2. they try to tailor their norms to others so as to enhance their persuasive
impact
3. recipients tend to adapt to the norms that are most advantageous to them
4. people in different sorts of relationships will preach different sorts of norms.
These operating mechanisms are, according to Krebs and Janicki, biologically
determined in that they are evolutionary adaptations to ensure survival.
Whether these mechanisms are truly biological or are solely culturally
determined artefacts is open to debate.
The development of reciprocity, empathy and other relationship-related
characteristics may be considered as part of a psycho-cultural mechanism of
learning morality and ethics. Our capacity to consider the needs of others is
enhanced by our cultural interactions with other people at a psychological level.
Most people learn how to make and sustain friendships, work relationships
and marriage relationships based on cultural concepts such as respect,
interdependence, empathy, cooperation and negotiation, amongst others. All
of these are partial foundations of a moral and ethical perspective, even though
they vary in how they are prioritised and expressed culturally.
Research suggests that humans can exhibit empathy (the capacity to
intellectually understand how another person feels) from an early age, which is
potentially open to development by other cultural factors. Work by a number
of psychologists has shown that even very young children have the capacity
for empathy.22 This runs contrary to earlier thought by Piaget,23 who believed
that very young children were essentially egocentric and unable to perceive
the world from the perspective of others. Hoffman devised a set of levels in the
development of empathy encompassing the first 12 years of life.24
21 Krebs, D. L. and Janicki, M. 2004, Biological foundations of moral norms, in M. Schaller and C. Crandall
(eds), Psychological Foundations of Culture, Lawrence Erlbaum Associates, Mahwah, NJ, pp. 12548.
22 For example, Hoffman, M. L. 1988, Moral development, in M. H. Bornstein and M. E. Lamb (eds),
Social, Emotional and Personality Development. Part III of Developmental Psychology: An Advanced Textbook,
Lawrence Erlbaum Associates, Hove, UK, pp. 497548; Zahn-Waxler, C., Radke-Yarrow, M. and Wagner, E.
1992, Development of concern for others, Developmental Psychology, vol. 28, part 1, pp. 12636.
23 Piaget, J. 1932, The Moral Judgment of the Child, Kegan Paul, London.
24 Hoffman, M. L. 1988, Moral development, in Bornstein and Lamb, op. cit.
103
A major player in the development of moral theory learning was Kohlberg.25

He developed a six-stage model of moral development, with successive stages
being more able to affect moral reasoning. His model focuses particularly on the
concept of justice and he believed that this development continues throughout
adult life. He further developed his model to accommodate soft developmentallevel changes in adults over the age of about thirty years, and acknowledged
that his stages do not provide a complete description of adult development from
stage four onwards26 (see Table 7.1). It must be noted, however, that Kohlbergs
stages refer not to specific beliefs but to underlying modes of reasoning,27 and
as such, there is no guarantee that an individual will reach the highest level
of reasoning. Also, it should be noted that Kohlbergs work focuses on moral
thought, not moral action, and we can probably all accept that the two do not
always go together.
Table 7.1 Kohlbergs Six-Stage Model of Moral Reasoning
Level 1:
Preconventional
morality
Stage 1: Punishmentobedience orientation (avoiding punishment for doing

wrong, but only because it is good for you)
Level 2:
Conventional
morality
Stage 3: Interpersonal concordance (being good to please other people

because that is the right thing to do)
Level 3:
Postconventional
morality
Stage 5: Social contract orientation (varying personal values are

recognised but are mediated by social agreement and a democratic
approach in society as a whole; the law is respected but is subject to
change if it is not just)
Stage 2: Instrumental relativist orientation (how will this decision/action be

good for me and my needs?)
Stage 4: Law and order orientation (abiding by laws for the good of
society, not just for the resulting benefits to yourself)
Stage 6: Universal ethical principle orientation (right decisions/actions

are defined by ethical principles that are universal and consistent, which
emphasise justice, reciprocity, rights and respect for human dignity in
society)
Source: Kohlberg, L., Levine, C. and Hewer, A. 1983, Moral Stages: A Current Formulation and A Response
to Critics, Karger, Basel.
Kohlbergs model also appears to provide evidence of the ways in which

people from different cultures make decisions,28 with different cultures passing
through the stages of the model differently, usually in the same order, but not
25 Kohlberg, L. 1971, Stages of moral development as a basis for moral education, in C. M. Beck, B. S.
Crittenden and E. V. Sullivan (eds), Moral Education: Interdisciplinary Approaches, University of Toronto Press,
Toronto, pp. 2392; and Kohlberg, L. 1981, The Philosophy of Moral Development: Moral Stages and the Idea
of Justice, Harper & Row, San Francisco; and Kohlberg, L., Levine, C. and Hewer, A. 1983, Moral Stages: A
Current Formulation and A Response to Critics, Karger, Basel.
26 Kohlberg et al., op. cit., p. 6.
27 Kohlberg, L. and Gilligan, C. 1971, The adolescent as a philosopher: the discovery of the self in a
postconventional world, Daedalus, p. 1051 ff.
28 Harkness, S., Pope Edwards, C. and Super, C. M. 1981, Social roles and moral reasoning: a case study in
a rural African community, Developmental Psychology, vol. 17, part 5, pp. 595603.
104
always to the same end. This means that people in any given culture only need
to achieve the level of reasoning that is necessary for their culture to operate
effectively.29 Nevertheless, even when different cultures have different attitudes
to an issue, they will express the same reasoning methods when coming to their
differing conclusions. I would suggest that individuals and groups may operate
at several levels simultaneously, depending on the multiple contexts they are
operating in at any one time (home and work, friends and work, friends and
home, and so on), and this may be a mechanism by which scientists manage
any clash between their private morals and public ethics. In addition, I would
suggest that although we may like to think we are all operating professionally
in Kohlbergs Stage 6, we are probably not, or at least, not all the time and in
all circumstances.
Kohlbergs work has been subject to criticism, including that his work is
culturally biased. Simpson30 said that Kohlbergs stage model is essentially
based on the Western philosophical tradition and that it is inappropriate to
apply it to non-Western cultures without adequate reflection on their different
moral outlooks; however, the model appears to remain a useful indicator of
the development of moral reasoning per se, and I would suggest that as long
as cultural variations are effectively and appropriately considered as variables
when applying the model, it retains useful credibility.
We must also consider the psychological effects or outcomes of our personal
cultural upbringing. A range of cultural factors may be explained by psychology
as well as by anthropology and the social sciences. Characteristics of cognition,
motivation and social interaction may answer questions about the origins and
development of any culture, and subsequently those subject to the influences
of a culture. For example, a society that values the rights of the individual
will produce people who think differently about any given scenario to people
who originate from a society that does not place the rights of the individual
high on the social agenda. Likewise, cultural views on gender, equality, age,
family structure, education, work, marriage, religion and science all influence
us heavily as we grow up, embedding within us a standard of normality by
which we measure the rest of the world. This is the basis of some of the criticism
of Kohlbergs stage model. It is probably only when we leave the confines of our
own cultural context that we are challenged to focus on our own reactions to
the others, and have to negotiate ways forward that enable us to coexist in a
mutually acceptable way. This can, of course, include the transition from the
private, domestic sphere to the professional, public sphere.
29Ibid.
30 Simpson, E. L. 1974, Moral development research, Human Development, vol. 17, pp. 81106.
105
Private morals, public ethics and dual-use/

biosecurity education
When considering the biosecurity and dual-use dilemma therefore, we need
to recognise that life scientists bring to their existing professional ethical
perspectives a range of prior moral perspectives that influence the ways in
which their professional ethical views and behaviour develop. There may
even be multiple levels of prior professional ethics and norms involved in this
equation. A practising scientist may have been required to adhere to different
norms in different laboratories and workplaces;31 s/he may have been required
in one place to prioritise safety over economics, but then be pressured to reverse
this elsewhere; leaders in one workplace may resist post-qualification research
training to the detriment of the team, whereas leaders in another may promote
such trainingthe scientist finds himself in the new workplace with his old
norm of resistance to training in place, and as a result, he finds he is viewed by
the new team as backward and unwilling to change, unless of course he manages
to revise his view on training. A number of cultural expressions may affect
this process, and although as scientists, individuals may all tend to publicly
express similar value sets, as scientists, they will be modified by the underlying
culturally mediated values of their earlier life experiences. This will occur at
both the individual and the group levels.
In order to understand this more fully, it is useful to consider why certain
behaviours and attitudes develop. Private morality, according to work by
Kohlberg, Wilson, Piaget, Catchpoole and others (above), involves characteristics
of altruism, care, empathy and a sense of justice. Krebs and Janicki32 suggested
that the way in which these are prioritised and acted on by a given group may
be an adaptive mechanism, with people choosing the adaptation that best
helps them to tackle given cultural problems and manage their own responses
in such a way as to enable them to work. For example, scientists views may
conflict with the institutional views of the workplace in which they operate,
but in order to keep their position and retain a reasonable lifestyle, they have
the choice to either conform or be released from their position, perhaps with
other penalties as well as the commensurate economic ones. There is arguably a
tendency to overlook such cultural pressures and to insufficiently recognise the
forces acting on scientists. An example of this may be seen in a situation that has
been recognised under the Biological Weapons Convention (BWC): sometimes
even well-intentioned international prohibition regimes can lead to confusing
situations. The BWC prohibition norms allowance for biotechnology research
carried out for peaceful purposes has actually enabled a number of countries
31 See, for example, Bezuidenhout, Chapter 17 in this volume.
32 Krebs and Janicki, op. cit.
106
to cloak their biological weapons programmes within seemingly legitimate

facilities.33 Scientists working in such conditions are faced with the need to
juggle their private values and professional ethics in order to reach a state of
equilibrium in which they can continue to work with the least possible stress
and dissonance to them personally.34 This example offers a partial illustration
and explanation of the different ethical values and behaviours that we see
apparently exhibited around the world todayin both developing and
developed countries. Cultures (national, political, ethnic and religious, in some
cases) prioritise the values that enable them to best tackle the challenges they
face, and at the micro-level, the individuals involved in these circumstances
have to prioritise and compromise their own values within these contexts in
order to survive. Martin wrote an excellent commentary on this, focusing
on those workers who chose not to compromise their own moral and ethical
values.35 I will return to this later.
So how do we draw together common ethical values from such disparate moral
backgrounds within the context of the risks of the dual use of scientific activity?
It may be argued that scientists already share a common culture, but Crandall
and Schaller36 have highlighted professional/personal value clashes even within
(and between) scientific cultures. They suggested that most scientists publicly
state that they hold to an explicit set of values including a commitment to sharing
universal truths, to sharing data with colleagues, to being disinterested in the
sense of being non-judgmental and open to whatever the evidence suggests, and
to a sceptical outlook that distrusts anything not supported by empirical data
and observation. They also found, however, that a hidden agenda is apparent
in the work of most scientists, which clashes with the explicit values of the
scientific method. They suggested that in order to survive in the competitive
world of science, scientists must employ a set of hidden values, including
33 Atlas, R. and Somerville, M. 2007, Life sciences or death sciences: tipping the balance towards life with
ethics, codes and laws, in B. Rappert and C. McLeish (eds), A Web of Prevention: Biological Weapons, Life
Sciences and the Governance of Research, Earthscan, London, pp. 1533.
34 See the examples of Experiments of concern given in the Fink Report, which states that review of
research activity by the scientific community would be necessary if it involves (amongst other processes)
demonstrating: how to render a vaccine ineffective; how to enhance resistance to antibiotics; or how to alter
the host range of a pathogen. All of these activities could be legitimate elements of peaceful research, but
who on a team engaged in such work is likely to raise any alarm if funding for the research and the team
could be put at risk? See National Research Council 2004, Biotechnology in An Age of Terrorism, [Fink Report],
The National Academies Press, Washington, DC, <http://www.nap.edu/catalog/10827.html>. See also the
discussion of the ethical implications of publishing the reconstruction of the 1918 flu virus when it was
published in 2005. See Tumpey, T. L., Basler, C., Aguilar, P., Zeng, H., Solorzano, A., Swayne, D., Cox, N.,
Katz, J., Taubenberger, J., Palese, P. and Garcia-Sastre, A. 2005, Characterization of the reconstructed 1918
Spanish influenza pandemic virus, Science, 7 October, pp. 7780. This paper was accompanied by an editorial
in Science discussing the decision to publish the research, bearing in mind concerns about the risks of misuse
of it.
35 Martin, B. 2007, Whistleblowers: risks and skills, in Rappert and McLeish, op. cit., pp. 3547.
36 Crandall, C. S. and Schaller, M. 2004, Scientists and science: how individual goals shape collective
norms, in Schaller and Crandall, op. cit., pp. 20124.
107
resistance to new ideas that challenge the status quo, the selective forgetting
of data that contradict personal views, favouring of work supporting their own
views and a generally conservative approach to their practice. Conflicts between
the personal and professional value sets will obviously arise here.
This conflict may be explained psychologically by cognitive dissonance
theory,37 which describes how, in situations in which an individuals behaviour
or beliefs must conflict with beliefs that are integral to his or her self-identity
(for example, when being asked to professionally prioritise a value that is not
equally prioritised in his private value set), a conflict will arise. This conflict
is referred to as cognitive dissonance. In order to resolve this dissonance, the
individual can either leave the situation that presents the source of the conflict
(in this case, the professional role) or reduce his attachment to the private value
that is being challenged or compromised. If he does the latter then he may ease
the conflict further by emphasising the positive aspects of the professional value
that is challenging him. Cognitive dissonance is clearly an important factor in
the decision-making process.
Three key psychological strategies are commonly used to alleviate cognitive
dissonance. The individual may decide to seek more supportive beliefs or
ideals that could outweigh the dissonance; in essence, this may mean masking
or suppressing the conflicting value and his response to it. This would, in
turn, potentially cause longer-term psychological stresses. The individual
may convince himself that the challenging belief is not really important, and
can therefore be ignored or at least given less priority. He may also begin to
manipulate or moderate the belief that is challenging him and seek to bring
it more into line with his other beliefs and action patterns. These strategies
involve rationalisation: the tendency to develop their own explanations as to
why they have made their choices in relation to the challenging belief or action.
Festingers cognitive dissonance theory has not been without its detractors.
Bem38 suggested that individuals attitudes may be changed by self-observation
of their own behaviours (rather than by any response to feelings of discomfort
in a challenging situation), followed by consideration of what attitudes caused
those behaviours, thus producing emotional responses. For some time it was
considered that cognitive dissonance and self-perception theories were in
competition; however, work by Fazio et al.39 identified attitude-congruent
37 See Festinger, L. 1957, A Theory of Cognitive Dissonance, Row & Peterson, Evanston, Ill.; and Festinger,
L. and Carlsmith, J. M. 1959, Cognitive consequences of forced compliance, Journal of Abnormal and Social
Psychology, vol. 58, pp. 20310.
38 Bem, D. J. 1967, Self-perception: an alternative interpretation of cognitive dissonance phenomena,
Psychological Review, vol. 74, part 3, pp. 183200; and Bem, D. J. 1972, Self-perception theory, in L.
Berkowitz (ed.), Advances in Experimental Social Psychology, Academic Press, New York, pp. 162.
39 Fazio, R. H, Zanna, M. P. and Cooper, J. 1977, Dissonance and self-perception: an integrative view of each
theorys proper domain of application, Journal of Experimental Social Psychology, vol. 13, part 5, pp. 46479.
108
behaviour and attitude-discrepant behaviour, with self-perception theory

explaining the former and cognitive dissonance theory explaining the latter.
Attitude-congruent behaviour was defined as any position within an individuals
scope of acceptable behaviour and attitude-discrepant as any position within an
individuals scope of rejected behaviour.
The conflicted scientist: An example

Now that we have considered biological and psychological explanations of moral
development, lets look at the scientist we mentioned at the start of the chapter.
This hypothetical, recently graduated biochemist was appointed to his junior
post as a research assistant on a team in a commercial organisation developing
vaccines to counter an emerging tropical disease.
Lets imagine that his team leader and his employers prioritise economic
and scientific advantages over what they perceive to be the requirements of
biosecurity; indeed, the organisation believes it is already accommodating
biosecurity concerns through its biosafety policy. In as much as it has been
addressed at all, biosecurity as a concept is generally considered to be irrelevant
in the organisations laboratories because the management does not believe
that their work presents any biosecurity risk. Moreover, it is vital that this
organisation maintains scientific and economic leads over its competitors.
Our scientist has been brought up at home to believe that it is socially necessary
to abide by the rules in life, so that everyone can live together in the community
in some sort of respectful harmony. At school and at university, he modified this
necessity into a more flexible suggestion, meaning that he continued to abide
by the rules (which he is quite good at doing, at least in as much as it benefits
him), but felt able to ignore some as it suited him. In his childhood, he learned
that respect for the quality of life of others, as well as his own, is a key value,
and that he should always do his best to ensure the safety and wellbeing of
others. He learned respect for his elders, politeness and the need to recognise
his own position in the food chain both socially and at home. At work as an
adult, he believes that his leaders adhere completely to what is often known as
the scientific method (not an official term, but one that implies a commitment
to empiricism, objectivity, truth, openness and so on; whether one accepts this
or not is an interesting discussion in itself), as he believes he has seen them
do so regularly. He generally accepts the values that his scientific leaders pass
on to him both consciously and unconsciously. He has somehow forgotten the
one or two instances in which he noticed his leaders taking shortcuts in an
experiment and ever so slightly exaggerating a finding in a paper. Seeing himself
109
as a qualified and employed scientist, his priorities are twofold: accepting and
promoting the scientific method as an independent, value-free and objective
concept, and making sure he gets on well with his workmates and bosses.
He can see that his organisations new vaccine has considerable economic
potential as well as humanitarian value; however, he is unexpectedly challenged
by the need to manage his response to some new valuesin this case, a new
version of the biosecurity normwhich have been presented to him through
an international report40 he has just read. He wishes to persuade his colleagues
of the new perspective that he has recognised because he is so persuaded by the
argument that he has prioritised this new ethical approach at a high position in
his own professional value set. He is concerned that one or more of the processes
involved in the vaccine research and development could be used against human
populations if they were to be used maliciously. He sees this very clearly once
he has thought about it, and is concerned that no-one at work has considered
this before. How can he proceed?
He tries to tell his colleagues about his concerns. One or two of them agree that
there may be some theoretical risk, but generally take the view that it will
never happen and tell him that he should forget about it. Some point out to him
that this new process will generate a lot of money for the organisation (and longterm employment) and he should not raise problems with it. He tries to forget
his concerns, but as he continues to work on the new process, he sees more and
more chances for it to be misused. Other researchers from different branches of
the organisation have had access to the laboratories during the work, and no
confidentiality mechanisms were in place that would have prevented them from
seeing the work, so the processes could already be repeated at other sites.
Our scientists research team leader is planning to publish the work, albeit not
in full, so as to protect the organisations economic advantage. Even so, key
processes and outcomes can be relatively easily worked out by anyone with a
basic knowledge of genetics and vaccine science from what is to be published.
In order to advance his position further in the discipline, his team leader is
also going to give some more detail in an international conference podium
presentation, where he will take questions and presumably give some answers.
Our scientist goes to see the team leader to ask him if it would be possible
to restrict the amount of information given out in the publication and at the
conference, as he has concerns about security and what could happen if the
information fell into the wrong hands (delegations from certain countries with
security problems will be present). His team leader is shocked and annoyed at
the suggestion that he could possibly be involved in work that could be used
to make, in effect, a biological weapon. He challenges our scientist to show him
40 For example, National Research Council, op. cit.
110
what is wrong with the work, and goes on to say that he cannot be responsible
for what someone else may do with his work. Our scientist is not sufficiently
up to speed with all the possible ways in which the work may be misused, but
he suggests that certain potential outcomes of the research could be used against
the interests of society. He finds it difficult to understand that his colleagues and
team leader cannot see the potential problems that may arise from the teams
work on interfering with the susceptibility of organisms to vaccines; they do
not seem to be concerned that it could be used against human communities. The
team leader then falls back on the argument of scientific freedom and the duty
of the scientist to share and to replicate work with other scientists. Moreover,
he also points out that the research is going to generate a major income stream
for the organisation. He suggests to our scientist that he might wish to consider
his junior position on the team, and points out that contracts will soon be up for
renewal. Over the next few weeks, the team leader begins to give more work to
other colleagues on the team and less to our scientist, who does not receive an
invitation to the conference at which the leader is to speak.
Our scientist manages to find a friend who is quite senior at another workplace,
with whom he shares his concerns in a general way. This new friend suggests
that he could think about what it is about his own values that is causing him so
much stress in this situation. He has another friend, who is not a scientist, who
says that he should ignore the stress and what others say, and simply focus on
himself and what he can do to change himself in order to manage better at work.
Our scientist is now in an ethically challenging situation and does not know
which way to turn. What should he do?
We can review our scientists position using the concepts that we have been
looking at in this chapter. His upbringing has given him a family-based, culturally
derived value set (a bio-cultural mechanism) that encourages him to abide by the
rules in order to maintain his social position and to gain advancement socially as
he fits in with his tribes both at home and at work (Wilsons theory). He has
grown up with this approach, which has worked so far, but if he has to challenge
it beyond certain narrow limits, he feels emotionally stressed (Griffiths and
Izards theories). He is aware that it is necessary to balance competing values
and tries his best through life to manage the periodic clashes between the need
to fit in and the need to maintain his own values (the theories of Sahlins, Gould
and Lewontin on cultural drivers overcoming genetic drivers). He feels that it
is possible to manage this sort of balance by everyone giving and taking in
order for everyones needs and values to be accommodated (Trivers reciprocal
altruism). He is aware that some people are manipulators and will take when
appearing to give, but as long as it all balances out he is happy and assumes
that everyone else is as well (Moores criticism of Trivers, and the work of Mauss
on gift-giving as a means to self-promotion). He is even of the mind, at times,
111
that Dawkins ideas on genes as drivers can be related to humans in the sense of
every man for himself, although he does not like this view of life, as it seems
uncaring and inhumane.
He has a number of options for dealing with these questions (psycho-cultural
mechanisms). He is vaguely aware that he manages the periodic clashes between
his personal and professional value sets by balancing his unconscious feelings of
guilt and shame for doing wrong if and when he steps out of line in the family
or at work (Williamsons theory). His mechanisms for managing these emotional
reactions are those described by Krebs and Janickihe always manages, usually
unconsciously, to use one of their four ways to make sure he can operate in the
right way in any given situation. He is, unconsciously, at different stages of
Kohlbergs model depending on which context he is operating in at any given
timefor example, at home, at work or at play.
Now that he has been confronted with the challenge at work that I have
described above, he is faced with a situation of cognitive dissonance. In order to
remain at work and to flourish in the way he wishes to professionally, he must
now either conform to the prevailing view at work by giving up his concerns
(or at least stop talking about them) or choose to leave and seek work elsewhere.
As a junior member of the team, he does not think he is in a position to influence
the team himself. If he were higher up the food chain, he may feel that this is
a third option.
Our scientist now has four options to mitigate his cognitive dissonance at work,
thereby relieving himself of any sense of ethical responsibility about the possible
outcomes of the scientific activity of his organisation. First, he may choose to
look for some other beliefs that will help him to mask the dissonance so he can
pretend it does not exist; he could do this by saying to himself that his actions
alone will have no effect in the big picture as he is only a junior researcher:
what do I know? Second, he may decide to go along with his colleagues and
persuade himself on some level that it will never happen; he could do this
by actively deciding that he is worrying over nothinghe may be the only
person who has seen the potential problems, so maybe he is blowing it all out of
proportion and he is worrying about nothing. Third, he may try to manipulate,
in his head, the challenges that the dissonance is causing, so that the views of
his leader and colleagues do not cause him so much difficulty; he may do this by
reassessing the value that he places on the views of his colleagues and leaders,
deciding that their views are not as important to him as he previously thought.
Fourth, he may choose to be a whistleblower and speak publicly in some way
about his concerns.
The first three approaches involve our scientist in a process of rationalisation
whereby he will be able to invent and justify his own explanation as to how
112
he has reached his eventual position of relative ethical comfort (or discomfort).
Unfortunately, all these actions are likely to lead to further stress, which may
debilitate him in the future. In the first instance, he is effectively degrading his
own value in the team as a scientist and as an individual. In the second, he is
subduing his own intelligence and ignoring what his own values and intellect
are telling him. In the third, he is downgrading the value of those to whom
he naturally looks for learning. All of these rationalisations ultimately mean
compromise of his personal and professional values and are highly likely to
result in further dissonance.
In terms of self-perception theory, all three strategies will simply move him
to various stages along the continuum of attitude-congruent (acceptable) and
attitude-discrepant (unacceptable) behaviour. If he makes himself comfortable
at one point, this means that he will thereby throw up a further dissonance that
requires another relocation on the continuum. This may progress to such an
extent, raising more and more stress, that our scientist may feel that he is in an
impossible situation if he is unable to convince his colleagues and leaders to take
his concerns about the research seriously and to act on them.
The fourth approachto become a whistlebloweris, unfortunately, not a
stress-free option, even though it may assuage his moral and ethical concerns.
Our scientist may well blow the whistle on the research, and be the cause of it
being abandoned or amended by his organisation. So far so goodhe may have
prevented a possible biosecurity threat from translating into an unwelcome
action; however, in the process of doing this, if we read Brian Martins work
on whistleblowing,41 we find that he is highly likely to be ostracised by his
workmates, sidelined (at best) in the organisation and, more likely, fired from
his post as a result. If not fired, he may find that his contract is not renewed as
the organisation is restructured. Either way, he now has no job, no references
and little hope of continuing to work in the field he had chosen to spend his
life in. This prevents or seriously delays him buying his first house, starting
a family and advancing up the career ladder. Maybe he should have kept his
mouth shut after all?
A way forward: Memes and norms

Given that modern-day whistleblowing is such a dangerous option for the
individual to take, and given also that we accept the concept of biosecurity
risks being inherent in biotechnology research, it seems that we need to find
a way forward that reduces the necessity for whistleblowing to take place and
41 Martin, op. cit.
113
reduces the risk of biosecurity lapses occurring. One such way is to develop and
foster a new norm of biosecurity; this would become, in time, a part of everyday
scientific practice, just as the biosafety norm is hopefully embedded in daily
activities.
This is best considered in the context of the ways in which scientists carry
out their daily activities and approach potential challenges in their work.
We have seen earlier in this chapter that there are two possible psychological
mechanismsbio-cultural and psycho-culturalthat may be usefully pursued
further when attempting to identify, understand and predict the changing
perspectives of scientists when faced with the challenges of biosecurity in their
professional and personal lives. We should therefore consider these potential
explanations in the derivation and development of value sets or norms when
developing a new ethic of biosecurity and dual use. What we are faced with is,
essentially, a relativist scenario not only between the differing values held by
different people in regard to specific concepts, but also in the differing values
held by individuals themselves in terms of specific concepts through conflict
between their private moral values and their public ethical values.
Now that we have seen the challenges faced by our hypothetical challenged
scientist (albeit supported by real-life examples in Brian Martins work), lets
look at how cultural ideals and values can be transmitted effectively, thus,
hopefully, avoiding or reducing the chances of his situation being repeated in
real life.
Norms may be defined as formal or informal rules that govern what is acceptable
in terms of behaviours, attitudes, beliefs and values within a group or culture.
They may be explicitly stated or implicitly approved by the majority; they may
lead to warnings or reproach for those transgressing minor normative rules,
or to more severe punishment for those guilty of the transgression of norms
considered to be of great consequence. Norms develop over time, and are subject
to a range of cultural influences. They are often closely tied to religious and
social forces, and are commonly associated with group or cultural identity. Once
this cultural identification element takes hold, it may be particularly difficult to
dislodge or change a norm. Considering that such norms are usually part of our
childhood development and therefore our moral learning process, we can see
how they may challenge later professional ethical values that may clash with
them. So how do we develop and implement a new norm? Maybe it is better to
start with a smaller task and look at what Dawkins refers to as memes, which
may be referred to as infant norms, for want of a better term.
114
Memes42 are units of cultural transmission that may be skills, knowledge, beliefs,
attitudes, behaviours or values. They are spread through observation and social
learning43 and have been called cultural software44 that may be passed on as
they are or in modified forms. Dawkins has, in the past, cited the fashion for
wearing a baseball cap back-to-front as an example of a meme that has spread
through certain sections of society. One could argue that a well-entrenched
meme becomes, in effect, a new norm, if it lasts long enough. Crucially, unlike
genes and norms, which can take long periods to change, memes can, apparently,
change very quickly. It may follow then that ethical attitudes, knowledge and
characteristics could also be transmitted as memes, which, in passing from
one person to another, will change slightly with each transmission, allowing
development and swift cultural change to occur:
The power of human reason made possible in part by the memes we
possess, is also the power to mutate those memes and create something
new from something old. We are not simply inheritors of a zealously
guarded patrimony but entrepreneurial products of a new cultural
software which will help constitute future generations of human
beings.45
In order for old memes to be transmuted into new memes, however, there must
be a dominant host culture that allows change to take place.
In order to enfranchise and empower all people there needs to be a mutual
exchange between equals. If some individuals or groups are marginalised or
excluded for any reason, be it at the political, legal, social or economic levels,
there is no basis for trust, mutual respect and cooperation.46
This is clearly the level at which we need to operate when developing a new
norm of biosecurity as it is embedded in the responsible conduct of science. In
any moves to take on board shared common ethical values, it is vital that we do as
much as possible to support all actors in the biotechnology field so that no-one is
left behind and marginalised. Such help does not need to be financial, although
it could be, at least in part. Help could be provided in the form of intellectual,
educational, training or other relevant forms of support as well. Once a norm
has become established, recognised and shared among growing numbers within
a community, it is important that we define appropriate penalties to be faced by
those breaking the norm. Next, we need to agree on the implementation of those
penalties; without imposed penalties, a norm, by its very nature, is effectively
42 Dawkins, op. cit.
44 Balkin, J. M. 1998, Cultural Software: A Theory of Ideology, Yale University Press, New Haven, Conn., and
London, p. 43, cited in ibid., p. 265.
45Ibid.
115
toothless. In todays world, it may be argued that the economic penalty is the
one of greatest effect; however, penalties need not be just economicbut that is
another argument for another day.
Lets now consider the education route as a norm developer. Education is itself a
memes transmitter. We absorb new memes through the values set by those who
pass the new values on to us. Over time, new memes can become new, deep-rooted
norms. Social norms (such as the baseball cap) may start as memes, developing
into expectations of behaviour, then into controls that regulate behaviour and
identity (the baseball cap again). Philosophical norms imply some obligation or
duty, carrying threatened or real sanctions if they are broken. There is cultural
pressure to abide within the norm, so culture provides a setting in which the
norm may survive and develop.
Ethics and ethical attitudes may be passed on as memes (quick-changing,
initially superficial cultural artefacts) and then develop over time into norms
(more deep-rooted values and attitudes within a culture). Ethics, as norms, can
be implicit or explicit. They can vary from group to group and place to place
and can change over time. It is this capacity to change that we must address
in order to develop a new ethics of biosecurity so that it becomes a dominant
scientific norm. The biosafety norm is based on the principle of containment
and safe management of harmful biological agents, but in such a way as to
allow ongoing scientific practice; the biosecurity norm needs to be built on
the principle of the containment and safe management of biological knowledge
and processes, also allowing ongoing scientific practice. Just as the biosafety
norm (when implemented properly) draws a line in the sand over which practice
cannot step, so must the biosecurity norm.
So how may we develop research ethics as memes-into-norms and persuade
others to sign up to them in the biosecurity context? This would mean taking a
norm of, say, no harm, and defining it in practical ways in order to teach the
principle. Catchpoole believes that new norms can be taught and developed as
memes. She suggests that education can choose which memes to transmit and
which it will not.47 Educational transmission of a commonly held value, such
as care for others, could be inserted as a norm as part of the building of a
multicultural ethic of biosecurity, with a common value set, around it:
Having care for one another and our world involves finding a balance
between the needs of self, close others and global others in ways that
respect basic human rights and yet which seek to encourage more than a
minimalist commitment to having care for one another By delineating
the nature of care itself and recognising the ubiquitous nature of power
47 Ibid., p. 126.
116
within relationships, an extended ethic of care provides a set of transcultural memes for both guiding and evaluating the development of the
ethical form of life in the private and public spheres.48
The memes of ethics could therefore be taught and learned as a form of cultural
transmission in a multicultural context if they are presented from the perspective
of a commonly held value such as care for others, which few cultures would
dispute.49 At the same time, we need, however, to recognise that a range of
cultural pressures may be brought to bear on the implementation of a no harm
or care for others value. It is one thing to state a belief in do no harm to others
publicly, but at the same time, be driven by a more pressing value to actually
prepare to do the oppositewe only need to look at those countries in violation
of the BWC with their peaceful purposes cloak hiding offensive weapons
programs to see real-life cases in action. This prioritisation of pressures is to be
the subject of further research.50 One could argue that the norm of no harm is
so routinely violated in science (for example, in the use of animals in product
testing) that it is diminished as an argument and is no longer worth pursuing
or developing in practice; however, this in itself is an ethical debate: we either
accept the norm of no harm as being something to aspire to or give in to an
anarchic world in which care for the other no longer has a place worth arguing
for. I would suggest that it is inside the norm of no harm that much future
research is required. We can readily see that many individuals and groups apply
this norm differently, simply by the way in which they define who is to be
actually protected from harm, in what circumstances and why. This is the area
in which we seem to make our moral and ethical judgments based on culturally
appropriate norms with which we face everyday life.
Conclusions
This chapter has shown that we need to appreciate the variety of private moral
backgrounds of individual scientists as we seek to identify common values with
which to form a new norm of biosecurity within the public culture of ethical
science.
It is probable that there is a biological foundation underpinning our moral
development from infancy into adulthood. On top of this genetic and physiological
framework, we can identify a range of psychological processes that further
48 Ibid., p. 47.
49 Ibid., p. 186.
50 Further culture-based research is being planned by Judi Sture and Masamichi Minehata of Bradford
Disarmament Research Centre, looking at how scientists and educators prioritise and manage pressures in
their everyday work.
117
develop within us an increased capacity for empathy and consideration of the

rights of others. Some of these processes have been identified and categorised
by Kohlberg, Hoffman and others. An increasing body of work has shown that
psychology is a powerful factor in the development of culture, and that it is
possible to understand the nature and growth of cultures and cultural changes
through the lens of psychology. This is an area ripe for further investigation in
the biosecurity context.
Each individual scientist engaged in work or research that has or may have
potential for misuse is faced with a moral-ethical dilemma of his own. His
private moral value set(s) may come into conflict with his publicly required
professional value set(s). Because they are culturally derived and embedded,
private moral values can be difficult to change, as they are usually driven or
controlled by one of five expressions of culturesocial organisation, economics,
technology, religion or languageall of which are interlocking and may be hard
to shift (these expressions of culture are commonly used by archaeologists in
understanding culture).51 We need to identify the key drivers from among these
five expressions that influence the private morality that scientists bring to their
professional ethics, in order to challenge them and move forward to a common
value set. In addition, we need to recognise how these drivers also inform public
ethics in the workplace. By considering a number of existing theories such as
cognitive dissonance and self-perception, against a biological background of
genetic and evolutionary tendencies to altruism, emotional reaction or selfinterest, it may be possible to delve into the processes that give rise to the
professional ethical value set(s) of scientists in any given cultural setting and
that inform their prioritisation of each of the five expressions of culture.
Against the background of a call for increasing awareness of dual-use issues,52
we may consider the following fields of activity and inquiry in order to move
forward. First, we need to help scientists recognise their own value sets and
identify how they may clash with both the values of the science they practice
and the need to protect against the risk of dual use. Second, we need to review
our institutional organisations and look for hidden curricula (confounding
values that compete implicitly with required values) in science at governance
and teaching levels that may be impeding the search for and implementation
of a common value set in biosecurity and dual use. Third, in our aim to embed
a new ethics of biosecurity and dual use within the culture of life sciences,
we need to consider a range of psychological drivers that influence others and
51 See Renfrew, C. and Bahn, P. 1991, Archaeology: Theories, Methods and Practice, Thames & Hudson,
London.
52 United Nations 2008, BWC/MSP/2008/5, p. 6.
118
ourselves. Fourth, we need to look out for threats or undue influence from the
five archaeological expressions of culture, as these may exert overt or covert
pressure on scientists.
Finally, by identifying even a single commonly held value, such as care for
others, it may be possible to start laying the foundations of a new value set that
will underpin our efforts in building a sustainable capacity in biosecurity in the
life sciences.
119
8. Responsible Dual Use

John Forge
Introduction
This chapter addresses the moral dimension of dual-use research. To set the
scene, I will begin by explaining what I take dual use to be. I understand a dualuse item to be something that has both a good or neutral (neither good nor bad)
use or application and a bad use. Three different categories of dual-use items
can be distinguished: research, technologies and artefacts.1 These are clearly
different sorts of things. Research is an activity, while technology is a form
of knowledgeknowledge of the techniques for the production of artefacts
whereas artefacts are objects. But it is also clear that these items are related:
research aims to give us technology, which in turn produces artefacts.2 It is the
last that normally has the immediate impact on the individual, be this good
or bad. For instance, research has led to methods for mass-producing cheap
ammonium nitrate, which has a primary use as a fertiliser and a secondary use
a bomb-making material. It is the substancethe artefactammonium nitrate
that is the fertiliser/bomb component, not the technology or the research. The
import of this is that when we talk of threats or risks associated with dual use,
we need to distinguish between different levels of risk or threat, depending
on the nature of the dual-use item, and between different ways of dealing
with these. For instance, substances like ammonium nitrate can be physically
contained, but the knowledge of how to manufacture it cannot.
My suggested definition of a dual-use item is therefore as follows:
An item (knowledge, technology, artefact) is dual use if there is a
(sufficiently high) risk that it can be used to design or produce a
weapon, or if there is a (sufficiently great) threat that it can be used in
an improvised weapon, where in neither case is weapons development
the intended or primary purpose.3
1 This is the classification that I gave in Forge, J. 2010, A note on the definition of dual-use, Science and
Engineering Ethics, pp. 11118. This paper was written as a response to a challenge by David Resnik to come
up with a definition of dual use that is neither too narrow nor too broad. I do not claim that my contribution
is the last word, but I do think it is along the right lines. I also think that it is enough by way of a starting
point for the present discussion. I would, however, note that the sense in which artefacts are said to be objects
is a special onesee ibid., p. 115.
2 There are of course other kinds of research and other kinds of technologies besides those that aim to
produce artefacts.
3 Forge, op. cit., pp. 11118.
121
There are several comments in order. First, I equate the badsecondary,

unintended, and so onuse with weapons development. It is certainly possible
to broaden the discussion of dual use to include other harmful activities, but
I think these would be more controversial.4 I should stress that by weapons
development I include all aspects of the design, production, testing and
commissioning of both weapons themselves and all the ancillary structures,
platforms and processes that are necessary for their use. Second, the reference
to risk in the definition points to the possibility of the production of a new or
improved weapon. The (by now) familiar examples of techniques of synthesing
pathogens are thus risky in this sense. Threats differ from risks here in that
they signify the intention to do something harmful, rather than the possibility
that something harmful might happen. I have coupled risk with the design
of (new) weapons but threats with improvised weapons. States threaten one
another, as North and South Korea are doing at present, though the threat here
is usually conditional.5 In regard to dual use, however, I think we are more
concerned at the moment with sub-state actors, such as terrorist organisations,
using or adapting existing items for improvised weapons, as has been done
with ammonium nitrate, than with states arming themselves with improvised
weapons, as states do not usually need to improvise weaponry.6 Finally, research
that is intended to produce a weapon and which, fortuitously, has a friendly
civilian application, is not dual use according to my definition: the bad use is
always the unintended one.
My focus here is with dual-use research, the dual-use item that appears
furthest removed from any immediate impact on the individual, and with the
responsibilities of those who undertake research that is, or could be, dual use.
So if a particular research project has some prospect or likelihood of leading to a
bad outcome then what do we expect of those who would conduct the project?
Perhaps if the researcher does not intend for the bad outcome to come about,
or if she does not foresee that it will, she has no responsibility for it? There are
well-known discussions of the doctrine of the double effect that could have
some application here.7
Another issue is if the likely bad effects seem to be outweighed by the prospect
of good outcomes. The consequentialist who sees moral action as reducible
to the production of what, on the whole, is the best outcome is likely to be
sympathetic to this possibility. I shall touch on these issues in what follows, but
4 And this is very much in the spirit of the early definition of dual use given by the Office of Technology
Assessment in the United States and subsequent definitions given by bodies such as the National Research
Council. See ibid., p. 112.
5 If North Korea shells South Korean islands again then the South will retaliate.
6 The extent to which we should therefore be cautious about undertaking dual-use research becomes
important. The relevance of the precautionary principle, which would seem to have application to dual use,
is the subject of other chapters in this volume.
7 See Suzanne Uniackes Chapter 10, in the present volume, for a discussion of this.
122
my main aim is to apply a particular way of understanding the responsibilities

of researchers that I proposed in a recent work, The Responsible Scientist,8 to
the question of dual use. These responsibilities are both backward looking
and forward looking, having to do with what researchers are responsible
for in the past and what we want to hold them to in the future. Both sorts of
responsibility are relevant to dual use. The upshot of the account that emerges
is a cautious and restrictive response to dual-use research; other accounts will
have other implications. I will not assume much familiarity with the topic of
responsibility or with moral philosophy, so much of what follows in section one
will be exposition. I hope the implications of this will become clear in sections
two and three. As a final comment by way of introduction, what follows is put
forward as a contribution to the ongoing discussion of dual use. The topic is still,
I think, at a stage where a variety of different approaches are worth considering,
in regard to both definitions and the moral and policy dimensions. I hope the
account of responsibility that I have developed can stand as one such approach.
Responsibility and research

There are two kinds of responsibility: 1) responsibility for past actions, actions
that an agent has already completed, or omitted to donamely, backwardlooking responsibility; and 2) responsibility for future actions or omissions
namely, forward-looking responsibility. As we shall see, these concepts are
relatedand both are relevant to scientific research. It is easiest to begin with
a simple example that has nothing to do with science: suppose it is the job of
a railway guard to make sure everyone is on the train and the doors are closed
before signalling to the train driver to leave the station. If the guard fails to
realise that a disabled person will take extra time to get on the train or if he omits
to make sure that a carriage door is closed then he will be held responsible for
any harm that subsequently occurs. This would not be true of a member of the
public. We would hope that a member of the general public would try to assist
the disabled person or close the carriage door, but, unlike the guard, she has
no special responsibility to do so. In this example, the special responsibility
in question can be termed role responsibility. It is (a kind of) forward-looking
responsibilitythat is, one that obliges the person in the role to conform to
certain expectations about his or her future actions in a certain sphere: train
guards, for example, are supposed to make sure trains are safe. There are
other jobs and professions that define roles that are such that the performers
are supposed to safeguard others from harm. These are not all of a piece. For
instance, those of the medical and legal professions are more complex than those
of train guards, lifeguards, and so on, which is one reason these roles have an
8 Forge, J. 2008, The Responsible Scientist, Pittsburgh University Press, Pittsburgh.
123
associated professional ethic. An important point here is that certain failures to

foresee or omissions to act that result in harm to others can only be attributed
to the agent if there is a corresponding forward-looking responsibility. I am not
obliged to keep an eye on toddlers at my local pool, but the lifeguard is, and if
she omits to do so then she is open to blame.
In my book, I argued that scientists have a kind of role responsibility, even, and
perhaps especially, scientists who claim to be doing pure research.9 I used this to
ground a wide account of backward-looking responsibility for science, one that
held that scientists could be held responsible for failures to anticipate where
their work might lead, and for omissions to act in certain ways, in addition
to being responsible for what they intended to do and for what they foresaw
would be outcomes of their work. These are, of course, complicated issues.10
For instance, it might be asked how could a scientist know where his research
might go, what kind of applications it could have in the future? How could
Newton know of all the applications of universal gravitation or Einstein of
relativity? I did not try to argue that all scientists can foresee all applications of
their work, for that would have been a hopeless task, but I did show that there
are circumstances where this is possible. For example, had all the information
about the properties of the uranium nucleus available in 1939 and 1940 been
published, there was a significant probability that this would have led to a more
vigorous effort on the part of the Nazis and others to develop an atomic bomb.
And as the gap between pure research and applied research narrows, as it is
surely doing more and more in an age when funding for science is increasingly
tied to applications, it is easier to look ahead. Indeed, to continue the theme of
funding, research proposals are very often obliged to list possible applications
as outcomes. I will come back to this point.
The basic reason all scientists have a kind of role responsibility is simply because
science affects people (SAP) and does so in profound ways. I contend that these
effects are not always good. I believe that all weapons of mass destruction,
especially nuclear weapons, have done us no good, are terrifyingly dangerous,
and that it would have been better had the Manhattan Project never been set
up and no-one else had attempted to make atomic bombs.11 Like all these issues,
this is somewhat (but I think in this case not very) controversial: some think that
the dropping of the atomic bombs on Japan was a good thing because it ended
9 I understand scientist in a broad sense, not restricted to the natural sciences. Most of the examples in the
book are from physics, which is my area of expertise.
10 So complicated that I needed to devote two chapters to deal with them. One of the important steps in
the argument here was to show that we can be responsible for what we dont knowwe can be responsible
for being ignorant, which is at odds with the classical account of the matter by Aristotle. See Forge, 2008,
pp. 1079.
11 I agree with the assessment that it was only in wartime that the vast expenditure would have been
justified for the first nuclear weapons project, so without the Manhattan Project there may well have been no
nuclear weapons.
124
World War II, and that nuclear weapons prevented the Cold War from becoming
a hot one. As we learn more about the intentions of the former Soviet Union,
the latter proposition is becoming increasingly less plausible. (I am also willing
to argue, though not here, that all scientific research for the ends of weapons
innovation is wrong.12) Whatever one makes of these particular examples,
even someone with blind faith in science would be hard put to maintain that
everything scientists do has turned out for good. It should be stressed that it is
not part of the present position that when something has gone wrong then this
has been done intentionally or that scientists are always to blame. And it is far
from the present position that science either has no effect on us or is bad overall.
Clearly, science has done a great deal of good, as is clear when we review the
history of medical science.13
Once we accept SAP, it seems clear that we would like science to somehow
maximise the good outcomes and minimise the bad ones. The first step, or
maybe the preliminary step, is to get scientists to accept SAP; by this I mean all
scientists and not just those working in applied areas. Surely this should not be
too difficult and is probably already true of the majority of scientists. That all
scientists should be aware that their work can affect others can be argued with
reference to what I call the changed context of science. Indeed, since World War
II, all working scientists really should know that no scientific research is pure
in the sense that it cannot in principle affect people. Once this is established,
we get on to the hard choices: just what kind of research should scientists be
encouraged to do, and what to avoid? Philosophers can have some influence on
the policy processat least they can in theoryby suggesting some general
guidelines. What I have tried to do is to provide a set of moral principles that
could be used to inform the choices made by scientists themselves, though
policymakers could also use them. There are different ways in which this can be
done, given what I have said so far about responsibility, and in the rest of this
section, I will outline the way I prefer.
The reader may be aware that there are two traditions in moral philosophy
traditions that seem sharply opposed. According to the consequentialist tradition,
the moral import of an action or choice resides solely in its consequences. The
nature of the act itself, or of how we describe or characterise it, counts for
nothing. Evidently, there will be issues here for the consequentialist, resembling
those discussed above, in regard to how consequences stretching into the future
can be determined and weighted with respect to their moral significance.
Leaving these aside, a consequentialist normative ethic will specify some
12 See my Designed to Kill: The Case against Weapons Research. Dordrecht: Springer, 2012.
13 But even here there is no universal agreement: some strange folk even think that the triple antigen
vaccination is wrong. My general point here is that science is a mixed blessing and that is why it is important
to determine just what the responsibilities of the scientist are, Forge, 2008, op. cit., pp. 2831.
125
property, to be taken as the good, and then require moral agents to maximise
this property by their actions. Famous nineteenth-century consequentialists
like Bentham and Mill took happiness to constitute the good, and thus moral
action was such as to maximise happiness, the agents included. The calculus
has to include consequences where some were made unhappy, and this means
that all sorts of scenarios can be constructed in which happiness is maximised
but some small number of unfortunates is made utterly and forever miserable.
Some of these scenarios are portrayed as counterexamples to the theory. My
own account is non-consequentialist in that it does not hold that consequences
are all that matter.14 I wont make any further criticism of the consequentialist
tradition here.
Non-consequentialist moral philosophy is usually expressed as a set of norms
or rules. What these are, how these are understood and how they are justified
distinguish different non-consequentialist accounts. Modern accounts do not
usually interpret the rules as absolutely binding, and here there is a break
with the Kantian tradition. Kant famouslyor notoriouslysaid that one
should not lie, not even to mislead a maniac looking for ones best friend with
malicious intent. The rules contained in modern accounts are thus often said
to be prima facie, or to lay down prima facie duties: they are to be obeyed in
the first instance, unless there are very good reasons not to obey them. As for
the content of these rules, I follow those like Bernard Gert who think that they
should forbid harming but not require any positive or benefiting action. Here
there is evidently a considerable difference with the traditional consequentialist.
One good reason to accept our view is that it is possible to impartially refrain
from harming everyone, but one cannot impartially benefit everyone. If I have a
limited amount of help to give, I must favour some, but moral action is supposed
to be impartial. Gerts moral philosophy, which I follow closely here, does not
exclude trying to do good; however, good action, which is understood to be
equivalent to the prevention of harm, is taken to be the content of moral ideals,
and hence agents are not bound by rule to prevent harm. It has to be said
that these are quite subtle matters and it is hard to do justice to them here.
Another way to put the difference between the rules and the ideals is as follows:
agents need to justify breaking the rulesrecall that these are not absolutely
bindingbut only encouraged to act in accord with the ideals. Gert refers to his
account as common morality.
To conclude this discussion of the moral basis of responsibilities of the scientist,
we should first note that this way of doing moral philosophy makes a lot of
sense. No person in her right mind wants to be harmed: everyone, wherever
14 Note that non-consequentialism is not the contrary of consequentialism, which would be that
consequences do not matter at all.
126
they come from, can agree with that.15 Thus everyone can accept a moral code
that proscribes harming, and agree that everyone would be better off with
less harming. Doing good in some positive sense is much more difficult to
universalise. Moreover, striving not to harm others is surely within the reach
of everyone, while looking to maximise the good consequences of ones actions
looks just too hard. Now scientists are in a special position because of the farreaching implications of their work, primarily through technology.16 Therefore
it seems fair to attribute to them special responsibilities, and in line with the
proposed moral position, these should enjoin scientists in the first place not
to do research that has harmful outcomes. I have taken this to mean above all
(and at the very least) that scientists should not engage in weapons research;17
however, I acknowledge that scientists should be encouraged to do research
that will have beneficial outcomes. In The Responsible Scientist, I therefore
maintained that the responsibilities of the scientist are two tiered. The first tier
comprises this demand to avoid research that can have harmful outcomes, while
the second encourages scientists to do research that prevents harm. Research
in the biomedical sciences could well fall into the latter category, as could
research that breaks down racial and sexual stereotypes and biases. These two
tiers correspond to the moral rules and the moral ideals of common morality.
Most of this section has been devoted to the forward-looking responsibility of
the scientist, and I will come back in the final section to see how this applies
to dual-use research. The next section, on the other hand, will be concerned
mainly with backward-looking responsibility.
Dual use and responsible research

Our interest here is with dual-use research, and not with other dual-use items,
although of course dual-use research is worrisome precisely because it can
lead to the latter. To begin with, then, suppose that a given line of research,
R1, is already established as dual use. By this I mean that it is known that its
implications are such that there are bad uses as well as good or neutral ones. It
may seem strange that anyone would work on such a project, and indeed it may
be hard to find many real-life examples; however, the enrichment or reprocessing
of spent nuclear fuel rods to make weapons-grade uranium and plutonium, and
also to make new, and high-power, fuel rods counts as an example, and thus
15 Gert holds that common morality gets purchase on all those who are rational in the sense of accepting
certain basic beliefs, such as that I am mortal, I can be harmed, I have interests, and so on. See Gert, B. 1998,
Morality: Its Nature and Justification, Oxford University Press, Oxford.
16 Science as a body of ideas has had, and still has, a huge impact on us; witness the work of the likes of
Galileo and Darwin.
17 See Forge, J. 2007, What are the moral limits of weapons research? Philosophy in the Contemporary
World, vol. 14 (Spring), pp. 7988; and Forge, 2008, op. cit. In Designed to Kill I argue that no one should
engage in weapons research.
127
experiments or even theoretical work on this topic count as dual use. There are
two reasons a scientist might think it is morally acceptable for her to work on
this kind of project in applied nuclear physics. She might think that in fact there
is no bad use, or that if she only intends to work on enrichment or reprocessing
insofar as it has a benign use in civilian power reactors and that it deals with
the disposal of hazardous waste, that is okay. The first of these reasons raises an
important general issue about dual usenamely, the basis on which the uses in
question are classified.18 The second raises issues that concern attributions of
backward-looking responsibility. I will address this second matter first.
The second issue can be dealt with by what I call the wide view of backwardlooking moral responsibility, given in The Responsible Scientist. The wide view
holds that scientists are responsible for what they foresee as well as for what they
intend (and they can also be responsible for what they should have foreseen). The
view expressed by our scientist abovethat one is only responsible for what one
intendsis what John Mackie called the straight rule of moral responsibility and
which I call the standard view. The latter term is appropriate, as the view seems
to have slipped in as a kind of orthodoxy, being endorsed by Peter Strawson and
John Austin, both, like Mackie, renowned Oxford philosophers, but without
convincing argument. At first sight it seems plausible, because intentions are
given as reasons for action: Why do you work on this project? To make better
nuclear fuel rods and remove hazardous nuclear waste. The obvious question,
then, is why we should only be responsible for actions that we have a reason
for doing.19 For those who do think that responsibility must be tied to reasons
for action, and I am not one of this number, it is easy to see how a foreseen but
not intended outcome or side-effect of an action can be included in a reason for
action as an explicit qualification. In this example, it can be done by adding
although I am aware that reprocessed fuel can be used in weapons and this
did not make any difference to my choice. By incorporating foreseen outcomes
in this way, we get what I have called the modified standard view.20 How we
might manipulate or change the standard view is really beside the point, for it is
simply a commonplace that we do hold people responsible for what they foresee
as well as what they intend, in spite of what Mackie, Strawson and Austin might
18 I raised the issue of the role of values in the definition of dual-use items and how different sets of values
could lead to different judgments about what counts as a dual-use item in Forge, 2010, op. cit., p. 117. I have
more to say about this question below.
19 In some ways it would be clearer if this were expressed in terms of interests. Our researcher has an
interest in disposing of hazardous waste, but no interest in making material for nuclear weapons; her reason
for undertaking the project cites the former as her reason for action. Nevertheless, she sees that her work
contributes to the latter end.
20 The modified standard view, and why it is not enough, is discussed at length in Chapter 5 of Forge,
2008, op. cit. Also, it is one thing to recognise that something is a commonplace and quite another to give
it a convincing philosophical rationalehence the length of Chapter 5 and the chapters before and after it.
128
have thought. And hence the response of our scientist that any other outcomes
of this work are beside the point because they are not what she intends will not
wash.21
On what basis do we decide whether something has both good and bad outcomes
or uses, and hence whether it is a dual-use item? And if there are alternatives that
lead to different judgments, which do we chose? These are important questions,
ones that I cannot address fully here. But for a start we can all agree that bad
outcomes are those that are, or have the means to be, harmful. Common morality
of course would have this implication. If we think of our moral philosophy
as expressing values then common morality elevates the value of not harming
above all others, and would not, for example, allow the harming of a few to
benefit many. Again, this is surely something that most of us would accept: if
we could benefit at the expense of others being harmed then most of us would
not accept such a deal if it were offered to us (at least I hope we would not). So
I assume again that something like the values expressed by common morality
provide a plausible and realistic value system. We will see in a moment how this
can affect classifications of items as dual use. First, however, note that there is
a difference between harmful acts, acts that directly harm moral subjects, and
outcomes of researchthe topic of the present discussionwhich are such that
they could be the means to harm. Provision of the means to harm is not the same
as harming, and if we are to try to attach the kind of strong moral prohibitions
about harming suggested here to activities that do not directly harm, it seems
that we need some further argument.
In fact I dont think that this is too hard to supply in cases of research whose
explicit aim is to design new and better weaponry.22 Scientists who engage in this
kind of work need justification for what they do: they need to give reasons why
it is acceptable for them to engage in an activity whose objective is to provide
tools whose purpose is to harm others. The typical justification is that we need
weapons to defend ourselves from others, from the bad guys. This is similar
in form to the typical justification that we give when we feel we need to harm
othersnamely, that we need to do so to prevent harm to others and ourselves.
The acceptability of that justification will depend on the circumstances: clearly,
inflicting a great deal of harm to prevent a small amount of harm will not do.
The harm inflicted should in some way be commensurate with that prevented.
But this judgment is more difficult to sustain when it comes to deciding whether
making new and improved weaponsnamely, making the means to harm rather
than harming directlyis justified. Are they really necessary? Will others
21 The challenge, of course, is, as in the previous footnote, to provide the philosophical argument.
22 Not too hard, but quite lengthy: in Forge, 2008, op. cit., pp. 1558, I argue for a means principle that is
such as to transfer responsibility for bad outcomes using artefactsweapons, for instanceto the designer
of the artefact.
129
acquire them and use them for evil purposes? What alternatives are there to
weapons research? I think these questions are extremely hard if not impossible
to answer in ways that support weapons research (namely, yes, no, none), and so
I think that weapons research is an activity that should be undertaken only in
exceptional circumstances.23
In this way we can see how to go about addressing questions about matters
that are morally suspect, such as weapons research. When the topic is dual
use, one of which involves weapons, for instance, the questions become more
complicated, but I think we can now see just where the complications arise.
The assumption is that R1 is such that we know it can have bad uses. Any
research that can be harmful needs to be justified, and so far we have focused on
justifications that do not take into account any good use (that is not a product
or function of the bad use). A good use, on the present system of values, is one
that prevents harm; in the case of R1 this is the disposal of hazardous waste.
The role this will play is thus as part of the justification of the project. What
the present system of values and justifications requires for dual-use research is
that the bad use is offset by considerations that appeal to the good use, and
any other relevant considerations, and that these good uses must involve the
prevention of harm. Where the good use does not prevent harm, where it is
neutral as evaluated on the prevent system, it cannot figure in the justification
of the project. In regard to a project like R1, it may therefore even be useful
to proceed as follows: justification needs to take the form as if it were the bad
use that is the main object of the investigationthat this is foreseen but not
intended is irrelevant hereand the good use is to be cited as a reason not to
abandon the project.
Dual use, responsible research and uncertain

outcomes
Now we need to introduce a further complication. Suppose project R2 is such
that it might have a bad use, if it were to fall into the wrong hands, if it turned
out in a certain way, if certain people used it as a point of departure for other
research, and so onin other words, R2 is risky. Consider a response here
resembling that made above about intentionthat it is not the researchers job
to look ahead and consider other possible uses of her research. None of these
potential outcomes needs to be taken into account; only the project itself and
its scientific outcomes need concern the researcher. As we have seen already,
this attitude is at odds with the responsibilities of the scientist in this day
and age. Science affects us and our environment, whether it is set up as pure
23 So exceptional as to never obtain in fact, though this is not the occasion to try to make that case.
130
or applied, and all scientists are obliged to try to look ahead and see where
their research is going. Everyone has responsibilities when they take on the
role of the scientist, and in this way scientists resemble our train guard.24 But
again there are difficulties: how are scientists able to look into the future and
see where their research will lead, for are not outcomes unpredictable? This
matter was raised above, where I conceded that some research is unpredictable
but also maintained that some is not. Moreover, in the present context where
science is heavily sponsored and funded, it can only operate where it promises
to give useful outcomes. Were this not true then institutions such as defence
departments would not spend more money than anyone on scientific research.25
Science is now far too big and expensive to be left undirected. Of course, when
scientists propose research projects and say what they hope to achieve, they
maintain that these are desirable outcomes, that their methods are proper and
so on. They do not, as a rule, list undesirable outcomes; these will not normally
be the focus of attention of researchers, but I do not see any reason they could
not be. I conclude that the present account of the responsibilities of scientists
has the implication that they are obliged to do their best to look ahead and try
to see if their work will lead to any bad outcomes.
These kinds of assessments involve judgments about risk and threat, and are
therefore more or less uncertain, as indeed are good outcomes, because all
research has some uncertainty associated with it. The present account implies
a cautious and conservative approachand this may not be welcomed by all
researchers. Consider the following familiar scenario: a research project is aimed
at uncovering the structure of a pathogen with the aim of finding a preventative
therapy or cure for the condition that it causes; however, this knowledge could
also be used for making the pathogen more virulent and resistant to the very
measures that the project was designed to put in place. The present account,
as we saw in section one, sees the prevention of harm as something to be
encouraged. This is the second tier of the two-tiered account of the forwardlooking responsibility of the scientist, and we can all agree that the prevention
of harm is a good thing. But the first tier forbids scientists from doing harmful
research, and that strongly proscribes any bioweapons research, including
bioweapons research that is unintended. Where does this leave the research
project? It does not follow that it should not be undertaken, but it does follow
that justification must be given. And here an assessment of the costs versus the
benefits is needed, as well as control of the results.
24 I discuss the case of Frederic Joliot-Curie, who published nuclear data in 1938, against the urging of
Szilard. Fortunately, the data were incorrect and it did not seem to interest the Nazi scientists; in Forge, 2008,
op. cit. Joliot-Curies actions were irresponsible even back then.
25 For many years the US Department of Defense has spent more money on scientific research than anyone
else.
131
If there is some risk that the project will be used for bad ends then it seems
clear that not only should the results not be published, but also they should be
tightly controlled. Michael Selgelid26 has raised the question of the censorship
of research in regard to dual-use questions, and has advocated a moderate and
balanced position. The present account will certainly suggest stricter controls
(perhaps another unwelcome implication). From the perspective of common
morality, agents are free to do whatever they wish, as long as this does not
harm others. Thus, scientists are free to do whatever research they like, but
the freedom to research, or research itself, is not given special value or status.
When there is a risk that research will cause harm then it is firmly proscribed.
And when research also has benefits then it should be conducted with effort to
minimise harms, and if this entails censorship then that is what is required for
the project to go ahead. Indeed, it seems that this is precisely the view that makes
the most sense. The response that the whole purpose of publication is to make
ideas and results open to all members of the community who can then build on
them will not do here. Networks of respected colleagues can be informed, but
the risks of open publication may be too great with dual-use research.
Conclusion
I conclude that the account of science and responsibility put forward in
The Responsible Scientist has relevance for the ethics of dual-use research, and
in two sorts of ways. In the first place, the theory of responsibility ties the
scientist to the outcomes of his or her work, and does so more tightly than
other viewpoints. For instance, an account that incorporated the straight rule
and did not see any special sort of forward-looking responsibility for scientists
would have quite different implications; however, I think that the topic of dual
use shows that we want a wide-ranging account of responsibility, and we want
researchers to think carefully about what they do. The second way in which the
account is relevant is in the way it sees responsibility as being discharged or
cashed out. This is with respect to the no-harming ethic of common morality.
I suspect that this will be the more controversial aspect of the present proposals.
26 Selgelid, M. 2007, A tale of two studies; ethics, bioterrorism, and the censorship of science, Hastings
Center Report, vol. 37, no. 3, pp. 3543.
132
9. An Expected-Value Approach to
the Dual-Use Problem
Thomas Douglas
In this chapter I examine how expected-value theory might inform responses
to what I call the dual-use problem. I begin by defining that problem. I then
outline a procedure, which invokes expected-value theory, for tackling it. I first
illustrate the procedure with the aid of a simplified schematic example of a
dual-use problem, and then describe how it might also guide responses to more
complex real-world cases. I outline some attractive features of the procedure.
Finally, I consider whether and how the procedure might be amended to
accommodate various criticisms of it. The aim is not to defend the procedure
in its original form, but to consider how far we must deviate from it to evade
the criticisms that I consider. I seek to show that, though it is necessary to
make substantial ammendments to the procedure, we need not eschew a role for
expected-value theory altogether. Even if one accepts the criticisms I discuss, it
is possible to defend a role for expected-value theory in responding to dual-use
problems.
The dual-use problem

Scientific work can typically be used in both good and bad ways, and sometimes
it is difficult or impossible for a scientist to prevent the bad applications without
also forestalling the good ones. For example, if a scientist publishes her findings,
thus enabling good applications of those findings, she may then have little power
to prevent bad applications of the same results. Scientists may thus face a tradeoff between enabling certain good applications of their work and preventing
certain bad ones. In a few cases, the risk of bad applications may be so high
that there is a genuine dilemma about how to resolve this trade-off. Consider the
following hypothetical scenario:
Its 2030. Designer viruses are now used to treat some cancers and
infectious diseases. But theyre expensive and difficult to manufacture.
You discover a new, cheap way to produce synthetic viruses using outof-date benchtop DNA synthesisers that are now ubiquitous, even in
developing countries. Youre excited about the discovery and hoping to
publish it in Nature. You think it could bring a wide range of medical
treatments, not to mention research tools, within the grasp of the
133
developing world. However, theres a catch: every major military and

terrorist group in the world has access to these obsolete synthesisers. It
would take only one malevolent agent and one such machine to produce
enough vaccine-resistant smallpox virions to devastate humanity.1
Arguably, it is genuinely unclear in this case whether you ought to publish your
discovery. If this is so, you may be faced with what has been termed the dualuse dilemma.
As a tentative starting point, we might define the dual-use dilemma as the
quandary that arises whenever: 1) a scientist faces a choice about whether or
not to produce or disseminate some scientific output, such as a piece of scientific
knowledge, or a physical technology; 2) that output could be used in both good
and bad ways, and the agent is not in a position to altogether eliminate the risk
of bad use without also reducing the likelihood of good use; 3) there is therefore
a trade-off between enabling the good application(s) and preventing the bad
one(s); and 4) it is consequently unclear whether producing the scientific output
is the right thing to do, morally speaking.
This definition captures some commonplace ideas about dual-use dilemmas;
however, there are, I think, good reasons to broaden the definition somewhat
before embarking on any ethical analysis.
First, though the dual-use dilemma is typically presented as a problem that
arises for scientists and in relation to particular scientific projects, similar problems
may arise for others in a position to support or impede particular projects,
and for those faced with decisions about what general stance or policy to take
on scientific work, or some particular class of scientific work. For example, a
policymaker considering whether to institute censorship of scientific journals
may also face a trade-off between enabling good applications of scientific
findings and preventing bad ones, though in this case the choice does not relate
to any particular scientific project. Given the similar structure of this kind of
ethical problem to paradigmatic examples of the dual-use dilemma faced by
individual scientists, it seems sensible to consider them together.
Second, as defined above, the dual-use dilemma arises because of a tension
between two ethical considerations: reasons to enable good applications of some
scientific output, and reasons to prevent bad ones. But in real-world cases, there
will be a number of other ethical considerations bearing on decisions relating to
the production or dissemination of scientific outputs.2 For example, a scientist
1 Adapted from Douglas, T. and Savulescu, J. 2010, Synthetic biology and the ethics of knowledge, Journal
of Medical Ethics, vol. 36, no. 11, pp. 68793.
2 Buchanan, A. and Kelley, M. 2013, Biodefense and the production of knowledge: rethinking the dual-use
problem, Journal of Medical Ethics, vol. 39, pp. 195204.
134
9. An Expected-Value Approach to the Dual-Use Problem
could have a reason to pursue scientific knowledge about some phenomenon

regardless of how that knowledge is likely to be used; she could have a reason
to pursue the knowledge because it would be intrinsically valuable, or simply
because she previously promised to do so. Similarly, a policymaker could have
reasons (not) to censor scientific publications that are independent of how the
censored information would have been usedfor example, she might have
reasons not to censor the information in order to respect freedom of speech. An
adequate formulation of the dual-use dilemma should accommodate the possible
existence of reasons besides those normally taken to generate the dilemma.
Third, according to the above definition, the dual-use dilemma involves making
a choice between exactly two alternatives. This feature is suggested by the term
dilemma. But it might be thought that such binary choices are in fact rather
rare. Often when we think there are only two alternatives on the table there
are actually more. For example, a scientist may think she is faced with a choice
between either publishing her research in full or not, when in fact there may
be several other optionsfor example, publishing elements of the research,
or publishing the research conditional upon its distribution being restricted.
Similarly, a policymaker may believe that there is a simple choice between
censoring scientific journals and not, when in fact there are many different
types of censorship (for example, self-censorship by journal editors, censorship
by peer reviewers, censorship by external agencies). It is tempting to say that,
when presented with a dual-use dilemma, an agent should in the first instance
try to find a way to escape the dilemma by finding some third, superior course
of action. Only if this is impossible should he grasp one horn of the dilemma.
Recognising the existence of further alternatives, however, does not necessarily
resolve the problem. Strictly speaking, when there are more than two alternatives
available, there can be no dilemma. However, there may still be a quandary, since
alternatives more likely to enable good applications may also be more conducive
to bad applications or be more problematic in other ways (for example, they
may be more resource intensive). We can imagine a case in which the more likely
an option is to allow good applications relative to alternative options, the more
likely it is to also enable bad applications. It may thus remain unclear which
option to select.
I wish to include within the scope of the discussion that follows ethical problems
that are similar to paradigmatic dual-use dilemmas but are not faced by
individual scientists, do not relate to specific scientific projects, involve choices
between more than two alternatives, and involve ethical considerations besides
reasons to enable good and prevent bad applications of scientific outputs. I will
do this by introducing the broader idea of a dual-use problem. I define this as the
quandary that arises whenever: 1) an agent faces a choice between two or more
alternatives that will influence the production and/or dissemination of some
135
scientific output, such as a piece of knowledge or a physical technology; 2) that

scientific output could be used in both good and bad ways; 3) there is a tradeoff between enabling the good application(s) and preventing the bad one(s),
since any alternative that increases the likelihood of the good use, relative to
some alternative, also increases the likelihood of the bad use; and 4) given this
trade-off and other normative considerations bearing on the choice, it is unclear
which alternative(s) it would be morally right for the agent to take.
Introducing an expected-value approach

Having just suggested that the dual-use problem should be understood broadly,
I am now, temporarily, going to narrow down on a simplified example of a
dual-use problem, so that we have something more tractable to work with. The
example is a schematic version of a putative dual-use problem faced recently by
scientists working on H1N1 avian influenza.3
Suppose that a scientist is deciding whether to embark, with her assistants,
on a research project that will investigate whether it is possible to mutate the
H1N1 virus so as to make it transmissible by air between humans. (Assume
that existing variants of the virus cannot be transmitted in this way.) A likely
outcome of the project is that the scientists will indeed discover a method for
rendering the virus air-transmissiblethat is, that will acquire knowledge
that would enable others to create such a virus. This knowledge could, let us
suppose, be used to develop a vaccine against the air-transmissible virus that
will save 5000 lives, or to facilitate a highly lethal act of biological terrorism in
which an air-transmissible variant of H1N1 is released and kills 100 000 people.
Suppose (implausibly) that these are the only two possible applications of the
knowledge, and that there are no other moral considerations bearing on the
scientists decision whether to embark on the project. To further simplify the
case, suppose that the development of the vaccine and the lethal biological attack
are events that can occur at most once each. Finally, suppose also that knowledge
about how to render H1N1 air-transmissible will certainly not be produced by
anyone other than the scientist and her team, so that if the scientist decides not
to pursue the project, the lethal H1N1 attack will certainly not occur, but nor
will a vaccine against the air-transmissible variant of H1N1 be developed.
The scientist is faced with the choice of taking a gamble. She can gamble on
bringing about the good outcomethe development of the vaccinebut it is
possible that the bad outcome, the lethal attack, will occur instead or as well.
Taking the gamble generates four possible future states of the world
3 See, for a brief description of the case, Evans, N. G. 2013, Great expectationsethics, avian flu and the
value of progress, Journal of Medical Ethics, vol. 39, no. 4, pp. 20913.
136
state I: vaccine developed; no bioterrorist attack

state II: bioterrorist attack; no vaccine developed
state III: vaccine developed and bioterrorist attack
state IV: no vaccine developed; no bioterrorist attack.
If the scientist does not take the gamble, state IV is guaranteed.
Assuming (again, implausibly) that the values of these states of the world can
be assessed simply by tallying up the number of lives saved by the vaccine and
subtracting the number of lives lost in a bioterrorist attack, these states can be
assigned the following values
state I: 5000 lives saved 0 lives lost = 5000
state II: 0 lives saved 100 000 lives lost = 100 000
state III: 5000 lives saved 100 000 lives lost = 95 000
state IV: 0 lives saved 0 lives lost = 0.
Suppose that the probability of developing a vaccine is rather high if the
project goes aheadsay, 0.5 (50 per cent). On the other hand, suppose that the
probability of the devastating bioterrorist attack, if the project goes ahead, is
lowsay, 0.05 (5 per cent). Then the probabilities of each of these states coming
about will be
state I: 0.5 x 0.95 = 0.475 (47.5 per cent)
state II: 0.5 x 0.05 = 0.025 (2.5 per cent)
state III: 0.5 x 0.05 = 0.025 (2.5 per cent)
state IV: 0.5 x 0.05 = 0.475 (47.5 per cent).
States II and III are both very bad. In terms of lives saved and lost, they are more
bad than states I and II are good. This counts in favour of abstaining from the
scientific project; however, given the probabilities specified above, states I and
IV are much more likely to come about than states II and III. This militates in
the opposite direction. How should A decide whether to take the gamble? One
approach would be to calculate the expected value of the gamble, and proceed
only if it exceeds the expected value of the alternativenamely, obtaining state
IV with certainty. (Readers familiar with expected value theory may which to
skip to the next section, Complicating the Picture.)
The expected value of an individual state is given by multiplying the value
of that state by its probability of occurring. The expected value of a gamble
over multiple possible states is given by summing the expected values of
137
the individual states. We can think of the expected value of the gamble as a
weighted average of the values of the individual states that may result, with the
weights given by the probabilities of those states coming about.4
To illustrate, suppose that you are considering whether to engage in a gamble
that involves tossing a coin. If the coin turns up heads, you win $10, if it turns
up tails, you lose $5. The probability of a heads is 0.5, and likewise for tails. If
you dont take the gamble, you win or lose nothing. Thus, the expected value of
the gamble will be given by
probability(heads).value(heads) + probability(tails).value(tails)
= 0.5 x $10 + 0.5 x $5
= $2.50.
The expected value of this gamble is $2.50. On the other hand, the expected
value associated with declining the gamble is $0, as, with certainty, one wins
nothing and loses nothing. Since the expected value of the gamble exceeds that
of the alternative, the expected-value approach under consideration will advise
you to take the gamble.
Now lets return to our schematic dual-use problem. The gamble faced by the
scientist has four possible outcomes: I, II, III and IV. So the expected value of
the gamble, Vgamble, will be given by
Vgamble = p(I).v(I) + p.(II).v(II) + p(III).v(III) + p(IV).v(IV)
where p(X) is the probability of state X and v(X) is the value of state X. (A
difference with the coin toss example, however, is that in this case, the values of
the states are supposed to reflect their overall value, not simply their value to the
agent, in this case, the scientist. Our question is not whether the scientist ought,
from a self-interested point of view, to undertake the research, but whether it
would be morally right for her to do so. Insofar as the value of an outcome bears
on the moral rightness of the action that brings it about, it is usually thought to
be the overall value that matters.)
Plugging in the values and probabilities that we assigned to these gambles, we
get the following result:
Vgamble = 0.475 5000 + 0.025 100 000 + 0.025 95 000 + 0.475 0

