Ethical Questions and International NGOs

ETHICAL QUESTIONS AND INTERNATIONAL NGOS
LIBRARY OF ETHICS AND APPLIED PHILOSOPHY

VOLUME 23
Editor in Chief
Marcus Düwell, Utrecht University, Utrecht, NL
Editorial Board
Deryck Beyleveld, Durham University, Durham, U.K.

David Copp, University of Florida, USA
Nancy Fraser, New School for Social Research, New York, USA
Martin van Hees, Groningen University, Netherlands
Thomas Hill, University of North Carolina, Chapel Hill, USA
Samuel Kerstein, University of Maryland, College Park
Will Kymlicka, Queens University, Ontario, Canada
Philip van Parijs, Louvaine-la-Neuve (Belgium) en Harvard, USA
Qui Renzong
Peter Schaber, Ethikzentrum, University of Zürich, Switzerland
Thomas Schmidt, Humboldt University, Berlin, Germany
For further volumes:

http://www.springer.com/series/6230
ETHICAL QUESTIONS AND
INTERNATIONAL NGOs
An exchange between
Philosophers and NGOs
Edited by
KEITH HORTON
University of Wollongong, NSW, Australia
CHRIS ROCHE
Oxfam Australia, Melbourne, VIC, Australia
With a Foreword by Peter Singer
123
Editors
Keith Horton Chris Roche
University of Wollongong Oxfam Australia
Fac. Arts 132 Leicester Street
School of English Literatures, Carlton VIC 3053
Australia
Wollongong NSW 2522 croche@oxfam.org.au
Australia
khorton@uow.edu.au
ISSN 1387-6678
ISBN 978-90-481-8591-7 e-ISBN 978-90-481-8592-4
DOI 10.1007/978-90-481-8592-4
Springer Dordrecht Heidelberg London New York
Library of Congress Control Number: 2010920234
© Springer Science+Business Media B.V. 2010

No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by
any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written
permission from the Publisher, with the exception of any material supplied specifically for the purpose
of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Foreword
The ethical case for international aid is straightforward. More than a billion people
live in extreme poverty. They live on less than US$1.25 a day. This kind of poverty
kills. UNICEF tells us that nearly ten million children die every year from avoidable,
poverty-related causes. When poverty doesn’t kill, it forces people to live hard lives,
lacking in freedom and dignity.
In addition to the standing situation of so many people living in extreme poverty,
millions of other people in poor countries need temporary assistance to cope with
a catastrophic flood, drought, cyclone or civil war, disasters that create needs that
their country cannot meet without help from abroad.
Our world also has about a billion people living in affluence, spending money on
a wide range of luxuries, large and small. The money they spend on a new car or
a vacation could save the lives of children dying of measles, malaria or diarrhoea,
or provide food and safe drinking water to those in need because of a local crisis.
$50 could restore the sight of someone who is blind because she has cataracts. $450
could repair an obstetric fistula that has turned a beautiful young woman into a
social outcast. $25 might pay for a microloan that enables a weaver to buy his tools,
thus avoiding the exorbitant cost of renting them that prevents him making a decent
living no matter how hard he works.
Can anything be more obvious than that in this situation those of us who have so
much should do something to help those who have so little?
The basic case for aid really is simple, and there are some situations where apply-
ing the funds received is also quite straightforward. But as this book very clearly
demonstrates, there are other situations where matters rapidly become more com-
plicated. A straightforward case for aid gives rise to a variety of moral dilemmas
about how best to help those in need. How can Non-Government Organizations
best demonstrate that their work is effective? To what extent should an organization
that is seeking to help ethnic minorities, or women, respect the local culture, which
may discriminate against these groups? To what extent should aid agencies become
involved in politics?
In the introduction that follows this preface, the editors outline these dilemmas
in more detail. The authors they have selected to discuss these ethical problems
bring together experience in the field and expertise in ethics. This mix of theory and
v
vi Foreword
practice is of course exactly what is needed if we are to do better in this area. This
book opens an important discussion that should accompany moves to extend and
improve aid to the poor.
Peter Singer
Acknowledgements
We are very grateful to the Australian Research Council Governance Research

Network and the Centre for Applied Philosophy and Public Ethics for funding the
workshop from which this book developed and to Oxfam Australia and La Trobe
University for their financial support in the production of this book. Thanks are also
due to Anna Roche for her editing.
We would also like to thank all those who are not publishing essays in this book
but attended part or all or the workshop and offered helpful comments, includ-
ing Christian Barry, Paul O’Callaghan, Rekha Nath, Paul Nichols, Brad Pettit,
Thomas Pogge, Wendy Rose, and Doris Schroeder. Particular thanks are in order
to Judy Mitchell who produced a set of remarkably coherent notes from what was
sometimes a complex and non-linear discussion!
Finally we would like to thank an anonymous referee for some excellent feedback
and useful recommendations for improving the text.
September 2009 Keith Horton and Chris Roche
vii
Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Keith Horton and Chris Roche
Ethical Obligations to the Poor in a World of Nation States . . . . . . . 13
Paul Ronalds
Human Rights, Development INGOs and Priorities for Action . . . . . 39
Kieran Donaghue
The Ethics of Taking Sides . . . . . . . . . . . . . . . . . . . . . . . . . 65
Peter Ellis
The Epistemic Problem: Potential Solutions . . . . . . . . . . . . . . . . 87
Keith Horton
The Seeming Simplicity of Measurement . . . . . . . . . . . . . . . . . 119
Chris Roche
Whose Impact, and Is It All About Impact? . . . . . . . . . . . . . . . . 147
Jamie Isbister
Compromised Humanitarianism . . . . . . . . . . . . . . . . . . . . . . 157
Garrett Cullity
Aid Agencies, States, and Collective Harm . . . . . . . . . . . . . . . . 175
Ramon Das
To Respect or Not to Respect . . . Ethical Dilemmas of INGO
Development Practitioners . . . . . . . . . . . . . . . . . . . . . . . . . 193
Conny Lenneberg
Ethical Behaviour in Non-government Organisations . . . . . . . . . . . 207
Linda Kelly
Afterword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
ix
Contributors
Garrett Cullity University of Adelaide, Adelaide, SA, Australia,

garrett.cullity@adelaide.edu.au
Ramon Das Victoria University of Wellington, Wellington, New Zealand,
ramon.das@vuw.ac.nz
Kieran Donaghue Centre for Applied Philosophy and Public Ethics, Australian
National University, Canberra ACT, Australia, kieran.donaghue@gmail.com
Peter Ellis Independent Monitoring and Results Advisor, NZAID,
ellisp@netspeed.com.au
Keith Horton University of Wollongong, Wollongong, NSW, Australia,
khorton@uow.edu.au
Jamie Isbister Caritas Australia, Sydney, NSW, Australia,
jamie.isbister@ausaid.gov.au
Linda Kelly Praxis Consultants Pty Ltd, Sutton, NSW, Australia,
linda.kelly@praxisconsultants.com.au
Conny Lenneberg World Vision Australia, Melbourne, VIC, Australia,
conny.lenneberg@worldvision.com.au
Chris Roche Oxfam Australia, Carlton, VIC, Australia, croche@oxfam.org.au
Paul Ronalds World Vision Australia, Melbourne, VIC, Australia,
Paul.Ronalds@worldvision.com.au
xi
Introduction
1 Introduction
In recent decades there has been a great expansion in the number, size and influence
of International Non-Governmental Organisations involved in international relief
and development (henceforth simply ‘INGOs’). These changes have led to increased
scrutiny of such organisations, and this scrutiny, together with increasing reflection
by INGOs themselves and their staff on their own practice, has helped to highlight
a number of pressing ethical questions such organisations face. Should they attempt
to provide emergency assistance even when doing so risks helping to fuel further
conflict, for example? How should they manage any differences between their values
and those of the people they seek to benefit? How open and honest should they be
about their own uncertainties and failures?
Such questions are difficult and controversial. INGOs need to address them
though, for they cannot avoid making decisions about the matters that give rise to
those questions. Of course, they also need to take care to ensure that doing so doesn’t
lead to the ‘paralysis of analysis’. Decisions cannot be postponed indefinitely, even
if there is persisting uncertainty about the consequences of different actions or poli-
cies, or about what principles should inform their choices. That, however, provides
no reason not to tackle the relevant questions, including questions concerning what
one should do when one has such uncertainties.
There is an emerging literature on such questions. The complex humanitarian
emergencies of the 1980s and 1990s—above all, the Rwandan genocide and its
aftermath—have prompted a relatively lively debate about humanitarian aid, includ-
ing its ethical aspects.1 There is also a growing literature on ‘development ethics’,2
and some discussion of ethical issues in the broader literature on development.3
Further, there is some discussion of the ethical issues faced by INGOs in particular
in the literature on such organisations that has emerged since the early 1990s.4
Nevertheless, the ethical issues raised by development and humanitarian aid
have not yet attracted nearly as much concentrated attention as the ethical issues
K. Horton (B)
University of Wollongong, Wollongong, NSW, Australia
e-mail: khorton@uow.edu.au
K. Horton, C. Roche (eds.), Ethical Questions and International NGOs, Library of 1

Ethics and Applied Philosophy 23, DOI 10.1007/978-90-481-8592-4_1,

C Springer Science+Business Media B.V. 2010
2 K. Horton and C. Roche
in many other fields, such as medicine, business and the environment. While each
of these fields has given rise to several journals (such as The Journal of Medical
Ethics, Business Ethics Quarterly, and Environmental Ethics), there is still no jour-
nal devoted to development ethics or humanitarian ethics or, more broadly, the ethics
of aid, for example.
It was in part with the aim of stimulating more work on these issues that the
editors of this volume organised the workshop—held at Melbourne University in
July 2007—that led to this book. The immediate aim was to bring together a group
of people—mostly reflective practitioners and moral and political philosophers—
to discuss some of the specific ethical questions that INGOs face. The rationale
for the presence of the practitioners is obvious enough: they are the ones who have
immediate experience of the conflicts such questions raise, and the ones who have to
make the decisions about how to respond to them. Moral and political philosophers,
for their part, spend their working lives studying moral and political issues, and
this naturally led us to hope that they might have something useful to contribute to
discussions about such issues in the context of international relief and development.
This book is the main result of that exercise.5 The central chapters consist of
papers presented at the workshop, revised in the light of discussion at the workshop
and of feedback from the editors. We also gave some time at the workshop to more
general discussion about issues concerning what the most pressing of ethical ques-
tions INGOs face are, and what the most productive ways of tackling them might
be. In the Afterword we summarise some of this discussion, together with some
reflections by the editors of this book.
In Section 3, we provide brief summaries of the chapters. First, though, we set
out a general classification of some of the ethical questions INGOs face.
2 Ethical Questions and INGOs: A Classification
What, then, are the key ethical questions that INGOs face? The field here is a large
and messy one, and there are a number of different ways of carving it up. Below is a
sketch of one way to do so, with six broad categories, a number of subcategories in
each of those categories, and examples of questions in each of those subcategories.
The categories in question are not mutually exclusive; indeed, there is a great deal
of crossover between different categories. Nor, of course, are they exhaustive; fur-
ther categories could be added, as well as more subcategories. Even this partial and
incomplete classification does something to bring out, however, the range of ethical
questions INGOs face.
2.1 Ethical Questions Raised by Specific Kinds of Activity INGOs

Undertake
2.1.1 Emergency Relief

Should the right to emergency assistance be treated as absolute? What if the price of
providing such assistance is silence concerning human rights abuses, for example,
Introduction 3
or there is a high risk that doing so will help to fuel further conflict? When might
protection be more important than the delivery of humanitarian assistance, and what
rôle can INGOs be expected to play in this?
2.1.2 Service Delivery

Given the risks of such activities (fostering dependency, undermining local solu-
tions, letting governments off the hook, etc.), should they be replaced entirely with
activities aimed at lasting solutions? What if the risks in question seem fairly small,
lasting solutions are hard to find, and people would suffer greatly without immediate
help? Does a rights-based approach assist in addressing these questions? If so, how
exactly?
2.1.3 Development
Is there reliable data showing that (certain kinds of) development activities or pro-
cesses are more successful than others? If so, should INGOs be more prepared to
stop activities or processes that are relatively less successful, or for which evidence
is harder to come by? If not, what justifies INGOs in continuing with such activ-
ities or processes, given that they may not just be less successful but have serious
negative effects? What, in any case, gives INGOs the right even to aim at signifi-
cant changes—social, cultural, and political—in the societies they work in? Does
the claim that they are promoting human rights supply an adequate answer to this
question?
2.1.4 Advocacy
Should more (or fewer) resources be devoted to lobbying, advocacy, and campaign-
ing? Do INGOs have a greater moral responsibility to change the policies of their
home governments than those of others? What gives INGOs the right to engage
in such work? What conditions must they meet before they can be said to speak
on behalf of the poor? Given the difficulties associated with attributing policy or
practice change to specific NGO actions—let alone the methodological challenges
of assessing the diverse impact of policy and practice changes on the lives of peo-
ple living in poverty—what conditions have to be met for INGOs to be justified in
diverting resources from more direct relief or development activities to advocacy
activities? When is it morally unacceptable not to speak out?
2.2 Ethical Questions Concerning the Selection of Activities

and Areas of Work
2.2.1 Strategic Choice

How should INGOs make such strategic choices concerning what general kinds
of activity (relief, service delivery, this or that kind of development, advocacy) to
engage in, and where? By a calculation of what is likely to have the best conse-
quences overall? If so, how should they make such calculations? On the basis of
need, human rights, comparative advantage or cost effectiveness? Or by some other
criteria? If so, what criteria? Is avoiding harm ethically more important than failing
to help?
2.2.2 Specific Choices

Whichever general kinds of activity an NGO chooses to employ, how should they
decide which particular instances of that kind to select? Again, by a calculation of
what is likely to have the best consequences overall? Or by other criteria? Should
they focus their efforts on those who are worst off, for example, even if it is easier
and cheaper (and hence more efficient) to bring about improvements in the con-
ditions of those who are not so badly off? How much influence, if any, should
contributors to such agencies or other stakeholders have on these decisions? How
much influence, if any, should those they aim to help have? What should INGOs do
if the views of different stakeholders clash?
2.2.3 The Risk of Negative Effect

Experience shows that even promising activities can have negative effects. How
should INGOs respond to such risks? If the risk is high enough, should they hold
back? Or should they go ahead if the positive effects are expected to outweigh the
negative by a sufficient amount? If so, how much might a ‘sufficient’ amount be, and
how might it be calculated? Does it make a difference whether those negative effects
come about as a direct result of the activity of the NGO, for example, or indirectly?
Does it make a difference whether those who will suffer if those negative effects
come about make an informed decision to consent to the activity in question? Does
it make a difference if those negatively affected are from less poor communities?
2.3 Ethical Questions Concerning the Relationship Between

INGOs, Their ‘Partners’ and Those They Aim to Benefit
2.3.1 Participation
Must INGOs always ensure that their partners and those they seek to benefit partic-
ipate in any decisions that concern them? If so, what exactly does this requirement
come to in practice and how would one assess whether it has been met? Whether or
not any such participation is necessary, is it also sufficient for an activity to be legit-
imate, or morally acceptable? Or is a more thoroughgoing handing-over of power
essential so that the men and women they seek to benefit do not merely participate
in decision-making, but lead the process?
Introduction 5
2.3.2 Autonomy
More broadly, is respecting the autonomy of partners and those they seek to
benefit—roughly speaking, their right to determine for themselves how to operate
and indeed live—an important value in development (and relief) work? If so, what
constraints might respecting this value place on INGOs? Does it mean, for example,
that INGOs should only seek to support initiatives that people living in poverty have
started by themselves?
2.3.3 Cultural Differences

What should INGOs do when local norms or cultural practices clash with their val-
ues? Should INGOs always respect such norms and practices and if so what might
such respect come to in practice? Does the notion of human rights offer any help in
answering these questions? Does the fact the culture is dynamic and often locally
contested make a difference?
2.3.4 Accountability
Should INGOs be more accountable to their partners and those they aim to benefit
than they currently are? If so, in what ways precisely? Might sanctions be appropri-
ate if things go wrong, for example? How are these accountabilities to be balanced
with other accountabilities INGOs have (see Section 2.4 below)?
2.3.5 Systems and Procedures

What adjustments in the systems and procedures of INGOs might be necessary if
they were to truly seek more effective and equal relationships with their ‘partners’
and those they seek to benefit?
2.4 Ethical Questions Concerning the Relationship Between

INGOs and Their Contributors and the Wider Public
in Northern Countries
2.4.1 Accountability
In what ways should INGOs be accountable to their contributors and supporters?
Should they always make evaluations of their activities available, for example, or
at least summaries of those evaluations? If not, how can private contributors and
potential contributors have confidence that the work INGOs are doing is achieving
good results? What other accountability measures might be considered necessary?
2.4.2 Openness
More generally, how open and honest should INGOs be about their activities?
Should they be required to publish their accounts in such a way that really makes
clear how much of their income they spend on different activities, for example? Is
it justifiable to present their activities in a way that will appeal to contributors, even
if such a presentation is not strictly accurate? Might it be permissible to exaggerate
the risk of a potential emergency in order to raise extra funds, for example? Is it
reasonable to expect INGOs to be completely open and honest about the problems
and complexities of aid? What if more openness and honesty would lead to fewer
donations—would this be a sufficient reason to maintain the status quo?
2.4.3 Fundraising
Which methods of fundraising are morally acceptable? Is child-sponsorship accept-
able, for example? If not, why not? What about pictures of starving children?
Relatedly, is it morally acceptable for INGOs to ensure they have a presence in
places where the media will be present, in order to maintain their profile, even if
they could be doing more good elsewhere?
2.4.4 Active Engagement

Should INGOs seek to engage supporters in ways that allow them to become active
citizens in a process of social change rather than passive donors to a distant agency?
If so, how might they best do so?
2.4.5 Institutional Imperatives

More broadly, how should INGOs manage the balance between what are sometimes
called ‘institutional imperatives’, such as keeping the funds flowing in and surviv-
ing as an institution, on the one hand, and ‘developmental imperatives’, as specified
in their mission statements, on the other? How far do institutional imperatives con-
tribute to, or constrain, INGOs’ abilities to be true to the answers to the questions
listed in Sections 2.4.1, 2.4.2, and 2.4.3?
2.5 INGOs and Politics
2.5.1 Being Political

Should INGOs adopt specific political stances or orientations? If so, how open
should they be about doing so? Should they state their political orientation in the
same kind of way that political parties do, for example? What implications might
this have for their sources and types of funding?
2.5.2 Legitimacy
Do INGOs have the right to intervene in political processes at home or abroad? If
so, what gives them that right? Conforming to certain rules, regulations or stan-
dards? Representing a certain constituency? Their experience on the ground? Their
successful performance? Or something else? If so, what exactly?
Introduction 7
2.5.3 Internal Politics

Should INGOs be (more) democratic? Is there something wrong if their internal
processes don’t reflect what they are fighting for in their work (for example, by being
hierarchical or authoritarian or non-democratic)? If so, what kind of fault would they
be displaying? Hypocrisy? A lack of integrity? Or might there be legitimate reasons
for such apparent inconsistencies? If so, what are those reasons?
2.5.4 Funding from Governments of Developed Countries

Should INGOs take money from governments? If so, how should they handle the
familiar risks of doing so—becoming more reluctant to criticise government policy,
adopting government agendas and practices, and so on? Alternatively, should they
refuse all government funding in order to remain independent, and therefore able to
act freely? Might this be one area where a division of labour between INGOs would
be helpful—some accepting government money and others not?
2.5.5 Relations with Governments of Developing Countries

How should INGOs deal with governments and other authorities that are not sym-
pathetic to their aims? Does it make a difference if the government in question is
democratically elected? Again, is there one best policy that all INGOs should fol-
low in this area, or would it make more sense for different INGOs to follow different
policies?
2.5.6 Acting Illegally

Are there circumstances where breaking the law would be the right thing for INGOs
to do? If so, what are these circumstances? What might be the limits to this?
Would one stop at civil disobedience, for example, or are there circumstances where
supporting violence might be acceptable?
2.6 Ethical Questions Related to Broader Questions of Individual

and Collective Governance, Structure and Accountability
2.6.1 Governance
What new forms of governance, structure and accountability might be appropriate
for INGOs, given the ethical issues raised above? Are improvements in the internal
governance and accountability structures of such organisations likely to prove suffi-
cient to respond to these problems? Or will more energetic self-regulation, codes
of conduct, external governance, or external regulation of such organisations be
necessary? If so, what form should such regulation or governance take?
2.6.2 Corporatisation
Is the increasing growth and ‘corporatisation’ of INGOs changing their values, and
diminishing their ability to act flexibly, be responsive and adapt quickly? Is the asso-
ciated concern with branding and profile leading to a greater focus on institutional
rather than humanitarian or developmental imperatives? Is the adoption of largely
western corporate management models by INGOs antithetical to their mission?
2.6.3 Collective Action Problems

Some of the problems raised above have certain features of collective action
problems—e.g. avoiding ‘pornography of poverty’ fundraising. The obvious way
to tackle such problems is by binding agreements not to engage in practices that are
harmful at the collective level. What such agreements currently exist? How effective
are they? What other such agreements might be feasible? Again, what kinds of col-
lective governance structures would be needed to make them work? In particular,
what kind of collective agreements might be needed when the actions of individ-
ual INGOs might be positive but the combined actions of many INGOs negative?
Alternatively what kind of moral division of labour between INGOs might be appro-
priate (for example between advocacy or Human Rights INGOs and humanitarian
relief agencies in Darfur)?
2.6.4 Institutionalising Changes

More briefly and generally, what governance arrangements in individual agen-
cies might make it most likely that the answers to questions raised in Sections
2.6.1, 2.6.2, and 2.6.3 will be addressed, and what collective action is required for
individual agencies to be able (or willing) to do this?
3 The Papers in this Volume
Though the chapters in this volume do not of course address all of these questions,
they do offer insights into many of them, and there are some common areas touched
upon by several of the papers. These include:
• The need for more consideration of collective INGO agreements and processes,
particularly in relation to collective action problems, as well as more effec-
tive ‘moral divisions of labour’ between INGOs, particularly when a number of
different responses to a given situation might be required;
• The potential of ethical reflection and rights-based approaches to assist INGOs
in determining relative priorities between different types of activity;
• Ideas about how to address some of the complexities and challenges of knowing
and communicating ‘what works’ and what does not and thus making ethical
judgements on the basis of (cost-)effectiveness;
Introduction 9
• Suggestions about how asserting the importance of organisational values, adjust-

ing structure, and promoting learning and accountability can assist in developing
ethical and moral decision making and behaviour.
We finish here by briefly summarising each of the papers.