138
= 2500.
4 Some use the term expected value in a way that presupposes that the values of individual states are
measured in monetary terms. Expected value can then be contrasted with expected utility, which is instead
determined by the amount of utility contained in each state of the world. I use expected value in a way
that is neutral between different metrics of value such as monetary value, utility and health. I thus regard
expected utility as a species of expected value.
That is, an expected 2500 lives will be lost, on net. On the other hand, the expected
value of the alternativeobtaining state IV with certaintyis given by:
Valt = v(IV)

= 0.
The approach under consideration would advise taking the gamble if and only
if Vgamble exceeds Valt, and against taking the gamble if and only if Valt exceeds the
gamble. Given the probabilites and values that we have used, Valt exceeds Vgamble
so the approach will recommend against taking the gamblethat is, against
pursuing the scientific project.
Complicating the picture

The schematic dual-use problem set out above was highly simplified, but the
same basic approach could be taken to more complicated, actual dual-use
problems. Some revisions and qualifications will, however, be necessary.
For example, in actual cases, there will typically be some chance that the
scientific output in question will come about, and be used in good and/or bad
ways, even if the agent in question does not produce it: someone else might do
so. In producing the output, then, the agent is not replacing a certain outcome,
in which there is no chance of the good and bad applications in question, with
a gamble in which these outcomes might occur. Instead, she is replacing one
gamble with another. To accommodate this, we will need to amend the expectedvalue approach so that it compares the expected values of two different gambles.
In fact, in most cases, since there will be many different actions open to the
agent, we will have to compare the expected values of multiple gambles: there
will be a different gamble posed by each alternative.
In addition, each gamble will typically involve many more possible outcomes
than in the simplified example set out above. In an actual case where a scientist
is considering whether to try to develop an air-transmissible variant of the
H1N1 virus, there will be a variety of different ways in which the resulting
knowledge, if the project is successful, might be used. In addition to being used
to develop a vaccine against the new variant or to facilitate a bioterrorist attack,
it might also be used, for example, to publicly demonstrate the ease with which
terrorists might create a biological weapon, thus encouraging politicians to take
further steps to protect against such an attack, or to advance the understanding
of what makes viruses air-tranmissible in a way that allows the development of
vaccines for existing air-transmissible diseases. Moreover, there will typically
be many different forms that each of these kinds of outcome could take. For
139
example, there are many different levels of severity (in terms of both lives lost and
other negative consequences) that a bioterrorist attack could have. In addition,
though we assumed above that each outcome would occur either once or not at
all, in actual cases many of the important good and bad applications of scientific
output could occur multiple times. Finally, the development of a technology or
piece of scientific knowledge might have good and bad effects that are unrelated
to how it will subsequently be used (for example, one good effect might be
that there is now more intrinsically valuable knowledge; one bad effect might
be that valuable research funding is consumed).5 These complications all serve
to greatly multiply the number of permutations of different states of the world
that could result from the scientists decision, compared with the simplified
dual-use problem outlined above.
Moving from schematic examples to the real world would thus greatly
complicate any attempt to apply expected-value theory in order to resolve dualuse problems. However, it does not obviously raise any in principle difficulties.
One can still envisage an approach to such problems that would: 1) identify the
possible outcomes of each alternative course of action; 2) identify the possible
states of the world, each consisting of combinations of these outcomes;6 3)
explicitly assess the values and probabilities of these states of the world; 4)
use these probabilities and values to calculate the expected value associated
with each alternative; and 5) select the alternative associated with the highest
expected value. I will call this approach the expected-value procedure or EVP.
Strengths of the expected-value procedure

The EVP has several attractive features.
First, it captures some commonsense intuitionsfor example, that decisions
should be made by weighing the upsides and downsides of the various choices,
that upsides and downsides should be given equal weight, and that, other things
being equal, one should prefer an alternative with more valuable upsides, more
likely upsides, less disvaluable downsides, or less likely downsides.
Second, the expected-value approach is consistent with a major school of ethical
thought: consequentialism. Consequentialism is often taken to hold that an
5 The outcomes that I describe here would perhaps not normally be described as effects of the action
since their relationship to the action is arguably constitutive rather than causal. There is, however, arguably
nothing to prevent us from thinking of them as effects in order to include them within the scope of the
expected-value approach.
6 I leave it open whether, in identifying possible outcomes and states of the world, an agent applying
this procedure should seek to list all possible outcomes/states, the most likely outcomes/states, all reasonably
foreseeable outcomes/states or some other subset of possible outcomes/states.
140
act is morally right if and only if its consequences will be at least as good as
those of any alternative action. One variant of consequentialism holds that an
actions actual consequences are what determine its rightness. Others hold that
the actions foreseen, foreseeable or likely consequences are what are important.
Proponents of these latter variants typically hold that the goodness of a set
of foreseen/foreseeable/likely consequences is determined by its expected value.
Their variants of consequentialism thus imply that an action is right if and only
if it is associated with at least as much expected value as any alternative. Thus,
if one applies the expected-value procedure outlined above when faced with a
dual-use problem, one can think of oneself as explicitly trying to identify and
adopt the course of action that is morally right according to some variants of
consequentialism.7
Third, the expected-value approach has a track record. The methodology used
in the EVP is a methodology that has been widely employed, often under the
label risk analysis, risk management or costbenefit analysis, to inform
major corporate and government decisionsfor example, decisions between
alternative power generation projects or mining operations.8
Criticisms of the expected-value procedure

Many criticisms of expected-value theory, of its application to major public
and private decisions, and of consequentialism might also be adduced against a
proposal to use the EVP to confront dual-use problems. In this section I discuss
six such criticisms, in each case considering whether and how it might be
possible to ammend the EVP to accommodate the criticism.
My aim is not to defend the EVP tooth and nail, by rebutting all objections that
might be raised against it. Indeed, I do not comment on whether the criticisms I
discuss are persuasive. Rather, the aim is to explore how far we must deviate from
the procedure in order to evade these criticisms assuming they are persuasive.
My conclusion will be that we have to deviate from it to a rather great degree;
the approach I end up defending bears little resemblance to the EVP. However,
it does, I will suggest, retain at its heart an important role for expected-value
theory.
7 Note that consequentialists differ on what makes a given consequence good or valuable. The most wellknown variety of consequentialismutilitarianismholds that the goodness of a set of consequences is
determined by the total amount of welfare or wellbeing that it contains, but other variants of consequentialism
allow that other things besides wellbeing may be of value. For example, a fairer distribution of wellbeing may
be better than a less fair one.
8 For an early, but somewhat controversial, application of expected-value theory to the assessment of risks
from nuclear power generation, see Rasmussen, N. C. (chair) 1975, Reactor Safety Study: An Assessment of
Accident Risks in U.S. Commercial Nuclear Power Plants, WASH-1400, US Nuclear Regulatory Commission.
141
This strategy is motivated by the thought that some authors have, in other areas,
been too quick to move from the thought that a straightforward expected-value
approach faces fatal problems to the claim, often implicit, that expected-value
theory is not relevant at all to the problem at hand. Insofar as my argument
succeeds in showing that, at least in relation to the dual-use problem, one can
accommodate the most important objections to a straightforward expectedvalue approach without giving up entirely on expected-value thinking, I hope I
will have established that such a swift rejection of expected-value theory is not
justified in relation to the dual-use problem.
1. No adequate metric
An initial criticism of the EVP focuses on the need, in that procedure, to evaluate
various possible states of the world. It denies that there is any adequate metric
of value for making such evaluations.
In order to apply the expected-value procedure, we will need a cardinal metric
of value that can be used to evaluate the various states of the world that might
follow from different responses to a dual-use problem.9 Moreover, we will want
this metric to assign a value to every state of the world that is taken as an
input into the EVP, and to capture all of the morally significant features of each
state. Arguably, there is no such metric. One possibility would be to find some
cardinal metric of individual wellbeing. We could then sum the wellbeing of
all individuals in each state of the world and use this measure of aggregate
wellbeing to compare different states of the world. A problem, however, is that
it is controversial whether there is a common cardinal scale on which we can
measure and sum the wellbeing of different people. Some deny that there is any
cardinal metric that allows the wellbeing of different people to be measured in a
reliable way,10 others question whether interpersonal comparisons of wellbeing
have any meaning at all, implying that, as a matter of principle, there could
be no such metric.11 In addition, an aggregate wellbeing metric might fail to
capture some morally significant features of each state of the world. Some would
argue that, holding aggregate wellbeing constant, differences in the distribution
of wellbeing across individuals are morally significant. For example, some argue
that it is worse if a fixed amount of wellbeing is distributed less equally, justly
9 A cardinal metric is one that preserves orderings uniquely up to linear transformations. In a cardinal
metric, the interval between two points on the scale has a consistent meaning: it means the same regardless of
where those points fall on the scale.
10 See, for example, Jevons, W. S. 1970 [1871], The Theory of Political Economy, Penguin Books,
Harmondsworth, UK, Introduction.
11 See, for example, Robbins, L. 1935, An Essay on the Nature and Significance of Economic Science, 2nd edn,
Macmillan, London; Arrow, K. J. 1963 [1951], Social Choice and Individual Values, 2nd edn, Yale University
Press, New Haven, Conn., p. 9. Robbins allows that interpersonal comparisons of wellbeing have meaning as
disguised normative judgments, but denies that they have any descriptive meaning.
142
or fairly.12 Similarly, others would argue that, in assessing the value of a given
state of the world, it matters not just what distribution of wellbeing it contains,
but also how that distribution came about.
2. Empirical and evaluative uncertainty

A second criticism targets both the evaluation of different states of the world
and the assignment of probabilities to them.
In actual settings where dual-use problems arise, there will often be significant
uncertainty about the probabilities and values of possible outcomes, and indeed
about what outcomes are even possible. For example, we may simply lack any
good evidence of how likely a project investigating whether to render the H1N1
virus air-transmissible between humans is to succeed in realising that goal.
Similarly, we may lack any good evidence of how likely it is that knowledge
about such a virus would be used to produce biological weapons, of how likely
it is that such weapons would be deployed, and of what the likely effects of their
deployment would be. Finally, as an example of evaluative uncertainty, we may
lack evidence on how much disvalue the effects of a terrorist attack, even if they
could be fully specified, would have. Since, when faced with such uncertainty,
we will simply have to make imperfect estimates of probabilities and values, our
expected-value calculations will often give incorrect results.
Accommodating objections 1 and 2

Though both of these problems are serious and may give us some reason to
deviate from the EVP, it can be argued that neither of these concerns gives us
a clearly decisive reason to reject the use of expected-value theory altogether.
In response to the first of these concerns, it can be argued that, even if there is
no perfect value metric for employing in the expected-value procedure, there
are nevertheless adequate ones. Perhaps there is no cardinal metric of value that
captures all morally significant features of all of the different states of the world
that we wish to compare. But there are cardinal metrics that we could use to
evaluate some of the morally important features.
One approach, then, would be to simply retain the EVP and employ the best
cardinal metric that we have. We might thus fail to capture some morally
significant considerations, but we would quantitatively balance as many
considerations as it is possible to quantitatively balance.
12 See, for example, Temkin, L. 1993, Inequality, Oxford University Press, New York; Persson, I. 2008, The
badness of unjust inequality, Theoria, vol. 69, nos 12, pp. 10924.
143
It might be objected that this strategy would give disproportionate weight

to outcomes that can be rated on a cardinal scale at the expense of morally
important considerations that cannot be rated in this way. It would allow
cardinally measurable considerations to fully determine what course of action to
take, with other considerations given no influence at all; however, this problem
could be avoided by modifying the EVP so that the expected values that one
calculates on the basis of ones cardinal metric are treated as only an imperfect
indicator of what course of action to pursue. We could proceed by calculating
expected values using the best cardinal metric that we have, and then use the
values produced as an imperfect indicator of which course of action to take,
though we might sometimes wish to choose a course of action associated with
lower expected value than another because we believe it has virtues that are not
captured by the metric we have used.
Thus, suppose we wish to compare the value of a world in which a robust vaccine
for air-transmissible H1N1 is discovered, but there is also a major bioterrorist
attack, with a world in which neither of these things happens. I suggested
above that the scientist could compare the value of these with states of the
world by tallying up the number of lives lost and saved. In fact, there are likely
to be better cardinal measures available. For example, she could instead look at
the total number of quality-adjusted life-years (QALYs) that would be enjoyed
by the whole population in each possible state of the world resulting from each
of the available courses of action. This would allow her to capture at least some
of the possible effects of her actions on morbidity, as well as mortality. Clearly,
the QALYs measure would still be imperfect and would not capture many
evaluatively important effects of the two alternatives. For example, it would
not capture non-health-related outcomes such as the inconvenience and loss
of privacy caused by security measures that would be introduced as a result
of a major terrorist attack. But a determination of the expected level of QALYs
associated with each course of action nevertheless plausibly provides some
indication of which of these courses of action the scientist should pursue. The
scientist could use expected QALYs as an indicator of whether to pursue the
H1N1 research, and then consider whether factors not captured by this measure
would justify deviating from the course of action that it suggests. Exactly how
these other factors would need to be taken into account would of course need to
be determined. It is not clear how we should weigh considerations that cannot
be quantified using a shared cardinal measure. But this problem is not specific
to attempts to use expected-value theory to respond to dual-use problems. If
there are factors that cannot be rated on a shared cardinal measure, any proposal
for how we ought to respond to dual-use problems will need to determine how
these should be taken into account.
144
The concern about uncertainty can, I think, be accommodated in a similar way.

As with the concern about metrics, this worry does not take issue with the basic
idea underpinning EVPthe idea that our choices when confronted with dualuse problems should reflect the expected value of the available alternatives.
Rather, it points out a practical difficulty with trying to implement this idea.
But it remains plausible that trying to calculate the relevant expected values
would be desirable. We could calculate expected values for the available courses
of action using our best estimates of the values and probabilities of the relevant
outcomes, insofar as it is possible to assign values or probabilities at all. We
could then decide what course of action to pursue on the basis of these estimates
of expected value.
It might be argued that this approach would yield implausible results in some
cases. It is possible that two courses of action could have the same expected
value, though the probabilities and values of the outcomes associated with one
course of action are highly uncertain, and the probabilities and values associated
with the other course of action are certain. Arguably, in this case one should
prefer the latter course of action.
But again, this problem could be accommodated by treating the expected values
as only an imperfect indicator of what to do. This would retain some role for
expected values, but would allow us to favour a course of action with a lower
expected value than another because the other course of action could cause
outcomes whose probability or value is less certain.
The concerns about the lack of a cardinal metric and uncertainty suggest that
the EVP may not be able to capture all of the considerations that are relevant
to how one should act when faced with a dual-use problem: it does not capture
outcomes that cannot be evaluated on a common cardinal scale, nor does it
capture differences in the certainty of the values and probabilities of the possible
outcomes. It is, however, still plausible that we should adopt a procedure
that incorporates the most important elements of the EVP. More specifically,
it remains plausible that we should conscientiously calculate expected values
using the best metric available and take this as at least an imperfect indicator of
what to do. Call this the restricted expected value procedure or rEVP.
There are, however, two further criticisms that suggest it might be unwise even
to adopt the rEVP.
3. Bias
In attempting to identify and assign probabilities and values to different
outcomes, agents might have systematic tendencies to neglect or exaggerate
some considerations. For example, people might overstate the disvalue of
145
negative outcomes relative to the value of positive outcomes, or they might

exaggerate the probability of the most extremely valuable or disvaluable events
occurring. In some cases, the direction of a bias may depend on the particular
circumstances and psychology of the person(s) applying the rEVP. For example,
people may be inclined to relatively exaggerate the (dis)value or probability of
outcomes that they have experienced in the recent past, and so are associated
with readily available mental images.13 Or they might be prone to exaggerate
the (dis)value or probability of outcomes that will primarily affect them rather
than others (self-serving bias).14 They may also be prone to underestimate the
disvalue of harms that affect large numbers of people (scope insensitivity)15 and
susceptible to framing effects in which the way an outcome is presented affects
the probability or value assigned to it.16 Given these biases, the application of
the rEVP could be expected to give misleading results in many casesthat is,
the expected values that it takes as indicators may often be incorrect.
4. Demandingness
Applying the rEVP comprehensively and conscientiously might require
substantial time, effort, expertise and financial resourcesall resources that
could otherwise be spent on other worthwhile activities. It might also be
psychologically burdensome in the sense of requiring those who apply it to
override their natural inclinations (suppose that a journal editor strongly
committed to academic freedom decides, on the basis of the rEVP, that she
must heavily censor a submitted article). Applying the rEVP might, due to
the presence of such costs, have negative overall consequences even if it leads
decision-makers to make the best choices. The benefits of making the best
decisions might be outweighed by the costs associated with the means via
which those decisions were reached.
13 Lichtenstein, S., Slovic, P., Fischhoff, B., Layman, M. and Combs, B. 1978, Judged frequency of lethal
events, Journal of Experimental Psychology: Human Learning and Memory, vol. 4, no. 6, pp. 55178; Sunstein,
C. 2007, Worst-Case Scenarios, Harvard University Press, Cambridge, Mass., p. 6.
14 Babcock, L., Loewenstein, G., Issacharoff, S. and Camerer, C. 1995, Biased judgments of fairness in
bargaining, American Economic Review, vol. 85, no. 5, pp. 133743.
15 Desvousges, W. H., Johnson, F. R., Dunford, R. W., Boyle, K. J., Hudson, S. P. and Wilson, N. 1993,
Measuring natural resource damages with contingent valuation: tests of validity and reliability, in J. A.
Hausman (ed.), Contingent Valuation: A Critical Assessment, North-Holland, Amsterdam, pp. 91159;
Fetherstonhaugh, D., Slovic, P., Johnson, S. and Friedrich, J. 1997, Insensitivity to the value of human life: a
study of psychophysical numbing, Journal of Risk and Uncertainty, vol. 14, pp. 238300.
16 Tversky, A. and Kahneman, D. 1981, The framing of decisions and the psychology of choice, Science,
vol. 211, no. 4481, pp. 4538.
146

The problem of bias is likely to be a problem for many approaches besides the
rEVP. Any plausible approach to dual-use problems will require us to identify,
evaluate and assess the probability of at least some possible outcomes. Some
may permit us to do so in an implicit, subconscious way rather than an explicit,
conscious wayfor example, rather than adopting the rEVP we could simply
have experts make decisions based on their intuitions about the case at hand.
This does not, however, necessarily render such approaches less susceptible to
bias. Indeed, it may render them more susceptible. The process of assigning
values and probabilities may make our biases more obvious and allow us to
correct for them. Nevertheless, it may be possible to reduce the risk of bias by
deviating from the rEVP. Suppose it were known that people quite generally
tend to understate the probability of extremely bad states of the world, even
when they are aware of this and attempt to correct for it. We could to some
extent compensate for this bias by making some changes to the rEVP. We could
require, for example, that the most disvaluable states be given greater weight
than other states even if we judge their probability of occurring to be the same.17
Similar thoughts apply to the problem of demandingness. Again, this problem
is likely to apply to many approaches besides the rEVP; however, it may be
possible to reduce the burdens of deciding between alternative strategies by
deviating from the rEVP. One approach would be to apply simple heuristics (for
example, unless there is a clear and present risk of misuse, proceed as if there
were no risk), and resort to the rEVP only in hard casesfor example, cases that
are not covered by the simple heuristics, or cases in which heuristics conflict.
Alternatively, the problem of demandingness could be mitigated by ensuring
that the rEVP is applied by those able to apply it at least cost. For example,
it may be less burdensome for a committee of scientific, health and security
experts to assess the values and outcomes of synthetic biology research than it
is for individual scientists to do the same. Regulatory bodies could be charged
with applying rEVP, whilst individual scientists follow a simple heuristic such
as aggressively pursue your scientific goals unless this violates a regulation.
Concerns about bias and demandingness may justify deviation from the rEVP. But
note that neither justifies a wholesale rejection of the basic ideas underpinning
that procedure. In particular, both are consistent with retaining the idea that
the right course of action to take, when faced with a dual-use problem, is (the) one
associated with at least as much expected value as any available alternative (call
this the expected-value criterion or the EVC). The concerns simply show that
the procedure by which we make decisions when faced with dual-use problems
should not necessarily be to explicitly calculate expected values. Explicitly
17 This might, for example, involve applying a multiplier (> 1) to the most negatively valued outcomes.
147
attempting to satisfy the expected-value criterion might not be the best way
of in fact satisfying that criterion, or might involve costs that outweigh the
benefits of satisfying it.
This distinction between criteria of rightness and decision procedures has long
been emphasised by consequentialists when presented with concerns about
demandingness and bias and other related objections. Consequentialists have
argued that their theory provides an abstract description of which acts are
right and which are not. It does not provide a procedure via which to decide
how to act.18 In fact, consequentialism is consistent with adopting some other
ethical theory as a procedure for guiding decisions. In an ideal world where
we responded to our evidence in a perfectly rational and costless way, we could
simply apply consequentialism as a decision procedure. But the actual world
is not like that. In the actual world it may, according to consequentialists, be
best not to make decisions by explicitly calculating which act will have the best
consequences.
Drawing a distinction between criteria of rightness and decision procedures
allows us to retain the basic idea underpinning EVP (and thus rEVP) while
deviating from these procedures. More importantly, it provides us with a
higher standardthe EVCagainst which we can measure alternative decision
procedures. Other things being equal, we should prefer a decision procedure
that comes closer than some alternative procedure to recommending the courses
of action that are right according to the EVCthe courses of action associated
with the highest expected value. In an ideal world, the best decision procedure
would be the EVP or rEVP. In the actual world, it may be some amendment of
these procedures, or even a wholly different approach. Either way, it is plausible
that we should keep the EVC in mind as a standard by which to judge competing
decision procedures.
I now turn to consider two criticisms that suggest that even the EVC will need
to be rejected.
5. Rational risk aversion

The expected-value procedure, and the criterion that underpins it, is blind to
the distribution of value across possible states of the world. Suppose that a
18 See, for example, Austin, J. 1954 [1832], The Province of Jurisprudence Determined, H. L. A. Hart (ed.),
Weidenfeld, London, p. 108; Mill, J. S. 1985 [1861], Utilitarianism, in Collected Works. Volume X, John M.
Robson (ed.), University of Toronto Press, Toronto, ch. 2, pp. 20360; Sidgwick, H. 1907, The Methods of
Ethics, 7th edn, Macmillan, London, p. 413; Bales, R. E. 1971, Act-utilitarianism: account of right-making
characteristics or decision making procedure? American Philosophical Quarterly, vol. 8, pp. 25765; Parfit,
D. 1984, Reasons and Persons, Clarendon Press, Oxford, pp. 249, 3143; Railton, P. 1984, Alienation,
consequentialism, and the demands of morality, Philosophy and Public Affairs, vol. 13, pp. 13471, at pp.
1658. The terminology decision procedure and criterion of rightness is due to Bales, op. cit.
148
scientist is considering whether to publish a scientific finding that has a 50 per

cent chance of being used in a good way and a 50 per cent chance of being used
in a bad way. Suppose further that the value of the good outcome is 100, and of
the bad outcome is 100. Then there will a 25 per cent chance of realising a state
of the world with a value of 100 (good outcome, no bad outcome), a 25 per cent
chance of realising a state with a value of 100 (bad outcome, no good outcome),
and a 50 per cent chance of realising a world with a value of zero (because both
the good outcome and the bad outcome occur or neither occurs). Suppose that
the scientist does not publish the result: there is a 100 per cent probability that
neither the good nor the bad outcomes will occur. In this case, the expected
value associated with publishing the result is zero, as is the expected value
associated with not publishing it. Thus, the EVC will hold that both courses
of action are right. But, arguably, the right thing to do in this case is to abstain
from pursuing the project since this alternative is associated with lower (indeed
zero) risk: if the scientist abstains from the project, a world with a value of 0 will
obtain with 100 per cent probability, whereas if she pursues the project there is
an expected value of zero, but a significant risk of winding up in a world whose
value is 100. If we ought to be risk averse, as some have argued,19 we should
reject EVC in favour of an alternative criterion that penalises riskier alternatives.
6. Agent-relativity
The EVC is also blind to the way in which an agent posed with a dual-use
problem contributes to a good or a bad outcome; it focuses only on the probability
and value of the outcome. But, according to many nonconsequentialist ethical
theories, an agents relation to an outcome is also important. Consider a case in
which a scientist performs and publishes some piece of research in synthetic
biology that is intended to develop a new cure for cancer but could also be
misused in unjustified biowarfare. It might be argued that this possible negative
outcome should be discounted relative to at least some of the other outcomes
because
1. It is not caused by the scientist, but is merely allowed to occur. In contrast,
at least some positive outcomes of the researchfor example, the production
of intrinsically valuable knowledgemight be said to be caused by the
scientist.20
2. It is not intended, but is merely foreseen, by the scientist. By contrast, the
possible positive outcome of developing a cure for cancer is intended.21
19 See, for example, Hansson, S. O. 2003, Ethical criteria of risk acceptance, Erkenntnis, vol. 59, no. 3, pp.
291309.
20 This claim could be grounded on the doctrine of doing and allowing.
21 This claim could be grounded on the doctrine of double effect. See Suzanne Uniackes Chapter 10, in this
volume.
149
3. The occurrence of the bad outcome depends on the subsequent immoral

actions of another agent: the person who actually engages in unjustified
biowarfare. Thus, even if the scientists actions resulted in unjustified
biowarfare, the scientist would not be the primary wrongdoer, but would at
most be an accomplice to the wrongdoing. By contrast, some other outcomes
of the work, such as the production of intrinsically valuable knowledge, may
occur without requiring the intervention of other moral agents. The scientist
might be said to be the principal agent implicated in bringing about these
outcomes.

If these criticisms are well founded, the EVC should be rejected since the right
strategy to adopt when faced with a dual-use problem will not necessarily be the
one that offers the greatest expected value. It may, for example, be the one that
minimises risk, or that minimises the likelihood of intentionally causing harm.
I am not in a position to assess the arguments for and against agent-relativity or
the rationality of risk aversion here. But I will briefly consider how a proponent
of the EVC might seek to accommodate these criticisms. She might argue that,
even if we accept agent-relativity and that it is rational to be risk averse, it may
be that the EVC still serves as a useful starting pointas a default position from
which alternative criteria and procedures for assessing responses to dual-use
problems might be derived. Thus, we would assume that all equally likely and
equally valuable outcomes should be given the same weight unless some sound
nonconsequentialist argument could be supplied for prioritising or discounting
outcomes to which we bear a certain relationship. Similarly, we would take a
risk-neutral approach unless a sound argument for risk aversion could be found.
One reason to take an expected-value-based approach as a starting point is that
it captures certain core considerations while taking all other possible factors to
be irrelevant until proven otherwise: everyone should agree that in confronting
dual-use problems, it matters what good and bad outcomes might result, how
likely they are, and how good or bad they are. An expected-value approach
captures the relevance of these considerations. Some might argue that other
factors, such as risk and the agents relationship to the good and bad outcomes,
are also important; but this would be controversial, so it might seem appropriate
to adopt, as a default position, a decision procedure that does not take them into
account.
A second reason to adopt an expected-value approach as a point of departure
is that the formal framework of expected-value theory may provide a helpful
tool for thinking about other approaches as well. As we saw above, the expected
value of a gamble over a number of different states of the world can be thought
150
of as a weighted average of the values of those states, with the weights given
by their probabilities. But we could also weight the values using further
factors to accommodate agent-relativity or rational risk aversion. Consider a
nonconsequentialist approach that takes into account the intentions of an agent
posed with a dual-use problem as well as the consequences of her decision.
It may be helpful to think of such an approach as a variant of the expectedvalue approach that weights the values of different possible states of the world
according to both their probability and whether they were intended.22 Adopting
such a formalised approach may help to ensure that we are clear and explicit
about precisely how much the intentions of the agent matter.
Conclusions
I have outlined an approach to dual-use problems that involves: 1) identifying
possible outcomes of each alternative course of action; 2) identifying possible
states of the world that might result, each consisting of a combination of these
outcomes; 3) explicitly assessing the values and probabilities of these states of
the world; 4) using these probabilities and values to calculate the expected value
associated with each alternative course of action; and 5) selecting the alternative
with the highest expected value.
This expected-value procedure is attractive because it captures some
commonsense intuitions about how to respond to risks and benefits, is consistent
with a major school of ethical thought (consequentialism), and has a track record
in government and corporate decision-making. It is, however, also susceptible
to several criticisms, which may justify deviation from it. Criticisms adverting
to the lack of an adequate metric for ranking states of the world and to empirical
and moral uncertainties suggest that we may need to regard expected values
as providing only an imperfect indication of which course of action to pursue.
Criticisms appealing to bias and demandingness suggest that, even if the right
course of action to take when faced with dual-use problems is always the one
that maximises expected value, it may be best to adopt a decision procedure
that does not involve calculating expected values at all. Finally, criticisms based
on agent-relativity and the rationality of risk aversion go even further: they
suggest that in some cases the right course of action will not be the one that
maximises expected value.
I have not assessed whether these criticisms are persuasive. But I have tried to
show that, even if they are, there may be some value in retaining the expected-
22 The expected value assigned to an alternative would then be relative to the agent making the decision
between alternatives; it would no longer be a value that could be ascribed from any standpoint.
151
value criterion as a default position from which deviations must be justified.

Thus, there may remain an important role for expected values in thinking about
dual-use problems.
Further questions
Suppose that we do keep the EVC in mind, at least as a starting point for further
discussion. Two important further questions arise.
First, what outcomes should be included among the positive and negative
outcomes of a course of action? For example, should the production of knowledge
be included as a positive outcome independent of any positive applications of
that knowledge? This will depend on the controversial question of whether
knowledge has any intrinsic value.
A second question is, to the extent that anyone should ever explicitly assess the
expected value associated with a particular strategy for preventing misuse, who
should do it? For example, should the decision be made by some form of expert
committee, or should it be made by a political or representative body?
I leave these questions as potential topics for future discussion.
152
10. The Doctrine of Double Effect

and the Ethics of Dual Use
Suzanne Uniacke
Is the doctrine of double effect (DDE) relevant to dual-use dilemmas? Can
consideration of the DDE make a significant contribution to the ethics of dual
use? Several writers assume that dual-use dilemmas are instances of double
effect.1 Certainly this particular connection is strongly suggested by the terms
dual use and double effect and also by the structure of dual-use dilemmas.
A dual-use dilemma is said to arise when an action or activity such as research
in the life sciences or publication of that research can have both good and bad
effects: alongside its intended good outcome, for example, the enhancement of
knowledge or human improvement, there can also be a foreseeable bad effect
for example, assisting in the production of bioweapons. Since the doctrine, or
principle of double effect as it is sometimes called, is specifically intended as a
guide to decision-making in ethically difficult cases where an action or course of
action with an intended good effect can also produce a foreseen bad effect, it is
natural to consider whether the moral guidance that the doctrine offers in cases
of double effect is also applicable to dual-use dilemmas.
In order to answer the questions posed above we need an understanding of
the DDE and its intended role in practical ethics. I provide these in the next
section of the chapter. According to the account of the DDE that I shall outline,
dual-use dilemmas are not paradigm instances of double effect. Nonetheless,
as I go on to explain, because dual-use dilemmas have significant features in
common with typical instances of double effect, some important considerations
that can arise in relation to the DDE and its practical application are important
to providing a satisfactory account of the ethics of dual use.
The doctrine of double effect

The DDE is concerned with the application of normative moral theory to
practical moral problems. It takes its rationale from the view that certain actions
are morally objectionable because of the types of actions they are. Those who
invoke the DDE in particularly difficult cases hold that what an actor intends
that is, what she aims to achieve by her actioncan itself be highly significant
1 See, for example, Briggle, A. 2005, Double effect and dual use, in C. Mitcham (ed.), Encyclopaedia of
Science, Technology and Ethics. Volume 2, rev. edn, Macmillan, London, pp. 5436.
153
to the morality of what she does. This view can take absolutist and nonabsolutist forms. In its absolutist form, it holds that certain types of actions,
such as intentionally killing an innocent person, are intrinsically wrong and
always morally impermissible (absolutely prohibited). A non-absolutist version
maintains a very strong moral constraint against certain types of actions such
as those that involve the intended killing of an innocent person; on this view,
such actions are always intrinsically morally objectionable even if in extreme
circumstances they might be necessary in the absence of any morally acceptable
alternative.2
Not all prominent contemporary normative moral theories regard an actors
intention as directly morally significant in this way. An entirely outcomeoriented (consequentialist) theory such as utilitarianism, for instance, does
not share the assumptions on which the DDE is founded and thus regards the
DDE as irrelevant to the moral evaluation of actions that have both good and
bad effects (including instances of dual use). Furthermore, some of those who
accept or are sympathetic towards the moral assumptions behind the DDE
nevertheless regard the DDE itself as problematic in some respects. There is an
extensive critical philosophical literature on issues surrounding the DDE, which
obviously cannot be explored here.3 Since the principal purpose of this chapter
is to address the relevance of the DDE to the ethics of dual use, I shall assume
as a modus operandi that there is a morally significant difference between a bad
effect of an action that the actor intends and one that she foresees but does not
intend and that this distinction can be directly relevant to the moral evaluation
of some actions that have both good and bad effects. This assumption can be
disputed but it represents a widely accepted view.
A key to understanding the nature and purpose of the DDE can be found in
its origins. St Thomas Aquinas appealed to a distinction between an intended,
as opposed to a foreseen, effect of an action in his explanation of why it can
be permissible to kill another person in self-defence.4 Homicide in self-defence
posed a problem for Aquinas because he held the more general position that it
is never permissible for a private person to engage in intentional killing. Does
this view imply that I cannot legitimately use lethal force on an unjust attacker
if it is necessary to save my own life? In responding to this question, Aquinas
claimed that the act of fending off an attacker can have two effects: a good
effect (saving ones own life), which the self-defending actor intends, and also
2 This latter position is sometimes referred to as threshold deontology. See Alexander, L. and Moore, M.
2007, Deontological ethics, in Stanford Encyclopedia of Philosophy, Stanford University Press, Stanford,
Calif., <http://plato.stanford.edu/entries/ethics-deontological/> (viewed 6 February 2012).
3 See, for example, McIntyre, A. 2009, Doctrine of double effect, in Stanford Encyclopedia of Philosophy,
op. cit., <http://plato.stanford.edu/entries/double-effect/> (viewed 6 February 2012).
4 Aquinas, St Thomas 1966, Summa Theologiae, vol. 38, Blackfriars edn, Eyre & Spottiswood, London, 2a,
2ae, 64, article 7.
154
10. The Doctrine of Double Effect and the Ethics of Dual Use
a bad effect (the attackers death), which the actor can foresee but need not
intend. Thus, Aquinas maintained, homicide in genuine self-defence does not
contravene an absolute prohibition against intended killing.
The distinction between an actions intended effects and those effects that the
actor foresees but does not intend is central to what later came to be known as
the doctrine of double effect. Subsequent to Aquinass discussion of homicide
in self-defence, the DDE was developed as part of natural-law reasoning about
the morality of a range of actions that can have both good and bad effects, where
the bad effect is something that it would be wrongful to intend. (Those who
invoke the DDE in cases of double effect need not share Aquinass view that
homicide in self-defence against an unjust attacker is always unintended killing
or that it must be justified as such.5 In contemporary applications of the DDE
the foreseen bad effect in question is usually the death of an innocent person.)
In its traditional form, the DDE is now a general principle of practical ethics
that holds that in some circumstances and under specific conditions it is morally
permissible to cause a foreseen bad effect of a type that is always morally wrong
to intend. These conditions are
1. the act itself must be morally good or at least indifferent
2. the agent must not positively will (intend) the bad effect
3. the good effect must be produced directly by the action, not by the bad effect
4. the good effect must be sufficiently desirable to compensate for the bad
effect.6
Although the DDE has its origins in Thomistic and natural-law ethics it has been
taken up more widely as part of secular moral thinking by those who maintain
that whether or not an actor intends to bring about a particular bad outcome,
such as an innocent persons death, is itself a morally significant feature of her
action that can make a difference to the morality of what she does. For instance,
the DDE is frequently invoked in a number of prominent contemporary contexts,
most notably in relation to the ethics of war and to issues of medical ethics. These
applications include the foreseen killing of noncombatants as an incidental
effect of aiming at a military target, the use of triage in hospital emergency
rooms and on battlefields, some cases of risky surgery, and the administration
of increased doses of pain relief that suppress respiration.7 It is perhaps worth
pointing out here that within its natural-law context, the DDE was developed
5 For a critical discussion, see Uniacke, S. 1994, Permissible Killing: The Self-Defence Justification of Homicide,
Cambridge University Press, Cambridge, ch. 4.
6 A fuller statement can be found in the New Catholic Encyclopedia, 1967, vol. 4, McGraw-Hill, New York,
pp. 10202.
7 See Uniacke, S. 2007, The doctrine of double effect, in R. Ashcroft et al. (eds), Principles of Heath Care
Ethics, 2nd edn, Wiley & Sons, Chichester, UK.
155
and is characterised as a guide to decision-making in morally difficult cases

where an action will have both good and bad effects; however, many recent
secular philosophical discussions seem to assume that the DDEs plausibility
depends on its ability to give a definitive answer about morally permissible
action in every conceivable problem to which it might be applied.
As noted earlier, the DDE derives from the view that irrespective of their overall
effects on the world, some actions are morally objectionable as the types of
actions they arefor instance, as instances of intentional killing. There are
obvious practical difficulties for this view, especially (but not exclusively) in
circumstances of emergency where the alternative actions available to an agent
are severely restricted. Consider the following example. My cars brakes fail on
a hill and I need to swerve to the left in order to avoid oncoming traffic and also
some children who are on a pedestrian crossing in front of me. If I swerve to the
left I will hit and could seriously injure or kill a pedestrian who has just stepped
out from the curb. Is it permissible to swerve the car in these circumstances?
(Since I will hit oncoming traffic and the children if I do not swerve, if it is not
permissible to swerve, will I act impermissibly whatever I do?) In this particular
example the DDE says it is permissible to swerve. My swerving the car is not
in itself a morally impermissible type of act. (The morality of swerving a car
depends on the circumstances and on why I swerve.) If I swerve the car to
the left in these particular circumstances, this would have two distinguishable,
independent effects: an intended good effect (avoiding oncoming traffic and the
children) and also a foreseen bad effect (hitting a pedestrian) that is strictly
incidental to what I intend to do in swerving the car.
According to the DDE, a foreseen effect is strictly incidental to the actors
intention if it is not part of what the actor aims to achieve either as an end or
as a means of achieving an end in the circumstances.8 If at the time of swerving
the car I were to ask myself whether my intention would be fulfilled if, against
expectation, I happen not to hit or kill the pedestrian, I can honestly answer
yes. The presence and position of the pedestrian and any harm he might suffer
are irrelevant both to my swerving to the left in order to avoid oncoming traffic
and the children and also to how I achieve that aim in the circumstances. (It
would be misleading not to point out here that there are vexed philosophical
issues about intention and its moral significance in relation to the DDE.9 To
address those deeper issues would take us too far into a discussion of the DDE
itself. Suffice to say that despite these critical issues there are cases of double
8 Incidental effects of an action are not necessarily bad or unwelcome. Consider, for example, taking on hard
manual labour solely in order to earn money and becoming physically very fit as a result.
9 See, for example, Kamm, F. M. 2000, The doctrine of triple effect and why a rational agent need not intend
the means to his end, Proceedings of the Aristotelian Society, Supplementary Volume, vol. 74, no. 1, pp. 2139;
McIntyre, A. 2001, Doing away with double effect, Ethics, vol. 111, no. 2, pp. 21955; and McIntyre, 2009,
op. cit.
156
effect, including the example just described, where a foreseen bad effect of an
action is clearly incidental to what the actor aims to achieve in acting as she
does.) While the actors intention is central to the guidance offered by the DDE
it is not the sole consideration. For an act of double effect to be permissible the
intended good effect must also be sufficiently morally weighty in comparison
with the foreseen bad effect to warrant causing the bad effect. The DDEs fourth
condition requires a judgment of proportionality. This particular condition
could permit my swerving the car in order to avoid hitting a greater number of
innocent people in the circumstances described above, but it would not permit,
for example, my swerving a car in the direction of a pedestrian in order to avoid
hitting a dog or damaging property.
An obvious question is why we should bother with a distinction between
intention and foresight in circumstances such as these. Why not simply consider
something like the DDEs fourth condition and compare the probable outcomes
of the available alternatives (for example, to swerve or not to swerve) and say
that someone who is faced with making such a decision should act so as to save
as many innocent people as possible? Those who think that the DDE is morally
significant in such cases would reply that if we simply appeal to a principle that
tells us to save as many innocent people as possible and do not also invoke a
distinction between an effect that an actor intends as opposed to an effect that
she (merely) foresees, we commit ourselves to killing some innocent people as
a means of saving others, and this is morally unacceptable. It is one thing that I
put the life of a pedestrian in grave danger in swerving the car in order to avoid
hitting a greater number of people; it would be another thing to cause the death
of innocent person as a means of preventing harm to a greater number by, for
example, using an innocent person as a human shield.
A related point is that according to the DDE, actions of double effect that
meet its four conditions are morally permissible. (They are not necessarily the
right thing to do all things considered or what the actor is morally required to
do.) If under the fourth condition the foreseen bad effect is disproportionate
in relation to the intended good effect then the action under consideration is
held to be impermissible. The fourth condition does not represent an all-thingsconsidered judgment based on the actions predicted or actual overall outcome.
This point of clarification is important in the context of a discussion of the
application of the DDE to the ethics of dual use because dual-use dilemmas are
often explicitly discussed solely in cost/benefit terms, as being entirely a matter
of whether the overall likely benefits of an activitys good use can be traded
off against and would outweigh the risks of its malevolent or negligent use. As
with actions of double effect, however, the basic question posed by a dual-use
157
dilemma is, I take it, whether it would be morally permissible to engage in the
activity in question given the risks of its misuse, and not whether it would be
morally right or morally obligatory to do so.
Double effect and dual use

It is easy to see why one might regard dual-use dilemmas as instances of double
effect. A dual-use dilemma arises when an action or activity has both an intended
good use and a possible malicious or negligent use. Structurally this looks very
much like a case of double effect; however, dual-use dilemmas are also dissimilar
to paradigm instances of double effect in a number of respects. For instance, in
the above example my decision whether to swerve the car is a forced choice in
circumstances in which I have little time for deliberation and extremely limited
options. This is the situation in many instances of double effect whereas dualuse dilemmas are usually not like this. But we can set this particular dissimilarity
aside; although the DDE is often applied to situations of forced choice where the
actor has little time for reflection or practical manoeuvre, this is not necessary
to double effect, and very familiar applications of the DDE in medical contexts
and in war do not display these features.
Other dissimilarities between double effect and dual use are, however, more
significant. In typical instances of double effect the actor foresees the bad
effect as morally certain or highly probable; the bad effect is usually also an
unavoidable and a direct effect of the action that produces the good effect.
(This is true of my swerving the car in the above example, for instance.) These
particular features have implications for what is required under the DDEs four
conditions, fuller statements of which say that in coming to a judgment under
condition (4) a foreseen bad effect that is morally certain to occur should weigh
more heavily (ceteris paribus) than one that is merely probable; they also specify
that if the actor could obtain the intended good effect without also producing
the foreseen bad effect he should do so. (A reasonable interpretation of this
latter directive will inevitably involve considerations of comparative cost and
risk, but it clearly implies that, for example, if I swerve the car I must try to
warn the pedestrian of the danger by, say, tooting the cars horn.) By contrast,
an actor facing a dual-use dilemma who foresees a bad use as a possibility is
often unable to ascertain the probability of its occurring; this is partly because
the bad use, if it occurs, is indirect and dependent upon the further agency of
another person.
The resolution of dual-use dilemmas must obviously address the extent to which
an actor can reasonably foresee the harmful use of something that she intends
for a good purpose. Such judgments are notoriously difficult, particularly
158
because they involve the actions of others, and this epistemological difficulty
is prominent in the literature on the ethics of dual use in relation to the socalled precautionary principle.10 What is less often highlighted is that dual-use
dilemmas also involve the complex question of the moral significance of another
persons further agency in relation both to the degree to which I ought to take
this into account in deciding what it is permissible for me to do and to the
degree to which by acting in a certain way I am morally obliged to try to prevent
the further agency of another person or its bad effects. Above I have brought
out this moral complexity by identifying a number of dissimilarities between
dual-use dilemmas and typical cases of double effect. Might considerations that
inform the DDE and the debate surrounding its application nonetheless assist us
in reasoning about this moral issue in relation to dual use? I will suggest in the
next section that they can.
Prior to that discussion, as a crucial first step, we must ask whether the DDE
is applicable to cases in which the foreseen bad effect is due to the agency of
another person. (If it is not then dual-use dilemmas are not instances of double
effect.) I see no reason in principle why it cannot be. Consider a modified version
of the earlier example in which my cars brakes fail, but where I will hit the
pedestrian only because when I swerve to the left he will be jostled into the
path of my car by someone else. We can also think about the DDE in relation to
a case of duress. Say the Gestapo threatens to torture or kill an innocent hostage
unless I tell them the whereabouts of a Jewish family in hiding. In resisting this
threat I foresee but I do not intend the injury that the Gestapo will inflict on an
innocent hostage, even if I could prevent it by giving in to the threat. In a third
type of case, we might apply the DDE to my decision to work as a bartender,
where in taking up such a position I foresee that despite taking due care, at
some future point I will almost certainly serve alcohol to someone who will then
drive while intoxicated and possibly injure or kill someone.
If the DDE is indeed applicable to examples such as the three just described, in
which a foreseen bad effect is due to the agency of another person, we cannot
conclude simply on this basis that the DDE is therefore applicable to dual-use
dilemmas. This is because there are also various respects in which each of these
three examples differs structurally from a dual-use dilemma. For instance, in
the first, modified car-swerving example, although the bad effect is contingent
upon the intervening agency of another person (who jostles the pedestrian into
the path of my car), nonetheless I will directly hit the pedestrian by swerving
10 This is a general principle that states that if an action or activity has a suspected risk of causing serious
harm to the public or to the environment, if experts disagree about whether the action or activity will or is
likely to cause this harm, it is up to those proposing to undertake the action or activity to prove that it will
not cause this harm. The implication of the principle is that without such proof they should not proceed.
159
my car.11 In the second, duress example, the act in question (my refusal to reveal
the familys whereabouts to the Gestapo) is a negative action, as opposed to
something that I do; furthermore, even if I could protect the innocent hostage
by telling the Gestapo what they want to know, the injury inflicted on the
hostage will be wholly due to the actions of another agent (the Gestapo). In the
third, bartender example, the action in question is my assumption of a longerterm activity, as opposed to an individual action of double effect. Nonetheless,
the third example is probably structurally the closest to an instance of dual
use of the three examples, in that it involves positive action on my part with a
foreseen bad effect that is both indirect and due to the agency of another person.
More importantly, the third example also shares a crucial feature with dual-use
dilemmas that is missing in the second, duress example, but is present in the
first, modified car-swerving example. And this is that in the circumstances as
they actually occur, my positive action plays a specific causal role in bringing
about the bad effectnamely, it enables or aids another person to do something
wrongful (to jostle the pedestrian into the path of my car/to drive while
intoxicated/to create a bioweapon). (In the second example, on the other hand,
my refusal to give in to the Gestapos threat does not enable them to torture or
kill the innocent hostage: it does not provide them with the means.) For this
reason, in the first and the third examples the description of the foreseen bad
effect of my positive action includes enabling or aiding the wrongful action of
a third party, and this is also true of dual-use dilemmas where what I do (for
example, publish research) enables or aids its malicious or negligent use.
The fact that in each of these cases the further actions of another agent are
properly regarded as a foreseen bad effect of my action presses the following
question: under what conditions am I responsible for the actions of another
person in the sense that I must regard what he or she does as an effect of what
I do for which I can be accountable? This is a broad and complex question that
clearly cannot be addressed in detail in this chapter. Nevertheless, in the next
section I shall identify considerations relevant to the DDE that can also shed
important light on how we should answer this question in the case of dual-use
dilemmas.
Responsibility for a foreseen bad effect

The DDE does not say that a person is not responsible for the foreseen bad effect
of his action provided this effect is unintended and proportionate to an intended
11 Whether we would regard the jostling as (merely) part of circumstances in which I swerve the car, as
opposed to a primary cause of the pedestrians being hit, could depend on whether it was accidental or
inadvertent (simply a result of a crowded footpath, for instance), as opposed to an intended, negligent or
malicious action.
160
good effect. On the contrary, according to the DDE the actor is responsible for the
bad effect in an important sensenamely, that she must take it very seriously
into account in deciding how to act and she must morally account for having
brought it about or not prevented it when she might have done otherwise.12 As
an upshot of this responsibility for the bad effect, the actor must try to achieve
the good effect without the bad if she can possibly do so. It might uncritically
be assumed that this type and degree of responsibility for the bad effect are
due to the fact that in typical cases of double effect although the bad effect is
unintended it is both a foreseen and a direct effect of what the actor chooses to
do. Certainly these features strengthen the case for attribution of responsibility
for the bad effect. But they are not necessary to such attribution. In the first
(modified car-swerving) and the third (bartender) examples discussed earlier,
the foreseen bad effect of my action incorporates the agency of another person,
and this is something that I must take into account in deciding how to act and for
which I can be called to account. To be sure, in the first example, the bad effect
is directly and strongly connected to my swerving the car, but the connection
is less direct in the third example in which my serving someone alcohol might
be a sufficient but not a necessary condition of his being intoxicated and also
an indirect cause of his driving in this condition. In both these examples,
attribution of responsibility for the bad effect to my own agency is due in large
part to the enabling role that my positive action plays in the further agency of
another person. And this will also be true of cases of dual use.
At this point I would like to suggest that I am responsible for the bad effect of
another persons further agency, in the sense that I must take it into account in
deciding how to act and I can be accountable for it (and will bear a degree of
culpability for it in the absence of justification or excuse) if at the time of my
own prior action: 1) I can reasonably foresee that another agent will bring about
this effect; and 2) in acting as I do I provide someone else with the means or the
opportunity to bring about this effect.13 Discussions of dual-use dilemmas tend
to focus on the first of these two conditionsthat is, on the extent to which
another persons further agency is reasonably foreseeable and on the probability
of the suspected bad effect actually occurring. In drawing attention to the
significance of the second condition, I hope to emphasise that where these two
conditions are met, although the bad effect is indirect and due to the agency of
12 Adherents to the DDE regard our obligation to prevent harm as generally less stringent than our obligation
not to do harm. Nonetheless, they also hold that it can be wrong not to prevent harm to an innocent person,
particularly where this is an intended effect of ones inaction. It can be important to the permissibility of
some instances of non-prevention of harm (for example, some decisions not to rescue) that the harm although
foreseen is not intended.
13 Note these two conditions say if and not only if. Some people would also regard me as responsible
in this sense for what the Gestapo does to the innocent hostage in the second example. But a persons
responsibility for a foreseen outcome of her negative action is a much more contentious issue, especially where
it also involves the further agency of someone else, and neednt be taken up here.
161
another person, my contribution to bringing about the bad effect is strong in

virtue of the fact that I both foresee and enable it. Under these two conditions, I
am responsible for the bad effect in the robust sense outlined above. Moreover,
responsibility in this sense also generates an agent-related obligation on my
part to try to prevent the bad effect if possible. Although the bad effect, if it
occurs, will be an indirect effect of my action that is due to the further agency of
someone else, because it is nonetheless something that I have enabled or aided,
my obligations in this regard are stronger than any general obligations one
might have to prevent harm or wrongdoing by others.
Concluding remarks
I have maintained in this chapter that dual-use dilemmas are not paradigm
instances of double effect and I have identified significant similarities and
dissimilarities between the two. All the same, our critical thinking about whether
the DDE and its conditions are indeed relevant to dual-use dilemmas has served
to highlight some morally important aspects of the ethics of dual use that might
otherwise not receive sufficient critical attention. These include the relevance
of a distinction between foresight and intention to the moral permissibility
of an action with both good and bad effects; the conditions under which we
are accountable for the further agency of other people who bring about a bad
effect; and the relatively stringent agent-related obligation to prevent harm or
wrongdoing by others where ones action would provide them with the means
or the opportunity. A suitably morally complex ethical evaluation of dual use
needs to address these various issues. But in a more sceptical vein we can ask
whether we need to refer to the DDE in order to do this.
The DDE is centrally concerned with the moral relevance of a distinction between
intention and foresight, since this distinction is held to be crucial to permissible
action in cases of double effect. It is clear in the case of a dual-use dilemma
that the foreseen bad use is not intended by the agent who faces the dilemma
(although it may well be intended by the further agent who engages in the bad
use). According to the DDE, the fact that an actor does not intend a foreseen
bad effect of her action can be highly significant to the moral permissibility
of what she does. This then will also be a consideration in deliberation about
the permissibility of activities that are potentially dual use. The DDE also
emphasises that in cases of double effect the fact that the bad effect is foreseen
is itself morally significant to the permissibility of the action. The actor must
take the bad effect seriously into account in her deliberations and her action
must be justified in terms of a good effect that is sufficiently morally important
to warrant causing the bad effect. And as I have argued above, this can be so in
cases of double effect where the bad effect will be indirect and due to the further
162
agency of another person. Furthermore, we can be responsible in the sense of

being accountable for the harmful or wrongful actions of other people if by our
own actions we have enabled or aided them. This means that our obligation as
agents not to do harm extends to the obligation not to provide others with the
means of doing harm; while not an absolute obligation, it is nonetheless a strong
one. It also means that our obligation to prevent harm being done by others
becomes a more stringent, agent-related obligation when the harm in question
is something that we ourselves have provided others with the means of doing.
These obligations are particularly salient to the ethics of dual use.
163
11. Unknowns in Dual-Use Dilemmas

Michael Smithson
Dual-use dilemmas are defined as a consequence of the potential for the same
piece of research to be used for harm and for good. Miller and Selgelid1 advise
that fine-grained ethical analyses of dual-use research in the biological sciences
would seek to quantify actual and potential benefits and burdens, and actual and
potential recipients/bearers of these benefits and burdens. These analyses would
also identify a range of salient policy options. Desirable as such quantification
may be, the path to it is obstructed by several yawning abysses in the form of
unknowns. If unresolved or ignored, these unknowns can render fine-grained
analysis and quantification impossible or arbitrary.
This chapter investigates these unknowns and presents some approaches for
dealing with them or, at least, taking them into account. These approaches are
grounded in subjective expected utility (SEU) theory, whose primary tenet is
that rational agents weigh up the potential consequences of acts by summing
the products of the probability of every possible outcome and its utility. At least
some of the probabilities and utilities might be based on subjective assessments,
whence the S in SEU. SEU is employed here as a prescriptive or benchmark
framework. My primary intent is to ask what an SEU-rational agent would
conclude or choose, so that human decision-makers can knowledgeably decide
whether to take the agents advice on board or reject it.
Our survey of unknowns comprises three sections
1. dilemma structures
2. state space indeterminacy
3. imprecision and biases in judgments.
The first section examines dual-use dilemmas from the viewpoint of the standard
social dilemmas framework. The primary purpose is to ascertain when a dual-use
dilemma is a mixed-motive game and therefore a genuine dilemma, and when it
is a trade-off. Dilemmas pose difficulties for rational self-interest that trade-offs
do not. The second section begins with the observation that dual-use dilemmas
often are not limited to considering just two possible uses and may instead
involve an indeterminate number of uses. Likewise, the number of response
options also may be a matter of choice. In other words, the use and response of
1 Miller, S. and Selgelid, M. 2007, Ethical and philosophical consideration of the dual-use dilemma in the
biological sciences, Science and Engineering Ethics, vol. 13, p. 542, emphasis in the original.
165
state spaces are indeterminate. Both state spaces have consequences for decisionmaking and these are elaborated in this section. The third section begins by
pointing out the dangers in restricting judged probabilities and utilities to
falsely precise representations and describing the decisional consequences of
imprecision. It then brings in psychological considerations such as tendencies
towards overconfidence in predictions and confirmation bias.
The discussion of these issues involves a bit of mathematics and some technical
definitions. Both the mathematics and technicalities are necessary, and these are
far from merely academic matters. Dilemmas were not well understood until a
mathematical framework was developed for describing them and distinguishing
them from non-dilemmatic trade-offs, and that development transformed the
economics of public goods and common-pool resources as well as our ideas about
individual versus collective rationality. It also provided heretofore unachievable
insights into the evolution of cooperation, and has spawned a vast literature
in economics, political science and psychology. Likewise, not knowing what
all the possible outcomes are is commonplace in real-world decision-making in
the face of a largely unknowable futureand is routinely ignored by decisionmakers and standard decision frameworks. A systematic consideration of the
consequences thereof includes asking which options (and how many) should be
on the table for responding to dual-use dilemmas. The framework presented
in this chapter has been applied to debates about this issue in law, medicine
and related policy matters. Finally, a careful assessment of the consequence of
uncertainty requires understanding that there are different kinds of uncertainty,
with distinct consequences for reasonable decision-makers. A growing literature
on this topic includes demonstrations of its practical impact in domains such as
insurance.
One reviewer of an earlier draft of this chapter declared that I indulged in makebelieve that mathematics can contribute to our understanding of dual-use
dilemmas and what to do about them. I have three rebuttals to this dismissive
remark. First, as I hope the previous paragraph has made abundantly clear, the
mathematics in this chapter are about as far from make-believe as it is possible
to be; they concern some of the most important uncertainties in real-world
decision-making. Second, without even the simple mathematics that informs
this chapter, it is nigh well impossible to properly understand and deal with
those uncertainties; words simply will not suffice. Finally, failing to distinguish
a trade-off from a dilemma, failing to take into account the fact that we do not
know what new technologies will be invented and yet often underestimate the
likelihood of their emergence, not bothering to think carefully about whether
we need two, three, 10 or 1000 alternative ways of responding to dual-use
166
dilemmas, and ignoring the ways in which we are systematically biased in

our judgments about unknownsnow that would be genuine make-believe
indeed.
Dilemmas or trade-offs?
When are the dilemmas actually social dilemmas, as opposed to trade-offs?
Genuine social dilemmas are harder to resolve than trade-offs. They also present
a fundamental difficulty for rational, self-interested agents because the pursuit
of self-interest in a social dilemma leads to the destruction of the common good.
Moreover, the structure of a social dilemma partly determines the approaches
needed to resolve it.
First, social dilemmas are social. They involve a game structure comprising at
least two decision-makers. Some dual-use dilemmas do not readily yield such
a game structure because they are cast as single-agent decisions. An example
is the concern that research conducted for beneficial purposes might be used
by secondary researchers or other users to construct bioweapons. If these users
would not be able to exploit the research if it were not conducted then the
situation reduces to a single-agent decision:
Research potential benefits and risk of exploitation
versus
No research potential costs and no risk of exploitation.
While this decision may be difficult, it is not a social dilemma or even a dilemma
in the sense of damned if you do and damned if you dont. Instead, this is
arguably a trade-off wherein each option combines potentially strong positive
and strong negative consequences.
Dual-use dilemmas can become social dilemmas involving multiple agents if the
decisions made by each agent alter the consequences for all of them. Biological
research as an arms race is perhaps the most obvious example. For instance, if
researchers in country A revive an extinct pathogen and researchers in country
B do not, country A temporarily enjoys a tactical advantage over country B while
also risking theft or accidental release of the pathogen. If country B responds
by duplicating this feat then B regains equal footing with A, but has increased
the overall risk of accidental release or theft. For country A, the situation has
worsened not just because it has lost its tactical advantage but also because
the risk of release has increased. Conversely, if A restrains from reviving the
167
pathogen then B may play A for a sucker by reviving it. It is in each countrys
self-interest to revive the pathogen in order to avoid being trumped, but the
collective interest resides in minimising the risk of accidental or malign release.
A similar example of a social dilemma is where countries A and B are considering
whether to eliminate their respective stockpiles of smallpox. The payoff matrix
is shown in Table 11.1. The entries are
R = reward
T = temptation
S = sucker
P = punishment.
This matrix enables a definition of a social dilemma. A social dilemma exists if
these four conditions hold

R>P
R>S
2R > T + S
T > R or P > S.
Table 11.1 Pay-Off Matrix for a Two-Agent Game

B
A
Eliminate
Retain
Eliminate
Ra, Rb
Sa, Tb
Retain
Ta, Sb
Pa, Pb
Source: Authors representation.
There are three well-known dilemma structures, depending on how each

countrys decision-makers rank-order the consequences of the countries joint
decisions
chicken: T > R > S > P
prisoner: T > R > P > S
trust: R > T > P > S.
We do not require quantification of the matrix entries; they only need to have
a complete ordering for each player. We will denote the best outcome by 4
and the worst by 1. Of course, it is possible for the structure to differ between
the two countries. In Table 11.2, the structure is Chicken for country A and
Prisoner for country B.
168
Table 11.2 Chicken and Prisoners Dilemma Combination

B
A
Eliminate
Retain
Eliminate
3, 3
2, 4
Retain
4, 1
1, 2
Table 11.2 makes it easy to see the roles played by greed and fear in a social
dilemma. Each country can obtain its best outcome (rated 4) by retaining their
supply if the other country eliminates theirs. Country Bs worst outcome (rated 1)
and country As second-worst result are if each eliminates supply while the other
retains theirs. If both act on fear and/or greed and retain their supplies then the
joint outcome is the worst of all four (rated 1 for A and 2 for B).
Different structures yield distinct pressures for and against eliminating smallpox
stockpiles. A cooperation index is
which provides an overall indication of motivation for elimination. All else

being equal, Prisoner will have a smaller value for K than Chicken or Trust.
The cooperation index, in turn, may be decomposed into a Fear and Greed
component
where
and
Thus, in Trust and Prisoner Kf > 0 whereas in Chicken Kf < 0, while in Chicken
and Prisoner Kg > 0, whereas in Trust Kg < 0. In Chicken, Greed is the component
detracting from motivation to eliminate stockpiles, and in Trust, Fear is the
detractor. Prisoner is the only dilemma in which both the Fear and the Greed
components exceed zero, so that both detract from motivation to eliminate. This
is why K generally is lowest for Prisoner.
Another crucial characteristic of a dilemma is the public versus private
nature of the consequences. This strongly influences whether institutional
solutions such as privatisation are potential solutions for social dilemmas.
A good is subtractable if its use by one agent decreases the potential for its
169
use by another. Attention is subtractable (devoting attention to one thing

decreases the attention that can be given to others) whereas information is nonsubtractable (simply acquiring information does not decrease its availability to
others). A good is excludable if access to it can be restricted. Secrets and legally
proprietary information are fairly excludable, whereas unsecured information
on the Internet is not. Goods are privatisable insofar as they are excludable and
subtractable.2
Public goods (and bads) are strongly non-subtractable and non-excludable.
The open-access and communalistic norms of scientists render research outputs
a public good. A virulent, easily transmissible pathogen quickly can become
a public bad. Common-pool resources, on the other hand, are goods that are
subtractable but non-excludable. Air or water quality is an example of a
common-pool resource, and the diffusion of nanomaterials could threaten either
of these. Toll goods are those, like proprietary information, that are excludable
but non-subtractable. And finally, truly private goods are those, like wellguarded smallpox supplies, that are both excludable and subtractable.
The temporal dimension also can play an important role in dilemmas.3 A
large literature indicates that repeated dilemmas are more easily solved than
one-shot dilemmas.4 Repeated dilemmas permit agents to learn, build trust or
negotiate and verify compacts, whereas these are considerably more difficult in
one-shot dilemmas. Consider cat out of the bag (COB) consequences: it takes
only one instance of the research to yield the potential for misuse or accident;
subsequent research replication usually does not increase those risks. The COB
risk associated with a particular research project can be the basis of a one-shot
dilemma; however, if we consider the potential for multiple research efforts to
throw up COBs then we have the makings of a repeated dilemma. Packaging oneshot dilemmas into a common category reframes them as repeated dilemmas,
enhancing the chances of solving them.
Finally, it should be noted in passing that we have implicitly assumed that
both agents know not only their own outcome preferences but each others as
well. Of course, it is also crucial to take into account each agents perception
of the others pay-offs, because those determine what each agent believes the
others (rational) motives and best moves will be. Referring back to Table 11.2,
if country As intelligence is that country Bs pay-off matrix is identical to As
(that is, Chicken instead of Prisoner) then A will underestimate Bs motivation
2 Ostrom, E., Gardner, R. and Walker, J. 1994, Rules, Games and Common Pool Resources, University of
Michigan Press, Ann Arbor.
3 For example, Smithson, M. 1999, Taking exogenous dynamics seriously in public goods and resource
dilemmas, in M. Foddy, M. Smithson, S. Schneider and M. Hogg (eds), Resolving Social Dilemmas: Dynamic,
Structural, and Intergroup Aspects, Psychology Press, Philadelphia, pp. 1732.
4 Danielson, P. 1992, Artificial Morality: Virtuous Robots for Virtual Games, Routledge, London.
170
for retaining smallpox supplies. This is because Bs actual outcome ranks for
retention are {4, 2} and for elimination they are {3, 1}, whereas A will believe
they are {4, 1} and {3, 2} respectively.
Obviously there is much more to determining the nature of a dual-use dilemmas
structure than has been dealt with in this section. The intention here is merely
to provide a starting point by posing the question of whether a structure
constitutes a social dilemma and, if so, what kind of social dilemma the structure
corresponds to. The crucial difference between a social dilemma and a tradeoff is that a social dilemma entails a conflict between individual and collective
interests that does not appear in trade-offs. It is plausible, therefore, that the
policies and procedures for dealing with dual-use dilemmas also will need to
distinguish between the two.
Partition indeterminacy
Nearly all formal decision-making frameworks, including SEU, assume
that all possible options and outcomes are known. In other words, the state
space is predetermined. The nature of innovative research implies that in at
least some dual-use dilemmas that assumption is untenable on three counts.
First, the potential outcomes of research often are not completely known. The
accidental creation of a mousepox super strain5 is a case in point, as is the
current state of ignorance regarding potential consequences of third or fourthgeneration nanomaterials. Second, the uses of research outputs also sometimes
are unanticipated. Witness the applications in cryptography of number theory,
a sub-discipline that once was held up as the epitome of pure mathematics
beyond reach of any applicability. Third, the variety of responses to the threat
of research misuse is not predetermined. The first two sources of state space
indeterminacy are matters to be taken into account by those who make judgments
and decisions, and this is the topic of the next subsection. The third, however,
can be a matter of choice, and this is discussed in the subsection thereafter.
Unknown outcomes and consequences

In most standard probability theories, on the grounds of insufficient reason,
a probability of 1/J is assigned to J mutually exclusive possible events when
nothing is known about the likelihood of those events. For example, in a race
involving three greyhounds, an agent who knows nothing about any of the
dogs would assign a value of 1/3 to the probability of each greyhound winning.
Moreover, even under alternative probability assignments the probabilities
5 Miller and Selgelid, op. cit., pp. 52380.
171
of the J events must sum to 1, meaning that the entire probability mass is
concentrated on that set of events. Thus, a more knowledgeable rational agent
who has assigned a probability of 1/2 to the first dog winning and 1/4 to the
second dog is compelled to assign the remaining 1/4 to the third.
The number of possible elementary events or states in a space is determined by
the partition of that space. The greyhound race has been partitioned into three
outcomes: dog one wins, dog two wins or dog three wins. Were we to allow
ties, the partition would expand to J = 7. The ignorant agent now would assign
a probability of 1/7 to each dog winning, and the more knowledgeable agent
could distribute the remaining 1/4 probability across the remaining five events
instead of having to allocate it all to the third dog winning. Thus, probability
assignments are partition dependent.
When partitions are indeterminate, partition dependence poses a problem for
subjective probability assignments. This is not the same problem as unknown
probabilities over a unique and complete partition (for example, where we know
that there are only red and black marbles in a bag but do not know how many
of each).6 It is more profound. In the absence of a uniquely privileged partition,
there is no defensible prior probability distribution to be constructed.
Two separable problems for partitions may arise. One is an incomplete account
of possible events. A unique and complete partition might be attainable in
principle, but we lack the necessary information. The other problem is the
absence of a privileged partition even when one has a complete account of
those possibilities. Shafer7 presented an example of this problem as a motivation
for the belief functions framework. He asked whether the probability of life
existing in a newly discovered solar system should be partitioned into {life, no
life} or {life, planets without life, no planets}. This issue arises naturally when
a decision must be made that involves a threshold or an interval on a continuum.
We shall revisit this particular problem in the next subsection.
Returning to the first problem, the most common situation confronting judges
or decision-makers is partial knowledge of the possible outcomes. We may know
some of the potential uses and misuses of a new biotechnology but not all of
them. We might even be willing to assign a subjective probability that party
X will misuse this technology in ways we can anticipate. But what probability
should we assign to X misusing the technology in ways we havent anticipated?
Likewise, what probability should we assign to party Y finding a new way to
use the technology for good?
6 Smithson, M. 2009, How many alternatives? Partitions pose problems for predictions and diagnoses,
Social Epistemology, vol. 23, pp. 34760.
7 Shafer, G. 1976, A Mathematical Theory of Evidence, Princeton University Press, Princeton, NJ.
172
Smithson8 presents strategies for dealing with partition dependence,

distinguishing those that apply when a privileged or at least agreed-upon
partition is attainable from those that apply when it is not.
1. Where a privileged or agreed-upon partition is attainable
a. de-biasing strategies
b. establishing criteria for choosing partitions.
2. Where there is no privileged or agreed-upon partition
a. using diverse partitions
b. modelling partition-dependence effects
c. using (non-standard) probabilistic frameworks that avoid partition
dependence.
De-biasing strategies are needed because human judges are strongly influenced
by partitions in their subjective probability assignments. Two important
manifestations of partition dependence are distorted judgments of likelihoods
of compound events and anchoring on an ignorance prior. A classic study9
concerning peoples assignments of probabilities to possible causes of a given
outcome (for example, an automobile that will not start) revealed that possible
causes that were explicitly listed received higher probabilities than when the
same causes were implicitly incorporated into a catch-all category of additional
causes. The effect has since been referred to as the catch-all underestimation
bias and also sometimes the pruning bias.10
Likewise, it has been empirically demonstrated11 that subjective probability
judgments are typically biased towards the ignorance prior determined by the
partition salient to the judge. That is, people anchor on a uniform distribution of
1/J across all J possible events, even when taking into account prior evidence of
how likely each event is. Because those adjustments typically are insufficient,12
judges intuitive probability assignments are biased towards probabilities of 1/J.
Criteria for choosing partitions and methods for exploring diverse partitions are
not well established. One recently proposed set of criteria will be elaborated in
8 Smithson, 2009, op. cit., pp. 34760.
9 Fischhoff, B., Slovic, P. and Lichtenstein, S. 1978, Fault trees: sensitivity of estimated failure probabilities
to problem representation, Journal of Experimental Psychology: Human Perception Performance, vol. 4,
pp. 33044.
10 Russo, J. E. and Kolzow, K. J. 1994, Where is the fault in fault trees? Journal of Experimental Psychology:
Human Perception Performance, vol. 20, pp. 1732.
11 Fox, C. R. and Rottenstreich, Y. 2003, Partition priming in judgment under uncertainty, Psychological
Science, vol. 13, pp. 195200.
12 Tversky, A. and Kahneman, D. 1974, Judgment under uncertainty: heuristics and biases, Science,
vol. 185, pp. 112431.
173
the next subsection,13 but these have limited scope. Other criteria could be linked
with strategies for manipulating and exploring judgment biases in informative
ways. As a simple example, expert judges estimating probabilities of adverse
consequences arising from the revival of an extinct pathogen could be randomly
assigned to one of two conditions: a twofold partition (consequence versus no
consequence) or a J-fold partition (a list of anticipated consequences plus a
catch-all category for unanticipated ones). Partition dependence would predict
that the average probability of an adverse consequence in the first condition
should be less than the average sum of the probabilities across the J consequence
categories in the second condition. The results would yield fairly defensible
lower and upper expert estimates of the probability of adverse consequences.
More sophisticated experimental designs would enable the construction and
estimation of relevant partition-dependence effects.
Finally, let us briefly consider non-standard probability frameworks that are not
partition dependent. These have appeared in the growing literature on generalised
probability theories, and also in behavioural economics.14 Walley15 argues on
normative grounds that imprecise probability frameworks can avoid partition
dependence entirely. He proposes that when judges are permitted to provide
a lower and an upper probability judgment (that is, imprecise probabilities)
every ignorance prior should consist of vacuous probabilities {0, 1}. In the
greyhound race example, the ignorant agent could assign a lower probability
of 0 and an upper probability of 1 to every event regardless of whether the
partition is threefold or sevenfold. The lower and upper probabilities of the first
dog winning would be 0 and 1 regardless of the partition, thereby avoiding
partition dependence. Walley developed an updating method (the imprecise
dirichlet model) that is partition independent and has generated interest within
the community of imprecise-probability theorists.
That said, recent studies16 experimentally demonstrated that naive judges are
just as strongly influenced by partitions when making imprecise probability
judgments as they are when making precise probability judgments. Moreover,
they demonstrated that many judges anchor on 1/J as the midpoint of their
lower and upper probability judgments. No applicable de-biasing strategies have
13 Smithson, M. 2006, Scale construction from a decisional viewpoint, Minds and Machines, vol. 16, pp.
33964; and Smithson, 2009, op. cit., pp. 34760.
14 For example, Grant, S. and Quiggin, J. 2004, Conjectures, refutations and discoveries: incorporating
new knowledge in models of belief and decision under uncertainty, Paper presented at the 11th International
Conference on the Foundations and Applications of Utility, Risk and Decision Theory (FUR XIParis), under
the joint auspices of the Ecole Nationale Suprieure dArts et Mtiers (ENSAM) and the Ecole Spciale des
Travaux Publics (ESTP), Paris, 2 July.
15 Walley, P. 1991, Statistical Reasoning with Imprecise Probabilities, Chapman Hall, London; and Walley,
P. 1996, Inferences from multinomial data: learning about a bag of marbles, Journal of the Royal Statistical
Society, Series B, vol. 58, pp. 334.
16 Smithson, M. and Segale, C. 2009, Partition priming in judgments of imprecise probabilities, Journal of
Statistical Theory and Practice, vol. 3, pp. 16982.
174
yet been reported. Nevertheless, the possibility remains that allowing judges to
express one kind of uncertainty (imprecision in their probability assignments)
may militate against the impact of another kind (partition indeterminacy).
How many options?

Policies regulating responses to dual-use dilemmas could be limited to two
optionsfor example: laissez faire and bans. But what about a third option,
such as oversight by a regulatory body? Or more than two additional options?
Are there criteria that could indicate how many options a rational agent should
prefer? How would we know whether each option was worth retaining? This
appears to be a relatively unexplored topic, but reasonably important given
that this is one aspect of dual-use dilemmas where policy and decision-makers
actually have choices. It is directly related to partition indeterminacy because
we are constructing a partition of a space of possible acts.
In the context of legal standards of proof, a typical threshold probability of
guilt associated with the phrase beyond reasonable doubt is in the [0.9, 1]
range.17 For a logically consistent juror, a threshold probability of 0.9 implies
the difference between the utility of acquitting versus convicting the innocent
is nine times the difference in the utility of convicting versus acquitting the
guilty.
Connolly demonstrated that the utility assignments to the four possible outcomes
(convicting the guilty, acquitting the innocent, convicting the innocent, and
acquitting the guilty) that are compatible with such a high threshold probability
are counterintuitive. Specifically, if one does want to have a threshold of 0.9,
one must be prepared to hold the acquittal of the guilty as highly desirable, at
least in comparison to the other available outcomes.18 He also showed that more
intuitively reasonable utilities lead to unacceptably low threshold probability
values.
Smithson19 showed that the incorporation of a third middle option (such as
the Scottish not-proven verdict) with a suitable threshold can resolve this
quandary, permitting a rational (subjective expected utility) agent to retain a
high conviction threshold and still regard false acquittals as negatively as false
convictions. The price paid for this solution is a more stringent standard of
proof for outright acquittal.20 The main point here is that a consideration of
17 Connolly, T. 1987, Decision theory, reasonable doubt, and the utility of erroneous acquittals, Law and
Human Behavior, vol. 11, pp. 10112.
18 Ibid., p. 111.
20 For evidence that this also is what humans do, see Smithson, M., Gracik, L. and Deady, S. 2007, Guilty,
not guilty, or ? Multiple verdict options in jury verdict choices, Journal of Behavioral Decision Making,
vol. 20, pp. 48198.
175
preferences as expressed by the relative positions of utilities can aid in the

choice of a partition of acts, due to the connection between these utilities and
the threshold probabilities that determine when one act is chosen over another.
Applying Smithsons framework to dual-use dilemmas, consider the simplest
set-up in which either some kind of misuse of a research output occurs or no
misuse occurs. Suppose we must make a decision regarding the fate of a potential
research project (for example, whether to prohibit it or allow it to proceed),
and we wish to do so on the basis of an estimated probability that the research
output could be misused. Let us assume that choices will affect the utility of
the no-misuse outcome because of inhibited scientific progress and/or resource
expenditure in security arrangements. Let us also assume that the utility of the
misuse outcome also will be affected by choice because the same considerations
will be combined with the consequences of misuse, even if they are dwarfed by
the latter.
Suppose we have a J-fold partition of acts Rj, for j = 0, 1, 2, , J-1. There are
two possible outcomes: no misuse and misuse. The act Rj has a utility Hj if there
is no misuse and a utility Gj if there is misuse. We assume that the acts Rj are
ordered so that Hj > Hj1 and Gj1 > Gj for any j. A straightforward argument
shows that if the odds of no misuse exceed an odds threshold defined by
then the decision-maker should prefer act Rj over Rj1. The odds threshold
Wj1j therefore is determined by the ratio of utility differences.
Table 11.3 Twofold Partition of Acts
R1
R0
Laissez F.
Prohibit
No misuse
H1 = 1
H0
Misuse
G1 = 0
G0
The simplest set-up of this kind is shown in Table 11.3. There are two possible
acts: prohibition or laissez faire. Without loss of generality we may assign
H1 = 1 (the best possible outcome) and G1 = 0 (the worst). Therefore, the odds
threshold is
It immediately follows that if G0 < q0 for 0 < q0 < 1 then

176
H0 > 1 q0/w01.
Suppose we also wish to restrict w01 > y0 > 1. This should seem reasonable,
because we are merely restricting the odds-of-no-misuse threshold to be above 1.
Then
For example, if q0 = 0.1 and y0 = 10 then H0 > 0.99; and in fact if q0 = 1 and
y0 = 100 then we also have H0 > 0.99. Thus, no misuse under prohibition has
nearly as high a utility as no misuse under laissez faire, implying that prohibition
hardly decreases utility at all. Moreover, in the special case where prohibition
obviates misuse so that G0 = H0, a high odds threshold yields a correspondingly
high value for G0 and H0. For instance, y0 = 10 implies G0 and H0 both must
exceed 10/11.
The problem is the inability to simultaneously have a high value of y0, a low q0
and a relatively low H0. The chief result is that a high (and therefore cautious)
odds-of-no-misuse threshold for invoking the prohibition of research requires a
belief that prohibition results in only a very small decrease in utility relative to
the improvement in the (dis)utility of misuse. As in the legal standard-of-proof
case, this difficulty arises because we have only two possible acts. A way around
this is to introduce a third act (middle option). Let us call it Regulate. Table
11.4 shows the utility set-up for this threefold partition.
Table 11.4 Threefold Partition of Acts
R2
R1
R0
Laissez F.
Regulate
Prohibit
No misuse
H2 = 1
H1
H0
Misuse
G2 = 0
G1
G0
The w01 threshold now determines when the Regulate option is chosen over
Prohibit, and a new threshold, w12, determines when Laissez-Faire is chosen
over Regulate. Now, H1 > (y1 q1)/y1 implies
which in turn implies

H0 > 1 (G1 q1)/w12 q1/y1.
177
Setting w12 = 5 and G0 = 0.5, for instance, and using the settings q1 = 0.1 and
y1 = 10, gives
H0 > 1 (0.5 0.1)/5 0.1/10 = 0.91
If we are willing to lower the threshold to w12 = 2 and increase G0 to 0.68 then
H0 > 1 (0.68 0.1)/2 0.1/10 = 0.7.
The threefold partition therefore can express a belief that outright prohibition
could substantially negatively affect research (in this last example, a decline in
utility from 1 to 0.7). Nevertheless, there are limits if we take certain additional
constraints into account. It seems reasonable to stipulate that misuse cannot
yield a greater utility than no misuse, so we impose the constraint G0 < H0. As
mentioned earlier, the case where G0 = H0 corresponds to the situation where
prohibition of research eliminates the possibility of misuse of its outputs, so that
there is no difference between the no misuse and misuse states. The restriction
G0 < H0 and the constraint w01 = 1 imply that H0 > 1/2. Higher odds thresholds
increase the lower bound on H0. It is easy to prove that the general relationship
is w01 = x implies H0 > x/(x + 1) = p01, the corresponding probability threshold.
In the two examples above, w01 = 5 implies H0 > 5/6 and w01 = 2 imply H0 > 2/3.
Thus, extreme cases where prohibition of further research would hardly alter
the (dis)utility of misuse of an existing technology impose severe restrictions on
the utility if there is no misuse. Table 11.5 shows a set-up like this, with similar
low values of G1 and G2. We would be inclined to set the odds thresholds w01
and w12 to be very highsay, w01 = 100 and w12 = 1000. The result would be
that H1 and H0 both would be very close to 1: H1 = 0.99999 and H0 = 0.99989.
Therefore, a substantial difference between H1 and H0 (say, due to the inhibition
of scientific progress) can only arise if there is a substantial difference between
G1 and G0 and relatively low threshold odds of no misuse, w01.
Table 11.5 Extreme Disutility of Misuse
R2
R1
R0
Laissez F.
Regulate
Prohibit
No misuse
H2 = 1
H1
H0
Misuse
G2 = 0
G1= 0.01
G0= 0.02
Are there sets of utilities and threshold odds that could satisfy the intuition
that some security measures should be in place when there is only a very small
chance of misuse, but that severe restrictions on research will have a substantial
impact on scientific progress? What would these look like? Table 11.6 illustrates
a set-up similar to an earlier example that is compatible with these intuitions.
178
The bottom row shows the odds thresholds. Resetting w12 to values greater than
10 has relatively little impact on w01 (or alternatively on utilities G0 and H0 if
we wish w01 to remain at 2) because H1 is already close to 1 and G1 is close to 0.
And of course it is possible to solve for H1 and G1 such that w12 takes a specific
value greater than 10 while w01 is unaffected and remains at 2. Thus, in the
threefold partition of acts we are free to set w12 to very conservative (high)
values while still retaining flexibility regarding w01 or the utilities it comprises.
Table 11.6: Extreme Disutility of Prohibition
R2
R1
R0
Laissez F.
Regulate
Prohibit
No misuse
H2 = 1
H1= 0.99
H0= 0.6933
Misuse
G2 = 0
G1= 0.1
G0= 0.6933
Odds threshold
w12 = 10
w01 = 2
The set-up is greatly affected, however, by changes in w01 because of the

relationship described earlier between w12 and the lower bound on H2.
Increasing w01 from 2 to 5, as mentioned earlier, raises the lower bound on H0
from 2/3 to 5/6. Preferences and intuitions regarding these effects will need to
be guided by a sense of how harmful potential misuses are under prohibition
versus regulation versus laissez faire in comparison with the loss of potential
knowledge and benefits when research is prohibited versus regulated. These
comparisons are admittedly not easy to make, let alone quantify. Nevertheless,
decisional thresholds do need to be set, and setting them in a considered manner
requires comparisons of this sort.
Therefore some considerations about utility scales are appropriate to conclude
this subsection. The utility scales used here are not absolute, or even ratiolevel. They have neither an absolute zero nor a fixed upper bound. At best,
they are interval-level scales, meaning that the difference between two utility
assignments (for example, H2 H1) is a ratio-level scale. Recall that a ratio
comparison of two such differences, (Gj1 Gj)/(Hj Hj1), determines the odds
threshold wj1j. Smithson21 defines two kinds of risk-orientation bias in the
utility differences when utilities are restricted to the [0, 1] interval. A-bias is
measured by the sum of the log of the odds thresholds and refers to greater risk
aversion to one outcome than the other. In our examples thus far, all wj1j > 1,
indicating greater risk sensitivity to misuse than to no misuse. R-bias, on the
other hand, is measured by

179
and compares gains and losses in utility as the decision-maker moves from one
act to another. A positive sum indicates greater risk sensitivity in choosing
between acts for high js and a negative sum indicates greater risk sensitivity
in choosing between acts with low js. In Table 11.6 these log ratios are 3.39
and 1.78, so there is greater risk sensitivity in choosing between Regulate and
Prohibit than between Laissez Faire and Regulate. This is simply due to the
greater changes in H and G utilities as we move from Regulate to Prohibit.
Finally, given that the utility scales have no absolute lower or upper bounds, a
reasonable question to ask is whether some bounds are more useful or sensible
than others. The [0, 1] interval probably is not well suited to human judgments
because it lacks two features that have psychological significance: a reference
point representing the status quo and a distinction between being better off
or worse off than the status quo. A well-established empirical and theoretical
literature22 informs us that people judge the utility of future outcomes relative
to a reference point (usually the status quo) instead of in absolute terms, and
that they are more sensitive to losses than to gains.
Table 11.7 presents one way of rescaling Table 11.6 according to these
considerations. Suppose we assign 0 to represent the status quo and represent
the maximal loss by 100. Suppose also that we believe misuse of a research
output under laissez faire would yield a loss that is 10 times the magnitude of
the gains that could be realised if no misuse occurred. Then G2 = 100$ and
H2 = 10. The odds thresholds in Table 11.6 partially determine the remaining
utility assignments. We require more than one constraint, so let us repeat the
loss due to misuse being 10 times the gain with no misuse under Regulate. The
end result reveals that we believe we will be worse off than the status quo under
the Prohibit option whether there is misuse or not, but that will be our best
option if the odds of misuse are shorter than 2 to 1.
Table 11.7 Rescaled Utilities from Table 11.6
R2
R1
R0
Laissez F.
Regulate
Prohibit
No misuse
H2 = 10
H1= 9.9
H0= 26.4
Misuse
G2 = 100
G1= 99
G0= 26.4
22 Beginning with Kahneman, D. and Tversky, A. 1979, Prospect theory: an analysis of decision under
risk, Econometrica, vol. 47, pp. 26391.
180
Imprecision and bias in judgments

Probability and utility judgments regarding dual-use dilemmas ultimately must
be made by human judges, and this last section discusses the most important
issues regarding human judgments of this kind. We begin by considering issues
of imprecision and conflict in judgments, and subsequently discuss relevant
human tendencies towards overconfidence in predictions and confirmation bias.
Even when it is foreseeable, the probability of the misuse of a new technology
or research output and the severity of its consequences almost never are known
precisely, nor is there usually a consensus on their magnitudes. Imprecision
and conflict are very likely to pervade judgments of probability and utility
in dual-use dilemmas. These uncertainties must not be denied or ignored;
decision-makers will treat falsely precise estimates as if they really are precise
and decisions based on them will be far from robust. At the very least, decisions
and their criteria should be subjected to sensitivity analyses to ascertain which
components are the most affected by altering parameter values. In the preceding
section, for instance, we saw that the three-option set-up in Table 11.6 was
robust against changes in w12 but sensitive to changes in w01.
I shall leave conflict aside as even a brief treatment of it is beyond the scope of
this chapter, except to note in passing that some psychological investigations
indicate that people prefer dealing with vague but consensual opinions to
precise but disagreeing ones.23 Thus, imprecision is viewed as a less severe kind
of uncertainty than conflict.
Nevertheless, imprecision complicates decision-making. A precise probability
assigned to the misuse of a technology either exceeds or fails to exceed a
decisional threshold of the kind discussed in the preceding section, so the choice
among alternatives is clear. Precise probabilities bring decisiveness with them;
however, a probability interval may lie entirely below or above the threshold,
or may include it. Standard decision frameworks for imprecise probabilities
treat the lower bound as the probability to use in betting on misuse and the
upper bound as the probability to use in betting against it. Therefore, these
frameworks claim there is no basis in the probabilities themselves for preferring
the alternative on either side of a decisional threshold if the probability interval
straddles it.
Suppose, for instance, that the set-up in Table 11.6 is our decisional guide
and we are confronted with a potential technological development for which
23 Smithson, M. 1999, Conflict aversion: preference for ambiguity vs. conflict in sources and evidence,
Organizational Behavior and Human Decision Processes, vol. 79, pp. 17998; Smithson, 2006, op. cit., pp.
33964; and Cabantous, L. 2007, Ambiguity aversion in the field of insurance: insurers attitude to imprecise
and conflicting probability estimates, Theory and Decision, vol. 62, pp. 21940.
181
experts estimate the odds of misuse to be somewhere between 5 and 50 to 1.

This interval includes the threshold w01 = 10, so should we choose Regulate or
Laissez Faire? If we can defer this decision pending more information, should
we do so? This issue is an active topic of research and attempting a resolution of
it is beyond the scope of this chapter, but the main purpose in raising it here is
to point out that because imprecision really matters, decision-makers must work
out how they will treat imprecise estimates differently from precise ones.
We now turn to probability judgments themselves. There is a large body of
empirical and theoretical work on subjective probability judgments, but
discussion will be restricted to just two judgment biases that are directly
relevant. The first of these is probability weighting, which may be summarised
by saying that people overweight small and underweight large probabilities.
Note that this does not mean that people are necessarily under or overestimating
the probabilities, but instead treating them in a distorted fashion when making
decisions based on them. Rank-dependent expected utility theory24 reconfigures
the notion of a probability weighting function by applying it to a cumulative
distribution whose ordering is determined by outcome preferences. Cumulative
prospect theory25 posits separate weighting functions for gains and losses.
Two explanations have been offered for the properties of probability weighting
functions. The first26 is diminishing sensitivity to changes that occur further
away from the reference points of 0 and 1. The second is that the magnitude
of consequences affects both the location of the inflection point of the curve
and its elevation. Large gains tend to move the inflection point downward and
large losses move it upward.27 Diminishing sensitivity has an implication for
judgments and decisions based on imprecise probabilities as well as precise
probabilities. A change from 0.01 to 0.05 is seen as more significant than
a change from 0.51 to 0.55, but a change from 0.51 to 0.55 is viewed as less
significant than a change from 0.95 to 0.99. An implication is that for decisional
purposes people might view a probability interval [0.01, 0.05] as less precise
than [0.51, 0.55], and so on. The prospect of large losses (as in the misuse of
biotechnology) will exaggerate these effects for low probabilities. This issue is
important for dual-use dilemmas because at least some of the possible outcomes
under consideration will have extreme probabilities attached to them.
The second relevant bias concerns confidence judgments and the elicitation
of prediction or confidence intervals from human judges. Numerous studies
24 For example, Quiggin, J. 1993, Generalized Expected Utility Theory: The Rank Dependent Model, Kluwer,
Boston.
25 Tversky and Kahneman, op. cit., pp. 112431.
26 Camerer, C. F. and Ho, T. H. 1994, Violations of the betweenness axiom and nonlinearity in probability,
Journal of Risk and Uncertainty, vol. 8, pp. 16796.
27 For example, Etchart-Vincent, N. 2004, Is probability weighting sensitive to the magnitude of
consequences? An experimental investigation on losses, Journal of Risk and Uncertainty, vol. 28, pp. 21735.
182
demonstrate that both novices and experts tend to be overconfident in the

sense that they construct prediction intervals that are much too narrow for their
confidence criteria. A typical discrepancy is that when asked to construct an
interval that has a 90 per cent probability of including the correct prediction
the actual hit rate is less than 50 per cent.28 Recent findings,29 however, have
suggested that when presented with prediction intervals, people do not
overestimate their coverage rates. The take-home lesson from this literature is
that asking experts to estimate how likely is the probability of, say, the theft of
smallpox supplies from a particular source between two values will yield more
well-calibrated results than asking the experts to construct, say, a 95 per cent
confidence interval for that probability.
Finally, the catch-all underestimation bias described earlier is a special case of
confirmation bias. This is a largely unconscious tendency in human information
processing and judgment such that people seek out and pay more attention to
information that confirms their beliefs than to disconfirming information. In the
catch-all underestimation bias, confirmation bias manifests itself as a tendency
to underestimate the likelihood of novel or unanticipated events. Unfortunately
these are exactly the kinds of events that policy planners and decision-makers
must be on the lookout for in dealing with dual-use dilemmas. There are few
recommendations on record for militating against confirmation bias. One is to
construct inclusive teams containing members with diverse backgrounds and
viewpoints and ensure that decision-makers and planners listen attentively
to those members with whom they disagree; however, this seemingly obvious
strategy is complex and deceptively difficult to implement.30 Another is the
use of formal analyses, simulations and models to reveal consequences or
possibilities that our preconceptions render invisible to us.31
Conclusions?
This chapter largely neglects ethical considerations, which may seem odd given
the predominantly ethical nature of dual-use dilemmas. Ethical considerations
have been set aside to enable a focus on some prerequisites for a fine-grained
analysis of dual-use dilemmasnamely, a systematic investigation of specific
28 For example, Russo and Kolzow, op. cit., pp. 1732.
29 For example, Winman, A., Hansson, P. and Juslin, P. 2004, Subjective probability intervals: how to cure
overconfidence by interval evaluation, Journal of Experimental Psychology: Learning, Memory, and Cognition,
vol. 30, pp. 116775.
30 See, for example, Brown, V. A. 2010, Collective inquiry and its wicked problems, in V. A. Brown,
J. Russell and J. Harris (eds), Tackling Wicked Problems through the Transdisciplinary Imagination, Earthscan,
London, pp. 6184.
31 See Lempert, R., Popper, S. and Bankes, S. 2002, Confronting surprise, Social Science Computer Review,
vol. 20, pp. 42040.
183
unknowns that such an analysis would have to contend with. My hope is that
ethicists will find useful guidance in this investigation, avoiding some of the
pitfalls and traps awaiting the unwary.
Some pertinent unknowns have not been dealt with here, so this chapter cannot
be taken as anything like an exhaustive survey. Nevertheless, we have examined
types of unknowns that are beyond the purview of standard decision theories,
such as state space indeterminacy and imprecision. We have seen that there
are genuinely different kinds of unknowns, not just different sources of the
same kind, and that these play distinct roles. One of the key emergent points
is that many of the unknowns in dual-use dilemmas (and in so-called wicked
problems) are interconnected. They can be traded against one another, and how
one unknown is dealt with has ramifications for other unknowns. Allowing
imprecision in probability assignments, for instance, offers a way of handling
state space indeterminacy. Conversely, choosing the right number of options
can rectify incompatibilities between preferences and decisional probability
thresholds.
We may never be able to attain precise quantification of costs, benefits and
probabilities of outcomes arising from dual-use dilemmas, so a fine-grained
analysis in that sense also is unachievable. After all, accidental findings and
consequences are legion in cutting-edge research and development, and so
irreducible unknowns such as the catch-all underestimation problem are likely
to dog policy formation and decision-making alike. Moreover, as I have argued
elsewhere,32 if we value creativity, discovery and/or entrepreneurship then we
shall have to tolerate at least some irreducible unknowns.
Nevertheless, as Head33 has pointed out, great uncertainty alone is not sufficient
to render a problem wicked in the sense used in most of the literature on
that topic. Wickedness also requires complexity and divergent or contradictory
viewpoints about the nature of the problem and preferences regarding alternative
outcomes. I have tried to show here that even rather simple formal analyses in
the form of thought experiments can frame and structure dual-use dilemmas in
useful ways that avoid some aspects of wickedness, so that at least some of our
psychological foibles can be taken into account and even overcome.
32 Smithson, M. 2008, The many faces and masks of uncertainty, in G. Bammer and M. Smithson (eds),
Uncertainty and Risk: Multidisciplinary Perspectives, Earthscan, London, pp. 1326.
33 Head, B. W. 2008, Wicked problems in public policy, Public Policy, vol. 3, pp. 10118.
184
12. Moral Responsibility, CollectiveAction Problems and the Dual-Use

Dilemma in Science and Technology1
Seumas Miller
The dual-use dilemma

The so-called dual-use dilemma arises in the context of research in the sciences
as a consequence of one and the same discrete piece, or ongoing program of
scientific research, intentionally undertaken for good ends having the potential
to be intentionally used to cause great harm.2 So there is a primary user who
creates new knowledge or designs new technology for goodfor example,
discovers how to aerosolise chemicals for use in crop-dusting. But there is also
a secondary user who uses the knowledge or technology for some harmful
purposefor example, uses the newly discovered process of aerosolisation to
weaponise chemicals.
Many, if not most, so-called dual-use dilemmas are not really dilemmas in the
narrow sense of being situations involving two options that are equally ethically
problematic. In the first place, the dilemmas in question could be tri-lemmas;
indeed, there could be four or five or some very large number of options, all
of which are equally ethically problematic. In the second place, the options
are not generally equally ethically problematic. Certainly, there are ethical
considerations for and against each of the options, however, it may well be that,
all things considered, one of the options is morally preferable to the others and
that this is relatively obvious to any rational, morally sensitive person. The
point is rather that there are at least some significant moral costs associated with
each of the available options.
Naturally, many, if not most, scientific discoveries and especially new technologies
have dual-use potential in the trivial sense that they could be used by someone
for some malevolent purpose. Indeed, any newly designed object, such as the
first baseball bat, has dual-use potential in this trivial sense. After all, baseball
1 This chapter is a condensed version of material provided by the author in Work Package 3 (WP3) of the
European Unions Seventh Framework project, Synth-Ethics. (WP3 was jointly authored by Seumas Miller
and Michael Selgelid.)
2 Miller, S. and Selgelid, M. 2008, Ethical and Philosophical Consideration of the Dual Use Dilemma in the
Biological Sciences, Springer, London, ch. 1.
185
bats can be used to hit people over the head, as well as for the enjoyment of
playing baseball; however, it is implicit in the use of the term dual use in play
in the academic literature that the potential harm in question is of a very great
magnitudefor example, the potential of nuclear fusion to lead to the creation
of the hydrogen bomb, and the potential of genetic engineering to lead to a
super virus. Moreover, it is also implicit that the dual-use potential in question
is not simply a repeat application of existing science or technologyas is the
case with the construction of the first baseball batbut genuinely new science
or technology that can be used to provide a qualitatively or quantitatively new
means of harming, for example: the invention of the first explosives, or creating
a highly virulent form of an existing much less virulent pathogen. Hitting a
person over the head with a solid object has been done since time immemorial,
so hitting someone over the head with a baseball bat is hardly a novel means
of harming. By contrast, blowing someone up with gunpowder was initially a
novel means of harming. Moreover, unlike the use of a baseball bat, the use of
an explosive device, such as a 10-tonne bomb, could harm on a very large scale.
Note that accidents involving science and technology, even accidents on a very
large scale such as the Chernobyl disaster, are not dual use in my sense since there
is no secondary evil user, although they may involve unethical behaviour such
as negligence with respect to safety precautions. Nor are weapons designed as
weaponsfor example, gunsinstances of dual-use science and/or technology.
For even if their harmful use is intended to be ultimately for the good, such
weapons are in the first instance designed to harm; their use to harm is not a
secondary, but rather a primary, use.
One paradigmatic case of dual-use research was the biological research on the
mousepox virus. The dilemma was as follows.
Option 1: The scientists ought to conduct research on the mousepox virus
and do so intending to develop a genetically engineered sterility treatment
that combats periodic plagues of mice in Australia.
Option 2: The scientists ought not to conduct the research since it might lead
(and in point of fact did lead) to the creation of a highly virulent strain of
mousepox and the consequent possibility of the creationby, say, a terrorist
group contemplating a biological terrorist attackof a highly virulent strain
of smallpox able to overcome available vaccines.
It is a dilemma since there are two options with good reasons in favour of both
and it is an ethical dilemma since these options are morally significant. In
essence the dilemma involves a choice between intentionally doing good and
foreseeably providing others with the means to do evil.
A second and more recent paradigm of dual-use research is the biological
research done on a deadly flu virus (A [H5N1]), which causes bird flu. Scientists
186
12. Moral Responsibility, Collective-Action Problems and the Dual-Use Dilemma in Science and Technology
in the United States and the Netherlands created a highly transmissible strain
of this virus, albeit, as it emerged, a strain that is not as deadly as ordinary
H5N1. Crucially, the work was done on ferrets, which are considered a very
good model for predicting the likely effects on humans.
As with the mousepox case, there are two options, ethically speaking.
Option 1: The scientists ought to conduct research on the bird flu virus and
do so intending to develop vaccines against similar, but deadly, naturally
occurring and artificially created strains of H5N1.
Option 2: The scientists ought not to conduct the research since it will lead
to the creation of a virus that is transmissible to humans, and is unlikely to
lead to the development of vaccines against similar, but deadly, naturally
occurring and artificially created strains of H5N1.
Notice here that the scientific claim in option two that the research is unlikely
to lead to the development of vaccines contradicts the corresponding claim in
option one. In this respect the dilemma is unlike that posed by the mousepox
research. Moreover, if this scientific claim made in option two is correct then the
justification offered in option one for conducting the research collapses.
In such dual-use cases, the researchersif they go ahead with the research
will have foreseeably provided the means for the harmful actions of others and,
thereby, arguably infringed a moral principle (albeit their infringement might
in some cases be morally justified). The principle in question is the principle of
what we might refer to as the no means to harm principle.3 Roughly speaking,
this is the principle that rules out providing malevolent persons with the means
to do great harma principle that itself ultimately derives from the more basic
principle: do no harm.
The no means to harm principle (NMHP) is the principle that one should not
foreseeably (whether intentionally or unintentionally) provide others with the
means to intentionally do great harm and it assumes: 1) the means in question
is a means to do great harm; and 2) the others in question will do great harm,
given the chance.
As with most, if not all, moral principles, the NMHP is not an absolute principle
and, therefore, it can be overridden under certain circumstances. For example,
it is presumably morally permissible to provide guns to the police in order that
they can defend themselves and others. Moreover, as is the case with most, if
3 This principle, or similar ones, is familiar in a variety of ethical contexts. See, for example, Scanlon, T.
1977, A theory of freedom of expression, in R. M. Dworkin (ed.), The Philosophy of Law, Oxford University
Press, Oxford.
187
not all, moral principles, the application of the NMHP is very often a matter of
judgment. In the case of NMHP the need for judgments depends in large part on
the uncertainty of future harms.
The dual-use dilemma is a dilemma for researchers, governments, the community
at large, and for the private and public institutions, including universities and
commercial firms that fund or otherwise enable research to be undertaken.
Moreover, in an increasingly interdependent set of nation-statesthe socalled global communitythe dual-use dilemma has become a dilemma for
international bodies such as the United Nations. The dilemma is perhaps most
acute in those areas of science and technology that operate on an engineering
or construction modelfor example, synthetic biology, nanotechnology (as
opposed to a description model, which restricts itself to the description of
pre-existing entities and their causal and other relationshipsfor example,
astronomy).
Scientific research and weapons of mass

destruction
The history of science and technology is replete with examples of scientific
research being used intentionally or unintentionally to create weapons,
including weapons of mass destruction (WMD).
Scientists have developed biological, chemical and nuclear weapons. Such
weapons include the following historical examples: the mustard gas used by
German and British armies in World War I; the aerial spraying of plagueinfested fleas by the Japanese military in World War II that killed thousands of
Chinese civilians; the dropping of atomic bombs on Hiroshima and Nagasaki by
the US Air Force in World War II; the large-scale biological weapons program
in the Soviet Union from 1946 to 1992; the biological weapons program of the
apartheid government in South Africa; and the use of chemical agents against
Kurds by Saddam Husseins Iraqi regime in 1988.
In recent years there have been a number of high-profile defections of scientists
from Western countries to authoritarian states with WMD programs. For
example, Dr Abdul Qadeer Khan joined and in large part established Pakistans
nuclear weapons program after working for Urenco in the Netherlands, and
Frans van Anraat (also from the Netherlands) went to Iraq to assist Saddam
Husseins WMD program in producing mustard gas.
Moreover, in recent years there have been a number of acts, or attempted acts,
of bioterrorismnotably, by the Aum Shinrikyo in Japan (they attempted to
188
acquire and use anthrax and botulinum toxin, and actually carried out a number
of terrorist attacks using the chemical sarin gas), by al-Qaeda (they attempted
to acquire and use anthrax) and by US Government employees in the so-called
Amerithrax attacks (involving the actual use of anthrax).
Given that a small number of animal, human and plant pathogens are readily
obtainable from nature, and that bioterrorists with some microbiological training
could use these to inflict casualties or economic damage, and especially given
the new possibilities provided by synthetic biology (involving the creation of
pathogens de novo), evidently there is a non-negligible bioterrorist threat, and
it is increasing.4 This threat is perhaps greatest in unregulated environments,
including in weak or failing states in which well-resourced international
terrorist groups are allowed to flourish.
Moreover, it would be naive to assume that the scientific community can be
entirely trusted to regulate itself in relation to dual-use problems. After all,
thousands of scientists have worked in the abovementioned and other WMD
for example, biological weaponsprograms, and in doing so have had as their
institutional collective end the production of biological weapons. Accordingly,
these scientists are directly collectively morally responsible for the existence
of those weapons and, in the case of scientists working for authoritarian
governments, for enabling authoritarian regimes to possess them. Moreover, on
some occasions, as already noted, WMD have actually been used; accordingly,
the scientists involved in the development of these WMD are morally implicated,
even if only indirectly, in the harms caused by such use.
Scientific freedom and joint action

According to scientist cum philosopher Michael Polanyi:
The existing practice of scientific life embodies the claim that freedom
is an efficient form of organization. The opportunity granted to mature
scientists to choose and pursue their own problems is supposed to result
in the best utilization of the joint efforts of all scientists in a common
task. In other words: if the scientists of the world are viewed as a team
setting out to explore the existing openings for discovery, it is assumed
that their efforts will be efficiently coordinated if only each is left to
follow his own inclinations. It is claimed in fact that there is no other
4 Miller, S. 2009, Terrorism and Counter-Terrorism: Ethics and Liberal Democracy, Blackwell, London, ch. 7.
189
efficient way of organizing the team, and that any attempts to coordinate
their efforts by directives of a superior authority would inevitably
destroy the effectiveness of their cooperation.5
Polanyis view is that each scientist acts freely but does so
1. on the basis of the work of past scientists
2. with constant reference and adjustment to the work of other contemporary
scientists
3. in the overall service of a collective end of comprehensive knowledge (in the
sense of understanding) of the scientific phenomena in question.
So his conception is one of individual scientific freedom in the overall context
of intellectual interdependence in a joint epistemic project.
The epistemic character of scientific research is obvious: science aims at
knowledge or, more broadly, at understanding. What is perhaps not quite so
obvious, however, is that scientific work is, therefore, epistemic action. We
often contrast the acquisition of beliefs, knowledge and other cognitive states
with actionthe notion of action being reserved for behaviour or bodily action.
Action is intentional and voluntary; on the other hand, knowledge acquisition, it
is often held, is unintentional and involuntary. Consider perceptual knowledge.
If an object, say, a piece of litmus paper placed in a liquid solution in a test tube,
turns blue before our eyes then we come to believe that it has turned blue; we
have no choice in the matter, or so it seems.
This focus, however, on perceptual knowledge of ordinary middle-sized objects
provides a distorted picture of the scientific epistemic project for reasons that
will become clear in what follows. A further point to be made here picks up on
the joint or cooperative nature of scientific work described above. If scientific
research is epistemic action and, at the same time, it is joint action then scientific
research is joint epistemic action.
Naturally, this being an epistemic academic project, the participants are governed
by epistemic principles (for example, replication of experiments), peer-review
processes to ensure quality control and uncensored publication.
The key notion in play here is what I am referring to as joint epistemic action.
What is joint action?
Joint action consists of multiple individual actions performed by multiple agents
and directed towards a collective endfor example, a team of workers building
the Empire State Building; a team of terrorists destroying the Twin Towers,
5 Polanyi, M. 1951, The Logic of Liberty: Reflections and Rejoinders, Routledge & Kegan Paul, London, p. 34.
190
killing thousands; and a team of scientists discovering the cure for cancer.6 A
collective end is an individual end that each of the participating agents has, but
it is an end that no one agent acting alone realises on his or her own. So each
agent acts interdependently with the other agents in the service of the same,
shared end: the collective end. Again, consider the collective end of a security
organisation such as the Federal Bureau of Investigation (FBI) whose members
may be jointly working to prevent harmnotably, great harms planned by
criminal organisations such as terrorist groups; or consider a team of scientists
working feverishly to develop an antidote for some infectious disease that is
reaching epidemic proportions.
Joint actions exist on a spectrum. At one end of the spectrum there are joint
actions undertaken by a small number of agents performing a one-off simple
action at a moment in timefor example, two lab assistants lifting some
equipment onto a bench. At the other end of the spectrum there are large
numbers of institutionally structured agents undertaking complex and often
repetitive tasks over very long stretches of historyfor example, those who
built the Great Wall of China, or biological scientists developing vaccines.
Joint activity within institutions typically also involves a degree of competition
between the very same institutional actors who are cooperating in the joint
activity. For example, life scientists within a university who are engaged in joint
research and teaching activity in the service of the universitys institutional
goals are also competing for academic status within the relevant international
scientific community, and competing in part on the basis of their individual
contribution to those same institutional goals.
Moreover, in many institutional settings organisations compete with one
anotherfor example, business organisations in market settings. Here there is
joint activity at a number of levels. For one thing, each competing organisation
(for example, a single corporation) comprises a team of individual agents who
are cooperating with one another and jointly working to secure the collective
ends of the organisation (for example, a biotech company trying to maximise
market share). For another thing, each team (for example, each corporation) is
engaged in joint compliance with the regulatory framework that governs their
competitive market behaviourthat is, each complies with (say) the regulations
of free and fair competition interdependently with the others doing so and in
the service of ensuring the ongoing existence of the market in question. This
is consistent with the existence of a regulator who applies sanctions to those
organisations which breach the regulations, including safety regulations that
might be regarded as a costly and unnecessary impost on business (for example,
the select agent rule, some mandatory safety procedures in laboratories, and
6 Miller, S. 2001, Social Action: A Teleological Account, Cambridge University Press, Cambridge, ch. 1.
191
screening all orders of DNA sequences); the last compliance mechanism is an

add-on to the fundamental underlying structure of interdependence of action
in the service of collective ends that is constitutive of market mechanisms.7
Let us consider further the notion of a joint action and the correlative notion of
a collective end. As stated above, joint actions involve multiple agents with the
same endfor example, to build a house (the collective end) or map the human
genome. Note the following points.
First, each agents individual action is a (possibly small) causal contribution
to the collective endfor example, building the Great Wall of China, or
mapping the human genome.
Second, each agents individual action or omission is performed on
condition that others perform their contributory actions/omissions; there is
interdependence of action.
Third, each has the collective end only on condition others have the collective
end; there is interdependence of ends.
Fourth, what the collective end is and that it is being pursued are matters of
mutual true belief among the participants (A and B mutually truly believe
that p if and only if A believes truly that p, B believes truly that p, A believes
that B believes that p, and so on).8
Fifth, collective ends are purely conative states; they are not affective states
such as feelings or desires. Accordingly, we need to distinguish the mental
states constitutive of joint actions (that is, intentions, ends and beliefs)
from the mental states that might motivate some joint actions (for example,
feelings and desires).
Joint actions can realise collective ends that are also goodsnamely, collective
goods. Examples of such collective goods are a law-abiding society and
an economically viable biotechnology sector. At an organisational level, a
collective good might be the realisation of a collective end that consists in harm
minimisation or prevention. Thus a biotechnology firm might have as one of
its collective ends to avoid any major industrial accidents or to prevent any
serious security breaches. Note that in my sense of the term collective good,
a collective good is simply a good that is produced by joint action directed
to a collective end the realisation of which consists in the provision of that
good. Such joint action includes action that consists in joint compliance with
safety and security procedures that has as a collective end the collective good
of prevention of harm.
7 Miller, S. 2010, The Moral Foundations of Social Institutions: A Philosophical Study, Cambridge University
Press, Cambridge, pp. 502, and ch. 2.
8 Miller, 2001, op. cit., pp. 569.
192
Such collective goods are not necessarily reducible to an aggregate of individual

benefits. Relational goods produced by joint activitysuch as social harmony
and mutual scientific knowledge (each knows that p and each knows that each
knows that p, and so on)are cases in point.
Some collective goods are intrinsic goods and are jointly pursued for their own
sakethat is, they are not pursued merely as a means to an individual end.
Various kinds of collective interest, such as the national interest or the interests
of the scientific community, are examples of this.
Moreover, a belief in the value of collective goods can motivate action irrespective
of individual self-interestfor example, a soldier giving his life in the national
interest, or a whistleblower blowing the whistle on a secret biowarfare program.
It will be evident from the above that I am distinguishing between self-interested
reasons (or motives) for individual and joint actionfor example, so-called
sticks and carrotsand moral reasons (or motives) for action (including joint
action), and claiming that the latter are not reducible to the former (and vice
versa).
It will be further evident that I hold that moral reasonsfor example, a belief
that biological warfare is wrong, or a belief in the common goodcan motivate
in and of themselves. So individual self-interest is not the only motive for action.
Moreover, even when the motive of self-interest is present, which it obviously
typically is, it is not necessarily the dominant motivation.
Armed with this general characterisation of joint action, collective goods and
so on, let us now turn to joint epistemic actionthe salient form of joint action
involved in scientific activity.
Joint epistemic action

As we have seen, epistemic actions are actions of acquiring knowledge. Here
we can distinguish between so-called knowledge-that and knowledgehow, the former being propositional knowledge (knowledge of the truth
of some proposition), the latter being practical knowledge (knowledge of
how to undertake some activity or produce some artefact). The definition of
propositional knowledge, in particular, is philosophically controversial, but let
us assume for our purposes here that someone, A, has knowledge that p if and
only if A has a true belief that p and A has a justification for believing that p,
which does not rely on some other false belief.9
9 See, for example, Moser, P. 1989, Knowledge and Evidence, Cambridge University Press, Cambridge.
193
The methods of acquiring propositional knowledge are manifold but for scientific
knowledge they include observation, calculation and testimony. Moreover, the
acquisition of these methods is very often the acquisition of knowledge-how
for example, how to calculate, how to use a microscope, how to read an x-ray
chart.
In the case of the engineering sciences there is an even more obvious and
intimate relationship between propositional and practical knowledge, since both
are in the service of constructing or making things. Thus in order to build an
aeroplane engineers have to have prior practical (how-to) knowledge and that
practical knowledge in part comprises propositional knowledgefor example,
with respect to load-bearing capacity. Moreover, this engineering model has
increasing applicability in new and emerging sciences such as synthetic biology
and nanotechnology. In the case of synthetic biology, for example, scientists can
develop new vaccines, enhance the virulence and transmissibility of existing
pathogens and even create new pathogens (albeit, presumably using elements of
existing pathogens as building blocks).
What counts as sufficient evidence for the possession of knowledge varies from
one kind of investigation and one kind of investigative context to another.
Thus a scientist seeking a cure for cancer would need his or her experimental
results to be replicated by other scientists and the putative cure (say, a new
drug) would need to be subjected to clinical trials before being made widely
available. A detective investigating a series of murdersfor example, by the
Yorkshire Ripperwill be focused not only on physical evidence but also on
motive (a mental state) and opportunity. Moreover, the evidential threshold for
being found guilty is beyond reasonable doubt.
Whereas the acquisition of practical knowledge is readily seen as emanating
from action and, indeed, as being a species of action (knowledge-in-action), the
acquisition of propositional knowledge is a different matter; however, coming to
truly believe that p on the basis of evidencethat is, propositional knowledge
acquisitionis action in at least three respects.
First, the agent, A, makes a decision to investigate some matter with a view
to finding out the truth; the action resulting from this decision is epistemic
action. For example, a detective intentionally gathers evidence having as an
end to know who the serial killer of prostitutes in Yorkshire isthat is, who the
Yorkshire Ripper is. Thus the detective gathers physical evidence in relation to
the precise cause and time of death of the Rippers victims; the detective also
interviews people who live in the vicinity of the attacks, and so on. Here A has
decided that A will come to have a true belief with respect to some matter, as
opposed to not having any belief with respect to that matterfor example, a
true belief with respect to who the Yorkshire Ripper is. As decision is between
194
coming to have true belief and being in a state of ignorance, and in conducting
the investigation A has decided in favour of the former. Similarly, a scientist
seeking to discover the genetic structure of some organism makes a decision to
come to have a true belief with respect to this matter rather than remaining in
ignorance.
Second, the agent, A, intentionally makes inferences from As pre-existing
network of beliefs; these inferences to new beliefs are epistemic actions. For
example, a forensic scientist might infer the time of death of a murder victim on
the basis of her prior belief that rigor mortis sets in within 10 hours after death.
Third, in many cases A makes a judgment that p in the sense that when faced
with a decision between believing that p and believing that not p, A decides
in favour of p; again, A is performing an epistemic action. For example, our
detective, A, intentionally makes an evidence-based judgment (mental act) that
Sutcliffe is the Yorkshire Ripper (as opposed to that Sutcliffe is not the Yorkshire
Ripper) and does so having as an end the truth of the matter. Here, A is deciding
between believing that p and believing that not p; but A is still aiming at truth
(not falsity). A is not deciding to believe what he thinks is false. Similarly, our
forensic scientist makes an evidence-based judgment in relation to the cause of
death of the victim having as an end the truth of the matter. Here the scientist
is deciding between believing that the cause of death was x and believing that
the cause of death was not x (but was, say, y).
As is the case with non-epistemic action, much epistemic actionwhether it
is propositional or practical epistemic action or, more likely, an integrated mix
of bothis joint action, that is, joint epistemic action. Joint epistemic action is
knowledge acquisition involving multiple epistemic agents seeking to realise a
collective epistemic end. For example, a team of scientists seeking knowledge of
the cure for cancer is engaged in joint epistemic action.10
In cases of joint epistemic action there is mutual true belief among the epistemic
agents that each has the same collective epistemic endfor example, to discover
the cure for cancer. Moreover, there is typically a division of epistemic labour.
Thus in scientific cases some scientists are engaged in devising experiments,
others in replicating experiments, and so on. So, as is the case with joint action
more generally, joint epistemic action involves interdependence of individual
action, albeit interdependence of individual epistemic action.
As we saw above, knowledge of the cure for cancer, for example, is joint epistemic
action that involves a collective epistemic end and also involves a division of
epistemic labour and interdependence of epistemic action. The further point
to be made here is that there is interdependence in relation to such collective
10 See Miller, 2010, op. cit., ch. 11.
195
epistemic ends. This is because, given the need for replication of experiments
by others, each can only know that p is the cure for cancerto continue with
our examplegiven that others also know this, that is, there is interdependence
in relation to the collective end of knowledge.
A collective epistemic end can be both a collective intrinsic goodand thus
an end in itselfand the means to further ends. Knowledge of the cure for
cancer is a case in point. Such knowledge consists of propositional and practical
knowledge: knowledge of the cure for cancer and knowledge of how to produce
it. This knowledge, however, has as a further (collective) end: the actual
production of the cure (say, a drug). And this end has in turn a still further
endnamely, to save lives.
If knowledge of the cure for cancer is a collective end in itself then it is not
simply a means to individual endsnamely, each having as an end that he/she
knows the cure for cancer. Rather it is mutually believed that knowledge of the
cure for cancer is a collective intrinsic good. In my view, however, moral beliefs
can have motivational force.11 In that case the mutual belief that knowledge of
the cure for cancer is a collective intrinsic good can have motivational force.
It follows from this thatas we saw with joint action more generallyjoint
epistemic action can be collectively self-motivating and does not necessarily
have to rely on prior affective states such as desires.
Layered structures of joint epistemic action

Organisational action typically consists of what elsewhere I have termed a layered
structure of joint actions.12 One illustration of the notion of a layered structure
of joint actions is an armed force fighting a battle. Suppose at an organisational
level a number of joint actions (actions) are severally necessary and jointly
sufficient to achieve some collective end. Consider an army fighting a battle.
Here the action of the mortar squad destroying enemy gun emplacements, the
action of the flight of military planes providing air cover and the action of the
infantry platoon taking and holding the ground might be severally necessary
and jointly sufficient to achieve the collective end of defeating the enemy; as
such, these actions taken together constitute a joint action. Call each of these
actions level-two actions, and the joint action that they constitute a level-two
joint action. From the perspective of the collective end of defeating the enemy,
each of these level-two actions is an individual action that is a component
of a (level-two) joint action: the joint action directed to the collective end of
defeating the enemy.
11 See, for example, Benn, S. 1988, A Theory of Freedom, Cambridge University Press, Cambridge, ch. 2.
12 Miller, 2001, op. cit., ch. 5.
196
Each of these level-two actions is, however, already in itself a joint action with
component individual actions; and these component individual actions are
severally necessary and jointly sufficient for the performance of some collective
end. Thus the individual members of the mortar squad jointly operate the
mortar to realise the collective end of destroying enemy gun emplacements.
Each pilot, jointly with the other pilots, strafes enemy soldiers to realise the
collective end of providing air cover for their advancing foot soldiers. Further,
the set of foot soldiers jointly advances to take and hold the ground vacated by
the members of the retreating enemy force.
At level one there are individual actions directed to three distinct collective
ends: the collective ends of (respectively) destroying gun emplacements,
providing air cover, and taking and holding ground. So at level one there are
three joint actionsnamely, the members of the mortar squad destroying gun
emplacements, the members of the flight of planes providing air cover, and the
members of the infantry taking and holding ground. Taken together, however,
these three joint actions constitute a single level-two joint action. The collective
end of this level-two joint action is to defeat the enemy; and from the perspective
of this level-two joint action, and its collective end, these constitutive actions
are (level-two) individual actions.
Importantly for our purposes here there are layered structures of joint epistemic
action. Consider a crime squad, comprising detectives, forensic scientists and so
on, attempting to solve a crime.
At level one, a victim, A, communicates the occurrence of the crime (say, an
assault) and description of the offender to a police officer, B. But A asserting
that p to B is a joint epistemic action; it is a cooperative action governed by
conventionsthe convention that the speaker, A, tells the truth and the hearer
trusts the speaker to tell the truth.13
Also at level one, a couple of detectives interviews the suspect to determine
motive and opportunity; the detectives are cooperating with one another in the
performance of a joint epistemic action the collective end of which is to discover
motive and opportunity.
Finally, at level one, a team of forensic scientists analyses the available physical
evidencefor example, the DNA of the blood samples of the offender found on
the victim are matched to the suspects DNA; the forensic scientists are engaged
in joint epistemic action to determine whether there is or is not a DNA match.
These three level-one joint epistemic actions are constitutive of a level-two joint
epistemic actionnamely, the level-two joint epistemic action directed towards
13 Miller, 2010, op. cit., ch. 11.
197
the collective end of determining who committed the crime. Accordingly, when
each of the level-one joint epistemic actions is successfully performed then the
level-two joint epistemic action is successfully performedthat is, the crime
squad solves a crime.
Now consider an example of a large scientific project conducted by a number of
cooperating organisations and hundreds of scientists over many yearsnamely,
the human genome project. The project involved multiple connected goals
collective endsand multiple layered structures of joint action, including joint
projects in publishing, undertaken to realise those goals.
In fact most organisations are hierarchical institutions comprising task-defined
roles standing in authority relations to one another, and governed by a complex
network of conventions, social norms, regulations and laws. Consider a science
department in a university or the forensic laboratory in a police organisation:
both comprise heads of department, scientists, laboratory assistants and so on,
and the work of both is governed by scientific norms of observation, replication
of experiments, and so on. So most layered structures of joint action, including
joint epistemic action, are undertaken in institutional settings, and scientific
joint epistemic action is not an exception.
Institutions have de facto purposes/strategic directionsthat is, collective ends,
such as to maximise shareholder profit (corporations), find a cure for cancer
(university research team), or build an atomic bomb (military organisation).
Institutions also have specific structures (hierarchical, collegial and so on)
and they have specific cultures (for example, a competitive, status-driven
ethos).14 In this connection consider scientific activityfor example, biological
research, undertaken in three different institutional settings: that of the
university, the commercial firm and the military biodefence organisation. Some
of the principal purposes/strategic directions (collective ends) of commercial
firmsfor example, to maximise shareholder profitsare quite different from,
and possibly inconsistent with, those of universities (for example, scientific
knowledge for its own sake), and quite different again from those of military
research establishments (for example, to save the lives of our military personnel).
Again, the hierarchical structures within a military research establishment are
quite different from the more collegial structures prevailing in universities; and
the structure of commercial firms is quite different again. The general point to be
made here is that scientific activity is not only a form of complex joint activity (a
layered structure of joint epistemic action); it is activity that is inevitably shaped
by the non-scientific institutional setting in which it is conductedthat is, by
the specific collective ends, structure and cultures of particular institutions.
14 Miller, 2010, op.cit., Introduction.

198
Here we also need to stress the distinction between the de facto institutional
collective end, structure and/or culture and what it ought to be; cultures,
for example, can vary greatly from one organisational setting to another,
notwithstanding that the type of institution in question is the same or very
similar.
In the light of the above, we can distinguish the normative account of science
as a joint intellectual activityfor example, aimed at knowledge for its own
sakefrom science as a means to broader social ends (for example, vaccines to
save lives). Moreover, we can distinguish both from the normative account of
specific institutions in which science exists principally as a meansfor example,
commercial firms (vaccines to make profit) and a military biodefence organisation
(vaccines to save the lives of our military personnel). Importantly, in the context
of discussion of dual-use concerns, we can distinguish within the normative
account of science (both at the level of joint intellectual activity and at the
level of specific institutions) between its positive ends (for example, knowledge
for its own sake and knowledge as a means to combat disease) and negative
ends (for example, harm prevention). Earlier we discussed the no means to
harm principle (NMHP)namely, the principle that one should not foreseeably
(whether intentionally or unintentionally) provide others (for example,
bioterrorists) with the means to do great harm. Clearly the means in question
is essentially scientific or technological knowledge and this knowledge is the
product of joint epistemic action (indeed, joint epistemic action undertaken in
the context of multiple layered structures of joint epistemic action). Accordingly,
harm prevention in relation to dual-use concerns is, or at least morally ought to
be, a joint enterprise with respect to joint epistemic action; it is something that
biological scientists as members of their scientific community (or communities)
and specific institutions (for example, biology departments in universities,
biotech companies) morally ought to jointly address including (presumably) by
way of education and regulation of their potentially harmful joint epistemic
action. But let us be clear on the moral responsibilities in play here.
Collective moral responsibility

As we have seen, dual-use issues in science are inherently ethical, and scientific
activity is essentially joint epistemic activity undertaken in various institutions
with possibly divergent normative underpinnings. Given the multilevel,
cooperative character of scientific activity and given its ethical or moral
significance, a key theoretical normative notion in play here is that of collective
moral responsibility.
199
Let us distinguish between natural, institutional and moral responsibility and,

in respect of responsibility, between individual and collective responsibility. I
note that the notions of natural, institutional and moral responsibility are not
mutually exclusive. I also note that agents responsible for some outcome might
have both individual and collective responsibility for that outcome.
An agent, A, has natural responsibility for some action, x, if A intentionally
did x for a reason and x was under As control. Bench scientists engaging in
routine scientific researchfor example, replication of experimentshave
natural responsibility for their actions. Moreover, such actions might not have
any obvious moral implications.
Agent A has institutional responsibility for action x if A has an institutional
role that has as one of its tasks x. Thus, for example, laboratory assistant, A,
has the institutional responsibility to clean the test tubes; moreover, A has this
responsibility even if A does not in fact do this.
What of moral responsibility? Roughly speaking, agents have moral responsibility
for natural or institutional actions if those actions have moral significance. So
if A is naturally or institutionally responsible for x (or for some foreseeable
outcome of x, O) and x (or O) is morally significant then, other things being
equal, A is morally responsible for x (or O) and, other things being equal, can
be praised/blamed for x (or O).
Note that other things might not be equal if, for example, A is a psychopath
(and, therefore, incapable of acting in a morally responsible fashion) or if A
does something wrong but has a good excuse (and, therefore, ought not to be
blamed).
Note also that if O involves some intervening agent, B, who directly causes O,
then A may have diminished moral responsibility for O.
Let us now consider collective moral responsibility. In essence, the account of
collective moral responsibility mirrors that of individual moral responsibility,
the key difference being that the actions in question are joint actions, including
joint epistemic actions.
Accordingly, if agents A, B, C and so on are naturally or institutionally responsible
for a joint (including epistemic) activity, x (and/or some foreseeable outcome of
x, O), and x (and/or O) is morally significant then, other things being equal, A,
B, C and so on are collectively (that is, jointly) morally responsible for x (and/
or O) and, other things being equal, can be praised or blamed for x (and/or O).
The other things being equal clauses function here as they did in the above
account of individual moral responsibility. Moreover, as was seen to be the
case with individual moral responsibility, if there are additional intervening
200
(individual or joint) actions then those jointly responsible for the joint action in
question, and its outcome, may have diminished moral responsibility. Scientists
who engage in dual-use research that is subsequently used in the construction
of WMD may well have diminished responsibility for the harm caused by those
WMD. Diminished responsibility is not, however, necessarily equivalent to no
responsibility. Further points to be made here are as follows.
First, each agent may have full or partial moral responsibility for x jointly
with others for the joint action x and/or its outcome. If, for example, five men
each stab a sixth man once, killing him, each is held fully morally (and legally)
responsible for the death even though no single act of stabbing was either
necessary or sufficient for the death. In some cases each agent might have full
moral responsibility (jointly with others) for some outcome, Onotwithstanding
the fact that each only made a very small causal contribution to the outcome
in large part because each is held to have prior full institutional (including legal)
responsibility (jointly with others) for O.
On the other hand, each agent might have partial and minimum moral
responsibility jointly with others if each only makes a very small and incremental
contribution as a member of a very large set of agents performing their actions
over a long periodfor example, the scientists who worked on the human
genome project.
Second, we need to distinguish cases in which agents have collective moral
responsibility for some joint action or its outcome from cases in which agents
only have collective moral responsibility for failing to take adequate preventative
measures against O taking place. Many untoward dual-use cases are of the latter
kind.
On the other hand, agents may not have any collective moral responsibility
with respect to some foreseeable morally significant outcome, O, if O has a low
probability, takes place in the distant future and involves a large number of
intervening agents.
The collective moral responsibilities of scientists are multiple. Scientists have
a collective institutional (professional) and moral responsibility as scientists to
acquire knowledge for its own sake.
Scientists functioning in universities also have a collective institutional and
moral responsibility to acquire knowledge for the good of humanityfor
example, vaccines for poverty-related diseases.
201
Scientists functioning in commercial firms might have a collective institutional

and (contractually based) moral responsibility to acquire (say) knowledge of
vaccines for rich peoples diseasessince that is a commercial imperative of
their employer and they are being paid to do just that.
Scientists functioning in biodefence organisations have a collective institutional
(and moral?) responsibility to acquire knowledge of vaccine-resistant pathogens
if this is a national security imperative of their employerthat is, the government.
As human beings, scientists have a collective moral responsibility not to provide
the means for others to intentionally do great harmfor example, the means to
allow others to drop atomic bombs on Hiroshima and Nagasaki or engage in
biowarfare.
Moreover, these various collective institutional and moral responsibilities may
be inconsistent with one anothernotably, the collective moral responsibilities
scientists have as human beings and the institutional responsibilities that they
might have as members of military research organisations.
Regulatory frameworks
In the light of the dangers stemming from the dual uses of science and technology
there is a need to consider a range of regulatory measures to reduce the risks.
Such measures include the imposition of limits on dual-use experiments and on
the dissemination of potentially dangerous information resulting from dual-use
discoveries. These measures themselves exist on a spectrum ranging from the
least intrusive/restrictive to the most intrusive/restrictive.15
Some specific regulatory measures that might be considered include the
following ones.
1. Mandatory physical safety and security regulation: Should there be
regulations providing for physical safety and security in relation to the
storage of, the transport of and physical access to samples of pathogens,
equipment, laboratories and so on? And should compliance with such
regulations be mandatory? In both cases the answer is presumably in the
affirmative.
2. Licensing of dual-use technologies/techniques: Should there be mandatory
licensing of dual-use technologies/techniques/pathogen samples? Only certain
laboratories in the public sector and the private sector might be licensed to
15 For a detailed treatment of these and other options, see Miller and Selgelid, 2008, op. cit.
202
engage in research involving the use of certain dual-use technologies and

licences for DNA synthesisers might be required.
3. Mandatory education and training: Given the potential harms arising
from, for example, experiments of concern, it is clear that some process of
education and/or training for relevant researchers and other personnel is
called for.
4. Mandatory personnel security regulation: Doubtless it is prudent, indeed it
is a moral requirement, that access to virulent pathogens be disallowed to a
researcher diagnosed as a psychopath or to a known member of a terrorist
organisation.
5. Censorship/constraint of dissemination: The question of whether research
findings ought to be freely disseminated, censored or their dissemination
in some lesser way restricted is an extremely difficult issue and it is by no
means obvious who the ultimate decision-maker ought to be. A relevant
important distinction here is that made above between first-tier and secondtier dual-use research. For example, first-tier research findings might need to
be disseminated in such a way that anyone being informed of these findings
would not be able to replicate the experiments that enabled the results
reported in the findings. Indeed, this is precisely what was recommended
at one point (for example, by the US National Science Advisory Board for
Biosecurity) in relation to the dual-use bird flu research mentioned at the
outset of this chapter.
Let us assume that a range of regulatory options ought to be pursued, at the
institutional (university, commercial firm, government research laboratory),
national and international levels (for example, Biological Weapons Convention
verification processes). There remain some in-principle obstacles to the
establishment of adequate measures to deal with the dual-use problem in science
and technology and a number of these stem from various perverse incentive
structures that derive from collective-action problems. In the final section of
this chapter, I will briefly discuss some of the more salient of these.
Collective-action problems
Thus far we have characterised the scientific enterprise as essentially a joint
epistemic one: the emphasis has been on intellectual cooperation to achieve
common scientific (epistemic) goals in an institutional context of minimal
restrictions on the freedom of individual scientists. This picture, however, while
acceptable as far as it goes, is an oversimplification. Specifically, it obscures the
competitive dimension of scientific activity and, in particular, it masks various
203
collective-action problems arising from such competition. This is important for

our purposes here, not least because it casts the dual-use problem in a somewhat
different light.
On the purist (as we might call it) model of scientific activity as joint epistemic
action performed under conditions of scientific freedom, the dual-use problem
arises only because scientific research undertaken for the benefit of humankind
can be misused by others for harmful purposes. Accordingly, there is a need to
monitor dual-use research and erect safeguards against misuse by malevolent
individuals and groupsfor example, lone wolf malcontents, nihilistic
terrorist groups, rogue states, and so on.
Notice, first, that there is here an implicit additional assumptionnamely,
that scientific activity will be undertaken in the first place in order to benefit
humankind. This is, as we have seen in relation to WMD programs, not
necessarily the case. On the other hand, WMD research is not dual use in our
sense (unless military research undertaken for protective purposes yet having
the potential, as is probable, to be misused for aggressive purposes is to be
regarded as dual use).
Notice, second, that much scientific work, including in the biological sciences,
is not undertaken under conditions of scientific freedom. Consider research
undertaken in the private sector or for various government laboratories in, for
example, authoritarian states such as China. Which research is undertaken, and
whether or not it is published, is not necessarily or even typically a decision
made by individual scientists or, indeed, by groups of scientists. Rather these
are commercial decisions made by managers or they are decisions made by
government officials in the national interest (presumably). Accordingly, it is
simply not true that scientific work, including scientific work in the biological
sciences, is necessarily, or even typically, conducted under conditions of
scientific freedom.
But to return to the main point at issuenamely, competition and, relatedly,
collective-action problems: in the biological sciences, as elsewhere, there is
competition between individual scientists, between scientific institutions and
between nation-states.
It is self-evident that there is competition between, for example, biotechnology
companies in the private sector. Moreover, governments compete insofar as they
have an interest in promoting their own biotech industries and, more generally,
insofar as they want to ensure that they do not fall behind in research and
development (R&D) in the biological sciences (not least for military reasons).
Further, even in the case of scientific work undertaken under conditions of
204
scientific freedom (for example, in universities), there are important elements of

competitionfor example, between rival teams of scientists in competition for
status and (relatedly) for scarce funding.
As suggested above, competition in these various sectors gives rise to a variety
of collective-action problems that have important implications for the dual-use
issue.
First, in the private sector there are collective-action problems arising from
commercial competition. As already noted, many scientists work in commercial
firms in which there is an imperative to maximise profit. In such a context of
fierce commercial competition restrictions on dual-use research may handicap
an organisation. This is a collective-action problem insofar as an organisation
all things considered and given appropriate weightought to choose not to
perform a particular dual-use experiment on the grounds that the potential
harm to humankind resulting from this kind of experiment might outweigh the
potential benefits to humankind. All things might not be considered, however,
or given appropriate weight. Specifically, the firm might give excessive weight
to its commercial interests, especially if it believes that some other competing
firm is likely to be less scrupulous and go ahead with the experiments in
question. In short, in dual-use cases where discretionary judgment is called for,
the judgment might be skewed by considerations of commercial self-interest
in a fiercely competitive commercial environment. I note that commercial selfinterest may well be dominant in such cases, notwithstanding the commitment
of individual scientists to the NMHP. For one thing, it is not necessarily the
decision of the scientists, who are, after all, mere employees; and for another,
their self-interest as employees might align them with the firms commercial
interest, especially given the relative ignorance of scientists of security issues.
Second, in the university sector there are collective-action problems arising
from competition for status. As already noted, many scientists working in
the university sector are engaged in a competition for status (and for scarce
funds to undertake projects by means of which they can achieve status), both
for themselves as individuals and on behalf of the institutions they work for.
Accordingly, there is an analogue of the above-described collective-action
problem in the private sector. In dual-use cases where discretionary judgment
is called for, the judgment might be skewed by considerations of individual or
institutional self-interest in a competitive environment, albeit the competition
in question is primarily for status (and scarce funds to achieve status).
Third, in the government sector there are collective-action problems arising
from competition among nation-states. As noted, in the past and, indeed, in
the present there has been a variety of arms racesfor example, the nuclear
arms racein which scientists played a central role. The problem here is that
205
national self-interest is pitted against humanitys collective interest in a context

in which there is no enforceable international law; evidently nation-states
cannot effectively collectively self-regulatehence the WMD programs of the
United States, Iran, North Korea and so on. The inability or unwillingness to
collectively self-regulate exists, however, at least potentially, in relation to the
dual-use problem and does so independently of any inherent desire on the part
of nation-states to maintain WMD programs. We saw above that the self-interest
of individual scientists and the institutions in which they work (for example,
commercial firms and universities) can under conditions of fierce competition
lead to collective-action problems in relation to dual-use research. Nationstates are, however, themselves in competition with one another, and it is in the
economic, military and so on interests of nation-states to support their own R&D
in science and technologythat is, to support the work of their commercial firms
and universities, and to do so in the face of foreign competition. Accordingly,
we cannot necessarily look to individual governments to regulate adequately
the scientific research in their own institutions, at least if adequately in this
context refers to an all-things-considered, morally informed decision made in
the long-term interests of humankindas opposed to a decision made in the
(possibly short-term) national interest.
A fourth collective-action problem arising from competition is of a somewhat
different kind: it is a species of the generic problem of freeriding. It is the
possibility of the untoward consequences of scientific freeriding, so to speak. Let
us assume that Polanyis scientific freedom modelfor example, no censorship
is in fact the best model to acquire new knowledge; those operating entirely
outside the model cannot compete. Accordingly, so the argument runs, the
good guys (for example, the scientists making vaccines within the framework
of scientific freedom) stay ahead of the bad guys (for example, the scientists
weaponising pathogens outside the framework of scientific freedom); the bad
guys are always playing catch-up. Contrary to this argument, however, it might
be claimed that a well-qualified national cohort of bad guys can always free ride
but then get ahead of good guysfor example, scientists in an authoritarian
state with biodefence projects benefit from the work of those in the scientific
freedom model but dont share their own work.
206
13. Biosecurity and the Just-War

Tradition1
Biosecurity and the just-war tradition occupy separate worlds. In debates and
discussions about biosecurity and dual use, no references are made to justwar criteria. And in textbooks on just-war tradition, biosecurity and dual use
hardly get any attention. This chapter will deal with the question of whether
this separation is justified. First, an oversight will be provided of the just-war
tradition and more especially of recent developments within it. Attention will be
given to which questions this tradition deals with, why and how. The answers it
offers will then be applied to biosecurity and dual-use issues.
Just-war tradition: Between past and present2

Thinking about what counts as a just war can truly be called a tradition. It
has its roots in authors such as Aristotle, Cicero, Augustine, Thomas Aquinas,
Grotius, as well as Spanish theologians such as Vitoria and Suarez, and the Swiss
diplomat de Vattel. They and many other philosophers, theologians and lawyers
have contributed to debate about how to approach the notion of just in the
context of conflict.3 In recent years, debate has continued through authors such
as Paul Ramsey, James T. Johnson, Michael Walzer, Jeff McMahan andagain
many others.4
Just-war tradition as a name for this normative approach to conflict is preferable
to just-war theory. The use of the word theory implies a framework was developed
at a certain time and has remained more or less static. In fact, just-war thinking
has evolved and has been influenced by political, social, technological and
military developments. The political situation of the Roman Empire cannot be
compared with that of the Italian city-states or the Spanish colonial expeditions.
1 The views expressed in this chapter are those of the author and do not represent those of the Royal
Netherlands Academy of Arts and Sciences.
2 See also: van der Bruggen, K. 2009, Other wars, other norms? in G. Molier and E. Nieuwenhuys (eds),
Peace, Security and Development in an Era of Globalization. The Integrated Security Approach Viewed from a
Multidisciplinary Perspective, Martinus Nijhof, Dordrecht/Leiden/Boston, pp. 35576.
3 An anthology of relevant texts is published by: Reichberg, G. M., Syse, H. and Begby, E. (eds) 2006, The
Ethics of War. Classic and Contemporary Readings, Blackwell, Malden, Mass.
4 Classic studies are: Walzer, M. 1977, Just and Unjust Wars. A Moral Argument with Historical Illustrations,
Basic Books, New York; Johnson, J. T. 1981, Just War Tradition and the Restraint of War, Princeton University
Press, Princeton, NJ; Johnson, J. T. 1984, Can Modern War Be Just? Yale University Press, New Haven, Conn.
207
Also technical developments have impacted thinking. A well-known example

involves a high-tech weapon of the Middle Ages: the crossbow. During the
Second Lateran Council, its use was prohibited, at least against Christians.5 This
ban can be seen as an early application of the principle of proportionality, at least
for Christians. The example shows that debates about the moral acceptibility of
the use of weapons are longstanding. This certainly holds true for biological
and chemical weapons. May and Crookston write that for 2,500 years the use
of poisons during battles has been forbidden.6 The moral and legal rejections
of poisons sometimes were even more rigorous than those of other means. From
Roman authors such as the historian Valerius Maximus, we know the saying
armis bella non venenis geri (wars are fought with weapons, not with poison).7
In other words: poison was not seen as a legitimate weapon. This prohibition
did not mean that no poisons have been used in times of war though. History
provides many examples of poisoning:
Evidence can be found for the existence of forms of chemical and
biological warfare in ancient and classical times. The evidence for
chemical and toxin warfare is the clearest. Solon of Athens is said
to have used hellebore roots (a purgative) to poison the water in an
aqueduct leading from the Pleistrus River around 590 B.C. during the
siege of Cirrha. Writings of the Mohist sect in China dating from the
Fourth Century B.C. tell of the use of ox-hide bellows to pump smoke
from furnaces in which balls of mustard and other toxic vegetable
matter were being burnt into tunnels being dug by a besieging army
to discourage the diggers. The use of a toxic cacodyl (arsenic trioxide)
smoke is also mentioned in early Chinese manuscripts. Sparta used the
toxic smoke generated by burning wood dipped in a mixture of tar and
sulfur during one of its periodic wars with Athens.8
This overview can be extended with examples from the Middle Ages until the
IranIraq war in the 1980s.9 In spite of the use of poisons and other biological
agents during wars, the saying of Valerius Maximus has always kept a kind of
validity. Biological (and chemical) weapons never reached the status of normal
weapons, comparable with swords, spears, crossbows and later firearms. Using
biological weapons always had the smell of doing something foul. An example
is the praise of Cicero for a Roman general who refused the offer of a deserter to
5<http://www.nuclearmuseum.org/online-museum/article/waging-peace-the-challenge-of-nuclearstewardship/> (viewed 2 January 2011).
6 May, L. and Crookston, E. 2008, Introduction, in L. May (ed.), War. Essays in Political Philosophy,
Cambridge University Press, Cambridge, p. 2.
7 Valerius Maximus, Factorum Et Dictorum Memorabilium Libri Nouem, <http://www.thelatinlibrary.com/
valmax6.html> (viewed 11 January 2011).
8 <http://www.cbwinfo.com/History/History.html#0001> (viewed 2 January 2011).
9Ibid.
208
13. Biosecurity and the Just-War Tradition
poison King Pyrrhus.10 And Samuel von Pufendorff (163294) wrote that the
more civilized nations condemn certain ways of inflicting harm on an enemy:
for instance the use of poison.11
Back to just-war thinking in general. New times and new situations ask for
revisiting what moral and legal norms should be weighed and how in seeking
to answer the question if (and under what conditions) the use of violence is
justified. Sometimes developments lead to more than marginal adjustments
of assessments. So after World War II, the issue was posed of whether the
invention of nuclear weapons led to novel considerations in relation not only
to the possible use of these weapons, but also to deterrence. Is it, for instance,
morally justified to threaten their use if this would go against all prevailing
criteria regarding what is appropriate in terms of violence? What if deterrence
is the only way to prevent the actual use of nuclear weapons? Such complex
questions raised by this case could lead to the conclusion that the just-war
tradition has lost its relevance and significance.12 But such a view is too rash
given it is based on only onehowever important and shockingdevelopment
in the waging of war. In the heydays of the Cold War, (too) many so-called
conventional wars were fought in which just-war criteria could and should
have provided good guidance. More than this: just-war thinking could provide
criteria for moral argument on the paradoxes of nuclear deterrence, as is shown
in the development of a theory of justified deterrence.13
Just-war criteria
In the just-war tradition, a distinction is made between the so-called jus ad
bellum and the jus in bello. The jus ad bellum indicates if and when it can be
justified to start a war. The criteria one to five below refer to this jus ad bellum.
The sixth one belongs to both domains and the remaining two criteria deal with
the question of the jus in bello: what behaviour is permitted or forbidden during
a war. More recently a third domain has been added: the jus post bellum. There
is much debate about whether this new distinction is useful and necessary for
judging modern wars. Elsewhere I have questioned the utility for a jus post
10 Cicero, On duties, in Reichberg et al., op. cit., p. 53.