Paul Ronalds frames the debate by asking whether the world’s wealthier states
have ethical obligations to those beyond their borders, and if so what the nature
and extent of those obligations are. He discusses a number of different answers to
these questions, focusing mainly but not exclusively on cosmopolitan proposals. In
order to obtain enough political support to make such proposals politically realistic,
he argues, the cost of such measures must not be excessive, and citizens must be
able to see some connection between such expenditures and their own self-interest.
And he argues that both of these requirements would be met by an obligation to
meet the basic needs of the global poor. He concludes that such a requirement,
then constitutes a moderate position which is both ethically desirable and politically
feasible.
Kieran Donaghue asks what implications having a rights-based approach to
development might have. (This would fall mainly under Section 2.2.2 in the out-
line sketched above.) He begins by sketching the influential account of basic rights
given by the philosopher Henry Shue, and then asks how that account might help
INGOs committed to a rights-based approach to determine their moral priorities. He
argues that account implies that in general such INGOs should focus on basic rights
(that is, those rights that are necessary to the enjoyment of other rights), and that
they should give duties to protect rights-holders from being deprived of their rights
by third parties priority over the duty to aid those who have been deprived, though
he also acknowledges a number of qualifications to this generalisation. Finally, he
applies this analysis in some detail to a number of practical choices INGOs face.
Peter Ellis focuses on a set of questions concerning INGOs and politics (hence
Sections 2.5, especially 2.5.1). He notes that INGOs have recently been criticised
on political grounds both in recipients’ countries, for being tools of imperialist gov-
ernments, and in donor countries, for being illegitimate political advocates. Such
criticisms, he adds, are amplified by government funding of INGOs. The appropriate
response, he argues, is for INGOs simply to acknowledge the political nature of their
activities, and then to begin thinking about the ethical challenges this implies. This
will include considering what sources they derive their political legitimacy from,
and what ethical standards should regulate the political aspects of their work. He
also sketches a number of precepts that might be put in a code of conduct covering
those aspects, including issues such as transparency and organisational integrity.
The next three chapters focus mainly on a set of related questions concerning
how INGOs might demonstrate the impact they are having, especially to the public
in developed countries (Sections 2.4, especially 2.4.1).
Keith Horton writes from the point of view of individuals living in developing
countries who have no special expertise on aid. In order to determine whether or not
they should give to INGOs, he argues, such individuals need to form some idea con-
cerning how good or bad the effects of the kinds of work such agencies do are. It is
very difficult, however, for them to do so. Horton calls this the ‘Epistemic Problem’,
and argues that it has a number of unfortunate consequences. Accordingly, he then

goes on to consider a number of measures that might be taken to make it easier
for potential contributors to find out how good or bad the effects of the work aid
agencies do are.
Chris Roche begins by acknowledging it is argued by some that those living in
developed countries have an obligation to give to INGOs only if they are able to
satisfy themselves that the effects of the work of such INGOs are sufficiently posi-
tive. And then he focuses on what INGOs’ response to this challenge should be. He
begins by discussing a number of recent initiatives to increase accountability and
standards initiatives in the humanitarian sector, in each case considering how such
initiatives might be extended to other sectors, such as development and advocacy.
He then asks why the evidence concerning impact and effectiveness is currently so
weak. After summarising some of the main challenges and approaches to learning
and accountability in INGOs, he discusses three broad responses to those problems:
the ‘Scientific’ or’ ‘Measurement’ solution, the ‘Adaptive Systems’ solution, and
the ‘Political’ or ‘Power-based’ solution, summarising the advantages and disad-
vantages of each. Finally, he considers the implications all this may have for INGOs
facing the challenge sketched above.
Like Chris Roche, Jamie Isbister writes from a practitioner’s perspective.
Though he acknowledges the force of demands for INGOs to demonstrate what
impact they are having, he sounds a warning note. First, he articulates a number
of serious problems involved in answering questions about impact in relation to
complex activities like those INGOs engage in. Then he goes on to argue that the
increasing pressure to demonstrate impact is having a number of negative effects on
INGO practices, and especially on their values. The accountability mechanisms that
currently exist, he argues, tend to be over-bureaucratic, and few are directed towards
those communities that are most directly affected by INGO activities. Indeed, many
of these mechanisms, he argues, risk moving INGOs further from being accountable
to such communities. He concludes with a number of suggestions concerning how
the demand to demonstrate impact might be satisfied without such negative conse-
quences, emphasising in particular evaluations and accountability frameworks that
focus on demonstrating alignment with INGO values.
Next, there follow two papers that focus on collective action problems in relation
to emergency aid (predominantly Sections 2.1.1 and 2.6.3).
Garrett Cullity begins his paper by reviewing a number of contrasting views
concerning what agencies should do if acting on a ‘humanitarian imperative’ risks
being exploited by others, or reinforcing the structures of disempowerment that cre-
ate destitution. Then he distinguishes three different aspects of morality, focusing
especially on the ‘morality of cooperation’, which comprises the ways in which we
ought to treat each other when pursuing common projects. And he then goes on
to explain the bearing that the morality of cooperation has on humanitarian action,
focusing especially on situations in which, if each agency does what would pro-
duce the best results available to it in the context, they still collectively contribute
to making things worse than they otherwise would be. This analysis, he argues that
can clarify just what is right and wrong about each of the views sketched at the
beginning of his paper.
Introduction 11
Ramon Das takes the framework adopted by Cullity as his starting point. He
argues that there are strong grounds for thinking that aid agencies are members
of larger groups that are collectively harming the very individuals such agencies
are trying to help or protect. If this is right, how might it affect what such agencies
should do? Das argues that because of this, aid agencies are morally required to take
on a much more politically activist role than they have to this point. In particular,
in the medium to longer term, they are morally required to engage cooperatively in
political advocacy toward ending the harms perpetrated by those larger groups to
which they belong, even if other agencies do not.
Respect for local people and their culture is a value that virtually all INGOs
emphasise. At the same time, many INGOs have a radical social change agenda.
Conny Lenneberg asks how it might be possible to reconcile these two tendencies
in an ethically acceptable way. She reviews the literature on these issues and dis-
cusses of some real life cases, emphasising that development practitioners inevitably
bring their own values to their development work, and the consequent need to be
transparent about what those values are. She also she puts forward a number of
conclusions as a basis for further debate. These conclusions feature an emphasis
on social justice as the ethical basis for INGOs’ actions, and include a set of ques-
tions to guide critical reflection about the legitimacy of proposed social development
interventions.
Linda Kelly focuses on the fundamental question: What is it for an INGO to
behave ethically? On the basis of case studies of three INGOs that she takes to
exemplify high ethical standards, she picks out two answers to that question. The
first is that ethical INGOs behave with integrity, in that they do what they claim to
do, even when doing so is difficult or involves considerable costs. And the second
(which Kelly thinks might be even more important) is that they act in the way that is
likely to be most effective, again even when doing so is difficult or costly. Kelly also
draws a connection between these two answers, arguing that effectiveness depends
not only on what INGOs do, but on who they are as organisations, and whether
there is consistency between their organisational values and how they operate. She
finishes by suggesting that in order to meet these ethical criteria, INGOs may need
to change themselves and their systems in quite fundamental ways
Notes
1. For introductory reviews, see e.g. Collins (1998) and Weiss (1999). For book-length studies,
see e.g. de Waal (1997) and Terry (2002).
2. See e.g. Gasper (1994), Dower (1997) and Crocker (1998) for introductory accounts, and
Goulet (1995), Gasper (2004), and Crocker (2008) for more detailed studies.
3. See e.g. Quarles van Ufford and Giri (2003).
4. See e.g. Korten (1990), Clark (1991, 2003), Smillie (1995), Fowler (1997), Edwards (1999),
Riddell (2007), and the papers collected in Edwards and Hulme (1992, 1996), Hulme and
Edwards (1997), and Bebbington et al. (2008).
5. The only previous book focused on the ethical questions INGOs face that is the product of an
encounter between reflective practitioners and moral and political philosophers (among oth-
ers), to our knowledge, is Bell and Coicaud (2007). As one might expect that volume and this
one engage with many common themes, but there are also a number of differences. Bell and
Coicard (2007) includes within its remit organisations that have their origin as human rights
organisations, such as Amnesty International and Human Rights Watch, for one thing, while
our volume does not. Consequently, that volume gives more attention to issues concerning
civil and political rights. Our volume, on the other hand, gives more attention to issues such
as INGO accountability and effectiveness, the dilemmas of humanitarian relief, and questions
associated with the phenomenon of INGO growth and organisational structure.
References
Bebbington, A., S. Hickey, and D.C. Mitlin, eds. 2008. Can NGOs make a difference? The
challenge of development alternatives. London: Zed Books.
Bell, D.A. and J.-M. Coicard. 2007. Ethics in action: The ethical challenges of international human
rights nongovernmental organizations. Cambridge: Cambridge University Press.
Clark, J. 1991. Democratizing development: The role of voluntary organizations. London:
Earthscan.
Clark, J. 2003. Worlds apart: Civil society and the battle for ethical globalization. West Hartford:
Kumarian Press.
Collins, C. 1998. Critiques of humanitarianism and humanitarian action. In Humanitarian coordi-
nation: Lessons learned. New York: Office for the Coordination of Humanitarian Affairs.
Crocker, D.A. 1998. Development ethics. In Routledge encyclopedia of philosophy, ed. E. Craig,
vol. 3. London: Routledge.
Crocker, D.A. 2008. Ethics of global development: Agency, capability, and deliberative democracy.
Cambridge: Cambridge University Press.
De Waal, A. 1997. Famine crimes: Politics and the disaster relief industry in Africa. London:
African Rights and the International African Institute.
Dower, N. 1997. Development ethics: An overview. In Encyclopaedia of applied ethics, ed.
R. Chadwick. San Diego CA: Academic Press.
Edwards, M. 1999. Future positive. London: Earthscan.
Edwards, M. and D. Hulme eds. 1992. Making a difference: NGOs and development in a changing
world. London: Earthscan.
Edwards, M. and D. Hulme eds. 1996. Beyond the magic bullet: NGO performance and
accountability in the post-cold war world. West Hartford: Kumarian Press.
Fowler, A. 1997. Striking a balance. London: Earthscan.
Gasper, D. 1994. Development ethics: An emergent field? In Market forces and world development,
R. Prendergast and F. Stewart eds. London: Macmillan.
Gasper, D. 2004. The ethics of development—From economism to human development. Edinburgh:
Edinburgh University Press.
Goulet, D. 1995. Development ethics: A guide to theory and practice. Muscat: Apex Press.
Hulme, D. and M. Edwards, eds. 1997. NGOs, states and donors: Too close for comfort?
Basingstoke: Palgrave Macmillan.
Korten, D. 1990. Getting to the 21st century: Voluntary action and the global agenda. West
Hartford: Kumarian Press.
Quarles van Ufford, P. and A.K. Giri. 2003. A moral critique of development. Boca Raton, FL:
Taylor and Francis.
Riddell, R. 2007. Does aid really work? Oxford: Oxford University Press.
Smillie, I. 1995. The alms bazaar: Altruism under fire—Non-profit organizations and international
development. London: IT Publications.
Terry, F. 2002. Condemned to repeat? The paradox of humanitarian action. Ithaca, NY: Cornell
University Press.
Weiss, T.G. 1999. Principles, politics, and humanitarian action. Ethics and International Affairs
vol. 13, 1–22.
Ethical Obligations to the Poor in a World
of Nation States
Paul Ronalds
Can we find in the human ‘thou’, a fragment of the divine

‘Thou’? Can we recognise God’s image in one who is not in my
image? (Sacks 2002, p. 17)
1 Introduction
Aid activists and International Non-Government Organisations (INGOs) often
describe the low levels of official development assistance (ODA) provided by some
of the most developed countries like Australia, the United States and Japan as a
‘moral failure’ on the part of their government leaders.1 Such claims make for good
sound bites on the six o’clock news but is there really any philosophical basis for
them? While the laws and institutions of many of the most developed countries
implicitly acknowledge moral obligations between citizens, evidenced by signifi-
cant internal redistribution of wealth through progressive taxation systems, national
health and education schemes and social welfare, the Westphalian System2 has not
traditionally imposed on states or their citizens any general responsibility for the
welfare of those living in other states.3 This dichotomy between insiders and out-
siders permits enormous differences in welfare to exist between citizens of different
countries.
Compare, for example, East Timor and Australia. We would certainly condemn
someone who explicitly argued that the life of an East Timorese was worth less than
that of an Australian.4 Yet in East Timor around 9 percent of children die before their
first birthday (UNDP 2006, p. 8), while in Australia the equivalent figure is 0.5 per-
cent (ABS 2004). In East Timor, more than 50 percent of people do not have access
to safe drinking water (UNDP 2006, p. 1), while in Australia almost 100 percent
have such access (UNICEF 2001). Sixty four percent of the East Timor population
P. Ronalds (B)
World Vision Australia, Melbourne, VIC, Australia
e-mail: Paul.Ronalds@worldvision.com.au


14 P. Ronalds
suffers from food insecurity (UNDP 2006, p. 2) while more than 3 million tonnes
of food is thrown out by Australian homes and businesses each year (Dasey 2005).5
These statistics lead us to ask some key ethical questions. What moral significance
does the Timor Sea have that allows the government of Australian and by implica-
tion in a democratic country, its citizens, to see these differences in human welfare
as acceptable? Put more generally, to what extent do the most developed countries
owe obligations to those outside of their communities, especially the world’s poor?
At the international level, should obligations be limited to not harming other people
or should we also have duties of assistance that, for example, ensure all peoples’
basic needs are met?6
The answers to these questions are important to INGOs for both theoretical and
practical reasons. As Peter Ellis’ paper makes clear, it is important that INGOs clar-
ify the basis for political actions like lobbing states for increased expenditure on
ODA. Arguing that increased ODA is an international ethical obligation helps to
defend INGOs against critics who contend that INGOs are merely self interested
and that prioritising increased aid over other budget items is pandering to narrow
sectorial interests.7 Secondly, the answers will help INGOs decide what is the most
legitimate basis for their advocacy: ethical obligations, human rights and justice or
beneficence and charity. Thirdly, the answers to these questions have ramifications
for how INGOs approach related issues such as humanitarian intervention and aid
effectiveness. Lastly, while this paper focuses on the ethical obligations of states,
campaigns by INGOs to increase ODA in democracies will only be politically effec-
tive if they understand how to mobilise public opinion in support of their goals.
For INGOs, this requires an appreciation for the inter relationship between moral
philosophy and politics.
INGOs also need to understand the impact globalisation, ‘the expanding scale,
growing magnitude, speeding up and deepening impact of transcontinental flows
and patterns of social interactions’ is having on these arguments (Held and McGrew
2002, p. 1). The almost instantaneous transmission of events around the world
through television and the internet has raised our awareness of the needs of those
living in poverty. In our own living room, we can now come face to face with the
need of our fellow humans living on the other side of the world. The result is that, in
Australia, more people than ever believe they have a responsibility to do what they
can about global poverty. In one survey, the number of people expressing this senti-
ment has increased to 47 percent from 39 percent in 2001 (World Vision Australia
2005, p. 18).8 Interestingly, the greatest shift in attitudes was seen within the 14–
17-year-old bracket, with 21.7 percent more young people feeling responsible than
two years previously (World Vision Australia 2005). Similar increasing humanitar-
ian concern is evident in other countries, as demonstrated by the political pressure
to increase aid levels in Europe and to intervene in the affairs of other countries in
situations of gross violations of human rights.9 Globalisation has also increased our
ability to respond to people’s needs. Many years of economic growth in wealthy
countries has significantly improved real wages, giving us a greater financial capac-
ity to respond generously to other’s needs. Technological improvements have also
improved our ability to respond. For example, through air travel, pre-positioned
Ethical Obligations to the Poor in a World of Nation States 15
goods and global communications, World Vision now aims to respond to disasters
anywhere in the world within 48 hours. On the other hand, nationalism with its
emphasis on the ‘uniqueness of a particular nation and on the historical and cul-
tural differences’ of peoples’ remains strong and seems to be ‘intrinsically opposed
to a universalist approach which stresses what all people share in common’ (Carter
1997, p .69). Nationalism’s ongoing strength is evidenced by the numerous demands
of peoples around the world for secession, or for greater autonomy within states, or
in protests about migration or free trade. According to Halliday (2005, p. 522), there
is now ‘much more emphasis on the importance of what distinguishes people’. This
may, in part, be a response to the profound sense of change and loss of identity
many people feel in our modern world and that they often attribute to the impact of
globalisation.
This essay therefore examines arguments about whether ethics has any place
in international relations and, if it does, what is the nature and extent of the most
developed countries’ ethical obligation to the world’s poor, particularly in light of
how globalisation is changing the way developed states and their citizens perceive
their connectedness with the least developed countries and their citizens. Although
it focuses principally on cosmopolitan approaches,10 it also considers realist, liberal
and more radical ideas. It argues that not only is the failure by many of the World’s
wealthiest states to meet the UN target of contributing 0.7 percent of GNI to alle-
viating extreme poverty unethical, it is unsustainable in an increasingly globalised
world. Instead, we should actively encourage the development of a cosmopolitan
ethic that requires the most developed countries like Australia to have a positive,
moral responsibility to adequately support measures to achieve the basic subsistence
and security needs of all people.
2 The Role of Ethics in International Relations

Some early theorists of the nation state, such as Machiavelli, Bodin and Hobbes
argued that while ethics may be relevant to domestic politics, it has no role to play in
international relations. Such views were in contrast to the earlier ‘medieval concep-
tion of the world as a universal moral order’ based on universal principles of natural
law (Beitz 1979, p. 406). More recently, these views, categorised by Beitz (1979, p.
405) as ‘international moral scepticism’, have formed the basis for realism’s rejec-
tion of the role of ethics in international relations. Under the most extreme form of
this view, national interest is the sole determinant and driving force of state action
leaving no relevance for questions of ethics. Consequently, the anarchical nature of
international society never permits states to raise issues like human rights above the
need for self preservation. According to this view, the only basis for foreign aid is
self interest.11 For example, foreign aid may make the donee state more amenable
to giving the donor trade or other concessions such as military bases or, by address-
ing poverty in another country, you may reduce the impact of terrorism, climate
change, pandemics or refugees on your own country. Neo-liberalism took a similar
16 P. Ronalds
view.12 However, as realism’s dominance began to be challenged after the end of

the cold war, there was a renewed interest in the ethical aspects of international rela-
tions issues. While this interest may have been initially dismissed as theoretical or
even utopian, by the early 1990s foreign policy dilemmas like humanitarian inter-
vention, gave such issues practical significance for policy development.13 This trend
has continued into the new millennium. For example, on 27 March 2006, then Prime
Minister Tony Blair, in a speech to the Australian Parliament, stated that ‘we cannot
say we favour freedom but sit by whilst millions in Africa die and millions more are
denied the very basics of life’ and he called for ‘global alliance for these global
values’.14 However, in my opinion, while globalisation may have increased our
interest in ethics at the international level, the realist’s dichotomy between domestic
politics and international relations was always unsustainable, from both a theoreti-
cal and practical perspective. For example, global institutions that fail to deal with
extreme inequality are increasingly likely to be seen as illegitimate.15 This illegit-
imacy was partly responsible for some of the widespread discontent with global
institutions, seen for example in the protests against the WTO in Seattle in 1999.
And as Shue says, while much of the discourse on ethics is indeed about individual
behaviour, to be meaningful, it must consider their context as members of fami-
lies, classes, genders and states, especially in a globalised world. In addition, many
other aspects of ethics are directly concerned with foreign policy and international
institutions (Shue 1995, pp. 453–461).
Of course, one might concede that ethics is relevant to international relations
while arguing that the proper objects of moral concern are states rather than indi-
viduals. Beitz (1979, pp. 407–408) describes this view as the morality of states.16
Support for this view is particularly strong among pluralists, those who value order
above justice in international relations and view such order as the best way to protect
cultural diversity within states. As Nicholas Wheeler (2002, p. 11) argues, pluralists
see attempts to protect human rights through, for example, humanitarian interven-
tion as placing in jeopardy the structure of inter-state order, a ‘violation of the
cardinal rules of sovereignty, non-intervention, and non-use of force’. In contrast,
‘solidarists’ look to strengthen the legitimacy of international society by deepening
its commitment to justice. Solidarists’ view of a society of states is one in which
states accept not only a moral responsibility to protect the security of their own cit-
izens, but also the wider one of ‘guardianship of human rights everywhere’ (Bull
1966, p. 63). They argue that if pluralist ‘diversity entails that states have the right
to mistreat their populations, then it is difficult to see why such diversity is to be
valued’ (Brown 1992, p. 125). They conclude that ‘states that massively violate
human rights should forfeit their right to be treated as legitimate sovereigns, thereby
morally entitling other states to use force to stop the oppression’ (Wheeler 2002,
pp. 12–13). Support for the views of both solidarists and pluralists can be found in
international relations. However, in my view we are seeing an increasing number
of situations where solidarist principles are receiving broad international support.
For example, in Somalia in 1992, the UN Security Council was prepared to override
notions of sovereignty, albeit in the absence of any form of national government,
and support armed humanitarian intervention to protect the delivery of international

aid.17 While such support was not as forthcoming in other similar situations, such
as Kosovo in 1999,18 at the World Summit in September 2005, a majority of the
150 leaders present accepted collective ‘responsibility to help to protect popula-
tions from genocide, war crimes, ethnic cleansing and crimes against humanity’
(United Nations 2005). Such a statement clearly represents a significant break from
the principle of non-intervention and a move towards solidarist principles and sug-
gests that support for the solidarist position has continued to evolve and is now at
least tentatively accepted by a majority of the General Assembly.
From an ethical perspective, the morality of states view is more internally con-
sistent and sustainable than that of realists. As Beitz (1979, pp. 407–408) argues,
the morality of states approach treats states in international society like people in
domestic society, ‘as autonomous sources of ends, morally immune from external
interference, and morally free to arrange their internal affairs as their governments
see fit’ and ‘each state is assumed to have an initial right to the wealth of its territory
and is not subject to any restrictions in subsequent market transactions’. It therefore
might be understood as the ‘international analogue of nineteenth-century liberalism’
(Beitz 1979). Unsurprisingly then, traditional liberals, such as John Rawls, like plu-
ralists of other theoretical persuasions, regard distributive justice19 as inappropriate
to international relations, placing order ahead of justice in the international arena
(Brown 2002, p. 168). The arguments of Rawls and other liberals and neo-liberals
are discussed more comprehensively below.
Ultimately, however, the attractiveness of the morality of states approach from an
ethical perspective requires either that proponents convince us that political bound-
aries have either an intrinsic ethical value or that bounded political communities in
the form of nation states are the most practical way to ensure a ‘good life’. The
difficulty of demonstrating any intrinsic ethical value in state boundaries leads most
proponents to take the latter course.20 There are certainly some good arguments in
favour of retaining state borders based on practical administration, preserving cul-
ture and the advantages that bonds of sentiment between citizens generate (O’Neill
1994). However, in my view, these arguments do not prevent an obligation aris-
ing for some form of international distributive justice. Certainly supporters of a
more cosmopolitan form of liberalism argue that a liberal theory that has nothing to
say about the global inequalities in wealth lacks merit. And while liberal national-
ists may have once been able to argue that citizens of one state were unwilling to
redistribute resource to citizens in much poorer states, proponents of cosmopoli-
tanism can now point to the ‘moral concern with welfare on a world scale’ as
evidence of a growth of . . . cosmopolitan moral awareness’ (Bull 1984, p. 13).
Growing out of an increasing level of interdependence (from increasing globalisa-
tion), modern communication technologies, and international travel, they argue that
in some respects this now requires that the world be treated as ‘though it was a single
society’.21
Even if it is accepted, however, that ethics are relevant to issues in interna-
tional relations and that individuals rather than states are the proper objects of
moral concern, the extent and the ethical nature of those obligations remains to
be determined. It is to this issue that we now turn.
18 P. Ronalds
3 Ethical Basis and Extent of Obligations to the World’s Poor

In western thought, the Greeks were the first to contend that one’s ethical obligations
extended beyond the borders of one’s political community.22 Although the cynic
Diogenes first claimed that he was a ‘citizen of the world [kosmopolites]’, it was the
stoics of the third century that argued for positive ethical obligations to non-citizens
as well as citizens. 23 This cosmopolitan thinking may have significantly influenced
the early Christian church. For example, positive obligations towards non-citizens
are evident in the Parable of the Good Samaritan (Luke 10:25–37).24 Christians
owed obligations to Caesar but the greater work for human goodness is removed
from traditional politics to the sphere in which the people of all nations can become
fellow citizens with the saints (Stanford 2007; Ephesians 2:20).25 This influence
continued into the middle ages: Beitz characterizes the ‘natural law theories of such
medieval Catholic writers as Suarez and Vitoria’ as cosmopolitan as well as the
secular international theory of Grotius (Beitz 1979, p. 408). Many ideas of important
eighteenth and nineteenth century thinkers were also cosmopolitan. For example,
Immanuel Kant in Perpetual Peace (1891) claimed that a violation of rights in one
part of the world is felt everywhere.
Like the morality of states view, cosmopolitan theories reject international skepti-
cism and believe ‘that moral judgments about international relations are meaningful’
(Beitz 1979, p. 408). Their approach opens up the state to external moral assess-
ment and prioritises the moral concern of individual people over all other forms of
human association. However, the extent of the obligations and the political impli-
cations vary considerably between different forms of cosmopolitanism. At one end
of the spectrum is a theorist like Thomas Pogge who argues that global poverty is
primarily due to institutional arrangements in developing countries for which their
political and economic elites bear responsibility and global institutions for which
governments and citizens of affluent countries bear responsibility. He argues that
poverty is caused by these global institutions and this forms the basis for imposing
an obligation on developed countries to either reform such institutions or otherwise
make restitution. Essentially, Pogge combines a very weak form of cosmopolitanism
with an argument that the most developed states are complicit in creating global
poverty to derive the ethical basis for an obligation to the world’s poor. A number
of arguments are advanced by critics of Pogge’s approach. Firstly, it is argued that
Pogge over emphasises the impact of historical processes on local poverty. While
it is certainly true that in many cases colonialism, slavery, unfair trade concessions
and the like significantly contributed to the poverty in the world’s least developed
countries, in other cases, many previously colonised countries have achieved signif-
icant poverty reduction with the current global structures and despite their colonial
history.26 In addition, while the harm done by some colonising countries may give
rise to special obligations towards their former colonies, this does not of itself give
rise to a general obligation on the part of all of the most developed countries to
re-distribute resources. Secondly, while Pogge (2005a, p. 722) rightly rejects argu-
ments that global poverty is based on national factors alone,27 I believe he goes
too far in the other direction, generally downplaying the importance of domestic
factors such as climate, geography, civil war and poor governance which have enor-
mous impact on peoples’ wealth.28 Even if one accepts Pogge’s arguments that
developed countries and their citizens have an ethical obligation to make resti-
tution for their part in imposing an unjust institutional order, if non-institutional
factors play a significant role in contributing to poverty then it is quite conceiv-
able that many of the global poor would still not enjoy basic human rights since
the obligation to compensate is limited to the amount of harm for which devel-
oped countries are responsible.29 Thirdly, critics like Risse argue that Pogge fails
to demonstrate that current global institutional arrangements actively cause excess
poverty, rather than passively fail to prevent it. Certainly, there is little doubt that
global organisations do not work equally for all. For example, the rules of the World
Trade Organization (WTO) and its predecessor the General Agreement on Trade
and Tariffs (GATT), better suit the economic interests of the most developed states
rather than the interests of developing states.30 The benefits of these rules are rein-
forced by the economic power of wealthy developed countries and even when a
WTO case is brought and won by a developing state, a wealthy county’s economic
power makes it very difficult to force it ‘to implement an adverse decision that it
is determined to ignore’ (Winham 2005, p. 109).31 However, this does not satisfy
Pogge’s own test: to impose responsibility on the most developed states for the
world’s poor through a breach of the do no harm principle, it is not sufficient to
show that the WTO favours developed states, that it fails to alleviate poverty or that
another system would reduce poverty more.32 Pogge’s test requires some positive
act or intention, not merely omission.33 As Pogge himself admits, he is propos-
ing a relatively narrow duty, ‘tightly limited in range . . . subject matter. . . and in
demandingness’.34 Pogge’s argument (2005b, p. 61) is appealing because a broader
audience is likely to accept some level of responsibility to the global poor based on
the do no harm principle.35 It also highlights the importance of institutional reform
for the alleviation of global poverty and the role that citizens of affluent countries
have in advocating for it. However, in order to avoid imposing a positive obligation,
I think Pogge has to resort to a much more complex argument to make the case for
change.36 Other cosmopolitan writers do not follow this approach and I will now
examine their arguments.
In a recent article, Andrew Linklater (2006, p. 338) also relies on the do
no harm principle but adopts a much broader definition of harm to invoke cos-
mopolitan ethics ‘demanding positive duties as well as respect for negative rules
of forbearance’. Relying on Adam Smith’s view (1790, p. 79) that ‘the viola-
tion of justice is injury’ he distinguishes simple and complex moral responsibility.
Simple responsibility, reflected in most obligations imposed on states by interna-
tional obligations, merely requires avoiding direct harm. Complex responsibility
involves being more ‘reflective about the ways in which conduct may uninten-
tionally harm other human beings, whether now or in the future’ (Linklater 2006,
p. 338). Examples of complex responsibility can be found in the demands by
non-governmental organisations for socially responsible investment, fair trade and
ethical tourism in order to avert unintended harm (Linklater 2006). He concludes
that the negative obligations, usually associated with the do no harm principle
20 P. Ronalds
cannot be reduced to mere rules of forbearance, ‘they also require duties of assis-
tance, steps to dismantle “coercive regimes” and efforts to create global political
arrangements which can represent the voices of vulnerable people and communi-
ties’ (Linklater 2006, p. 343). While acknowledging such obligations are beyond
the requirements of most current international obligations,37 this is the direction
in which ‘international society must develop if it is to build on past achievements
in making a global version of the harm principle more central to its organisation’
(Linklater 2006, p. 343). Linklater’s argument may be a neat intellectual exercise
but in my view, ultimately seeks to use the do no harm principle for a purpose
too far removed from its traditional liberal meaning. While this may be consistent
with Linklater’s desire to achieve freedom from hunger ‘through discourse ethics
rather than directly through an appeal to fundamental human rights and basic human
needs’ (Linklater 1999, p. 173)38 I find Linklater’s earlier (1998) arguments for a
form of cosmopolitanism that creates duties, including duties to reduce material
inequality to outsiders by increasing the boundaries of political community, more
attractive
Like Linklater’s earlier work, other cosmopolitan writers argue for positive
duties of assistance across state boundaries. According to Held (2004, p. xi), the
cosmopolitan view is that national, ethnic or gendered boundaries ‘should not
determine the limits of rights or responsibilities for the basic human needs’. For
example, he is critical of the narrow conception of security adopted by the US after
11 September 2001 rather than a broad conception ‘based on establishing the essen-
tial conditions for human well being and development’ (Held 2004, p. xiii). Held’s
views echo earlier arguments by scholars such as Henry Shue and Charles Beitz.
For example, as demands for a new international economic order were replaced by
the language of rights, Henry Shue (1983) called for a universal obligation to meet a
person’s ‘basic rights’ which included a minimum level of security and subsistence.
In my view, this limited obligation should recognise that all people have the same
intrinsic worth while still acknowledging that they may be unequal in terms of talent
or in their contributions to society and still allowing different levels of wealth to exist
between people. Such a cosmopolitan obligation, of course, does not exclude more
onerous obligations to those with whom we are in closer relationships—such as our
family, friends, local community or fellow citizens.39 What must be provided is the
minimum required for a person’s human dignity to be honoured. Selnzick (1994, p.
483) describes it as ‘the basic conditions that make life possible, tolerable and hope-
ful’ and that maintains people’s dignity and integrity.40 While this conclusion may
have originally been based in ethics, the most developed states now have strong self
interest reasons for supporting it. In a globalised world, some key public goods can
only be delivered globally and international social unrest impacts domestically. As
a result, social democracies cannot effectively balance their twin commitments to
market orientated policies and social cohesion without a strong cosmopolitan ethic
(Held 2004, pp. 15–16).41
Finally, there are utilitarian scholars like Australian Peter Singer who go much
further than other cosmopolitans like Held and believe in ‘only one standard, that
it is right to do what will have the best consequences’ (Singer 2002, p. 153).42
From a utilitarian perspective, the ‘best consequences’ are those that result in the
greatest good for the greatest number. Singer (2002, pp. 163–4) argues that ‘national
sovereignty has no intrinsic moral weight’ and is merely a secondary principle,
based on lessons from history that respecting national sovereignty promotes peace
between states. For Singer (2002, p. 187), a globalised world is an opportunity to
re-imagine our sense of community, potentially extending it to a world community.
Accordingly, Singer (1977, p. 37) argues that ‘we ought to prevent as much star-
vation as we can, up to the point at which we can do no more without sacrificing
something’ comparable in moral significance to the lives that would be saved as a
result of our preventive action.43 In my opinion, Singer’s view is problematic from
both theoretical and practical perspectives. Firstly, from a particularist perspective,
Singer’s approach is not consistent with ‘common sense morality’ and is unable to
reconcile obligations owed to those who are in a special relationship with us from
those owed to others who are more distant (Brown 2002, pp. 166–7).44 Secondly,
since utilitarian proposals seek redistribution beyond that merely required to meet
basic needs such as freedom from hunger they also appear impractical. As social
liberals like Rawls are right to argue, large scale wealth redistribution of the order
of magnitude undertaken by many developed states is dependent on a sufficient
sense of community which is still lacking at the international level. By comparison
to the high levels of wealth transfers envisaged by utilitarian proposals, proposals
to ensure only the basic needs of all peoples are met require the transfer of about
one percent of developed countries’ GDP, a fraction of the amount distributed by
modern welfare states.45 Secondly, it is by no means certain that distribution of this
magnitude is practically feasible even if it is desirable.46 Despite Singer’s claim that
‘there is no real dilemma’ about how to distribute aid funds ‘so long as we bear in
mind that our goal is to prevent as many people as possible from starving’ (Singer
1977, pp. 45–6), as Beitz (1979, p. 422) rightly points out:
In fact, there are many real dilemmas, not only about distributing aid funds but, more gen-
erally, about preventing as many people as possible from starving. Massive cash transfers
may succeed only in removing incentives for increasing indigenous food production, and
even institutional reforms . . . may only reinforce the structural inequalities found in many
poor societies.
Finally, the large redistributions of wealth proposed by Singer are not the only
means of addressing an ethical obligation to the world’s poor. While in my view
aid is certainly a key component, it is only one of a number of measures that could
contribute to fulfilling the most developed countries’ obligation.
If a cosmopolitan obligation to met basic needs is conceded it logically raises a
number of practical questions about the extent and nature of the obligation, many of
which require INGOs to wrestle with some very complicated policy issues. Firstly,
do all states have the same obligation or do some have more responsibility than oth-
ers? In my view, a general obligation should be imposed on all states but they should
be expected to respond according to their capacity and some may have additional
ethical responsibility based on the past conduct.47 Since the most developed states
have the greatest wealth and they can respond without any significant detriment to
their own citizens’ overall welfare, they should be expected to meet the majority of
22 P. Ronalds
the required financial resources. However, developing countries will often be able
to contribute in other ways and should be expected to do so. Similarly, it seems
reasonable for there to be obligations on those countries to whom resources are
transferred to use them effectively. However, if such reciprocal obligations exist
and a state fails to observe them, does that absolve the most developed coun-
tries of their ethical obligation or should they then suspend a state’s sovereignty
in order to directly assist in developing the requisite domestic institutions? As can
be seen, if a cosmopolitan obligation to meet people’s basic right does exist and
it is taken seriously, it could have profound implications for traditional notions of
state sovereignty, making it effectively subject to the quality of its domestic insti-
tutions. Arguably, this is, in effect, how sovereignty began to be re-interpreted by
some members of the international community during the 1990s. There was a grow-
ing expectation, especially in western countries, that the international community
should intervene, by force if necessary, to prevent or end widespread and grave
violations of fundamental human rights.48 For example, in the case of Somalia in
1992/1993, mentioned above, American public support for humanitarian interven-
tion, together with potential domestic political advantage for President Bush (Sr.)
(Wheeler 2002, pp. 178–182), led to a US offer to lead a UN approved coalition
to create a secure environment for the delivery of aid. With UN Security Council
resolution 814 (approved on 26 March 1993), it then went even further: the revised
mandate included assisting the people of Somalia in ‘rehabilitating their political
institutions and economy’. Unfortunately, this ‘experiment’ was cut short by the
evaporation of US public support for the intervention following the death of eigh-
teen American soldiers in the ‘Black Hawk Down’ incident on 3 October 1993.
The withdrawal of US troops after the incident effectively led to the collapse of the
UNOSOM II mission. While the Somalian situation was unusual due to the lack
of any form of national government, the later examples of Bosnia in 1995 and of
Kosovo in 1999 demonstrates that a group of states may be prepared to act even
where it overrides a state’s sovereignty. Such examples allow us to conceive of a
cosmopolitan obligation to ensure people’s basic needs are met, rather than one that
merely seeks to try to meet those needs. Of course, there are a plethora of ethi-
cal and practical problems to the emergence of such an international obligation to
ensure that people’s basic needs are met. The first is the continuing pluralist reluc-
tance to override state sovereignty, except perhaps in extreme cases. Although the
Somalian example took place in the absence of any national government, Indian
and Chinese representatives on the Security Council were still careful to emphasise
that the Somalian situation was ‘unique’, ‘extraordinary’ or ‘exceptional’ (Wheeler
2002, p. 186). While NATO received Security Council authority to conduct air
strikes against Bosnian Serb targets in 1995, it could not obtain similar approval
in 1999 for its air strikes against Serbian targets over Kosovo, although NATO’s
actions were approved of or acquiesced to by a majority of Security Council mem-
bers (Wheeler 2002, p. 242). At best, any norm of humanitarian intervention remains
contentious. 49 Secondly, in situations where the national interests of a country are
not perceived to be at stake, the public are far less likely to accept the financial
and military costs of a conflict than when a state’s national interests are directly
involved. The increasing public pressure in western states to intervene to stop gross
abuses of human rights has been accompanied by a similarly increasing expecta-
tion that technological advances in armaments since at least the first Gulf War mean
that such humanitarian interventions can be conducted without the significant loss
of their military personnel. This in turn leads to military tactics by the intervening
power that increase civilian casualties, that often does not prevent atrocities or mass
expulsions50 and that may actually prevent the long term resolution of the under-
lying causes of the human rights abuse.51 It also demonstrates the gap between
western leaders’ rhetoric and their commitment to sacrifice their soldiers for non-
citizens in danger (Wheeler 2002, p. 243). Thirdly, there is also the vexed question
of whether and in what circumstances armed force can ever really serve human-
itarian ends.52 That said, the emerging concept of the Responsibility To Protect
(RTP) 53 and the General Assembly’s apparent support for it,54 does have the poten-
tial to both reconstitute states’ interests and provide a principled way to balance state
sovereignty and an ethical obligation on the most developed countries to meet peo-
ples basic needs.55 However, even where the international community is prepared
to override sovereignty and commit military and other resources to restoring peace,
the task of rebuilding a country is incredibly difficult. Again, in Somalia, the UN
forces did not sufficiently understand the implications of the complex clan struc-
ture and failed to engage local communities and their leaders in developing new
political structures that could support the establishment of law and order.56 Given
the complexity of these issues, a detailed examination of an obligation to ensure
basic needs must be deferred until another time. Nonetheless, this brief considera-
tion allows us to at least envisage how an ethical obligation to provide resources to
meet people’s basic needs combined with a set of principles, such as the RTP, that
provide a process for overriding state sovereignty in certain circumstances could
potentially transform the ethical duty from a mere obligation to try to meet peo-
ple’s basic needs to a much stronger obligation to ensure that they are met. It also
demonstrates the importance of INGOs thinking through the logical consequences
and ethical implications associated with their demands for increased ODA.
Although the line between cosmopolitan and non-cosmopolitan proposals is
often blurred, there are a number of non-cosmopolitan approaches to this issue that
are worthy of exploration. The first is proposals by liberals, such as John Rawls that
suggest that the obligations of ‘well ordered’ societies to assist ‘burdened societies’
is limited to assisting them create ‘the appropriate political culture and a rational
and reasonable approach to their affairs’ (Brown 2002, p. 177; Singer 2002, pp.
195–196). Once countries attain such a state, any moral obligations to them cease
(Patten 2005, p. 19). Accordingly, pursuant to liberal reasoning, distributive justice
is generally seen as inappropriate in international relations (Brown 2002, p. 168).
This liberal ‘middle’ view is criticised from both sides of the debate. On the one
hand, as indicated above, cosmopolitan scholars argue that a liberal theory that has
nothing to say about the global inequalities in wealth lacks merit.57 On the other,
neo-liberals, whose ideas achieved a level of international ascendency during the
1970s and 1980s, argue that even social liberal’s proposals are counter-productive to
reducing global poverty. Neo-liberals claim that their policy agenda, often referred
24 P. Ronalds
to as the Washington Consensus, requires national and global institutions to be