11 von Pufendorf, S. 2006, On the duty of man and citizen, in Reichberg et al., op. cit., p. 459.
12 A common way of thinking during the Cold War was that a nuclear war never could be a just war
and thatby consequencejust-war tradition had lost its significance. Often this view was promoted by
pacifists, who had already disputed just-war thinking by definition. But another group was the so-called
nuclear pacifists, who limited their view to the conclusion that a just nuclear war was a contradiction in terms.
13 van der Bruggen, K. 1986, Verzekerde Vrede of Verzekerde Vernietiging. Ontwikkeling van een Theorie
vanGerechtvaardigde Afschrikking [Assured Peace or Assured Destruction. Development of a Theory of Justified
Deterrence], Kok, Kampen.
209
bellum.14 Setting aside this wider question though, the jus post bellum will not
be considered in this chapter, because it is not very relevant for biosecurity
issues.
Jus ad bellum
1. The war must be waged by a legitimate government
This criterion expresses the assumption that only a legitimate government has
and should have the monopoly on violence. In a sovereign state the government
and the government alone is allowed to use violence. This may seem obvious,
but in practice things often are not so obvious, because the legitimacy of a
government is not always undisputed. Much violence (for example, in civil
wars or anticolonial wars) stems from conflicts over who owns the legitimate
authority. Often it works out that legitimacy can be decided only post hoc: the
winner takes all. In addition, the notion of legitimate authority had varying
meanings over time. Thomas Aquinas had a much different concept than Hugo
Grotius, and also in our days definitions are changing. But the fact remains that
the notion of legitimate authority has a certain core holding. By convention
actually since the Peace of Westphalia in 1648sovereign states hold the
monopoly on violence.15 States and government embody the public domain and
because of that they differ from all other (private) groups and entities. Given
all disputes and discussions about sovereignty, this first just-war criterion
certainly is not ideal, but something better is not available. If the public domain
collapses, the war of all against all remains a possibility, even today. And in
such a situation sovereign authority will be sought. Recent examples can be
found in so-called failed states, where in practice the central government has
virtually disappeared. Many people in that situation eventually accept the
authority of terrorist movements, becausein the view of powerless groups
their violent and arbitrary exercise of power is always better than the absence
of any authority. To give an example: it is no wonder that in Afghanistan many
people accept (which is not the same as support) the authority of the Taliban
instead of what they see as a condition of anarchy.
2. The war must be waged for a just cause
Fighting war for a just cause is the core of the jus ad bellum. Influenced by
the Charter of the United Nations (1945), the interpretation of what counts
under this category has become more and more restricted in recent times. Only
a reaction against aggression or an intervention that is sanctioned by the UN
14 van der Bruggen, 2009, op. cit.
15 Schrijver, N. 1998, Begrensde soevereiniteit350 jaar na de Vrede van Munster [Limited
sovereignty350 years after the Peace of Mnster], Transaktie, vol. 27, no. 2, pp. 14174.
210
Security Council is taken as a legally acceptable base for waging war. This means
thatto give some examplesretribution or recoveries of previously suffered
injustices are not acceptable justifications.
In a sense the interpretation has also been extended. Since the end of the Cold
War the idea has become widespread that the international community has
the rightif not the dutyto use military force to intervene in cases of gross
violations of human rights. The intervention in Kosovo in 1999 serves as one
example, even though a UN Security Council authorisation was missing. A
further extension came after 11 September 2001, when the war against terror
was added to the list of just wars by many people. The invasion of Afghanistan
to expel the Taliban regime was not explicitly mandated by the UN Security
Council, but the American action was widely seen as a legitimate form of selfdefence supported by resolutions that were adopted after 9/11.16 In relation to
biosecurity the question arises if and under which conditions the development
or possession of biological weapons can be a reason to start a war. This is not
a purely academic question as is shown below in the heated debates about the
AmericanBritish intervention in Iraq.
3. The war must be waged with a right intention
This criterion is in line with the previous one, but it goes further. It says that it
is not justified to have secondary intentions when fighting a war for a just cause.
The only intention should be that just cause (as resisting the aggressor). If the
defender succeeds in that objective, he must stop fighting. He is not allowed
to aim at military or economic destruction of the opponent. According to the
classical tradition, feelings of revenge also should play no role. Most recent wars
have shown that standard proves unrealistic. In fact, in most wars, intentions
other than the official ones can be discerned. Economic (oil), political or
personal secondary intentions are never far away. If and how these secondary
intentions played a part in the Iraq warformally intended to remove biological
and chemical weaponswill be illustrated below.
4. All other means of conflict resolution must be exhausted; war is
the last resort
This criterion is designed to prevent governments from taking up arms too
quickly. The question that arises is: when is too quickly and who is to decide?
Parties that are contemplating force would likely argue that it is the only
solution. The relevance of this criterion is that it is forcing the aggressor to
make the case for why aggression is necessary. Is a military intervention the
16 For example, United Nations Security Council (UNSC) Resolution 1373 (2001): <http://daccess-dds-ny.
un.org/doc/UNDOC/GEN/N01/557/43/PDF/N0155743.pdf?OpenElement> (viewed 11 January 2011).
211
last remaining option? Are all other means really exhausted? These kinds of
questions have to be asked and answered. The problem is that the answers to
these and similar questions can never be given with mathematical certainty.
Fortunately, history also shows many examples where war could be prevented.
Humankind has repeatedly escaped the battlefield.17 Of course we owe this to
several causes, but certainly the message of the just-war traditionand more
specifically of this criterionplayed a role in the idea that starting a war is a
decision that has to be avoided if possible. This can be illustrated by the way
political leaders try to persuade others that their war is justified and that they
do not have any other possibility. One need not believe what is said to still note
the importance of making a justification. Peace is the rule; war (unfortunately
all too often) the exception.
5. There must be a reasonable chance that the intended purposes
are reached through the war
The purpose of this criterion is to prevent a government from starting a war
if it is clear beforehand that the intended targets cannot be reached. At least
one objection to this criterion rises immediately: it seems to support the side
of the strongest party. Yet military superiority is not always a guarantee for
victory. The most famous example in recent history is the Vietnam War. Despite
overwhelming military superiority, the Americans did not manage to reach their
goals. Especially in a humanitarian intervention, the militarily stronger party
does not always have the greatest interest in going on with fighting. The tragedy
of Srebrenica seems an example. If the UN Protection Force (UNPROFOR, the
UN troops in Bosnia) had used all available resources, the fall of Srebrenica could
have been prevented. Among other factors, the fact that the self-interest of the
Netherlands and the Dutch military was not at stake certainly played a role. The
professional soldiers of Dutchbat had signed up for the army and realised that
an ultimate consequence of this was to die, [b]ut dying for Srebrenica. It was
not worth it.18
6. The objectives and means of war must be in a reasonable
relationship to each other. This is the principle of proportionality
The principle of proportionality can relate to both the jus ad bellum and the
jus in bello. In the first case, the principle refers to waging the war: is there a
proportionate relationship between its (expected) cost and the aim to be reached?
During the war the principle can be applied to judge if a concrete action or the
17 Gerrits, A. and de Wilde, J. 2000, Aan het slagveld ontsnapt. Over oorlogen die niet plaatsvonden, Waburg
Pers, Zutphen. This book is about wars that did not take place.
18 Westerman, F. and Rijs, B. 1997, Srebrenica. Het zwartste scenario [Srebrenica. The Blackest Scenario],
Atlas, Amsterdam/Antwerpen, p. 11. Quote from a Dutchbat soldier.
212
use of a certain kind of weapon is proportionate in relation to the goals to be

reached. In both cases the discussion will concentrate on the issue of weighing
and judging the factors that determine the aims and the means. Who is to decide
on this and on the basis for weighing factors? This problem becomes even more
complicated because it is almost impossible to know beforehand what the
costs of a war or even a concrete action during a war will be. Who would have
predicted that hundreds of thousands of French, British and German soldiers
would be killed in the trenches of World War I? When these soldiers marched
into the battlefields in August 1914, they expected to be home at Christmas in
the same year, after having won a frischen und frhlichen Kriege. Proportionality
can become a salami-like criterion: the lives already lost become an argument for
going on, because not doing so would mean that these victims lost their lives in
vain.19 So the limits of what still is seen as proportional shift.
7. The use of force should be limited to the minimum that is
necessary in order to resist the aggressor
This criterion has a certain affinity with the just mentioned proportionality
principle. But it goes further by suggesting even less than proportionally
acceptable violence has to be used, if that is enough for achieving the aims of
war. Again the same questions return as in the preceding two criteria: what
is minimal violence and who is to decide? Once started wars have their own
momentum that often leads to more escalation. Von Clausewitz, the famous
Prussian theorist of war, assumed that wars were by definition absolute and that
they should be.20 Only politics could prevent that absolute character.
Indeed, the influence of politics in the recent wars of intervention has been
great, but that has not always led to restraint. Regarding the Kosovo war there
has been much criticism of the decision to carry out aerial bombardment for
strategic reasons on some civilian targets in Yugoslavia. First there was the
question of whether it was necessary from a military perspective. The attack on
the TV studio in Belgrade is one of the most infamous examples. This brings us
to the final criterion of the just-war tradition to discuss here: the principle of
noncombatants.
19 This was the reason that some parents of Dutch soldiers who had been killed in Afghanistan were among
the people who resisted the return of Dutch troops. The same argument can be read in the memoir of George
W. Bush, where he describes his meetings with parents of killed soldiers: George W. Bush 2010, Decision
Points, Crown, New York, ch. 12, pp. 35594.
20 von Clausewitz, C. 1982, Over de oorlog, Het Wereldvenster, Bussum, p. 15.
213
8. A distinction should be made between military and civilians;

the latter group may not be involved in the fight. This is the
noncombatants principle
The noncombatants principle has always played an important role in the just-war
tradition and it still does. This principle has been translated in international
treaties like the Geneva conventions. What is important is that not only citizens
belong to the category of noncombatants, but also soldiers who are not actively
involved in the fight (for instance, those wounded, and prisoners). Developments
in modern technology have, however, led to wars that are conducted in ways
different than in the time when the noncombatants principle was developed.
Massive battles seem at least in the Western world to belong to the pastand
whoever remembers the images of the trenches of World War I can of course
only be happy with it. But it begs the question of the actual meaning of the
noncombatants principle.
Todays wars by major military powers are often wars at a distance. There is
hardly direct physical contact with the combatants of the counterparty and
most governments are reluctant to create situations in which such a direct
confrontation is provided. The discussion about whether or not to proceed with
a ground war during the intervention in Kosovo (1999) is such an example.
At the other end of the spectrumthat of the noncombatantsthere have
been developments also. To give only one instance: the distinction between
combatants and noncombatants coincided for centuries with the difference
between civilians and soldiers. That is not true anymore. Most terrorists do not
belong to the military, but they surely are to be categorised as combatants.
Despite this noncombatant criterion, the general trend is that attacks in which
civilians are deliberately targeted have become the normal picture of modern
wars. Names of cities such as Guernica, Dresden and Hiroshima suffice as an
illustration. And even when citizens are not deliberately chosen targets,
questions can be askedfor instance, regarding the concept of collateral
damage. A question in this context is whether this so-called collateral damage
is too easily accepted. How unintended is unintended if you know that the
chosen method of attack will lead almost by definition to civilian casualties or to
victims because of destroyed infrastructure? Walzer stated in his famous Just and
Unjust Wars that collateral damage in such a case cannot be morally justified.21
More recently, however, he was less convinced of this judgment. In a lecture in
Amsterdam in 2007, Walzer states that the distribution of responsibility is the
ultimate moral factor in judging if the killing of civilians can be justified. As an
example he gives his view on the Lebanon war (2006):
21 Walzer, 1977, op. cit., pp. 1514.

214
In the Lebanon war, the Israeli army caused most of the civilian deaths,
but some (or many) of the villages it attacked were being used by
Hezbollah as bases for rocket attacks on Israeli cities, so the greater
part of the responsibility for civilian deaths in those villages lay with
Hezbollahas did the greater part of the responsibility for the war
itself, which began with a rocket barrage and a Hezbollah raid across
the international frontier. Israel is responsible for deaths caused by
unjustifiable bombings like the Qana raid or by the cluster bombs used
late in the war. But (again) it shouldnt be the proportionality argument
that guides our judgment of those deaths; they were wrong whether or
not they were disproportionate. In Vietnam, Kosovo, and Lebanon, it is
the balance of responsibility that is morally determinative.22
Just-war tradition and biological weapons:

The Iraq war of 2003
How can a link be made between just-war principles and the problems of
biological weapons, ormore generallybiosecurity? At first sight the jus ad
bellum is not directly linked to questions of biosecurity. Seen in the widest sense
of the term though, some recent international conflicts, and more especially the
Iraq war (2003), do make such a link. Albeit this is a link that evokes many
questions!
Officially, one of the main arguments the United States and the United Kingdom
used to justify their attack on Iraq was the consideration that Saddam Hussein
had weapons of mass destruction (WMD), and as part of this arsenal, biological
weapons. In his oral evidence before the Chilcot Inquiry Committee in 2010,
former UK prime minister Tony Blair confirmed that he still maintains the
possession of WMD by Iraq as a justification for the war against Saddam.23
Is the development or possession of biological weapons by a country a just
cause for starting a war? Here it will be argued that this is not the case. The
development of WMDobjectionable as it may be on legal or moral grounds
is not equal to aggression as defined in the UN Charter. This means that the only
justification for an intervention in Iraq had to be found in a resolution under
Chapter 7 of the UN Charter. This is what the United States and the United
Kingdom tried to realise in the months before they invaded Iraq. When such a
resolution appeared to be unattainable, however, they decided to start the war
22<http://www.ru.nl/soeterbeeckprogramma/terugblik/terugblik-2007/teksten-2007/thomas_more_
lezing_0/> (viewed 12 November 2010).
23 <http://www.iraqinquiry.org.uk/media/45139/20100129-blair-final.pdf> (viewed 4 October 2010).
215
without a (new) resolution. Attempts were made to justify this by an appeal

to the so-called revival argument. This contended that the whole range of
UN resolutions against Saddam since 1991 could be legitimating for the war.
This argument was defended by Peter Goldsmith, the then attorney-general for
England and Wales.24 Because the Dutch Government took his arguments on,
the Dutch Committee of Inquiry on the War in Iraq analysed this argument in
detail. They concluded that it was not a valid way of reasoning on the basis of
public international law.25
Of course this argument is based on legal judgments. A case on purely moral
grounds could lead to a different judgment. This is acknowledged by the Dutch
Committee of Inquiry:
[S]ome defend the position that a basis in international law alone cannot
be the deciding factor for the justification of international action by
states. The observance of international law rules is very important, so
runs the argument, but cannot always be decisive. Sometimes, in an
international conflict, values of such importance are at stake that states
can feel compelled to act even when this may not be according to the
prevailing international law. However, these are exceptional situations
in which very compelling moral imperatives apply. The legal maxim,
need before law, and the international law concept state of necessity
also offer safety valves within the law for finding a way out of this
situation. It is striking that the debate on the possible justification for
the Iraq war never went down this path.26
In other words: none of the members of the coalition of the willing used
this need for law argument. Such an argument usually is used as a reason for
human intervention in case of genocide. It has never been used in the case of the
(suspected or expected) development of weapons of mass destruction.27
In the debates about the Iraq war the question has often been asked whether
considerations or intentions other than the official ones played a part in the
decision for intervention. And, if so, can these intentions be seen as proper?
Many commentators have pointed to other motivations: the importance of oil
interests, thehardly hiddenwish for regime change and last but perhaps
24 Iraq resolution 1441 adviceoriginal memo, BBC News, 7 March 2003, <http://news.bbc.co.uk/1/
shared/bsp/hi/pdfs/28_04_05_attorney_general.pdf> (viewed 4 October 2010).
25 Dutch Committee of Inquiry on the War in Iraq 2010, Report of Dutch Committee of Inquiry on the
War in Iraq, Chapter 8: The basis in international law for the military intervention in Iraq, Netherlands
International Law Review, vol. LVII, no. 1, pp. 1268.
26 Ibid., pp. 8790.
27 Although not all (applications of) biological weapons lead to mass destruction, I will follow the usual
terminology for chemical, biological, radiological and nuclear (CBRN) weapons as weapons of mass destruction.
216
not least the desire of President George W. Bush to finish the job of his father.
Moreover, it is well known that Tony Blair had an almost religious zeal to restore
human rights in Iraq.
Were all other means of conflict resolution exhausted and was war indeed the
last resort?
This third criterion of the just-war tradition played an important role in the
discussions about the Iraq war. Opponents did not stop to mention the efforts
of the International Atomic Energy Agency (IAEA) and the UN Monitoring,
Verification and Inspection Commission (UNMOVIC) in the search for possible
WMD. Both organisations had confidence in the success of their efforts. Because
of that they were opposed to the strategy of Bush and Blair, which was directed
at starting a conflictalmost independently of the results of the IAEA and
UNMOVIC. The United States and the United Kingdom had, however, developed
their plan of attack and they were not willing to make that plan dependent on
the possible results of the inspections. The planning of the invading countries
was related more to practical circumstances in and around Iraq.28
Looking at the Iraq war, it seems that biological weapons hardly played any role
in the final decision. But in the declaratory policy to justify the war as well as
in the long lead-up to the war, the issues certainly were important. It helped to
make war acceptable: if a country possesses and even uses chemical or biological
weapons, it must be a very despicable regime and thus war is an acceptable way
to get rid of that regime. In other words: the (supposed) possession of WMD as
such is seen a casus belli.
The next criterion (number five) that should be taken into account is that there
must be a reasonable chance that the intended purposes are reached by waging
war. If the purpose of a war is eliminating possible stocks of WMD or dual-use
capacities, the feasibility of this purpose is not only dependent on the outcome
of the war. Of course it is helpful and perhaps even necessary to have military
superiority in those areas where the WMD or other materials are stockpiled,
but in addition an intensive scientific survey has to be set up. Anyone who
takes a look at the complexity of the activities of the UN Special Commission
(UNSCOM), UNMOVIC, the IAEA andafter the warthe Iraq Survey Group
gets an impression of the work that has to be done to find, identify and eventually
eliminate WMD materials. The inspectors Hans Blix (UNMOVIC) and Mohamed
El Baradei (IAEA) had repeatedly declared that they would be able to finish their
complex job without military intervention. This makes the argument that war is
necessary to reach the goals of eliminating biological (and nuclear or chemical)
weapons at least implausible. Assuming the hidden goals or intentions of the
28 Woodward, R. 2004, Plan of Attack, Simon & Schuster, New York.
217
Iraq war, it can be argued that the goal of regime change indeed was reached.
Within two months the regime of Saddam Hussein collapsed. But, as we know
now, this was not the end of the fighting.
What can be said of proportionality (criterion six) as an argument for whether
or not to start a war? Proportionality as a criterion of the jus ad bellum
becomes relevant ifaccording to the other ad bellum criteriastarting a
war can be justified. Proportionality in that case is an added criterion to judge
if the relation between purposes and means is balanced. There is no need to
consider the proportionality criterion if it has already been determined that the
other ad bellum arguments define a war as unjust. Given the fact that despite
all counterarguments the Iraq war took place, the jus in bello context of the
proportionality criterion is applicable. But how to apply it to the specific aspect
of biosecurityor broader biological weaponsis not very obvious. The Iraq
waronce startedled to many actions that were and could not be foreseen.
This war is no exception to the rule that a war creates its own judgments on
proportionality: the longer a war lasts, the less some actions are appreciated as
disproportionate. The formal link with the search for biological weapons drifted
out of sight and out of mind.
Finally, both parties have violated the noncombatant principle many, many
times during the Iraq war. Many civilians were killed by Iraqi militants as well
as by coalition troops.29 But it is not possible to link these violations and the
biosecurity issue. The only thing that can be said is that this warat least
partiallywas based on handling the problem of weapons of mass destruction;
however, this was a very exceptional situation. This evokes the question if
and how just-war criteria are relevant for everyday biosecurity policy. This
question will be addressed by looking for elements of this policy to which justwar criteria could or should be applied.
Biosecurity, dual use and just-war tradition

The Iraq war was an exceptional event in the way in which the (alleged)
development and possession of WMD led to an armed conflict. In this section,
attention will be paid to some more mundane aspects of biosecurity, where there
is no imminent threat of war. Is an appeal to criteria of the just-war tradition in
these circumstances helpful?
29 The number of civilian deaths is registered at <www.iraqbodycount.org> (viewed 12 January 2011).

See also: Hsiao-Rei Hicks, M., Dardagan, H., Guerrero Serdn, G., Bagnall, P. M., Sloboda, J. A. and Spagat,
M. 2009, The weapons that kill civiliansdeaths of children and noncombatants in Iraq, 20032008, New
England Journal of Medicine, vol. 360, pp. 15858.
218
Suspicion of development of biological weapons

The cases of Iraq and alsomore recentlyIran, North Korea and Syria are
about states that are involved in or at least suspected of developing nuclear,
biological or chemical weapons. What if the main actors threatening to develop
WMD are non-state actors? Is it allowable to attack a state if inhabitants of this
state are suspected of acting in such a way? Is there a right for third parties to
start a war and, if so, under what conditions? At first sight, this right does not
exist. Two scenarios can be discerned. First, it is possible that the government
of a country itself is developing, producing, stockpiling or otherwise acquiring
biological weapons. If this country is a state party of the BWC, it should act in
accordance with Article 1 of the convention.
The second possibility is that the development, production, stockpiling or
acquisition of biological weapons is taking place within the territory of a state
or under its jurisdiction or control. In that case Article 4 of the BWC is violated.
It seems evident that the first party that is to act in such a case is the involved
national state. Each government has the duty to prevent the misuse of biological
agents. The state should do all that is possible to put an end to this situation.
This has been arranged in the BWC and in many more treaties and agreements,
such as UN Security Council Resolution 1540.
The BWC also indicates what has to be done if there is a breach of an obligation
of the convention. This is described in Articles 6 and 7 of the convention:
Article VI
(1) Any State Party to this convention which finds that any other State
Party is acting in breach of obligations deriving from the provisions of
the Convention may lodge a complaint with the Security Council of the
United Nations. Such a complaint should include all possible evidence
confirming its validity, as well as a request for its consideration by the
Security Council.
(2) Each State Party to this Convention undertakes to cooperate in
carrying out any investigation which the Security Council may initiate,
in accordance with the provisions of the Charter of the United Nations,
on the basis of the complaint received by the Council. The Security
Council shall inform the States Parties to the Convention of the results
of the investigation.
Article VII
Each State Party to this Convention undertakes to provide or support
assistance, in accordance with the United Nations Charter, to any Party
219
to the Convention which so requests, if the Security Council decides

that such Party has been exposed to danger as a result of violation of
the Convention.
Articles VI and VII determine that it is the UN Security Council which has to
decide if any measures will be taken. This means that any attack by other states
to end such a violation is not justified, unless there are other reasons that justify
such an attack (such as an imminent threat or a resolution of the UN Security
Council that justifies the use of any other means).
Of course there are other measures that can be taken to prevent a violation of
Articles I and IV. Most important is to look for peaceful solutions: cooperation,
helping to counter terrorist threats, education and training, and so on. These
measures are especially fitting and relevant in cases where states are not the
ones violating the BWC, but groups or organisations within a country. The
importance of these measures has been stressed by Resolution 1540 of the
Security Council on the nonproliferation of weapons of mass destruction. In
this resolution attention is paid explicitly to the possibility that states need help
in realising the goals of this resolution. The council
[r]ecognizes that some States may require assistance in implementing the
provisions of this resolution within their territories and invites States in
a position to do so to offer assistance as appropriate in response to specific
requests to the States lacking the legal and regulatory infrastructure,
implementation experience and/or resources for fulfilling the above
provisions.
This resolution has been adopted under Chapter VII of the UN Charter, so it has
an obligatory character, but it does not entail any direct or indirect legitimation
for using violence.
Biodefence as bio-offence
According to Article I, BWC state parties are allowed to undertake activities for
prophylactic, protective or other peaceful purposes. This implies that biodefence
is allowed as far as this is limited to protective or peaceful purposes. But who is
to decide what purposes are protective and peaceful? In practice biodefence can
coincide with bio-offence. And even this could be defended with an argument
that is derived from the debate on nuclear deterrence. The argument could
be that having the ability to produce biological weapons will deter another
state or non-state actor from using their biological weapons because of fear of
retaliation. This of course was the logic of the nuclear-deterrence policy during
the Cold War. In those days many debates were devoted to the question of
whether nuclear deterrence was justified from a moral point of view: was it
220
allowable to threaten using a weapon that clearly should lead to a violation of

the proportionality and the noncombatant principles? Nothing like consensus
was reached in this debate between people who were of the opinion that it could
never be right to threaten with these WMD and others, who defended their
view that deterrence was the only way to prevent the use of nuclear weapons.30
If this latter view could be defended from a moral point of viewand there
are some convincing arguments for itthen that was only possible in very
specific circumstances. This includes the bilateral relationship between the
United States and the Soviet Union during the Cold War in combination with
the strategic importance of nuclear weapons. Biological weapons never had the
same strategic importance in practice. During the Cold War there was never a
situation that gave reason for a deterrence strategy with biological (or chemical)
weapons. In fact, the United States and the Soviet Union even agreed on the
BWC during the heyday of the Cold War, although the Soviet Union for at least
some years still went ahead with expanding its programin part because they
thought the United States was doing the same.
If biological deterrence was not defendable during the Cold War, the same
holds today. There can be no strategic or political argument that overrules the
moral inhibitions of the proportionality principle and noncombatants principle.
Moreover, there is reason to believe that such a policy of biological deterrence
would undermine the BWC. A possible reasoning that biological deterrence is
not directed at other states (let alone state parties of the BWC), but at non-state
actors, is untenable. Terrorist groups are almost certainly not deterred by fear
of retaliation. Besides, it is almost impossible to react with a targeted action.
Terrorists are often not directly linked with a specific area or state. Of course
it is conceivable, as happened after the 9/11 attacks, to attack the (presumed)
host state of the terrorists; however, the case of the military actions against
Afghanistan and al-Qaeda could be spoken of as proportionate and targeted
actions, but using biological weapons is almost by definition disproportionate
and untargeted.
In summary: biodefence may be allowed according to the BWC, but the margins
of this research are limited. Caution is required, especially since the risk of dual
use of biodefence is not at all imaginary. The anthrax letters of 2001 (which
allegedly came from a biodefence laboratory) are an already classic example.
And of course there is the risk of accidents. The Sverdlovsk accident (1979)
30 In 1996 the International Court of Justice gave this judgment on nuclear weapons: the threat or use of
nuclear weapons would generally be contrary to the rules of international law applicable in armed conflict,
and in particular the principles and rules of humanitarian law. However, in view of the current state of
international law, and of the elements of fact at its disposal, the Court cannot conclude definitively whether
the threat or use of nuclear weapons would be lawful or unlawful in an extreme circumstance of self-defence,
in which the very survival of a State would be at stake. International Court of Justice 1996, Legality of the
Threat or Use of Nuclear Weapons, Advisory Opinion of 8 July 1996.
221
suffices to draw attention to these kinds of risk. Although such accidents do not
take place during a war, the collateral damage that almost by definition will be
caused by biological weapons can be seen as a violation of the noncombatants
principle of the just-war tradition.
Conclusion
This chapter started with the observation that biosecurity and the just-war
tradition occupy separate worlds. This observation has been confirmed. There
are not many overlaps between both. But although the overlaps are few, some
clear lines can be drawn between the two.
Biological weapons are in the category of weapons of mass destruction, but
biological weapons have much less military value today in deterrence and in
practice than nuclear weapons. The most important example of a war that was
waged for reasons that were linked to biological weapons was the Iraq war of
2003. But this link existed more on paper and in the declaratory policy than in
reality. As far as the argument was used, it cannot be accepted as a just cause
for the war.
This does not mean that developing and storing biological weapons are justified
from a just-war perspective. Most of these weapons (certainly the ones with
contagious agents) are by definition indiscriminate, and using them would be
a violation of the noncombatants principle. The same argument also applies to
a possible bioterrorist use of biological agents. And for the category of nonindiscriminate biological weapons, the argument against their use can be found
in the inhumane character of these means. From a more military point of view,
the argument of Valerius Maximus (armis bella non venenis geri: wars are fought
with weapons, not with poison) still can be seen as valid.
The consideration that possession of biological weapons could be legitimated for
reasons of deterrence is refuted by the current political and military situation.
Because of the BWC, which became possible because of the limited military
value of biological weapons, there is no credible reasoning that these weapons
have a deterring function.
222
14. The Precautionary Principle and

the Dual-Use Dilemma
Steve Clarke
Three precautionary principles

The precautionary principle (PP) is a conceptual tool used to guide decisionmaking in the management of risk.1 It has been widely taken up in environmental
law, and is now being applied in a variety of contexts, including the regulation
of potentially dangerous technologies. It was first developed in Sweden and the
former West Germany in the late 1960s,2 was explicitly used in West German
environmental law by the 1980s3 and has become increasingly influential
in many countries since then, particularly in Europe.4 That the PP is usually
referred to as the PP might seem to suggest that there is a canonical formulation
of this principle. But this is not the case. There are at least 20 different versions
of the PP5 and new ones appear on a regular basis. The fact that there are
many different versions of an abstract principle is perhaps not surprising in
and of itself. What is somewhat surprising is that these different principles do
not appear to be variants of a more general principle. Instead, they are only
loosely associated with one another. What they have in common is a shared
history and the fact that they all advance precaution in some way. As we will
see, they involve at least three very different approaches to the advancement of
precaution, which can be understood in terms of their differing relationships
with costbenefit analysis (CBA) (also known as benefitcost analysis).
Development of the PP was initially motivated by dissatisfaction with CBA,
which was the dominant conceptual tool used in risk management until the rise
of the PP, and which continues to be very widely applied. The application of
1 The PP is also sometimes understood, more broadly, as a set of guidelines for structuring the deliberative
processes involved in risk management. See, for example, Rappert, B. and Moyes, R. 2010, Enhancing the
protection of civilians from armed conflict: precautionary lessons, Medicine, Conflict and Survival, vol. 26,
no. 1, pp. 2447.
2 See Sunstein, C. 2005, Laws of Fear: Beyond the Precautionary Principle, Cambridge, Cambridge University
Press, p. 16.
3 Majone, G. 2002, What price safety? The precautionary principle and its policy implications, Journal of
Common Market Studies, vol. 40, pp. 89109.
4 The PP was referred to in 27 resolutions of the European Parliament between 1992 and 1999, is referred to
in the 1992 Maastricht Treaty on the European Union, and appeared in a draft constitution for the European
Union. See Sunstein, op. cit., p. 17.
5 See ibid., p. 18.
223
CBA involves attempting to determine the probability of benefits occurring, and

the probability of costs being incurred, as well as determining the relative sizes
of the benefits and costs of a particular course of action and balancing these.
This calculation is compared with the relative balance of costs and benefits
for alternative courses of action from which the option with the best overall
balance (adjusting for the probability of these occurring) is selected.6 From the
1960s, environmental lawyers, policymakers and activists became increasingly
dissatisfied with many of the decisions that were made with the use of CBA.
One source of dissatisfaction was that, in many actual applications, only potential
costs that were established with full scientific certainty were considered. A
precautionary corrective to this tendency was the development of versions of
the PP that were intended to guide the use of CBA so as to ensure that potential
costs other than those established with full scientific certainty were given
due consideration.7 A good example of this form of the PP is Principle 15 of
the Rio Declaration on Environment and Development: In order to protect the
environment, the precautionary approach shall be widely applied by States
according to their capabilities. Where there are threats of serious or irreversible
damage, lack of full scientific certainty shall not be used as a reason for
postponing cost-effective measures to prevent environmental degradation.8
A second complaint about decisions guided by CBA was that in some of these
it was implicitly assumed that the onus of proof of the existence of a cost fell
on critics of an activity. A precautionary response to this tendency was to
supplement CBA with a second step specifying conditions under which the
onus of proof lay with proponents, rather than critics, of an activity. The wellknown Wingspread Statement is a good example of this sort of PP: Where an
activity raises threats of harm to the environment or human health, precautionary
measures should be taken even if some cause and effect relationships are not
fully established scientifically. In this context the proponent of the activity,
rather than the public, should bear the burden of proof.9
Neither insistence on full scientific certainty nor placement of the onus of proof
on critics is intrinsic to CBA; so the above versions of the PP supplement CBA,
6 This decision-making process is philosophically problematic in various ways, most obviously because
benefits often appear to be incommensurable with risks. For a discussion of further philosophical problems
with CBA, see Hansson, S. O. 2007, Philosophical problems in costbenefit analysis, Economics and
Philosophy, vol. 23, pp. 16383. For a defence of CBA against various criticisms, see Schmidtz, D. 2001, A
place for costbenefit analysis, Philosophical Issues (A Supplement to Nous), vol. 11, pp. 14871.
7 For more on the history of the development of this type of precautionary principle, see Magnus, D. 2008,
Risk management versus the precautionary principle: agnotology as a strategy in the debate over genetically
modified organisms, in R. N. Proctor and L. Schiebinger (eds), Agnotology: The Making and Unmaking of
Ignorance, Stanford University Press, Stanford, Calif., pp. 25065.
8 United Nations Environment Programme, 1992.
9 Wingspread Statement on the Precautionary Principle 1998, <www.gdrc.org/u-gov/precaution-3.html>
(viewed 1 November 2010).
224
14. The Precautionary Principle and the Dual-Use Dilemma
rather than replacing it with an alternative. What have come to be known as

strong versions of the PP (sPP), however, have been developed with the intention
of replacing CBA, at least under certain circumstances, with an altogether
different approach to risk management that does not involve weighing the costs
and benefits of a particular policy and comparing these weightings with those
of possible alternative policies. Instead, it involves an exclusive focus on the
potential costs of a particular policy. An example of sPP is the Final Declaration
of the First European Seas at Risk Conference, 1994: If the worst case scenario
for a certain activity is serious enough then even a small amount of doubt as to
the safety of that activity is sufficient to stop it taking place.10
Here we are instructed to focus our attention exclusively on costs and ignore the
potential benefits of particular activities when making policy, no matter how
significant these potential benefits might be.
We have focused on the above three examples of the PP because these are clear
examples of different types of PP that are intended to advance precaution in
particular ways. Not all versions of the PP are as clear as these. Indeed, some
are couched in highly nebulous language that makes it hard to see what their
creators are trying to achieve, beyond conveying enthusiasm for precaution. We
will go on to examine an example of the PP, created especially for application
in dual-use contexts,11 which suffers from this problem. Jordan and ORiordan12
embrace the vagueness of (many versions of) the PP, which they see as a
virtue that enables those who employ it to be more politically effective. This
may be true if the PP is understood, as Jordan and ORiordan understand it,
as an intellectual tool of protest movements; however, its vagueness becomes
a vice when the PP is used to try to steer policy.13 Indeed, scholars who are
concerned about effective precautionary regulation sometimes complain that
while adoption of the PP changes conceptualisations of risks, it does not appear
to have had a clear effect on regulatory practice.14
I have analysed the PP in terms of its different possible relationships to CBA.
Some commentators, such as Sandin,15 might want to object to the direct
10 Cited in Sunstein, op. cit., p. 29.
11 Kuhlau, F., Hoglund, A. T., Evers, K. and Eriksson, S. 2011, A precautionary principle for dual use
research in the life sciences, Bioethics, vol. 25, no. 1, pp. 18.
12 Jordan, A. and ORiordan, T. 1999, The precautionary principle in contemporary environmental policy
and politics, in C. Raffensperger and J. A. Tickner (eds), Protecting Public Health and the Environment:
Implementing the Precautionary Principle, Island Press, Washington, DC, pp. 1535.
13 Clarke, S. 2005, Future technologies, dystopic futures and the precautionary principle, Ethics and
Information Technology, vol. 7, pp. 1216.
14 ORiordan, T. and Cameron, J. (eds) 1994, Interpreting the Precautionary Principle, Earthscan, London;
and Eckley, N. and Selin, H. 2004, All talk, little action: precaution and European chemicals regulation,
Journal of European Public Policy, vol. 11, pp. 78105.
15 Sandin, P. 1999, Dimensions of the precautionary principle, Human and Ecological Risk Assessment,
vol. 5, pp. 889907.
225
comparison of any form of the PP with CBA, taking the view that, while CBA has
been developed to be applicable in circumstances where we need to make policy
decisions under risk, the PP has been developed to address circumstances of
uncertainty. The distinction between risk and uncertainty goes back to Knight.16
In his terminology, situations of risk are circumstances where the probabilities
of possible outcomes can be specified, on the basis of reliable evidence, and
situations of uncertainty are circumstances where the probability of possible
outcomes cannot be specified, on the basis of reliable evidence. Flipping a
normal coin creates an instance of risk without uncertainty as we know all
the possible outcomes of a coin flip (heads and tails) and we are warranted in
specifying particular probabilities for these outcomes. Speculation about the
details of a possible afterlife is a case of uncertainty without risk. There are
many possible accounts of the ways in which an afterlife might be experienced,
but we appear to lack good grounds for assigning probabilities to any of them.
While coin flips are pure instances of risk and speculations about the afterlife
seem to be pure instances of uncertainty, these are exceptional cases. The
majority of real-world cases where decisions involving probabilities need to be
made involve a mixture of risk and uncertainty. Consider a couple of real-world
decisions: 1) an insurer who writes home insurance policies will try to price
those policies on the basis of assessments of the risks to which particular houses
are exposed. It is not possible, however, to anticipate all such risks and work
out exactly how likely these are, so there is inevitably an element of uncertainty
in such assessments. Exact probabilities can, of course, be assigned to possible
outcomes, but such assignments will involve a degree of stipulation. 2) A
patient who is contemplating an operation will want to know what the risks
involved are and the surgeon who is set to conduct the operation will generally
try to transmit this information to the best of her ability. But while it is possible
to anticipate many of the risks involved in a complex operation, estimates of
how likely these are to occur in particular contexts will inevitably involve a
degree of speculation. In many real-world circumstances, we attempt to turn
uncertainties into risks, however, this typically involves some speculation and
so an element of uncertainty remains.17 Most real-world circumstances involve
a mix of risk and uncertainty and the PP and CBA can both be applied to these.
So it is appropriate to make direct comparisons of CBA and the PP in dealing
with most real-world cases.
16 Knight, F. 1921, Risk, Uncertainty, and Profit, Hart, Schaffner & Marx, Boston.
17 Knights view of risk is based on the (I think plausible) presupposition that we try to assign objective
probabilities to the world, and that when we are unable to do so, we are left with uncertainties: LeRoy, S.
F. and Singell, L. D. 1987, Knight on risk and uncertainty, The Journal of Political Economy, vol. 95, pp.
394406. Some commentators, such as Friedman, who take the view that we only ever assign subjective
probabilities to the world, hold that we never need to have recourse to the Knightian concept of uncertainty
at all. Friedman, M. 1962, Price Theory: A Provisional Text, Aldine, Chicago.
226
Precaution, paradox and bias

Strong versions of the PP (sPP) have attracted much controversy and have been
subjected to an apparently devastating form of criticism, which is that they
typically lead to contradictory policy recommendations, if applied consistently,
and are therefore paralysing.18 This is because viable alternatives to a policy
typically have risks associated with them and if all of these are given due
consideration under sPP then policy paralysis will result. Consider, for example,
the application of sPP to the possible development of nuclear power plants,
intended to address an impending shortage of energy in a particular country.
Clearly there are risks associated with building and running nuclear power
plants, so an application of sPP would lead to the recommendation that we do
not develop nuclear power plants. But what will happen if we fail to develop
nuclear power in this context? Viable alternatives seem also to involve risks. If
we develop new sources of power that are based on fossil fuels then we increase
the chance and severity of climate change. So sPP precludes the development
of new fossil fuel-based power sources. Another alternative is not to provide
additional power but to try to convince people to use less energy. But this
alternative involves the risk of social instability, so sPP precludes the policy of
not developing new sources of power. If the three options considered above are
our only viable alternatives then the consistent application of sPP necessarily
provides incoherent policy recommendations, precluding all viable courses of
action, and also precluding inaction.
One way that advocates of sPP have attempted to avoid paralysing paradoxical
outcomes is by insisting that only risks over a certain threshold of significance
(understood in terms of probability of occurrence and/or seriousness of costs
involved) be considered in formulations of the PP.19 Indeed our exemplar of
sPP, the Final Declaration, mentions risks being serious enough to warrant
consideration (although it does not specify how likely it is that these will be
realised), so it might be supposed that it is an instance of sPP that can avoid
paradoxical outcomes. The advocate of this strategy to defend sPP from
the change of paradox needs to make a choice about where to set the levels
of significance and/or likelihood of occurrence of risks. If these are set too
high then sPP will not be applicable to many of the risks that advocates of
precaution have wished to apply it to. On the other hand, if they are set too low
then the problem of paradox wont be avoided and sPP will continue to issue
contradictory advice.20 In effect this is a Goldilocks strategy: it might work
18 Manson, N. A. 2002, Formulating the precautionary principle, Environmental Ethics, vol. 24, pp. 263
74; and Sunstein, op. cit.
19 Sandin, P., Peterson, M., Hannson, S. O., Rudn, C. and Juthe, A. 2002, Five charges against the
precautionary principle, Journal of Risk Research, vol. 5, pp. 28799.
20 See Clarke, op. cit., p. 126.
227
if we know in advance where to set levels of significance, but because we are

dealing with the management of uncertain outcomes we usually dont have any
reliable way of knowing where to set such thresholds, so there is not usually any
reason to be confident that our strategy has succeeded. There have been other
attempts to evade or resolve the paradox of sPP,21 but they do not appear to be
successful either.22
If sPP leads to incoherent policy recommendations and causes policy paralysis
then it seems odd that this is not often noticed by those who attempt to apply
sPP, such as those who attempt to apply the Final Declaration. The reason this
is not often noticed is that when sPP is applied it is usually applied selectively.
We are able to apply sPP when we focus on one possible policy outcome and
fail to consider alternatives; and it seems that we have a propensity to do this.
This propensity is encouraged by the structure of the PP. Whereas CBA involves
an explicit consideration of alternative policies, most versions of the PP are
designed to be applied to one policy option at a time, and do not involve explicit
comparison with alternatives. According to Sunstein,23 there is also a deeper
reason we tend to consider risks selectively when applying the PP, which is that
we are hostage to a variety of significant cognitive biases that have the effect
of blinding us to the fact that we are typically faced with risks however we
decide to act (and even if we dont act). Sunstein mentions various sources of
cognitive bias that have this effect including probability neglect, a belief in the
benevolence of nature, loss aversion and systems neglect;24 however, the factor
he considers the most significant in encouraging a selective approach towards
risk is the widespread use of the availability heuristic.25
The availability heuristic is a rule of thumb that people intuitively use to estimate
the magnitude of particular risks. If I am asked how likely an earthquake, a
flood or a bushfire is in Oxford, I will tend to make intuitive assessments of the
likelihood of these events by seeing how readily I can bring to mind episodes in
which such events have taken place in Oxford and will adjust my estimates in
accordance with the availability to me of instances of such events. Availability
21 Gardiner, S. M. 2006, A core precautionary principle, Journal of Political Philosophy, vol. 14, pp. 3360;
and Weckert, J. and Moor, J. 2006, The precautionary principle in nanotechnology, International Journal of
Applied Philosophy, vol. 20, pp. 191204.
22 Clarke, S. 2009, New technologies, common sense and the paradoxical precautionary principle, in P.
Sollie and M. Duwell (eds), Evaluating New Technologies: Methodological Problems for the Ethical Assessment
of Technological Developments, Springer, Dordrecht, pp. 15973.
23 Sunstein, op. cit.
24 Ibid., p. 37.
25 See ibid., p. 5. For more on the availability heuristic, see Tversky, A. and Kahneman, D. 1982, Judgment
under uncertainty: heuristics and biases, in D. Kahneman, P. Slovic and A. Tversky (eds), Judgment under
Uncertainty: Heuristics and Biases, Cambridge University Press, Cambridge, pp. 321; and Kahneman, D. and
Frederick, S. 2002, Representativeness revisited: attribute substitution in intuitive judgment, in T. Gilovich,
D. Griffin and D. Kahneman (eds), Heuristics and Biases: The Psychology of Intuitive Judgment, Cambridge
University Press, Cambridge, pp. 4981.
228
is influenced by both familiarity and salience.26 The more familiar I am with a

class of events, all things being equal, the greater will be my intuitive assessment
of its likelihood. If a class of events is particularly salient to meif, say, I have
personal experience of a flood or have just watched a documentary about the
danger of flooding in the Thames Valley, where Oxford is locatedthen that
will also increase my intuitive estimate of the likelihood of flooding in the area.27
When particular risks are highly available to us they tend to crowd out other
risks that we would otherwise be inclined to consider. After the events of 11
September 2001, ordinary estimates of the dangers associated with air travel
rose dramatically, particularly in America, and this led many people to alter
their travel plans and drive or take a train rather than fly. It seems that such
decisions, made at that time, were a response to the high availability of the
dangers of air travel, which crowded out consideration of the risks involved
with other forms of transport. But the risks of alternative travel choices were
significant. It is estimated than an extra 350 road fatalities occurred in America
in the final three months of 2001 as a result of people avoiding air travel in the
aftermath of the events of 11 September. This is a higher number of deaths than
the number of deaths of passengers and crew in all four of the crashed flights of
11 September combined (266 deaths).28
Dual use and lessons from debates about the

precautionary principle
A dual-use dilemma arises when a piece of researchtypically in the life
scienceshas potential benefits as well as the potential to cause harm.29 If we
go ahead and conduct this research then we may enjoy the benefits that might
result from it, but only if we are willing to bear the risk of potential harms. If
we do not conduct such research then potential harms are avoided but benefits
are also forgone. How are we to decide what to do? The most obvious way to
attempt to make such decisions is by applying CBA, under which we attempt to
weigh the potential benefits of use of the potential piece of research in question
against the potential harms. It has recently been suggested, however, that the PP
26 See Sunstein, op. cit., p. 37.

27 The passage of time will tend to decrease both familiarity and salience, and thereby lead to a decrease in
intuitive estimates of likelihood.
28 Gigerenzer, G. 2004, Dread risk, September 11 and fatal traffic accidents, Psychological Science, vol. 15,
pp. 2867.
biological sciences, Science and Engineering Ethics, vol. 13, p. 524.
229
could be usefully applied in dual-use contexts.30 Even more recently, a specific

formulation of the PP for application in dual-use life-science research has been
developed, which we will go on to consider.
If a PP is to be applied to dual-use dilemmas then I would urge those who
are attempting to apply it to be as clear as they can about what role the PP
is supposed to play in the resolution of dual-use dilemmas. As we have seen,
it might be expected to play (at least) three very different roles in decisionmaking. First, the PP might be used to ensure that some harms, which might not
be properly considered, are properly considered in applications of CBA in dualuse contexts. Second, it might be invoked in an attempt to specify where the
onus of proof lies in respect of either the significance of particular harms or the
viability of particular remedies for those harms. Third, it might be invoked as an
alternative to CBA. There is no one way to be cautious, and it is important that
we try to be clear about how we intend to be cautious, if we are to do so in dualuse contexts. It is also important that we explain why we need to be cautious in
this or that particular way. What is the particular failing of current approaches
to dual-use dilemmas that the PP is supposed to help address; and are there any
other possible responses to this problem that we might also consider?
As we have seen, one of the key problems with some versions of the PP (sPP) is that
application of these depends on us adopting a selective approach to risk. We can
adopt a selective approach to risk and be oblivious to the fact that we are doing
so because of widespread cognitive bias that affects ordinary assessments of
risk. We should ask ourselves whether framing a problem as a dual-use dilemma
makes us more or less susceptible to adopting selective attitudes towards risk.
We should also ask ourselves if applying sPP (or other versions of the PP) adds to
such problems or reduces these. When we are considering whether to develop a
new drug, framing our choice as a dual-use dilemma encourages us to focus our
attention on a comparison of the consequences of going ahead and developing
the drug, which will involve potential benefits as well as potential costs, with
the consequences of not developing the drug. We may be in a position, however,
to develop a variety of different drugs that provide overlapping benefits and
involve overlapping risks. Rather than considering choices about which of these
to develop and which not to develop as a series of isolated dilemmas, it may be
more sensible to try to make a comparison of the overall benefits and costs of
each, before deciding which options to pursue.
Framing a problem as a dual-use dilemma may well increase our propensity to
adopt selective approaches to risk. When we frame a potential development as a
dual-use dilemma (or any other sort of dilemma), we are encouraging a focus on
30 See Rappert, B. 2008, The benefits, risks and threats of biotechnology, Science and Public Policy, vol.
35, p. 40.
230
a choice between exactly two alternatives; however, as is the case with the above
example, there may be contexts in which we may be able to make comparative
choices that are more complicated than simple dilemmas, and such possibilities
may be obscured from us by framing our choice as a dilemma.31
Framing a problem as a dual-use dilemma encourages us to consider a choice
between two alternatives in isolation from other possible choices. And at least
some versions of the PP (sPP) can only be applied if we make a decision about
risks in isolation from consideration of alternatives. Other versions of the PP,
including the Wingspread Statement and the Rio declaration, are also structured
around consideration of the risks that are associated with a particular activity
and do not encourage us to consider risks involved with alternatives to that
activity. The concern here is that if the PP is applied to the dual-use dilemma
then the tendency of both of these conceptual structures to focus our attention
on a particular policy option, to the exclusion of consideration of other options,
may reinforce one another. So we have a reason to be especially wary about
applying PP when that is combined with the framing of options as dilemmas.
A suggested precautionary principle for dualuse contexts

Kuhlau et al. suggest the following formulation of the PP for application in dualuse life-science research: When and where serious and credible concern exists
that legitimately intended biological material, technology or knowledge in the
life sciences pose threats of harm to human health and security, the scientific
community is obliged to develop, implement and adhere to precautious measures
to meet the concern.32
This is a very vague formulation of the PP and a key problem here is to understand
what sort of PP Kuhlau et al.33 intend us to apply. The most straightforward
reading is that they are pointing out that when the risk of serious costs is present,
serious remedies will be required if these are to be addressed. The problem with
this suggestion is that it is hard to see what work is being done by the PP, as
an implication of ordinary CBA is that when the risk of serious costs is at stake
serious remedies will be required if these are to be addressed.
31 This problem may become resolved as language evolves. The word dilemma is sometimes used these
days in a way that is meant to be inclusive of tri-lemmas, quadri-lemmas, and so on. If this usage becomes
sufficiently common then framing a decision as a dilemma will no longer encourage a forced choice between
exactly two alternatives.
32 Kuhlau et al., op. cit., p. 6.
33 Ibid., pp. 18.
231
But perhaps Kuhlau et al. mean to offer the scientific community a reminder
that there are threats to human health and security that can arise from lifescience research and they are insisting that the scientific community considers
these when deciding whether or not to conduct particular research that may
lead to harms. On this second reading, their version of the PP is supposed to
function like the Rio declaration and is designed to ensure that, when costs and
benefits are weighed up, the potential costs of certain risks are not excluded
from consideration. If this is the sort of reading that is intended then it would
be useful to know what motivates Kuhlau et al. to suppose that there is a danger
of these concerns not being considered, when consideration of them is expected
under the application of ordinary CBA. The Rio declaration was developed
in response to a history of significant risks being ignored on the (fallacious)
grounds that if these had not been established with full scientific certainty
then they should not be considered at all. But is there a reason to think that
threats to human health and security are liable to be ignored by the scientific
community?34 If there is no particular reason to believe that this may happen
then it seems that Kuhlau et al.35 have created a gratuitous version of the PP that
performs no needed function.
On a third possible reading, Kuhlau et al. intend a strong version of the PP.
Their wording is ambiguous, but they may intend that the scientific community
only allows research in the life sciences to go ahead once all concerns regarding
possible threats to human health and security have been met, no matter how
potentially beneficial such research may be. If they do intend this strong
reading then it would be useful to know how the following two questions can
be answered. First, it would be useful to know why they think we should accept
this strong precautionary approach. If the benefits of some piece of research are
judged to outweigh the risks, and if all significant possible costs and benefits
have been considered, why shouldnt we accept the risks of conducting such
research, in order to try to reap potential benefits? Second, we need to be told
how the problem of paradox is to be avoided.36 In many situations there will be
risks involved in not conducting research in the life sciences. If we do not take
the risks involved in developing new vaccines for currently lethal diseases, for
example, we implicitly accept the risks of people continuing to die as a result of
34 Tom Douglas suggests that there is a tradition in the scientific community of denying that scientists
should be held responsible for the ways in which scientific knowledge is employed and that Kuhlau et al.
may be intending to issue a special reminder to scientists, as a way of trying to overcome the influence of
this tradition. Also Selgelid suggests that biological scientists and bioethicists have a history of ignoring the
dual-use potential of genetics. See Selgelid, M. 2010, Ethics engagement of the dual-use dilemma: progress
and potential, in B. Rappert (ed.), Education and Ethics in the Life Sciences, ANU E Press, Canberra, pp. 2334.
35 Kuhlau et al., op. cit.
36 Kuhlau et al. consider some objections to the PP, including the objections that it stifles scientific
development, it lacks practical applicability and that it is poorly defined and vague. They do not consider the
applicability of these charges to specific versions of the PP and they do not consider the charge that (strong
versions of) the PP leads to contradictory policy recommendations.
232
these diseases. It looks like Kuhlau et al.s version of the PP (on a strong reading)
is susceptible to the problem of there being risks on all sides and so it looks
like it leads to paradoxical recommendations, if applied consistently.37
The PP can be useful in particular contexts when it has specific uses. There
may be particularly good uses for a PP developed to suit dual-use contexts.
Unfortunately, Kuhlau et al.38 are not clear about the uses to which they wish
to put their version of the PP and it looks like their version is not designed to
suit any specific use. If others wish to develop new versions of the PP for dualuse contexts then I would urge that they use precise language and specify what
their version of the PP is intended to achieve.
37 Thanks to Linsey McGoey, Brian Rappert and Tom Douglas for some very helpful comments on an earlier
version of this chapter.
38 Kuhlau et al., op. cit.
233
Part III: Ethical Practices
15. Scientific Control

Over Dual-Use Research:
Prospects for Self-Regulation
David B. Resnik
Introduction
In the past decade, scientists, policymakers, ethicists and citizens have become
increasingly aware that scientific research that promotes public health and
safety has the potential to be used for terrorist, criminal or other malevolent
purposes.1 The phrase dual-use research refers to research that may have
beneficial as well as detrimental consequences. The National Science Advisory
Board for Biosecurity (NSABB), a US Government committee that provides advice
to researchers and federal agencies, has defined dual-use research of concern as
research that, based on current understanding, can be reasonably anticipated to
provide knowledge, products, or technologies that could be directly misapplied
by others to pose a threat to public health and safety, agricultural crops and
other plants, animals, the environment or materiel.2 Examples of published
research that has raised dual-use issues include research on a mousepox virus
that could be used to enhance the virulence of the human smallpox virus,3 a
study showing how to manufacture a polio virus from available sequence data
1 Atlas, R. 2002, National security and the biological research community, Science, vol. 298, pp. 753
4; Atlas, R. and Dando, M. 2006, The dual-use dilemma for the life sciences: perspectives, conundrums,
and global solutions, Biosecurity and Bioterrorism, vol. 4, pp. 27686; National Science Advisory Board for
Biosecurity (NSABB) 2007, Proposed Framework for the Oversight of Dual Use Life Sciences Research: Strategies
for Minimizing the Potential Misuse of Research Information, National Science Advisory Board for Biosecurity,
Bethesda, Md, <http://oba.od.nih.gov/biosecurity/pdf/Framework%20for%20transmittal%200807_Sept07.
pdf> (viewed 30 December 2011); Selgelid, M. 2007, A tale of two studies; ethics, bioterrorism, and the
censorship of science, Hastings Center Report, vol. 37, no. 3, pp. 3543; Resnik, D. B. and Shamoo, A. S.
2005, Bioterrorism and the responsible conduct of biomedical research, Drug Development Research, vol. 63,
pp. 12133; Selgelid, M. 2009, Governance of dual use research: an ethical dilemma, Bulletin of the World
Health Organization, vol. 87, pp. 7203; National Research Council 2004, Biotechnology Research in the Age of
Terrorism, The National Academies Press, Washington, DC; National Research Council 2006, Globalization,
Biosecurity and the Future of the Life Sciences, The National Academies Press, Washington, DC; Resnik, D. B.
2010, Can scientists regulate the publication of dual use research? Studies in Ethics, Law, and Technology, vol.
4, no. 1, article 6, <http://www.bepress.com/selt/vol4/iss1/art6> (viewed 30 December 2011).
2 NSABB, op. cit.
3 Jackson, R., Ramsay, A., Christensen, C., Beaton, S. and Hall, D. 2001, Expression of mouse interleukin-4
by a recombinant ectromelia virus suppresses cytolytic lymphocyte responses and overcomes genetic
resistance to mousepox, Journal of Virology, vol. 75, pp. 120510.
237
and mail-order supplies,4 a study on the genetics of human smallpox virus that
could be used to develop a strain of Vaccinia (the virus used in smallpox vaccine)
that would overcome the immune systems natural defences,5 a paper describing
how a terrorist could poison the US milk supply with botulinum toxin,6 and a
paper demonstrating how to reconstruct the 1918 Spanish influenza virus from
published sequence data.7
Preventing the use of scientific research for malevolent purposes raises dilemmas
for ethics and public policy. Although most of these issues have existed in some
form or another since the dawn of science, the globalisation of scientific research,
advances in biotechnology and the spectre of bioterrorism create problems for
bioscientists and other researchers that are complex, difficult and urgent.8 The
fundamental dilemma related to dual-use research is how to protect society
from harm without unduly hampering the advancement of science.9 While
some restrictions on research are widely recognised as necessary to prevent the
misuse of science, administrative, legal and bureaucratic oversight of research
can hinder collaboration, the sharing of data and materials, and publication.10
Policymakers have explored two forms of oversight of dual-use research:
governmental control and self-regulation.11 Government oversight mechanisms
that have been used or proposed include12
regulations, such as the Patriot Act, that control access to and transfer and
storage of dangerous biological, chemical or radiological materials13
laws related to the transfer of technology across national borders14
4 Cello, J., Paul, A. and Wimmer, E. 2002, Chemical synthesis of poliovirus cDNA: generation of infectious
virus in the absence of natural template, Science, vol. 297, pp. 101618.
5 Rosengard, A, Liu, Y., Nie, Z. and Jimenez, R. 2002, Variola virus immune evasion design: expression of
a highly efficient inhibitor of human complement, Proceedings of the National Academy of Sciences, vol. 99,
pp. 880813.
6 Wein, L. and Liu, Y. 2005, Analyzing a bioterror attack on the food supply: the case of botulinum toxin in
milk, Proceedings of the National Academy of Sciences, vol. 102, pp. 99849.
7 Tumpey, T. M., Basler, C. F., Aguilar, P. V., Zeng, H., Solrzano, A., Swayne, D. E., Cox, N. J., Katz, J. M.,
Taubenberger, J. K., Palese, P. and Garca-Sastre, A. 2005, Characterization of the reconstructed 1918 Spanish
influenza pandemic virus, Science, vol. 310, pp. 7780.
8 Atlas and Dando, op. cit.
biological sciences, Science and Engineering Ethics, vol. 13, pp. 52380.
10 Ibid.; Atlas and Dando, op. cit.
11 Miller and Selgelid, op. cit.; Atlas and Dando, op. cit.
12 Miller and Selgelid, op. cit.
13 Malakoff, D. 2002, Biological agents: new U.S. rules set the stage for tighter security, oversight, Science,
vol. 298, p. 2304; Bhattacharjee, Y. 2010, Biosecurity: new biosecurity rules to target the riskiest pathogens,
Science, vol. 329, pp. 2645.
14 National Research Council, 2006, op. cit.
238
15. Scientific Control Over Dual-Use Research: Prospects for Self-Regulation
classification of government-sponsored research that could threaten national

security if disclosed15
restrictions on government funding of research with potential dual-use
implications16
restrictions on immigration of scientists with questionable backgrounds
censorship of non-classified research that poses a threat to national security.17
Self-regulation by the scientific community might include18
control over the dissemination of materials and technology by researchers
and their institutions19
review of research by committees charged with overseeing research20
education and training in responsibilities related to dual-use research21
development of professional ethics codes and institutional policies that
address dual-use research22
journal review of dual-use research.23
There is currently a vigorous debate over whether government oversight, selfregulation by the scientific community or some combination is the best way to
control dual-use research.24 The main argument for government regulation is
that self-regulation by the scientific community has significant limitations and
is often ineffective.25 There is some justification for this view. First, scientists
(and non-governmental research organisations) lack the legal authority to
adequately control some types of research. For example, scientists do not have
the authority to classify information, restrict access to dangerous materials or
prevent the transfer of technologies across national borders.26 Second, scientists
often lack the expertise or resources needed to make decisions pertaining to
dual-use research.27 For example, scientists reviewing dual-use research for
15 Resnik, D. B. 2009, Playing Politics with Science, Oxford University Press, New York; National Research
Council, 2004, op. cit.
16 Resnik and Shamoo, op. cit.
17 van Aken, J. 2006, When risk outweighs benefit: dual-use research needs a scientifically sound risk
benefit analysis and legally binding biosecurity measures, EMBO Reports, vol. 7(SI), pp. S1013; Miller and
Selgelid, op. cit.
18 NSABB, op. cit.
19 Atlas and Dando, op. cit.
20 NSABB, op. cit.
21 Somerville, M. and Atlas, R. 2005, Ethics: a weapon to counter bioterrorism, Science, vol. 307, pp.
18812.
22 Rappert, B. 2004, Responsibility in the life sciences: assessing the role of professional codes, Biosecurity
and Bioterrorism, vol. 2, pp. 16474.
23 van Aken, J. and Hunger, I. 2009, Biosecurity policies at international life science journals, Biosecurity
24 Miller and Selgelid, op. cit.; van Aken, op. cit.; Resnik, 2010, op. cit.; Atlas and Dando, op. cit.
25 Miller and Selgelid, op. cit.; Selgelid, op. cit.; van Aken, op. cit.
239
an institutional biosafety committee (IBC) may not have access to information

related to national security interests or threats. Third, scientists have a vested
interest in promoting research that may conflict with their social responsibility
to prevent harmful uses of research.28 Journals may be more interested in
publishing research that will generate interest and discussion than in preventing
the dissemination of information that could threaten society.
The argument against government regulation is that it may pose a significant threat
to the advancement of science. Freedom of inquiry, association and expression
are essential to scientific research.29 Government restrictions on science and
technology can undermine collaboration, creativity, criticism, publication and
sharing of data, materials and methods. Especially worrisome are restrictions
on publicationthe lifeblood of science.30 Since the 1980s, the official policy
of many governments, including the US Government, has been not to interfere
with the publication of non-classified scientific research;31 however, concerns
about dual-use research have led some commentators to question the wisdom of
this policy.32 The NSABB has reviewed papers with dual-use implications, but
its opinions are purely advisory and do not determine publication decisions.33
Government restrictions on access to research materials can also have negative
impacts on science, since they can interfere with collaborations among
researchers, especially international ones. In the United States, scientists must
submit to background checks in order to access dangerous biological materials
classified as select agents,34 which has generated concerns about the potential
for racial or ethnic discrimination.35 Any type of government restriction on
scientific research also raises the possibility of politically motivated interference
in the scientific process.36
I will not discuss or assess government oversight of dual-use research in depth,
since the aim of this chapter is to examine the prospects for self-regulation
of dual-use research; however, the conclusions reached in this chapter may
have implications for government oversight, since one of the main arguments
for government involvement is that attempts at self-regulation are likely to be
ineffective. This chapter will describe the different mechanisms that scientists
might use to control dual-use research and assess the potential effectiveness
28Ibid.
29 Resnik, 2009, op. cit.
30 Journal Editors and Authors Group 2003, Uncensored exchange of scientific results, Proceedings of the
National Academy of Sciences, vol. 100, p. 1464.
33 NSABB, op. cit.
34 Bhattacharjee, op. cit.
240
of these mechanisms, including: control over transfer of materials, review

by institutional committees, education and training, professional codes, and
journal policies.
Control over the transfer of materials

Every day, scientists send materials and technologies, such as reagents, cell lines,
pathogens and laboratory animals, to other scientists. Most institutions sign a
material transfer agreement (MTA) when they transfer materials. MTAs state
the terms and conditions for using such materials, and can be enforced through
contract law. Although laws and regulations govern the transfer of some types
of materials, such as select biological agents and radioactive substances, they do
not cover all the types of materials that might be used for harmful purposes, as
some materials might not be deemed dangerous enough to warrant regulations,
and other dangerous materials might not yet have been identified (as such). For
example, when investigators published research on the 1918 Spanish influenza
virus, this pathogen was not on the list of select agents, and it was not until later
that the list was amended to cover this material.37
Scientists can play an important role in preventing the transfer of dangerous
materials to individuals or organisations with reprehensible motives by
attempting to determine whether the request for materials comes from a
legitimate, respected researcher or whether it comes from a suspicious individual
or organisation. They can ask for curriculum vitae, research protocol and other
information necessary to verify the requestors expertise, background and
affiliations. If they have any doubts about the legitimacy of the request, they
can delay the transfer until they are satisfied that the requestor is a responsible
scientist, or they can consult institutional officials on how to proceed.
Whether scientific control over the transfer of materials and technologies will
be an effective mechanism for preventing the use of these items for malevolent
purposes remains to be seen. For this mechanism to be effective, investigators
must first be aware of the dual-use implications of transferring materials and
technologies. A survey of attitudes of US life scientists concerning dual-use
research indicates that most are aware of biosecurity issues, although a majority
considers the risk of a bioterror attack to be small. A small percentage of those
surveyed indicated that they had refused to collaborate with some individuals
due to biosecurity concerns. 38 Although many scientists are aware of dual-use
37 National Select Agent Registry 2010, <http://www.selectagents.gov/Select%20Agents%20and%20
Toxins%20List.html> (viewed 30 December 2011).
38 Committee on Assessing Fundamental Attitudes of Life Scientists as Basis for Biosecurity Education
2009, A Survey of Attitudes and Actions on Dual Use Research in the Life Sciences: A Collaborative Effort of
241
issues, education (discussed below) can help to raise awareness. Investigators

must also be willing to invest some of their time and effort into verifying the
legitimacy of individuals or organisations requesting materials or technologies.
Investigators who have extensive responsibilities related to research, teaching
or administration may feel that they do not have enough time to devote such
attention to requests for materials and technologies. Educational sessions may
be able to convince investigators of the importance of this issue.
Review by institutional committees

Several different institutional committees may have the opportunity to review
research with dual-use implications. These committees can decide whether the
research should be approved, disapproved or approved with additional oversight
or restrictions to prevent harmful consequences. Most of the discussion thus
far has focused on the role of IBCs in addressing the dual-use implications
of research. IBCs are responsible for reviewing research involving hazardous
biological materials, such as pathogens and toxins, as well as recombinant
DNA research, including experiments to transfer genes to plants, animals and
humans.39 IBCs are in charge of protecting researchers and the public from risks
related to research on hazardous biological materials. IBCs include experts in
microbiology, genetics, pathology, immunology and other disciplines with the
requisite expertise. IBCs are a natural place for the review of research with dualuse implications, given their function and expertise.40 Institutional animal care
and use committees (IACUCs) review research involving research on animals.
The main function of IACUCs is to protect the welfare of laboratory animals, and
to ensure that investigators are using methods that minimise pain, suffering and
distress.41 Though IACUCS lack the mandate and expertise to review dual-use
issues, they could refer any research that raises concerns to an IBC. Institutional
review boards (IRBs) review research involving human subjects, and their main
function is to protect the rights and welfare of individuals participating in
research. IRBs, like IACUCs, lack the mandate or expertise to review dual-use
issues, but they could make a referral to an IBC.42
the National Research Council and the American Association for the Advancement of Science, The National
Academies Press, Washington, DC, <http://books.google.com/books/about/A_Survey_of_Attitudes_and_
Actions_on_Dua.html?id=2-MLru0bHkQC> (viewed 30 December 2011).
39Cornell University 2007, Institutional Biosafety Committee, <http://www.ibc.cornell.edu/
responsibilities> (viewed 30 December 2011).
40 NSABB, op. cit.
41 Shamoo, A. S. and Resnik, D. B. 2009, Responsible Conduct of Research, 2nd edn, Oxford University Press,
New York.
42 Resnik, D. B. 2010, Dual-use review and the IRB, Journal of Clinical Research Best Practices, vol. 6, no.
1, <http://www.firstclinical.com> (viewed 30 December 2011).
242
Although these three committees have the ability to identify scientific research
with dual-use implications, inevitably some research may fall through the cracks,
such as biomedical research that does not involve dangerous pathogens or animal
or human subjects; chemical research; and engineering research. For example,
the study of how to infect the US milk supply with botulinum toxin would
not have been reviewed by any IBC, IACUC or IRB.43 Because these committees
may lack the expertise to assess the dual-use implications of research that falls
within their purview, the NSABB has recommended that institutions consider
appointing committees that focus on dual-use research.44 It is not known
how many institutions have created such committees, although the National
Institutes of Health (NIH) intramural program has a dual-use committee.45
Review of research by institutional committees is likely to be one of the most
effective ways of preventing harms related to dual-use research. First, most
institutions already have IBCs, IACUCs and IRBs or equivalent committees.
Second, since these committees are accustomed to dealing with a variety of
ethical, legal and social issues related to research, it will not come as a surprise
to most committee members that they should be mindful of dual-use concerns.
Third, these committees are likely to know how to access the appropriate
institutional officials (for example, vice-president for research, compliance
officer) for dealing with dual-use issues. Although institutional committees
are well positioned to deal with dual-uses issues, it is not known how often
these committees encounter research that raises dual-use concerns, whether
they recognise the dual-use implications, or how they typically respond to
these situations. More research is needed to get a better understanding of the
effectiveness of institutional committees of dealing with dual-use issues.
Education and training

The NSABB46 and other commentators47 have recommended that investigators,
students and trainees receive education concerning their responsibilities related
to dual-use research. Education should cover topics pertaining to dual use, such
as select agents, control of dangerous materials and the potential consequences
of research and publication. Many funding organisations and institutions
already require or support some type of education in responsible conduct of
research (RCR) to promote scientific integrity. In the United States, the NIH
and the National Science Foundation (NSF) mandate research ethics training
43
44
45
46
47
Wein and Lu, op. cit.

NSABB, op. cit.
J. Schwartz, Personal Communication, 15 December 2009.
NSABB, op. cit.
Resnik and Shamoo, op. cit.; Sommerville and Atlas, op. cit.
243
requirements for students supported with grants funds.48 Education in RCR varies
considerably. Some institutions require students to complete online modules
in RCR, while others require attendance at classes, workshops or seminars.49
RCR education typically addresses fabrication, falsification, plagiarism, data
management, conflict of interest, authorship, publication, mentoring and
peer review, but this list could be expanded to include dual-use issues. It is
not known how many institutions have conducted education and training in
dual-use research, and more research is needed on this topic. In 2009, the NIH
required that all intramural researchers and trainees receive education on their
social responsibilities related to dual-use research. The educational materials
developed by the NIH included an overview of dual-use research as well as
several case studies.50
While education in dual-use issues is likely to play an important role in
preventing the misuse of scientific research or materials, its effectiveness has been
questioned. According to some studies, RCR education does not decrease the
prevalence of negative behaviour, such as data falsification or plagiarism.51 Other
studies suggest that RCR education can promote awareness of ethical issues and
knowledge of ethical concepts, but does not influence attitudes or behaviour.52
Although there have been no published studies on the effectiveness of dual-use
education and training, it is plausible to hypothesise that these efforts may have
mixed results, given the scientific communitys experiences with RCR education
and training. A survey of life scientists indicates they support education and
training initiatives that cover dual-use research;53 however, more research is
needed on the effectiveness of dual-use education, and the effectiveness of
different pedagogical techniquesfor example, seminars, online training, and
so on.
48 Shamoo and Resnik, op. cit.