designed in a way that results in free and open markets with a minimum of taxes
and regulations so as to attract investment and stimulate economic growth. This
policy prescription, it is argued, will maximise economic growth that will, in turn,
trickle down to the world’s poorest. On this analysis, the principal causes of poverty
are ‘government policies that distort economic incentives, inhibit market forces and
actually work against economic development’ (Gilpin 2001, p. 311). Underpinning
neo-liberal economic theory is the notion of rational economic actors seeking to
maximise their individual utility. This approach usually excludes any notions of
distributive justice since it leads to extensive government intervention and requires
individuals to make an ethical decision that they have sufficient resources to meet
their needs and that it is preferable that other resources be used to meet other peo-
ple’s needs.58 Neo-liberals argue that their policies are supported by empirical data
that shows that globalisation, by which they generally mean economic liberalisation,
has significantly reduced the number of people living in extreme poverty. However,
neo-liberal arguments are unconvincing for at least three reasons. Firstly, the evi-
dence that global poverty has been falling is contested. As Wade (2005, p. 293)
argues based on an analysis of World Bank data, there ‘is a non-trivial probabil-
ity that the liberal argument is wrong’.59 Secondly, real economies do not seem to
operate anything like neo-liberal models: market failure is more prevalent; spill over
effects provide incentives to cluster some economic activities together (Wade 2005,
pp. 310–11), and some countries level of development is so low they cannot easily
take advantage of globalisation. Thirdly, real people do not always act in a ratio-
nal way and notions of morality are a very important determinant in most people’s
decision making.
In response to the broad based critique of neo-liberal’s policies, economists like
Jeffrey Sachs and Joseph Stiglitz have sought to strike a balance between the ‘two
extremes of abandonment of neo-liberalism and total reliance on the market’ (Gilpin
2001, p. 340). Both Sachs and Stiglitz see development as a far more complex pro-
cess than neo-liberals’ one size fits all approach. In Sachs’ view, the standard policy
prescriptions of organisations like the IMF during the 1980s and early 1990s failed
to appreciate the complexity of poverty and were unsuited to developing country
contexts. Instead, he promotes a more comprehensive approach he calls ‘clinical
economics’. Sachs credits globalisation with significant positive impact on poverty,
especially in China and India but argues that it needs to be more enlightened, a
globalisation of ‘democracies, multilateralism, science and technology and a global
economic system designed to meet human needs’ (Sachs 2005a, p. 358). Similarly,
Stiglitz (2006, p. xii) argues that ‘one size fits all solutions do not, can not, cap-
ture’ the complexities of development. While they both believe that globalisation
can contribute to a reduction in poverty and that markets are a critical part of devel-
opment, unlike neo-liberals, they believe that government intervention can have a
positive impact, especially in relation to poverty reduction. While neither Sachs nor
Stiglitz expressly rely on cosmopolitan reasoning, their support for much higher lev-
els of global aid, debt relief and a fairer trade system lead to very similar practical
conclusions.
There are two further, more radical, approaches to global poverty that are worth
considering. The first are social democratic proposals like those of Vandana Shiva
that reject economic globalisation entirely rather than seeking to merely reform it.
Shiva (2000 and 2005) argues that any understanding of human beings that is, at its
heart, economic, is unsustainable. Like Pogge, Shiva believes that poverty is caused
by the appropriation of resources by developed states. However, unlike Pogge who
wants to use globalisation to reduce inequality, she argues that the rules of globali-
sation are inherently inconsistent with justice and describes globalisation as ‘a war
against nature and the poor’ (Shiva 2000, p. 5). Secondly, there is Garrett Hardin’s
lifeboat ethics. Hardin argues against helping the poor using consequentialist rea-
soning, suggesting that to do so provides the wrong incentive for population control,
leading to further environmental degradation and ultimately results in the death of
those in the life boat (people living in developed countries).
Both approaches have attracted significant criticism. Shiva’s arguments seem
implausible and Hardin’s simplistic. Hardin’s lifeboat metaphor: fails to take into
account continued technological innovation (significant since 1974); does not seem
to appreciate that the causes of poverty are much broader than merely population
growth or that population growth slows as income increases; and ignores the signif-
icant resources expended in developed countries that, if redirected, would not have
a significant impact on the welfare of the rich.60 Some may also argue that even if
we cannot save everyone, provided we develop fair procedures for deciding between
people, there is not necessarily moral failure. In a globalised world, the metaphor
also fails to appreciate the inevitable connectedness of those in the lifeboat and those
swimming in the water. Finally, it also requires us to choose between two extreme
choices: providing unrestricted aid or providing no aid. In reality, well targeted aid,
especially for health and education, is likely to reduce population growth and ulti-
mately create new ‘lifeboats’. Vandana’s proposals require the complete dismantling
of an increasingly entrenched economic system and ignore the beneficial effects on
employment and real wages from trade.
The issue of the effectiveness of aid is also often cited as a reason for not act-
ing, even where a theoretical ethical obligation to the world’s poor is conceded.61
The argument made is that practical implementation of aid programs is either
pointless or, worse, counter productive. These arguments are usually based either
on a view about how economic development occurs (such as arguments by neo-
liberal economists that the effect of aid is to distort markets and crowd out private
investment) or about the limitations of existing government and non-government
institutions. In my view, the arguments that aid does not work are not supported by
the evidence. A report released by World Vision Australia (2006), while recognising
that many human development improvements are due to a complex mix of domestic
action, technological improvements and economic growth, nonetheless points to the
impact of aid, especially in areas such as health and education. Similarly, the gov-
ernments of Malaysia, Singapore, Taiwan and Mauritius all recognised that their
long term economic well being required big investments in public health, supported
by overseas aid, to prevent tropical diseases from sapping their economies. These
26 P. Ronalds
investments proved to be a key to their strong economic growth (Diamond 2005,

p. 523). While well targeted and carefully implemented aid can work, there is no
doubt that there are many examples where aid has little or no effect, or worse, had
an adverse impact.62 However, in my view, this does not mean we should abandon
aid but that we endeavour to develop ever more effective means for its implementa-
tion. A variant of this aid effectiveness argument is that it is most efficient for each
group to look after their own. Epistemologically, this view argues that local people
are in the best position to know how to solve their own problems. Such a view calls
for a moral division of labour in which each group is entrusted with looking after
the welfare of its own members as the most efficient way to maximise the welfare
of all. While I agree that local people need to be intimately involved in develop-
ing solutions to the problems that they face and that too often local knowledge and
resources are ignored, local people can be unaware of the systemic causes of their
poverty or the full range of solutions to address it. Even where people understand
the causes of their poverty and have the political will to address it, they may lack
the resources to implement the solutions. It is also true that many poor communities
can benefit from learnings derived from solving poverty in other communities.63
Many of the criticisms of the effectiveness of aid or of international institutions
that deliver it sometimes seem to suggest that because aid’s impact is uncertain,
that there is corruption or some other failure on the part of the beneficiary or
that international institutions are not perfect, that it is somehow better not to act.
Chris Brown (2007, p. 5), for example, recently argued that ‘a great deal of mod-
ern writing on humanitarian intervention and global distributive justice . . . refuses
to accept that sometimes there are no unambiguously right answers’.64 He cites
as examples the ‘clear clash of duties and values’ involved in humanitarian inter-
vention (Brown 2007, p. 9) and what he sees as Singer’s failure to understand the
‘tragedy’ involved in trade offs implicit in the redistribution he proposes. INGOs
maybe partially to blame for the criticisms levelled at aid, often failing to articu-
late the difficulty of their mission or to explicitly recognise the impact on people
in developed countries (often the most vulnerable in their own societies) of adopt-
ing their policy recommendations.65 It is therefore important to acknowledge that
the policies advocated by Held and supported by this author do involve trade-offs
and will undoubtedly leave some worse off. One can imagine, for example, that by
increasing Australia’s aid budget, there is less money for Australian health care that
may result, in the extreme, in the deaths of some Australians. Similarly, Australian
may have to accept a marginally lower standard of living to ensure that the impact
of climate change on the poorest, for whom it will undoubtedly have the greatest
adverse impact, is reduced (IPCC 2007). While the adverse consequences for some
people in the most developed countries of an obligation to meet the basic needs of
the world’s poor must not prevent us from acting to assist a much greater number
of people, it does, once again, reinforce the importance of INGOs thinking through
the competing ethical issues carefully. It also highlights the need for INGOs who
are serious about achieving policy change to consider the political ramifications of
their ethical positions. 66 This is the final issue to be examined by this essay.
4 Political Support for Obligations to the World’s Poor

While ethics should not be about what is most popular and we should expect our
leaders to show both moral leadership and courage, INGOs cannot be blind to pol-
itics. To be able to describe our political leaders’ decision not to meet the UN’s
0.7 percent target for ODA as a ‘moral failure’, requires, in fairness to them, that
the proposals by cosmopolitan scholars to not merely be philosophically correct but
also politically realistic. In the past, this is the basis on which many cosmopolitan
proposals have been criticized: because they lack political realism they are not plau-
sible. While this may have been true in the late 1960s and 1970s when international
relation scholars began to pay more attention to philosophical problems and before
the end of the cold war (Beitz 1979, pp. 404–405), it is not clear that this is still the
case today.
In my view, to generate sufficient political support in liberal democracies so that
government leaders can implement an obligation to meet the basic needs of the
world’s poor requires, firstly, that citizens can see at least some connection between
their government’s expenditure and their own self interest and secondly, that the
cost is not prohibitive. In my view, both of these requirements can now be met.
Modern globalisation means that domestic societies are increasingly vulnerable to
external factors. The only way to effectively ensure public goods like a clean envi-
ronment, global security67 and protection from pandemics will be to move to a
post Westphalian system where we recognise Australians are inextricably linked to
East Timorese, to Papua New Guineans and increasingly to Afghans and Somalis.
This view is accepted by the Australian Government’s aid agency, AusAID. While
acknowledging the ethical imperative of providing aid because ‘it is the right things
to do’ (AusAID 2007),68 also argues for an official aid program on the basis that it
is in Australia’s self interest in a globalised world:
Australia helps partner governments improve law and order, prevent and recover from con-
flict, and manage a range of transnational threats such as people trafficking, illicit drugs,
HIV/AIDS and other communicable diseases. By helping to build stronger communities
and more stable governments we improve our own economic and security interests.
This view is consistent with Australian citizens growing understanding of the link
between ensuring the basic needs of those living overseas and their own security and
well being.69 Of course, where aid is provided on the basis of national interest, there
is a far greater likelihood that it will fail to achieve development outcomes, or worse,
have an adverse impact. From an ethical perspective, while the egoist argument for
humanitarian action asserts that there is a strong convergence between our welfare
and that of the world’s poor, to take ethics seriously is to accept ‘the possibility that
at least sometimes the best course to follow, all things considered, is not the course
that would most advance whichever interests one happens to be attached to, like
the interests of one’s own nation’ (Shue 1995, p. 456). On the other hand, if ethics
always required you to go against your own interests it would be impossible to be
ethical.
28 P. Ronalds
Secondly, the elimination of extreme poverty which would significantly con-

tribute to realising many global public goods can be achieved for a relatively small
cost, generally calculated at less than 1 percent of the most developed countries’
GNI. For example, AusAID (2007) argues that the cost of its 2007–2008 program70
equates to around A$2.40 for each Australian citizen or ‘about the cost of a loaf of
bread a week’. This amounts to around 1 percent of Federal Government expen-
diture compared to the 42 percent spent on social security and welfare. World
Vision estimates that even if ODA was increased to the 0.7 percent GNI target,
it would still only cost around 3 percent of Federal Government expenditure (World
Vision Australia 2007, p. 28). Opinion polls consistently show strong public sup-
port for Australia’s aid program. For example, a May 2007 survey by World Vision
(2007, p. 24) found 83 percent of 600 respondents ‘support’ or ‘strongly sup-
port’ the Australian Government providing overseas aid (Quantum Market Research
2002)71 As a result, I believe there is a sufficient sense of common humanity (or
Rawls’ sense of community) to make this level of expenditure politically feasible.
Combined with the growing sense of inter-connectedness, it is arguable that lim-
ited cosmopolitan proposals such as a positive obligation to meet all peoples’ basic
needs are now not just politically possible but almost a political imperative.72
5 Conclusion
Contemporary globalisation has changed the world. Subsistence farmers in Laos

can now watch images of Hollywood glamour on satellite television. Farmers and
pastoralists in the Horn of Africa create migration pressures in developed countries
as they fight over land made increasingly marginal by climate change. Urbanisation
combined with imported agricultural techniques like intensive poultry farming in
South East Asia significantly increases the risk of pandemics that modern trans-
portation could spread around the world in 48 hours. The result is that the traditional
Westphalian view of state sovereignty looks increasingly unsustainable. While very
few people would argue that national borders are now irrelevant, cosmopolitan pro-
posals like those of Held that impose a positive, moral responsibility to meet the
basic subsistence and security needs of all people, irrespective of nationality, seem
more appropriate, even if morally imperfect. The alternative cosmopolitan approach
of Pogge, that seeks to impose obligations on developed states through a negative
duty rather than a positive one, seems less attractive and more contentious. Once it
is accepted that most poverty is not the result of the actions of developed countries,
a mere duty not to harm means states do not have any responsibility to ameliorate
extreme poverty. Accordingly, morality requires a positive obligation to rescue be
imposed at the international level.
Our increasing interdependence is undoubtedly leading to a growing cosmopoli-
tan awareness. The result is that what may have seemed implausible 25 years ago
now seems obvious. This is already starting to occur in relation to climate change
where the refusal of the Australia under the Howard Government to ratify the Kyoto
protocol became a key reason for its defeat by the Rudd Government. In my view,
as globalisation continues, pressure for such co-ordinated global action in relation

to extreme poverty will also become overwhelming.
Notes
1. According to the latest figures from the OECD (2007), total ODA from members of the
Development Assistance Committee fell by 8.4 percent in real terms on 2006 levels. This rep-
resents 0.28 percent of members’ combined Gross National Income (GNI). In 2007, the only
countries to reach or exceed the target of 0.7 percent of GNI agreed to at the 2002 Monterrey
Conference were Sweden, Luxembourg, Norway, the Netherlands and Denmark. By compar-
ison, the US’ ODA/GNI ratio was 0.16 percent, Japan’s was 0.17 percent and Australia’s was
0.3 percent (although Australia has now set a timetable to increase ODA to 0.50 percent of
GNI by 2015).
2. The treaties of Munster and Osnabruck, signed in 1648 following the end of the Thirty Year
War, are traditionally seen as creating a new era in international relations. From these treaties,
collectively known as the Peace of Westphalia, there emerged international norms and prac-
tices that by the second half of the twentieth century had spread from Europe around the
world.
3. Some obligations are imposed by international law in particular circumstances, such as obli-
gations imposed by international humanitarian law in times of war. Authors such as Carol
Lancaster (2007) also argue that there is an emerging norm in international law for wealthy
states to provide resources for the social and economic development of poor countries. See
also Alston (2005).
4. Such arguments would be contrary to fundamental international documents like the 1948
Universal Declaration of Human Rights and the United Nations Charter that proclaim the
equality of all people. They would also be contrary to the rhetoric of leaders of the most
developed countries. For example, in proclaiming Human Rights Day on 10 December 2004,
President George Bush stated that ‘Freedom and dignity are God’s gift to each man and
woman in the world’. See also the Joint Statement by the European Union and the United
States: “Working Together to Promote Democracy and Support Freedom, the Rule of Law
and Human Rights Worldwide”, 20 June 2005.
5. This equates to approximately 25 percent of the total purchased by Australian households.
6. The do no harm principle can be contrasted with a need-based view, which supports ‘positive’
human rights, such as rights to have one’s basic needs fulfilled (Patten 2005, p. 20)
7. For example, think tanks like the Institute of Public Affairs accuse INGOs of ‘being self-
serving, unrepresentative, non-transparent, dishonest, and unaccountable interest groups that
lack professional or political legitimacy’ (Mendes 2005).
8. Similarly, a 2006 Lowy Institute Poll (2006) found that 80 percent of Australians said that
‘combating world hunger’ was a very important foreign policy goal for Australia.
9. A recent opinion poll by the Chicago Council on Global Affairs and WorldPublicOpinion.org
found that, in each of the eleven countries surveyed, many more people favoured than were
opposed to the proposition that ‘the UN Security Council has the responsibility to authorise
the use of military force to protect people from severe human rights violations such as geno-
cide, even against the will of their own government’. This included approval from 76 percent
of Chinese, 74 percent of Americans, and for example 69 percent of Palestinians, 64 percent
of Israelis, 54 percent of French and Poles, and 51 percent of Indians.
10. Beitz (1999b) provides one of the best descriptions of cosmopolitanism: ‘Cosmopolitanism
stands in contrast to the notion that the boundaries between states, nations or societies have
deep moral significance. It holds that each person is equally a subject of moral concern (or,
that in the justification of choices of action or policy, the interests of each person affected
should be taken equally into account), and that spatial proximity or shared membership are
not in themselves sources of moral privilege. On a cosmopolitan view, there is no funda-
mental moral discontinuity between domestic and international society because states qua
states have no special standing; if individuals have more extensive responsibilities to their
30 P. Ronalds
own compatriots than to foreigners, or if states are entitled to be treated with some special
respect by others in the international arena, this should be explained in a way that is consis-
tent with the basic conception of a single moral realm in which each individual is equally
worthy of concern and respect’. Cosmopolitanism should not be confused with universalism.
Cosmopolitans respect cultural diversity, although most would argue there are limits to the
‘moral validity of particular communities’: (Held 2004, p. 172).
11. This could be described as a group egoist argument in favour of aid. It is also consistent with
some aspects of the idea of ‘good international citizenship’ pioneered by former Australian
foreign minister, Gareth Evans, which ‘rejects the assumption that the national interest always
pulls in the opposite direction to the promotion of human rights’: (Wheeler & Dunne 1998,
p. 848).
12. As Brown (2002, p. 73) argues both neo-realism and neo-liberalism make it very difficult to
ask the questions of what obligations citizens of one state have to the rest of humanity because
the ‘first premise of both of these approaches is that states are rational egotists’. He argues
that this is the case despite neo liberals who assume that states seek absolute gains and neo
realists who assume they seek relative gains.
13. Of course, the claim that ethical reflection is merely utopian is usually made out of ignorance.
It fails to appreciate that most ethical reflection is normative rather than utopian (although it
can sometimes be both). As Shue says (1995, p. 454), ‘it is about what would be better, not
about what would be best’. In addition, foreign policy that is not grounded in ethics risks
lacking legitimacy, either in the eyes of citizens or foreigners or both. As Booth et al., (2001,
p. 2) argue, ‘Foreign policy is always an ethics-in-action; one can no more conceive foreign
policy quarantined from ethical considerations and implications than one can conceive foreign
policy quarantined from power considerations and implications. A point of view rejecting the
unity of politics and ethics is itself a particular ethical perspective’.
14. Tony Blair (2007) expressed similar sentiment in relation to climate change, saying ‘We
know we have a clear interest in combating climate change; but we feel it too, as a moral
duty to successive generations as well as our own’ and claimed that there was a remarkable
‘degree of consensus around a values-based international agenda . . . There is a true sense of
global responsibility’. The ethical dimension to Britain’s foreign policy was announced by
Robin Cook on taking office as foreign secretary in 1997 (Wheeler & Dunne 1998) and has
continued despite the events of 11 September 2001.
15. According to Beitz (1979, p. 418), ‘if international economic institutions (e.g., the interna-
tional market) systematically yield significantly inegalitarian distributions, then we might
regard them as illegitimate and in need of reform.’
16. See also Holzgrefe’s distinction (2003, p. 51) between individualists and collectivists as part
of his ethical categorisation.
17. UN Security Council resolution 794, approved 3 December 1992, authorised the use of ‘all
necessary means to secure as soon as possible a secure environment for humanitarian relief
operations in Somalia’.
18. Although even in this case, a Security Council resolution condemning NATO’s action was
defeated 12 votes to 3 (Wheeler 2002, pp. 278–281).
19. Defined as the principles by which the benefits and burden of economic activity is distributed.
20. Arguably, the pluralist approach was always a practical response to geo-political realities
rather than a principled choice. As Suganami (2002, p. 13) argues, a close reading of Headley
Bull’s writing ‘reveals that solidarism is an interpretation of the existing international
society’ rather than a value judgment. On this interpretation of Bull’s work, while it may
have been right to characterize a solidarist view as ‘premature’ (Bull 1966, p. 73) in 1966,
this is far less certain today.
21. See Brown’s discussion (2002, p. 173) of the work of Charles Beitz.
22. Although Rabbi Jonathan Sacks (2002, p. 12) claims that it was ‘religion that first taught
human beings to look beyond the city-state, the tribe and the nation to humanity as a
whole’. Certainly, the ‘Golden Rule’ of ethics, which can be broadly stated as ‘treat others
as you would like to be treated.’, when extended beyond one’s own political community,
is cosmopolitan in nature and can be found in nearly all of the major religions and cul-
tures including the Jewish (Leviticus 19:33–34), Hindu (Mahabharata 5:15:17), Christianity
(Matthew 7:12) and Confucianism. On the other hand, as Rabbi Sacks acknowledges, the
view that those ‘who do not share my faith, or my race or my ideology, do not share my
humanity’ is responsible for the deaths of more people than any other belief (Sacks 2002, p.
45).
23. Diogenes Laertius VI 63. Diogenes was seeking to avoid any service owed to the Sinopeans,
hence his claim was a negative one (Stanford 2007). See also Carter (1997, p. 70).
24. In Matthew 5:43–44 in the Sermon on the Mount, Jesus goes further than requiring ‘mere’
justice towards strangers, but talked about loving neighbours and enemies.
25. See also Paul’s insistence (Galatians 3:28) that ‘there is neither Jew nor Greek . . . we are all
one in Christ Jesus’.
26. Dependency theories that argued that the development of ‘the North’ entailed the underdevel-
opment of ‘the South’ have generally been discredited (Risse 2005, p. 13)
27. An argument he associates with John Rawls (1993, p. 77).
28. For example, he argues that ‘most of the severe poverty today would be avoided if the design
of the global order were just’ (Pogge 2005b, pp. 76–77). In contrast, Jared Diamond (2005)
argues that differences in people’s environments are the most important determinant of the
course of their history. Paul Collier (2007) identifies four ‘traps’, all of which are domes-
tic, that he argues are behind the poverty of the world’s ‘bottom billion’. Sachs (2005b, p.
44) emphasises domestic causes of poverty, such as ‘geographic isolation, disease and nat-
ural hazards’ that prevent the world’s poor from getting their ‘first foot on the ladder of
development’.
29. For Pogge’s response to this criticism, see Pogge 2005b, pp. 66–67.
30. For example, agreements are often vague where it benefits developed states and precise where
it disadvantages developing states (Wade 2003, p. 630; Pogge 2005a, p. 724).
31. One example is the US effectively ignoring the WTO’s 2004 ruling that its payments to cotton
farmers were illegal.
32. Mathias Risse (2005, pp. 9–11) argues that ‘The global order is not fundamentally unjust;
instead, it is incompletely just’ and that the global order ‘can plausibly be credited with the
considerable improvements in human well-being that have been achieved over the last 200
years’. He specifically discusses the WTO which he argues is an improvement for the world’s
poor on bilateral trade agreements or no rules at all.
33. States and their citizens will only be held accountable where they ‘cooperate in imposing an
institutional order on those whose human rights are unfulfilled’ that ‘foreseeably produces
avoidable human rights deficits’ (Pogge 2005b, pp. 60–61).
34. On the other hand, Pogge (2005b, p. 72) appears to suggest that merely by taking advantage
of, for example, lower global commodity prices, there is a sufficient positive act to create an
obligation of compensation.
35. An obligation based on the ‘do no harm’ principle is generally acceptable to liberal thinkers
while, as Pogge (2005b, p. 69) argues, ‘libertarians and their sympathizers tend to be quite
sceptical of any and all stringent positive duties’.
36. It should be noted that Pogge (2005c, p. 5) does not argue that positive duties ‘to rescue
people from life-threatening poverty’ do not exist, merely that ‘it can be misleading to focus
on them when more stringent negative duties are also in play’.
37. Arguably, the Kyoto Protocol which seeks to address climate change acknowledges com-
plex responsibility in environmental matters and may demonstrate a change in international
thinking due to a significant increase in transnational challenges in a globalised world.
38. For a discussion of discourse ethics, see Devetak (2001, p. 173).
39. To this extent, such a duty may not be inconsistent with some forms of nationalism.
40. See also Sacks (2002, p. 206).
41. In this regard, Held’s arguments combine cosmopolitanism with realist self interest and begin
to resemble Gareth Evans’ concept of Good International Citizenship.
32 P. Ronalds
42. Beitz (1979, p. 419) argues that Singer’s approach is a non-utilitarian form of consequential-
ism because it measures the goodness or badness of the consequences of action in terms of
‘moral importance’ rather than happiness or utility.
43. Singer’s arguments are based on an earlier (1972) article. See also Beitz (1979, p. 416).
44. The response of utilitarians is to argue for two levels of utilitarianism, an intuitive one where
our relationships will guide our calculation of the greatest good and a critical level of morality
that tests out institutions (Hare 1981; Singer 2002, pp. 175–176). While I agree with Brown’s
conclusion, I prefer Rabbi Sacks’ approach to resolving the tension between the particular and
the universal, the individual and the state or between the tribe and humanity. He acknowledges
our moral obligations to all of humanity—the sanctity of life, the dignity of the human person,
the right to be free, to be no person’s slave or the object of someone else’s violence. However,
we have thicker or greater, context bound moral obligations, ‘which confers on us loyalties
and obligations to the members of our own community that go beyond mere justice’ (Sacks
2002, p. 57).
45. AusAID estimates that in Australia 42 percent of Federal Government expenditure is spent
on social security and welfare compared to around 1 percent on overseas aid (AusAID 2007).
46. The expansion of the European Union provides a real world case study. The strain that the
expansion of the EU’s welfare to its new members has created both politically and practically
suggests that a global system many times larger is not yet politically or practically feasible.
Of course, as information systems develop and other technological advancements are made
this may change.
47. For example, the harm done by some colonising countries may give rise to additional
obligation.
48. See the recent opinion poll by the Chicago Council on Global Affairs and
WorldPublicOpinion.org discussed at footnote 9.
49. Wheeler admits that a number of important members of the international community such as
China, Russia and India oppose such a norm and Brown argues that creating a norm of human-
itarian intervention requires a broader base than affluent western states of Europe and North
America: Brown (2002), p. 157. Following the Kosovo War, however, Kofi Annan (1999)
declared that there was a ‘developing international norm’ to forcibly protect civilians who
were at risk from genocide and large scale killing. According to Evans and Sahnoun (2002,
pp. 1–2), this led to ‘cantankerous exchanges in which fervent supporters of intervention on
human rights grounds, opposed by anxious defenders of state sovereignty, dug themselves
deeper and deeper into opposing trenches’.
50. The decision to rely solely on air power in Kosovo may have intensified the death and
expulsion of Kosovar Albanians (Wheeler 2002, p. 269).
51. For example, in Somalia, US tactics undermined local support reducing their capacity to
negotiate a long term political sentiment.
52. For a good illustration of how vexed the competing moral arguments can be in the context of
a humanitarian intervention, see Wheeler (2002, pp. 281–284).
53. The RTP concept arose out of a report in December 2001 from the International Commission
on Intervention and State Sovereignty, co-chaired by Australia’s Gareth Evans and Algerian
diplomat, Mohamed Sahnoun.
54. The UN General Assembly appeared to at least partially endorse the RTP at the UN’s 60th
Anniversary World Summit in September 2005 by including references to it in the Summit
Outcome Document at paras 138–139. For a review of the implications of this inclusion, see
Bellamy (2006).
55. Perhaps surprisingly, African countries have taken a lead in this area by creating a regional
mechanism for humanitarian intervention in Article 4 of the Constitutive Act of 2002 and by
creating the Peace and Security Council in 2004. For a recent discussion of this mechanism,
see Bellamy (2006, pp. 157–162).
56. According to Human Rights Watch (1993) ‘little evaluation of local needs took place before
attempts were made to set up the regional and district councils’ and there is ‘little indication
that UNOSOM’s political office consulted or worked through any of the local voluntary
organizations that sprang up in many areas, and which often operated across clan lines.’
57. For a more extensive critique of Rawls and similar scholars who emphasise justice between
societies (described as social liberals) compared with cosmopolitan justice, a concern about
justice between persons, see Beitz (1999). See also Brown (2002, p. 177–178).
58. On the other hand, such a decision would be consistent with the alternative liberal view,
described by Burchill (2001, pp. 29–30) as the ethical view in liberalism or ‘ethical liberalism’
The result is that whether or not you think that liberalism can engage with the global poverty
debate will depend on the type of liberalism that you subscribe to.
59. See also Reddy and Pogge’s papers on this theme, available at www.socialanalysis.org . Both
argue that the World Bank data underestimates the level of poverty and that the purchasing
power parity approach used by the World Bank should be replaced with one that emphasizes
achieving basic needs such as a minimal level of nutrition. And, as Grewal argues, even
if poverty has fallen over recent decades, neo-liberals still need to demonstrate that other
policies would not have been more effective: Grewal (2006, pp. 252–253). On the other hand,
the World Bank’s Global Monitoring Report 2007 on the MDG’s says the proportion of the
population in developing countries living on $1/day fell from 29 percent in 1990 to 18 percent
in 2004, and is projected to fall to 12 percent by 2015.
60. As Singer (1999) argues, ‘The average family in the United States spends almost one-third of
its income on things that are [not] necessary to them . . . so much of our income is spent on
thing not essential to the preservation of our lives and health.
61. Keith Horton’s paper refers to not knowing whether aid is effective as the ‘epistemic
problem’, although the paper is dealing with private giving rather than ODA.
62. While it is not the purpose of this essay to provide a detailed defence of the effectiveness
of aid, it is appropriate to mention some of the recent books that address this issue. These
include: Dichter (2003), Easterly (2006), Maren (2001), Terry (2002), Hancock (1994), Sogge
(2002), Waal (1998), Reiff (2003), and Riddell (2007).
63. This is not accepted by everyone: ‘particularists, especially communitarians and postmodern
relativists, reply that universalism masks ethnocentrism and (Northern) cultural imperialism’,
(Crocker 1998).
64. Beitz (1979, p. 418) makes a similar point (even though he is one of the cosmopolitan writers
than Brown believes fails to identify this issue): ‘But the arguments for global redistribution
are not therefore pointless, for they imply that the prevailing distribution is morally unac-
ceptable and, thus, that the existing institutional structure of the world economy is morally
illegitimate. While the best solution to this problem may be practically unavailable, there may
be second-best solutions, involving, for example, conventional measures such as foreign aid’.
65. Brown (2007, p. 9) cites the example of trade policy: ‘genuinely free trade clearly involves
rich countries sacrificing the interests of domestic workers in the medium term while pro-
tectionism clearly damages the interests of workers in low-wage countries . . . to act is to do
wrong to someone’. While Brown’s argument may over state the case and ignore the positive
impact on some domestic workers of lower prices and new jobs or the costs of protectionism,
the underlying point still holds.
66. As Brown (2002, p. 185) argues, there is a role for ideal theory in establishing a ‘goal to
aim for, but this needs to be combined with a more overtly political account of how the
present system of international economic inequality came into existence, whose interests it
serves, and how, politically, it can be adapted or changed to serve the interests of the poor and
downtrodden’; or as Rawls (1999, p. 128) puts it, ‘political philosophy provides a long term
goal of political endeavour’.
67. Collier (2007, p. 139) argues, the security benefit alone of aid often exceeds its cost. He argues
that in post conflict situations, ‘the security benefits alone are more than enough to justify a
large aid program’. In other circumstances, the security benefits reached perhaps half of the
cost of the aid, before counting the benefits to developed countries of reduced drug trafficking
and terrorism.
34 P. Ronalds
68. Although at least one commentator on AusAID’s recent white paper suggests that Australia’s
aid budget is ‘still driven more by self interest than altruism’ and is ‘as much about protecting
Australia’s boundaries in the age of terrorism and trans-national crime as . . . about helping
those neighbours’ (Baker 2004). This is probably even more true of USAID’s program.
69. As indicated in the introduction to this paper above, recent surveys by World Vision suggest
that Australians have a growing sense of responsibility for global poverty.
70. Based on an ODA of A$3,155.3 million.
71. Other research that supports this includes 2002 research of young people commissioned by
World Vision that included a question about support for an additional tax of 1 percent for the
world’s poor. Thirty three percent of respondents indicated they were ‘very likely’ to support
such a tax and 49 percent ‘quite likely’ suggesting there is political support for a tax to help
the poor.
72. Both of Australia’s two main political parties entered the 2007 Federal election campaign
promising to increase ODA should they be elected.
References
ABS, 2004. 4829.0.55.001—Health of children: A snapshot, 2004–2005. http://www.abs.gov.au/
AUSSTATS/abs@.nsf/mf/4829.0.55.001/. Accessed 15 Apr 2008.
Alston, P. 2005. Ships passing in the night: The current state of the human rights and development
debate seen through the lens of the millennium development goals. Human Rights Quarterly
27(3): 755–829.
Annan, K. 1999. Address to the United Nations General Assembly, Sep 1999.
AusAID. 2007. About Australia’s aid program. http://www.ausaid.gov.au/makediff/default.cfm.
Accessed 15 Apr 2008.
Baker, M. 2004. Hardly the fairy godfather: Australia’s assistance to poorer neighbours is based
on its own need. The Age 29 Apr.
Beitz, C. 1979. Bounded morality: Justice and the state in world politics. International
Organization 33(3): 405.
Beitz, C. 1999a. International liberalism & distributive justice. World Politics 51: 269–296.
Beitz, C. 1999b. Philosophy of international relations. In Routledge Encyclopedia of Philosophy,
ed. E. Craig. London: Routledge.
Beitz, C. 1999c. Social and cosmopolitan liberalism. International Affairs 75(3): 515–529.
Bellamy, A. 2006. What will become of the responsibility to protect. Ethics and International
Affairs 20(2): 143–169.
Blair, T. 2007. Speech to the World Economic Forum, Davos, 27 Jan.
Booth, K., T. Dunne, and M. Cox 2001. How We Might Live: Global Ethics in a New Century.
Cambridge: Cambridge University Press.
Brown, C. 1992. International Relations Theory: New Normative Approaches. Hemel Hempstead:
Harvester Wheatsheaf.
Brown, C. 2002. Sovereignty, Rights, and Justice: International Political Theory Today.
Cambridge: Polity Press.
Brown, C. 2007. Tragedy, tragic choices and contemporary international political theory.
International Relations 21(1): 5–13.
Bull, H. 1966. The grotian conception of international society. In Diplomatic Investigations, eds.
H. Butterfield and M. Wright. London: Allen & Unwin.
Bull, H. 1984. Justice in international relations. The Hagey Lectures, University of Ottawa,
Waterloo.
Burchill, S. 2001. Liberalism, In Theories of International Relations, ed. S. Burchill. New York:
Palgrave MacMillan.
Carter, A. 1997. Nationalism and global citizenship. Australian Journal of Politics and History
43(1): 67–81.
Collier, P. 2007. The Bottom Billion: Why the Poorest Countries are Failing and What Can Be
Done About It. New York: OUP.
Crocker, D. 1998. International Development Ethics. Paper Delivered at the Twentieth World
Congress of Philosophy, in Boston, Massachusetts. http://www.bu.edu/wcp/Papers/OApp/
OAppCroc.htm. Aug 10–15.
Dasey, D. 2005. How our greed costs millions every year. The Sun-Herald May 8.
Devetak, R. 2001. Critical theory. In Theories of International Relations, ed. S. Burchill. New
York: Palgrave MacMillan.
Diamond, J. 2005. Guns, Germs and Steel: The Fates of Human Societies. New York: Norton &
Company.
Dichter, T. 2003. Despite Good Intentions: Why Development Assistance to the Third World Has
Failed. Amherst, MA: University of Massachusetts Press.
Easterly, W. 2006. The White Man’s Burden: Why the West’s Efforts to Aid the Rest Have Done So
Much Ill and So Little Good. New York: Penguin.
European Union & The United States. 2005. Working Together to Promote Democracy and Support
Freedom, the Rule of Law and Human Rights Worldwide. Joint Statement by the European
Union and the United States, 20 June.
Evans, G. and M. Sahnoun 2002. The responsibility to protect. Foreign Affairs 81(6): 1–2.
Gilpin, R. 2001. Global Political Economy: Understanding the International Economy. Princeton,
NJ: PUP.
Grewal, D. 2006. Is globalization working? Ethics and International Affairs 20(2): 247–259.
Halliday, F. 2005. Nationalism, In The Globilization of World Politics, eds. J. Baylis and S. Smith,
3rd edn. Oxford: OUP.
Hancock, G. 1994. The Lords of Poverty: The Power, Prestige, and Corruption of the International
Aid Business. London: Atlantic Monthly Press.
Hardin, G. 1974a. Lifeboat ethics: The case against helping the poor. Psychology Today Sep,
38–43.
Hardin, G. 1974b. Lifeboat ethics. In Environmental Ethics: Readings in Theory and Application,
ed. L. Pojman. Stamford, CT: Wadsworth/Thomson Learning.
Hare, R.M. 1981. Moral Thinking: Its Levels, Method and Point. Oxford: Clarendon Press.
Held, D. 2004. Global Covenant: The Social Democratic Alternative to the Washington Consensus.
Held, D. and A. McGrew 2002. Globalisation/Anti-globalisation. Cambridge: Polity Press.
Holzgrefe, J. 2003. The humanitarian intervention debate. In Humanitarian Intervention, Ethical,
Legal and Political Dilemmas, eds. J.L. Holzgrefe and R.O. Keohane. Cambridge: Cambridge
University Press.
Human Rights Watch. 1993. Somalia: Human rights developments. http://www.hrw.org/reports/
1994/WR94/Africa-08.htm#P356_163056. Accessed on 25 May 2007.
International Commission on Intervention and State Sovereignty. 2001. The Responsibility to
Protect, Ottawa.
IPCC. 2007. Climate Change 2007: Impacts, Adaptation and Vulnerability—Summary for
Policy Makers. Contribution of Working Group II to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change, Geneva, Intergovernmental Panel on Climate
Change, 22 Apr, 2007. http://www.ipcc.ch/SPM13apr07.pdf.
Kant, I. 1891. Kant’s Principles of Politics, Including His Essay on Perpetual Peace. A
Contribution to Political Science, (trans. W. Hastie). Edinburgh: Clark.
Lancaster, C. 2007. Foreign Aid: Diplomacy, Development, Domestic Politics. Chicago: The
University of Chicago Press.
Linklater, A. 1998. The Transformation of Political Community: Ethical Foundations of the Post
Westphalian Era. Cambridge: Polity Press.
Linklater, A. 1999. Transforming political community: A response to the critics. Review of
International Studies 25: 165–175.
Linklater, A. 2006. The harm principle and global ethics. Global Society 20(3): 329–343.
36 P. Ronalds
Lowy Institute. 2006. Lowy Institute Poll 2006, Australia, Indonesia and the World. Sydney: Lowy
Institute.
Maren, M. 2001. The Road to Hell: The Ravaging Effects of Foreign Aid and International Charity.
New York: Free Press.
Mendes, P. 2005. The hidden agenda of the NGO wars. Canberra Times 4 Mar.
OECD. http://www.oecd.org/document/17/0,2340,en_2649_201185_38341265_1_1_1_1,00.html.
O’Neill, O. 1994. Justice and boundaries. In Political Restructuring in Europe: Ethical
Perspectives, ed. C. Brown. London: Routledge.
Patten, A. 2005. Should we stop thinking about poverty in terms of helping the poor? Ethics and
International Affairs 19(1): 19–27.
Pogge, T. 2005a. Recognised and violated by international law: The human rights of the global
poor. Leiden Journal of International Law 18: 717–745.
Pogge, T. 2005b. Severe poverty as a violation of negative duties: reply to the critics. Ethics and
International Affairs 19(1): 55.
Pogge, T. 2005c. World poverty and human rights. Ethics and International Affairs 19(1): 1–7.
Quantum Market Research. 2002. Report Card 2002 Youth Research, Melbourne, July
(Unpublished research commission by World Vision Australia)
Rawls, J. 1993. The law of peoples. In On Human Rights: The Amnesty Lectures of 1993, eds.
S. Shute and S. Hurley. Oxford: Basic Books.
Rawls, J. 1999. The Law of Peoples. Cambridge MA: Harvard University Press.
Reiff, D. 2003. A Bed for the Night: Humanitarianism in Crisis. New York: Simon & Schuster.
Riddell, R. 2007. Does Aid Really Work? Oxford: OUP.
Risse, M. 2005. Do we owe the global poor assistance or rectification? Ethics and International
Affairs 19(1): 9–18.
Sachs, J. 2005a. The End of Poverty. London: Penguin.
Sachs, J. 2005b. The end of poverty. Time Magazine 14 Mar.
Sacks, J. 2002. The Dignity of Difference: How to Avoid the Clash of Civilisations. London:
Continuum Books.
Selznick, P. 1994. The Moral Commonwealth. Berkley CA: University of California Press.
Shiva, V. 2000. Globalisation and poverty, Reith Lectures, BBC, London. http://news.bbc.co.uk/
hi/english/static/events/reith_2000/lecture5.htm .
Shiva, V. 2005. Making Poverty History and the History of Poverty. New Delhi: Navdanya.
http://www.navdanya.org/articles/end-poverty.htm .
Shue, H. 1983. Basic Rights: Subsistence, Affluence and US Foreign Policy. Princeton, NJ:
Princeton University Press.
Shue, H. 1995. Ethics, the environment and the changing international order. International Affairs
71: 453–461.
Singer, P. 1972. Famine, affluence and morality. Philosophy and Public Affairs 1(3):
229–243.
Singer, P. 1977. Reconsidering the famine relief argument. In Food Policy: The Responsibility of
the United States in the Life and Death Choices, eds. P. Brown and H. Shue . New York: Free
Press.
Singer, P. 1999. The singer solution to world poverty. New York Times Sep 5.
Singer, P. 2002. One World: The Ethics of Globalisation. Melbourne: Yale University Press.
Smith, A. 1790. The Theory of Moral Sentiments. 6th edn. London: A. Millar.
Sogge, D. 2002. Give And Take: What’s the Matter with Foreign Aid? London: Zed Books.
Stanford Encyclopaedia of Philosophy. 2007. http://plato.stanford.edu/entries/cosmopolitanism /.
Accessed 31 May 2007.
Stiglitz, J. 2006. Making Globalization Work. Melbourne: Penguin.
Suganami, H. 2002. The international society perspective on world politics reconsidered.
International Relations of the Asia 2: 1–28.
Terry, F. 2002. Condemned to Repeat? The Paradox of Humanitarian Action. New York: Cornell
University Press.
UNDP. 2006. Paths Out of Poverty: Integrated Rural Development, UNDP Human Development
Report 2005. New York: UNDP.
United Nations. 2005. United Nations Summit Outcome Document, Paragraphs 138–139, New
York. http://www.un.org/summit2005/documents.html.
United Nations’ Children’s Fund (UNICEF). 2001, State of the World’s Children 2002, Geneva.
http://www.unicef.org/sowc02 /.
Waal, A. 1998. Famine Crimes: Politics & the Disaster Relief Industry in Africa. Bloomington:
Indiana University Press.
Wade, R. 2003. What strategies are viable for developing countries today? The World Trade
Organisation and the shrinking of development space. Review of International Political
Economy 10(4): 621–644.
Wade, R. 2005. Globalization, poverty and inequality. In Global Political Economy,
ed. J. Ravenhill. Oxford: Oxford University Press.
Wheeler, N. 2002. Saving Strangers. Oxford: Oxford University Press.
Wheeler, N. and T. Dunne. 1998. Good international citizenship: A third way for British foreign
policy. International Affairs 74(4): 847–870.
Winham, G. 2005. The evolution of the global trade regime. In Global Political Economy, ed.
J. Ravenhill. Oxford: Oxford University Press.
World Vision Australia. 2005. Island Nation or Global Citizen. Melbourne: World Vision
Australia.
World Vision Australia . 2006. Does Aid Work? Melbourne: World Vision Australia, Apr.
World Vision Australia. 2007. Island Nation or Global Citizen. Melbourne: World Vision
Australia.
Human Rights, Development INGOs
and Priorities for Action
Kieran Donaghue
1 Introduction
International NGOs (INGOs) dedicated to securing the vital interests of vulnera-