49Ibid.
50 Schwartz, op. cit.
51 Anderson, M. S., Horn, A. S., Risbey, K. R., Ronning, E. A., de Vries, R. and Martinson, B. C. 2007, What
do mentoring and training in the responsible conduct of research have to do with scientists misbehavior?
Findings from a national survey of NIH-funded scientists, Academic Medicine, vol. 82, pp. 85360.
52 Schmaling, K. B. and Blume, A. W. 2009, Ethics instruction increases graduate students responsible
conduct of research knowledge but not moral reasoning, Accountability in Research, vol. 16, pp. 26883;
Plemmons, D., Brody, S. and Kalichman, M. 2006, Student perceptions of the effectiveness of education in
the responsible conduct of research, Science and Engineering Ethics, vol. 12, pp. 57182; Powell, S., Allison,
M. and Kalichman, M. 2007, Effectiveness of a responsible conduct of research course: a preliminary study,
Science and Engineering Ethics, vol. 13, pp. 24964.
53 Committee on Assessing Fundamental Attitudes of Life Scientists as Basis for Biosecurity Education, op.
cit.
244
Professional codes
The NSABB54 and other commentators55 have recommended that professional
associations should develop codes of ethics (or conduct) to promote selfregulation of dual-use research. In response to the dual-use concerns that arose
in 2001, the American Society for Microbiology (ASM) revised its code of ethics,
which now includes the following paragraph:
ASM members are obligated to discourage any use of microbiology
contrary to the welfare of humankind, including the use of microbes
as biological weapons. Bioterrorism violates the fundamental principles
upon which the Society was founded and is abhorrent to the ASM and
its members. ASM members will call to the attention of the public or
the appropriate authorities misuses of microbiology or of information
derived from microbiology.56
Other professions with codes that address dual-use issues include the
International Union of Biochemistry and Molecular Biology (IUBMB)57 and
the American Medical Association (AMA).58 It is not known how many other
professions have such codes but development thus far has been sparse.59 The
NSABB has published some considerations for developing a code of conduct
related to dual-use research in the life sciences, which address such topics
as proposing, managing and conducting research; collaborations; public
communications; and mentoring.60
The effectiveness of professional codes at promoting ethical conduct has been
the subject of much debate.61 Some argue that professional codes are merely
symbolic statements of values that have little effect on behaviour. Others
argue that codes provide useful guidance for members of the profession and
help promote public trust by defining standards of conduct;62 however, most
professional codes lack enforcement mechanisms. Those that are the most
effective are linked to professional licensure or certification, which can result
54 NSABB, op. cit.
55 Rappert, op. cit.
56 American Society for Microbiology (ASM) 2005, Code of Ethics, <http://www.asm.org/ccLibraryFiles/
FILENAME/000000001596/ASMCodeofEthics05.pdf> (viewed 30 December 2011).
57 International Union of Biochemistry and Molecular Biology (IUBMB) 2005, Code of Ethics, <http://www.
babonline.org/bab/babcethics.pdf> (viewed 30 December 2011).
58 American Medical Association (AMA) 2004, Code of Medical Ethics, Opinion 2.078Guideline to Prevent
Malevolent Use of Biomedical Research, <http://www.ama-assn.org/ama/pub/physician-resources/medicalethics/code-medical-ethics/opinion2078.shtml> (viewed 30 December 2011).
59 Rappert, B. 2007, Codes of conduct and biological weapons: an in-process assessment, Biosecurity and
Bioterrorism, vol. 5, pp. 110.
60 NSABB, op. cit.
61 Rappert, 2007, op. cit.
62 Bayles, M. 1988, Professional Ethics, 2nd edn, Wadsworth, Belmont, Calif.
245
in suspension of ones licence or disbarment for misbehaviour. For example,

lawyers who violate ethical standards set by their bar association can have their
licence to practise law revoked.63 Since licensure by a professional association
is not necessary to conduct many types of scientific research, scientific codes of
ethics usually lack the coercive power to control behaviour. A microbiologist
who violates the ASM code of ethics can still practise microbiology even if he
is sanctioned by the organisation. Despite these limitations, many argue that
scientific professional codes can still play an important role in guiding conduct,
even if they lack coercive power.64
Institutional policies
The NSABB has also recommended that institutions develop policies addressing
dual-use issues, complimenting existing RCR policies.65 Many scholars and
commentators have argued that institutional policies, combined with education
and training, can help promote ethical conduct in research;66 however, it is
difficult to assess the effectiveness of institutional policies due to social, economic
and cultural differences among institutions. Studies have shown that there is
considerable variation in the content of some institutional ethics policies, such
as conflict-of-interest and misconduct rules.67 In the United States, the content of
institutional policies is largely driven by government mandates, such as NIH or
NSF rules that require institutions to establish conflict-of-interest, misconduct,
animal research and human subjects research policies as a condition of obtaining
funding.68 Federal agencies could promote institutional policy development by
requiring that funding recipients have dual-use policies; however, so far no
federal agency has done so.
Institutional dual-use policies could play a key role in preventing the
misuse of scientific work for malevolent purposes. Policies could address
controlling, securing, transferring and accessing hazardous materials, as well
as publishing research that could be readily used to cause significant harm to
63 Rotunda, R. 2007, Legal Ethics, 3rd edn, West Publishing, St Paul, Minn.
64 Shamoo and Resnik, op. cit.; Rappert, 2007, op. cit.
65 NSABB, op. cit.
66 Institute of Medicine 2002, Integrity in Scientific Research: Creating An Environment that Promotes
Responsible Conduct, The National Academies Press, Washington, DC; Geller, G., Boyce, A., Ford, D. E. and
Sugarman, J. 2010, Beyond compliance: the role of institutional culture in promoting research integrity,
Academic Medicine, vol. 85, pp. 1296302.
67 Cho, M. K., Shohara, R., Schissel, A. and Rennie, D. 2000, Policies on faculty conflicts of interest at US
universities, Journal of the American Medical Association, vol. 284, pp. 220328; McCrary, S. V., Anderson,
C. B., Jakovljevic, J., Khan, T., McCullough, L. B., Wray, N. P. and Brody, B. A. 2000, A national survey of
policies on disclosure of conflicts of interest in biomedical research, New England Journal of Medicine, vol.
343, pp. 16216; Lind, R. A. 2005, Evaluating research misconduct policies at major research universities: a
pilot study, Accountability in Research, vol. 12, pp. 24162.
246
public health, society, security or the environment. In theory, institutional

policies could be more effective than professional codes, because institutional
policies are enforceable. Scientists or students who violate university policies
can be suspended, demoted, fired or dismissed. It is not known how many
institutions have developed dual-use policies, but probably very few have, since
universities usually do not develop ethical oversight mechanisms in the absence
of legal or financial threats.69 Government action could spur the development of
institutional policies.
Journal policies
Scientific journal dual-use review policies are another form of self-governance.
Journals have been at the centre of the dual-use controversy, as several papers
published by prominent journals have been controversial. For example, several
members of the US Congress criticised journal editors for publishing the paper
on how to manufacture a polio virus mentioned earlier. The US Department of
Health and Human Services (DHHS) asked the editors of the Proceedings of the
National Academy of Sciences (PNAS) not to publish the paper on contaminating
the US milk supply, also mentioned earlier.70 The editors of PNAS met with
DHHS representatives prior to publication to discuss the benefits and risks of
public dissemination of the findings. The Department of Homeland Security
(DHS) asked the NSABB to review the paper on reconstructing the 1918
pandemic influenza virus prior to publication in Science. Although the NSABB
voted unanimously in favour of publication, the editor of Science publicly
stated he would have ignored the NSABBs recommendations if had been against
publication.71
Recently, the NSABB recommended that journals omit important details from
research that used genetic-engineering techniques to create a mutated form of
an avian flu virus, A(H5N1), which could be transmitted between mammalian
species, including humans. Key details would be available only to responsible
scientists. Currently, the virus can only be transmitted from birds to other
mammalian species, not between members of mammalian species. Six hundred
people have contracted the virus since 1997, and more than half of them have
died. The journals reviewing the research, Science and Nature, had not made a
decision concerning publication as of the writing of this chapter.72
69Ibid.
71 Kennedy, D. 2005, Better never than late, Science, vol. 310, p. 195.
72 Grady, D. and Broad, W. 2011, Seeing terror risk, U.S. asks journals to cut flu study facts, The New York
Times, 20 December 2011, p. A1.
247
Journals began developing dual-use review policies in the early part of the
twenty-first century, following the publication of several controversial papers.73
Some of the leading journals with dual-use review policies include Science,
journals published by the Nature Publishing Group (NPG)that is, Nature,
Nature Biotechnology, and so onand journals published by ASM.74 Review
policies include an additional level of review for papers that raise dual-use
concerns. Outside reviewers with expertise in national security, terrorism or
other relevant subjects may be asked to assist with the review. The outcome
of the review could include a recommendation to publish, to not publish or
to publish with appropriate revisions, such as restrictions on access to key
information.75 The NPGs dual-use policy provides that:
Nature journal editors may seek advice about submitted papers not
only from technical reviewers but also on any aspect of a paper that
raises concerns As in all publishing decisions, the ultimate decision
whether to publish is the responsibility of the editor of the Nature
journal concerned. The threat posed by bioweapons raises the unusual
need to assess the balance of risk and benefit in publication. Editors
are not necessarily well qualified to make such judgements unassisted,
and so we reserve the right to take expert advice in cases where we
believe that concerns may arise. We recognize the widespread view that
openness in science helps to alert society to potential threats and to
defend against them, and we anticipate that only very rarely (if at all)
will the risks be perceived as outweighing the benefits of publishing a
paper that has otherwise been deemed appropriate for a Nature journal.
Nevertheless, we think it appropriate to consider such risks and to have
a formal policy for dealing with them if need arises. The editorial staff of
Nature journals maintains a network of advisers on biosecurity issues.
All concerns on that score, including the commissioning of external
advice, will be shared within an editorial monitoring group consisting
of the Editor-in-Chief of Nature publications, the Executive Editor of the
Nature research journals, the Chief Biological Sciences Editor of Nature,
and the chief editor of the journal concerned. Once a decision has been
reached, authors will be informed if biosecurity advice has informed
that decision.76

74 van Aken and Hunger 2009, op. cit.
75 Report of the NSABB Working Group on Journal Review Policies 2010, NSABB Board Meeting, 19 October
2010,
<http://oba.od.nih.gov/oba/biosecurity/meetings/Oct2010/Journal_Review_Policies_Slides.pdf>
(viewed 30 December 2011).
76 Nature Publishing Group (NPG) 2010, Biosecurity Policy, <http://www.nature.com/authors/editorial_
policies/biosecurity.html> (viewed 30 December 2011).
248
There have been three published studies to date on the dual-use review policies
of scientific journals. In 2009, van Aken and Hunger published a survey of
28 major life-science journals that regularly publish research that may raise
biosecurity issues. They found that 25 per cent of these journals had dual-use
review policies.77 In October 2010, a working group from the NSABB reported
their findings at an NSABB board meeting. Members of the working group
reported on their discussions with the editors of 18 high-impact life-science
journals that had published dual-use review policies online. Members of the
working group found that different models of dual-use review were in place,
and that editors felt they needed additional guidance.78
Though these two studies provide some useful information about the dual-use
policies of scientific journals, they do not provide systematic information about
journal policy development because the sampling was focused (not random) and
the sample sizes were small. To overcome these limitations, my colleagues and
I conducted a larger survey of biomedical journals. We drew a random sample
of 400 journals from the ISI Web of Knowledge Journal Citation Reports 2009
Edition. We eliminated journals that had little chance of reviewing dual-use
research. We sent out an email survey to the editors and reminders after seven
and 14 days if we did not receive a response. Of these, 155 journals responded to
our survey (response rate 39 per cent). Only 7.7 per cent said they had a formal
(written) dual-use policy; 72.8 per cent said they had had no experience with
reviewing dual-use research in the past five years, 5.8 per cent indicated they
had some experience, and 21.8 per cent did not give a definite answer to this
question. We attempted to determine whether several variables were associated
with having a dual-use policy. Belonging to the NPG was the most significant
predictor of having a policy (positive association). Having experience with
reviewing manuscripts with dual-use implications was also positively associated
with having a policy.79
Our research indicates that less than 8 per cent of biomedical journals have a
dual-use policy. This is much lower than the percentage reported by van Aken
and Hunger. This is an important finding, since it shows that the scientific
community has not made much progress in an important area of self-regulation.
Another important finding is that most journals have not had any experience
with reviewing research with dual-use implications. Indeed, some editors said
they had not developed a policy because they saw no need for one, and others
had not even heard of the term dual use. Some editors said they were planning
to develop a policy after they learned about our study.80
77 van Aken and Hunger, op. cit.
78 Report of the NSABB Working Group on Journal Review Policies, op. cit.
79 Resnik, D. B., Barner, D. D. and Dinse, G. E. 2011, Dual use policies of biomedical journals, Biosecurity
80Ibid.
249
Conclusion: Prospects for self-regulation

So what are the prospects for self-regulation of dual-use research? It is difficult
to say at this point, because there has been very little empirical research that
addresses this question. While a survey of life scientists indicates that most
are aware of biosecurity issues, research on biomedical journals indicates that
only a small percentage has developed dual-use review policies. Also, very few
professional associations have developed dual-use policies. We lack systematic
data on other important aspects of self-regulation, such as institutional policies
and oversight, and education and training efforts. Clearly, more research is
needed to better understand how the scientific community is responding to the
dual-use dilemma, and any conclusions drawn at this point about prospects for
self-regulation should be viewed as tentative.
Looking to the past as a predictor of the prospects for self-regulation, we find
that scientists have had some successes and some failures. One of the best
examples of successful self-regulation would be the management of the risks
related to recombinant DNA. In the early 1970s, scientists began to conduct
experiments involving the transfer of DNA to micro-organisms. The public
became gravely concerned about the threat of superbugs escaping from the
laboratory and causing a modern plague. The NIH responded to the publics
anxiety by forming the Recombinant DNA Advisory Committee (RAC) in 1974
to oversee NIH-funded recombinant DNA research. Scientists who were at the
forefront of recombinant DNA research met in Asilomar, California, in February
1975 to discuss the risks of their work and make some recommendations to
protect laboratory workers, the public and the environment from biohazards.
The recommendations included a number of different safety procedures, such as
only using organisms that are unable to survive outside the laboratory, physical
containment protocols and following good laboratory practices. The scientists
also recommended a voluntary moratorium on experiments that were deemed
too risky and said that work should move forward cautiously.81 In 1976, the RAC
published its first set of guidelines for NIH-funded recombinant DNA research.
Among the RACs most important recommendations was that institutions
establish biosafety committees to oversee recombinant DNA research. Although
the RAC guidelines apply only to NIH research, scientists who are not funded
by the NIH have also often followed the RACs recommendations.82 While the
government has played an important role in helping to promote responsible
81 Berg, P., Baltimore, D., Brenner, S., Roblin III, R. O. and Singer, M. F. 1975, Summary statement of the
Asilomar Conference on Recombinant DNA Molecules, Proceedings of the National Academy of Sciences, vol.
72, pp. 19814.
82 Kimmelman, J. 2009, Gene Transfer and the Ethics of First-in-Human Research: Lost in Translation,
Cambridge University Press, Cambridge.
250
recombinant DNA research, scientists have taken it upon themselves to establish

and follow safety standards. Although public anxieties about genetic engineering
persist, the safety record of recombinant DNA research is excellent.83
An example of not-so-successful self-regulation would be research involving
human subjects. Prior to World War II, there were no widely recognised ethical
standards for research involving human subjects. Scientists and physicians
appealed to their own sense of right and wrong in treating human beings
used in research.84 In most cases, research was ethical and humane; however,
there were many abuses involving risky procedures, experimentation without
consent and the use of vulnerable populations, such as children, prisoners
and mentally disabled adults, as research subjects. The worst of these abuses
occurred in Germany under the Nazi regime, when scientists and physicians
used concentration camp prisoners as research subjects. These experiments
often produced severe pain and suffering and often resulted in death. Informed
consent was out of the question.85 After the war ended, German physicians and
scientists were tried for war crimes related to human experimentation. In 1947,
the Nuremberg War Tribunals adopted the Nuremberg Code, the worlds first
ethical guidelines for research with human subjects, to use as a basis for trying
physicians and scientists accused of war crimes.86 While the Nuremberg Code
provided important guidance for researchers, it did not constitute government
oversight because it established an ethical standard, not a legal one. The first
steps towards government oversight of research with human subjects in the
United States occurred in 1966, when the NIH adopted rules for research that
it sponsors. In 1971, the Food and Drug Administration (FDA) adopted human
subjects research rules.87
The US Government started to play a much more prominent role in the oversight
of research with human subjects in the 1970s, when Congress held hearings
on the Tuskegee syphilis study and other ethical abuses in research involving
human subjects. The Tuskegee syphilis study, in which researchers investigated
the natural course of untreated syphilis in black men, was funded by the
Public Health Service and ran from 1932 to 1972. It included 399 patients with
untreated syphilis and 400 controls. The subjects were not told that they were
participating in a research study. They were told only that they were getting
medical treatment for bad blood. The investigators also took steps to prevent
the syphilitic patients from receiving penicillin, an effective treatment, when it
83Ibid.
85Ibid.
86 Nuremberg Code 1947, Directives for Human Experimentation, <http://ohsr.od.nih.gov/guidelines/
nuremberg.html> (viewed 30 December 2011).
251
became available in the 1940s.88 In 1973, Congress responded to the abuses of

human research subjects by passing the National Research Act (NRA), which
established the National Commission for the Protection of Human Subjects in
Biomedical and Behavioral Research to study the ethics of research with human
subjects and authorised federal agencies to draft regulations. The national
commission published its findings, known as the Belmont Report, in 1979,
which provided the conceptual underpinning for a major revision of the federal
research regulations in 1981, which have changed little in the past 30 years.89
Many other governments, following the United States lead, regulate research
involving human subjects. Some states, such as California, have also adopted
their own laws dealing with human experimentation.90
While recombinant DNA research has avoided a great deal of government
regulation, research with human subjects has evolved from self-regulation to
robust government regulation. Why have these two areas of research taken
different trajectories? One reason human subjects research has been highly
regulated is that it directly impacts human rights and wellbeing. When people
die or become gravely ill as a result of an experiment, the public becomes upset
and laws are passed. While the risks of recombinant DNA research to laboratory
workers are concrete and realistic, other risks are abstract and theoretical. The
public may be less apt to react to risks that are yet to be realised. Another reason
human subjects research is highly regulated is that scientists have not done a
good job developing and following their own ethical standards. Laws related to
research with human subjects have been passed in response to highly publicised
scandals when investigators failed to follow ethical rules.91 Not surprisingly,
investigators who work with human subjects often complain that regulations
and rules create needless administrative work that impedes research.92
Will dual-use research become externally regulated or will self-regulation prevail?
Much depends on how scientists respond to opportunities for self-regulation
presented by dual-use dilemmas. If scientists take these issues seriously and
develop professional codes, journal policies, institutional oversight mechanisms,
educational and training programs, and other forms of self-regulation, they may
be able to avoid a high degree of government oversight. If, however, scientists
fail to take appropriate steps towards self-control, a high degree of external
control may result. A tragic event related to dual-use research, such as transfer
of a dangerous pathogen to a terrorist who uses it to develop and deploy a lethal
weapon, could motivate governments to impose stronger controls on dual88 Jones, J. H. 1981, Bad Blood, Free Press, London.
90Ibid.
91Ibid.
92 Gunsalus, C. K. 2004, The nanny state meets the inner lawyer: overregulating while underprotecting
human participants in research, Ethics and Behavior, vol. 14, pp. 36982.
252
use research. Hopefully, scientists will be motivated to take appropriate steps

towards self-regulation before an incident occurs that compels governments to
impose extensive restrictions.
Acknowledgments
This research was sponsored by the National Institute of Environmental Health
Sciences (NIEHS), National Institutes of Health (NIH). It does not represent the
views of the NIEHS, NIH or US Government.
253
16. Contrasting Dual-Use Issues in

Biology and Nuclear Science
Nicholas G. Evans
Dual use and nuclear science

The so-called dual-use dilemmawhich arises when scientific research,
materials or technologies can be used to both benefit and harm humanity1is
not an altogether new phenomenon. Early in the twentieth century, nuclear
sciencehere, the study of the properties of atomic nucleiraised similar
concerns about materials, technologies and knowledge that had the potential to
be used to benefit humanity, but also posed serious risks of misuse. Comparing
the experiences of the nuclear sciences with those of biology in regulating dualuse materials, knowledge and technology will be the focus of this chapter.
Nuclear science raises a number of dual-use issues. Nuclear science has many
beneficial uses, the best known being nuclear-fission (the process by which
a nucleus is split into two or more smaller parts) reactors. These reactors are
useful in power generation, and are also used to create radioisotopeselements
that are subject to radioactive decay that have a variety of uses described below.
Nuclear fusion (the process of fusing two lighter nuclei into a heavier nucleus)
is currently being explored as a cleaner, more plentiful fuel source that could
become an important source of energy in years to come.
Non-fissile radionuclidesradioactive substances not capable of energetic
fissionhave a range of useful and beneficial medical, agricultural,
construction, military and aviation applications, to name a few. Radioisotopes
have therapeutic applications in medicine, as their emissions can be used to
bombard and kill cancerous cells with radiation. Radioisotopes are also used in
diagnostic testing, for example, where the emissions of ingested radioisotopes
are plotted to gain information about physiological processes. In agriculture,
radionuclides are used to create blight or weather-resistant crops, and kill
parasites in food storage. In construction, radioisotopes are used to verify the
1 This corresponds with a common definition of dual use, such as that found in Miller and Selgelid. It,
as others have discussed previously, does not deal with all definitions of the dual-use dilemma, but for the
purposes of this chapter I restrict my analysis to the explicitly normative definition of dual use. See Miller,
S. and Selgelid, M. J. 2008, Ethical and Philosophical Considerations of the Dual-Use Dilemma in the Biological
Sciences, Springer, Dordrecht; Selgelid, M. J. 2010, Ethics engagement of the dual-use dilemma: progress and
potential, in B. Rappert (ed.), Education and Ethics in the Life Sciences, ANU E Press, Canberra, pp. 2334.
255
integrity of metals by detecting the scattering of emitted particles to determine

the internal structure of metal or concrete under load. Depleted uraniuma
byproduct of the enrichment process where the uranium-235 used in nuclear
weapons and reactors is separated from non-fissile uranium-238is incredibly
dense and malleable, making it an ideal form of armor plating in tanks. In
aviation, depleted uranium is used to weight propeller rotors. Depleted uranium
also provides an ideal form of radiation shielding in reactors due to its high
rate of capture of alpha, beta and gamma radiation, as well as neutrons. A layer
of depleted uranium in a reactor can function as reactor shielding, and, as the
depleted uranium captures neutrons and is converted into plutonium, can be
used as reactor fuel.
The study of the atom in nuclear science and its connection to quantum mechanics
and materials science have also led to the development of the materials and
electronics that presently dominate the technological landscape. Many scientific
innovations that are now ubiquitous in contemporary society (for example, the
personal computer) owe their genesis to the pioneers of nuclear physics.
Nuclear science also has well-known harmful uses that pose a risk of being
used by malevolent actors. Nuclear fission is the driving mechanism behind
the atomic bombs dropped on Hiroshima and Nagasaki. Nuclear fusion is
the mechanism that, combined with fission, drives the super or H-bomb.
Contemporary research into either nuclear fission or nuclear fusion can enable
the creation and development of these weapons.
Many radioisotopes are highly toxic and can cause radiation sickness if ingested
by or embedded in a person for an extended period. Radioisotopes packed
around conventional explosives, known as dirty bombs, can cover a large area
in radioactive ash. Survivors of such a blast would have to deal with radioactive
particulates embedded in their bodies as a result of the explosion, infrastructure
affected would need decontamination or have to be destroyed and removed,
and the treatment requirements and potential for contamination in caring for
the injured would increase significantly.2 Radionuclides can cause radiation
poisoning or other health problems if releasedaccidentally or intentionally
into water supplies, or as a gas or powder. The murder of former KGB agent and
defector Alexander Litvinenko, who died of radiation poisoning after ingesting
significant amounts of polonium-210, is an example of malevolent uses of
radioisotopes.3
2 For example, Zimmerman, P. D. and Loeb, C. 2004, Dirty bombs: the threat revisited, Defense Horizons,
vol. 38, pp. 111.
3 Stoll, W. 2007, 210polonium, Atw. Internationale Zeitschrift fr Kernenergie, vol. 52, no. 1, pp. 3941.
256
16. Contrasting Dual-Use Issues in Biology and Nuclear Science
Wither nuclear science?

Despite the dual-use potential of nuclear science, there has been little examination
of the parallels between nuclear science and biology in current debates about
the dual-use dilemma.4 One important reason is that, from the outset, discourse
has neglected or downplayed the relevance of the nuclear sciences to the debate
in the life sciences. In 2004, a canonical report by the National Research Council
chaired by Gerald Fink titled Biotechnology Research in An Age of Terrorism (the
Fink Report) stated:
The nature of the biotechnology problemindeed the nature of the
biological research enterpriseis vastly different from that of theoretical
and applied nuclear physics in the late 1930s. The contrast between what
is a legitimate, perhaps compelling subject for research and what might
justifiably be prohibited or tightly controlled cannot be made a priori,
stated in categorical terms, nor confirmed by remote observation.5
The Fink Report argued that the primary mechanism of arms control in the case
of nuclear weapons, materials control, is not applicable to the life sciences.6
The report highlighted differences between the properties of nuclear fissile
materials and biological pathogens, which it claimed rendered materials control
impractical as a regulatory solution (Table 16.1).
A follow-up report in 2006, Globalization, Biosecurity, and the Future of the
Life Sciences (the Lemon-Relman Report), gave a similar argument to the Fink
Report regarding dual use:
Clear thinking on the issue must proceed from an understanding of
the significant differences among [biological and nuclear] weapons.
Although there are lessons to be learned from the history of and our
experience with nuclear weapons technology, many of the differences
between the nuclear and biological realms are too great to adopt a similar
mix of nonproliferation, deterrence, and defense.7
4 Notable exceptions include Selgelid, M. J. 2007, A tale of two studies: ethics, bioterrorism, and the
censorship of science, The Hastings Centre Report, vol. 37, no. 3, pp. 3548; Finney, J. L. 2007, Dual use:
can we learn from the physicists experience? in B. Rappert (ed.), A Web of Prevention: Biological Weapons,
Life Sciences and the Governance of Research, Earthscan, London, pp. 6776; Forge, J. 2008, The Responsible
Scientist, University of Pittsburg Press, Pittsburg; Miller and Selgelid, op. cit.
5 National Academies of Science 2004, Biotechnology Research in An Age of Terrorism, The National
Academies Press, Washington, DC, p. 23.
6 Ibid., pp. 813.
7 National Academies of Science 2006, Globalization, Biosecurity, and the Future of the Life Sciences, The
National Academies Press, Washington, DC, p. 46.
257
Table 16.1 Characteristics of Fissile Materials and Pathogens

Fissile materials
Biological pathogens
Do not exist in nature
Generally found in nature
Non-living, synthetic
Living, replicative
Difficult and costly to produce
Easy and cheap to produce
Not diverse: plutonium and highly enriched

uranium are the only fissile materials used in
nuclear weapons
Highly diverse: more than 20 pathogens are

suitable for biological warfare
Can be inventoried and tracked in a

quantitative manner
Because pathogens reproduce, inventory

control is unreliable
Can be detected at a distance from the

emission of ionising radiation
Cannot be detected at a distance with

available technologies
Weapons-grade fissile materials are stored at

a limited number of military nuclear sites
Pathogens are present in many types of

facilities and at multiple locations within a
facility
Few non-military applications (such

as research reactors, thermo-electric
generators, and production of radioisotopes)
Many legitimate applications in biomedical

research and the pharmaceutical/
biotechnology industry
Source: National Academies of Science 2004, Biotechnology Research in An Age of Terrorism, The National
Academies Press, Washington, DC.
The above conclusionechoing the Fink Reportindicates that, due to the

marked differences between the materials in question, methods of prevention
used in the nuclear sciences are unsuitable to regulate biological science and
technology. Moreover, the current culture of biology is not one that is conducive
to these regulations.
Yet, I wish to contend, the lack of immediate compatibility between regulatory
measures does not entail that the differences between fields are too great for
anything valuable to be learned from nuclear history when reflecting upon
how dual-use biology should be governed. It is doubtful that the regulatory
framework used in nuclear sciences would provide a good model to use even
if biology and nuclear sciences were more similar. No-one wants another arms
race. That said, even just the lessons learned from the regulatory and policy
mistakes in the nuclear sciences would be usefulbut I hope to show that we
can learn far more than that.
The purpose of this chapter is to compare and contrast regulation of the dual-use
dilemma in the nuclear and biological sciences. This will not be a comprehensive
comparisonthe history of the nuclear sciences from its beginning in the early
twentieth century through to the political implications of nuclear science in the
twenty-first century is too large for a single chapter. Rather, I will provide a
series of examples from the nuclear sciences that invites exploration of different
ethical issues of dual use in the life sciences. I also view this chapter as a call for
258
further investigation of the nuclear sciences and their history to understand the
pitfalls of regulating science and technology, and apply lessons learnt to dual
use in the life sciences.
Finally, it is not my intention to claim that biology and the nuclear sciences
are the same regarding the dual-use dilemma. I will show where the issues
presented by dual-use materials, knowledge, techniques or technologies in
each field overlap, and where they come apart. The differences between nuclear
science and biology do not preclude an analysis of their similarities; likewise,
those differences may be important in and of themselves.
Second, the examples I have chosen serve to highlight some less-explored
features of the dual-use dilemma in the life sciencesfeatures I believe deserve
significant exploration. Though we may continue to worry about the potential
for state-sponsored biological weapons programs, a primary concern in the
dual-use dilemma is that of small groups or individuals committing acts of
bioterrorism. My examples are chosen to identify issues in regulating these
small-scale operations. This does not mean that there is nothing to be learned
from the nuclear sciences about state-level regulations, or that my analysis
doesnt cover these issues at all. Rather, my focus is meant to target the trajectory
I see the (rapidly) evolving life sciences following, in an effort to identify issues
that will arise or are already coming to light regarding dual-use research.
Materials control
Materials control is an important part of preventing the misuse of science
and technology. If access to the constituent materials of biological or nuclear
weapons is restricted, their use can be restricted presumably to good uses by
good and responsible actors. As the subject of the Fink Report that led to their
conclusion that the nuclear sciences cannot inform our thought on the dual-use
dilemma, materials control is a good first point of inquiry for this chapter. The
differences between fissile materials and biological pathogens listed in Table
16.1 are a paradigmatic way of contrasting nuclear science and biology. The
differences between the materials used in both fields present a picture in which
nuclear science is far removed from biology.
There are definitely some pathogens that possess the same type of threat as
nuclear weapons. A smallpox attack could potentially cause up to 167 000
deathsgreater than the number who died when the United States dropped a
nuclear weapon on Hiroshima.8 An anthrax attack on an unprepared population
could cause more than half a million deaths, but only if the emergency response
8 See Samuel, G., Selgelid, M. J. and Kerridge, I. 2010, Back to the future: controlling synthetic life sciences
trade in DNA sequences, Bulletin of the Atomic Scientists, vol. 66, p. 10.
259
to the act was mismanaged.9 The botulinum toxina biological toxin and the
most potent poison in the worldreleased into the milk supply of the United
States could kill up to 200 000 people, the majority children or the elderly.10
Most recently, there are concerns that genetically modified avian influenza
could kill millions.11
Those listed above, however, are some of only a few pathogens capable of the
devastation similar to that visited on Hiroshima. Moreover, comparisons to
Hiroshima are dated and misleading: the W76 warhead, a mainstay of the US
nuclear arsenal, has a yield of 100 kilotons,12 roughly five times that of the
Hiroshima bomb. The W76 bomb is, further, small in yield compared with others
in the US and Russian arsenals. Nuclear weapons range from tactical warheads
of a fraction of a kiloton, to megaton (where one megaton is equivalent to 1000
kilotons) scales.13
Biological attacks, further, suffer from a greater range of contingencies than
nuclear attacks, and are prone to more variable yields than nuclear weapons. Bad
weatheror, in the case of spores that are sensitive to sunlight, good weather
can seriously mitigate the effect of a bioweapons attack, to say nothing of other
factors. Damage caused by biological weapons can be further mitigated by
strong public health policies, smart post-attack vaccinations, and so on.14 This is
contrasted with the indiscriminate and destructive nature of nuclear weapons,
which are more or less impervious to weather conditions or most attempts to
defend against them.
This is not to say that biological pathogens should be treated less seriously than
fissile materials. A key difference listed in the Fink Report is that the destructive
potential of nuclear weapons is balanced to an extent by the difficulty of
procuring fissile materials. Bioweapons, on the other hand, are likely to be more
attainable and more augmented for causing death in years to come. For the
most part, however, the distinction between the (broad) class of things called
9 Wein, L. M., Craft, D. L. and Kaplan, E. H. 2003, Emergency response to an anthrax attack, Proceedings
of the National Academies of Science, vol. 100, no. 7, pp. 434651. It is worth, in the interest of disclosure,
noting that 660 000 people are predicted to die in the modelled attack in the case where only individuals
symptomatic with the disease are given antibiotic treatment. Wein et al. are quite clear in their study that
there are a number of factors that, taken together, could lower this death count to approximately 10 000.
10 Wein, L. M. and Liu, Y. 2005, Analyzing a bioterror attack on the food supply: the case of botulinum
toxin in milk, Proceedings of the National Academies of Science, vol. 102, no. 28, pp. 99849.
11 Enserink, M. 2011, Grudgingly, virologists agree to redact details in sensitive flu papers, Science Insider,
20 December 2011, <http://news.sciencemag.org/scienceinsider/2011/12/grudgingly-virologists-agree-to.
html> (viewed 1 February 2012).
12 Where 1 kiloton = 1000 tonnes of TNT equivalent energy of explosion.
13 Norris, R. S. and Kristensen, H. N. 2007, Estimates of the U.S. Nuclear Weapons Stockpile [2007 and 2012],
<http://www.fas.org/programs/ssp/nukes/publications1/USStockpile2007-2012.pdf> (viewed 24 February
2010).
14 See, for example, Wein et al., 2003, op. cit.; Chen, L.-C. et. al. 2004, Aligning simulation models of
smallpox outbreaks, Intelligence and Security Informatics, vol. 3073, pp. 116; Wein et al., 2005, op. cit.
260
biological pathogens and the (narrow) class of things called fissile materials
is inaccurate; we are working with two classes of materials for which the
comparisons we can make are not strong ones.
Radionuclides
A better comparison, I contend, is between biological pathogens and
radionuclides described above. Though radionuclides more generally are not
dual-use materials of the same degree as fissile materialswhich form a small
but significant set of radionuclidesradionuclides generally pose a series of
dual-use issues, some of which I have described above.
Radionuclides are significant for two reasons. First, much as we think there
are dual-use issues with life-sciences materials, technology and research that
are worthy of attention even though they are not dangerous on the scale of,
say, weaponised smallpox, there are dual-use issues in the nuclear sciences that
do not involve nuclear weapons. Second, the way fissile materials were and
are regulated is, as the Fink Report argued, somewhat unique. It is useful to
examine the regulation of other areas of the nuclear sciences, which may have
experienced issues with regulation more in line with what we would expect
from the life sciences.
Though nowhere near as powerful as a thermonuclear detonation, or the expert
use of weaponised smallpox, dirty bombs would cause damage comparable
with less spectacularand paradigmaticexamples of bioterrorism. First, the
Amerithrax attacks that occurred following the 11 September 2001 attacks on
the World Trade Centre killed five and infected a further 17 people. Second, the
attempted attack by the Aum Shinrikyo cult through the dispersion of anthrax
failed,15 but could have easily inflicted hundreds of casualties had it worked;
the Sarin gas attack in 1995 on the Tokyo subway by the same cult killed 12
and injured thousands.16 Finally, the lacing of salad bars with salmonella by
the Bagwasrinesh cult in Oregon did not kill anyone, but hospitalised 45 and
infected 750.17 These cases are of serious concern, but the number of casualties
(or hospitalisations, in those cases that lacked fatalities) is much smaller than
from a thermonuclear explosion.
Sufficient exposure to radioisotopes can also lead to radiation poisoning. The
most high-profile recent case of this is the death of KGB defector Alexander
Litvinenko from a lethal dose of polonium-210. Polonium-210 is found in
15 See, for example, Miller, J., Broad, W. and Engelberg, S. 2001, Germs: Biological Weapons and Americas
Secret War, Simon & Schuster, New York.
16 Tu, A. T. 1999, Overview of the sarin terrorist attacks in Japan, in A. T. Tu and W. Gaffield (eds), Natural
and Selected Synthetic Toxins: Biological Implications, ACS Symposium Series, 20 December, pp. 30417.
17 Miller et al., op. cit.
261
antistatic brushes used in photography, and in the United States is sold in

quantities large enough to be used to administer a fatal dose to someone. Large
quantities of powdered radionuclides, even those that emit alpha radiation
(which is typically considered harmless as it cannot penetrate the skin),
distributed in a water supply could cause significant harm.
The malevolent use of radionuclides in dirty bombs or as poisons makes many
industrial, medical and commercial technologies dual use. Determined terrorist
groups presently could accrue large amounts of radionuclides through looting
old or abandoned radiotherapy devices.18 In 1987, for example, two men stole
a capsule containing caesium-137 from an abandoned hospital site in Goinia,
Brazil.19 The amount of caesium they stole would cause a significant number
of casualties and economic damage if used in a dirty bomb. It is believed
that a terrorist organisation would be able to easily procure quantities of
radionuclides similar to those in Goina, if not larger amounts, through the
purchase of radioactive refuse or radionuclides from former Soviet states that
once functioned as disposal sites for highly radioactive nuclear waste.20 UN
inspectors have raised concerns that there are still highly problematic quantities
of unguarded, unaccounted for radiological material in the world.21
One might object that the issues involved with securing biological pathogens are
not the same as those with radionuclides. After all, influenza if left absent a host
can die out relatively quickly, whereas caesium-137 will continue to exist and
decay with a half-life of approximately 30 years until, via beta and gamma decay,
it has turned into the stable barium-137. I dont deny that there are differences
in the practicalities of securing different types of dual-use materialbut this
is both within and between fields. What is more interesting are the challenges
presented by attempts to subvert regulations, especially when we consider
materials thatbefitting their classification as dual useare used in public,
commercial or industrial applications.
Acquiring materials piecemeal

Even if existing large sources of radioactive material were secured there is a
possibility that someone could acquire significant nuclear materials to, for
example, create a dirty bomb by acquiring dangerous radionuclides piecemeal.
By this, I mean that dangerous radioactive compounds could be acquired in
18 Zimmerman and Loeb, op. cit.; International Atomic Energy Agency (IAEA) 2002, Inadequate Control of
Worlds Radioactive Sources, <http://www.iaea.org/newscenter/pressreleases/2002/prn0209.shtml> (viewed
24 February 2011).
19 International Atomic Energy Agency (IAEA) 1988, The Radiological Accident in Goinia, <http://wwwpub.iaea.org/mtcd/publications/pdf/pub815_web.pdf> (viewed 24 February 2011).
20 Zimmerman and Loeb, op. cit.
21 IAEA, 2002, op. cit.
262
quantities small enough or clandestine enough to avoid the oversight of a body

like the Nuclear Regulatory Commission (NRC). To illustrate this, consider the
case of David Hahn, who at seventeen years of age, attracted the attention of
the NRC and the Environmental Protection Agency for acquiring dangerous
amounts of radium, americium and thorium, when he irradiated his mothers
potting shed in an attempt to create a nuclear breeder reactor.22 Though radium
and americium are both problematic, as they could be used in a dirty bomb,
Davids acquisition of significant quantities of thorium is especially worrisome,
as thorium-232 can be converted into uranium-233, a fissile material.
Hahn failed in his attempts to create a reactor (breeders require liquid sodium as
a heat-transfer mechanism and are notoriously difficult to create even in ideal,
well-financed conditions). What he did succeed in was creating a highly active
neutron source. Hahns neutron source was made by encasing americium that
he had extracted from smoke detectors (and, later, radium extracted from the
paint used in antique clocks) in lead with a pinhole in it, and using a strip
of beryllium to convert the alpha radiation from the americium/radium into
neutrons. Hahn targeted his neutron source at a grid of thorium-232 bricks he
had purified from the ash of the mantles of camping lanterns. His purification
method was relatively sophisticated considering his resources; he heated bars
of lithium sourced from lithium-ion batteries surrounded by the thorium ash,
causing the thorium to reduce and lithium to oxidise, leaving lithium oxide ash
and pure thorium.23
Hahns story is important because it shows how little training and money
are needed to extract radioisotopes from commonplace technologies. Smoke
detectors, camping lanterns and antique clocks do not typically fall into the
purview of dual-use technologies, and yet they were sufficient for a seventeenyear-old to acquire thorium in quantity and purity that he (had he bought
the thorium) would have violated materials control regulations of the NRC;
his experiment caused enough of a hazard to warrant a Superfund hazardous
materials clean-up.
The case of David Hahn raises a problem of acquisition of materials from existing
technologies and natural sources. Piecemeal acquisition of dual-use materials
in the life sciences has already received some attention.24 There are common
and naturally occurring viruses that could conceivably function as a natural
template for a modified, virulent pathogenic organism. For instance, a potential
terrorist could attempt to synthesise genetically modified influenzaa presently
22 See Silverberg, R. 2004, The Radioactive Boy Scout: The Frightening True Story of a Whiz Kid and his
Homemade Nuclear Reactor, Random House, New York.
23 Ibid.
24 Samuel et al., op. cit. See also National Academies of Science 2010, Sequence Based Classification of Select
Agents: A Brighter Line, The National Academies Press, Washington, DC.
263
technically difficult propositionor they could wait until flu season to gain
living native samples and modify them. Anthrax, to use another example, is
still a common disease amongst cattle in developing countries.
The advent of do-it-yourself (DIY) or garage biology, where individuals
practise biology at home or in private hobby groups, further undermines the
efficacy of materials control. As the cost of sequencing and synthesis continues
to decline, and the methods of DIY biologists in procuring samples either
individually or through communal open labs (biological commons) or creating
their own technologies become increasingly sophisticated,25 materials control
will become progressively more costly, or even completely ineffective. Current
regulatory approaches should take this into account when attempting to design
overlapping regulatory measuresthe so-called web of prevention26to
prevent the misuse of biotechnology.
Moreover, if present expectations for biology are fulfilled, biological materials
stand to become a ubiquitous part of modern lifestyles not in the way of nuclear
materials, but in the way of microchips.27 Many of these biological materials will
not pose a significant dual-use threat, but there is a good chance that many of
them will. Understanding how biological materials will appear in the everyday
world and how motivated individuals might use them for malevolent or just
manifestly reckless purposes is an important step in understanding dual use
beyond the paradigm of state-level weapons conventions, deterrence and a fieldwide security enterprisea paradigm that is already recognised as untenable or
insufficient for the dual-use threat in the life sciences.
Restricting access to technology

Another regulatory option is to restrict access to dual-use technologies.
Biological technologies such as modern DNA sequencers have serious dual-use
potential. DNA sequencers are able to sequence code in aid of doing beneficial
research or creating non-pathogenic varieties of viruses for study or medicine;
however, they may also find use in developing pathogens for use in harming
othersfor example, in manufacturing novel or existing pathogens to use in
biological weapons.
25 See Carlson, R. 2010, Biology is Technology: The Promise, Peril, and New Business of Engineering Life,
Harvard University Press, Cambridge, Mass.
26 Feakes, D., Rappert, B. and McLeish, C. 2007, Introduction: a web of prevention? in B. Rappert and C.
McLeish (eds), A Web of Prevention: Biological Weapons, Life Sciences and the Future Governance of Research,
Earthscan, London, pp. 114.
27 See Carlson, op. cit.
264
Restricting access to technologies, much like materials control, is no less sure a

strategy than materials control against talented, curious or motivated individuals.
Famously, in 1976, Dr Abdul Qadeer Khan stole the plans for a centrifuge from
his workplace at Urenco in the Netherlands. This theft, and Khans later work,
would ultimately mark the genesis of the Pakistani atomic weapons project.28
In todays communicative environment, there are many more opportunities for
determined individuals like Khan to enable the proliferation of technologies.
The ubiquity of portable electronic storage devices and the potential of online
so-called cloud storage today make information containment very difficult.
In weighing the benefits created by restricting technologies against the costs
of such policies, we must also account for the possibility that determined
individuals, as in the case of David Hahn, will simply adopt indigenous or
DIY methods to achieve their goals. This is already an emerging phenomenon
in biology; garage biologists have created DIY and low-cost solutions to
expensive technologies in biology, such as Lava Amp, a low-cost, USB-powered
PCR thermocycler.29 For those devices that cant be made at a low cost, secondhand laboratory equipment is now available online. The sophistication and
availability of low-cost or DIY biological laboratory tools are only likely to
increase in the near future.
The creativity and drive to create lower-cost, more accessible tools in biology
are surely things to be valued, for their own sake and for the gains we could
have with a larger and more accessible research culture. This rapid lowering of
access barriers to technologies, however, comes at a cost to security. That is, as
biological technologies become ubiquitous, indigenous approaches to dual-use
technologies will emerge as much from small enterprises as from large projects.
While in the nuclear sciences, it required a state to manufacture indigenous
technologies where restrictions on dual-use technology were already in place,
individuals may conceivably be able to manufacture their own technologies.
These gains through low-cost, accessible technologies are surely not the only
things we value. These gains must still be weighed against other things we value,
such as our collective wellbeing, how that collective wellbeing is distributed
(that is, equality), if those who are harmed as a result had the capacity to
understand the risks of the proliferation of these technologies and consent to
their proliferation (that is, fairness), and how tightly we think these matters
should be regulated in the public interest (that is, liberty). Running roughshod
over these complex ethical issues reduces us to a simple appeal to utility, and
28 See Corera, G. and Myers, J. J. 2006, Shopping for Bombs: Nuclear Proliferation, Global Insecurity, and the
Rise and Fall of A. Q. Khans Nuclear Network, Oxford University Press, London.
29 Lava Amp, <http://www.lava-amp.com/> (viewed 27 February 2011).
265
a narrow sense of utility at that. Most of us, reflecting on the nuclear sciences,
would admit that the nuclear sciences, even if they have brought gains to our
lives, have brought gains unequally and at times unfairly.
Restriction of information
At times, it may be that the knowledge produced by a particular piece of
research is dual use. The polio virus case, for example, functioned as a proof
of principle that viruses could be synthesised de novo. This is independent of
which virusor strain of a particular virusis to be synthesised. It means that
vaccines could be constructed rapidly, but also that virulent pathogens could
be made. Even then, this could be good, enabling more individuals to study
a virus with an eye to creating medicines against it; however, the knowledge
could be also used to harm others. Jeronimo Cello and Eckard Wimmer, two
of the researchers who worked on the polio virus synthesis, noted that one of
the important outcomes of the research was confirming that viruses, in fact,
behave in a chemical fashion and can be synthesised like chemicals.30 It is
this knowledge, we might think, which is dangerous as much as any physical
materials used in the research.
One option is to censor information. We could restrict what projects are
undertaken, or we could restricteither temporarily or permanently, depending
on the seriousness of the dual-use concernthe publication of research. For
example, a 2005 study in the Proceedings of the National Academies of Science
modelling a terrorist attack in which botulinum toxin was released in the US
milk supply was embargoed for approximately a month as the journal, the
National Academy of Sciences and government officials debated the security
concerns in releasing such a report.31 In 2011, the National Science Advisory
Board for Biosecurity (NSABB) recommended the censorship of two studies
in which highly pathogenic H5N1 avian influenza virus was modified to be
transmissible between ferrets (close immunological analogues to humans),
though this recommendation was later reversed.32
Regulating the production of such knowledge by restricting its creation or its
dissemination is typically met with suspicion or outrage. Denying scientists
30 Selgelid, M. J. and Weir, L. 2010, Reflections on the synthetic production of poliovirus, Bulletin of the
Atomic Scientists, vol. 66, no. 3, pp. 19.
31 Alberts, B. 2005, Modeling attacks on the food supply, Proceedings of the National Academy of Sciences,
vol. 102, no. 29, pp. 97378. See also Wein, L. M. 2008, Homeland security: from mathematical models
to policy implementation: the 2008 Philip McCord Morse lecture, Operations Research, vol. 57, no. 4, pp.
80111.
32 Evans, N. G. 2013, Great expectationsethics, avian flu and the value of progress, Journal of Medical
266
their right to freedom of speech and their right to freedom of intellectual

inquiry is seen as depriving them of basic rights. Moreover, there is concern
that censorship or restriction of knowledge will not only deprive scientists of
their rights, but also prevent beneficial research from occurring. Until the avian
flu recommendation, scientific self-censorship was considered the norm for
regulation of dual-use information.33
Though an important consideration, it would be a mistake to take self-governance
on its own as a feasible regulatory strategy; no field demonstrates this more
aptly than the nuclear sciences. For example, in 1939, Enrico Fermi, working
with Szilard, showed that a nuclear chain reaction was experimentally possible.
Szilard pressed Fermi at Columbia University and Frederic Joliot in Paris to
keep their results secret. Fermi agreed hesitantly to comply on the condition
that others followed suit,34 but Joliots team rejected Szilards proposal, Joliot
stating that if he withheld publication an implication would be that Hitler
succeeded in destroying another precious liberty.35
Following publication of this information in May 1939, a September meeting
of the German military began an investigation into the plausibility of a chain
reaction leading to weapons36 and banned exports of uranium from Germancontrolled Czechoslovakia, effectively beginning the arms race for the bomb.37
This is a prime example of why relying on self-imposed censorship of results
the kind mentioned in contemporary reports on the dual-use dilemma38
is problematic. When regulatory measures are self-imposed without a way
of binding members of a community to a decision, single individuals can
undermine the greater projects of a community without good justification.
Most of us would, I think, say that Joliot was unjustified in his assessment of
the situationsurely if Hitler had succeeded in creating atomic weapons, more
than just Joliots liberty to publish would have been at risk.
Admitting that scientific self-censorship is not enough, there is genuine
concern about permitting government control over scientific information.
Allowing governments to control scientific information can be both ineffective
and prone to abuse. For example, in 1979, the Department of Energy and the
US Government attempted to censor the Progressive magazine and suppress
information relating to the construction of the hydrogen bomb. This censorship
was attempted under the auspices of the born secret doctrine, a policy of
33 National Academies of Science, 2004, op. cit., p. 101.
34 Rhodes, R. 1995, The Making of the Atomic Bomb, Simon & Schuster, New York, p. 295.
35 Goldsmith, M. 1976, Frdric Joliot-Curie: A Biography, Lawrence & Wishart, London, p. 74.
36 Powers, T. 2000, Heisenbergs War: The Secret History of the Atomic Bomb, Penguin Books, London, p. 65.
37 Weart, S. 1976, Scientists with a secret, Physics Today, vol. 29, no. 2, p. 28.
38 For example, National Academies of Science, 2004, op. cit. See recommendation three of the Fink Report.
See also recommendations three and four of National Academies of Science, 2006, op. cit.
267
secrecy in the Atomic Energy Act of 1954, restricting: 1) the design, manufacture
or utilisation of atomic weapons; 2) the production of special nuclear material
(uranium-233 or 235 or plutonium in natural or enriched states); and 3) the use
of special nuclear material in the production of energy.39
The article of concern, written by Howard Morland, was based on information
in the public domain.40 Morland had not, in writing his article, used any
information that was secret. Much was still considered born classified, but had
existed in the public domain long enough that it was more or less commonly
accessible knowledge. The information of concern was the description of
radiation implosion, the technical process that allowed the superbetter
known as the hydrogen bombto operate.41
Morlands stated intentions in releasing the article were to promote discussion
about the arms race and the extensive secrecy of the nuclear complex.42 In this
sense, the knowledge was used in a manner that many would consider to be
good. Nuclear secrecy, while justified in certain areas, has been criticised for
subverting discourse on arms reductions and control of nuclear weapons.43
The US Governments reasons for attempting to censor Morlands article were
that public knowledge was not complete. Potential builders needed more than
just what was publicly availableknowledge that Morlands article, they
claimed, provided. The court, however, rejected this claim, stating that any
deduction that Morland had made regarding the technical mechanisms of the
hydrogen bomb were clearly independently deducible: four countries already
had independently created hydrogen bombs.
Lessons for biology

The lesson of nuclear science is that restrictions on information, technology
or materials are not, as is commonly described, tensions between freedom and
control. Rather, the tension is all about controlnamely, whose control should
take priority. The above cases highlight the problems that arise when individual
scientists or governments have unilateral control over the dissemination
of information. Despite the increasing proliferation of information and
communications technologies, we should resist the idea that individuals having
39 US Congress, The Atomic Energy Act of 1954, ss. 115.

40 Knoll, E. 1979, The secret revealed, The Progressive, vol. 43, pp. 12; Morland, H. 2005, Born secret,
Cardozo Law Review, vol. 26, p. 1401.
41 Knoll, op. cit.; Morland, op. cit.
42 See Symposium on the progressive case, Cardozo Law Review, vol. 26, no. 4 (Special Issue 2005).
43 See, for example, Rotblat, J. (ed.) 1982, Scientists, the Arms Race, and Disarmament: A Pugwash/
UNESCO Symposium, Taylor & Francis, London; Masco, J. 2002, Lie detectors: on secrets and hypersecurity
in Los Alamos, Public Culture, vol. 14, no. 3, p. 441.
268
unilateral control over the dissemination of scientific information is always

better than allowing government control over some scientific information; and
even if it is, that individual control is the best strategy all things considered.
Rather, we should aim to moderate between these two positions. What the
balance between these two extremes should be is a live issue. Moreover, it
is an ethical issue, not just one of efficiency or technical capacity to control
information (though it surely is that as well). Yet just as we would deny the
desirability of subjecting science to centralised, complete government control,
we can deny the desirability of individuals making unilateral decisions about
scientific information.
Further, we can broaden our view to include the views and concerns of citizens.
Much of the debate around dual use stems from dual use beingby definition
beneficial to humanity. Consulting humanity on what constitutes beneficial
would presumably be a step in the right direction. Morlands contention was that
lacking in information due to the secrecy of the nuclear complex, the public was
not able to engage with decision-making about the role of the nuclear sciences in
public life. This, then, leads us to discussions about education.
Education
Education is advocated as a worthwhile strategy to deal with the dual-use
dilemma.44 Typically, the proposed venues for ethics education are institutions
of higher learning, where professional or apprentice practitioners can be trained
in the ethical implications of their discipline. The aim of this education, then, is
to engage bench scientists or students of science in the ethical issues associated
with their research.
There are, however, other types of education that are crucial to mitigating the
harms of the dual-use dilemma, as demonstrated by the nuclear sciences. These
measures are associated with the education of the public in both ethics and
scientific literacy. Recall that Morlands contention in the Progressive case was
that public engagement and reasonable knowledge of the technical as well as the
political aspects of the regulatory debates and structures in the nuclear sciences
were necessary to functioning scientific policy.
Moreover, ordinary public citizens may at times be the only witnesses to
problematic acts that are not the purview of conventional regulations. Without
appropriate education, ordinary public citizens are not well placed to know what
44 See, for example, Rappert, op. cit.; National Science Advisory Board for Biosecurity (NSABB), Strategic
Plan for Outreach and Education on Dual Use Research Issues, National Science Advisory Board for Biosecurity,
Bethesda, Md.
269
to look for or, sometimes, even what they are looking at. One of the observations
that came from the David Hahn case is that, had they been suitably attentive,
Davids parents, teachers and classmates should have known something was
amiss. Particularly at the point at which David began showing up with burns
caused by exposure radiation, the public who interacted with David, we might
think, should have become concerned enough to notify someone.45
Literature on dual use in the life sciences has discussed the meritsalbeit to a
lesser degree than the education of lab scientistsof public outreach.46 Such
outreach typically aims to create a culture of acceptance of novel technologies.
The Biotechnology Industry Organization (BIO), for example, notes that 31 per
cent of Americans surveyed by the Robert Wood Johnson Foundation are at
least somewhat concerned about becoming ill from a bioterror attack.47 The aim
of public outreach is often to calm fears about biotechnology, often by showing
how the emerging science will contribute to security.48
Another reason to pursue extensive public education is that as the life sciences
become ever more ubiquitous in the technological and social landscapes, the
publics comprehension may become an important part of security. In the life
sciences, this is particularly important as garage or DIY biology comes into
its own. DIY biologists are biologists who pursue biological research from their
homes, usually as a hobby or as part of an entrepreneurial activity. While DIY
biologists can create communities that have strong commitments to safety and
security, such as the organisation DIYBio,49 this is not a necessarymuch less
mandatorypart of being a DIY biologist. It remains to be seen how such a
community can deal with members who would actively seek to cause harm, or
perform reckless or dangerous experiments.
It may be that in some cases individuals engaged in DIY biology who pursue
reckless or malevolent acts are only subject to observation by their peers,
friends and family. In these cases, oversight is only plausible if those around an
individual engaged in reckless or malevolent activity are suitably knowledgeable
to recognise that something is amiss. If the public is not educated, activity done
outside official laboratories may be overlooked with disastrous consequences.
Moreover, lack of public education may lead to powerful stakeholders dominating
regulatory discourse in ways that are not conducive to the promotion of the good
45 See Silverberg, op. cit., ch. 8 and Epilogue.
46 For example, Presidential Commission for the Study of Bioethical Issues 2010, New Directions: The Ethics
of Synthetic Biology and Emerging Technologies, Washington, DC.
47 Biotechnology Industry Organization (BIO) 2010, Healing, Fueling, Feeding: How Biotechnology is
Enriching Your Life, Biotechnology Industry Organization, Washington, DC, pp. 2324, 29 n. 63.
48 Carlson, op. cit.; Carlson, R. 2003, The pace and proliferation of biological technologies, Biosecurity and
Bioterrorism: Biodefense Strategy, Practice, and Science, vol. 1, no. 3, pp. 20314.
49 DIYBio, <http://diybio.org> (viewed 27 February 2011).
270
of society as a whole. A longstanding criticism of the nuclear sciences is that

regulatory debates were dominated by militant voices.50 In a similar fashion,
much has been made recently of the potential for biology to create a new form
of bio-economy.51 Without input from a range of actors, we might see money
come to dominate the debate in the life sciences, much as the military came to
dominate the debate in the nuclear sciences.
Conclusion: From ethical comparison to ethical

practice
In this chapter, I have presented a number of cases that inform our thinking on
issues created by, and issues that arise in dealing with, the dual-use dilemma.
This, as I noted at the beginning of this chapter, is not a comprehensive account
of the history of the nuclear sciences that might apply to the biological sciences.
It is a snapshot to demonstrate the issues that can emerge from a comparison
between the two, and lessons learnt about the success or failure of various
attempts at regulation.
The above cases, I propose, provide four points that ought to guide our ethical
practice
1. regulation of objects or information provides a barrier to access, but cannot
be future-proof
2. the genius of the determined should not be underestimatedmotivated and
gifted individuals and groups will work around technical or political hurdles
3. restriction of information should seek to moderate between the risks of
government and of self-imposed controls on information
4. education must focus on public education as well as the education of scientists.
These recommendations are not particularly controversial, but nor should they
be. The history of regulating the malevolent use of science and technology is
one that is as old as science and technology. In addition to any suggestions that
directly bear on the types of regulation we should adopt, nuclear science serves
as a reminder of the pitfalls regulation can succumb to.
A common theme in these points is thatas the nuclear sciences learned
the hard waytraditional regulatory arguments that pit absence of formal
50 Rotblat, op. cit.; Chan, A. H. 1993, Team B: the trillion-dollar experiment, Bulletin of the Atomic
Scientists, vol. 49, no. 3, pp. 247, <http://www.channelingreality.com/NWO_WTO/Team_B_The_trilliondollar_experiment.pdf> (27 February 2011).
51 Carlson, 2010, op. cit.
271
regulation against absolute government control are deeply problematic. Nuclear

science, I believe, tells us that all levels of stakeholder need to be engaged,
most importantly the public. Public citizens are the end users and beneficiaries
of biotechnology, and are likely to be victims of the accidents and misdeeds
that motivate biosecurity debates. As biology becomes increasingly corporate
and ubiquitous amongst private practitionersthe aforementioned garage
biologiststhe public not only has the most to gain or lose from biology, but
will also become witness to the misuse or abuse of biology and biotechnology.
It is therefore imperative that regulation and biosecurity take into account the
role the public will play.
This public education is surely to the benefit of everyone. For scientists, it
presents an ever more educated public who values science and the benefits it
brings. For a government, it presents a well-educated public who is better able
to contribute to public life, and better able to understand the consequences
of the misuse of biology, and react accordingly (for example, by supporting
appropriate policy).
A better-educated public also provides the grounding for what I will refer to
as open security. As the pace of technological change increases, maintaining
security will increasingly rely on surveillance, rather than attempting to
build unwieldy regulations. In a world where the average personal computer
can run complex simulations of nuclear weapons or epidemiological analysis;
where desktop biological synthesis devices are not merely speculative but
being discussed as real technologies;52 where individuals are endlessly and
perpetually connected, debates about security must honestly weigh the tradeoffs and costs of restrictions to materials and information. As I have discussed
often in this chapter, restrictions are not future-proof, and are only valuable
when their benefits and costs are weighed against other important social goods.
Increasingly, the cost of even small amounts of justified control53 is prohibitive.
Life will overtake security if security does not become adaptive.
Finally, it is important to take a broader view of my analysis. First, I have argued
that a number of the usual suspects in regulation are problematic, but not
necessarily regulatory methods particular to nuclear science. Current discourse
in the life sciences, from the Fink Report on, still treats materials control,
technology control, information control, education and so on as primary methods
of controlling the dual-use threat. But these methods were not foolproof in the
nuclear sciences; today they are almost invitations to practitioners in the life
52 Carlson, R. 2007, Laying the foundations for a bio-economy, Systems and Synthetic Biology, vol. 1, no.
3, pp. 10917, <http://www.ncbi.nlm.nih.gov/pubmed/19003445>.
53 As opposed to unjustified amounts of control, which we might think is easier to affect, but costs a
population much more.
272
sciences to innovate around regulation. The nuclear sciences, in their failure to

regulate the spread of dual-use technologies, should serve as a point of inquiry
to figure out what doesnt work, and why.
Second, we should think beyond the nuclear sciences and begin drawing
lessons of regulation from other technological cultures. If the genetic age stands
to replace the information age, we should look at lessons learnt in the past 10
years of online culture to determine how regulation succeeded and failed to
address the malevolent use of online technologies. The more information we can
draw on, the more adaptive regulatory discourse can be.
Nuclear science is a broad and complex field with a long history. How much we
can learn from it is as much a matter of how much care we take to investigate
this history and apply lessons learnt to our current thinking and practice. As
the debate over dual use in the life sciences is one in which we wish to prevent
catastrophe, it seems prudent to take any source of guidance we can. Of those
experiences in regulating science, none should be more important than the
hard-earned lessons of nuclear science.
273
17. Considering Contextuality in

Dual-Use Discussions:
Is There a Problem?
Louise Bezuidenhout
Introduction
As dual-use ethics continues to grow as a topic of discussion, a number of
features are increasingly becoming identifiable in the discourse. While many
of these have been well discussed in a number of other volumes,1 this chapter
focuses on a little-examined characteristic: how issues relating to contextuality
in life-science research are currently addressed in dual-use ethics.
The issue of contextuality in dual-use ethics is an interesting topic for
consideration because it may be simultaneously argued that there is too much
focus as well as too little. Those suggesting that dual-use ethics has been
predominantly context driven will point to the central role that the web
of prevention rhetoric and policy development have had in the evolving
discussions on responsibility. And they would not be wrong; indeed, much
of the discourse in dual-use ethics has evolved out of control and regulatory
discussions and continues to be strongly influenced by them.
The alternativethat current approaches to dual-use ethics are largely decontextualisedis more difficult to defend (and ultimately much less popular).
As this chapter will elaborate, however, examining the contextual oversights in
dual-use ethics raises some extremely important considerations. In particular,
these considerations shed light on how and why scientists in developing
countries remain marginalised from dual-use ethics discourse and are unlikely
to gain more prominence if current approaches are continued.
In order to elaborate on this position, the chapter will start by briefly examining
dual-use ethics in light of the latter position. It will then go on to highlight
how assumptions made through this position may impact on scientists in nonWestern countries. These issues are supported by fieldwork observations from
1 Such as National Research Council 2011, Challenges and Opportunities for Education about Dual-Use Issues
in the Life Sciences, The National Academies Press, Washington, DC.
275
dual-use ethics research conducted in Africa. The chapter will conclude by

highlighting some areas that need to be further examined if dual-use ethics is to
become a globally debated topic in the scientific community.
Contextuality in dual-use ethics

Dual-use ethics is an expanding field of study following the recent widespread
endorsement of increasing capacity in ethics education for the life sciences
particularly from the West. Since 2001 international bodies such as the UN Policy
Working Group on the United Nations and Terrorism, national organisations
like the British Medical Association, and international agencies including the
International Committee of the Red Cross, have made calls for the enhanced
education of scientists, administrators, physicians and others about the potential
for destructive application of the biomedical and medical sciences.2
Importantly, many of these calls have emphasised the need for promoting
ethical decision-making, which serves to indicate a responsibility for scientists
beyond legal and regulatory compliance.3 In Globalization, Biosecurity, and the
Future of the Life Sciences,4 the report commissioned by the National Research
Council of the United States, it was argued that it was prudent to establish
a decentralized, globally distributed, network of informed and concerned
scientists who have the capacity to recognize when knowledge or technology
is being used inappropriately or with the intent to cause harm.5 This implies
that the familiarisation of scientists with the legal requirements of bio-risk
management is the starting point for ethics education, rather than an end in itself,
and that the development of a sense of individual responsibility is desirable
within the science community.6 Indeed, as the National Science Advisory Board
2 Rappert, B. and Davidson, E. M. 2008, Improving oversight: development of an educational module on

dual-use research in the west, in Uganda National Academy of Sciences Promoting Biosafety and Biosecurity
within the Life Sciences: An International Workshop in East Africa, Uganda National Academy of Sciences,
Kampala, p. 127.
3 The extent of individual responsibility of scientists within dual-use dilemmas is a subject of considerable
debate, but will not be addressed here due to space limitations. It is well reasoned in Miller, S. and Selgelid, M.
J. 2007, Ethical and philosophical considerations of the dual-use dilemma in the biological sciences, Science
and Engineering Ethics, vol. 13, pp. 52380.
4 Committee on Advances in Technology and the Prevention of their Application to Next Generation
Biowarfare Threats 2006, Globalization, Biosecurity, and the Future of the Life Sciences, National Research
Council, Washington, DC.
5Ibid.
6 Uganda National Academy of Sciences 2008, Promoting Biosafety and Biosecurity within the Life Sciences:
An International Workshop in East Africa, Uganda National Academy of Sciences, Kampala, p. 6. This is in line
with Article IV of the Biological and Toxin Weapons Convention.
276
17. Considering Contextuality in Dual-Use Discussions: Is There a Problem?
for Biosecurity (NSABB) stated: an enhanced culture of awareness is essential

to an effective system of oversight and is a critical step in scientists taking
responsibility for the dual-use potential of their work.7
Current approaches to dual-use ethics discourse and

pedagogy
Current approaches to dual-use ethics have therefore focused predominantly on
identifying and understanding the responsibilities that scientists have towards
ameliorating the dual-use potential of their research. These discussions of
responsibility have been heavily influenced by the development of the web
of prevention model that promotes multiple stakeholders in biosecurity. This
approach dates to the early 2000s when the International Committee of the Red
Cross (ICRC) launched an initiative on Biotechnology, Weapons and Humanity,
calling for the reaffirmation of norms against biological weapons and better
controls on potentially dangerous biotechnology.8 This web of prevention
emphasised the crucial need for the involvement of security, health and judicial
communities in addressing the dual-use issue.9
Crucially, the web of prevention concept has built on existing biosafety
and biosecurity initiatives to include security, law-enforcement and lifescience organisations, and the coordination of international oversight. In one
formulation it was suggested that any web of prevention include (and/or
improve) initiatives such as
export controls
disease detection and prevention
effective threat intelligence
biosafety and biosecurity initiatives
international and national prohibitions
oversight of research
education and codes of conduct.10
Dual-use ethics has played an important role in shaping understanding of the
notions of both distributed responsibility and partial responsibility that the
7 National Science Advisory Board for Biosecurity (NSABB) 2007, Report of the NSABB Working Group on
Oversight Framework Development, National Science Advisory Board for Biosecurity, Bethesda, Md.
8 Feakes, D., Rappert, B. and McLeish, C. 2007, Introduction: a web of prevention? in B. Rappert and C.
McLeish (eds), A Web of Prevention: Biological Weapons, Life Sciences and the Future Governance of Research,
Earthscan, London, pp. 114.
9 Ibid.; and International Committee of the Red Cross (ICRC) 2003, Biotechnology, Weapons and Humanity,
International Committee of the Red Cross, Geneva.
10<http://www.brad.ac.uk/bioethics/EducationalModuleResource/EnglishLanguageVersionofEMR/>
(viewed 25 August 2013), see lecture 2.
277
web of prevention model engenders for scientists. When considering the notion
of partial responsibility, it is important to note that dual-use ethics has widely
endorsed the idea that although (t)he misapplication of peacefully intended
research may cause moral distress among scientists it is difficult to argue
that researchers should (solely) be held morally accountable for harm caused
by unforeseen acts of misuse.11 Indeed, scientists are usually suggested only to
have a limited amount of responsibility regarding the reuse of their data, and
bioterrorist activities are thought of as beyond the responsibility of most life
scientists either to prevent or to respond to.12
This notion of partial responsibility is intimately connected to the idea of
distributed responsibility. As suggested by Ehni,13 only a mixed authority
which is constituted by the scientific community together with governmental
bodies, but with the participation of scientists meeting their responsibilities so
far as possible, can solve the problem. Indeed, the web of prevention model
engages a wide range of stakeholders who bear some responsibility towards
addressing and controlling the dual-use potential of the life sciences, including
security, health and judicial communities. The web of prevention model has been
very influential in structuring discussions regarding dual-use controls and most
commonly includes a number of different areas of interventions including public
health initiatives, security surveillance, biosafety and biosecurity controls, and
the education of scientists and the development of codes of conduct.14
The web of prevention model has thus been influential in promoting the idea
that scientists bear only partial responsibility for dual-use issues, and that they
cannot be expected to address the dual-use potential of their research alone.
While there has been general agreement on this, there remains considerable
discussion on how this idea of a partial responsibility may be understood.
In recent years, there have been a number of attempts to determine lists of
(conditional) duties for scientists that will clearly elucidate the expectations
that they have towards dual-use concerns. These, as promoted by Kuhlau et al.,15
may be summarised as follows
the duty to prevent bioterrorism
the duty to engage in response activities
the duty to consider the negative implications of their work
11 Kuhlau, F., Eriksson, S., Evers, K. and Hoglund, A. T. 2008, Taking due care: moral obligations in dual
use research, Bioethics, vol. 22, no. 9, pp. 47787 at p. 483.
12 Ibid., p. 477.
13 Ehni, H.-J. 2008, Dual use and the ethical responsibility of scientists, Archivum Immunologiae Et
Therapiae Experimentalis, vol. 56, pp. 14752 at p. 151.
14 As discussed in Rappert and McLeish, op. cit.
15 Kuhlau et al., op. cit., pp. 4836.
278
the duty not to publish or share sensitive information

the duty to oversee or limit access to dangerous materials
the duty to report activities of concern.16
These duties (and similar ones, such as Ehnis)17 are predominantly presented
as deontological duties and thus come with two important characteristics.
First, these duties are usually presented as globally applicable (or at least, any
discussion to the contrary is absent) and may be applied in any laboratory
context around the world. Second, by virtue of being deontological, these duties
are presented as moral obligations for scientists. Thus, there is the (explicit or
implicit) understanding that the failure to fulfil them has ethical import.
This duty approach has made a considerable impression on current dual-use
ethics, and formulations similar to the one presented above are often used in
ethics pedagogy to inform discussion on responsibility and expectations.18 This
presents an interesting contrast in current dual-use educational initiatives,
and for ethics discourse in general. Although there is considerable discussion
about the interplay between the (contextual) web of prevention and the
ethical responsibilities of scientists, and although there is extensive debate on
contextually suitable styles of pedagogy,19 dual-use responsibilities continue
to be presented to scientists as a list of globally applicable duties with little
discussion about how they are applied in a contextual fashion. The next section
considers this idiosyncrasy in further detail.
Web of prevention aside

Of course these duties are an excellent means of presenting dual-use
responsibility discourse to scientists, and should not be viewed otherwise;
however, that is not to say that this approach is not without its problems. In
order to understand what these problems are, it is important to go back to these
duties and examine them properly in light of the broader web of prevention.
In particular, it becomes crucial to ask: what expectations about the research
environment do these duties make?
16 Ehni proposed similar duties (Ehni, op. cit., p. 150): not to carry out a certain type of research;
systematically to anticipate dual-use applications in order to warn of dangers generated by them; to inform
public authorities about such dangers; not to disseminate results publicly, but keep dangerous scientific
knowledge secret.
17Ibid.
19 There is a considerable amount of discussion about how dual-use ethics should be taught and who should
be teaching the scientists. For an extensive discussion, see Rappert, B. (ed.) 2010, Education and Ethics in the
Life Sciences, ANU E Press, Canberra.
279
The duties for scientists, as proposed by Kuhlau and her colleagues above, were
subjected to a number of different conditions20
it must be within their professional responsibility
it must be within their professional capacity and ability
it must be reasonably foreseeable
it must be proportionally greater than the benefits
it must be not more easily achieved by other means.
A brief survey of these conditions, it must be noted, does not make provision
for any deficiencies in research environments, lack of support or problems with
carrying out the duties. Rather, these conditions seem to delineate student
scientists from principle investigators, and technicians from researchers. Thus,
these conditions provide little in the way of support for scientists working under
research conditions that may be markedly different from the Western norm.21
It therefore becomes important to question whether dual-use responsibility
duties such as those proposed above make implicit assumptions about research
environments and the implementation of webs of prevention. If the duties
discussed above are therefore re-examined in light of this, a number of key
considerations are noted. First, the duty to prevent bioterrorism, while a
laudable goal, may be seen to be largely dependent on the existence of a web of
prevention in action. Without the combined efforts of the security, health and
judicial stakeholders, it is difficult to conceptualise how this duty may be carried
out. Without the integrated involvement of governmental and international
bodies,22 it is difficult to conceptualise how such a duty would be acted upon.
Second, the duty to engage in response activities raises important questions
about the responsibility of scientists in the absence of coordinated activities.
Is it their responsibility to lobby for the establishment of response activities,
or have they fulfilled their obligations solely due to the absence? Furthermore,
if scientists are obligated to engage in response activities, are they similarly
obliged to be involved in those not created in their own milieu? Are scientists,
for example, in developing countries morally obliged to actively participate in
any Western response activity, or is that in fact a form of ethical imperialism?
Third, although the duty to consider the negative implications of their work
may at the outset be seen as self-explanatory, it is vital to consider that risk
and benefit are interpreted quite differently around the world. Therefore, it is
20 Kuhlau et al., op. cit., pp. 4812.
21 As will be discussed in the following sections, these differences could be in regulatory controls, funding,
extra-laboratory service provision, governmental involvement and support, and access to the international
life-science community.
22 As proposed by Miller and Selgelid, op. cit.
280
highly likely that scientists in non-Western countriesparticularly countries

experiencing food insecurity, considerable healthcare challenges and (nonbioterrorism) security issuesmay view the negative implications of their work
in a totally different light to their Western colleagues.23 Thus, are scientists to
consider what they perceive to be the negative (and positive) implications of their
work, or simply the negative implications according to a Western perspective?
The third, fourth and fifth duties (not to publish or share sensitive information,
to oversee or limit access to dangerous materials and to report activities of
concern) all depend on the provision of national structures that will allow
scientists to report concerns, control their research and manage their security.
In the absence of such structures, it becomes crucial to question where
scientists responsibilities lie. Furthermore, will issues such as the fear of
losing international funding or collaboration mean that, in such circumstances,
scientists will not seek international alternatives?
Even the briefest of critiques of these duties in light of differing research
contexts highlights the importance of a more deeply contextualised dual-use
ethics discourse. As not all research environments are equal in their social,
physical and extra-laboratory support provisions, it becomes important to ask:
what happens when scientists cannot fulfil these duties through no fault of their
own?
Properly considering research environments

Such considerations, of course, present difficulties to any notion of global duties
for dual-use responsibility amongst scientists. In turn, it may be suggested
that this presents a crossroads to dual-use ethics discourse. Either the notion
of global duties must be abandoned in favour of more contextually sensitive
suggestions or the global duties must be thoroughly excavated to eliminate the
implicit expectations that they contain about research environments.
Such debates are, of course, extremely complicated and beyond the scope
of this chapter. Nonetheless, although not offering answers, the rest of the
chapter will concern itself with highlighting why such issues need to be taken
extremely seriouslyparticularly in the realm of dual-use ethics pedagogy. If
one considers the issues raised above, a number of questions immediately spring
to mind. These may include whether it is unfair to expect scientists to act as
whistleblowers when the likelihood of losing their job is near 100 per cent;
whether it is feasible to expect scientists to report their concerns when there
23 Bezuidenhout, L. (forthcoming), Moving life science ethics debates beyond national borders: some
empirical observations.
281
are no structures in place for them to do so; and whether, in the face of extreme
public health crises in many countries, the risk of losing funding outweighs any
threat of terrorism and thus perceptions of risk.
These concerns all require serious consideration. They suggest not only that
achieving a common culture of awareness and a shared sense of responsibility24
amongst the global scientific community may be more complicated than initially
envisioned, but also that current methods of raising dual-use awareness may
alienaterather than incorporatescientists from non-Western research
environments. If scientists are presented with such duties during dual-use ethics
education without accompanying discussion on the strengths and limitations of
implementing them contextually, it is just possible that such initiatives may do
more harm than good.
Such hesitations relate to another chapter in this volume, by Judi Sture, which
examines the concept of ethical erosion within communities of learners. It is
possible that the presentation of idealised duties or unattainable standards
of behaviour may significantly detract from attempts to engage scientists in
discourse about dual-use responsibilities. Indeed, studies with scientists in a
number of African laboratories strongly suggest that the wholesale importing
of Western ethical approaches to teach dual use to these scientists was limited
in success.25
These issues are further complicated by the lack of capacity in most developing
countries to invest in home-grown ethics initiativesat least for the moment.
Thus, as it stands, within developing countries ethics education often remains
largely in the hands of foreign funding agencies or interest groups. It is thus
plausible to reiterate the question: is a lack of understanding of the structure
of research within developing countries hampering efforts to build capacity
within ethics?
Researching in developing countries

One of the reasons that these issues of contextuality are so poorly represented in
dual-use ethics discourse is because of its historical legacy. Dual use has really
only become a topic of concern in the life sciences since the terrorism events of
2001. In the subsequent years of the war on terror, the majority of dual-use
discussion, quite naturally, occurred in the United Kingdom, the United States
and a small number of other developed Western countries.
24 National Science Advisory Board for Biosecurity (NSABB) 2006, Globalization, Biosecurity and the Future
of the Life Sciences, The National Academies Press, Washington, DC.
25 Bezuidenhout, op. cit.
282
Within these countries, the physical research environments of laboratories

and the regulations governing biosafety and biosecurity have high degrees of
similarity. Indeed, widely endorsed prerequisites for biosafety and biosecurity
provisions26 have further strengthened the harmony and standardisation
between these laboratories. Thus, in the case of most dual-use discussions, the
presence of a minimum level of biosafety and biosecurity regulation (that is
implemented effectively) is not discussed because it is already in place.
Furthermore, when one considers the process of daily life-science research in
light of any web of prevention, it becomes evident that a number of additional
assumptions are also made about research environments because of key similarities
between laboratories in the United States, the United Kingdom and the European
Union. These include issues such as
the existence of core funding for research facilities
stable, adequate and reliable water, electricity and transport services
stable, adequate and reliable postal, telecommunications and internet
provision
skilled support staff, such as technicians and those who maintain or repair
equipment
efficient and informed customs controls
national regulations governing and guiding biosafety and biosecurity
measures
capacity and protection for the reporting of misconduct.
Unfortunately, outside a Western context, such provisions should not be
automatically assumed. Indeed, many laboratories around the world represent
extremely different working conditionsones that challenge these assumptions.
Many laboratories in developing countries, as will be discussed below, struggle
daily with problems that range from a lack of core funding for facilities to
unreliable electricity supplies.
Despite these different working conditions, it is vital to note that the research
in these laboratories should not necessarily be considered unethical, unsafe or
insecure. Rather, scientists have often found innovative ways to work around
the limitations of their environments27 and produce high-quality research;
26 Laboratory biosafety includes areas such as recruiting and retaining qualified individuals, training,
laboratory work practices (for example, disinfection, waste handling, material control and accountability),
personal protective equipment, medical surveillance, maintenance, access controls, self-assessments,
documentation, corrective actions, reporting requirements, and incident response plans. Attempts to
standardise such procedures have been spearheaded by the World Health Organisation (WHO), which has
published guidelines such as World Health Organisation (WHO) 2004, Laboratory Biosafety Manual, World
Health Organisation, Geneva.
27 There are a number of innovative low-cost alternatives to many biosafety and biosecurity requirements
while still ensuring that the laboratories comply with international standards.
283
however, what does need to be considered is whether the current Westerncentric approach to dual-use responsibility as presented in current ethics
education may potentially alienate scientists by presenting them with duties
they cannot fulfil.
Research environments in developing countries

Literature on laboratory life in developing countries is comparatively scarce.
Indeed, the majority of discussion of laboratory environments, provisions
for research and national support structures often comes from research and
development (R&D) focused policy reports aimed at capacity building and
investment. These reports tend to focus more on the shortcomings of these
research environments rather than innovative alternatives; however, they
provide some important considerations about research environments in
developing countries.
The issues identified in these reports include28
small, undifferentiated institutions
lack of funding and lack of effective mechanisms for utilisation of funds
high teaching burden
lack of experienced mentors due to brain-drain
history of poor investment in higher education and research
lack of buy-in by institutions for new initiatives (such as centres of excellence)
need for networking and networks
corruption, mismanagement and institutional rigidity
lack of governmental support, funding and control
lack of vetted information about possible collaborators and institutions
need for strong administrative and managerial skills
inadequate resources and allocation thereof.
Furthermore, very few developing countries currently contribute even 1 per
cent of their gross domestic product (GDP) to science R&D. Because of this
lack of governmental involvement, many of these laboratories lack dedicated
core funding, meaning that all day-to-day research expenses must somehow be
28 As informed by the fieldwork; Kiringia, J. M., Wambebe, C. and Baba-Moussa, A. 2005, Status of
national research bioethics committees in the WHO African region, BMC Medical Ethics, vol. 6, no. 10; Fine,
J. C. 2007, Investing in STI in sub-Saharan Africa: lessons from collaborative initiatives in research and higher
education, Global Forum: Building Science, Technology and Innovation Capacity for Sustainable Growth and
Poverty Reduction, Washington, DC; and Council on Health Research for Development (COHRED) 2010, Fact
Sheet: NEPAD-COHRED Strengthening Pharmaceutical Innovation in Africa, COHRED, Johannesburg.
284
covered by grants for dedicated projects. This has far-reaching implications, as

often salaries for researchers and technicians are not guaranteed by the research
institutions.
Taken together, such characteristics raise some important considerations for
the duties presented above. In particular, the disjunction between the ideal
and real research environments as well as the distance between the desired
and actual behavioural outcomes may have significant impacts on attempts to
engage developing-country scientists in dual-use discussion.
Between 2011 and 2012, I conducted a large number of interviews with
scientists in a range of African life-science laboratories.29 This research, which
is also discussed in a number of publications,30 aimed to examine whether
developing-country scientists engaged easily with the dual-use debate as it is
currently presented in educational modules. This study thus, incorporating
the issues raised above, questioned whether the current approaches to dualuse responsibility were too Westernised for scientists in these countries and
needed re-examining.
The majority of the participants in this study had problems with the concept
of dual use as it was presented in current educational modules. In particular,
the strong focus on bioterrorism and the perceived lack of sensitivity towards
the problems they experienced in their daily research were often used as
justifications for this negative reaction. The two following sections detail a
couple of the most prevalent responses: first, that dual-use was not a problem for
African scientists, and second, that even if it was a problem, there was nothing
that could be done about it.
Why is dual use my problem?
Many of the scientists interviewed approached the concept of dual use as a
problem far removed from their personal research. While many of the scientists
expressed a strong academic interest in the problems associated with dual use,
they nonetheless maintained that it was not a problem for African scientific
research. In many interviews participants made comments similar to: I think
that in Africa we just dont deal with such questions. I think its more in the
domain of the Western world, America, UK, where the threat of bioterrorism is
a very real threat and so I think this issue is poignant there.
Some of the reactions, on the other hand, were more extreme and at least 20 per
cent of interviewees expressed hostility regarding the concept and its presence
29 At least 40 interviews in four different sites (three countries) with participants ranging from postgraduate
students to heads of departments and institute directors.
30 See Bezuidenhout, op. cit.; and Bezuidenhout, L. (forthcoming), Ethics in the minutiae: examining the
impact of daily laboratory processes on ethical behaviour and ethics education.
285
in international debates. In many cases, as below, scientists expressed frustration

at what they saw as a skewed Western perspective of the state of science,
emphasising the other serious issues that they felt should take precedence. One
participant was emphatic in their opinion, stating:
I thought it was totally irrelevant and paranoid on the part of the
Western world for this threat that often doesnt materialise and its just
huge amounts of money that go into fighting this phantom threat where
I feel like we have more important things to do here as were in the
middle of such a huge HIV and TB epidemic and we just want to get
on with doing the research. It was not an issue that Id ever considered
before and quite frankly I dont feel its very relevant.
Many similar responses mentioned that dual use was just not a topic for
discussion because of the considerable healthcare and food-security challenges
within these countries that the scientists were attempting to address through
their research. Such perceptions of risk and benefit obviously differ from the
expectations of current dual-use discourse, and may significantly challenge how
the duty to consider the negative implications of their work31 was interpreted
by these African scientists. Furthermore, one must question how the duty to
prevent bioterrorism might be interpreted if, as with the participant quoted
above, scientists perceived the topic to be totally irrelevant and paranoid.
Such considerations have far-reaching consequences for the success of ethics
education and the perpetuation of dual-use awareness. Lack of personal buyin to the concept of dual use due to differing perceptions of risk and benefit
has severe consequences for ethics pedagogy, as students will struggle to make
the connection between the information received during their instruction and
their daily behaviour. This undermines the concept of a culture of responsible
awareness and questions the effectiveness of concepts such as codes of conduct
for the life sciences.
Nonetheless, during this fieldwork a number of the participants manipulated the
dual-use concept to fit the concerns within their own environment. Thus, still
taking dual use to refer to the potential for beneficial scientific research to be
misused for nefarious purposes by a third party,32 the scientists connected the
harm caused by a third party to fearmongering within irresponsible scientific
journalism instead of bioterrorism. By contextualising the concept within their
own research environment, the participants were then able (and willing) to
critically re-examine the concept and its utility within their daily research.33
31 One of the duties proposed by Kuhlau et al., op. cit.