ble human beings are commonly divided into: human rights INGOs (e.g. Amnesty
International and Human Rights Watch); humanitarian/relief organisations (e.g. the
International Committee of the Red Cross, Medicins Sans Frontieres); and devel-
opment INGOs (e.g. Oxfam, World Vision, Plan International, Save the Children
Fund, CARE).
These categories are not mutually exclusive, with many development INGOs also
working in the relief field. However, until relatively recently human rights and devel-
opment INGOs moved in largely separate universes defined by distinct intellectual
traditions. This began to change in the 1990s, with (i) human rights INGOs pay-
ing increasing attention to economic and social rights, in addition to their work on
civil and political rights, and (ii) a number of development agencies and scholars
exploring rights-based approaches to development. However, collaboration among
human rights and development INGOs, although growing,1 remains the exception
rather than the rule. And it remains unclear to what extent the rights-based approach
amounts to much more than (to use an ungenerous characterisation) a rhetorical
ploy to benefit from the moral and political appeal of human rights discourse while
changing little in development practice (Uvin 2002; Slim 2002).
I believe that a rights approach to development, drawing on a conceptual analysis
of rights and their attendant duties, can have important practical implications for
development INGOs. In what follows I will seek to substantiate this claim by means
of an examination of the analysis of basic rights provided by the philosopher Henry
Shue.
But before turning to Shue, it will be useful to explore briefly the background to
the distinction between human rights and development.
K. Donaghue (B)
Centre for Applied Philosophy and Public Ethics, Australian National University,
Canberra ACT, Australia
e-mail: kieran.donaghue@gmail.com


40 K. Donaghue
2 Human Rights and Development
2.1 Human Rights: Conceptual and Ideological Divides

The separation between human rights and development INGOs reflects a debate
that has taken place since the second world war about the respective status of
civil and political rights (hereafter CP rights) and economic, social and cultural
rights (hereafter ESC rights), each codified in their respective covenants (UN 1966a,
UN 1966b). The position associated with the rich (‘Western’) world has been that
ECS rights are in a crucial sense aspirational, a political ideal dependent for their
fulfilment on the wealth of respective countries. This point is reflected in the ref-
erence in the International Covenant on Economic, Social and Cultural Rights to
the duty to meet these rights subject to the availability of resources, a qualifi-
cation absent from the International Covenant on Civil and Political Rights (UN
1966a, Article 2.1). Advocates of this viewpoint tend to see CP rights as nega-
tive, requiring for the most part forbearance from duty holders and therefore largely
independent of resource constraints. ECS rights in contrast they see as positive,
requiring substantial resources for successful outcomes. Moreover, so the propo-
nents, violations of CP rights can normally be clearly specified and traced to the
wrongful actions of particular agents, and remedies for these violations are there-
fore easy to designate: the cessation of the wrongful behaviour. According to this
view it is not possible to allocate the duties associated with ECS rights in the clear
and concise way that it is with CP rights, so talk about ECS rights is loose talk.
Such ‘rights’ represent aspirations rather than rights properly so called (Cranston
1973).
While many (although by no means all) Western voices were sceptical of ECS
rights, there was a widespread belief in Western countries that a large part of the
substance of these ‘rights’—access to the material means of a good life through
economic growth—was attainable by the poor of the world. A common Western
view in the aftermath of the second world war (the memory of the 1930s depression
notwithstanding) was that the rich world had accumulated the technical and man-
agerial knowledge and the institutions needed to abolish poverty, and for reasons of
self-interest and common humanity it should share this knowledge and these insti-
tutions, together with financial resources as needed, with the poor world. The word
encapsulating this thinking was ‘development’. Poverty was the result of under-
development, a failure to modernise, a default position if modernisation had failed
to occur. Those who had arrived (the developed world) could and should help those
who had fallen behind. But this was not a matter of right in any strong sense. The
poor world had no right to develop, and the rich world had no duty—certainly not in
any more than the vaguest sense—to contribute to the development of others. The
development of the ‘third world’ was throughout the cold war primarily a political
issue, with official (state-financed) development assistance largely a tactic to reduce
the chances that the developing world would ‘fall’ to communism (Rostow, 1960;
Easterly 2002).
Human Rights, Development INGOs and Priorities for Action 41
2.2 A Right to Development

The Western view of the secondary character of ECS rights was challenged by the
global ‘south’ which, beginning in the 1970s, mounted a campaign in favour of ECS
rights in the context of its call for a ‘new international economic order’. This cam-
paign culminated in 1986 with the adoption by the UN General Assembly of the
Declaration of the Right to Development. However, key Western countries either
voted against the Declaration (the United States) or abstained (eight others), and
the process of providing the Declaration with operational weight has been largely
unsuccessful, mired in politicised debates about the West’s responsibilities and obli-
gations in respect of the poverty of developing countries (Nelson and Dorsey 2003;
Marks 2004). The right to development has played a negligible role in shaping the
development assistance industry.
2.3 Understanding Development in Terms of Rights
Alongside arguments over whether development is properly conceived of as the

substance of a human right, the last decade has seen attempts by both scholars
and practitioners to use the language of human rights to explicate the concept of
development. UNDP devoted its 2000 Human Development Report to the topic
(UNDP 2000), and some bilateral aid donors and NGOs have incorporated rights
language into the conceptualisation of their development work (DfID 2000; Oxfam
International 2000).
Advocates of the human rights approach identify a number of distinct ways in
which human rights are relevant to development and poverty reduction. Constitutive
relevance refers to the ways in which poverty can be defined in terms of non-
fulfilment of rights, such as the rights to an adequate standard of living, including
health and education. Instrumental relevance refers to ways in which the fulfilment
of rights can promote the cause of poverty reduction, even when these rights are
not part of the definition of poverty. Instrumental relevance can be causally based,
for example when CP rights such as rights to freedom of expression affect the way
governments protect, or fail to protect, access to economic resources. Instrumental
relevance can also be evaluatively based, as when particular rights are needed to
guarantee participation in the process whereby a society determines which features
are to count as basic. Finally, human rights can have relevance by constraining the
actions that governments and other agents can legitimately take to reduce poverty.
Forced sterilisation, for example, while it might be undertaken in the context of a
population program designed to protect people’s economic rights, violates the rights
associated with personal integrity and is therefore morally proscribed (Hunt et al.,
2004).
A further theme in the human rights approach to development discussion is the
importance to poverty reduction of the empowerment of the poor and the ways
in which rights support empowerment. Rights provide the poor with entitlements,
42 K. Donaghue
which are claims that they can make against others. The response of those against
whom claims are made is thereby taken out of the domain of charity and given a
much more demanding moral foundation. Rights holders need not, indeed should
not, be supplicants. The recognition of this is itself empowering—or so it is argued.
And given the widespread existence of legally binding human rights instruments,
claims consequent upon the existence of rights can, in theory and sometimes also in
practice, be pursued through legal channels (Hunt et al., 2004).
Rights language also provides a strong foundation for the accountability that is
increasingly called for from organisations working to promote development and
reduce poverty. To the extent that these organisations are responding to the duties
correlative to the rights of the poor, they have a clear responsibility as duty-holders
to articulate how they understand these duties and to what extent they are meeting
them. They also agree that their work should be measured against recognised stan-
dards, namely those articulated in international human rights instruments (Nelson
and Dorsey 2003).
2.4 Human Rights and States
Human rights as defined and expressed in the Universal Declaration are primar-
ily addressed to individual states as relevant duty holders, and to issues where the
actions of individual states in meeting these duties are seen as properly matters of
international concern (Beitz 2004). In terms of rights and duties as they apply to
development and poverty reduction, it is again commonly assumed that the primary
duty holder is the national state, although other (particularly developed) states and
intergovernmental organisations are also recognised as having important respon-
sibilities, both in helping to create an international environment conducive to the
realisation of rights and in aiding those whose rights are violated. The extensive
work that has been done to encapsulate human rights in legally-binding national
and international human rights instruments reflects the overall focus on states, pri-
marily the individual national state but secondarily the international community of
states, as the principal agents of human rights implementation.
It might seem that this focus on states as primary duty bearers must inevitably
marginalise INGOs in matters of human rights. The fact that rights-holders have
claims on duty-bearers is without doubt central to any human rights approach, and
INGOs committed to such an approach must be especially attentive to where specific
duties most logically lie and how their actions bear on the responsibilities of duty-
holders. This may mean that INGOs need to alter some of their practices, but it does
not mean (or so I will argue) that they need be reduced to marginal players when it
comes to human rights.
3 Henry Shue on Basic Rights and Duties
3.1 Human Rights and Correlative Duties—Conceptual Issues

The philosopher Henry Shue has developed an analysis of moral rights and their
correlative duties that I find attractive. For Shue a moral right ‘. . . provides (1) the
rational basis for a justified demand (2) that the actual enjoyment of a substance
be (3) socially guaranteed against standard threats’ (Shue 1980, p. 13). A right is a
claim for which reasons in the form of principles can be adduced. Such principles
justify the claimant in demanding (not just asking) of others that his or her rights
be respected—with the justifiability of the demand being central to human dignity
(Feinberg 1973, cited in Shue 1980, p. 13). For a right to be respected the actual
substance of the right (that which the right is a right to, for example freedom of
speech) must be respected. And for this respect to obtain it is not enough for a
person simply to enjoy the substance of a right. A right is properly respected, Shue
maintains, only if society has made provisions to protect the rights-holder against
standard threats to the substance of the right, where standard threats are those that
can reasonably be foreseen and protected against. One has no right to immortality,
for example, because it is not within the power of any human agent, individual or
collective, to protect against all threats to human life.
Shue suggests that being socially guaranteed is probably the most important fea-
ture of a right, because it highlights the correlative duties which each right entails
and underlines the weight attached to these duties. He writes:
Suppose people have a right to physical security. Some of them may nevertheless choose to
hire their own private guards, as if they had no right to social guarantees. But they would
be justified, and everyone else is justified, in demanding that somebody somewhere make
some effective arrangements to establish and maintain security. Whether the arrangements
should be governmental or non-governmental; local, national, or international; participatory
or non-participatory, are all difficult questions. . . But it is essential to a right that it is a
demand upon others, however difficult it is to specify exactly which others. (Shue 1980,
p. 16)
Correlative duties have a threefold structure, which Shue characterises as duties
I. not to deprive rights holders of the substance of their rights;

II. to protect rights holders from deprivation by third parties; and
III. to aid those who have been deprived.
Shue fills out this classification by elaborating duties of types II and III in the
following way. The type II duty to protect from deprivation comprises: (a) enforcing
the duty not to deprive; and (b) designing institutions that avoid providing strong
incentives to violate the duty not to deprive. The type III duty comprises the duty
to aid those who are deprived of the substance of their rights and who: (a) are one’s
special responsibility, e.g. one’s children; (b) are deprived of the substance of their
rights because of a social failure in the performance of duties I and II; and (c) are
deprived of their rights because of some natural, i.e. non-social, circumstance (Shue
1980, pp. 52–57).
Shue claims a number of advantages for this analysis. First, he argues that it
shows to be misguided ‘. . . the common notion that rights can be divided into rights
to forbearance (so-called negative rights), as if some rights have correlative duties
only to avoid depriving, and rights to aid (so-called positive rights), as if some rights
have correlative duties only to aid’ (Shue 1980, p. 53). Shue’s suggestion that it is
duties, not rights, that can be divided into the positive and the negative, and that
44 K. Donaghue
all rights, whether CP or ECS, have correlative duties that include action (protect-
ing from violation and aiding the violated) as well as forbearance (refraining from
violating) undermines the key argument of those who would sharply distinguish
between CP and ECS rights.
The second point emphasised in Shue’s analysis is the social or institutional
dimension of human rights, since he stresses that the duties correlative to rights
require the creation of institutions if these duties are to be fulfilled successfully.
This is evident in the typology of duties discussed above, in particular duty II to
protect rights-holders from the violation of their rights (Shue 1980, p. 17).
In the Afterword to the 2nd edition of Basic Rights, Shue acknowledges that his
account of the role of institutions in the 1st edition is far too negative. ‘Certainly it is
a good thing to produce institutions that avoid incentives to violate rights . . . Merely
eliminating evil, however, is far too negative and narrow’ (Shue 1996, p. 159). The
Afterword carries the heading ‘Right-grounded duties and the Institutional Turn’,
and gives a much more positive take on the role of institutions in protecting rights.
As a corollary of the institutional turn, Shue’s analysis rebuts arguments that
human rights talk is necessarily abstract and therefore unable to give direction to
action. If rights require social guarantees against standard threats, a great deal of
social scientific knowledge is needed both to identity standard threats and to deter-
mine how best to provide individuals with institutionalised protection against these
threats (Shue 2004). This point will be elaborated on below.
Shue acknowledges that his analysis is limited in one crucial respect. While it
identifies the categories of duties that are correlative to rights, the theory is silent
on how these duties are to be allocated to particular agents. Shue accepts that the
national state is generally thought to have the prime responsibility for protecting
the rights of those who live within its borders, but argues that this view is based
on a theory of sovereignty, not on a theory of rights per se. Similarly, questions of
who bears responsibility to aid those who have been deprived of their rights is not
determined by a theory of rights and attendants duties, but ‘. . . is largely determined
by theories of moral responsibility which set the proper bounds of self-interest, by
theories of citizenship which draw the appropriate line between compatriots and
strangers, by theories of the state which draw the appropriate line between individual
responsibility and government responsibility, and so forth’ (Shue 1984, p. 91). For
Shue it is only in respect of the duty not to violate that the allocation of correlative
duties is clear, because these duties are universal. Every agent has the ceteris paribus
duty not to violate the rights of others—either directly or through complicity with
the violator(s). It is only where duties not to violate rights clash with each other that
one or other of these duties can be properly put aside.
I think that Shue’s theory has broader implications than he suggests, at least in
respect of protection. As we have seen, he argues (convincingly in my view) that
protection involves the provision of social guarantees against standard threats. It
would seem reasonable to think that social guarantees would best be provided by
some kind of collective social agent, whether this be the individual state, as current
international political arrangements suggest would be logical, or some other collec-
tive agent. And this agent, whatever its precise constitution and reach, would need to
be able to exercise some form of compulsion over its members to be able to provide
the requisite guarantees. I doubt that voluntary compliance with this agent’s dictates,
or voluntary compliance with social norms, would be sufficient, because the possi-
bility of large-scale defection would undermine the security that a social guarantee
is designed to provide. While much of the work of respecting rights might be done
by compliance with social norms, for a social guarantee to exist there would need to
be a coercive mechanism to dissuade defections before they occur and punish them
afterwards. If no such mechanism existed, the state could reasonably be accused of
not doing all it could to protect rights. These considerations suggest that voluntary
organisations, such as NGOs, are not well placed to bear the duty to protect human
rights.
This raises the question whether INGOs, to the extent that they are involved in
promoting human rights, are consigned to the role of aiding those whose rights
have been violated. I will argue below that this does not follow. But first I need to
complete the discussion of Shue’s theory of rights.
3.2 Basic Rights
Shue argues that certain rights are fundamental in a way that others are not. The for-
mer he designates ‘basic rights’. These he defines as those rights whose enjoyment
is ‘. . . essential to the enjoyment of all other rights’ (Shue 1980, p. 19). The right to
physical security is one such right. It is a basic right because:
[n]o one can fully enjoy any right that is supposedly protected by society if somebody can
credibly threaten him or her with murder, rape, beating, etc., when he or she tries to enjoy
the alleged right. . . If any right is to be exercised except at great risk, physical security must
be protected (Shue 1980, p. 21).
Since to enjoy a right is in Shue’s view to be able to exercise it secure from

standard threats, one cannot enjoy any right if one is not secure from the threat of
physical violence. The right to physical security is therefore a basic right. One may
have been able on a particular occasion to protest peacefully against an action of the
government, but if it is a regular occurrence for such demonstrations to be broken up
violently either by government forces or by others without the government taking
preventative action, then one could not be said to have enjoyed the right to peaceful
demonstration. Shue maintains that being secure from physical aggression is not just
a means to the enjoyment of the right to assemble; being secure from aggression is
part of what it means to enjoy assembling as a right. ‘If one cannot safely assemble,
one is not free to assemble’ (Shue 1980, p. 26).
The right to subsistence (the right to a level of consumption ‘. . . needed for a
decent chance at a reasonably healthy and active life of more or less normal length,
barring tragic interventions’ (Shue 1980, p. 23)) is similarly, according to Shue, a
basic right. He writes:
46 K. Donaghue
No one can fully, if at all, enjoy any right that is supposedly protected by society if he or
she lacks the essentials for a reasonably healthy and active life. Deficiencies in the means of
subsistence can be just as fatal, incapacitating, or painful as violations of physical security.
The resulting damage or death can at least as decisively prevent the enjoyment of any right
as can the effects of security violations (Shue 1980, p. 24).
But what if someone has the essentials of a healthy life without enjoying these
as a right? Would such a person be able to enjoy the right to (say) free assembly,
assuming that they enjoyed the right to physical security? Shue would say ‘no’,
because this person would be open to intimidation through a threat to withdraw the
essentials of life, ‘. . . and credible threats can paralyse a person and prevent the
exercise of any other right as surely as actual beatings and actual protein/calorie
deficiencies can’ (Shue 1980, p. 26). To intimidate, such threats would have to be
credible, and it is possible to imagine circumstances in which powerful individuals
or groups might be immune to such threats. But these would be exceptions.2 For
the vast majority of individuals and groups, as Shue writes, ‘[c]redible threats can
be reduced only by the actual establishment of social arrangements that will bring
assistance to those confronted by forces that they themselves cannot handle’ (Shue
1980, p. 26). These would include social arrangements to ensure that subsistence is
guaranteed, thereby making the right to subsistence a basic right.
Some liberties are also, Shue argues, basic rights. Freedom to participate effec-
tively in influencing fundamental choices among social institutions and policies and
in influencing the way in which institutions operate and policies are implemented,
is an example. To the response that many people may not be interested in such par-
ticipation, Shue replies that this does not argue against such participation being a
basic right. To the response that a paternalistic dictatorship could ensure that rights
to security and subsistence are enjoyed, and that participation cannot therefore be
a basic right, he stresses the demand element that a right contains (a demand the
fulfilment of which should be socially guaranteed). If one is not able to make this
demand known to those who bear the duty to ensure that its enjoyment is socially
guaranteed, then one is not in a position to exercise the relevant right. Without the
freedom to participate, therefore, while one might enjoy security and subsistence,
one could not enjoy these as rights. If you are not able to forcefully demand that the
duty to protect be met, then you are at the discretion of others’ good will. ‘And to
enjoy something only at the discretion of someone else, especially someone pow-
erful enough to deprive you of it at will [e.g. the state], is precisely not to enjoy a
right to it’ (Shue 1980, p. 78).
Using a similar line of argument Shue shows that freedom of physical movement
should also be designated a basic right. But he doesn’t claim that the list he provides
is exhaustive. What he is particularly keen to show is that some rights from each of
the CP and ESC camps can reasonably be seen as basic in his sense.
3.3 Basic Rights, the Structure of a Moral Right

and the Complexity of Duties
Shue ties his account of basic rights to his analysis of the general structure of a moral
right by claiming that the violation of basic rights represents one of the standard
threats to all rights. He states: ‘The fulfilment of a basic right is a successful defence
against a standard threat to rights generally. This is precisely why basic rights are
basic. . . If the substance of a basic right is not socially guaranteed, attempts actu-
ally to enjoy the substance of other rights remain open to a standard threat like the
deprivation of security or subsistence’ (Shue 1980, p. 34).
This foreshadows a point that Shue takes up in greater detail in the Afterword
to the second edition of Basic Rights, where he discusses the claim by authors such
as Jeremy Waldron and Thomas Pogge that there need not be a simple one-to-one
relationship between a given right and a particular duty or set of duties. Waldron
talks of ‘successive waves of duties’ which may well be needed if a right is to be
properly protected (Waldron 1993, cited in Shue 1996, p. 56). Pogge distinguishes
between an interactional and an institutional conception of rights, with the former
matching duties to rights in a direct manner, while the latter argues that actions not
directly related to a given right may be necessary to ensure its enjoyment (Pogge
1995, cited in Shue 1996, p. 156). Shue supports Pogge’s ‘indirection’ thesis, which
he sees as fully compatible with his own views on the interdependence of rights, the
complex nature of duties, and the consequent need for social scientific (means/ends)
and philosophical (fairness in allocation of duties) thinking about how best to meet
the duties correlative to rights. He writes:
One cannot stay at the purely conceptual level, reasoning simply that if there is a right to
have x, there must be a duty for others to provide x. There may instead be a duty to stay out
of people’s way while they take x for themselves, or a duty to teach them to read so they can
figure out how to make or grow x, or a duty to let them form a political party so that they
can effectively demand that the government stop exporting x (instead of having the CIA arm
their police so that they can suppress all dissent). Sometimes there is a duty to provide x to
them, conditionally or unconditionally. . . Often the means to the fulfilment of one right will
include the fulfilment of other rights, because rights may be of great instrumental value in
the fulfilment of other rights regardless of whether they are of intrinsic value as well. (Shue
1996, p. 164)
4 Basic Rights, Correlative Duties and the Moral Priorities

of INGOs
I believe that Shue’s discussion of the nature of a moral right and of basic rights in
particular, including his typology of duties correlative to rights,3 provides a useful
set of tools to think about the work of INGOs and analyse their priorities. I propose
to begin with self-styled human rights organisations, where the relevance of Shue’s
work should be relatively easy to demonstrate, before moving on to development
INGOs.
4.1 Human Rights INGOs: Addressing Economic as Well as Civil

and Political Rights
INGOs such as Amnesty International (AI) and Human Rights Watch (HRW) focus
on duties of type II, i.e. they work to improve the protection of rights-holders from
48 K. Donaghue
the violation of their rights. Typically their priority is the first of Shue’s two sub-
categories under (II), namely pressuring powerful duty holders not to violate rights
(duty a). Commonly they rely heavily on the methodology of naming and sham-
ing those agents which violate rights (often governments violating the rights of
their own citizens) to achieve this end. But not all human rights INGOs adopt this
methodology. Some (the Danish Institute for Human Rights is an example) work
with governments with dubious human rights records to strengthen the institutions
responsible for ensuring that duty (I) is not violated (Bell and Coicaud 2007). This
work is an instance of duty II(b), building institutions that guard against the viola-
tion of the relevant right(s). Questions of methodology aside, in Shue’s terms both
types of cases involve the identification of governments as a major standard threat
to enjoyment of rights. However, this is not always the case. Some NGOs lobby
private sector organisations such as mining or sportswear companies with the aim
of improving their human rights practices.
Typically human rights organisations started out with a focus on civil and politi-
cal rights, or a sub-set thereof (Goering 2007). Over time mandates have broadened
to include a wider range of civil and political rights, and in some cases to include
certain economic, social and/or cultural rights. Pressure for these changes has come
from a variety of sources: from within the staff and/or support-base of the INGOs
themselves, from partner organisations particularly in developing countries and
to some extent from those whose rights the organisations are pledged to protect.
Typical arguments have focused on the interdependence of rights and the resultant
artificiality of giving all one’s attention to some while leaving others aside (Goering
2007).
These changes in mandates have been inevitably accompanied by substantial
debates among and between board members, staff and supporters of the organ-
isations concerned. These debates have usually involved an acceptance that the
scope of the rights on which the organisations focus should be enlarged, in recog-
nition of an arbitrariness associated with the current focus. But this acceptance has
been tempered by a concern that expanding the scope will dilute the organisation’s
effectiveness.
The experience of Human Rights Watch is worth considering in this context
(Roth 2007a). Its founder was a strong advocate of the primacy of CP rights. On his
departure the organisation signalled a preparedness to consider widening its man-
date, and this has progressively happened over the last decade or so. However, HRW
has not been prepared to change its naming and shaming methodology—indeed, the
remarks of Executive Director Kenneth Roth give no suggestion that the possibility
that the methodology might change has been seriously considered. In HRW’s view,
to be effective this methodology requires that there be a clearly identifiable rights
violation, that the violator also be clearly identifiable, and that a remedy for the vio-
lation be apparent and that the steps needed for this remedy be clear and achievable.
If these circumstances obtain, HRW is prepared to take up the case, organisational
resources permitting. The conditions underlying the naming and shaming method-
ology fit very well with Shue’s account of the structure of a moral right. A situation
where a violation, a violator and a remedy are clearly identifiable meets the notion
of a standard threat, protection against which is a crucial duty correlative to a moral

right.
HRW insists that the methodology does not restrict it to working only on CP
rights. But the methodology does require that the actions to be protested are clearly
illegal, discriminatory or arbitrary. Where actions at issue concern differences over
the allocation of resources where reasonable arguments can be mounted for and
against a particular line of action, Roth argues that the naming and shaming method-
ology cannot be successfully employed. To the extent that a country’s economic
policy displays features of this latter kind, it is not a matter for HRW.
HRW’s position has been criticised on the grounds that, in focusing on particular
rights (those that can be effectively defended by means of the naming and shaming
methodology) it inevitably downgrades those rights it does not address. Given that
contentious resource allocation issues are often at play, or at least often believed to
be at play, when it comes to rights to adequate nutrition, health care and education,
this means, so the criticism, that HRW effectively disbars itself from defending these
rights. In practice if not in intention, therefore, HRW perpetuates the Western liberal
view that CP rights should take precedence over ECS rights (Chandhoke 2007).
HRW’s response to the criticism is to state that the naming and shaming method-
ology is an effective instrument and that effectiveness is crucial if HRW is to
maintain its moral capital (Roth 2007b). But effectiveness itself is clearly not suf-
ficient, for if it were then effective action to rectify relatively trivial wrongs could
be seen as having high priority. But HRW would argue that the rights violations it
addresses are of great moral significance, and that it is fully justified when choosing
which rights it will defend to rely upon a particular methodology that it believes
has proved its effectiveness. Undoubtedly at play here is the view that one human
rights organisation cannot be expected to cover the whole field, that specialisation is
inevitable and, within reason, likely to promote effectiveness, and that not engaging
with certain aspects of the human rights agenda need not carry the implication that
these aspects are of secondary importance.
4.2 Priorities Among Rights and Rights-Based Duties
The previous discussion suggests that INGOs need to consider at least two factors
when determining priorities:4 (i) the moral weight of the issues they are addressing;
and (ii) the extent to which they can act cost-effectively (achieving desired results
at low relative cost).
In terms of the issue of moral weight, I think Shue’s analysis helps on two fronts.
First let us take the three-fold typology of duties correlative to human rights. I
would suggest that, other things being equal, the ideal situation would be one in
which the first duty (not to deprive others of their rights) is universally honoured,
since this would mean few if any rights are violated (the only violations would be
those, if indeed there are such, solely attributable to natural, i.e. non-social, factors).
However, it is unrealistic to expect or to rely on this. And even if it obtained, it
50 K. Donaghue
would not be sufficient to ensure the enjoyment of rights, given that it is essential to
enjoying something as a right, rather than simply happening to enjoy it, that social
guarantees against standard threats to the enjoyment of the substance of the right are
in place. This suggests that the duty to protect has at least the same moral weight as
the duty not to deprive (Shue 1980).
Shue himself, at least in the first edition of Basic Rights, does not appear to see
any moral priority among the duties correlative to rights, arguing throughout his
book that all three duties are necessary if rights are to be enjoyed. He seems to
conceive of the duty to aid those deprived of their rights as in effect providing a
social guarantee against deprivation. An initial deprivation is made good through
aid, which thereby abolishes the deprivation. But if aid is guaranteed, then it seems
to me to be best thought of as part of the duty to protect—a second order protection,
as it were, to be called on when first-order protective measures fail. I think it is more
natural to talk about the duty to aid only when a right has truly been violated. Aid in
this sense is by definition a stop-gap measure. If it were not, it would be an aspect
of protection.5
If aid is thought of in the way I am suggesting, then a priority ordering between
types II and III duties does suggest itself, at least in the sense that a world in which
rights are effectively protected would seem to be morally preferable to one in which
violations occur and the victims are subsequently provided with aid to redress (to
the extent possible) the effects of the violations.
It is worth noting that, in the practical part of Basic Rights, duties to aid receive
the bulk of Shue’s attention. But in the Afterword to the second edition he labels
the focus on duties to aid ‘unfortunate’ ‘. . . above all because it obscured the impor-
tance of the second kind of duty, the duty to protect’ (Shue 1996, p. 159). Later in
the Afterword Shue writes: ‘. . . for rights bearers, earlier is of course always better.
Every devolution to a later duty leaves violated rights, and violated people, in its
wake. Prevention is always better than cure, even when there is something resem-
bling a cure. . . It is not permissible to substitute fulfilment of duties to aid that came
into play only because victims went unprotected, for fulfilment of back-up duties to
protect’ (Shue 1996, pp. 173–174). This suggests that a priority ordering between
duties of types II and III that was absent from the first edition of Basic Rights is
present in the second.
But there are a number of factors which complicate any move from the sup-
posed moral preferability of one state of affairs to another to moral claims about
how agents should act with respect to these states of affairs. The first complication
is urgency. As Shue points out, when a right, particularly a basic right, has been
violated there is often an urgent need to provide the victim with the substance of the
right if permanent damage is not to result. He states that this greater urgency does
not mean that the duty to aid is ‘more compelling’ than the other two categories of
duty (Shue 1980, p. 62). But clearly the question of urgency must be addressed by
anyone who wanted to argue that the duty to aid is less compelling than the other
two types of duty.
Related to urgency is the issue of the extent and nature of need. Those whose
basic rights have been violated are reasonably seen as among the worst off of
all human beings. Not only are their basic needs unmet, but they are unmet not
because of the exigencies of nature (although these might play a part) but because
of shortcomings of human duty holders. Unmet need is thereby compounded by a
fundamental lack of respect. For this reason the moral demands placed upon us by
those whose basic rights have been violated might appear to be more onerous than
those coming from any other source. Aiding the deprived might therefore be seen as
of higher moral priority than any other category of action.
An initial response might be that, to the extent that there are people whose basic
rights have been violated, aid is clearly necessary to make good the violations to the
extent possible. But it remains the case that it would be morally better if these viola-
tions did not take place, and that therefore protective action to reduce the likelihood
of violations, if it has a reasonable chance of success, must not be overwhelmed by
the need to aid the deprived.
A further complication in the move from a desired state of affairs to implications
for action is that particular agents may better placed to fulfil type III rather than type
II duties. Their expertise might lie in type III work. In this situation, the morally
correct approach would surely be to focus on type III duties. There is clearly a
trade-off at play between the moral importance of particular states of affairs and the
capacity of different agents to perform different kinds of actions cost-effectively.6
Indeed, it might be concluded from the point made above about the inability of
organisations based in voluntary action to provide the social guarantees needed for
the protection of rights that these organisations should concentrate on aiding those
who rights have been violated. I will argue below that this is not necessarily the case.
A related objection is this. Surely it is the case that, despite the best endeavours
of those who would protect human rights, some people will have their rights vio-
lated. Therefore there need to be organisations, probably specialist organisations,
which are able to provide effective aid in such circumstances. What is needed is a
division of labour between organisations addressing the full range of duties correl-
ative to basic rights—relief as well as development agencies, agencies focusing on
(particular) CP rights and those focusing on (particular) ECS rights. If so, even if a
state of affairs of effective rights protection is morally preferable to one of violations
plus compensation, it may make little sense to say that one set of actions (protection
against violations) is morally preferable to another (helping the violated).
I would respond that, for agents who have the power to protect rights-holders
against violations, the preferability of a state of affairs of rights protection to one of
violations plus aid means that there is a strong presumption that these agents will
act protectively in respect of human rights. Certainly there could still be situations
where such agents might, justifiably from a moral point of view, decide to help
victims rather than protect the vulnerable (for example where interventions designed
to protect might be far more costly than those designed to aid, as might be the
case where one state intervenes in the affairs of another), but the moral priority of
protection means that the agent would face a burden of proof to justify the decision
to aid rather than to protect. The greater the extent to which help to victims would
be likely to fall short of fully compensating for violations, the heavier the burden of
proof.
52 K. Donaghue
The situation is far less clear for those agents—and one might think that
INGOs fit here—whose power to protect rights holders is uncertain. Urgency and
extent/nature of need considerations seem to me to justify relief agencies focus-
ing on aiding the deprived in situations where significant rights violations have
occurred. However, for development INGOs urgency and related considerations
have much less salience, since these organisations are explicitly concerned to
address the underlying causes of poverty rather than to ameliorate its effects. As
a result it is reasonable to think that the moral preferability of a state of affairs of
rights protection over one of violation plus aid has moral implications for the actions
of development INGOs. It could still be the case that cost-effectiveness considera-
tions mean that development INGOs might be justified in aiding those whose rights
have been violated, for example by providing them with services (e.g. education
and health) that should be guaranteed by the state. But the burden of proof would be
on development INGOs to show why they are in the service-delivery business and
not in the business of human rights protection. The greater the ability of develop-
ment INGOs to play an effective rights-protective role, and the greater the extent to
which service delivery by INGOs would reduce the likelihood that the social guar-
antees necessary to the enjoyment of rights will be forthcoming, the greater this
burden of proof. I will return to this issue below.
Shue’s analysis is useful on a second front, namely his discussion of basic rights.
Basic rights establish a moral minimum: they define ‘. . . everyone’s minimum rea-
sonable demands upon the rest of humanity’ (Shue 1980, p. 19). Questions about
the nature of the good life for human beings, about the ‘moral heights’ to which
human beings might aspire, are not disavowed by Shue but are simply not relevant
to his task of tracing the moral minimum. To the extent that INGOs see themselves
as engaged primarily in helping to shield the vulnerable from at least some of the
more common and more devastating of life’s threats, then Shue’s basic rights pro-
vide a clear delineation of the territory on which they should focus. Together with
my argument about the priority of type II duties, this leads to a first approximation of
moral priorities for development INGOs: focus on basic rights (i.e. those rights that
are fundamental to the enjoyment of other rights), and focus on Shue’s duties I and
II (to avoid violation and to protect from violation). Let us now see how this first
approximation might assist development INGOs to think about how they allocate
resources.
4.3 Development INGOs, Duties and Moral Priorities

To be able to successfully protect from deprivation, development INGOs need to
identify where the principal threats to the enjoyment of economic and social rights7
stem from and what form they take. As suggested by Shue, these threats may take the
form of potential violations of civil and/or political rights. Only when this work of
identification is done can INGOs determine how best to combat these threats. I sug-
gest that, in requiring those who would protect economic and social rights to identify
standard threats to their enjoyment, the human rights approach helps to bring into
focus important resource-allocation questions facing development INGOs.
4.3.1 Protecting Against Violation of Economic and Social Rights:

Implications for Development INGOs
Where threats to the enjoyment of rights issue from a single and readily identifiable
source (paradigmatically a government), and where they involve acts or omissions
they are clearly illegal or in other ways discriminatory or arbitrary, they provide a
potential target for an INGO. As discussed above, this is the situation where agen-
cies such as Human Rights Watch are most at home. A question for development
INGOs is: how relevant to the objectives of development work are threats of this
kind?
Kenneth Roth of HRW argues that there is a wide range of issues pertaining to
various economic and social rights where what is at issue is primarily a matter of
policy rather than resources, and which for this reason lend themselves to protective
action based on a naming and shaming methodology. He cites a number of recent
HRW reports dealing with issues such as ‘. . . South Africa’s ideological (and hence
arbitrary) refusal to provide postexposure prophylactic drugs for victims of sexual
violence’ and ‘. . . Uganda’s inadequate measures to address sexual and domestic
violence, with resulting exposure of the victims to HIV’ (violations of the right to
health); ‘. . . Indonesia’s discriminatory refusal to apply its labour code to domestic
workers’ (violation of basic labour rights); and ‘. . . gender discrimination in inher-
itance laws in Kenya’ (violation of among other things the right to housing) (Roth
2007a, pp. 178–179).
To such cases might be added that of corrupt or incompetent local elites, who
either steal or mismanage the resources needed to abolish poverty. HRW’s naming
and shaming methodology may be effective against such threats to the enjoyment
of rights. INGOs could also assist by building the capacity of local organisations to
identify corruption and mismanagement and mobilise local action to hold corrupt
and/or incompetent local elites to account.8 There is of course nothing to guarantee
that such measures will be effective. But given the real possibility that the actions
of elites (primarily governments) constitute a (if not the) major standard threat to
the enjoyment of (at least some) economic and social rights, development agencies
should examine carefully the extent to which this is the case in a given situation and
what counter actions they might best take.
A development INGO might reply that, just as HRW is committed to its nam-
ing and shaming methodology and allows this methodology to set limits to the
work it does, so it has the development project as its main tool, and the nature of
the development project sets limits on the sorts of activities a development INGO
can usefully pursue. I suggest two responses to this. One is that the development
project is a multi-purpose tool which can be used among other things to build the
capacity of development partners to hold rights-violators to account. Another is that
many development INGOs do not in fact limit themselves to development projects,
54 K. Donaghue
but spend time and money in policy and advocacy work designed to change the
behaviour of agents which are in a position to violate and/or protect human rights.
The implication is that development INGOs should look carefully at the extent to
which the development issues they address relate in significant part to arbitrary,
discriminatory and/or incompetent actions of national and local governments or
other agents. I recognise that some INGOs are increasingly doing this (Brouwer
et al. 2005; Oxfam International 2006b), but a human rights approach, drawing on
my take on Shue’s classification of duties, provides a clear rationale for why such
an approach should be given high priority by all development INGOs—it focuses
very clearly on protecting rights holders against violations, rather than waiting for
violations to happen and then providing aid to those affected.
The issue of cost effectiveness is central to this as to any other approach. If the
approach is not cost effective, it is of little moment that it addresses issues of high
moral weight. But the fact that an approach addresses issues of high moral weight
means that its cost effectiveness should be very carefully assessed. I can’t offer
any answers in this paper on the cost effectiveness of naming and shaming and
related approaches. But hopefully what I have said suggests that the cost effective-
ness of these approaches warrants careful analysis by development INGOs (Nelson
and Dorsey 2003).
4.3.2 Protecting Against International Threats

In identifying the major standard threats to the enjoyment of basic rights, there is
no good reason to stop at the actions of national and sub-national agents. Could
it not be the case that standard threats to the economic and social rights of those
living in developing countries include arbitrary, discriminatory and/or incompetent
actions of rich country governments, their agents (principally inter-governmental
organisations) or those headquartered in their jurisdictions (multinational corpora-
tions)? If this is so,9 then action to remove or at least to minimise these threats,
if it were cost effective, would be of great moral importance. Action at this level
would be of particular relevance to INGOs, since they may well be (directly
or indirectly) complicit through their support for their home governments and/or
their home governments’ support for them—developed country governments are,
after all, significant financiers of INGOs.10 Moreover, removing externally-imposed
impediments has the advantage of not requiring any form of direct help for the
intended beneficiaries, thereby by-passing the significant difficulties associated with
providing help in a way that is compatible with the autonomy of those being helped
(Ellermann 2001), as well as avoiding issues associated with interfering with the
political processes of another country.
These considerations provide development INGOs with very strong reasons for
examining whether the economic and social rights of the poor in developing coun-
tries are threatened by the unjustifiable actions of developed country agents, and
to what extent they (INGOs) are in a position to influence such actions. Certainly
some INGOs are engaged in work to change the behaviour of agents within devel-
oped countries in areas such as the arms trade (Control Arms 2003), debt (Jubilee
Australia, May 2007), trade rules (Make Trade Fair, May 2007) and access to
medicines (MSF, May 2007), although I suspect that this work claims only a rel-
atively small share of total INGO resources. The moral priority of protecting human
rights as against helping the deprived, and the particular moral weight attached to
avoiding complicity in causing harm, imply that INGOs need to put significant effort
into assessing the soundness of arguments of writers such as Pogge concerning the
harm done by the developed world to the developing, and to their own capacity to
mitigate this harm.
Most development INGOs claim that they seek a relationship of equality and
cooperation between themselves and their donors on the one hand and their con-
stituencies in developing countries on the other. I suggest that this goal behoves
INGOs to look carefully at any responsibility of their home governments for the
problems they are seeking to address in the developing world. The more that devel-
opment INGOs operate on the assumption that the key problems they confront are
caused solely or primarily by factors endogenous to developing countries, or that,
where the actions of external agents are at play, compensating for these actions is all
that can be expected of INGOs, the more difficult it is likely to be to achieve such a
relationship.
Admittedly this raises potential difficulties, in that any increase in the level of
criticism of their home governments by INGOs might impact negatively on the level
of funds made available by these governments for INGO programs. Similarly, devel-
opment INGOs which directly criticise developing country governments or (more
realistically) support their developing country partners to do so might find their
freedom to work in the countries concerned severely constrained. Development
INGOs might use these considerations to argue for maintaining the division of
labour between themselves and human rights INGOs. Leave human rights INGOs
to get involved in politics, to name and shame governments or in other ways hold
them to account, while development INGOs get on with the job of development,
with concrete actions that have tangible benefits for poor people. The following sec-
tion addresses the conception of development at play here, and whether a division
of labour of this kind is sensible.
4.3.3 Is Lack of Development a Standard Threat to the Enjoyment of Rights?