32 As proposed by Miller and Selgelid, op. cit.
33 This is extensively examined in Bezuidenhout, Ethics in the minutiae, op. cit.
286
Even if dual use is a problem, what can I do about it?

Another characteristic of many of the participants interviewed was that
they removed themselves from responsibility for the dual-use problem by
emphasising their lack of agency for engaging with any sort of dual-use control.
This was often related to a number of characteristics within their environments.
In many cases, participants suggested that [b]eing alone you cant change the
system [and] you just become a problem to the institution. It has to come from
above. If you come somewhere personally and tell them about standards you are
causing problems. The participants commonly associated these feelings with
issues within their research environments.
Those at all the field sites mentioned the difficulties of getting reagents and
samples in and out of their countries. They mentioned that poorly trained border
officials, unrealistic foreign export and import requirements and poor transport
infrastructure significantly complicated their daily research. Statements such
as [i]t already takes four to six weeks to get a delivery through, so any extra
restrictions will make it even worse regularly appeared in the interviews. In
such cases, it must be asked whether expecting scientists to raise awareness of
dual-use concerns34 and draw attention to the negative implications of their
work35 are remotely feasible.
At one of the field sites I was made aware of the fact that the government did not
provide any core funding for the facility, and that the entirety of their running
costs and research budget came from foreign grants. This, fieldwork participants
suggested, raised two important problems for the dual-use control duties. First,
as the possibility of losing their funding had implications far beyond stopping
a project, many participants stated that they would be unwilling to report
activities of concern36 due to the possibility that misunderstandings might shut
us down entirely. Second, the participants highlighted that any expectations
of improving security or limiting the access to dangerous materials37 were
unrealistic as general facility maintenance and improvement were not included
in project-specific grants. Thus, the participants viewed themselves as in a
difficult position in which not raising dual-use concerns was indeed the lesser
of two evils.
In all of the facilities participants also regularly asked me what response
activities they had a duty to engage in, and to whom they could report activities
of concern. In both cases it was patently clear that a lack of government
involvement and low levels of institutional buy-in for dual-use control meant
34
35
36
37
One of the duties proposed by Kuhlau et al., op. cit.

One of the duties proposed by ibid.
287
that there were often no answers to be had. It was my impression that presenting
these duties without a proper, contextually considered understanding of how it
may be implemented often turned the participants off the dual-use discussion in
its entirety, as it was once again perceived as not a problem for Africa.
If, as is the case in dual-use ethics, the duties continue to be presented as moral
obligations, it is also easy to see how these scientists are placed in ethically
untenable positions that compromise their ethical development. Thus, structural
issues within the research environment, if unaddressed, have the potential to
undermine ethical training. Continually facing deficits in the ethical conflicts
between expected duties and the characteristics of the research environment
may cause frustration and resignation amongst the scientists and lower the
potential for them to get involved in ethical discussions. It must be asked how
to reflect these environmental issues within ethics training.
Starting to re-contextualise dual-use ethics

This chapter thus presents a contrasting perspective of dual-use ethics discourse
and pedagogy to prevailing norms. Based on the theoretical analysis and the
fieldwork discussed here and elsewhere it strongly suggests that the current dualuse ethics discourse is largely de-contextualised due to the absence of discussion
on variations in research environments. This has significant implications for
discussions on responsibility and duty rhetoric and consequently for the
building of a common culture of awareness and a shared sense of responsibility
within the global community of life scientists.38
The majority of responsibility and duty rhetoric within dual-use discussions
is strongly influenced by the web of prevention model of dual-use oversight.
Despite a strongly deontological, global approach to these discussions, the
chapter highlighted that these duties have a number of implicit assumptions
about the laboratory (and extra-laboratory) environments in which lifescience research takes place. Unsurprisingly, due to the historical legacy
of the dual-use discussion, these assumptions are based predominantly on a
Western understanding of a minimum level of research environment. Such
an environment includes key contributions from the government, the research
institution and the general surrounding society.
As this chapter points out, however, such an environment only reflects a
portion of laboratories around the world, and in many different countries highquality research is occurring under significantly different conditions. In such
38 National Science Advisory Board for Biosecurity 2006, Globalization, Biosecurity and the Future of the
Life Sciences, Washington D. C., The National Academies Press. See pp 5.
288
cases, the implicit assumptions about research environments inherent within

ethics discourse may serve to alienate scientists from engaging within the dualuse debate. Paradoxically, as the life sciences become increasingly global, these
previously marginalised communities of scientists are precisely the ones that
urgently require representation.
So, it remains to be asked, what can be done? As mentioned above, it may be
that dual-use ethicsparticularly responsibility ethicsneeds to be critically
re-evaluated. Either the notion of global duties for dual-use control must be
abandoned in favour of more contextually sensitive suggestions or the global
duties must be thoroughly excavated to eliminate the implicit expectations
that they contain about research environments. Of course, this is a complicated
discussion and will require considerable attention from the dual-use ethics
community.
Much else, however, can be done. Within much dual-use discussion there is a
tendency to talk about research environments as largely homogenous bodies.
Although, of course, much of the discourse remains centred in (and on) Western
countries, developing a sense of awareness of the heterogeneity of research
environments (even within these countries) cannot help but strengthen current
discourse. Furthermore, cultivating an awareness of the challenges faced
by developing-country researchers will no doubt facilitate their inclusion in
international discourse.
Within ethics pedagogy, any gestures towards contextual sensitivity will no
doubt be much appreciated. In particular, educationalists should consider
how discussions about implementing duties into daily life may be fostered by
pedagogical initiatives. With regards to developing-country scientists, however,
future educational initiatives should also consider whether developing more
context specific case studies39 can really fix current problems, or whether
considerable and sustained attention needs to be paid to the problems identified
above.
It has often been quoted that dual use cannot be addressed within national
borders or in institutions, but requires a regionalindeed an international
response due to the nature of the threat and the characteristics of modern life
sciences. In order to adequately realise this need, a global body of dual useaware scientists is vital. It is therefore becoming increasingly important that the
previously marginalised groups of scientists from developing countries become
key players in the development of a truly international dual-use discussion.
In order to do so, the first step is to recognise the differences inherent in others.
39 National Research Council, op. cit.

289
Part IV: Ethical Futures
18. Exploring the Role of Life

Scientists in Combating the
Misuse of Incapacitating
Chemical and Toxin Agents
Michael Crowley
Introduction
Over the past two decades there has been a revolution in the life sciences
with extremely rapid advances in genomics, synthetic biology, biotechnology,
neuroscience and the understanding of human behaviour. The speed of
progress is staggering. For example, in 1999 a special meeting of the National
Academies of Sciences and the Society of Neuroscience noted that [t]he past
decade had delivered more advances than all previous years of neuroscience
research combined.1 Many of these developments have great potential to
benefit humankindin, for example, the production of more effective, safer
medicines.2 Concern has been raised, however, by a growing number of those
in the scientific and medical communities regarding the dual-use nature of
certain advances with the consequent danger of the new technologies being
misused for the development of a new range of chemical or biological weapons.
Meselson has stated that [d]uring the century ahead, as our ability to modify
fundamental life processes continues its rapid advance, we will be able not
only to devise additional ways to destroy life, but also be able to manipulate it
including the processes of cognition, development and inheritance.3
And he added: A world in which these capabilities are widely employed for
hostile purposes would be a world in which the very nature of conflict had
radically changed. Therein could lie unprecedented opportunities for violence,
coercion, repression or subjugation.4
1 Society of Neuroscience 1999, Neuroscience 2000: A New Era of Discovery, Symposium organised by the
Society of Neuroscience, Washington, DC, 1213 April.
2 Andreasen, N. 2004, Brave New Brain: Conquering Mental Illness in the Era of the Genome, Oxford
University Press, USA.
3 Meselson, M. 2000, Averting the hostile exploitation of biotechnology, The CBW Conventions Bulletin,
no. 48, pp. 1619.
4 Ibid., pp. 1619.
293
294
Substance P
Fentanyl
Carfentanil
Cyanide
Phosgene
Infect
Source: Adapted from: Pearson, G. 2002, Relevant scientific and technological developments for the first CWC Review Conference: the BTWC Review Conference experience, CWC
Review Conference Paper, no. 1, University of Bradford, UK, p. 5.
Poison
Plague
Tularemia
Midazolam
Anthrax
Dexmedetomidine
Rickettsia
Etorphine
Viruses
Bacteria
Traditional
biological
weapons
Remifentanil
Modified/tailored
bacteria and viruses
Genetically modified
biological weapons
Mustard
Staphylococcal enterotoxin B
(SEB)
Toxins
Nerve agents
Neurokinin A
Bio-regulators
peptides
Industrial pharmaceutical
chemicals
Classical chemical
weapons
Table 18.1 (Bio)chemical Threat Spectrum Chart
18. Exploring the Role of Life Scientists in Combating the Misuse of Incapacitating Chemical and Toxin Agents
As the ongoing revolution in the life sciences has proceeded, the boundary
between chemistry and biology, and consequently the distinction between
certain chemical and biological weapons, has become increasingly blurred.
Rather than thinking of chemical and biological weapons threats as distinct,
certain analysts including Aas,5 Dando,6 Davison7 and Pearson8 believe it is more
useful to conceptualise them as lying along a continuous biochemical threat
spectrum. This chapter will focus upon research and development of those midspectrum agents (pharmaceutical chemicals, bio-regulators and toxins featured
in Table 18.1) that some consider as having potential utility as incapacitating
weapons (incapacitants). The chapter will employ a holistic arms control (HAC)
approach to examine the potential dangers and proposed utility of such agents
and explore the obligations and opportunities for the life-science community to
ensure that such agents are not utilised for hostile purposes.
Incapacitants: A primer
Although certain states and multilateral organisations such as the North Atlantic
Treaty Organisation (NATO)9 have sought to characterise incapacitants, there is
currently no internationally accepted definition for these chemical agents. Indeed
certain leading international experts believe that such a technical definition is
not possible.10 Whilst recognising the contested nature of this discourse, as a
provisional working description, they can be considered as substances whose
chemical action on specific biochemical processes and physiological systems,
especially those affecting the higher regulatory activity of the central nervous
5 Aas, P. 2003, The threat of mid-spectrum chemical warfare agents, Prehospital and Disaster Medicine, vol.
18, no. 4, pp. 30612.
6 Dando, M. 2007, Scientific outlook for the development of incapacitants, in A. Pearson, M. Chevrier and
M. Wheelis (eds), Incapacitating Biochemical Weapons, Lexington Books, Lanham, Md, p. 125.
7 Davison, N. 2009, Non-Lethal Weapons, Palgrave Macmillan, Basingstoke, UK, pp. 1067.
8 Pearson, op. cit.
9 NATO defines an incapacitant as a chemical agent which produces temporary disabling conditions which
(unlike those caused by riot control agents) can be physical or mental and persist for hours or days after
exposure to the agent has ceased. Medical treatment, while not usually required, facilitates a more rapid
recovery. North Atlantic Treaty Organisation (NATO) 2000, Glossary of Terms and Definitions (AAP-6 (V),
Modified version 02), 7 August 2000.
10 A report of an expert meeting organised by Spiez Laboratory concluded that because there is no clearcut line between (non-lethal) ICA [incapacitating chemical agents] and more lethal chemical war-fare agents,
a scientifically meaningful definition cannot easily be made. One can describe several toxicological effects
that could be used to incapacitate, but in principle there is no way to draw a line between ICAs and lethal
agents. See Mogl, S. (ed.) 2011, Technical Workshop on Incapacitating Chemical Agents, Spiez, Switzerland,
89 September; Spiez Laboratory 2012, op cit., p. 10; The Royal Society 2012, Brain Waves Module 3:
Neuroscience, Conflict and Security, Science Policy Centre, The Royal Society, London, pp. 445; Organisation
for the Prohibition of Chemical Weapons (OPCW) 2013, Report of the Scientific Advisory Board on Developments
in Science and Technology for the Third Special Session of the Conference of the States Parties to the Chemical
Weapons Convention, Third Review Conference RC-3/DG.1, 819 April 2013, 29 October 2013, p. 4.
295
system, produce a disabling condition (for example, can cause incapacitation or

disorientation, incoherence, hallucination, sedation or loss of consciousness) or,
at higher concentrations, death.11
There is a wide variety of agents that could potentially be utilised as
incapacitants including anaesthetic agents, skeletal muscle relaxants, opioid
analgesics, anxiolytics, antipsychotics, antidepressants and sedative-hypnotic
agents,12 many of which are currently legitimately utilised by the medical or
veterinary professions.13 Table 18.2 summarises the results of a literature study
and analysis of biomedical research into a range of pharmaceutical agents,
published in 2000, by the Applied Research Laboratory and the College of
Medicine at Pennsylvania State University, to identify the range of drug classes
that had potential utility as incapacitants. The study clearly illustrates and
indeed actively explores the potential dual-use applications of drugs initially
developed for medical purposes, noting that:
It is well known that for every one new compound successfully
proceeding from the discovery phase through all phases of clinical trials
and on to market, perhaps hundreds, if not thousands, of compounds
are discarded or shelved by the pharmaceutical industry [for example,
as a result of their side effects] However, in the variety of situations
in which non-lethal techniques are used there may be less need to
be concerned with side-effects; indeed, perhaps a calmative may be
designed that incorporates a less than desirable side-effect as part of
the drug profile.14
Furthermore, the study recommends explicit collaboration in this area,
stating that it may be appropriate to develop a working relationship with the
pharmaceutical industry to better incorporate their knowledge and expertise in
developing a non-lethal calmative technique.15 The ethical implications of such
relationships are, however, not explored.
11 Adapted from Pearson et al., op. cit., p. xii. Incapacitants have also been called advanced riot-control
agents, biochemical agents, biotechnical agents, calmatives, incapacitating biochemical weapons and
immobilising agents.
12 See, for example: Lakoski, J., Bosseau Murray, W. and Kenny, J. 2000, The Advantages and Limitations of
Calmatives for Use As A Non-Lethal Technique, College of Medicine Applied Research Laboratory, Pennsylvania
State University.
13 See Aas, op. cit., p. 309.
14 Lakoski et al., op. cit., p. 48.
15 Ibid., p. 48.
296
Figure 18.2 Indicative Drug Classes and Agents with Potential Utility as
Incapacitants
Drug class
Selected compounds
Site of action
Benzodiazepines
Diazepam
GABA receptors
Midazolam
Etizolam
Flumazenil (antagonist)
Alpha2 adrenergic receptor
agonists
Clonidine
Dexmedetomidine
Alpha2 adrenergic
receptors
Fluparoxan (antagonist)
Dopamine D3 receptor agonists
Pramipexole
D3 receptors
Cl-1007
PD 128907
Selective serotonin reuptake
inhibitors
Fluoxetine
5-HT transporter
Sertraline
Paroxetine
WO-09500194
Serotonin 5-HT1A receptor

agonists
Buspirone
5-HT1A receptor
Lesopitron
Alnespirone
MCK-242
Oleamide
WAY-100, 635
Opioid receptors and mu agonists
Morphine
Mu opioid reception
Carfentanil
Naloxone (antagonist)
Neurolept anaesthetics
Corticotrophin-releasing factor
receptor antagonists
Propofol
GABA receptors
Droperidol and fentanyl

combination
DA, NE and GABA

receptors
Phencyclidines
Opioid receptors
CP 154,526 (antagonist)
CRF receptor
NBI 27914 (antagonist)

CRF-BP (binding protein)
Cholecystokinin B receptor
antagonists
CCK-4
CCKB receptor
Cl-988 (antagonist)
Cl-1015 (antagonist)
Source: Adapted from Lakoski, J., Bosseau Murray, W. and Kenny, J. 2000, The Advantages and Limitations of
Calmatives for Use As A Non-Lethal Technique, College of Medicine Applied Research Laboratory, Pennsylvania
State University, pp. 1516.
297
There is a long history, dating from the late 1940s, of certain state programs
attempting to develop incapacitant weapons employing a range of pharmaceutical
chemicals or toxins.16 Analysis of open-source information from the mid
1990s onwards indicates that a number of states including China,17 the Czech
Republic,18 Russia19 and the United States20 appear to have conducted research
relating to incapacitants and/or possible means of delivery at some stage during
this period. It is, however, difficult to establish the current situation, and certain
states that have previously shown an interest in developing such agentssuch
as the United Stateshave recently declared that no such activities currently
take place.21
According to the International Committee of the Red Cross (ICRC): There is
clearly an ongoing attraction to incapacitating chemical agents but it is not
easy to determine the extent to which this has moved along the spectrum from
academia and industrial circles into the law enforcement, security and military
apparatuses of states.22
16 See, for example: Crowley, M. 2009, Dangerous Ambiguities: Regulation of Riot Control Agents and
Incapacitants under the Chemical Weapons Convention, University of Bradford, UK; Dando, M. and Furmanski,
M. 2006, Midspectrum incapacitant programs, in M. Wheelis, L. Rzsa and M. Dando (eds), Deadly Cultures:
Biological Weapons Since 1945, Harvard University Press, Cambridge, Mass.; Davison, op. cit.; Furmanski,
M. 2007, Historical military interest in low-lethality biochemical agents, in Pearson et al., op. cit.; Pearson,
A. 2006, Incapacitating biochemical weapons: science, technology, and policy for the 21st century,
Nonproliferation Review, vol. 13, no. 2.
17 Crowley, op. cit., p. 82; Guo Ji-Wei and Xue-sen Yang 2005, Ultramicro, nonlethal and reversible: looking
ahead to military biotechnology, Military Review, JulyAugust, as cited in Pearson, A. 2007, Late and postCold War research and development of incapacitating biochemical weapons, in Pearson et al., op. cit.
18 Hess, L., Schreiberov, J., Mlek, J. and Fusek, J. 2007, Drug-induced loss of aggressiveness in the
macaque rhesus, Proceedings of 4th European Symposium on Non-Lethal Weapons, 21st23rd May 2007,
Ettlingen, Germany, European Working Group on Non-Lethal Weapons, Pfinztal: Fraunhofer ICT, V15; Hess,
L., Schreiberova, J. and Fusek, J. 2005, Pharmacological non-lethal weapons, Proceedings of the 3rd European
Symposium on Non-Lethal Weapons, 10th12th May 2005, Ettlingen, Germany, European Working Group on
Non-Lethal Weapons, Pfinztal: Fraunhofer ICT, V23; Davison, N. and Lewer, N. 2006, Bradford Non-Lethal
Weapons Research Project (BNLWRP)Research Report No. 8, University of Bradford, UK, p. 50.
19 Klochikin, V., Pirumov, V., Putilov, A. and Selivanov, V. 2003, The complex forecast of perspectives of
NLW for European application, Proceedings of the 2nd European Symposium on Non-Lethal Weapons, 13th14th
May 2003, Ettlingen, Germany, European Working Group on Non-Lethal Weapons, Pfinztal: Fraunhofer ICT;
Klochinkhin, V., Lushnikov, A., Zagaynov, V., Putilov, A., Selivanov, V. and Zatekvakhin, M. 2005, Principles
of modelling of the scenario of calmative application in a building with deterred hostages, Proceedings of
the 3rd European Symposium on Non-Lethal Weapons, 10th12th May 2005, Ettlingen, Germany, European
Working Group on Non-Lethal Weapons, Pfinztal: Fraunhofer ICT.
20 Crowley, op. cit., pp. 768; Davison, op. cit., pp. 10542; Furmanski, op. cit.; Pearson, 2007, op. cit.;
Furmanski and Dando, op. cit.
21 Organisation for the Prevention of Chemical Weapons (OPCW), Executive Council 2013, Statement by
Ambassador Robert P. Mikulak, United States Delegation to the OPCW at the Seventy Second Session of
the Executive Council, Seventy-Second Session EC-72/NAT.8, 67 May 2013, 6 May 2013. In his statement,
ambassador Mikulak declared [i]n this context, I also wish very clearly and directly to reconfirm that the
United States is not developing, producing, stockpiling, or using incapacitating chemical agents.
22 International Committee of the Red Cross (ICRC) 2010, Expert Meeting: Incapacitating Chemical Agents,
Implications for International Law, Montreux, Switzerland, 2426 March, p. 3.
298
Box 18.1 Contemporary Czech Republic research into incapacitating

chemical agents
The Czech Republic has had a longstanding research program into incapacitants dating from
at least 2000, part of which was funded by the military.a In May 2005, at the Third European
Symposium on Non-Lethal Weapons, Czech researchers delivered a paperb describing their
investigations over several years, administering rhesus monkeys with various pharmacological
cocktails in order to determine which combinations and doses resulted in fully reversible
immobilization. The paper also described how [f]ully reversible analgesic sedation was
tested in man, utilising the triple combination of dexmedetomidine, midazolam and fentanyl
given to patients undergoing surgery, and a second combination of dexmedetomidine,
midazolam and ketamine, which was tested on 10 nurses.c
In a 2007 follow-up paper, Czech researchers described how they decided to test new
combinations [of drugs] for suppression or complete abolition of aggressive behaviour in
macaque monkeys.d The researchers claim that the results can be used to pacify aggressive
people during medical treatment (mental disease), terrorist attacks and during production
of new pharmacological nonlethal weapons.e In July 2010, Czech researchers published
a paper describing their studies inducing immobilisation in orang-utans and chimpanzees
utilising a naphthylmedetomidine-ketamine-hyaluronidase combination.f Although the results
of this research were presented in terms of facilitating the relocation and painless medical
examination of the animals, such research may also potentially be applicable to incapacitant
development. Czech researchers have also investigated a number of alternative means of
agent delivery including via inhalation administration, which was initially tested on rats
and then on human volunteers,g who were reported to have been children in hospital.h
Researchers have also explored conjunctival, nasal, transbucal, sublingual and transdermal
administration.i
According to Davison and Lewer, research to develop sedative and anaesthetic agent combinations
for use as weapons had been funded by the Czech Army under Project No: MO 03021100007. See
Davison and Lewer, op. cit., p. 50.
a
Hess et al., 2005, op. cit.
Ibid., pp. 89.
Hess et al., 2007,op. cit., p. 6.
Ibid., p. 7.
Hess, L., Votava, M., Schreiberov, J., Mlek, J. and Horek, M. 2010, Experience with a
naphthylmedetomidine-ketamine-hyaluronidase combination in inducing immobilization in
anthropoid apes, Journal of Medical Primatology, vol. 39, no. 3 (June), pp. 1519.
f
Hess et al., 2005, op. cit., pp. 1112.
Davison, op. cit., p. 128.
Hess et al., 2005, op. cit., pp. 1014.
Potential dangers and proposed utility

Proponents of incapacitants have promoted their development and use in
certain law-enforcement scenarios (such as hostage-taking situations) where
there is a need to rapidly and completely incapacitate a single or a group of
individuals without causing death or permanent disability. Incapacitants have
also been raised as a possible tool in a variety of military operations, especially
in situations where combatants and noncombatants are mixed.23
23 See, for example: Fenton, G. 2007, Current and prospective military and law enforcement use of chemical
agents for incapacitation, in Pearson et al., op. cit., pp. 10323; Whitbred, G. 2006, Offensive use of chemical
299
A broad range of observers, howeverincluding scientific and medical

professionals, arms-control organisations, international legal experts, human
rights monitors and humanitarian organisations, as well as a number of statesis
highly sceptical about the development and utility of incapacitants, highlighting
the fact that such weapons are not inherently nonlethal, even if they were to be
used with a nonlethal intent. In their 2003 study conducted under the auspices
of the Federation of American Scientists, Klotz et al. developed a predictive
model illustrating why seemingly non-lethal incapacitating agents may be
quite lethal in actual use.24 In their conclusion, they stated: We have shown,
at least within the approximations of our simple (but generous) two receptor
equilibrium model, that even with a therapeutic index of 1,000 (above any
known anaesthetic or sedative agent), a chemical agent used as an incapacitating
weapon can be expected to cause about 10% fatalities.25
Furthermore, as Pearson has noted, even such predictive modelling will
potentially underestimate fatalities when an incapacitant is used in real-life
situations where there is uncontrollable variability both in terms of exposure
(uneven concentration and exposure time) and within the target population
(age, size, gender, health status and individual susceptibility).26 As a result of
such considerations, the British Medical Association believes:
The agent whereby people could be incapacitated without risk of death
in a tactical situation does not exist and is unlikely to in the foreseeable
future. In such a situation, it is and will continue to be almost impossible
to deliver the right agent to the right people in the right dose without
exposing the wrong people, or delivering the wrong dose.27
Similarly, Klotz et al. concluded that genuinely non-lethal chemical weapons
are beyond the reach of current science.28
technologies by US special operations forces in the global war on terrorism, Maxwell Paper, no. 37, Air
University Press, Maxwell Air Force Base, Ala. It should be noted that other authors have questioned the
utility of incapacitants in certain proposed scenarios such as premeditated hostage situations, due to the
availability of countermeasures. See Wheelis, M. 2007, Non-consensual manipulation of human physiology
using biochemicals, in Pearson et al., op. cit., p. 6.
24 Klotz, L., Furmanski, M. and Wheelis, M. 2003, Beware the sirens song: why non-lethal incapacitating
chemical agents are lethal, Federation of American Scientists Paper, <http://www.fas.org/bwc/papers/sirens_
song.pdf> (viewed 8 February 2012).
25 Ibid., p. 7.
26 Pearson, 2007, op. cit., p. 70.
27 British Medical Association (BMA) 2007, The Use of Drugs As Weapons: The Concerns and Responsibilities
of Healthcare Professionals, BMA, London, p. 1.
28 Klotz et al., op. cit., p. 1.
300
Box 18.2 Use of Chemical Incapacitant by the Russian Federation

Concerns about incapacitants were heightened following the use of a presumed derivative of
fentanyl by Russian security forces to free more than 800 hostages held by heavily armed
Chechen separatists in the Dubrovka Theatre in Moscow, in October 2002.a According
to reports, 30 minutes after an incapacitant was pumped into the theatre, the building
was stormed by Russian Spetsnaz special forces who killed all of the Chechen hostage
takers, including those left unconscious from the incapacitant, in apparent contravention
of international humanitarian law.b Although the hostages were released, more than 120
died as a result of the direct effects of the agent used or of airway constriction due to their
incapacitation. An undetermined, but large, additional number of hostages suffered long-term
damage, or died prematurely in the years after the siege.c
Treatment of the hostages who had been poisoned was delayed and compromised by the
refusal of the Russian authorities to state publicly what type of incapacitant had been
used in the theatre for four days after the siege had ended.d On 30 October 2002, the
Russian health minister, Yuri Shevchenko, identified the incapacitating agent as a mixture
of derivative substances of the fast action opiate Fentanyl.e Shevchenko further stated that
I officially declare: chemical substances which might have fallen under the jurisdiction of
the international convention on banning chemical weapons were not used during the special
operation.f The minister refused, however, to be more precise about the chemicals used even
on 11 December 2002 when faced with a parliamentary question. He said it was a state
secret.g In 2012, results of trace analysis undertaken in the United Kingdom of extracts
of clothing and urine from survivors indicated that the aerosol comprised a mixture of two
anaesthetics, carfentanil and remifentanil.h At the time of writing, the Russian authorities
have still not publicly stated exactly what chemical or chemicals were used.
For descriptions of the incident, see, for example: Koplow, op. cit.; Pearson et al., op. cit.; British
Broadcasting Corporation (BBC) 2004, Horizon: The Moscow Theatre Siege, BBC 2, 15 January, <http://
www.bbc.co.uk/science/horizon/2004/moscowtheatretrans.shtml> (viewed 30 July 2009); Amnesty
International 2003, Rough Justice: The Law and Human Rights in the Russian Federation, October,
Amnesty International, London, AI Index EUR 46/054/2003.
a
Specifically, the prohibition against attacking those recognised as hors de combat. See: Henckaerts, J.
2005, Study on customary international humanitarian law: a contribution to the understanding and
respect for the rule of law in armed conflict, International Review of the Red Cross, vol. 87, no. 857, p.
203.
b
Wheelis, M. 2010, Human impact of incapacitating chemical agents, inICRC, op. cit.; Levin, D. and
Selivanov, V. 2009, Medical and biological issues of NLW development and application, Proceedings
of the 5th European Symposium on Non-Lethal Weapons, 11th13th May 2009, Ettlingen, Germany,
European Working Group on Non-Lethal Weapons, Pfinztal: Fraunhofer ICT, V23, p. 7.
c
See, for example: Human Rights Watch 2002, Independent commission of inquiry must investigate
raid on Moscow theater: inadequate protection for consequences of gas violates obligation to protect
life, Press release, 30 October 2002, Human Rights Watch.
d
Russian experts discuss use of fentanyl in hostage crisis, ITAR-TASS, [from Moscow in English], 2112
hrs GMT, 30 October 2002, FBIS-SOV-2002-1030, as cited by Perry Robinson, October 2007, op. cit.
e
Alison, S. 2002, Russian confirms siege gas based on opiate fentanyl, [from Moscow for Reuters],
1257 hrs ET, 30 October 2002, as cited in Perry Robinson, 2007, op. cit.
f
Amnesty International, op. cit., p. 53.
Riches, J., Read, R., Black, R., Cooper, N. and Timperley, C. 2012, Analysis of clothing and urine
from Moscow theatre siege casualties reveals carfentanil and remifentanil use, Journal of Analytical
Toxicology, vol. 36, pp. 64756.
h
301
In addition, Nixdorff and Melling have surveyed the potential long-term

physiological consequences of exposure to incapacitants. Although insufficient
research has been undertaken to produce conclusive results, they believe that:
Numerous human and animal studies have shown that exposure to
incapacitating biochemical agents may induce heterogeneous cognitive
and physiological impairments and [may] lead to long term health
effects. This is even more pronounced when exposures to incapacitating
agents are combined with other factors such as stress or activation of the
immune system.29
Even if all technical barriers to the development of a truly nonlethal incapacitant
were overcome, there are a number of serious risks and damaging consequences
that could follow from the development of such weapons. These include the
following.
Creeping legitimisation: Perry Robinson believes that attempts by certain
states, particularly the United States, to legitimise the development and use
of incapacitants threaten to erode the norm against the weaponisation of
toxicity.30 He believes that this creeping legitimisation presents the greatest
danger to the existing prohibitions on chemical and biological weapons and
to the re-emergence of chemical and biological warfare.31
Proliferation and legitimisation by states: Pearson has warned that efforts
to develop incapacitating weapons may well gather steam as more nations
become intrigued by them and, observing the efforts of Russia and the United
States, become convinced not only that effective and acceptably non-lethal
incapacitating agents can be found, but that their use will be legitimized.32
Proliferation to, and misuse by, non-state actors: Analysts have highlighted
the potential utility of incapacitants to a range of non-state actors including
criminals, terrorists, paramilitary organisations and armed factions in failing
or failed states many of whom would not feel as constrained as states by
international law and concerns about lethality.33
Use as a lethal force multiplier: There are concerns that incapacitants will be
used by both military and law-enforcement agencies, not as an alternative
29 Nixdorff, K. and Melling, J. 2007, Potential long-term physiological consequences of exposure to
incapacitating biochemicals, in Pearson et al., op. cit., p. 165.
30 Perry Robinson, J. 2007, Categories of challenge now facing the Chemical Weapons Convention,
52nd Pugwash CBW Workshop: 10 Years of the OPCW: Taking Stock and Looking Forward, Noordwijk, the
Netherlands, 1718 March, p. 20; Perry Robinson, J. October 2007, Non-lethal Warfare and the Chemical
Weapons Convention, Further Harvard Sussex Program submission to the OPCW Open-Ended Working Group on
Preparations for the Second CWC Review Conference, Harvard Sussex Program.
31 Perry Robinson, 2007, op. cit., p. 19.
32 Pearson, 2006, op. cit., p. 172.
33 See, for example: ibid., p. 169; Wheelis, M. and Dando, M. 2005, Neurobiology: a case study of the
imminent militarization of biology, International Review of the Red Cross, vol. 87, no. 859, p. 564.
302
to lethal force, but as a means to make lethal force more deadly. During the
October 2002 Moscow theatre siege, those Chechen hostage takers who were
rendered unconscious by the incapacitant were then reportedly shot where
they lay by Russian forces rather than being arrested.34
Facilitation of torture and other human rights violations: As well as potentially
being utilised for torture and ill treatment of individuals, incapacitants could
also facilitate repression of groups by, for example, allowing the capture, en
masse, of large numbers of people participating in peaceful demonstrations.
Militarisation of biology: Analysts35 have warned that the continuing
utilisation of the life sciences in the development of incapacitants could
potentially open the way to more malign objectives, such as the widespread
repression of entire populations.
Camouflage for lethal chemical weapons programs: States could exploit the
limited transparency mechanisms required under the Chemical Weapons
Convention (CWC), for incapacitants and other toxic chemicals designated
for use in law enforcement, to hide illicit activities.36
Confusion between lethal and nonlethal chemical weapons: A state using
a nonlethal incapacitant during an armed conflict may be perceived by
another party as having used a lethal chemical weapon and thus initiate an
escalating cycle of retaliation leading to actual use of lethal chemical agents.37
Advances in science and technology

In the light of the previous research that has been conducted by certain
states into incapacitants, the potential application of the current advances in
the life sciences, particularly genomics, biotechnology, synthetic biology and
neuroscience,38 to incapacitating weapons development is a cause for concern.
Trapp has highlighted the potential implications of the misuse of such research:
The explosion of knowledge in neuroscience, bioregulators, receptor
research, systems biology and related disciplines is likely to lead to the
discovery, amongst others, of new physiologically-active compounds
that can selectively interfere with certain regulatory functions in the
brain or other organs, and presumably even modulate human behavior
34 Koplow, D. 2006, The Russians and the Chechens in Moscow in 2002, in Non-Lethal Weapons: The Law
and Policy of Revolutionary Technologies for the Military and Law Enforcement, Cambridge University Press,
Cambridge, pp. 10013.
35 BMA, op. cit., p. 1; Perry Robinson, October 2007, op. cit., p. 32; Wheelis and Dando, op. cit., pp.
55371.
36 Perry Robinson, 2007, op. cit., p. 31.
37 Pearson, 2006, op. cit., p. 170.
38 See, for example: The Royal Society 2011, Brain Waves, Module 1: Neuroscience, Society and Policy,
Science Policy Centre, The Royal Society, London; The Royal Society, 2012, op. cit.; Andreasen, op. cit.;
Society of Neuroscience, op. cit.
303
in a predictable manner. Some of these new compounds (or selective

delivery methods) may well have a profile that could make them
attractive as novel candidate chemical warfare agents.39
Wheelis and Dando had previously surveyed developments and future trends
in neurobiology and concluded that there were indications that military interest
was already directed towards the next generation of substances affecting the
brain and central nervous system:
In addition to drugs causing calming or unconsciousness, compounds
on the horizon with potential as military agents include noradrenaline
antagonists such as propranolol to cause selective memory loss,
cholecystokinin B agonists to cause panic attacks, and substance P
agonists to induce depression. The question thus is not so much when
these capabilities will arisebecause arise they certainly willbut
what purposes will those with such capabilities pursue.40
Indeed in 2005, Boken, a toxicologist with the Croatian Ministry of Defence,
published a paper warning of the potential use of a range of bio-regulators in
warfare and terrorist activities:
Recent years have seen a rapid advance in the discovery of new
bioregulators, especially of the incapacitating ones, in the understanding
of their mode of action and synthetic routes for manufacture. Some of
these compounds may be many hundreds of times more potent than the
traditional chemical warfare agents. Some very important characteristics
of new bioregulators that would offer significant military advantages
are novel sites of toxic action; rapid and specific effects; penetration
of protective filters and equipment, and militarily effective physical
incapacitation.41
Advances in discovery or synthetic production of potential incapacitating
agents have occurred in parallel with developments in particle engineering and
nanotechnology that could allow the delivery of biologically active chemicals
to specific target organs or receptors. The implications of this were highlighted
in the 2008 report by the National Research Council on Emerging Cognitive
Neuroscience and Related Technologies,42 which warned that nanotechnologies
could be used to overcome the bloodbrain barrier and thereby enable
39 Trapp, R. 2010, Incapacitating chemical agents: some thoughts on possible strategies and
recommendations, in ICRC, op. cit., p. 65.
40 Wheelis and Dando, op. cit., p. 10.
41 Boken, S. 2005, The toxicology of bioregulators as potential agents of bioterrorism, Arh Hig Tokiskol,
vol. 56, pp. 20511.
42 National Research Council 2008, Emerging Cognitive Neuroscience and Related Technologies, The National
Academies Press, Washington, DC.
304
unparalleled access to the brain. Nanotechnologies can also exploit existing

transport mechanisms to transmit substances into the brain in analogy with
the Trojan horse.43 The report also highlighted the potential threats resulting
from developments in nanotechnologies or gas-phase techniques that allow
dispersal of highly potent chemicals over wide areas. It noted that at present
pharmacological agents are not used as weapons of mass effect, because their
large-scale deployment is impractical as it is currently impossible to get an
effective dose to a combatant. The report states, however, that technologies
that could be available in the next 20 years would allow dispersal of agents in
delivery vehicles that would be analogous to a pharmacological cluster bomb
or a land mine.44 Despite the interest in the development of incapacitants by
certain states, and the ongoing advances in relevant science and technology
with dual-use application, the international governmental and scientific
communities currently seem unwilling or unable to establish and implement
effective regulatory controls in this area.
The following sections of this chapter seek to apply a holistic arms-control
approach to this issue in order to explore potential routes for effective regulation
of such agents.
Applying a holistic arms-control approach to

the regulation of incapacitants
For many years the governmental and non-governmental arms-control
communities have sought to develop strategies to combat the proliferation of
chemical and biological weapons to state and non-state actors. Recognising
that reliance upon a single disarmament or arms-control agreement alone
would not guarantee success, scholars have explored a number of concepts
seeking to broaden the range of possible regulatory mechanisms. Utilising
and building upon such work, particularly the concepts of preventative arms
control45 and webs of prevention (or protection),46 the author has sought to
develop a holistic arms control (HAC) framework for regulation. Although the
proposed HAC analytical framework concentrates upon existing arms-control
43 Ibid., p. 135.
44 Ibid., p. 137.
45 See, for example: Altmann, J. 2006, Preventive arms control: concept and design, in J. Altmann,
Military Nanotechnology: Potential Applications and Preventive Arms Control, Routledge, London.
46 See, for example: Pearson, G. 1998, The vital importance of the web of prevention, in Proceedings of
the Sixth International Symposium on Protection against Chemical and Biological Warfare Agents, Stockholm,
1115 May; Pearson, G. 2001, Why biological weapons present the greatest danger, Seventh International
Symposium on Protection against Chemical and Biological Warfare Agents, Stockholm; McLeish, C. and
Rappert, B. (eds) 2007, A Web of Prevention: Biological Weapons, Life Sciences, and the Governance of Research,
Earthscan, London.
305
and disarmament measures,47 it attempts to widen the range of applicable

mechanisms for regulation, and also the nature of the actors involved in such
regulatory measures.
Consequently, HAC can be thought of as a framework for analysis to aid the
development of a comprehensive, layered and flexible approach to arms control
that
is developed for and unique to the specific type of weapon or technology
under consideration rather than for a broad grouping of weapons
is potentially applicable to all stages of a weapons existence (that is, research,
development, mass production, stockpiling, deployment, use, transfer and
destruction)
is responsive to developments in science and technology (and will be able to
regulate weapons that have not yet been invented)
seeks to identify the types of permissible and non-permissible weapons,
acknowledges where ambiguity lies, and seeks to develop mechanisms for
resolving such ambiguities
seeks to clearly identify existing constraints upon the permitted use of
weaponsthat is, legitimate targets (for example, whether this would
include armed combatants, terrorist organisations, criminals, civilians),
legitimate types of operations (for example, whether this would include law
enforcement, military operations other than war, armed conflict) and how
such operations should be conducted (rules of engagement); acknowledges
where existing ambiguity lies and highlights potential mechanisms for
resolving such ambiguities
is responsive to developments in the nature of the use/misuse of weapons in
practice
is not necessarily limited to a single existing arms-control or disarmament
treaty, but actively explores and seeks to incorporate states existing
responsibilities under the full range of relevant international law and
applicable agreements
is responsive to developments in international law, particularly those limiting
types and use of weapons
incorporates measures to facilitate effective national implementation as well
as verification, enforcement and transparency mechanisms
47 This chapter does not, therefore, explore a range of parallel processes that can potentially play important
roles in preventing or ameliorating the effects of incapacitant weapons attack, including broadband defence
measures, strengthening public health surveillance and response, and so on. For further discussion of these
issues, see, for example, Pearson, 1998, op. cit.; Pearson, 2001, op. cit.; McLeish and Rappert, op. cit.
306
recognises that states are the prime actors in existing regulatory regimes,
and allows for and encourages participation by the full range of relevant
stakeholders.
Applying this approach to the case of incapacitants, a HAC regime can be
envisioned comprising
state-led activities
adherence to comprehensive legal prohibitions against chemical and biological
weapons (CBW) enshrined in the Geneva Protocol, the Biological and Toxin
Weapons Convention (BTWC) and the Chemical Weapons Convention (CWC)
adherence to international humanitarian law (notably, the four Geneva
conventions and two additional protocols) and international human rights
law (including the Convention against Torture, the International Covenant
on Civil and Political Rights and the Universal Declaration on Human Rights)
adherence to other relevant international law and agreements including the
Rome Statute of the International Criminal Court, the Single Convention on
Narcotic Drugs and the UN Convention on Psychotropic Substances
effective monitoring, verification, investigation and enforcement of the above
obligations
application of stringent export controls and interdiction measures.
engagement by civil society
conducting societal monitoring and verification
developing a culture of responsibility amongst the scientific and medical
communities built upon strong normative and ethical standards
developing and advocating mechanisms to strengthen the regime.
Whilst certain authors have previously examined the application of a range of
state-led mechanisms for the regulation of incapacitants,48 the importance of the
life-science communitys engagement in this issue has been underexplored. In
the following sections, the author will, therefore, briefly examine the potential
application of the two most pertinent arms-control regimesthe Chemical
Weapons Convention and the Biological and Toxin Weapons Convention
before concentrating upon the potential roles that civil society, and particularly
48 See, for example: Crowley, op. cit.; Casey-Maslen, S. 2010, Non-Kinetic-Energy Weapons Termed NonLethal: A Preliminary Assessment under International Humanitarian Law and International Human Rights
Law, Geneva Academy of International Humanitarian Law and Human Rights, Geneva; ICRC, op. cit.;
International Committee of the Red Cross (ICRC) 2012, Toxic Chemicals As Weapons for Law Enforcement:
A Threat to Life and International Law? Synthesis paper, September, International Committee of the Red
Cross, Geneva; International Committee of the Red Cross (ICRC) 2013, Incapacitating Chemical Agents: Law
Enforcement, Human Rights Law and Policy Perspectives, Expert Meeting, Montreux, Switzerland, 2426 April
2012, ICRC, Geneva, 13 January 2013.
307
the scientific and medical communities, can play in preventing the misuse of
biomedical research for hostile purposes, most notably, the development of
incapacitating weapons.

Article I of the BTWC declares that:
Each State Party to the Convention undertakes never in any circumstances
to develop, produce, stockpile or otherwise acquire or retain:
1. Microbial or other biological agents, or toxins, whatever their origin
or method of production, of types and in quantities that have no
justification for prophylactic, protective or other peaceful purposes.
2. Weapons, equipment or means of delivery designed to use such agents
or toxins for hostile purposes or in armed conflict.49
Article I, together with the extended understandings agreed at successive BTWC
review conferences,50 makes it clear that the convention is comprehensive in its
scope and that all naturally or artificially created or altered microbial and other
biological agents and toxins, as well as their components, regardless of their
origin and method of production are covered. Because some possible candidate
incapacitants, such as bio-regulators including neurotransmitters, could be
considered biological agents or toxins, a range of such incapacitants would be
covered by the BTWC.
Although the BTWC does appear to cover certain incapacitants, there are
ambiguities regarding the nature and scope of such coverage. For example,
although the use of incapacitants of a biological origin in armed conflict or for
hostile purposes would be banned, the delineation between prohibited hostile
purposes and permitted peaceful purposes has not been fully established
under the convention. Consequently, it is unclear how the use of incapacitants
of biological origin for military operations other than war (MOOTW), would
49 Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological)
and Toxin Weapons and their Destruction, 1972, 1015 UNTS 163, art. 1.
50 See, for example, Seventh BWC Review Conference Final Document (2012): 1. Conference declares that
the Convention is comprehensive in its scope and that all naturally or artificially created or altered microbial
and other biological agents and toxins, as well as their components, regardless of their origin and method
of production and whether they affect humans, animals or plants, of types and in quantities that have no
justification for prophylactic, protective or other peaceful purposes, are unequivocally covered by Article
I. 2. The Conference reaffirms that Article I applies to all scientific and technological developments in the
life sciences and in other fields of science relevant to the Convention 3. The Conference reaffirms that the
use by the States Parties, in any way and under any circumstances, of microbial or other biological agents or
toxins, that is not consistent with prophylactic, protective or other peaceful purposes, is effectively a violation
of Article I. United Nations 2012, Final Document of the Seventh Review Conference, Geneva, 522 December
2011, BWC/CONF.VII/7, 13 January 2012, art. I, paras 13.
308
be regulated by the BTWC. To date, there have been no determinations of these

issues by the BTWC states parties.51 Further important limitations on the value
of the BTWC (and its control regime) as a tool to regulate incapacitants arise
from its current lack of effective verification and compliance mechanisms, and
also the absence of an international organisation that could coordinate such
activities and facilitate implementation by states parties.52

Although certain states in their background scientific papers to the seventh
BTWC Review Conference highlighted the potential dangers of the misuse of
biologically active agents such as bio-regulators and peptides that could be
used as incapacitants, currently discussions on the regulation of such weapons
have largely concentrated on the Chemical Weapons Convention (CWC). The
CWC prohibits the development, production, stockpiling, transfer and use of
chemical weapons.53 In addition, it requires that all existing stocks of chemical
weapons54 and chemical weapons production facilities be destroyed.55 The treaty
is of unlimited duration and is designed to be far more comprehensive in scope
and application than any prior international agreement on chemical weapons.
It is overseen by its own treaty body, the Organisation for the Prevention of
Chemical Weapons (OPCW), including a technical secretariat of more than
500 inspectors, scientists, legal experts and ancillary staff headed by the
director-general, which carries out the daily work of monitoring, verifying and
facilitating implementation of the convention.
Although the convention prohibits chemical weapons, it allows for the controlled
peaceful use of toxic chemicals. Article II.2 of the convention defines a toxic
chemical as any chemical, regardless of its origin or method of production,
which, through chemical action on life processes, can cause death, temporary
51 Analysis was undertaken of all relevant documents pertaining to this issue publicly available up to 10
September 2013.
52 Although there is no equivalent of an OPCW for the BTWC, the Sixth BTWC Review Conference
decided to create and fund a (three-person) Implementation Support Unit (ISU) within the UN Office for
Disarmament Affairs (UNODA) of the UN Office at Geneva. The ISU was launched in August 2007 and its
mandate was renewed and extended by the Seventh BTWC Review Conference to run until 2016. The ISU
provides administrative support to, and prepares documentation for, meetings agreed by the BTWC Review
Conference. The ISU also facilitates communication among states parties, international organisations, and
scientific and academic institutions, as well as NGOs. It also acts as a focal point for submission of information
by and to states parties, and will support, as appropriate, the implementation by the states parties of the
decisions and recommendations of the Sixth and Seventh BTWC Review Conferences. The ISU, however, has
no authority to undertake verification or compliance activities.
53 Convention on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and
on their Destruction (Chemical Weapons Convention), 1993, art. I.1.
54 Ibid., art. I.3.
55 Ibid., art. I.4.
309
incapacitation or permanent harm to humans or animals.56 The convention

therefore covers a wide range of chemicals within its scope of regulation
including certain chemical agents that could be used as incapacitants.
To determine whether the use of a toxic chemical such as an incapacitant would
be in conformity with the CWC, the intention or purpose for its use needs to be
determined. Under Article II.1 of the convention, chemical weapons are defined
as toxic chemicals or their precursors, except where intended for purposes not
prohibited by the Convention, as long as the types and quantities are consistent
with such purposes.57
Such purposes not prohibited are defined under Article II.9 and include:
(c) Military purposes not connected with the use of chemical weapons
and not dependent on the use of the toxic properties of chemicals as a
method of warfare;
Law enforcement including domestic riot control purposes.58
It is therefore clear that the use of toxic chemicals such as incapacitants for
purposes not provided for in Article II.9 (for example, as a method of warfare)
would be prohibited, as would development, production, acquisition,
stockpiling, retention or transfer of these chemicals for such purposes (under
Article I of the CWC).
There are, however, a number of ambiguities in the CWC and limitations in
its current implementation that could seriously restrict its ability to effectively
regulate incapacitants. Although the use of toxic chemicals is permitted for
law enforcement as long as the types and quantities are consistent with such
purposes,59 there is no definition of law enforcement in the convention.
Furthermore, no OPCW policymaking organ has made any interpretative
statements elaborating the scope or nature of permitted law-enforcement activities
or regarding which toxic chemicals (if any)save riot-control agentscould
be used for such purposes. Consequently, the extent to which incapacitants
could be used (if at all) for activities such as counterterrorist operations is
contested.60 Unfortunately, and despite attempts made by a number of CWC
states parties to raise this issue at the 2013 Third Review Conference,61 there
310
56 Ibid., art. II.2.

60 For divergent argumentation on this issue, see: Fidler, D. 2007, Incapacitating chemical and biochemical
weapons and law enforcement under the Chemical Weapons Convention, in Pearson et al., op. cit.; von
Wagner, A. 2007, Toxic chemicals for law enforcement including domestic riot control purposes under the
Chemical Weapons Convention, in Pearson et al., op. cit.
61 See, in particular, the statements by Germany, Switzerland, the United Kingdom and the United States
made on 8 and 9 April 2013 in the General Debate of the Third Special Session of the Conference of the States
Parties to Review the Operation of the Chemical Weapons Convention (Third Review Conference), The Hague,
819 April 2013, all available from the OPCW web site: <www.opcw.org>.
has been a collective failure by the CWC states parties and policymaking organs
to effectively address the regulation of incapacitants under the convention. It
is therefore left to individual states parties to interpret the scope and nature of
their obligations with regard to the regulation of such agents.
Box 18.3 The ICRC and Incapacitants: A call to action
A range of civil society organisations and researchers has urged the international governmental
communityparticularly the OPCW and the CWC states partiesto take action to address
the dangers of the potential proliferation and inappropriate use of incapacitants.a The ICRC,
however, has noted that [a]lthough there have been exhortations by some States and by
some in academic circles to address these issues, there has been little or no movement to
date in the relevant multilateral fora.b Consequently in March 2010, the ICRC convened the
first of two expert meetings to explore the implications of these issues for international law.
The meeting brought together a group of 33 government and independent experts who were
joined by ICRC staff members.c For details of the second ICRC meeting, see: ICRC, 2013,
op. cit. In the subsequent report of the meeting, the ICRC urged [s]tates to give greater
attention to the implications for international law of incapacitant chemical agents. The
organisation also noted that [t]here is currently an opportunity to address preventatively
the challenges and risks identifiedd (emphasis added). In addition, the report concluded that
[t]here is a clear need to tackle the issues raised by incapacitating chemical agents in
appropriate fora engaging a broad range of experts including policy makers, law-enforcement
professionals, security personnel, military personnel, health professionals, scientists and
lawyers with IHL [international humanitarian law], human rights and disarmament expertisee
(emphases added).
See, for example: Crowley, op. cit., esp. pp. 57101, 11719; Pearson et al., 2007, op. cit.; Perry
Robinson, 2007 and October 2007, op. cit.
a
ICRC, 2010, op. cit., p. 7.
Ibid., p. 3.
ICRC, 2010, op. cit., p. 75.
Ibid., p. 75.
Engagement by the medical and scientific

communities
Given the evident ambiguities and limitations of the existing state-centric
chemical and biological weapons control regimes, compounded by their
inadequate implementation by certain states parties, a great responsibility
lies upon the life-science community as a whole and individual researchers in
particular to ensure that their activities do not contribute to the development of
a new generation of chemical and biological weapons, including incapacitants.
Furthermore, it can be argued that there is an obligation upon life scientists to
move beyond such personal regulation and to take a far more proactive role
in monitoring and regulating incapacitating agent research and combating the
potential misuse of such agents. This section will explore potential avenues for
greater engagement and action through the application of societal verification as
311
a complement to the existing official verification mechanisms, the development

of a culture of responsibility amongst the scientific and medical communities,
and finally explore the possible roles that scientists, academics and other civilsociety actors can play in informing and influencing the actions of states in this
area.
Societal monitoring and verification

Although there is no agreed formal definition of societal monitoring and
verification, Diseroth describes societal verification as
connot[ing] the involvement of civil society in monitoring national
compliance with, and overall implementation of, international treaties
or agreements. One important element is citizens reporting of violations
or attempted violations of agreements by their own government or
others in their own country A more recent development is civil
society monitoring of global compliance with international agreements.
In contrast to official verification organisations employing professional
experts, societal verification may involve the whole of society or groups
within it.62
The establishment of a global societal monitoring and verification network
involving large numbers of civil society actors resident in all states party to
relevant chemical or biological treaties who are able to monitor their states
implementation of treaty obligationsappears to be unlikely in the near to
medium terms. As Rotblat acknowledges, [e]ven if governments were persuaded
to pass laws to make reporting legitimate, which itself would be a revolutionary
development and counter to existing practice in arms control and disarmament
policy and the practice of many states, this goes so much against traditional
loyalties that it would require a considerable educational effort to induce people
to act on it voluntary.63 Consequently, he believes that implementation of
societal verification requires a change in certain attitudes of the general public,
which may take time.64
A more limited form of societal verification, however, can be envisaged,
comprising a smaller number of activist researchers who have access to the
relevant technical expertise and can, at the very least, undertake open-source
monitoring and analysis, and potentially conduct field missions. Due to resource,
personnel, political and security constraints, such groups are likely to be limited
62 Deiseroth, D. 2000, Societal verification: wave of the future? in VERTIC, 2000 Verification Yearbook, p.
265.
63 Rotblat, J. 1993, Societal verification, in J. Rotblat, J. Steinberger and B. Udgaonkar, A Nuclear-WeaponFree World: Desirable? Feasible? Westview Press, Boulder, Colo., p. 108.
64 Ibid., p. 105.
312
in terms of the countries from which they can operate and consequently the
quantity and quality of information they are able to receive, particularly from
inaccessible regions and closed or semi-closed authoritarian countries.
Open-source monitoring and analysis

A small number of academic and non-governmental organisations65 have
undertaken monitoring of open-source data, often utilising a range of national
oversight and transparency mechanisms, to obtain information relating to the
research, development and utilisation of incapacitants.66 Such open-source
monitoring and analysis are time-consuming, resource intensive and the
information obtained is often limited as a result of national security restrictions,
commercial confidentiality considerations and limited access to research
published in certain countries. In addition, there is much inaccurate or biased
reporting disseminated by both proponents and opponents of such weapons.
Despite the methodological difficulties and the limitations in the information
obtained, such work is vital to the formation of an informed public discourse
on the existing threats and potential dangers of the proliferation and misuse
of these weapons. In addition it can also help in the development of timely
and realistic publicly available threat assessments relating to R&D, deployment
or utilisation of such weapons in specific countries. Furthermore, information
derived from civil society research can be sent to relevant intergovernmental
organisations, most notably the OPCW.
Field missions and witness testimony

Independent scientists, health professionals and non-governmental
organisations (NGOs) can sometimes collect their own information, first hand,
from onsite investigations or may be able to utilise information (for example,
witness testimony) and analyse materials (for example, weapons shells, clothing
fragments, soil samples) obtained from other civil society actors operating in
the field (for example, journalists, national NGOs). There are several potential
constraints upon such investigations including access, logistics and translation;
safety considerations for researchers and witnesses; and difficulties ensuring
chain of custody, as well as establishing the representativeness of the information
65 See, for example, the publications of: Biological Weapons Prevention Programme, Bradford Non-Lethal
Weapons Research Project; Centre for Arms Control and Non-Proliferation; Federation of American Scientists;
Harvard Sussex Project; International Network of Engineers and Scientists for Global Responsibility; Pugwash
Conferences on Science and World Affairs; and the Sunshine Project.
66 For discussion, see: Crowley, M. 2010, Monitoring and opposing the misuse of incapacitantsexploring
the potential roles for independent scientists, in J. Finney and V. Slaus (eds), Assessing the Threat of Weapons
of Mass Destruction, Nato Science for Peace and Security Series E, Human and Societal DynamicsVolume
61, IOS Press, Amsterdam, pp. 11432.
313
obtained. Despite such constraints, material collected during field missions

can provide information that could not be obtained by any other meansfor
example, allowing identification of toxic chemical agents utilised during a
military or law-enforcement operation.
Box 18.4 Independent Analysis of a New Riot-Control Agent Used in the
West Bank
In July 2005, the Israeli Army reportedly employed a new riot-control agent against Palestinian
and Israeli civilian protesters that resulted in severe skin injuries. The Israeli Army refused
to identify the agent; however, scientists based in the United Kingdom obtained one of the
munitions utilised, and following physical and chemical analyses, were able to identify the
contents as capsaicin with an inert carrier and a dispersal agent.a The results were found to
correspond with the commercially available Pepperball Tactical Powder. The paper noted
that [s]kin injuries of the severity described had not previously been reported with this
agent, and would be difficult to manage for clinicians who were unaware of the nature of the
agent.b As well as alerting clinicians to the nature and effects of chemical agents they may
face in the future, such research can also help to identify possible international transfers of
chemical agents and devices.
Hay, A., Giacaman, R., Sansur, R. and Rose, S. 2006, Skin injuries caused by new riot control agent
used against civilians on the West Bank, Medicine,Conflict and Survival, vol. 22, no. 4 (October
December).
a
Ibid.
Building a culture of responsibility within the lifescience and biomedical communities

In its 2004 public statement Preventing hostile use of the life sciences, the
ICRC declared: If measures to prevent the hostile use of advances in the life
sciences are to work, a culture of responsibility is necessary among individual
life scientists. This applies whether these scientists are working in industry,
academia, health, defense or in related fields such as engineering and information
technology.67
According to the ICRC, such a culture of responsibility is also needed within
the institutions that employ scientists and fund research in the life sciences.68
Similar calls to the scientific and medical communities have also been made by
the states parties to the BTWC and the CWC at review conferences.69
67 International Committee of the Red Cross (ICRC) 2004, Preventing Hostile Use of the Life Sciences: From
Ethics and Law to Best Practice, 11 November 2004, International Committee of the Red Cross, Geneva,
<http://projects.exeter.ac.uk/codesofconduct/ Chronology/Principles_Actionpoints_11Nov04.pdf> (viewed
14 January 2012).
68Ibid.
69 For example, see: Third Review Conference of the Parties to the Convention on the Prohibition of the
Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their
Destruction, Final Document, Part II, BWC/CONF.III/23, 92 September 1991, p. 3; Report of the Second Special
Session of the Conference of the States Parties to Review the Operation of the Chemical Weapons Convention
(Second Review Conference), 718 April 2008, RC-2/4, 18 April 2008.
314
The remaining sections of this chapter will explore the current range of
initiatives being undertaken by those in the scientific and medical communities
to nurture a culture of responsibility, beginning with the growing recognition
of the dual-use dilemma and the consequent requirement for effective oversight
of research. This will then be followed by a discussion of the potential utility of
oaths, codes and pledges and the parallel processes of education and awarenessraising in building the appropriate norms of behaviour for the scientific and
biomedical communities. The practical application of such principles by
individual scientists through such practices as whistleblowing will be explored
as well as the duty of individual scientists to inform the policies and practices
of governments in this area.
Research oversight and the dual-use dilemma
To date, much of the discourse amongst the life-science community concerning
how best to combat the proliferation and misuse of chemical and biological
weapons has concentrated on tackling the dual-use dilemma and has been
framed in terms of regulating the actions of individual life scientists conducting
academic research projects and publishing academic articles. Highly
influential in this discourse have been the 2003 Fink Report (Biotechnology
Research in An Age of Terrorism)70 and the 2006 Lemon Report (Globalisation,
Biosecurity and the Future of the Life Sciences), both produced under the auspices
of the National Research Council of the US National Academies. Both reports
highlighted the importance of taking a comprehensive approach to analysing
dual-use research of potential concern. The Lemon Report recommended
the adoption of a broadened awareness of threats beyond the classical select
agents and other pathogenic organisms and toxins, so as to include, for
example, approaches for disrupting host homeostatic and defense systems and
for creating synthetic organisms.71 The broad threat spectrums enunciated
by both reports, particularly that of the Lemon Committee, appear to capture
incapacitants within their scope.
As a result of the concerns and recommendations outlined in the Fink and Lemon
reports and the work of others,72 a range of oversight structures and processes
70 National Research Council 2004, Biotechnology Research in an Age of Terrorism, [Fink Report], The
National Academies Press, Washington, DC, p. 114.
71 National Research Council 2006, Globalisation, Biosecurity and the Future of the Life Sciences, [Lemon
Report], The National Academies Press, Washington, DC, p. 216.
72 A range of analysts has subsequently highlighted the need to address further potential actors of concern
beyond individual life-science researchers. Garfinkel et al., for example, defined three major points for
potential policy interventionnamely: commercial firms that sell synthetic DNA (oligonucleotides, genes or
genomes) to users; owners of laboratory bench-top DNA synthesizers, with which users can produce their
own DNA; the users (consumers) of synthetic DNA themselves and the institutions that support and oversee
their work. See: Garfinkel, M., Endy, D., Epstein, G. and Friedman, R. 2007, Synthetic Genomics: Options for
Governance, J. Craig Venter Institute, Rockville, Md, and San Diego, Calif.
315
was established by governments, scientific bodies, academic institutions,

funders and publishers to review potential dual-use research, assessing the risks
and benefits of such research to determine whether they need to be modified or
withdrawn.73
Analysing the application of oversight measures in practice, however, Rappert
believes that such procedures rarely conclude that manuscripts, grant
applications or experiment proposals should not be undertaken or restricted.
In 2009, van Aken and Hunger analysed the application of biosecurity policies
agreed by a group of 32 influential science journals under which manuscripts
could be modified or rejected where the potential harm of publication outweighs
the potential societal benefits.74 Despite such policies having been established
in 2003, van Aken and Hunger found that no manuscript has ever been rejected
on security grounds.75 Rappert believes the same could be said of those funders
who have established submission-oversight systems.76 Furthermore, Rappert
believes that even more notable with these review processes is the infrequency
with which they have identified items of concern in the first place.77 Whilst
information relating to the research controls of government departments
(especially defence-related ones) is not readily available, Rappert believes that in
relation to universities and other publicly funded agencies, it seems justifiable
to conclude thatbarring dramatic changesoversight processes will identify
little research as posing security concerns and will stop next to nothing.78
Others have criticised the voluntary nature of the existing controls on lifescience dual-use research. For example, commenting upon the release of a draft
of the National Science Advisory Board for Biosecurity (NSABB) Proposed
Framework for the Oversight of Dual Use Life Sciences Research: Strategies
for Minimizing the Potential Misuse of Research Information,79 the Sunshine
Project director, Hammond, stated that the NSABB is divorced from reality if
its members believe that another set of voluntary NIH [National Institutes of
Health] guidelines is sufficient, and would be remotely effective, at preventing
73 For a discussion of such initiatives, see: Rappert, B. 2008, The benefits, risks, and threats of
biotechnology, Science & Public Policy, vol. 35, no. 1, pp. 3744.
74 Journal Editors and Authors Group 2003, Uncensored exchange of scientific results, Proceedings of the
National Academy of Sciences, vol. 100, no. 4, p. 1464.
75 van Aken, J. and Hunger, I. 2009, Biosecurity policies at international life-science journals, Biosecurity
and Bioterrorism, vol. 7, no. 1, pp. 6172.
76 Rappert, B. 2010, Introduction: education as, in B. Rappert (ed.), Education and Ethics in the Life
Sciences: Strengthening the Prohibition of Biological Weapons, ANU E Press, Canberra, pp. 89. Rappert
cites the UK Biotechnology and Biological Sciences Research Council, the UK Medical Research Council,
the Wellcome Trust, the Centers for Disease Control, and the Southeast Center of Regional Excellence for
Emerging Infectious Diseases and Biodefence.
77 Ibid., p. 9.
78 Ibid., p. 9.
National Science Advisory Board for Biosecurity, Bethesda, Md.
316
dual-use disasters.80 Research conducted by the Sunshine Project from 2004

to 2007 indicated that many US organisations obliged to follow NIH guidelines
did not do so.81 A 2007 Sunshine Project survey discovered that 18 of the top20 US biotechnology companies did not comply with existing voluntary NIH
biotechnology guidelines.82 Instead of a voluntary approach, Hammond stated
that [e]ffective federal management of dual-use risks requires making safety
and security oversight truly mandatory and subject to the sobering light of
public scrutiny.83
As well as concerns about the implementation of such voluntary oversight
systems in practice, a further concern relates to the limited range of issues being
considered by such bodies. The discourse and much of the current activity
appear to be concentrated upon preventing the diffusion of dual-use knowledge,
skills and materials to various non-state actors with malign intentprincipally,
terrorist organisations. Insufficient attention has been given to utilising existing
dual-use monitoring mechanisms or adopting additional processes to specifically
combat the misuse of dual-use expertise in state programs, even though national
CBW research and development programs arguably pose a greater danger to the
CWC and BTWC than non-state actors.
Oaths, codes and pledges for the life-science community
One approach to building a culture of responsibility has been through the
development of a range of non-binding ethical codes, codes of conduct and
oaths or pledges. An early advocate of such activities was Joseph Rotblat, who
declared in his 1995 Nobel acceptance speech that:
The time has come to formulate guidelines for the ethical conduct of
scientists, perhaps in the form of a voluntary Hippocratic Oath At
a time when science plays such a powerful role in the life of society,
when the destiny of the whole of mankind may hinge on the results of
scientific research, it is incumbent on all scientists to be fully conscious
of that role, and conduct themselves accordingly. I appeal to my fellow
scientists to remember their responsibility to humanity.84
The development of ethical codes of conduct became one of the priority areas
during the BTWC inter-sessional process. Subsequently, initiatives supporting
80 Sunshine Project 2007, Earth calling NSABB: voluntary compliance wont work, Press release, 17 April
2007.
81 See, for example: Sunshine Project 2004, Mandate for Failure: The State of IBCs in an Age of Bioweapons
Research, <http://www.sunshine-project.org/biodefense/ibcreport.html> (viewed 11 January 2012).
82 Supporting correspondence available at <http://www.sunshine-project.org/publications/pr/support/
deregistries.pdf> (viewed 11 January 2011).
83 Sunshine Project, 2007, op. cit.
84 Rotblat, J. 1995, Remember Your Humanity, Nobel Lecture, Oslo, 10 December 1995, <http://nobelprize.
org/nobel_prizes/peace/laureates/1995/rotblat-lecture.html> (viewed 11 January 2012).
317
such codes have been undertaken by a wide range of scientific associations and
organisations including the American Society of Microbiology,85 the National
Academy of Sciences, The Royal Society,86 the International Centre for Genetic
Engineering and Biotechnology, the International Union of Biochemistry and
Molecular Biology and the International Council for the Life Sciences. These
activities have been complemented and stimulated by the ICRC as well as the
work of individual scientists87 and academics.88
Such codes may well help to sensitise life scientists to the dangers of dual-use
research, and reinforce the importance of, and promulgate, ethical red lines
where the legal prohibitions or normative taboos are already clearly defined and
widely accepted. The effectiveness of such an approach, to date, has, however,
been questioned by a range of scholars including Corneliussen, Dando, Perry
Robinson and Rappert.89 One important limitation of the majority of such
initiatives is that the resulting codes are aspirational and non-binding in nature
with no clearly identified penalties elaborated for those individuals who breach
the prohibitions, or mechanisms established to monitor and enforce such
prohibitions.
Recognising the limitations of self-governance initiatives to effect change
in this area, some have called for codes of conduct to become binding, with
those breaching such codes facing sanction from their peers (or the state). For
example, Rotblat, in a letter to a Pugwash Workshop on Science, Ethics and
Society in 2004, stated his belief that:
[S]ome believe that the search for knowledge overrides all other
considerations and that scientists should be entitled to ignore the
ethical elements of their work The harm to society that has resulted
from such attitudes has brought science into disrepute, and action is
needed to restore the proper image of science. The introduction of a
hippocratic oath is our example of such action, but it should perhaps
be given more than a symbolic value. Perhaps the time has come for a
85 American Society of Microbiology (ASM) 2005, Code of Ethics (Revised and approved by ASM Council in
2005), American Society of Microbiology, Washington, DC.
86 The Royal Society 2005, The Roles of Codes of Conduct in Preventing the Misuse of Scientific Research, The
Royal Society, London.
87 See Atlas, R. and Somerville, M. 2007, Life sciences or death science: tipping the balance towards life
with ethics, codes and laws, in McLeish and Rappert, op. cit.
88 See, for example, Revil, J. and Dando, M. 2006, A Hippocratic oath for life scientists, EMBO Reports,
vol. 7.
89 Rappert, B. 2010, Biological Weapons and Codes of Conduct, <http://projects.exeter.ac.uk/codesofconduct/
Chronology/index.htm> (viewed 22 December 2011).
318
binding code of conduct, where only those who abide by the code should be
entitled to be practicing scientists, something which applies now to medical
practice [emphasis added].90
Examination of the literature, however, reveals that no such binding codes of
conduct prohibiting research, development or utilisation of incapacitating agents
have been established by any national or international scientific organisation to
date.
Furthermore, Rappert has noted that if codes are to go beyond reiterating
platitudes about the abhorrence of using modern biology toward malign ends,
then they are likely to confront major issues of controversy. For instance,
codes could comment on the acceptability of disputed attempts to develop
non-lethal incapacitating agents.91 Areas of dispute or controversy, such as
incapacitants, are the ones, however, where such codes remain silent or at best
provide ambiguous guidance. Similarly, as with dual-use research oversight,
whilst numerous codes condemn and seek to prevent the involvement of
scientists in the development of biological and chemical weapons by non-state
actors, it is questionable whether enough energy has been devoted to targeting
the involvement of life scientists in state-run weapons programs.
Corneliussen has asked
why voluntary self-governance regimesand codes of conduct in
particularare being given so much attention in policy discussions
about preventing the misuse of biological research when they appear to
have significant shortcomings in practice. Indeed, why have individual
scientists become the target of the policy discussions when it is generally
accepted within the disarmament community that the greatest risk of
misuse is at the level of national biological weapons programmes
Preventing these state-level programmes in the future should therefore
be a primary concern, rather than implementing codes of conduct for
life scientists.92
Corneliussen further contends that:
[T]he current sole focus on codes, and the extensive investment of
resources that accompanies it, might well serve to detract from other more
90 Rotblat, J. 2004, Letter to workshop, in Report of 2nd Pugwash Workshop: Science, Ethics and Society,
Ajaccio, Corsica, 1012 September 2004, <http://www.pugwash.org/reports/ees/corsica2004/corsica2004.
htm#attach2> (viewed 22 December 2011).
91 Rappert, B. 2005, Biological weapons and the life sciences: the potential for professional codes,
Disarmament Forum. Volume 1: Science Technology and the CBW Regimes, p. 56.
92 Corneliussen, F. 2006, Adequate regulation, a stop-gap measure, or part of a package? Debates on codes
of conduct for scientists could be diverting attention away from more serious questions, EMBO Reports, vol.
7 (Special Issue), p. 53.
319
crucial regulatory measures that target not only individual scientists but
also state programmes. Without this plurality of regulatory measures in
place, codes of conduct are doomed to fail.93
More recently there have been some indications of an increasing awareness
amongst certain life-science communities of the dangers of cooption into state
programs and initiatives undertaken to address these dangers (see Box 18.5).
Box 18.5 A Pledge for Neuroscientists
In January 2010, Dr Curtis Bell, senior scientist emeritus at the Oregon Health and Science
University, circulated a pledge intended to foster opposition amongst neuroscientists to
the application of neuroscience to torture and other forms of coercive interrogation or
manipulation that violate human rights and personhood.a According to Bell, such applications
could include drugs that cause excessive pain, anxiety, or trust, and manipulations such as
brain stimulation or inactivation.b Furthermore, signatories would oppose the application of
neuroscience to aggressive war illegal under international law. The pledge states that a
government which engages in aggressive wars should not be provided with tools to engage
more effectively in such wars. Neuroscience can and does provide such tools. Examples
include drugs which damage the effectiveness of soldiers on the other side.c
Under the pledge, neuroscientists commit to making themselves aware of the potential misuse
of neuroscience for violations of basic human rights or international law such as torture
and aggressive war and commit to refusing to knowingly participate in the application of
Neuroscience to violations of basic human rights or international law.d Bell acknowledges
that [s]igning this pledge will not stop aggressive wars or human rights violations, or
even the use of neuroscience for these purposes; however, he believes that by signing,
neuroscientists will help make such applications less acceptable.e
Bell, C. 2010a, Pledge by Neuroscientists to Refuse to Participate in the Application of Neuroscience to

Violations of Basic Human Rights or International Law, <http://spreadsheets.google.com/viewform?f
ormkey=dEF4RFhhSWZwNktCakYtbTdkd1cxckE6MA> (viewed 4 December 2011); Bell, C. 2010b,
Letter to International Society for Neuroethology, <http://neuroethology.org/cgi-bin/dada/mail.
cgi/archive/emaillist/20100106211821/> (viewed 4 December 2011); Bell, C. n.d., Responsibilities
of neuroscientists concerning aggressive war and torture, Poster display, <http://files.me.com/
curtiscbell/pu0ycj> (viewed 4 December 2011).
a
Bell, 2010a, op. cit.
Ibid.
Ibid.
Bell, C. 2010c, Neurons for peace: take the pledge, brain scientists, New Scientist, no. 2746
(8 February).
e
Despite the energy and resources expended upon the development and
promotion of codes, however, Rapperts stark conclusion is that efforts to
devise meaningful codes have largely floundered. Rappert believes that [i]n
no small part, this has been due to the lack of prior awareness and attention by
researchers as well as science organisations to the destructive applications of the
life sciences. Before codes can help teach, education is needed.94
93 Ibid., p. 54.
94 Rappert, 2010, op. cit., p. 14.
320
Education and awareness-raising

In 2005, following a series of interactive seminar discussions held with life
scientists at 15 UK universities, Dando and Rappert concluded there was little
evidence that participants regarded bioterrorism or bioweapons as a substantial
threat considered that developments in the life sciences research contributed
to biothreats were aware of the current debates and concerns about dual-use
research; or were familiar with the BWC.95 In 2006, Rappert et al. presented
a report to the Sixth BTWC Review Conference, which stated that despite the
recent international attention given to the problem of the potential misuse of the
life sciences, the initial UK findings were essentially replicated in later seminars
in the United Kingdom and in Finland, Germany, the Netherlands, South
Africa and the United States.96 Subsequent experience by Dando and Rappert
of carrying out seminars in 16 different countries with a few thousand life
scientists in over 110 different departments has consolidated these findings.97
In a 2009 Nature article, Dando highlighted the lack of engagement with this
issue among life scientists, which he considered alarming,98 specifically with
regard to the consequent dangers of the misuse of scientific and technological
advances for the development of incapacitant weapons. Dando and colleagues
have identified the current lack of adequate biosecurity teaching as an important
factor contributing to such low levels of awareness amongst life scientists.
Following their review of the effectiveness of existing education and awarenessraising initiatives, Rappert et al. conclude:
In such circumstances it is quite unrealistic to expect that simply,
for example, adding a lecture to a standard course in the life sciences
will make a great deal of difference in depth implementation of the
BWC within States Parties requires a significant effort on education
and outreach for such implementation to be effective. To achieve this,
a simple declaration as at previous Review Conferences about the
95 Dando, M. and Rappert, B. 2005, Codes of Conduct for the Life Sciences: Some Insights from UK Academia,
Bradford Briefing Paper No. 16 (2nd series), University of Bradford, UK, p. 23.
96 Rappert, B., Chevrier, M. and Dando, M. 2006, In-Depth Implementation of the BTWC: Education and
Outreach, Report presented at Sixth BTWC Review Conference, University of Bradford, UK, para. 89.
97 Whitby, S. and Dando, M. 2010, Biosecurity awareness-raising and education for life scientists: what
should be done now? in B. Rappert (ed.), Education and Ethics in the Life Sciences: Strengthening the Prohibition
of Biological Weapons, ANU E Press, Canberra. For a detailed discussion, see: Rappert, B. 2007, Biotechnology,
Security and the Search for Limits: An Inquiry into Research and Methods, Palgrave, Basingstoke, UK; Rappert,
B. 2009, Experimental Secrets: International Security, Codes, and the Future of Research, University Press of
America, Lanham, Md.
98 Dando, M. 2009, Biologists napping while work militarized, Nature, vol. 460, no. 7258, p. 951. Dandos
concerns are echoed in an accompanying Nature editorial entitled: A question of control: scientists must
address the ethics of using neuroactive compounds to quash domestic crises.
321
importance of education will be insufficient and States Parties will need

to take concerted action to ensure increased educational provision and
outreach.99
Similarly, Whitby and Dando believe that correcting this deficiency in educationand awareness-levels of life scientists will be a massive task.100 It is one that
will require action by a broad range of constituencies involved in life-science
education including, governments, bodies responsible for the administration of
standards in higher education, funders of life-science education, civil society
groups and NGOs involved in the production of educational material, and
teachers and trainers.101
Whilst a number of albeit relatively small and isolated initiatives are now being
undertaken to educate and raise awareness amongst life scientists regarding
dual-use dilemmas and the potential dangers of research being misused for
chemical or biological weapons development,102 there do not appear to be any
sustained activities designed to foster greater awareness and knowledge of such
issues beyond the life-science community. Rappert has noted that [s]cant efforts
made prior to 2001 (and even since) by scientists to popularise how their work
might aid the production of bioweapons indicate the historical pattern of not
seeking to foster wider debate and awareness.103
It is, however, worth considering whether and how those scientists, academics
and educators concerned about incapacitants and who are currently conducting
CBW education and awareness-raising activities can also engage with key civil
society actors in areas such as human rights, humanitarian law and medical
ethics who at present have limited or no knowledge of the dangers posed by the
potential harnessing of advances in chemistry and the life sciences to hostile
purposes. The education and engagement of such expert communities would
enrich and inform the existing discourse concerning incapacitants and broaden
the range of actors seeking effective restrictions of such weapons. In considering
such issues it may be worth exploring the roles of civil society awareness-raising
initiatives and public education in helping to build successful multidisciplinary
coalitions dedicated to addressing complex issues such as the prohibition of
antipersonnel landmines and cluster munitions, addressing climate change, and
promoting the establishment of the International Criminal Court.
99 Rappert et al., 2006, op. cit., paras 89, 90.

100 Whitby and Dando, op. cit., p. 182.
101 For further discussion, see: Dando, M. 2009, Dual-use education for life scientists, Disarmament
Forum, vol. 1, pp. 414.
102 Project on Building a Sustainable Capacity in Dual-Use Bioethics, available from Bradford University
web site, <http://www.brad.ac.uk/bioethics/About/> (viewed 4 December 2011).
103 Rappert, 2010, op. cit., p. 16.
322
Developing and applying ethical standards for health professionals

In addition to life scientists, the participation of physicians and other health
professionals in previous state-run incapacitant weaponisation programs has
been documented in a range of countries including the United Kingdom,
the United States and South Africa.104 More recently, a number of health
professionals in certain countries have voiced support for medical involvement
in the research and development of incapacitant weapons. Anaesthesiologists
who were engaged in the Czech Republics incapacitant weapons development
program have highlighted counterterrorism considerations: many agents used
in everyday practice in anaesthesiology can be employed as pharmacological
non-lethal weapons. An anaesthetist familiar with the pharmacokinetics and
pharmacodynamics of these agents is thus familiar with this use. As a result, he
or she can play a role in combating terrorism.105
The necessity of medical participation for the viability of incapacitant research
has been highlighted by Gross, who believes that chemical incapacitants are
among a limited range of nonlethal weapons that are in effect medicalized in
that they rely on advances in neuroscience, physiology, and pharmacology and
on the active participation of physicians and other medical workers.106
In comparison with the life-science community, amongst certain sectors, at
least, of the medical professional community, ethical discourse regarding the
involvement in development and use of nonlethal chemical weapons appears
to be more advanced. There are a number of ethical codes and declarations that
are potentially applicable and may well constrain the involvement of health
professionals in this area. First, there is a range of declarations and regulations
adopted by the World Medical Association (WMA) that guide health professionals
in situations of conflict and unrest107 and specifically prohibit their involvement
104 See, for example: Maclean, A. 2006, Historical Survey of the Porton Down Volunteer Programme, Ministry
of Defence, London, <http://www.mod.uk/NR/rdonlyres/7211B28A-F5CB-4803-AAAC-2D34F3DBD961/0/
part_iv.pdf> (viewed 17 August 2011), pp. 10942; Ketchum, J. 2006, Chemical Warfare: Secrets Almost
Forgotten, James S. Ketchum (self-published); Gould, C. and Folb, P. 2002, Project Coast: Apartheids Chemical
and Biological Warfare Programme, United Nations Institute for Disarmament Research, Geneva.
105 Hess et al., 2005, op. cit.
106 Gross, M. 2010, Medicalized weapons and modern war, Hastings Center Report, vol. 40, no. 1, pp.
345.
107 World Medical Association (WMA) 1956, Regulations in Times of Armed Conflict, Adopted by the 10th
World Medical Assembly, Havana, Cuba, October 1956, last amended by the 35th World Medical Assembly,
Tokyo, Japan, 2004, and editorially revised at the 173rd Council Session, Divonne-les-Bains, France, May
2006, para. 2.
323
in torture, ill treatment and other forms of human rights abuse.108 The WMA
has also sought to develop ethical guidelines prohibiting medical involvement
in the development of chemical or biological weapons.109
In addition, guidance has also been developed by certain national medical
associations and other medical bodies110 on the ethical considerations
surrounding health professionals involvement in specific weapons research
or weapons development more generally. A number of such bodies have also
established a range of mechanisms and structures to implement ethical standards
including ethics boards that have the authority to suspend or disbar physicians
from practising medicine in cases of extreme misconduct.
There are thus ethical frameworks and mechanisms in place that describe and
regulate the duty of health professionals to abide by and promote aspects of
human rights and international humanitarian law, and specifically prohibit
engagement in acts such as torture or the development of biological and
toxin weapons. Whilst these and other ethical standardsparticularly those
concerned with medical research involving human subjects111can in theory be
applied to the development and utilisation of incapacitants, this does not appear
to have occurred in a consistent manner to date. There are no widely accepted
guidelines specifically determining the permissibility or non-permissibility of
physician involvement in the development, testing or utilisation of so-called
nonlethal weapons in general and incapacitants in particular. Indeed the issue
appears, at present, to be both underexplored and contentious, with a spectrum
of opinions held by health professionals and medical ethicists.
Amongst national medical associations, it is the British Medical Association
(BMA) that has taken the lead in the development of ethical guidance for
the health community on the issue of nonlethal chemical weapons, and in
108 World Medical Association (WMA) 1975, Declaration of Tokyo, Adopted by the 29th World Medical
Assembly, Tokyo, Japan, October 1975, and editorially revised at the 170th Council Session, Divonne-lesBains, France, May 2005, and the 173rd Council Session, Divonne-les-Bains, France, May 2006.
109 World Medical Association (WMA) 1990, Declaration on Chemical and Biological Weapons, Adopted by
the 42nd World Medical Assembly, Rancho Mirage, Calif., October 1990, and rescinded at the WMA General
Assembly, Santiago, 2005; World Medical Association (WMA) n.d., Washington Declaration on Biological
Weapons, paras 18 and 19.
110 For example, the US Army textbook of military ethics urges medical professionals to stay in the
business of healing and not hurting, which includes not participating in or contributing to weapons research
and development. Frisina, M. E. 2003, Medical ethics in military biomedical research, in T. E. Beam and L. R.
Sparacino (eds), Military Medical Ethics, vol. 2, Office of the Surgeon General/Borden Institute, Washington,
DC, pp. 53361, as cited in Gross, op. cit., p. 37.
111 See, for example: World Medical Association (WMA) 1964, Declaration of Helsinki, Adopted by the
18th WMA General Assembly, Helsinki, , June 1964, and last amended at the 59th WMA General Assembly,
Seoul, October 2008, <http://www.wma.net/en/30publications/10policies/b3/index.html> (viewed 22 July
2011); Nuremberg Code, 1949, as detailed in Trials of War Criminals before the Nuremberg Military Tribunals
under Control Council Law, vol. 2, no. 10, pp. 1812, US Government Printing Office, Washington, DC, 1949,
available from: Office of Human Subjects Research, US National Institutes of Health, <http://ohsr.od.nih.gov/
guidelines/nuremberg.html> (viewed 22 July 2011).
324
particular incapacitants. In its 2007 publication Drugs As Weapons, which

explored the implications of incapacitant research, development and use,112 the
BMA declared that:
[D]octors should not knowingly use their skills and knowledge
for weapons development for the same reasons that these ethical
considerations oppose doctors involvement in torture and the
development of more effective methods of execution. In other words, the
duty to avoid doing harm rises above, for instance, a duty to contribute
to national security.113
The report specifically recommends that national organisations that represent
healthcare professionals should
[w]ork to promote the norms prohibiting the use of poisons, and therefore
the BWC and the CWC. They should further promote understanding
that the use of drugs as weapons would violate such norms Advocate
against the use of drugs as weapons and not be involved in the training
of military or law enforcement personnel in the administration of drugs
as weapons.114
Despite the activities of the BMA, the issue does not appear to have been
specifically addressed formally by the WMA during its general assembly or
in any other public WMA policy body.115 Given the importance of medical
participation to research, development, testing and utilisation of incapacitant
weapons, the development of clear guidance in this area is needed.
Harvard Sussex draft convention
The Harvard Sussex Project (HSP) has proposed an alternative approach to
addressing individual responsibility and culpability in the development or
misuse of chemical and biological weapons, through the development of criminal
sanctions. Meselson and Robinson have argued that [a]ny development,
production, acquisition, or use of biological and chemical weapons is the result
of decisions and actions of individual persons, whether they are government
officials, commercial suppliers, weapons experts, or terrorists. They contend,
however, that the BTWC and CWC are directed primarily to the actions of states,
and address the matter of individual responsibility to only a limited degree.116
112 BMA, op. cit.
113 Ibid., p. 20.
114 Ibid., p. 24.
115 A review was undertaken of all relevant publicly available World Medical Association documentation
from January 1990 to December 2011.
116 Meselson, M. and Perry Robinson, J. 2004, A draft convention to prohibit biological and chemical
weapons under international criminal law, Fletcher Forum of World Affairs, vol. 28, no. 1 (Winter), p. 57.
325
In order to address this failing, HSP has developed a Draft Convention to

Prohibit Biological and Chemical Weapons under International Criminal Law. If
enacted, the HSP draft convention would make it a crime under international
law for any person knowingly: to develop, produce, acquire, retain, transfer
or use biological or chemical weapons; to order, direct or knowingly render
substantial assistance to those activities; or to threaten to use biological or
chemical weapons.117
Under the HSP draft convention, each state party would be required to establish
jurisdiction with respect to such crimes according to established principles of
judicial law, and where the state has jurisdiction and is satisfied that the facts
warrant such action, to submit those cases to competent authorities for the
purpose of extradition or prosecution. Furthermore, with respect to the actual
use of CBWs, each state party would be required to establish jurisdiction over
all persons found on its territory regardless of their nationality or place of the
offence.118
The HSP draft convention is a civil society initiative that will require considerable
state support for it to be adopted by the international governmental community.
One potential route previously explored by the HSP is for a group of states to
submit the proposed convention or a similar draft in the form of a resolution for
consideration by the UN General Assembly [UNGA], seeking its referral to the
UNGA Sixth (legal) Committee for negotiation of an agreed text.119 Robinson has
stated that [o]nce we are satisfied that the political environment is favourable,
our plan is to convene an international conference that will bring together
policy makers, jurists and exponents of the Draft Convention.120
Nonparticipation/whistleblowing
Any serious attempt by state or non-state actors to develop new or indeed
existing chemical or biological weapons, be they considered nonlethal or
otherwise, would require the involvement of an array of scientists, engineers,
technicians and other ancillary workers. Whilst such staff are essential to
the development, production, stockpiling, transfer and use of such weapons,
they are also potentially capable of blowing the whistle on such research and
development programs through public denunciations, leaking information to
117 For the full text of the present draft of the convention, see <http://www.sussex.ac.uk/Units/spru/hsp/
Harvard-Sussex-Program-draft-convention-Text.html> (viewed 4 June 2010). Also see accompanying Draft
Legal Commentary, <http://www.sussex.ac.uk/Units/spru/hsp/Draft%20Convention%20supporting%20
docs/HSP%20papers/Legal%20Commentary.pdf> (viewed 4 June 2010).
118 Preamble to HSP Convention, <http://www.sussex.ac.uk/Units/spru/hsp/Harvard-Sussex-Programdraft-convention-Text.html> (accessed 4 June 2010).
119 Perry Robinson, J. 2011, Criminalization of biological and chemical armament, CBW Conventions
Bulletin, (Special Issue), February 2011, p. 4.
120 Ibid., p. 5.
326
journalists or by reporting concerns about potential or realised breaches of

national regulations or violations of international treaties directly to the relevant
national or international regulatory bodies.
In his Nobel acceptance speech, Rotblat stated that:
The purpose of some government or industrial research is sometimes
concealed, and misleading information is presented to the public. It
should be the duty of scientists to expose such malfeasance. Whistleblowing should become part of the scientists ethos. This may bring
reprisals; a price to be paid for ones convictions. The price may be very
heavy.121
Deiseroth highlights the particular vulnerability of whistleblowers:
Compared with normal citizens, employees are in a special situation
because they owe their employer a certain loyalty and, by law, are
normally not allowed to disclose internal or confidential information.
Whistle-blowers, therefore, need protection if they make a disclosure in
good faith and on the basis of reliable evidence.122
Similarly, Falter notes that it is neither realistic nor legitimate to put the full
burden of whistle-blowing and potential retaliation on individual scientists and
their moral sensibilities.123 Indeed, whilst it is the duty of individual scientists to
make known their concerns about the misuse of scientific research for activities
that breach ethical standards or international law, it is the responsibility of
the scientific community as a whole to ensure that such whistleblowers are
fully protected. This was recognised by the ICRC in their 2004 statement,
which declared that [t]hose working in life sciences who voice concern and
take responsible action require and deserve political and professional support
and protection, and the corresponding action point, which was to ensure
that adequate mechanisms exist for voicing such concerns without fear of
retribution.124
Although a number of states such as South Africa,125 the United Kingdom126
and the United States127 have legislation relating to whistleblowing activities
121 Rotblat, 1995, op. cit.
122 Deiseroth, op. cit., p. 266.
123 Falter, A. 2010, Including civil society into confidence building: protecting whistleblowers and societal
verification, in Finney and Slaus, op. cit., p. 289.
124 ICRC, 2004, op. cit.
125 Government of South Africa 2000, Protected Disclosures Act, <http://www.workinfo.com/free/Sub_for_
legres/data/Disclosure/protected.htm> (viewed 23 January 2012).
126 Government of the United Kingdom 1998, Public Interest Disclosure Act, available on the UK National
Archives web site: <http://www.legislation.gov.uk/ukpga/1998/23/contents> (viewed 23 January 2012).
127 Government of the United States of America 1989, US Federal Whistleblower Protection Act, 5 USC, s. 1201.
327
on their statute books, the effectiveness of such legislation and its enforcement
are variable.128 Furthermore, Martin, who has long experience of working with,
and seeking to protect, whistleblowers in many different spheres, believes that
the track record of whistle-blower protection measureswhistle-blower laws,
hot-lines, ombudsmen and the likeis abysmal. In many cases, these formal
processes give only an illusion of protection. Codes of ethics seem similarly
impotent in the face of the problems.129
It is important that independent scientists, health professionals and professional
bodiesin cooperation with human rights, civil liberties and whistleblowing
organisationspromote the establishment of truly effective mechanisms
under international and domestic law that provide legal protection against
discrimination and criminal prosecution for whistleblowers. Furthermore, given
the failings of the current systems of whistleblower protection, the life-science
and biomedical communities have a duty to support those individuals who
refuse to participate in what they consider immoral research and development
projects, and those who blow the whistle on such activities. A number of
scientific associations and professional bodies have mechanisms for promoting
ethical standards amongst their members, which can also be utilised to support
colleagues facing reprisals for acting ethically.130
Conclusion
In its public statement Preventing hostile use of the life sciences, from ethics
and law to best practice, the ICRC has noted that [a]dvances in the life
sciences hold great promise for humanity, but that there is also great risk if
these same advances are put to hostile use.131 Whether the life sciences will
be employed for the further development and proliferation of incapacitant
weapons with the consequent dangers of the misuse of such weapons for
internal repression or offensive military operations will depend to a significant
degree on the actions of individual scientists and biomedical researchers and on
the role of the life-science community as a whole. At a minimum this must entail
stringent personal and community reflection and regulation of the trajectories
128 See, for example: Calland, R. and Dehn, G. (eds) 2004, Whistleblowing around the World, Open
Democracy Advice Centre, Cape Town, & Public Concern at Work, London; Deiseroth, op. cit.; Devine, T.
1997, The Whistleblowers Survival Guide, Fund for Constitutional Government, Washington, DC; Devine,
T. 2004, Whistleblowing in the United States: the gap between vision and lessons learned, in Calland and
Dehn, op. cit.; Martin, B. 1999a, Suppression of dissent in science, Research in Social Problems and Public
Policy, vol. 7, pp. 10535; Martin, B. 1999b, The Whistleblowers Handbook, Jon Carpenter, Charlbury, UK.
129 Martin, B. 2007, Whistle-blowing: risks and skills, in McLeish and Rappert, op. cit., p. 6.
130 For example, the AAAS Science and Human Rights Program (SHRP), which works with scientists to
advance science and serve society through human rights, has successfully campaigned on behalf of a number
of scientific whistleblowers. For further information, see: <http://shr.aaas.org/>.
131 ICRC, 2004, op. cit.
328
and possible applications of relevant research to ensure it cannot be used for

hostile purposes. In addition, it can be argued that now is the time for the lifescience community to take a far more active and engaged stance on this issue
and join with the arms control, disarmament, human rights and humanitarian
law communities to explore the full range of voluntary, normative and legally
binding mechanisms that can be applied to protect humanity from the potential
of abuse of the relevant technologies they are developing, whilst preserving the
beneficial applications.
329
19. WHO Project on Responsible

Life-Sciences Research for Global
Health Security: Why Ethics and
How to Strengthen It?
Emmanuelle Tuerlings and Andreas Reis1
Introduction
There has been much discussion over the past years about global health security
and how to strengthen it. One area that has raised much activity revolves around
the risks posed by accidents and the potential deliberate misuse of life-sciences
research.2 Different actors have proposed a variety of measures to manage such
potential risks.3 Yet little information is available about the needs and capacities
of countries, laboratories and research institutions in this area.
The World Health Organisation (WHO) has developed a self-assessment
questionnaire for laboratory managers and researchers to assess their needs and
1 The author is a staff member of the World Health Organisation. The author alone is responsible for the
views expressed in this chapter and he does not necessarily represent the decisions or policies of the World
Health Organisation.
2 Research accidents are understood as research activities that may unexpectedly pose some risks via
accidental discoveries. World Health Organisation (WHO) 2010, Responsible Life Sciences Research for Global
Health Security. A Guidance Document, WHO/HSE/GAR/BDP/2010.2, World Health Organisation, Geneva.
3 See, for instance: National Research Council 2004, Biotechnology Research in An Age of Terrorism, Committee
on Research Standards and Practices to Prevent the Destructive Application of Biotechnology, The National
Academies Press, Washington, DC; National Science Advisory Board for Biosecurity (NSABB) 2007,Proposed
Framework for the Oversight of Dual Use Life Sciences Research: Strategies for Minimizing the Potential
Misuse of Research Information, National Science Advisory Board for Biosecurity, Bethesda, Md; Miller,
S. and Selgelid, M. 2008, Ethical and Philosophical Consideration of the Dual-Use Dilemma in the Biological
Sciences, Springer, Dordrecht; Report prepared by the Centre for Applied Philosophy and Public Ethics at
The Australian National University for the Australian Department of Prime Minister and Cabinet, National
Security Science and Technology Unit, November 2006; Steinbruner, J. et al. 2007, Controlling Dangerous
Pathogens. A Prototype Protective Oversight System, March, Center for International and Security Studies at
Maryland (CISSM), University of Maryland, College Park; InterAcademy Panel on International Issues (IAP)
2005, IAP Statement on Biosecurity, November; Statement on the scientific publication and security, Science,
vol. 299 (2003), p. 1149; United Nations Meeting of the States Parties to the Convention on the Prohibition
of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on
their Destruction; Managing Risks of Misuse Associated with Grant Funding Activities. A Joint Biotechnology
and Biological Sciences Research Council (BBSRC), Medical Research Council (MRC) and Wellcome Trust Policy
Statement, September 2005; Wellcome Trust 2005, Guidelines on Good Research Practice, November; Dando, M.
and Rappert, B. 2005, Codes of conduct for the life sciences: some insights from UK academia, Briefing Paper
No. 16 (Second series), May, University of Bradford, UK.
331
capacities in regards to responsible life-sciences research. This self-assessment

questionnaire is part of a larger project on Responsible life-sciences research
for global health security, which published several documents on this issue.4
The responsible life-sciences research framework and its associated selfassessment questionnaire are based on three pillars supporting public health:
research excellence, ethics, and biosafety and laboratory biosecurity. The biorisk framework for responsible life-science research is an integrated approach,
where each pillar is equally important and should complement and reinforce the
others. This chapter will focus on the ethics pillar because it is often regarded as
an area that is complex, underfunded and an obstacle to undertaking research.
This chapter starts by giving an overview of the bio-risk management framework
for responsible life-sciences research. It then underlines the role of ethics in
this area and it ends by discussing how the self-assessment questionnaire can
provide useful feedback to countries and institutions to identify needs and
capacities in ethics, and ways to strengthen them.
The bio-risk management framework for

responsible life-sciences research
The WHO project on responsible life-sciences research for global health security
aims at promoting excellent, safe, secure and responsible life-sciences research
through an integrated approach that recommends investing capacities in three
pillars supporting public health: research excellence; ethics; biosafety and
laboratory biosecurity.5
Pillar 1: Research excellence. This concerns fostering quality in lifescience activities, which is the basis for developing new treatments and
therapeutics, strengthening health research systems, and promoting public
health surveillance and response activities. These elements are essential to
protecting and improving the health and wellbeing of all people.
4 World Health Organisation (WHO) 2005, Life Sciences Research: Opportunities and Risks for Public Health.
Mapping the Issues, WHO/CDS/CSR/LYO/2005/20, World Health Organisation, Geneva, <http://www.who.int/
csr/resources/publications/deliberate/WHO_CDS_CSR_LYO_2005_20/en/index.html> (viewed 7 February
2012); World Health Organisation 2007, Scientific Working Group on Life Science Research and Global Health
Security. Report of the First Meeting, WHO/CDS/EPR/2007.4, World Health Organisation, Geneva, <http://
www.who.int/csr/resources/publications/deliberate/WHO_CDS_EPR_2007_4/en/index.html> (viewed 7
February 2012); World Health Organisation (WHO) 2008, Research Policy and Management of Risks in Life
Sciences Research for Global Health Security. Report of the Meeting. Bangkok, Thailand, 1012 December
2007 WHO/HSE/EPR/2008.4, World Health Organisation, Geneva, <http://www.who.int/csr/resources/
publications/deliberate/WHO_HSE_EPR_2008_4/en/index.html> (viewed 7 February 2012); World Health
Organisation (WHO) 2012, Informal Consultation on Dual-Use Research of Concern (DURC), World Health
Organisation, Geneva, 2628 February 2013, <http://www.who.int/mediacentre/events/meetings/2013/durc/
en/> (viewed 2 September 2013).
5 This section draws upon the following WHO guidance document: WHO, 2010, op. cit.
332
19. WHO Project on Responsible Life-Sciences Research for Global Health Security
Pillar 2: Ethics. This involves the promotion of responsible and good research
practices, the provision of tools and practices to scientists and institutions
that allow them to discuss, analyse and resolve in an open atmosphere the
potential dilemmas they may face in their research, including those related
to the possibility of accidents or misuse of the life sciences.
Pillar 3: Biosafety and laboratory biosecurity. This concerns the
implementation and strengthening of measures and procedures to: minimise
the risk of worker exposure to pathogens and infections; protect the
environment and the community; and protect, control and account for
valuable biological materials (VBM) within laboratories, in order to prevent
their unauthorised access, loss, theft, misuse, diversion or intentional release.
Such measures reinforce good research practices and are aimed at ensuring
a safe and secure laboratory environment, thereby reducing any potential
risks of accidents or deliberate misuse.
Figure 19.1 Bio-Risk Management Framework for Responsible LifeSciences Research

Source: World Health Organisation (WHO) 2010, Responsible Life Sciences Research for Global Health
Security. A Guidance Document, WHO/HSE/GAR/BDP/2010.2, World Health Organisation, Geneva.
This approach recognises that there is no single solution or system that will
suit all countries, institutions or laboratories. Each country or institution that
assesses the extent to which it has systems and practices in place to deal with
the risks posed by accidents or the potential deliberate misuse of life-sciences
research will decide which measures are most appropriate and relevant according
to their own national circumstances and contexts.
333
Likewise, this approach also acknowledges the fact that there is a multiplicity
of stakeholders developing different measures. The public health sector is one
stakeholder among others (for example, national academies of science, ethicists
and security communities), and coordination at national and international levels
is therefore crucial for addressing those types of risks.
One important advantage of this integrated framework is that it recommends
investing capacities in public health areas that either are already in place in
many countries or are being developed for strengthening research capacities
and for addressing laboratory bio-risks. This approach therefore promotes a
sustainable and effective way to address such bio-risks. Indeed it emphasises
that one of the most effective ways to prepare for deliberately caused disease is
to strengthen public health measures for naturally occurring and accidentally
occurring diseases.6 In a similar manner, the promotion of a culture of scientific
integrity and excellence is considered as one of the best protections against the
possibility of accidents and deliberate misuse of life-sciences research and offers
the best prospect for scientific progress and development. A culture of scientific
integrity and excellence can be strengthened through the development and
reinforcement of capacities in the three aforementioned public health pillars.
In adopting the bio-risk management framework for responsible life-sciences
research, countries and institutions are encouraged to consider
reinforcing public health capacities in terms of research for health, biosafety
and laboratory biosecurity management and ethics
investing in training personnel (laboratory staff and researchers) and students
in ethics, the responsible conduct of research, and biosafety and laboratory
biosecurity
ensuring compliance with biosafety and laboratory biosecurity
identifying multi-stakeholder issues, with different layers of responsibilities
and encouraging coordination among stakeholders
using existing mechanisms, procedures and systems and reinforcing local
institutional bodies (if they exist).
While recognising the above, countries and institutions may consider drawing
on a bio-risk management framework for responsible life-sciences research,
which has been developed to help countries and institutions identify where
public health resources can be used for addressing the above risks. To facilitate
such analysis, a self-assessment questionnaire on responsible life-science
research has been developed to support health researchers, laboratory managers
and research institutions to identify strengths and to address weaknesses in
6 World Health Assembly Resolution WHA55.16 of 18 May 2002, Global Public Health Response to Natural
Occurrence, Accidental Release or Deliberate Use of Biological and Chemical Agents or Radionuclear Material
that Affect Health, World Health Organisation, Geneva.
334
each of the three pillars. Among these three pillars, ethics is an important public
health area where countries and institutions can invest capacities to address
such bio-risks.
An ethics framework
Context and related initiatives
The potential duality of research and science, and related moral concerns, is
not a new issue. Following the invention of atomic weapons, many scientists,
including Einstein, Oppenheimer and Russell, voiced concern about the
potential dangers to humanity arising from scientific discoveries.7
Ethical issues arising in research and the corresponding obligations of scientists
have for a long time been one of the key areas of work on ethics of the UN
Educational, Scientific and Cultural Organisation (UNESCO). Already in 1974,
UNESCOs eighteenth general conference adopted the Recommendation on the
Status of Scientific Researchers,8 which enumerated ethical responsibilities and
rights of scientists.
In 1999, UNESCO organised the World Conference on Science, in Budapest,
which gave special attention to ethical principles and responsibilities in the
practice of science.
Since 2005, UNESCO has led a new project, with regional consultations, to
survey the field of science ethics and to evaluate the need for an update of the
1974 document.9
Until recently, however, the bioethics community concerned with research has
focused on the protection of human subjects in clinical trials, and other kinds of
ethical issues raised by genetics and related disciplines. It is only in the past few
years that ethicists have become more engaged with the scientific community
and security experts regarding the ethical questions raised by the potential
misuse of science.
7 See, for example: Russell-Einstein Manifesto and the Pugwash Conferences on Science and World Affairs,
<http://www.pugwash.org/about/manifesto.htm> (viewed 7 February 2012).
8 United Nations Educational, Scientific and Cultural Organisation (UNESCO) 1974, Recommendation
Adopted on the Report of the Commission for Science at the Thirty-Eighth Plenary Meeting on 20 November 1974,
Paris.
9 Scholze, S. 2006, Setting standards for scientists: for almost ten years, COMEST has advised UNESCO on
the formulation of ethical guidelines, EMBO Reports, vol. 7 (SI) (July), pp. S657.
335
Ethics is now recognised in the field as providing a useful framework for the
identification of dilemmas, and the discussion and evaluation of responsibilities
of different stakeholders. Ethics can help promote understanding of the nature
of decisions that researchers and other actors have to make.10 The discussion
of ethical issues can contribute to consensus-building among the stakeholders,
who sometimes have competing interests.
Ethics framework in the WHO project

WHO has, since its establishment in 1948, given great importance to ethical
considerations in its programs. In 1967, the World Health Assembly (WHA)
resolved that scientific achievements, and particularly in the field of biology
and medicinethat most humane scienceshould be used only for mankinds
benefit, but never to do it any harm.11 A WHO report in 2005 discussed in more
detail the ethical questions raised.12
Some of the basic dilemmas addressed in the current WHO project, Responsible
life science research for global health security, are
how to weigh the potential benefits of research against the risk of misuse
how to weigh the individual interests of researchers against the common
good of public health
how to best manage the risks associated with research, without hindering its
beneficial application to public health
what are the responsibilities of individual researchers, and of the scientific
community as a whole to society?
For the development of an ethics framework, it is helpful to analyse the
responsibilities and dilemmas of different stakeholders involved in life sciences,
and to develop corresponding tools to support them in making ethical choices.
Researchers, science societies and codes of conduct/

ethics
Individual scientists have been at the centre of the debate on ethical
responsibilities in dual-use research. On the one hand, they will aim to carry out
and publish beneficial research, while on the other hand they want to refrain
from research that could lead to the potential malevolent use of the results. They
10 WHO, 2007, op. cit.
11WHA20.54.
12 World Health Organisation (WHO) 2005, Life Sciences Research, WHO/CDS/CSR/LYO/2005/20, World
Health Organisation, Geneva, <http://www.who.int/csr/resources/publications/deliberate/WHO_CDS_CSR_
LYO_2005_20/en/index.html> (viewed 7 February 2012).
336
have a widely acknowledged obligation to carry out their research in accordance

with good research practices and laws and regulations, including codes (see
below). In addition, the debate on dual-use research has made clear that there
are also some expectations on the individual scientists to make judgments
about the outcomes and potential security implications of their research
projects; however, often researchers will not be security experts and may face
difficulties in making such assessments. Therefore, while education can help to
raise awareness among scientists, in case of doubt, they should be able to turn
to mechanisms for reporting and consultation, including for whistleblowing,
provided by their institutions.
Codes for researchers have for a long time been proposed as potentially useful
tools to address the potential misuse of life-science research at the level of the
individual scientists. In fact, codes for professional behaviour date back at least
two millennia (the Hippocratic Oath, Maimonides, and so on). Two types of
codes are usually distinguished
1. codes of conduct: these provide specific guidelines with respect to what is
considered appropriate behaviour
2. codes of ethics: these are more aspirational, setting forth the ideas to which
practitioners should aspire, such as standards of objectivity or honesty.
Some prominent examples of codes for scientists include those proposed by
scientific and academic societies (such as the American Society for Microbiology,
the Chinese Academy of Sciences, The Royal Society in the United Kingdom
and the US National Science Advisory Board for Biosecurity), and medical
associations (such as the World Medical Association and the American Medical
Association).13 Recently, the International Centre for Genetic Engineering and
Biotechnology has been undertaking a project to review codes of conduct.
There has been considerable debate on the effectiveness of codes of conduct/
ethics for researchers.14 Critics have insisted that there is little evidence about their
effectiveness in practice. For example, the Hippocratic Oath has not hindered
medical doctors from doing horrendous things to their patients. Scientists may
simply not comply with such non-binding instruments as codes, and terrorists
will not be deterred from using biotechnology with bad intentions. Even with
the best intentions, and fully aware of their duties, scientists are not security
experts and may thus not foresee the consequences of their experiments, which
might have dual-use potential.
13 For more examples, see WHO, 2010, op. cit., pp. 589.
14 Rappert, B. 2007, Codes of conduct and biological weapons: an in-process assessment, Biosecurity and
Bioterrorism: Biodefense Strategy, Practice and Science, vol. 5, pp. 14554.
337
On the other side, the proponents of such codesfor example, Margaret

Somerville and Ronald Atlas15have argued that they serve their purpose
in raising awareness, fostering norms among the scientific community and
establishing public accountability mechanisms. They can help sensitise
scientists who might be used by malevolent individuals or organisations and
the public health community to the risks of dual-use research. Codes can also
foster an environment in research institutions in which whistleblowing could
be encouraged. There is an ongoing discussion between the proponents of a
universal code, arguing that national codes of conduct will have little effect on
the global behaviour of life scientists, and those who prefer developing some
basic principles that would then be applicable to and implemented at local
settings, institutions or countries.
Education and training in biosecurity and ethics are other tools for raising
awareness among scientists. A report published in 2008 by the American
Association for the Advancement of Science (AAAS) examined such education
programs and emphasised their importance.16
Research institutions
As most research takes place in institutions, these also have important
responsibilities in the prevention of accidents and the deliberate misuse of
research. Research institutions have an obligation to ensure that research within
their premises is conducted according to national laws and relevant codes of
conduct. They should sensitise their researchers to dual-use issuesfor example,
through education and specific training, promoting discussion and reflection
on research practices, and creating a climate that encourages scientists to come
forward in case of difficulties. Procedures and standards for whistleblowing
should be in place for cases of scientific misconduct.
Another potentially useful approach to address the dual-use dilemma, which
has been proposed at the institutional level, is the ethical review of sensitive
research by an ethics committee. In practice, however, it is not clear which
committees could take on this role. Traditional research ethics committees review
research with human subjects, and not bench science, and their members are
not usually trained in biosecurity issues. With specific training, the members of
research ethics committees could probably be enabled to play this role. Another
15 Somerville, M. and Atlas, R. 2005, Ethics: a weapon to counter bioterrorism, Science, vol. 307, pp.
18812.
16 American Association for the Advancement of Science (AAAS) 2008, Professional and Graduate-Level
Programs on Dual Use Research and Biosecurity for Scientists Working in the Biological Sciences: Workshop
Report, American Association for the Advancement of Science, Washington, DC.
338
option would be to supplement institutional biosafety committees (IBCs) with

ethicists. A further difficulty to resolve is the proper identification of the
research proposals that should be subjected to such a review.
Publishers and journal editors

Freedom of intellectual inquiry, sharing of knowledge and publication are at
the very heart of science. So one of the most contentious debates about the
governance of dual-use research involves the extent to which the publication of
potentially dangerous information should be restricted, or censored.17
Publications enable progress in beneficial research and can inform about new
kinds of threats. In particular, sufficiently detailed materials and methods parts
of publications are essential for other scientists to reproduce and verify results.
Yet, these sections are precisely the ones that could potentially be misused by
ill-minded individuals. Publishers and journal editors have a responsibility to
prevent the publication of research results that may be misused. In February
2003, recognising this responsibility, a joint Statement on Scientific Publication
and Security of the Journal Editors and Authors Group was simultaneously
published by Science, Nature, the Proceedings of the National Academy of Sciences
and the American Society for Microbiology Journals.18 While reminding readers
of publishers unique responsibility for the integrity and the advancement
of research, the statement accepts that at times censorship of science may be
warranted: We recognize that on occasion an editor may conclude that the
potential harm of publication outweighs the potential societal benefits. Under
such circumstances, the paper should be modified, or not be published.19
Governments, international organisations and funding

bodies
Finally, governments, international organisations and funding bodies all face
dilemmas in managing the risks related to deliberate misuse of research. On
the one hand, they play a key role in promoting scientific development, and in
creating an environment that is conducive to stimulating life-sciences research.
On the other hand, they have important responsibilities in promoting global
health security and in minimising the potential harm to public health, at a
national and a global level.
17 Miller, S. and Selgelid, M. 2007, Ethical and philosophical considerations of the dual-use dilemma in the
biological sciences, Science and Engineering Ethics, vol. 13, pp. 52380.
18 Selgelid, M. 2007, A tale of two studies: ethics, bioterrorism, and the censorship of science, Hastings
Center Report, vol. 37, no. 3.
19 Journal Editors and Authors Group 2003, Statement on scientific publication and security, Science, vol.
299, no. 5610, p. 1149.
339
There are different risk-management options for governing life-science research

with dual-use potential. It is essential to find and maintain the right mix of
policies in order to allow the public health benefits of life-science research to
be maximised while minimising the risks. Besides the promotion of research
excellence, biosafety and laboratory biosecurity, an ethical framework is a key
component of a public health approach. While no magic bullet, codes of ethics
for scientists, as well as for publishers, can make essential contributions. Further
discussion about the usefulness and feasibility of an ethics review of sensitive
research is needed. Finally, assessments, awareness-raising and training at local
and facility levels are key to the implementation of effective policies. The WHO
self-assessment questionnaire can help countries and institutions to identify
these needs and capacities and take the necessary actions.
Ethics and the self-assessment questionnaire

The process of self-assessment starts with an identification of strengths,
weaknesses and gaps, and concludes with action to address the gaps and
weaknesses and to build on or consolidate the strengths. Going through this
process will first provide an assessment of the extent to which systems are
in place in the national public health system, in research institutions and in
laboratories to facilitate and ensure excellence in science, ethics, safety and
security.
Second, it will identify priority areas where action is necessary to ensure such
practices. In coordination with other stakeholders, the knowledge gained
through this process will help countries and institutions to align available
resources with local needs and circumstances. This is therefore an important
step in the identification of appropriate training and a sustainable way to
enhance global health security. This self-assessment tool could also be used
as a simple and cost-effective way for interested countries and institutions to
raise awareness about responsible life-sciences research. Health researchers,
laboratory managers and research institutions are the primary audiences for this
self-assessment questionnaire. It was not developed to evaluate the adequacy of
measures developed by other sectors (for example, security, academia, publishers
and editors). Yet, it recognises the importance of collaboration and coordination
between different sectors, especially when priority areas have been identified
and when action will unfold.
Methodology
The self-assessment questionnaire on responsible life-sciences research
lists several statements associated with each pillar, using the Likert scale. A
340
statement for which a respondent answered dont know or disagree may

mean that further information is needed in certain areas and may indicate some
weaknesses. On the other hand, responses of agree and strongly agree suggest
that appropriate measures are already in place.
The strengths and weaknesses of an institution can then be reported by pillars,
giving some overview of advantages and gaps. It is therefore strongly advised to
fill in the questionnaire as a whole and not in parts. The self-assessment tool is
a voluntary, anonymous and innovative tool in this field.
The ethics pillar lists some of the following statements to which respondents are
being asked to give their opinion20
education and/or training is offered on research ethics
appropriate ethical research guidelines and practices have been published
and implemented
adequate mechanisms exist for investigating and responding to nonadherence to ethical standards
researchers know how to assess whether the risk outweighs the benefit of
continuing with their research or activities
researchers are aware of and informed about national and international
conventions, laws and regulations related to their research
an ethics committee assesses research proposals involving human subjects
a review process exists to assess ethical issues raised by research proposals
not involving human or animal subjects
information about the national and international conventions and regulations
related to all fields of science is easily accessible.
The self-assessment questionnaire was piloted in a South African facility in
2009.21 A seminar on responsible life-sciences research and dual-use issues
was presented to the audience as a way to introduce the questionnaire to the
laboratory scientists and managers. The audience was also informed about the
purpose and content of the questionnaire, noting that this was an anonymous and
voluntary process. Of the 20 scientists present, 18 completed the questionnaire.
The respondents were coming from different laboratories, all part of the same
facility. Feedback on the questionnaire was provided to the respondents at the
end of the seminar.
20 WHO, 2010, op. cit.

21 Gould, C. 2011, Biosecurity: a public health approach to reducing risk, Research Report for the Wellcome
Trust Project on Building a Sustainable Capacity in Dual-Use Bioethics, <http://www.brad.ac.uk/bioethics/
Monographs> (viewed 22 September 2011).
341
The pilot study had some limitations. First, the sample size of the pilot study
is too small to draw conclusions that would be representative of the whole
facility. The collected data are therefore only illustrative of potential needs
and capacities. A second limitation of the self-assessment questionnaire is that
these questions may be considered subjective. The validity and reliability
of the subjective measures have been enhanced through multiple questions
measuring the same subjective state. The self-assessment questionnaire has
indeed been designed in a way that statements regarding the different pillars
can be analysed in combination, thereby reinforcing the data analysis. It should
be noted, however, that one important strength of the questionnaire is its ability
to identify differences in opinions about existing systems and measures so that
the discrepancies can be dealt withnot necessarily through a change in the
system itself but just with, sometimes, further discussion with the staff about
existing measures.
Results
The data collected by the pilot study regarding the ethics pillar showed the22
following.
Seventeen of the 18 respondents agreed or strongly agreed that education
and/or training is offered on research ethics. One respondent was undecided.
This is reinforced with the fact that 15 respondents agreed or strongly agreed
with the statement that ongoing research training takes place.
Likewise, a majority of respondents (15) agreed or strongly agreed that
research proposals raising ethical issues (whether or not human or animal
subjects are involved) are subject to review.
A majority of respondents, however, disagreed with the statement that
education and/or training is offered on dual use. Nine respondents disagree
or strongly disagreed, and seven respondents were undecided about the
statement. Only two agreed with it.
Eight respondents also disagreed with the statement that researchers are
aware of and informed about national and international laws and regulations
related to their research, and three respondents were undecided. This can
be read in conjunction with the fact that 14 respondents were undecided
or disagreed with the statement that information about the national and
international laws and regulations related to all fields of science is easily
accessible.
In addition, 13 respondents were undecided or disagreed with the statement
that whistleblowing mechanisms exist and make provision for the protection
22Ibid.
342
of whistleblowers. Related to this, 10 respondents were undecided or

disagreed with the statement that researchers have somewhere to turn for
competent advice if they have ethical, safety or security questions relating
to their research. Yet five respondents agreed or strongly agreed with the first
statement and eight respondents agreed with the second one.
Eleven respondents were undecided or disagreed with the statement that
adequate mechanisms exist for investigating and responding to nonadherence to ethical standards, and half of the respondents were undecided
or disagreed with the statement that appropriate ethical guidelines and
practices have been published and implemented. Yet seven respondents
agreed with the first statement and nine agreed with the second one.
Discussion
The pilot study at the South African facility illustrates the type of useful and
informative feedback the self-assessment questionnaire can provide to laboratory
managers and their staff.23 In this case, the pilot study may seem to indicate that
ethics training is an asset of the facility and that research proposals are being
regularly reviewed. But it also seems to point out some areas for action.
For instance, disagreements occurred in four areas
about the education and/or training on dual-use issues
about their awareness of and information regarding national and international
laws and regulations related to their research
about the existence of mechanisms for staff to report unlawful or irregular
conduct (that is, whistleblowing mechanisms) and where to turn for
competent advice if they have ethical, safety or security questions relating
to their research
about the publication and implementation of ethical guidelines and practices
and mechanisms for investigating non-adherence to ethical standards.
When a majority of respondents disagreed with a statement or was undecided,
this may point to a certain weakness that needs to be addressed. In this case, a
majority of respondents noted the lack of training and/or education on dual-use
issues.
Disagreement among respondents may indicate that there may be measures in
place but these are not sufficiently well communicated to all employees. This
may mean that ethical standards for research may be in place in the facility,
23 Although the pilot study implemented the whole self-assessment questionnaire, only the ethics data are
being reported in this chapter.
343
as well as mechanisms to deal with non-adherence to ethical standards, but

the employees should be better informed about them. Likewise, employees may
need to have more accessible information about national and international laws
and regulations related to their research as well as additional information as
to where to turn for competent advice if they have ethical, safety or security
questions associated with their research.
In summary, the analysis of the data associated with the ethics pillar tends to
suggest that the facility has important capacities in terms of ethics training and on
ethical review of research proposals while some needs may have been identified
in terms of dual-use training. As well, additional information may be provided on
the availability of competent advice; publication and implementation of ethical
guidelines and practices and mechanisms for investigating non-adherence to
them; and on national and international laws and regulations.
This process of self-assessment should therefore provide useful and practical
information for research institutions and their managers about research excellence
and management, ethics, biosafety and laboratory biosecurity practices within
their facilities. When comparing the results of the three pillars, the institution
can also identify capacities and needs for each pillar. It can then assess which
pillar looks stronger compared with the others and then identify priority areas
for action. One way to address potential needs might be to organise a seminar for
all staff to share the data analysis, to underline the strengths of the institution
and to propose measures to address identified gaps.
In this case, the identified needs could be addressed through a seminar that
would be aimed at sharing information about national and international laws
and regulations related to their research and the setting up of a web link to
encourage staff to make themselves aware of any changes in national and
international regulations. A seminar could also be held on the mechanisms
or processes in place in the institution if staff have ethical, safety or security
questions regarding their research and to make clear where the ethical standards
can be found and how whistleblowing mechanisms can be accessed. Information
regarding ethical standards can be distributed to the staff or could be put
on a web site easily accessed by the employees. Staff should also be able to
keep themselves regularly informed about any changes in ethical procedures.
Another way of addressing needs might be to invite other facilities and research
institutions to share information about their best practices, how they train their
staff on specific topics, how they have identified ways for making information
on certain issues available to staff and how to keep them regularly updated.
Finally the self-assessment questionnaire could be proposed again to the facility
employees after the implementation of activities that were aimed at addressing
needs to evaluate the progress made.
344
Self-assessment should therefore not be seen as an end in itself but as a

process for achieving better research management, enhanced ethics practices,
the implementation of biosafety and laboratory biosecurity measures and
procedures, and their continuous improvement. This would contribute
to fostering responsible life-sciences research activities in countries and
institutions and promoting public health. The questionnaire was also used in
Kenya, to assess the capacity of research and diagnostic laboratories in Nairobi
and surrounds.24
Ethics is essential to address the risks posed by life-sciences research. The pilot
study on the self-assessment questionnaire on responsible life-sciences research
illustrates how ethics needs can be identified and what kind of activities can be
put in place to address those needs.
24 Kenya E. et al., An Assessment of the Capacity of Research and Diagnostic Laboratories in Nairobi and
Surrounds, June 2012.
345
Conclusion
20. Ethics as
Brian Rappert
The contributions to On the Dual Uses of Science and Ethics have extended an
invitation to dwell on a matter of much importance: the unity of knowledge.
Each chapter has considered the potential for the skills, know-how, information
and techniques associated with modern biology to serve contrasting ends. Each
has spoken to steps that might be undertaken to prevent the deliberate spread
of disease. A recurring message has been that, to date, the discussion about the
dual-use potential of the life sciences has been characterised by silences and
absences. To be sure, while some researchers and policymakers have devoted
attention to this topic for years, overall the regard is patchy and problematic.
But as the title of this volume advocates, as we attend to the multiple purposes
served by the life sciences, parallel regard should be directed towards those
served by ethics. Another theme of On the Dual Uses of Science and Ethics has
been that, to date, the ethics discussion about the multiple potentials of the
life sciences has been characterised by silences and absences. This book has
sought to redress the state of analysis by bringing together individuals with
varied backgrounds. Some have been trained as ethicists but had given little
prior attention to the destructive applications of science, while others had long
mulled over this topic but not directly informed by the discipline of ethics.
As part of building the debate, the contributors have asked what role ethics
(particularly bioethics) might serve in averting the deliberate spread of disease.
Part reflection on the preceding chapters, part analysis of the wider literature,
and part programmatic agenda setting, this concluding chapter addresses how
the history, methods and practices of bioethics could enrich current dualuse deliberations. This is conceived as a two-way exchange. The queries,
uncertainties and anomalies raised by examining the unity of knowledge will
also be approached for what they tell us about the preoccupations of bioethics.
Through this the limits of bioethics, how it must adapt and what is at stake in
suggesting change is required will be considered.
In asking about uses, regard will also turn to possible abuses of bioethics.
Efforts to bring to bear the intellectual resources from this tradition come with
their own dangers. The assumptions informing normative approaches, how they
are positioned to justify courses of (in)action and the blind-spots of analysis
are just some of the reasons for holding together worries about the direction of
ethics and the life sciences. For instance, one of the common ways bioethics has
positioned itself as a discipline is as an intellectual response to new innovations
349
in science and technology.1 Ethicists then set out to offer needed analytical
responses to the latest challenges cropping up. Yet, this techno-origin myth2
risks becoming blind to the organisation, funding priorities, agendas and
purposes of ethics.
This chapter sets about to accomplish the above aims through a four-part
analysis. The next section begins by scrutinising assumptions informing
the contributions to On the Dual Uses of Science and Ethics as well as other
commentaries. Undertaking this critical turning back is vital if we wish to set
a productive program for the future. Section three asks how normative and
empirical approaches can be reconciled in efforts to prevent the destructive
application of the life sciences. The fourth section considers the potential
dangers associated with promoting greater bioethical attention to dual use.
Informed by the analysis that precedes it, the final section sets out a path for
future intellectual and practical engagement. That path entails attending to the
whys and hows associated with what is not: a) what is not recognised as posing
a concern in the first place or, if recognised at some level, b) what is not treated
as a problem, and c) if regarded as a problem then what is not acted upon.
Questioning beginnings
This section begins by undertaking a central task for robust ethical analysis:
scrutinising presumptions.
1. Categorical condemnation
The contributors to this volume have begun with a normative position that
has itself attracted little attention: the categorical unacceptability of biological
weapons. This overall evaluation is codified in the 1972 Biological and Toxin
Weapons Convention (BTWC) that forbids states from acquiring or developing
bio-agents and toxins for hostile purposes.
Within international diplomacy, policy deliberation and academic study, the
attribution of an abhorrent or an inhumane standing to these weapons is
commonplace (as in van der Bruggens chapter in this volume). While some
have impugned the political expediency of the categorical prohibition of
bioweapons vis--vis the laxer controls for nuclear weapons (which still enable
1 See Borry, P., Schotsmans, P. and Dierickx, K. 2005, The birth of the empirical turn in bioethics, Bioethics,
vol. 19, no. 1; and Kushe, H. and Singer, P. (eds), A Companion to Bioethics, Blackwell, London.
2 de Vries, R. 2007, Who will guard the guardians of neuroscience? Firing the neuroethical imagination,
EMBO Reports, vol. 8, S659.
350
20. Ethics as
retention by major military powers),3 biological weapons are rarely portrayed

as anything other than illegitimate today. That recent diplomatic accusations
about the existence of offensive programs all refer to clandestine ones illustrates
this exceptional status. With this now longstanding global rejection in place,
little in the way of justification is treated as necessary for condemnation. When
argument is given, it often amounts to a restatement of the cardinal assertion
that inflicting death and injury to life through life is qualitatively different than
other means of harming.4
Given this normative starting ought, it is hardly surprising that discussion
about the possible contribution of the life sciences to the spread of disease is
characterised by a binary language of use and misuse. This stark language
would be out of place for many elsewhere. The systematic engineering of
cross-fertilisation between civilian science and war-fighting capabilities has
been a longstanding mission of many national innovation policies since World
War II.5 When Colwell commented that weaponising bio-agents violates the
fundamental values of the life sciences that I and my colleagues hold dear: that
science is a vital tool for improving life and health of our planet and enhancing
our understanding of the natural world6 in a report for the US National Academy
of Science, the reference to the life sciences would no doubt be widely shared
in North America and much further afield. Generalising this sentiment against
weaponisation to science as a whole would, however, be far out of line with
scientific practice.
In the past though, the moral standing of what would today be labelled as
biological weapons was more contested. In a wide-ranging analysis, for instance,
Zanders argued that the condemnation of biowarfare over the ages has never
been uniform or absolute.7 Within the twentieth century, scientists, medics and
others have found ample justifications for their willingness to participate in
large-scale programs. Patriotism, civilian spin-offs and even the belief that (at
least some) bioweapons represented more humane alternatives were among the
3 See Falk, R. 2001, The challenges of biological weaponry, in S. Wright (ed.), Biological Warfare and
Disarmament, Rowman & Littlefield, London.
4 For an examination of the ethics of biological weapons, see, for example: Sims, N. 1991, Morality and
biological warfare, Arms Control, vol. 8, pp. 523; and Krickus, R. 1965, On the morality of chemical/
biological war, Journal of Conflict Resolution, vol. 9, no. 2, pp. 20010. Even in attempts to differentiate the
moral standing of types of biological weapons, the starting assumption is often that they all share dubious
qualities, as in Appel, J. M. 2009, Is all fair in biological warfare? The controversy over genetically engineered
biological weapons, Journal of Medical Ethics, vol. 35, pp. 42932.
5 James, A. 2007, Science & technology policy and international security, in B. Rappert (ed.), Technology
and Security, Palgrave, London.
6 Colwell, R. 2010, Understanding Biosecurity: Protecting against the Misuse of Science in Todays World,
National Research Council, Washington, DC.
7 Zanders, J. P. 2003, International norms against chemical and biological warfare, Journal of Conflict &
Security Law, vol. 8, no. 2, pp. 391410.
351
reasons.8 Prior to the ratification of the BTWC, in the 1960s scientific societies
(such as the American Society for Microbiology)9 found themselves embroiled
in bitter dispute about the rights and wrongs of offensive programs.
Recognition of this historical contestation is not taken as an opportunity to
reject the denunciation of bioweapons and the corresponding need to avert the
life sciences from aiding their development. In this sense, this chapter begins
where others in the volume have begun (and ended).
Noting this contestation though will be taken as an opportunity to reflect
on the basis for todays commonsense moral appraisals and thus how ethical
analysis can be made to matter. In relation to the topic of this volume, that
means questioning how norms, stigmas and taboos can be bolstered (see below).
Noting this contestation also will be taken as an opportunity to reflect on how
the boundary is set between what is and is not treated as a biological weapon.
As Whitby details in his chapter about plant inoculants, at times the distinction
can be fine to non-existent. The contingency associated with categorisations
is evident elsewhere. During ancient Greek and Roman times, debate about
the morality of deliberate poisoning and the contamination of water supplies
took place alongside debate about the morality of driving enemy troops into
mosquito-infested areas.10 Each of these represented a sort of weaponising of
nature, even if the last would not fit today under the term biowarfare. And
as yet another aspect of boundary drawing, while the use of pathogens as
weapons to inflict disease might be widely deplored, just when it is deemed
appropriate for biology to serve war-fighting is of much less accord. Whitmans
chapter indicated the range of bio-enabling capabilities that nanotechnology
offers militaries. The moral status of such applications is likely to be a matter of
disagreement in a way it would not be for other areas of nanotechnology given
the link to biology.
What such examples suggest is that the descriptive matter of how things become
understood needs to accompany the normative matter of how things ought to be
judged.
Take the example of bio-defence. As a number of contributors in this volume have
indicated, the acceptability of activities undertaken in the name of protection
is not a matter of unanimity. In no small part this has been due to disagreement
about what distinguishes the knowledge and techniques necessary for defence
from those of offence. As I have argued elsewhere, at stake in past disputes about
8 Balmer, B. 2002, Killing without the distressing preliminaries, Minerva, vol. 40, pp. 5775.
9 Cassell, G., Miller, L. and Rest, R. 1994, Biological warfare: role of scientific societies, in R. Zilinskas (ed.),
The Microbiologist and Biological Defense Research, New York Academy of Science, New York.
10 Mayor, A. 2009, Greek Fire, Poison Arrows and Scorpion Bombs, Overlook Duckworth, London.
352
20. Ethics as
what is ethically and legally permissible have been thorny questions about how
the purposes of activities can be discerned. Commentators have been divided on
whether purpose can be discerned from the characteristics of the activities or
whether their meaning has to be found in some wider context. In practice, for
many who subscribe to a contextual approach a sense of the context is often
mutually defined in relation to a sense of activities that are under scrutiny. For
instance, expansion of American military funding during the 1980s was one
occasion that featured heated dispute about the appropriateness of bio-defence.
As part of this debate, attempts to evaluate aerosolisation-testing centres were
subjected to contrasting assessments due to alternative characterisations about
the overall purpose of bio-defence programs. But on top of this:
[S]uch characterizations, in turn, [informed] determinations of intent.
Many critics of the US biodefense program treated the secrecy
surrounding it and the wider political posturing of US administrations
as indicators of the dubious intent of the undertaking. Seemingly
acceptable individual projects were re-considered for how they
might inform offensive weapons development. In turn, the number
and character of such ambiguous activities provided justification for
concerns about the ultimate aims served by the program as a whole.11
It is through suchwhat might well be deemed circularreasoning that
attempts to assess what is permitted often unfold. Herein, the trust held in a
presidential administration informed the evaluation of particular projects and
vice versa. In recognising such co-definition dynamics, it is possible to go
beyond simply noting the recurring difficulties with making decisions about
whether biodefense research can be ethically justified as truly defensive in
nature.12 Instead it is possible to examine how boundaries and evaluations are
established.
The previous paragraph indicated the need for empirically informed ethical
analysis in relation to matters of dual usea theme that will be taken up again
in the next section.
2. Neglected dis-ease
As mentioned at the beginning of this chapter, a motivation for this volume
and a conclusion supported by its contributorswas a sense that dual-use issues
had not hitherto been subjected to significant or sufficient ethical analysis.13
11 Rappert, B. 2005, Prohibitions, weapons and controversy: managing the problem of ordering, Social
Studies of Science, vol. 35, no. 2, p. 231.
12 King, N. 2005, The ethics of biodefense, Bioethics, vol. 19, no. 4, pp. 43246.
13 As also voiced in Chadwick, R. 2011, Bio- and security ethics: only connect, Bioethics, vol. 25, no. 1,
p. ii.
353
There are, however, past and present lively areas of deliberation related to the
overall themes of On the Dual Uses of Science and Ethics. The rights of research
subjects have long been an enduring topic of mainstream medical ethics. Human
experimentation in weapons programs has more than its share of flagrant
violations of rights.14 Just how much research and healthcare priorities ought
to pay to concerns about the deliberate spread of disease have animated much
debate recently, too, with a feared biosecuritisation of priorities.15 Many have
contended that the increasing funding directed at security post 9/11 has led to
skewed research and healthcare agendas. Herein pathogens that might kill in
warfare are given disproportionate resources compared with those that do kill
on a daily basis. What suffer are human welfare, global justice and international
security.16 How many lives, it might be asked, have been lost because of the
choice to spend scare resources on non-existent threats? In contrast, others
have advocated for the increasing inclusion of security within public health and
thereby reconceiving what is meant by both.17 Herein, not only do healthcare
systems need to do more to prepare for the intentional spread of disease, but also
security agencies need to recognise the gravity of natural disease for national
protection.18
These points about biosecuritisation suggest how the destructive use of the life
sciences overlaps with healthcare preparations for the outbreak of disease. To
the extent this is correct, turning the previous disconnect between ethics and
dual use into connect should be entirely feasible.
Affinity was a recurring theme of the chapter by Bartolucci and Dando, who
examined: 1) what neuroethics has to say about dual use; and 2) how neuroethics
has positioned itself as similar/different to other areas of ethics. The first of
these is relatively straightforward: neuroethics is notable for its lack of direct
engagement with the dual-use aspects of the life sciences. At the same time,
much of it could be pertinent. The second dimension is more complex. Debates
about distinctiveness have long been at the centre of positioning about whether
there is a need for neuroethics as a field of study in its own right. With the
increase of specialised publications and conferences, over recent years many
14 Harris, S. 1999, Factories of death, in T. Beauchamp and L. Walters (eds), Contemporary Issues in
Bioethics, 5th edn, Wadsworth, London, pp. 4708.
15 Fisher, J. and Monahan, T. 2001, The biosecuritization of healthcare delivery: examples of post-9/11
technological imperatives, Social Science & Medicine, vol. 72, pp. 54552; and Brown, T. 2001, Vulnerability
is universal, Social Science & Medicine, vol. 72, pp. 31926.
16 Enemark, C. 2009, Is pandemic flu a security threat?, Survival, vol. 51, no. 1, pp. 191214; and Choffnes,
E. 2002, Bioweapons: new labs, more terror? Bulletin of the Atomic Scientists, vol. 58, no. 5, pp. 2832.
17 See Fidler, D. and Gostin, L. 2008, Biosecurity in A Global Age, Stanford University Press, Stanford, Calif.
18 See, for example, de Waal, A. 2010, Reframing governance, security and conflict in the light of HIV/
AIDS: a synthesis of findings from the AIDS, security and conflict initiative, Social Science & Medicine, vol.
70, no. 1, pp. 11420.
354
20. Ethics as
have adopted a language of intellectual uniqueness, at least at times. One of the

dangers identified with this is that previous bioethical work becomes lost or
reinvented as neuroethicists divorce themselves from mainstream ethics.
Stepping back from neuroethics as a specific example, there is good reason
for contending that the likeness across topic areas will be tricky to pin down.
Establishing the extent of resemblance is practical judgment that must be argued
for, rather than simply being resolved once similarities (or differences) are
identified. Determining the extent of overlap between topics of ethics requires
identifying the key issues and defining how they should be understood. For
emerging or disputed topics thoughsuch as dual-use life sciencesthese can
be the very matters where accord breaks down.
I want to take the past two paragraphs as suggesting that the question of how
much the ethical issues associated with dual-use concerns are similar to those
elsewhere should be posed while also attending to implications of claims of
similarity or difference. So in pondering to what extent a distinctive dualuse ethics is required, the question of required for what? should loom
large. Attempts in relation to neuroscience and nanotechnology to stake out a
distinctive sub-branch of ethics, for instance, are part and parcel of building
professional identity and boundaries. While there was little in the way of calls
in this volume to clear a distinct space for dual-use ethics, the complexion of the
conversation might well change over time.
The implications of claims to similarity or difference figure in ongoing ethical
debates elsewhere. Take the area of public health again. Leslie Francis and
colleagues have argued that infectious disease has been largely absent from
the historical development of bioethics as a field.19 As a result, insufficient
attention has been given to the way contagious individuals are both patients
with illness and vectors for transmitting disease.20 Consequently, traditional
conceptualisations of the role for, composition of and need for informed consent
are inadequate. These fail to recognise how choices about treatment or nontreatment made by some affect others. Once this is acknowledged, an individual
patients desire need not be the prime consideration.21
Turning to the topic of this volume, such observations are also relevant in the
case of the deliberate spread of disease. A point to draw out is the way Francis
and colleagues insist that current deficits in bioethics cannot be alleviated by
19 Francis, L. P., Battin, M. P., Jacobson, J. A., Smith, C. B. and Botkin, J. 2005,How infectious diseases got
left outand what this omission might have meant for bioethics, Bioethics, vol. 19, no. 4, pp. 30722.
20 See as well Tausig, M., Selgelid, M. J., Subedi, S. and Subedi, J. 2006, Taking sociology seriously: a new
approach to the bioethical problems of infectious disease, Sociology of Health & Illness, vol. 28, no. 6, pp.
83849; and Selgelid, M. 2005, Ethics and infectious disease, Bioethics, vol. 19, no. 3, pp. 27289.
21 Likewise, Francis et al. argued past thinking about justice in health care in bioethics failed to acknowledge
the global origins of disease in anything like a robust manner.
355
a minor refinement of bioethical principles (such as autonomy). Instead they

argue that a fundamental rethink of the traditional liberal foundations of
bioethics is necessary. The principle of autonomy, for instance, needs to break
from a fixation on the preferences of competent and reasonable individuals.
Instead acknowledgment must be given to how the vulnerability of individuals
and communities derives from our embodied and relational existence. I take
the points above as indicating that alongside claims about what is or is not
missing from bioethics about dual use should be attention to how similarity and
difference are being conceived.
Meeting in the middle

The previous section examined two of the starting points for On the Dual Uses of
Science and Ethics: the normative condemnation of biological weapons and the
lack of robust bioethical engagement with dual-use life sciences.
Still another starting point for this volume was the belief that working towards
a bioethical engagement requires bringing together contributors from a variety
of disciplines. The composition of the chapters reflects that thinking. No single
field of studybe it ethics, social science, biology, and so oncould adequately
label the previous chapters. Moreover, a mix of aims has been pursued: clarifying
concepts and normative justifications, detailing how problems are recognised,
charting the state of development in science and technology, and so on. To some
extent, these varied activities have aligned with contrasting tasks: prescription,
description and assessment.
Another way of making the points above is to say that On the Dual Uses of Science
and Ethics has sought to clear a space for bioethical engagement within the
fraught terrain of empirical ethics. In recent years the umbrella term empirical
ethics has signalled the need to bring closer together research questions and
methodologies from multiple fieldssuch as ethics, sociology, bioethics, law,
applied ethics, epidemiology and anthropology. The aim has been to couple
normative analysis and empirical evidence. That has entailed finding some way
of combining contrasting preoccupations. While much of bioethics strives to
produce useful guides for making decisions in problematic-choice situations,
much of the empirical research in sociology and anthropology attends to how
notions of right and wrong are socially formed, policed and reproduced.
The need for coupling the normative and the empirical has been evident in this
volume in a number of respects.
Identification: In no small part, the very attention to dual use stems from
claims about the pace of development of biotechnology today and its
356
20. Ethics as
potential relevance for intentional destructive purposes tomorrow;22

however, as with other issues, the path from potential concern to publicly
recognised social problem is not one of necessity. Social, institutional and
cultural considerations are the reasons certain topics become treated as
serious problems. In the case of the topic of this book, the year 2001 remains
of paramount importance.
Specification: Moving from a general recognition of a problem to specifying
what is pressing and why are often informed by detailed argumentation. The
chapter by Kelle, for instance, delineates the multiple strands of synthetic
biology in order to map variations in kind and severity of potentials. As
he also argued, much of the policy attention to the governance of synthetic
biology has been limited to particular strands of itmost notably, DNA
synthesis. The reasons for and the implications of such narrow framings are
part of what empirical research needs to establish.
Appreciation: Part of ensuring that sensible and appropriate measures are
devised is understanding how practitionerssuch as laboratory researchers
and managersassess the possibilities of their work. As Connell so vividly
and honestly shares with readers, even those with longstanding experience in
preventing the deliberate spread of disease can lose sight of the implications
of their own work. Understanding the organisational and social reasons
for such blindness is vital in knowing how to respond. As Chambliss has
warned, everyday routines in organisations often delimit ethical scrutiny.23
Contextualisation: Benzuidenhouts chapter explored how recognition
and responses to dual-use issues are likely to vary systematically due to
research environments. Without regarding information about the day-to-day
experiences of those in labs, proposals for what needs to be done can be
compelling in the abstract but practically irrelevant.24
Evaluation: Assessing the measures undertaken to prevent the hostile use
of the life sciences requires empirical data.25 Without this, action can be
misplaced. For instance, the review of civilian research experiments, science
journal manuscripts and public grant proposals for their security risks has
excited much controversy.26 As the most prominent instance of this, in 2003,
22 See Rappert, B. 2007, Biotechnology, Security and the Search for Limits: An Inquiry into Research and
Methods, Palgrave, London, ch. 1.
23 Chambliss, D. F. 1996, Beyond Caring: Hospitals, Nurses, and the Social Organization of Ethics, University
of Chicago Press, London.
24 Context should not, however, simply be thought about as a static backdrop directing action. Financial
and organisational resource constraints motivate the search for different forms of research collaboration that,
in turn, structure possibilities for perception.
25 Douglas, T. and Savulescu, J. 2010, Synthetic biology and the ethics of knowledge, Journal of Medical
26 King, op. cit., pp. 43246; Frisina, M. 2006, The application of medical ethics in biomedical research,
Cambridge Quarterly of Healthcare Ethics, vol. 15, pp. 43941; and Tyshenko, M. 2007, Management of
natural and bioterrorism induced pandemics, Bioethics, vol. 21, no. 7, pp. 3649.
357
32 prominent (Western) journal editors committed themselves to enact peerreview procedures to assess the risks and benefits of individual manuscripts.
These were meant to determine whether articles needed to be modified
or withdrawn because the potential harm of publication outweighs the
potential societal benefits.27 One concern frequently expressed has been
that such measures would jeopardise the free flow of information that is vital
to civilian science. Such fears though are arguably misplaced for two reasons
1. experience from at least 2004 has indicated that the riskbenefit logic
of the formal review processes means that few manuscripts, grant
applications or experiment proposals are identified as posing dual-use
concerns (let alone are censored; there are no cases of civilian work being
categorically withheld)28
2. evidence from social studies of science has illustrated how the exchange of
resources and information in research is frequently subject to negotiation
in practice.29
Such observations raise concerns about the ultimate purposes and prospects of
formal oversight procedures. They might either miss important developments or
impose needless layers of bureaucracy.
Operationalisation: Policy initiatives can also be misjudged if they misconstrue
the unit at which ethical decisions are taken and moral standards enforced.
As Miller argues in Chapter 12, reducing ethical concerns to individual
decision-making would be fundamentally flawed. Science must be treated as
a group and community enterprise.
The need for ethical analysis informed by empirical research is readily
acknowledged today in bioethicsindeed, many contend it has always been
so.30 The rub is not with whether empirical and ethics can be placed side by
side, but rather how. The central difficulty is that of bringing together what
is and what ought to be. The facts of some issuefor instance, how those
in a lab regard their professional responsibilitiesdo not resolve what should
be the case.31 Clearly, it is not possible to justify standards for morality on
empirical data alone, even if in practice what is often shapes notions of what
ought to be.
27 Journal Editors and Authors Group 2003, Proceedings of the National Academy of Sciences, vol. 100, no.
4, p. 1464.
28 See Nightingale, S. 2011, Scientific publication and global security, JAMA, vol. 306, no. 5, pp. 5456;
and Rappert, B. 2008, The benefits, risks, and threats of biotechnology, Science & Public Policy, vol. 35, no.
1, pp. 3744.
29 Rappert, 2007, op. cit., ch. 2.
30 Herrera, C. 2008, Is it time for bioethics to go empirical? Bioethics, vol. 22, no. 3, pp. 13746; and Hurst,
S. 2010, What empirical turn in bioethics? Bioethics, vol. 24, no. 8, pp. 43944.
31 See Bosk, C. 2000, The sociological imagination and bioethics, in C. Bird, P. Conrad and A. Fremont
(eds), Handbook of Medical Sociology, Prentice Hall, Upper Saddle River, NJ, p. 403.
358
20. Ethics as
Recognising this though is just the start of unpacking how empirical study can
contribute to the normative. As de Vries and Gordijn argue, the conclusion of
an empirical-ethical inquiry is not necessarily a moral judgment or principle.
Nor are the conclusions of these studies necessarily based solely on empirical
results.32 In other words, the tasks undertaken as part of empirical ethics are
many and varied. Seeking some way to move from description to prescription is
just one possibility. Others include detailing conduct, understanding conditions
and processes of meaning making, acknowledging alternative ethical reasoning,
testing the feasibility of moral precepts, and refining ethical theory.
Ethics for and for

One succinct gloss of the previous sections would be this: ethics is contested
space and a contested label. Whether some concern is treated as ethical and
how, for instance, are part and parcel of the way issues become identified as
problems that demand action. With this recognition, the vocabulary of use and
abuse requires scrutiny. The previous sections included consideration of how
framing the potential of the life sciences in terms of destructive and peaceful
purposes is indebted to particular ways of thinking.
This section takes the use and abuse in a different direction. Instead of
focusing on the life sciences, it turns to the contrasting potentials of bioethics.
So rather than just treating research and innovation as the sources of problems
to which bioethics offers some sort of antidote, this section asks how pursuing
ethical analysis is associated with dangers. If dosage can be the only thing
that distinguishes a medicine from a poison then the same applies to ethical
prescriptions. Each of the possible positive utilities for bioethics given below
is questioned as to how it contains the seeds for dubious outcomes. In this
sense, seeking to input bioethics into current dual-use debates is treated as
dilemmatic.33
Advising
Bioethicists are often called upon to provide advice in controversies. The 2010
New Directions: The Ethics of Synthetic Biology and Emerging Technologies34 report
of a US presidential commission is one such instance. It recommended various
32 de Vries, R. and Gordijn, B. 2009, Empirical ethics and its alleged meta-ethical fallacies, Bioethics, vol. 23,
no. 4, pp. 193201.
33 Billig, M. 1996, Arguing and Thinking, Cambridge University Press, Cambridge; and Billig, M., Condo, S.,
Edwards, D., Gane, M., Middleton, D. and Radley, A. 1989, Ideological Dilemmas, Sage, London.
34 Presidential Commission for the Study of Bioethical Issues 2010, New Directions: The Ethics of Synthetic
Biology and Emerging Technologies, US Department of Health and Human Services, Washington, DC.
359
reforms to oversight and reporting procedures for research that were said to be
congruent with widely shared ethical principles. Through such advice giving,
ethics has a chair at the table of high-level public policy.
The need for considered ethical advice has been made before. Green called for
a portion of the multi-billion-dollar per year funding for bio-defence in the
United States to be set aside to examine its ethical, legal and social implications
(ELSI).35 This recommendation was based on the precedent of the Human
Genome Project. And yet, despite seeking to build on this example, Green also
acknowledged various problems with the Human Genome Project ELSI program:
its policy consequences, coordination and focus.
At a more theoretical level, attempts to advise can be misguided if they are
based on a faulty understanding of how analysisbe it ethical or empirical
is made to matter. In an idealised model of policy, analysis acts as an input
into rational decision-making processes through establishing facts and selecting
between options. Medical ethics and bioethics are arguably particularly aligned
with this inputting model given the centrality placed on decision-making. In
contrast, students of policymaking have forwarded a complex path between
knowledge and action, one characterised by iterative movements without
clearly demarcated beginnings or endings. Indeed, some have gone so far as to
question whether it would even be desirable for policymaking to be solely or
largely based on analysis.36 Instead of portraying it as some sort of authoritative
input, the role of analysis has been defined in more limited termssuch as
helping to identify concerns not widely recognised.
Therefore, a danger with greater bioethical scrutiny about dual use would be
that it invests a greater significance in the former than it can bear. In practice,
the arguments of ethics might well give a false air of justification to decisions
taken for altogether different reasons.37 Another potential problem is that of
misconstruing the nature of the choices faced. Hoffmaster and Hooker have
argued ethical dilemmas are often radically open-ended and indeterminate.
Herein,
not only is the solution unknown, but the problem itself is initially not
well defined, and the values that ought to drive its investigation and the
35 Green, S. 2005, E3LSI research: an essential element of biodefense, Biosecurity and Bioterrorism, vol. 3,
no. 2, pp. 12837.
36 Lindblom, C. and Cohen, D. 1979, Usable Knowledge, Yale University Press, New Haven, Conn.; Palumbo,
D. and Hallet, M. 1993, Conflict versus consensus models in policy evaluation and implementation, Evaluation
and Programme Planning, vol. 16, pp. 1123.
37 For a consideration of this point, see Winner, L. 1992, Citizen virtues in a technological order, Inquiry,
vol. 35, nos 34, pp. 34161.
360
20. Ethics as
valid methods to do so are unknown, unclear, or in dispute, as are the

set of applicable theoretical models, the solution set, and the criteria for
successful resolution.38
The danger of trying to apply bioethics to determine the proper course of action
in such situations is that the conventional concepts of bioethics can provide a
sterile and static understanding of the issues at stake. Any recommendations
that follow then are likely to have a stunted potential.
Guiding
If bioethics might not be thought of as wielding the decisive hand in policymaking,
a more modest goal would be to suggest that it provides guides for assessing the
morality of acts. Along this line, some have sought to derive general guidance
for scientists and research organisations from ethical principles.39 Kuhlau and
colleagues, for instance, outlined various criteria for identifying harm within
the moral responsibility of scientists.40 In line with the ethical principle of nonmaleficence, for instance, they argued that [r]esearchers should be responsible
not only for not engaging in harmfully intended activities but also for research
with harmful implications that they can reasonably foresee.41
The ability of ethics to derive moral obligations has been a topic of debate since
its inception as a field of study. Much of that discussion has turned on the
prospects for high-level principles to inform situated action. A frequent refrain
against principlism is that it is flawed because what it means to adhere to a
principle (such as autonomy or justice) is always indeterminate at some level.
As instances of ethical concern are never identical, the future application of
normative standards cannot be set out once and for all. Individuals must manage
the relevance of principles, what it means to follow or deviate from them, and
what consequences are likely to follow from violations. For instance, while
the obligation to prevent harm often figures as an ethical bar to bioweapons
development, this prescription does not have the same import for conventional
(read: commonly accepted) forms of weaponry.42 Enabling harm is a goal in
many areas of research. As such, dual-use discussions are often coloured by a
normative starting orientation about which principles matter and why, rather
than deriving a sense of the right course of action from the principles themselves.
38 Hoffmaster, B. and Hooker, C. 2009, How experience confronts ethics, Bioethics, vol. 23, no. 4, pp. 21425.
39 See Ehni, H. J. 2008, Dual use and the ethical responsibility of scientists, Archivum Immunologiae Et
Therapiae Experimentalis, vol. 56, pp. 14752; and Green, S., Taub, S., Morin, K. and Higginson, D. 2006,
Guidelines to prevent malevolent use of biomedical research, Cambridge Quarterly of Healthcare Ethics, vol.
15, pp. 4329.
40 Kuhlau, F., Erikson, S., Evers, K. and Hglund, A. 2008, Taking due care: moral obligations in dual use
research, Bioethics, vol. 22, pp. 47787.
41 Ibid., p. 481.
42 Rappert, B. 2012, How to Look Good in War, Pluto Press, London, ch. 6.
361
A concern with principlism then is that it has failed to acknowledge this

indeterminacy in favour of adopting a sense of certainty and inevitability. In
response, leading proponents have acknowledged the need for the specification
and interpretation of principles in relation to specific situations.43 Yet, whatever
the sophistication and subtlety within academic texts about how to do these
tasks, some have voiced concern about the blunt way in which principles figure
within day-to-day institutional practice.44
Deliberating
A different way of thinking about the utility of ethics is to focus on process. If
general ethical analysis has a limited ability to definitely determine the proper
course of action then it can help structure democratic deliberation.45 Herein, as
individuals schooled in conceptual analysis and argumentative logic, ethicists
can ensure the quality of discussions (even if they do not occupy elevated moral
positions). This procedural expertise might be exercised through clarifying
reasoning, ensuring consideration of neglected topics, making hidden values
explicit, providing comparative examples, and so on.
Each of these notionally procedural contributions though could be questioned
regarding how they import in (unrecognised) value commitments. The
prescriptive and procedural dimensions of bioethics are exemplified in current
disputes about the merits of the precautionary principle.46 Any discussion of
this notion needs to begin with the recognition that it comes in a multitude
of versions. While in general terms precaution is aligned with not requiring
conclusive demonstration of harm to prompt concern or even action, many
formulations of it exist.47 It is possible, for instance, to distinguish between
43 Beauchamp, T. 1995, Principalism and its alleged competitors, Kennedy Institute of Ethics Journal, vol.
5, no. 3, pp. 18198.
44 Marshall, P. 2001, A contextual approach to clinical ethics consultation, in B. Hoffmaster (ed.), Bioethics
in Social Context, Temple University Press, Philadelphia.
45 Boniolo, G. and di Fiore, P. P. 2010, Deliberative ethics in a biomedical institution: an example of
integration between science and ethics, Journal of Medical Ethics, vol. 36, pp. 40914.
46 In relation to how the prescriptive and the procedural meld, a concern for some is that attention to
procedure is a means of decision-making, at least by default. A criticism made of precautionary approaches
but one that could be made of any attempt to promote and conduct deliberationis that it ends up significantly
delaying decisions, and that such a delay thereby favours some over others. It is notable in this regard that
precaution has become a byword for inaction and hesitancy in many areas of science policy. See Ledford, H.
2011, Hidden toll of embryo ethics war, Nature, vol. 471, p. 279.
47 Stirling, A. 2008, Science, precaution, and the politics of technological risk converging: implications
in evolutionary and social scientific perspectives, Annals of the New York Academy of Sciences, vol. 1128,
pp. 95110.
362
20. Ethics as
argumentative, process-orientated kinds that establish guidelines for what sorts

of arguments are legitimate, and those prescriptive decision-orientated kinds
that resolve what action should be taken.48
As a decision rule for making political deliberations, this principle has its
detractors. While claiming the need to safeguard against harms in the face of
uncertainty and ignorance is reasonable, specific enactments of precaution are
seen as going too far. Clarkes chapter, for instance, offers a sustained critique
of the so-called strong version of the principle that exclusively considers
potential costs of a particular action (such as the 1994 Final Declaration of the
First European Seas at Risk Conference). Within this way of thinking, any
doubt about severe consequences is treated as providing adequate grounds
for stopping an activity from going aheadthe result being a paralysing
conservatism that can create larger risks than those forgone.49
Whatever the merits of formulations of the precautionary principle for making
decisions, as process guides they can structure how troublesome questions are
approached. For instance, with whom the burden of proof rests to substantiate
claims of harm or benefit, to whom and with what level of certainty are important
issues for any topic of controversy. Box 20.1 details some of the procedural
proposals for how evidence and onus could figure in responses to the destructive
potential of the life sciences. The relative merits of these alternatives could be
further informed by ethical analysis regarding their procedural dimensions.
Box 20.1 Handling Risks and Uncertainty in the Review of Research
The National Science Advisory Board for Biosecurity (NSABB) was in large part formed
following the recommendations of the US National Academies report Biotechnology Research
in An Age of Terrorism. A central task of NSABB is the development of recommendations
on guidelines for the oversight of dual-use research, including guidelines for the risk/benefit
analysis of dual-use biological research and research resultsa for the US Federal Government.
The guidelines for risk/benefit analysis and oversight represent attempts to define, evaluate
and handle concerns about the dual-use potential of research through the creation of formal
bureaucratic procedures.
The split NSABB has offered between research that might have some sort of dual-use potential
and that which is of concern has been of paramount importance. For the board, the term
dual-use research is used to refer in general to legitimate life sciences research that has the
potential to yield information that could be misused to threaten public health and safety and
other aspects of national security such as agriculture, plants, animals, the environment, and
materiel.b In contrast, dual use research of concern refers to the subset of life sciences
research with the highest potential for yielding knowledge, products, or technology that
could be misapplied to threaten public health or other aspects of national security.c
48 Sandin, P., Peterson, M., Hansson, S., Rudn, C. and Juthe, A. 2002, Five charges against the precautionary
principle, Journal of Risk Research, vol. 5, no. 4, pp. 28799.
49 See as well Harris, J. 2001, Introduction: the scope and importance of bioethics, in J. Harris (ed.),
Bioethics, Oxford University Press, Oxford; and Douglas and Savulescu, op. cit.
363
Because it is imagined that few experiments will need to be given security review, the
emphasis has been with devising a non-demanding tick-box first stage that should exclude
the majority of research from further formal consideration.d In this regard, NSABB has
proposed that the initial review of whether or not research is of concern be undertaken
by the principal investigator (that is, the senior project leader). Herein, this person would
ask of their work, based on current understanding, can [it] be reasonably anticipated to
provide knowledge, products, or technologies that could be directly misapplied by others to
pose a threat to public health and safety, agricultural crops and other plants, animals, the
environment, or materiel.e
To state that assessors must be able to reasonably anticipate a direct threat based on current
understanding sets a high threshold for proof. At this initial stage of the review process,
the determination of the status of research is not intended to impose significant demands
on principal investigators. Should research be found to match the criterion then it would be
subjected to institutional risk review.f
Such an approach can be contrasted with an alternative oversight model proposed by the
Center for International and Security Studies at Maryland (CISSM). This is envisioned as an
international legally binding system requiring the licensing of personnel and research facilities.
The Maryland system also involves independent peer review. An oversight body needs to
approve work going ahead, rather than the investigators making the initial determination. This
was justified on the basis that [i]n addition to having a self-interest in seeing their research
proceed, such individuals are also unlikely to have the security and other expertise necessary
to recognize the possible dual use risks of their work.g
The criteria proposed as part of the riskbenefit analysis in the Maryland system also go
further than the NSABB proposal. As part of assessing research, for instance, individuals
are required to consider whether the same experimental outcome could be pursued through
alternative means, whether the research is being done in response to a validated (credible)
threat, and whether it will yield results definitive enough to inform policy decisions. Such
questions place additional demands on those taking part in the assessment process to those
as part of NSABB recommendations. They also require forms of knowledge that the average
principal investigator is unlikely to posses. As another contrast to the NSABB proposals, the
Maryland one provides a metric for evaluating research based on the responses given to the
criteria mentioned.
At the heart of such alternative policy options is the matter of expertise and how this should
be exercised. While NSABB devolves much of the decision-making down to senior individual
scientists who are aided by others, the CISSM proposal places much more emphasis on a
diverse range of expertise structured through mandatory requirements.
a
CharterNational Science Advisory Board for Biosecurity, 16 March 2006, p. 1.
National Science Advisory Board for Biosecurity (NSABB) 2007, Proposed Framework for the
Oversight of Dual Use Life Sciences Research: Strategies for Minimizing the Potential Misuse of Research
Information, National Science Advisory Board for Biosecurity, Bethesda, Md, p. 4.
b
Ibid., p. 16.
As in comments made during National Science Advisory Board for Biosecurity, 20 March 2006.
NSABB, 2007, op. cit., p. 17.
Ibid., Appendix 4.
Harris, E. 2007, Dual use biotechnology research: the case for protective oversight, in B. Rappert
and C. McLeish (eds), A Web of Prevention: Biological Weapons, Life Sciences and the Governance of
Research, Earthscan, London, p. 120.
g
364
20. Ethics as
Legitimating
If legitimacy is a function of being in accordance with rules and procedures
justified by shared societal beliefs50 then one of the roles often sought of ethics is
to legitimate policies and practices. Whether through contributing expert moral
reasoning or facilitating deliberation, bioethicists (and others) are often called
up to ensure support for core social institutions. An often-voiced reservation
with such a purpose is that the basis for accord might be less than warranted or
genuine. In other words, rather than legitimacy being positively secured, it is
skilfully manufactured through a language supplied by ethics.
Chambliss, for instance, offers a highly critical evaluation of how medical ethics
and bioethics training figure within hospital practice. One of the starting points
for this is the view that organisational hierarchies are often the source of the
dilemmas experienced by professionals.51 Nurseswith relatively little formal
power in relation to doctors or administratorsoften struggle with the tension
between doing what they think is right and doing what conforms to official
policy. As a result, it is not enough for them to receive high-minded instruction
about moral principles in order for them to do what is right. For Chambliss,
ethical training in the form of abstract, principle-based talk cannot only be
irrelevant to lived experience, but it can also serve to give a false impression
that dilemmas and challenges are being acknowledged and addressed while, in
practice, ethics instruction reinforces relations of hierarchy. Winner too cautions
against the way moral categories and ethical arguments are often forwarded
without attention to the roles and institutions needed to enable notions of the
good to translate into deeds.52 In the absence of such opportunities, ethical
analysis all too easily validates the status quo.
More subtly than this, ethics can also act to reinforce the perception of shared
beliefs and values. Take the case of the previously mentioned dual-use review
procedures initiated by civilian journals, funders and research organisations
since 2004. While they differ in specifics, each proposes a riskbenefit
calculation. Expected societal gains from research are to be measured against
possible security threats, the balance between the two indicating what should
be done. As such, the review procedures are forwarded as embodying core
characteristics of rational decision-making. Ethicists have adopted and endorsed
this rationalistic framework of balancing risks and benefits.53
50 To paraphrase Green, P. and Ward, T. 2004, State Crime, Pluto, London, p. 3.

51 Chambliss, op. cit.
52 Winner, op. cit.
53 See, for instance, Krohmal, B. and Sobolski, G. 2006, Physicians and the risk of malevolent use of
research, Cambridge Quarterly of Healthcare Ethics, vol. 15, pp. 4414; and Kuhlau et al., op. cit.
365
What has not been elaborated is how risk could be calculated (see by way
of contrast the aforementioned CISSM system for an alternative framework).
Certainly no form of detailed calculus has been set out. Even putting to the side
a demand for exactitude, it is not at all clear, for instance, how reviewers could
specify the risks from the destructive use of fundamental scientific knowledge.
The users, situations, time frame, specific contribution of that knowledge, and
so on, are not well defined. The absence of significant past experience of the
deliberate spread of disease that might at least provide data points for assessing
risks and benefits also frustrates undertaking reviews. In the exceptionally
few instances of experiments where some science bodies have judged risks
in excess of benefitsas in the initial (but subsequently revised) NSABB
recommendations for publication redactions with research on the transmission
of a modified H5N1 virus in a ferret modeldetailed costbenefit calculations
have not been forwarded. Instead, general appeals to notions such as precaution
have been given.54
Perhaps most fundamentally, what positive values or negative implications might
be relevant for weighing are not clear. From one defensive standpoint, it might be
vital to identify and publish research that raises concerns (so as to confirm such
fears and to devise countermeasures). This would seem to be the rationale that
informed the Defense Advanced Research Projects Agency decision to fund the
synthesis of poliovirus.55 In Nancy Connells chapter, similarly, demonstrating
hyper virulence in a mouse model was seen as providing the grounds for future
follow-on funding. Thus, what should count as a problem is not simply poorly
defined, but also open for radically opposed interpretations in the first place.
Legitimation dangers do not just include giving undue credence to certain
outcomes. Another danger is perpetuating faith in the presumptions, beliefs
and competencies that underpin the proposal for what should be done. Rockel,
for instance, has offered a trenchant critique of the doctrine of double effect as
applied to modern warfare.56 As contended, its underpinning distinctions related
to foreseeable effects and intentions have provided the material for papering
over systematic deficiencies in military action. In relation to the themes of this
volume, it should also be noted that within the discussion of risks and benefits,
the non-destructive applications are often assumed to fall wholly on the plus
side. This thinking is in line with many public portrayals of science; however,
the extent to which biomedical research is linked to improvements in human
54 Cohen, J. and Malakoff, D. 2012, NSABB members react to request for second look at H5N1 flu studies,
Science, 2 March; and Imperiale, M. 2012, Presentation to Dual-Use Research and Biosecurity: Implications
for Science, Governance and the Law, The Hague, 12 March.
55 Selgelid, M. and Weir, L. 2010, The mousepox experience, EMBO Reports, vol. 11, pp. 1824, <http://
www.nature.com/embor/journal/v11/n1/full/embor2009270.html> (viewed 23 April 2003).
56 Rockel, S. 2009, Collateral damage: a comparative history, in S. Rockell and R. Halpern (eds), Inventing
Collateral Damage: Civilian Casualties, War, and Empire, Between the Lines Press, Toronto, pp. 196.
366
20. Ethics as
health can be questioned. It can even be counterproductive when it obscures

socioeconomic reasons for health problems [and] creates boundar[ies] to other
types of action that are more effective, efficient, and equitable.57
In considering the legitimating role of ethical analysis, it should be borne in
mind that the designation of a matter belonging to the domain of ethical
(or not) is part of what needs to be examined. As a case in point, Houtepen
examined how controversy about euthanasia in the Netherlands shifted over
time.58 Prior to the late 1960s physicians debated what should be done largely
among themselves. It was only in the late 1960s that others found a recognised
voice. Part and parcel of this was the explicit redefinition of euthanasia as a
matter of ethics, and, as such, a matter where varied perspectives from the
public at large had to be brought in. Later, with the introduction of formalised
clinical routines and policies informed by ethicists, the overt ethical framing
of practice waned. Beyond this specific case, what is deemed technical versus
ethical is a product of social negotiation that can readily work to exclude some
voices from being recognised. To what extent the dual-use review of research
activities is deemed a matter of ethics (which it largely has not been to date)
is tied to who needs to conduct such reviews. Whitman makes a related point
in this volume in recounting the consequences associated with how presenting
issues as dilemmas refracts our understanding of what is at stake.
Stigmatising
As contended previously, the categorical condemnation of bioweapons is
historically contingent and collectively produced. It is because ofnot in spite
ofthis thoroughgoing social basis that resistance would be offered to attempts
to sanction the employment of life-sciences knowledge and techniques for
destructive ends.59 Looking towards the future, given the numerous possibilities
for the malign applications outlined within the pages of this volume as well as
the difficulties of trying to enforce the prohibition through national security
and policing measures, stigmatisation is likely to be essential.
Taking this to be the caseand agreeing with the need to work against the
deliberate spread of diseaseimplies a certain agenda for bioethics: it should
work to find ways of strengthening and renewing stigmatisation. This needs to
take place in a manner sensitive to different possible belligerents: states, substate groups and lone individuals. How to prevent current efforts to develop
57 Sarewitz, D. 1996, Frontiers of Illusion, Temple University Press, Philadelphia, p. 150.
58 Houtepen, R. 1998, The social construction of euthanasia and medical ethics in the Netherlands, in R.
de Vries and J. Subedi (eds), Bioethics and Society, Prentice Hall, Upper Saddle River, NJ, pp. 11744.
59 In other words, the categorical nature of the prohibition in the BTWC and the 1925 Geneva Convention
is laudable not because it reflects an objective and essential truth, but rather because of the choices that
buttress it.
367
next-generation incapacitants (as discussed in the chapter by Crowley) from

leading to a wider normalisation of drugs as weapons, for instance, should be a
high priority.
While norms, stigmas and taboos have been subject to significant consideration
within the field of international relations in recent decades, much of that has
been of a classical, scholarly variety. Herein the relevance of positive normative
positions on the part of that writing is downplayed. By and large, international
relations scholars have sought to explain the formation of norms rather than
elaborating the practical skills necessary for bringing about reform. Therefore,
bioethics in its more applied forms could offer significant contributions to the
future of the prohibition against bioweapons.
Seeking such positive engagements for bioethics does not amount to promoting
blind faith in existing moral standards. This chapter has not sought such a
blind faith, even if it is always in danger of taking for granted certain normative
positions; however, doing so does require challenging the conventional way
stigma is regarded. In fields such as public health, stigma is routinely associated
with acts of prejudice and discrimination.60 HIV/AIDS would be a classic
example of where stigma leads to negative consequences. As a result, attempts
to make some scope for it as a tool in public health have been subjected to heated
criticism.61 Of course, one of the things that distinguishes talk of stigmatisation
in public health from international relations is the typical object of study: those
with illness versus state functionaries. Adopting the latter as the bearers of
negative distinction might well not animate fears about the victimisation.
Seeking to employ stigma within the international community in relation to
preventing the use of biological weapons comes with its own dangers though.
One is that the prohibition of these weapons is not neutral vis-a-vis the power
relations between nations. Insisting on the outright objectionable status of
one type of weapon while patchy controls exist for many other weaponsand
doing so in a world with starkly unequal distributions of powerserves some
more than others. Moreover, the shunning and integrating dynamics associated
with the stigma raise concerns about the importance attached to renunciation.
Undoubtedly, Libyas abandonment of its weapons of mass destruction
programs in 2003 helped secure a large measure of re-entry into the international
community. In the hindsight of 2013, the justifications for what was bought
from this act of disarmament seem questionable.
60 Meyer, I. and Stuber, J. 2008, Stigma, prejudice, discrimination and health, Social Science & Medicine,
vol. 67, pp. 3517.
61 See Bayer, R. 2008, Stigma and the ethics of public health: not can we but should we, Social Science &
Medicine, vol. 67, pp. 46372; as well as follow-on commentaries such as Burris, S. 2008, Stigma, ethics and
policy: a commentary on Bayers Stigma and the ethics of public health: not can we but should we, Social
Science & Medicine, vol. 67, pp. 4735.
368
20. Ethics as
Educating
Who needs to be educated, about what, how and by whom are longstanding
matters of commentary in ethics. The charge that those involved in medicine
and the life sciences are somehow lacking with regard to an appreciation of the
implications of their work is hardly unique to the topic of this volume.62 Yet
moving from such an appraisal to proposals for what needs to be done often
proves contentious. Approaches to ethics tuition differnotably, between
prescriptive, procedural and virtue-based varieties. Each of these is aligned
with distinct ways of thinking about individuals as moral agents, what count as
appropriate learning techniques and how value disagreement ought be handled.
Elsewhere I have considered the dilemmas, tensions and pitfalls of education
about the destructive use of life science.63 In this subsection I want to extend that
work by placing the issue of education within a wider political framework. Cribb
offers a valuable inroad into this by distinguishing types of health education.64
For him, a medical model treats education as a way of achieving health outcomes
through providing information that affects patient behaviour. A danger with
this is that healthcare workers assume a highly paternalistic role. Against the
medical model another would be to conceive of education as enabling people
to make informed choices based on their own values and preferences. A danger
with this model is that it treats individuals values and preferences as deriving
only from them as autonomous agents. An empowerment model, by contrast,
starts with asking about the factors that constrain individuals from realising
their preferences and then envisions education as part of overcoming those
barriers. In other words, the emphasis is with change rather than edification
for its own sake. A social-action model goes one step further by asking what
sort of structural changes in society (for instance, with regard to poverty and
social welfare) are necessary to achieve sought-for health gains. As with the
empowerment model, this one necessarily involves posing wider questions
about what needs reform. Individuals and groups require skills for participation
to affect change. In endorsing the social-action model, as with Chambliss,
Cribb counsels against divorcing education and training from institutional and
organisational conditions.
Such a typology offers a way of classifying the work that has taken place to date
with regard to dual-use education. The overwhelming orientation has been in
62 As, for instance, in the case of Sales, C. and Schlaff, A. 2010, Reforming medical education: a review and
synthesis of five critiques of medical practice, Social Science & Medicine, vol. 70, pp. 16658.
63 See Rappert, B. 2007, Education for the life sciences, in Rappert and McLeish, op. cit.; and Rappert, B.
2010, Introduction: education as , in B. Rappert (ed.), Education and Ethics in the Life Sciences: Strengthening
the Prohibition of Biological Weapons, ANU E Press, Canberra.
64 Cribb, A. 2005, Health and the Good Society: Setting Healthcare Ethics in Social Context, Oxford University
Press, Oxford.
369
line with the medical model. Herein, researchers or the public are expected to
take on board messages about the potential of the life sciences. The intention is to
help achieve certain thinking or behavioursuch as the competency to identify
research of concern. Situated within the international community of states, the
danger of paternalism typically associated with this model is compounded by
that of neo-colonialism. With much of the recent attention to dual use emanating
from North America and Europe, an obvious concern is that the agenda as well
as the ethical approaches employed to understand it (for example, individualist
and principle based) are indebted to strains of Western thinking.65 With the
effort dedicated to education as instruction, much less attention has been given
to empowerment or social-action models. Exceptions include the Kampala
Compact: The Global Bargain for Biosecurity and Bioscience and related DNA for
Peace report.66 Both proposed holding together biosecurity measures with social/
international development agendas. Brian Martin, too, spoke of the importance
of individuals and group empowerment in whistleblowing and dual use.67 In
the absence of such practical skills training, a danger of education is that it does
not enable positive reform. Such an outcome can lead to feelings of irrelevance,
indifference or frustration on the part of educators and the educated.
A non- research agenda

The previous sections have counselled the need for caution regarding the
commitments of our analysis: its starting points, the use of evidence and
argument, and the purposes to which it is put. In more or less direct ways,
in those sections I have suggested that among the prime challenges for ethics
include identifying moral issues and formulating them as problems in need
of redress.68 Both are inextricably tied to processes of categorising, labelling
and boundary-setting that help define but are also defined by social routines,
institutions and structures.
This section considers how these goals of identification and formulation can
be taken forward in a way that enriches bioethical engagement with dual-use
life science. The starting move in this is a shift, in a sense, backwards. Rather
than suggesting detailed engagement with this or that topic of controversy,
the proposal is to attend to the why and the how of what is not: what is not
65 For a more general discussion of this danger, see Widdows, H. 2007, Is global ethics moral neocolonialism? An investigation of the issue in the context of bioethics, Bioethics, vol. 21, no. 6, pp. 30515.
66 Available at: <http://www.utoronto.ca/jcb/home/documents/DNA_Peace.pdf> (viewed 4 April 2010).
67 Martin, B. 2007, Whistleblowers: risks and skills, in Rappert and McLeish, op. cit.
68 In line with Clouser, K. D. 1978, Medical ethics: some uses, abuses, and limitations, New England Journal
of Medicine, vol. 293, pp. 3847; and Hoffmaster, op. cit.
370
20. Ethics as
recognised in the first place or, if recognised at some level, what is not treated
as a serious problem; if regarded as a problem then what is not acted upon. In
short, the subject for scrutiny is the one of what isnt happening.
A rereading of the Australian IL-4 mousepox experiment illustrates the varied
relevancies of the non-: the potential for IL-4 to enhance virulence was
recognised prior to the mousepox publication by some experts, yet seemingly it
was not subject to much professional (let alone public) debate. In this respect,
what distinguishes the 2001 mousepox controversy from the majority of other
work with a dual-use potential is that the researchers actively voiced their
concerns beyond a closely knit expert coterie. Although counterfactual, it seems
doubtful that this work would have garnered anything like the same attention in
the absence of Ramshaws communication with New Scientist. The fraught path
to media agenda item is suggested by what happened subsequently. Despite
the profile of the researchers from the mousepox experiment, 9/11, the anthrax
attacks and much else besides, the publication of results of a follow-on study
indicating IL-4 modified mousepox resisted treatment with an antiviral agent
for smallpox garnered little notice. As another instance of the non-, only years
after the initial controversy did Ramshaw regard his research as entailing a sort
of weaponisation of sterilisation for rodents.
As well, on a different level, the way that mousepox has been one of only a
handful of so-called experiments of concern that are repeatedly put forward
speaks to the narrow, individual case-based approach that has come to dominate
framing how the life sciences might aid destructive purposes.69 Herein what
are held as mattering are the choices taken at critical ethical moments (for
example, should these results be published? Should experiments be approved?
And so on).
This framing has characterised recent debate regarding the justifications for
the proposed redaction of research undertaken by Dutch and American-based
researchers who mutated the H5N1 virus in a ferret model in such as way as
to enable it to transmit between mammals. While this case was a matter of
much controversy in early 2012, as with other experiments of concern, what
is perhaps most notable is its exceptionality. With this regard to a very limited
number of cutting-edge experiments, much less attention has been cast on what
the mundane commercialisation of science means for new biowarfare capacities.
69 As in World Health Organisation (WHO) 2011, Responsible Life Sciences Research for Global Health
Security, WHO/HSE/GAR/BDP/2010.2, World Health Organisation, Geneva, Part 2.
371
The non- as a non-issue?

Asking about what is not speaks to many of the themes raised previously in
this chapter: moral blindness, taken for granted meaning, the social construction
of moral reasoning, the lack of ethical scrutiny to the conditions that produce
ethical issues, and the need to move beyond a preoccupation with specific
decision points.
Noting such affinities prompts the wider question: to what extent has the non-
been addressed within bioethics? On the one hand, the case-based scenario
reasoning prevalent in bioethics is typically directed towards manifested
dilemmas and choices. Given the widespread technique of posing hypothetical
and real-life cases to ask what should be done, what is not taking place can
be sidelined. Attention rests with possibilities for action and agency in specific
scenarios. Certainly some approaches in bioethics, such as the case-based
casuistry ethics,70 are ill suited for directing themselves towards what is not
happening.
On the other hand, absences are also prevalent. Bioethical analysis generally
seeks to question the basis for what is treated as natural, inevitable, just so, and
so on. In some ways, the history of bioethics can be read as a history of seeking
to doubt prevalent moral conventions and priorities.
Within bioethics, non-action is often at the centre of dispute about what
is justifiable. In a review of ethical analysis of euthanasia, Holland mapped
out some of the contrasting orientations given to the distinction between
undertaking and refraining from action.71 As he argued, one common means of
differentiating between what is morally permissible and what is not is through
the language of killing versus letting dieor so-called active versus passive
euthanasia.72 Often the latter is treated as more justifiable than the former because
active euthanasia requires directed intervention. While passive euthanasia can
itself entail some sort of action (such as turning off life-support equipment),
that this is not the direct cause of death is held by some as justification for a
moral distinction between it and (the more problematic) active forms. Tooley
has countered such attempts to distinguish between killing and letting die. For
him, passive euthanasia should be regarded as morally equivalent. As such, it
is the lack of willingness on the part of doctors to intervene to hasten death
that should be seen as the prime problem. As Holland recounts, this analytical
70 Arras, J. 1991, Getting down to cases: the revival of casuistry in bioethics, Journal of Medicine and
Philosophy, vol. 16, pp. 2951.
71 Holland, S. 2003, Bioethics, Polity, London.
72 See Rachels, J. 1975, Active and passive euthanasia, New England Journal of Medicine, vol. 292, pp.
7890; and McLachlan H. V. 2008, The ethics of killing and letting die: active and passive euthanasia, Journal
of Medical Ethics, vol. 34, pp. 6368.
372
20. Ethics as
argument about moral equivalence has itself been disputed through the
contention that the widespread belief that a moral difference should be made
between killing and letting die itself provides an adequate basis for judging
which one is preferable.
In contrast, other ethicists have sought to move away from direct reference
to action/non-action as the basis for evaluating morality. One way that has
been done is by distinguishing between positive and negative duties. Herein
transgressing negative duties (such as refraining from killing) are treated as more
serious than positive ones (such as not intervening to prevent death). Yet this
is problematic because some actions (such as preventing someone from being
saved) cut across the starting distinction between killing and letting die. Still
other ethicists have advocated replacing the focus on action in debates about
euthanasia with that on agency and responsibility.73
The manner in which action/non-action is varyingly configured as relevant
speaks to the importance of how issues are identified and how they are
formulated. Action and inaction have been the locus for moral argument while
also being deemed somewhat beside the point. Thus when attending to dual use
vis-a-vis what is missing, in addition to considering the many ways (in)action
is said to matter, how the debate is framed must be considered: what is taken as
counting as (in)action, what evidence supports such claims, what implications
are said to follow.
This complicated picture of the coverage and place of the non- in ethics is
mirrored in the empirical social sciences. Again while fields such as sociology
and political science are generally preoccupied with what is taking place, what
is not happening has also figured as a subject of study. A recurring undercurrent
of the commentary by sociologists on bioethics is that it does not attend to the
structural and institutional conditions that delimit the possibilities available
to individuals. As such, sociological analysis often purports to attend to what
bioethics systematically ignores.
Another facet of the study of the non- in social research is the examination
of how social concerns about science are nullified. Much of this work starts
with Gieryns observations about how the boundaries between objectivity/
subjectivity, natural/social realms and expert/lay knowledge are routinely
managed within the practice and portrayals of scientists.74 Such boundary work
is part and parcel of how control is maintained over the goals and standards of
science. Along these lines, Cunningham-Burley and Kerr examined how adept
boundary work enabled geneticists to secure the cognitive authority necessary
to secure funding, while placing themselves as authority figures for speaking
73 Coggon, J. 2008, On acts, omissions and responsibility, Journal of Medical Ethics, vol. 34, no. 8.
74 Gieryn, T. 1999, Cultural Boundaries of Science, University of Chicago Press, Chicago.
373
about the social consequences of genetics, while also distancing themselves

from the responsibility for negative consequences.75 Specifically in relation to
the non-, Firth et al. have sought to chart how boundary work was part of the
creation of a settled morality76 in infertility clinics.77 Boundaries management
was central to securing agreement over many issues that were contentious
outside the lab. Ethical concerns were not identified with day-to-day practices
within clinics as part of what was referred to as the no ethics repertoire. The
result of both aspects of settled morality is that those in the clinics rendered
their practice immune from outside interference.
Non-groundings
As suggested in the previous subsection then, within both ethics and social
science, uneven regard is given to what is not taking place. As a result, attempts
in relation to the dual-use life sciences to combine normative justification with
empirical analysis face two types of problems: how the non- is treated within
fields of study and how these fields can be brought together. In relation to what is
absent, what really needs attention is the status quo and therefore what is likely
to foil efforts to move on from it. As part of this, the normative and the empirical
must seek to identify each others underlying assumptions. The non- as a topic
of study in this respect proves advantageous. This is so because attending to
what is absent requires not just clarifying thinking, but instead also inquiring
about the conditions under which quandaries arise and are structured.
If the non- has advantages as a topic for study in fostering this dialogue
between the empirical and the normative, it also has drawbacks. A prominent
one is its open-endedness. In de-anchoring analysis from something definite
to something that could be happening, the range of relevant considerations
multiplies manifold. Appeals to research-community interactions, time and
organisational constraints, awareness, professional socialisation,78 widespread
cultural myths and narratives,79 and so on, are among the reasons that could
be cited to explain the lack of professional attention to dual-use issues. And
since those considerations relate to what is not happening, proving their
counterfactual relevance is not straightforward. This in turn makes choosing
75 Cunningham-Burely, S. and Kerr, A. 1999, Defining the social, Sociology of Health & Illness, vol. 21,
no. 5, pp. 64768.
76 From Hoffmaster, B. 1990, Morality and the social sciences, in G. Weisz (ed.), Social Science Perspectives
on Medical Ethics, Kluwer Academic, Boston.
77 Frith, L., Jacoby, A. and Gabbay, M. 2011, Ethical boundary-work in the infertility clinic, Sociology of
Health & Illness, pp. 116.
78 As suggested in Sture, J. 2009, Educating scientists about biosecurity: lessons from medicine and
business, in Rappert, 2009, op. cit.
79 Gordon, D. and Paci, E. 1997, Disclosure practices and cultural narratives, Social Science and Medicine,
vol. 44, no. 10, pp. 143352.
374
20. Ethics as
between explanations a demanding task. While empirical and normative

argument can be marshaled to make a case that is persuasive for many, this is a
case that will need to be made. The non- in relation to dual use faces a similar
demand. Barring recourse to an objective sense of social problems that could be
read back to determine exactly how much of a concern is really posed by the
destructive use of science,80 normative and empirical questions can be raised
about claims regarding what should be recognised as a problem but is not.
Methodology
From the previous argument of this chapter it is possible to draw some important
conclusions for the study of what is not going on in relation to concerns about
the destructive use of the life sciences. First, while some of the existing literature
in bioethics and social sciences speaks to how issues are not recognised or
what actions are not taken, more systematic thought is needed. Second, with
the somewhat inevitable reliance on the counterfactual and the speculative,
arguments about the non- require careful justification. Although it might be
possible to make a case for why something is not happening that is persuasive
to many, it is likely to be disputed too. Third, a dialogue must be established
between the normative and the empirical. Locating a discussion of the non- of
dual use within the emerging literature about empirical ethics could provide
additional (normative) analytical resources for probing compared with those
typically called upon by social scientists.
In thinking in more specific methodological terms about how to study nonissues and non-actions, in this subsection I want to advance two considerations:
1) comparative examination, and 2) interventionist inquiry.
Comparative examination
One of the strategies used in researching the exercise of power has been to match
up situations that shared pertinent similarities in order to account for their
variations. For instance, the responses of communities affected by air pollution
have been juxtaposed to inquire about the reasons for those differences. In a
related vein, using time as the variable, periods of major social disturbance have
been examined for the opportunities created (and closed) for unconventional
ideas and practices.
80 For a consideration of objectivist and constructivist orientations to dual use as a social problem, see
Rappert, 2007, op. cit., ch. 1.
375
With respect to dual-use issues, it is possible to take inspiration from this

comparative strategy. For instance, a given line of experimentationsuch as
the insertion of IL-4 on pox virusescould be examined regarding why some
failed to raise dual-use concerns while others did.
To take comparison in a different direction, another tack would be to put
side-by-side efforts to dissociate science from bioweapons concerns actively.
For instance, in recent years an international do-it-yourself (DIY) community
has emerged, encouraging the formation of small-scale open-access biological
labs. While much of the DIY bio community emphasises the democratisation of
science in order to address problems ignored by corporations and universities,
the proliferation of capabilities beyond accredited labs has been repeatedly
associated with fears about bioterrorism.81 Leaders in the DIY bio community are
seeking actively to distance their work from such fears, in part by establishing
codes of conduct.82 Likewise, through innovations in art, the Critical Art
Ensemble has sought to debunk the link between biology and bio-threats.83
Such attempts to move from is is not could be compared with attempts to
move from is not is. Along these lines, civilian researchers who have raised
dual-use threats could be studied with a view to the reactions they experienced
from colleagues, funders and others. The pushback and resistance faced by
organisational whistleblowers would likely prove a salient comparison.84 In
the military area, the aim of US Defense Advanced Research Projects Agency
funding to link basic research to the protection against bioterrorism (as through
the synthesis of poliovirus that brought its own dual-use fears) would be one
such effort to move from is not to is.
Another comparative tact would be to juxtapose the evaluations made by
different communities. For instance, security experts could be enlisted to
identify lines of research they believe pose security risks but which have not
been the subject of much scrutiny to date. Then those working in the identified
areas could be approached to determine whether practitioners agreed with the
assessments of security experts, the extent to which researchers have identified
dual-use concerns with their work, and the reasons those concerns have (and
have not) been communicated. Much the same could be done for ethicists and
how their ranking of what ought to be a matter of concern compares with those
in the life sciences.
81 Ledford, H. 2010, Garage biotech: life hackers. Amateur hobbyists are creating home-brew molecularbiology labs, but can they ferment a revolution? Nature, vol. 467, pp. 6502; and Nature 2010, Garage
biology: amateur scientists who experiment at home should be welcomed by the professionals, Nature, vol.
467, p. 634.
82 As in the DIYbio Continental Congress held on 8 May 2011 attended by the author.
83 Critical Art Ensemble n.d.,Bodies of Fear in A World of Threat, <http://www.critical-art.net/mp.html>
(viewed 4 April 2010).
84 Martin, B. 2007, Whistleblowers: risks and skills, in Rappert and McLeish, op. cit.
376
20. Ethics as
Interventionist inquiry
That non-issues are non-issues and non-actions are non-actions speak to the way
in which a spirit of intervention needs to infuse their study.
To expand, the process of questioning practitioners about matters of dual use is
likely to be an act of questioning assumptions, priorities and world views. For
instance, since 2004 Malcolm Dando and I have conducted seminars for university
faculties and other public research centres in order to inform participants about
current life-science security developments as well as to generate debate about
how research findings should be communicated, whether experiments should
be subject to institutional oversight and what research should be funded.
More than 130 seminars have been undertaken in 15 countries (ranging from
the United Kingdom to Uganda and Japan to Argentina), with more than 3000
participants. While it has been possible to generate lively (but bounded)85
discussion about dual-use issues at these events, as interactions they required
careful management because they asked participants to think about their work
anew. As a result, what was said, how, to whom and when were all subject to
lengthy methodological consideration. The decision to run seminar discussions
akin to focus groups in which attendees were encouraged to deliberate with
each other was itself the result of the limitations experienced in a one-to-one
interview format.86
The manner in which probing about non-issues de facto amounts to a form of
intervention suggests the need to consciously attend to how this intervention
is conducted. Overall, what is required is a systematic process of planning
and execution that allows for learning and experimentation. As part of this,
inquiry should be thought of as a practical, intellectual, action-orientated
and consequentialist form of action.87 Kurt Lewins often quoted suggestion
that if you want to truly understand something, try to change it indicates
the potential for deliberate intervention to yield insights not readily obtainable
through unobtrusive means.
In fields of social science such as action research, such practical inquiry is
linked to the aim of transforming social relations.88 As with the empowerment
and social-action models noted in the previous section, the factors that
frustrate change require attention. This practical step entails incorporating
85 See, for instance, Rappert, 2007, op. cit., ch. 5.
86 Ibid., ch. 2.
87 Dewey, J. 1929, The Quest for Certainty, George Allen & Unwin, London.
88 See, for example, Ospina, S., Dodge, J., Godsoe, B., Minieri, J., Reza, S. and Schall, E. 2004, From consent
to mutual inquiry, Action Research, vol. 2, no. 1, pp. 4769; and Winter, R. 1996, Some principles and
procedures for the conduct of action research, in O. Zuber-Skerritt (ed.), New Directions in Action Research,
Taylor & Francis, London; and Winter, R. 1998, Managers, spectators and citizens, Educational Action
Research, vol. 6, no. 3, pp. 36176.
377
positive normative goals into the design of inquiry. Central to a robust process of
transformative intervention is to ensure that a conversation takes place between
the methods of inquiry and its normative aims. The latter should promote
scrutiny regarding the fallibility and commitments of the methods employed.
Also, the methods should enable the refinement and revision of what is held as
necessary and desirable. Doing so not only requires a certain kind of intellectual
understanding, but also practical skills.
What seems essential in studying what is absent is to find means of questioning
taken-for-granted assumptions about what counts as an ethical or a social
problem in the first place. Rather than going out and probing straightforwardly
overt, recognised issues widely labelled as ethical or contentious, research
techniques and strategies must help to cultivate thinking afresh in order to
avoid confirmation bias, to encourage alternative hypotheses and to embrace
negative evidence.
Inquiry about non-issues and non-actions then cannot be conceived simply as
an attempt to reveal holes in understanding. Instead, it must be a project of
questioning the historical, political and situational bases for how understandings
are formed and thereby what is counted as missing in the first place. As such,
ensuring inquiry interrogates its own starting points is vital. One interesting
direction to take the non- is to consider the hows and whys regarding ethicists
lack of engagement with dual-use life science. Some preliminary reflections
have already been given on this matter;89 however, undertaking systematic
empirical research might inform an understanding of not only the priorities and
presumptions of ethicists, but also therefore the likely limits of existing academic
disciplinary resources. Moreover, such a line of empirical investigation provides
an opportunity for bringing to the fore questions about how the normative and
the empirical can be combined.
At stake is the question of how facts, figures, concepts and arguments should be
made sense of in order to assess whether ethicists themselves have been remiss
for their past level of regard. The contention that some issues are being neglected
compared with others is commonplace in bioethics, with its attentiveness to
distributive justice. At times this extends to commentary on bioethics own
agenda.90 For both, the justifications for claims are open to disagreement in
relation to their underlying ethical assumptions. Appeals to consequences
versus duties versus rights, for instance, can result in far different assessments
about what is lacking from the agenda of bioethics. Each appeal is also reliant on
89 Selgelid, M. 2010, Ethics engagement of the dual-use dilemma: progress and potential, in Rappert, 2010,
op. cit.
90 For examples of this, see Selgelid, M. 2008, Ethics, tuberculosis and globalization, Public Health Ethics,
vol. 1, no. 1, pp. 1020; and Selgelid, 2005, op. cit.
378
20. Ethics as
different types of evidencing to substantiate neglect. While many or even most

bioethicists might agree that a particular topic is being relatively neglected,
the basis for this might well vary.
Complicating the situation further, varying appraisals of what counts as a
neglected non- will likely be part and parcel of meaningfully alternative ways
a given topic is framed. That then raises important questions about whether
commentators are orientating to neglect in realist or non-realist termsthat
is, whether they are assuming some definitive sense of what topic X is and how
it should be understood. There may well be meaningfully different topics X1, X2,
X3 and so on at play.91
In short, an empirical examination of how ethicists contend about whether and
how the dual-use aspects of the life scientists are being neglected could be a
way into examining the in-practice reasoning and bounds of bioethics.
If inquiry is linked to the aim of transforming social relations then another
reflexive dimension of the relationship between the normative and the empirical
opens up, one that again calls into question conceiving of the study of nonissues as the cataloguing of knowledge gaps. To make this point I want to return
to where this chapter began, with the categorical condemnation of biological
weapons. As contended earlier, assents to this denunciation are typically
accompanied by little explicit justification. As with many stigmas, the one
against biological weapons is often portrayed as self-evident.
To engage in explicit empirical-normative inquiry about what should count as a
problem might well cast doubt on this orientation. That could happen through
questioning arguments that biological weapons are really more inhumane or
indiscriminate than other weapons. If empirical-normative analysis were to
undermine the often visceral, intuitive reactions against biological weapons,
some would no doubt regard this as a deplorable outcome. As a result, not
undertaking analysis might well be judged as a wiser course of action. Whether
or not this appraisal holds depends on the goals sought from analysis. As ever,
then, the uses of science and ethics are topics for considered inquiry.
91 Further, then, whether some definitive sense of the topic needs to be established (and what that should
be) is a choice that needs to be addressed.
379
Appendix A
Who is working in neuroethics? Where are
they?
In a preliminary overview of the field, it was noticed that neuroethicists are not
saying enough about the problem of dual-use,1 a conclusion promptly validated
by Peter B. Reiner, Professor in the National Core for Neuroethics, who wrote:
The truth is that other than Jonathan Moreno, few neuroethicists have
applied serious scholarship to the issue of dual use. Of course, it is
a simple matter to just say no: neuroscience should only be used for
improving the quality of human life. But frankly, that is too simplistic.2
This appendix provides the reader with an overview of where neuroethics
is carried out, who is working in the field and what neuroethicists say about
dual use. Only some of the main institutions and networks as well as leading
neuroethicists are mentioned as representatives of the wider field.
Where
Center for Neuroscience & Society, University of Pennsylvania
The stated mission of the centre is to increase understanding of the impact of
neuroscience on society through research and teaching, and to encourage the
responsible use of neuroscience for the benefit of humanity. <http://neuroethics.
upenn.edu/>
Center for Cognitive Neuroscience, University of Pennsylvania
Penns Center for Cognitive Neuroscience is a multidisciplinary community
dedicated to understanding the neural bases of human thought. Their current
research addresses the central problems of cognitive neuroscience, including
perception, attention, learning, memory, language, decision-making, emotion
and development. Methods include functional neuroimaging, behavioural
testing of neurological and psychiatric patients, transcranial and direct-current
1 Dando, M. 2010, Neuroethicists are not saying enough about the problem of dual-use, Bulletin of the
Atomic Scientists.
2 Reiner, B. 2010, Neuroethics at the Core, 19 July, <http://neuroethicscanada.wordpress.com/2010/07/19/
neuroethics-of-dual-use/>.
381
magnetic stimulation, scalp-recorded event-related potentials, intracranial

recording, computational modelling, candidate gene studies and pharmacologic
manipulations of cognitive processes. <http://ccn.upenn.edu/>
Sage Centre for the Study of the Mind, University of Santa Barbara,
California
The Sage Center for the Study of the Mind at the University of California,
Santa Barbara, is designed to be a catalyst for interdisciplinary study of the
relationship of brain and mind. The centre integrates a wide range of scholarly
endeavours and technologies in the humanities, social sciences and the sciences.
<www.sagecenter.ucsb.edu/intro.htm>
Dana Foundation
The Dana Foundation is a private philanthropic organisation that supports
brain research through grants and educates the public about the successes and
potential of brain research. Dana produces free publications; coordinates the
International Brain Awareness Week campaign; supports the Dana Alliances,
a network of neuroscientists; and maintains a web site: <www.dana.org>.
The Dana Foundations science and health grants support clinical research in
neuroscience and neuro-immunology and their interrelationship in human
health and disease.
European Neuroscience and Society Network
The European Neuroscience and Society Network (ENSN) is the leading
European network for interdisciplinary discussions of the social implications of
the neurosciences. Funded by the European Science Foundation and convened
by researchers at the BIOS Centre, London School of Economics, the ENSN has
been established to serve as a multidisciplinary forum for timely engagement
with the social, political and economic implications of developments in the
neurosciences, a field that has experienced unprecedented advances in the past
20 years. <www.neurosocieties.eu/>
Neuroethics Society
The Neuroethics Society is an interdisciplinary group of scholars, scientists,
clinicians and other professionals who share an interest in the social, legal, ethical
and policy implications of advances in neuroscience. Their stated mission is to
promote the development and responsible application of neuroscience through
interdisciplinary and international research, education, outreach and public
engagement for the benefit of people of all nations, ethnicities and cultures.
<www.neuroethicssociety.org>
382
Appendix A
National Core for Neuroethics, University of British Columbia,

Vancouver, Canada
The National Core for Neuroethics hosted by the University of British Columbia
is an interdisciplinary research group dedicated to tackling the ethical, legal,
policy and social implications of frontier technological developments in the
neurosciences. The objective is to align innovations in the brain sciences with
societal, cultural and individual human values through high-impact research,
education and outreach. <www.neuroethics.ubc.ca>
Brain Research Centre, University of British Columbia, Vancouver,
Canada
The Brain Research Centre, located in Vancouver, Canada, is a unique partnership
between the Vancouver Coastal Health Research Institute and the Faculty of
Medicine at the University of British Columbia. The hospital has combined forces
with broad, multidisciplinary research expertise at the University of British
Columbia to advance knowledge of the brain and to explore new discoveries and
technologies that have the potential to reduce the suffering and cost associated
with disease and injuries of the brain. <www.brain.ubc.ca/>
The Oxford Centre for Neuroethics, University of Oxford
Established in January 2009, the Oxford Centre for Neuroethics aims to address
concerns about the effects neuroscience and neurotechnologies will have on
various aspects of human life. Its research focuses on five key areas: cognitive
enhancement; borderline consciousness and severe neurological impairment;
free will, responsibility and addiction; the neuroscience of morality and
decision-making; and applied neuroethics. It is the first international centre
in the United Kingdom dedicated to neuroethical research. It is founded by
the Wellcome Trusts Biomedical Ethics Strategic Awards Program. <www.
neuroethics.ox.ac.uk>
National Institute of Neuroscience, Italy
The Institute of Neuroscience of the National Research Council considers
itself to be one of the top institutions in the field of neuroscience in Europe. It
concentrates many of the most important Italian scientists involved in the study
of the nervous system, who are organised in research groups, located in Milan,
Padua, Pisa, Rome and Cagliari. The Institute of Neuroscience of the National
Research Council addresses all the principal topics in the study of the nervous
system, investigating the development and plasticity of the nervous circuitry;
vision and cognitive sciences; the mechanisms of memory and learning; as
well as those involving cellular transmission and neuronal communication;
383
neuromuscular and neuronal-glia interactions, and the neurobiological bases

of alcoholism and drug dependence. <www.cnr.it/istituti/Descrizione_eng.
html?cds=061>
Neuroscience Centre, Dartmouth (NCD)
The establishment of the Neuroscience Center at Dartmouth (NCD) in 2002
produced a new and unique interdisciplinary group whose mission is to foster
collaborative and interactive research and education in the neurosciences.
The NCD draws from its strengths in three key areas: clinical; cognitive and
behavioural; and molecular/cellular/systems neuroscience. It is the vision of
its researchers to produce and disseminate new knowledge, and in doing so
train and educate the next generation of neuroscientists. Interactions among
members of the neuroscience community are enhanced and foster a highly
interactive atmosphere through the development of this integrated centre. By
promoting multidisciplinary efforts in both basic and applied research, the
centres scientists will contribute to human health and wellbeing by increasing
our understanding of the mechanisms underlying nervous system function,
both in health and in disease. This will lead to valuable discoveries that translate
into novel pharmaceutical agents and therapeutic approaches for the treatment
of a variety of central nervous system diseases and disorders. <http://dms.
dartmouth.edu/>
International Neuroethics Network
The International Neuroethics Network (INN) was launched in 2005 at the Society
for Neuroscience Annual Meeting in Washington, DC. The networks vision is
to foster international collaboration in neuroethics through the identification
of common priorities and joint funding opportunities. The INNs objective is to
serve as a means of communication and support among neuroethicists all around
the world. <www.neuroethics.ubc.ca>
Stanford Program for Neuroethics, Stanford University
The Stanford Program for Neuroethics is a research team devoted to the new
field of neuroethics, with an initial focus on issues at the intersection of medical
imaging and biomedical ethics. These include ethical, social and legal challenges
presented by advanced neurofunctional imaging capabilities, the emergence
of cognitive enhancement neurotechnologies and pharmacology, self-referral
to healthcare and imaging services, incidental findings, and fetal MRI. New
initiatives are under way in regenerative medicine, neurogenetics and pediatric
neuroethics. Several program members are also involved in the John D. and
Catherine T. MacArthur Foundations Law and Neuroscience Project, including
Hank Greely, Emily Murphy and Teneille Brown. The project seeks to address
384
Appendix A
issues that neuroscience raises for our legal system through its three Research
NetworksDiminished Brains, Addiction and Decision Makingand its
Education and Outreach Program. <http://neuroethics.stanford.edu>
MacArthur Law & Neuroscience Project
The MacArthur Law and Neuroscience Project investigates the impact of modern
neuroscience on criminal law and in particular the diverse and complex issues
that neuroscience raises for the criminal justice system in the United States.
<www.lawneuro.org>
The Neuroethics Research Unit
The Neuroethics Research Unit is committed to training a new generation of
students in neuroethics through the conduct of collaborative interdisciplinary
research. Research interests include the ethical application of neuroscience
in research and patient care, empirical bioethics research, and pragmatism in
bioethics. It is based at the Montreal Institute of Clinical Research. <http://
www.ircm.qc.ca/microsites/neuroethics/en/index.html>
Novel Tech Ethics
The team at Novel Tech Ethics is committed to public discussion of the ethics
issues that affect all human beings. Key focal points of the Novel Tech Ethics
Research Team include
how are psychopharmacologies creating a new normal for human
behaviourand what is normal for humans when regenerative medicine
mixes and matches cells across species
how will the results emerging from neuroimaging studies and behavioural
genetics affect our understandingand social and legal enactmentof free
will and responsibility
what are the particular kinds of harm and benefit offered by neurological
treatment, neurological enhancement and neurological control, and how do
these challenge traditional notions and practices of risk assessment?
<www.noveltechethics.ca/site_events.php>
University of Wisconsin: Neuroscience and Public Policy Program
The program emerged from the recognition that the rapid advancement in
neuroscience demanded research neuroscientists to be trained to think critically
about both neuroscience and the making of public policy, and to have appropriate
skills, experience and networks to facilitate an effective integration of the two.
The program is hosted by the University of WisconsinMadison. A central
385
element of the program is the weekly Neuroscience and Public Policy Seminar,
which challenges students to synthesise information across neuroscience and
policy research. <http://npp.neuroscience.wisc.edu/index.html>
The Neuroethics New Emerging Team (NET)
The Neuroethics New Emerging Team (NET) is based at Dalhousie University,
Canada. Launched in 2003, the NET aims to undertake an interdisciplinary
study of, and disseminate their findings on, the ethical issues posed by advances
in neuroscience technology. It is funded by the Canadian Institutes of Health
Research. <http://www.neuroethics.ca/Neuroethics.ca>
Who
Colin Blakemore
Colin Blakemore is a neurobiologist at Oxford University. He is specialised
in vision and development of the brain. He also holds professorships at the
University of Warwick and the Duke UniversityNational University of
Singapore Graduate Medical School, where he is chairman of Singapores
Neuroscience Research Partnership. His research has been concerned with
many aspects of vision, the early development of the brain and plasticity of the
cerebral cortex.
Turhan Canli
Turhan Canli is Associate Professor at Stony Brook University. The work in Dr
Canlis laboratory focuses on the hormonal and neurogenetic bases of individual
differences in emotion and cognition. The research addresses these questions:
what are the biological mechanisms that can explain human personality? What
is the mechanism by which life experience, in interaction with genetic variation,
influences brain function to generate behavioural patterns that we associate with
certain personality traits? Do men and women differ in how their brains respond
to these genetic and experiential influences? Can this information be used to
identify healthy individuals at risk for psychopathology? To address these
questions, Dr Canlis team uses a number of different technologies: functional
magnetic resonance imaging (fMRI); transcranial magnetic stimulation (TMS);
molecular genetics; and hormone assays.
386
Appendix A
Arthur Caplan
Arthur Caplan is Emanuel and Robert Hart Professor of Bioethics and Philosophy
at the University of Pennsylvania. His research interests focus on transplantation
research ethics, genetics, reproductive technologies, health policy and general
bioethics.
Jocelyn Downie
Jocelyn Downie is Professor at Faculties of Law and Medicine, Dalhousie
University. She works at the intersection of law, ethics and health care. Her
research interests include womens health, assisted death, research involving
humans, and organ transplantation. Her work is interdisciplinary, collaborative
and geared both to contributing to the academic literature and to affecting
change in health law and policy at federal and provincial levels.
Martha J. Farah
Martha J. Farah is a cognitive neuroscientist at the University of Pennsylvania,
who works on problems at the interface of neuroscience and society, including
the effects of childhood poverty on brain development
the expanding use of neuropsychiatric medications by healthy people for
brain enhancement
novel uses of brain imaging in, for example, legal, diagnostic and educational
contexts
the many ways in which neuroscience is changing the way we think of
ourselves as physical, mental, moral and spiritual beings.
Kenneth R. Foster
Kenneth R. Foster is a professor of bioengineering. His research interests relate
to biomedical applications of non-ionising radiation from audio through to
microwave frequency ranges, and health and safety aspects of electromagnetic
fields as they interact with the body. For example, he examines the prospects
of workers in electrical occupations and the possibility (or lack of) cancer risk.
Another and somewhat broader topic of interest is technological risk and the
impact of technology (principally, electro-technologies) on humans. His goal in
this area is to examine technology, putting into perspective its relative risks and
benefits to society. What he hopes to impart is a better perception of the social
use of science.
387
Michael Gazzaniga
Michael Gazzaniga is a professor of psychology and the Director of the Sage
Center for the Study of the Mind at the University of California, Santa Barbara.
He oversees an extensive and broad research program investigating how the
brain enables the mind. Over the course of several decades, a major focus of
his research has been an extensive study of patients who have undergone splitbrain surgery that has revealed lateralisation of functions across the cerebral
hemispheres.
Henry Greely
Henry Greely is Deane F. and Kate Edelman Johnson Professor of Law. A
leading expert on the legal, ethical and social issues surrounding health law
and the biosciences, Hank specialises in the implications of new biomedical
technologies, especially those related to neuroscience, genetics and stemcell research. He frequently serves as an advisor on Californian, national and
international policy issues. He is chair of Californias Human Stem Cell Research
Advisory Committee and served from 2007 to 2010 as co-director of the Law
and Neuroscience Project, funded by the MacArthur Foundation. Active in
university leadership, Professor Greely chairs the steering committee for the
Stanford Center for Biomedical Ethics and directs both the law schools Center
for Law and the Biosciences and the Stanford Interdisciplinary Group on
Neuroscience and Society. Professor Greely serves on the Scientific Leadership
Council for the universitys interdisciplinary Bio-X Program.
Ronald M. Green
Ronald M. Green is Eunice and Julian Cohen Professor for the Study of Ethics and
Human Values. Ronald M. Green has been a member of Dartmouth Universitys
Religion Department since 1969; he also directs Dartmouths Ethics Institute,
a consortium of faculty concerned with teaching and research in applied and
professional ethics. Professor Greens research interests are in genetic ethics,
biomedical ethics and issues of justice in healthcare allocation. He is the author
of six books and more than 130 articles in theoretical and applied ethics.
Judy Illes
Judy Illes is Associate Professor of Pediatrics (Medical Genetics) and Director of
the Program in Neuroethics at the Stanford Center for Biomedical Ethics. She also
co-founded the Stanford Brain Research Center (now the Neuroscience Institute
at Stanford), and served as its first executive director between 1998 and 2001.
Today, Dr Illes directs a strong research team devoted to neuroethics, and issues
specifically at the intersection of medical imaging and biomedical ethics. These
include ethical, social and legal challenges presented by advanced functional
388
Appendix A
imaging capabilities, the emergence of cognitive enhancement technologies and

pharmacology, the commercialisation of cognitive neuroscience, and clinical
findings detected incidentally in research.
Jonathan Moreno
Jonathan Moreno is the David and Lyn Silfen University Professor of Ethics
and Professor of Medical Ethics and of History and Sociology of Science at
Pennsylvania University. He holds a courtesy appointment as Professor of
Philosophy. He is also a Senior Fellow at the Center for American Progress in
Washington, DC, where he edits the magazine Science Progress. He was a member
of President Barack Obamas transition team for the Department of Health and
Human Services. In the course of his career, Professor Moreno has applied
serious scholarship to the issue of dual use in neuroscience, revealing that
several of the new technologies are potentially applicable to medical therapy
or other peaceful purposes as well as combat, riot control, hostage situations, or
other security problems.
Stephen J. Morse
Stephen J. Morse is Professor of Psychology and Law in Psychiatry. Stephen J.
Morse is an expert in criminal and mental health law whose work emphasises
individual responsibility in criminal and civil law. Morse was co-director of the
MacArthur Foundations Law and Neuroscience Project and he co-directed the
projects Research Network on Criminal Responsibility and Prediction. Morse
is a diplomate in Forensic Psychology of the American Board of Professional
Psychology; a past president of Division 41 of the American Psychological
Association (the American Psychology-Law Society); a recipient of the American
Academy of Forensic Psychologys Distinguished Contribution Award; a
member of the MacArthur Foundation Research Network on Mental Health and
Law (198896); and a trustee of the Bazelon Center for Mental Health Law in
Washington, DC (1995present).
Peter B. Reiner
Peter B. Reiner is Professor in the National Core for Neuroethics and a member
of the Kinsmen Laboratory of Neurological Research, Department of Psychiatry
and the Brain Research Centre at the University of British Columbia. Dr Reiner
has a distinguished track record as a research scientist studying the neurobiology
of behavioural states and the molecular underpinnings of neurodegenerative
disease. Dr Reiner also has experience in the private sector, having been
president and CEO of Active Pass Pharmaceuticals, a drug discovery company
that he founded to tackle the scourge of Alzheimers disease. Upon returning to
389
academic life, Dr Reiner refocused his scholarly work in the area of neuroethics,
with interests in neuro-essentialism, the neuroethics of cognitive enhancement
and the commercialisation of neuroscience.
Paul R. Wolpe
Paul R. Wolpe is Associate Professor of Psychiatry in the Department of
Psychiatry at the University of Pennsylvania, where he also holds appointments
in the Department of Medical Ethics and the Department of Sociology. He is
President of the American Society for Bioethics and Humanities and is CoEditor of the American Journal of Bioethics. Dr Wolpe serves as the first Chief of
Bioethics for the National Aeronautics and Space Administration (NASA).
390

462759

Uploaded by

Copyright:

Available Formats

462759

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

462759

Uploaded by

Copyright:

Available Formats

ON THE DUAL USES OF SCIENCE AND ETHICS

PRINCIPLES, PRACTICES, AND PROSPECTS

ON THE DUAL USES OF SCIENCE AND ETHICS

Practical Ethics and Public Policy Monograph 4

Published by ANU E Press

National Library of Australia Cataloguing-in-Publication entry

9781925021332 (paperback) 9781925021349 (ebook)

Subjects: Science and civilization--Moral and ethical aspects.

Rappert, Brian, editor.

Selgelid, Michael J., editor.

Dewey Number: 501

Part I: Dual Use in Context

3. What Does Neuroethics Have to Say about the Problem of

4. Synthetic Biology as a Field of Dual-Use Bioethical Concern . . . 45

5. Crops Agents, Phytopathology and Ethical Review . . . . . . . . . 65

6. The Super TB Experiment: Evolution and Resolution of an

Part II: Ethical Frameworks and Principles

8. Responsible Dual Use. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

9. An Expected-Value Approach to the Dual-Use Problem. . . . . . 133

11. Unknowns in Dual-Use Dilemmas . . . . . . . . . . . . . . . . . . . 165

12. Moral Responsibility, Collective-Action Problems and the

13. Biosecurity and the Just-War Tradition. . . . . . . . . . . . . . . . 207

14. The Precautionary Principle and the Dual-Use Dilemma. . . . . 223

Part III: Ethical Practices

16. Contrasting Dual-Use Issues in Biology and Nuclear

17. Considering Contextuality in Dual-Use Discussions:

Part IV: Ethical Futures

19. WHO Project on Responsible Life-Sciences Research

On the Dual Uses of Science and Ethics

ethics debates, particularly examining how scientists in developing countries

Koos van der Bruggen

On the Dual Uses of Science and Ethics

On the Dual Uses of Science and Ethics

Andreas Alois Reis

On the Dual Uses of Science and Ethics

American Association for the Advancement of Science

On the Dual Uses of Science and Ethics

Nature Publishing Group

On the Dual Uses of Science and Ethics

United States of America

1. Ethics and Dual-Use Research

Nuclear physics and contemporary genetics

On the Dual Uses of Science and Ethics

The mousepox experiment

1. Ethics and Dual-Use Research

Reconstruction of 1918 Spanish flu

On the Dual Uses of Science and Ethics

pandemic strains and thus earlier implementation of protective measures. On

1. Ethics and Dual-Use Research

On the Dual Uses of Science and Ethics

1. Ethics and Dual-Use Research

Part I: Dual Use in Context

Nanotechnology development and the

On the Dual Uses of Science and Ethics

Nanotechnology: Pace, promise and peril

2. Nanotechnology and Dual-Use Dilemmas

conducted) on a huge range of potential developments seems as distant

6 Ibid., pp. 745.

On the Dual Uses of Science and Ethics

9 Royal Commission on Environmental Pollution, op. cit., p. 50.