The previous two sections have identified potential threats against which social
protection is needed if basic rights are to be enjoyed, namely the arbitrary, dis-
criminatory and/or incompetent actions of local, national and international agents,
primarily but not exclusively governments.
But for many of the proponents of development assistance it is the lack of
resources resulting from under-development, not (or at least not primarily) the arbi-
trary, discriminatory or incompetent actions of specific agents, that is the major
cause of global poverty, and hence of the violation of economic rights—although
most would now add that the weakness of developing country governments is one
expression of the lack of development, creating a vicious circle of low levels of
development and weak governments reinforcing each other (hence the recent focus
of development theorists and practitioners on good governance).
56 K. Donaghue
It is an empirical matter what the major standard threats to the enjoyment of

economic rights are in specific situations. The point of the discussion in the previ-
ous two sections is to suggest that actions by local, national or international actors
may in fact represent the major standard threats, and that it is therefore incumbent
upon development INGOs, not constantly but at reasonable intervals, to review the
evidence about this. To do so, they of course need to compare these threats with
those represented by a lack of development, in the sense of a lack of resources
(including the lack of institutional capacity to put available resources to effective
use). In so doing not only the impact of the threats on rights-holders needs to be
considered, but also the extent to which the various threats are remediable (recall
that being remediable is part of what it means for a threat to be ‘standard’). While
arbitrary, discriminatory or incompetent actors may be intransigent and resistant to
naming and shaming or other methods that might be brought to bear to change their
behaviour, it is at least reasonably clear what sorts of steps stand the most chance of
bringing about the desired behavioural change, and what the indicators of success
would be.
By contrast, the economic development of a country is exceedingly complex.
There is no ‘one key thing’ that, if addressed, will solve the development problem
(Easterly 2002). Rather, poor countries are poor because of a deep-seated set of
interrelated features, some of which are relatively easy to identify, some of which
are opaque, and where the causal relationships between the various facets of under-
development are often difficult to discern. These societies have for a range of reasons
not undergone the social transformations that enabled rich nations to increase rad-
ically the productivity of their labour which allowed them to create the wealth that
now underpins their extensive (although by no means total) guarantees of the eco-
nomic rights of their citizens.11 Whatever the exact diagnosis of the reasons for
underdevelopment in a given case, the forces at play are likely to resist manipulation,
particularly by outsiders (Collier 2007).
4.3.4 Development Assistance and Rights Protection

The complexity of development illustrates the difficulties facing development agen-
cies wanting to protect rights-holders from deprivation of their economic and social
rights through support for the development process. The challenge of development
as described stands in sharp contrast to that in which Human Rights Watch for exam-
ple would feel able to act, or indeed to any situation where the shortcomings of
identifiable agents are the major causes of rights violations. Where lack of devel-
opment due to deep-seated social and/or geographical factors is at play (certainly
where institutions are included as well as government policies) there is no clear
act of rights violation, no obvious violator, and no clear remedy. Protecting from
deprivation in such a situation would seem to require working for change along the
dimensions of policy, institutions and culture. This work is extremely complex, both
in terms of knowing what to do and how to determine if one is heading in the right
direction (Dichter 2003).
These considerations are similar to those adduced by those who would deny that
ECS rights should enjoy the same status as CP rights. Those adopting this position
would argue that the lack of development is not a standard threat to the enjoyment
of economic rights because, while it is in some sense remediable, the remedy does
not lie in the hands of any putative duty-holders.
These considerations suggest that development INGOs are limited in the extent
to which they can play a rights-protecting role. Their direct influence on the policies
of governments in the countries in which they work is likely to be small, as is their
capacity to play a significant role in institution building or cultural change. Does it
therefore follow that development assistance is a questionable strategy to protect the
vulnerable from violations of their economic rights? Is aiding the deprived the best
that development INGOs can do?
Although a number of INGOs now work across a range of modalities, the stan-
dard methodology of the development INGO remains the development project.
Development projects take various forms, but a distinction is usually made between
those which (i) focus on capacity building of partner organisations, groups or indi-
viduals in order to increase the ability of these agents to act on their own behalf; and
(ii) those which focus on the delivery of goods and services to end users. Individual
projects can, and often do, include both aspects.
I would argue that service-delivery projects are for the most part best thought of
as a response to the third category of Shue’s duties—they represent a form of helping
those whose rights have been violated rather than protecting rights-holders from
violation of their rights. In this sense they do not differ fundamentally from relief
work. Where successful they provide those deprived of their economic rights with
the substance of those rights or with the means to increase their chances of enjoying
the substance of these rights. But projects of this kind, I would suggest, often address
only marginally if at all the social guarantee aspect of rights highlighted by Shue.
A recent evaluation of Oxfam, the development INGO probably most committed
to a human rights approach, questioned whether ‘gap-filling’ (the evaluators’ term
for service delivery) is a legitimate function for an NGO which claims a rights-based
approach (Oxfam 2006a). Oxfam’s response acknowledged the need for greater clar-
ity in its approach to human rights, which in respect of service-delivery meant that
Oxfam ‘. . . will not be “gap-filling” although it may directly support the provision
of health and education services but only within a longer term agenda of engag-
ing duty bearers in taking up their responsibilities over time’ (Oxfam International
2006b, p. 2).
Projects which aid the deprived can undoubtedly be of great value—but they
also contain dangers to the broader agenda of rights protection. One such dan-
ger is exemplified by Bangladesh, where a substantial proportion of education and
health services are financed and provided by NGOs such as the Bangladesh Rural
Advancement Committee (BRAC). BRAC and similar agencies may well do a
good job of providing these services. However, their voluntary nature, their limited
accountability to those they serve, along with uncertainties about their funding and
their uneven coverage, means that the services they provide fall well short of con-
stituting a social guarantee. At the same time their service-provision work reduces
58 K. Donaghue
the pressure on the Bangladesh government to finance and/or provide such services
in a way that would constitute such a guarantee. The net result is that citizens have
no effective way of demanding that these services be provided as a matter of right
(Edwards 1999; Nelson and Dorsey 2003).
It is through their capacity-building work that INGOs would seem to make their
best claim to be addressing the protection of rights. While not particularly well
placed to support the creation of the institutional means for protection (supply-side
issues), INGOs do have a role in enhancing the self-protective capacity of individ-
uals and groups by (i) helping individuals/groups to strengthen their demands that
channels through which their views can be brought to the attention of (in particu-
lar) governments are created and maintained; and (ii) helping individuals/groups to
make effective use of such channels where they exist. This kind of helping spans the
divide between Shue’s duties of type II and type III. It is helping with a protective
dimension, based on the recognition that individuals can do much to protect them-
selves. This fits well with Shue’s stress on the right to representation as a basic right.
It might be thought that introducing the notion of self-protection must weaken
the emphasis that Shue’s account of a moral right gives to social guarantees. If
individuals can do much to protect their rights themselves, it would seem that social
guarantees are not always required. Furthermore, it would seem that an emphasis
on self-protection would sidestep many of the problems often associated with social
guarantees, in particular the tendency of such guarantees to weaken the incentives
for individuals to help themselves—leading to the sorts of welfare traps that, so
advocates such as Noel Pearson argue, have done much to weaken many indigenous
communities across Australia (Pearson 2006).
In the ‘Afterword’ to the second edition of Basic Rights, Shue shows himself to be
alert to these issues. He writes that ‘[a]rrangements for fulfilling rights should never
be arrangements to do for people what they can equally well do for themselves’
(Shue 1996, p. 167). Institutions that embody the social guarantees required for the
enjoyment of rights are, Shue suggests, not only institutions that ensure that the
substance of their rights can be enjoyed by those unable to help themselves, but
institutions that afford individuals effective channels whereby they can acquire the
substance of their basic rights through their own actions.
A social guarantee might therefore be characterised as a set of arrangements that
(i) provides a favourable context for individuals to act to secure the substance of their
basic rights (this includes removing institutional impediments to effective action and
providing institutional means for rights-holders to demand that relevant social actors
meet their rights-based duties); while (ii) ensuring that those unable to act effectively
on their own behalf are provided with the substance of their rights. Achieving a
workable balance between these two aspects of the institutional provision of social
guarantees is a major challenge facing welfare states the world over.
The social guarantees that are essential for something to be enjoyed as a right are
not all or nothing affairs. The policies and institutions which embody these guaran-
tees are built up over time and need on-going maintenance and improvement. The
supply of such guarantees is in significant part a function of the demand, and in
exercising this demand individuals and groups can contribute to the protection of
their rights.
Where does this leave development INGOs wishing to take a human rights
approach to their work? They need to be alert to policy or institutional impediments
to effective action on their own behalf by those with whom they work. A crucial
dimension of their work is to support their partners to identify and act to remove
these impediments. Advocacy in their own name, and (probably more importantly)
capacity building to enable their constituents to advocate effectively on their own
behalf, are the approaches most likely to achieve this objective. This will be the case
where the impediments are government policies, for example, or other factors that
are external to and opposed by those with whom the INGO is working. However,
where the impediments include institutionalised features of the groups with which
the INGO is working, for example inequitable and disempowering gender roles and
responsibilities, things are much more complex. I’m not sure that much useful of a
general nature can be said about these situations. Highly contextualised knowledge
about who the agents of change might be, how effective they are likely to be, and
how best they might be supported, is what is needed here.
If the policy and institutional environment provides sufficient space for rights-
holders to gain reasonably secure access to the substance of their basic rights
through their own efforts, there may be a case for development INGOs to focus
on enhancing the capacity of their constituents to make the best use of the oppor-
tunities which the environment offers. This could include, for example, helping to
increase the productivity of their labour through the provision of training or increas-
ing their economic security by improving access to financial services. But this
service-provision work, to take on a rights-protective character in Shue’s sense, must
build the self-protective capacity of those who receive the services. To the extent
that it provides goods and services without increasing the capacity of the recipients
to gain access through their own actions to the substance of their basic rights, this
work does not play a protective role. And to the extent that it deflects attention from
shortcomings in the policy and institutional environment that are both severe in their
impacts and amenable to change through advocacy and related means, it makes key
protective measures less likely.
At issue are matters about particular features of the world and how they might be
changed. These are matters of (often complex) fact that development INGOs need
to come to grips with. What are the standard threats to the enjoyment by the poor of
their basic rights? To what extent do the arbitrary, discriminatory and/or incompe-
tent actions of local, national or international actors constitute such threats, and to
what extent are these actions susceptible to countervailing measures by development
INGOs and by those on whose behalf they act? To what extent does the provision of
services by INGOs enhance the capacity of the recipients of these services to protect
their basic rights (rather than, for example, reducing the motivation to act effectively
on their own behalf), and to what extent does the provision of such services absolve
the primary duty holder (often the national state) from the responsibility for service
provision that meets the social guarantee requirement of a basic right?
60 K. Donaghue
5 Concluding Remarks
I have argued, on the basis of Shue’s typology, that the notion of basic rights
and attendant duties, and the centrality to rights of social protection against stan-
dard threats, can assist development INGOs in thinking about how to prioritise
their work. I have suggested that a world in which rights-holders have their rights
effectively protected is morally preferable to one in which violations occur but com-
pensation is forthcoming. I acknowledge there are complications when it comes
to setting out the implications of this preferability for different agents. For exam-
ple, the urgency often associated with helping the deprived, and the need for a
division of labour between agencies working on behalf of the poor, show that help-
ing those deprived of their basic rights remains crucially important from a moral
point of view. But those agents with the power to provide effective rights protec-
tion (I argue that this includes development INGOs, through their ability to thwart
would-be rights violators and their capacity to support the self-protective actions
of vulnerable rights-holders) must afford this protection very high priority, with
the phrase ‘very high priority’ implying that these agents are under a burden of
proof to justify in detail any actions that are not protective of rights. A commit-
ment to protection requires in turn a careful analysis of the major standard threats
to basic rights and the ways in which these threats might best be removed or at least
minimised.
There is much evidence to suggest that standard threats to the basic rights of the
poor in developing countries consist to a significant degree in the arbitrary, discrim-
inatory and/or incompetent actions of agents, particularly states, at local, national
and international levels. Development INGOs would do well to assess this evi-
dence carefully, and in addition to assess the extent to which they, their supporters,
their developing country partners and constituents can act cost-effectively to counter
these threats. The value of greater cooperation with traditional human rights INGOs
should be further explored in this context. Traditional development projects have a
role to play in protecting the basic rights of the poor, but these projects face impor-
tant constraining conditions: (i) they need to focus on building the capacity of the
poor to act self-protectively, in so doing avoiding the creation of disincentives for
the recipients to act in defence of their own rights where effective action is possi-
ble; and (ii) they should to the extent possible avoid making it easier for key duty
holders, principally national states, to abrogate their responsibility to provide secure
access to services that form the substance of basic rights, where individuals cannot
secure these rights effectively through their own actions.
Notes
1. An example is the Control Arms campaign, which involves collaboration between
Amnesty International, Oxfam and the International Action Network on Small Arms. See
http://www.controlarms.org/index.htm.
2. Nonetheless, the possibility that an agent could enjoy other rights without strictly speaking
enjoying the right to subsistence suggests that Shue’s position on basic rights needs to be
modified. Rather than basic rights being necessary to the enjoyment of other rights, they
might better be designated as those rights that are strongly supportive of other rights. James
W. Nickel (2007, pp. 89–90) suggests this modification.
3. In using the term ‘correlative duties’, I include the possibility that duties associated with a
particular right might involve ensuring that other rights are enjoyed.
4. I am here assuming that it is for the board of an INGO to determine its goals and modus
operandi and to seek resources from funders on this basis. I leave aside the difficult questions
that arise concerning how an organisation should respond if it determines that more funds
could be raised for an alternative (in its judgment morally inferior) agenda.
5. I am here referring to aid as used in Shue’s typology of duties. Aid as understood in everyday
language might address the duty to protect, for example by enhancing the capacity of duty
holders to meet their responsibilities, or by building the capacity of individuals to protect their
own rights. More on this below.
6. This is somewhat analogous to the point made by deontological moral philosophers that there
are constraints, deriving for example from the rights of others, that limit what agents may do
to bring about the best possible state of affairs.
7. Cultural rights are left to one side in this paper.
8. The International Budget Project, designed to build the capacity of civil society organisations
to influence government budget processes, is an example of an initiative of this kind. See
www.internationalbudget.org.
9. That it is so is argued forcibly by Thomas Pogge (2002 and 2005). Interestingly, Birdsall
et al. (2005), while arguing that factors internal to developing countries are in fact crucial in
determining their development performance, still acknowledge that measures with a strong
external component, such as reducing the supply of bribes from international companies
to corrupt governments, stopping arms sales to dangerous governments and improving the
transparency of payments from extractive industry companies to governments, do impact
significantly on development.
10. The fact that one is complicit in an activity harmful to others does not necessarily mean
that one should end this complicity rather than help the victims. As Thomas Pogge (2005)
points out, Oskar Schindler did more good by helping his Jewish workers than he would
have done by emigrating. Nonetheless, I would argue that complicity does place a burden of
proof on those involved to show, by an analysis of the nature of the wrongful actions, of their
complicity and of their capacity to help the victims, why avoiding complicity, if possible by
helping to stop the wrongful actions, should not be their primary moral concern. Emigrating
may have not been Schindler’s only realistic alternative to pursuing his business interests and
helping some of the victims of the Nazis. He may for example have had a much bigger impact
by putting his knowledge and contacts at the service of the German plots to overthrow Hitler.
11. I leave aside here questions about the extent to which the wealth of the West has depended on
an expropriation of resources from the parts of the world it once colonised.
References
Beitz, C. 2004. Human rights and the law of peoples. In The Ethics of Assistance: Morality and the
Distant Needy, ed. D. Chatterjee. Cambridge: Cambridge University Press.
Bell, D. and J.-M. Coicaud eds. 2007. Ethics in Action: The Ethical Challenges of International
Human Rights Nongovernmental Organizations. Cambridge: Cambridge University Press.
Birdsall, N., D. Rodrik, and A. Subramanian. 2005. How to help poor countries. Foreign Affairs
vol. 84(no. 4): 36–152.
Brouwer, M., H. Grady, V. Traore, and D. Wordofa. 2005. The experiences of Oxfam international
and its affiliates in rights-based programming and campaigning. In Reinventing Development?
Translating Rights-Based Approaches from Theory into Practice, eds. G. Paul and E. Jonathan.
London/New York: Zed Books.
62 K. Donaghue
Chandhoke, N. 2007. Thinking through social and economic rights. In Ethics in Action: The Ethical
Challenges of International Human Rights Nongovernmental Organizations, eds. D. Bell and
J.-M. Coicaud. Cambridge: Cambridge University Press.
Collier, P. 2007. The Bottom Billion: Why the Poorest Countries Are Failing and What Can Be
Done about It. New York: Oxford University Press.
Control Arms. 2003. www.controlarms.org/act_now.
Cranston, M. 1973. What Are Human Rights? London: Bodley Head.
DfID. 2000. Realising Human Rights for Poor People. London: UK Department for International
Development, http://www.dfid.gov.uk/pubs/files/tsphuman.pdf.
Dichter, T. 2003. Despite Good Intentions: Why Development Assistance to the Third World Has
Failed. Amherst/Boston, MA: University of Massachusetts Press.
Easterly, W. 2002. The Elusive Quest for Growth: Economists’ Adventures and Misadventures in
the Tropics. Cambridge, MA/London: MIT Press.
Edwards, M. 1999. Future Positive: International Cooperation in the 21st Century. London:
Earthscan.
Ellermann, D. 2001. Helping People Help Themselves: Toward a Theory of Autonomy-Compatible
Help. World Bank Policy Research Working Paper 2693.
Feinberg, J. 1973. Social Philosophy. Englewood Cliffs, NJ: Prentice-Hall.
Goering, C. 2007. Amnesty International and Economic, Social and Cultural Rights. In
Ethics in Action: The Ethical Challenges of International Human Rights Nongovernmental
Organizations, eds. D. Bell and J.-M. Coicaud. Cambridge: Cambridge University Press.
Hunt, P., M. Nowak, and S. Osmani. 2004. Human Rights and Poverty Reduction: A Conceptual
Framework. New York/Geneva: Office of the UN High Commissioner for Human Rights.
Marks, S. 2004. The human right to development: Between rhetoric and reality. Harvard Human
Rights Journal 17: 137–168, Spring.
Nelson, P.J. and E. Dorsey. 2003. At the Nexus of human rights and development: New methods
and strategies of global NGOs. World Development 31(12): 2013–2026.
Nickel, J.W. 2007. Making Sense of Human Rights, 2nd ed. Oxford: Blackwell.
Oxfam International. 2000. Towards Global Equity: Strategic Plan 2001–2004. Oxford: Oxfam
International, http://www.oxfam.org/en/files/strat_plan.pdf.
Oxfam International. 2006a. Promises to Keep: Report on the Evaluation of the Implementation
of Towards Global Equity, Oxfam’s Strategic Plan 2001–2006, Oxfam International,
http://www.oxfam.org/en/files/promises_to_keep.pdf .
Oxfam International. 2006b. Response to Report on the Evaluation of the Implementation of
‘Towards Global Equity’, Oxfam International’s Strategic Plan 2001–2006. Oxford: Oxfam
International, http://www.oxfam.org/en/files/promises_to_keep_oi_response.pdf.
Pearson, N. 2006. Road to responsibility. The Australian 30 Sep.
Pogge, T. 1995. How should human rights be conceived? Jahrbuch für Recht und Ethik 3: 103–120
(reprinted in Pogge 2002).
Pogge, T. 2002. World Poverty and Human Rights: Cosmopolitan Responsibilities and Reforms.
Pogge, T. 2005. Severe poverty as a violation of negative duties. Ethics and International Affairs
vol. 19(1): 55–83.
Rostow, W.W. 1960. The Stages of Economic Growth: A Non-Communist Manifesto. Cambridge:
Cambridge University Press.
Roth, K. 2007a. Defending economic, social and cultural rights: practical issues faced by an inter-
national human rights organization. In Ethics in Action: The Ethical Challenges of International
Human Rights Nongovernmental Organizations, eds. D. Bell and J.-M. Coicaud. Cambridge:
Roth, K. 2007b. Response to the critique of Neera Chandhoke. In Ethics in Action: The Ethical
Challenges of International Human Rights Nongovernmental Organizations, eds. D. Bell and
J.-M. Coicaud. Cambridge: Cambridge University Press.
Shue, H. 1980, Basic Rights: Subsistence, Affluence and US Foreign Policy (2nd ed 1996).
Princeton, NJ: Princeton University Press .
Shue, H. 1984. The interdependence of duties. In The Right to Food, eds. P. Alston and
K. Tomasevski. Boston, MA: M Nijhoff.
Shue, H. 2004. Thickening convergence. In The Ethics of Assistance, ed. D. Chatterjee. Cambridge:
Slim, H. 2002. A response to Peter Uvin—Making moral low ground: Rights as the struggle
for justice and the abolition of development. Praxis xvii: 1–11, http://fletcher.tufts.edu/praxis/
archives/xvii/Slim.pdf.
UNDP. 2000. Human Rights and Human Development. Oxford/New York: Oxford University
Press.
United Nations. 1966a. International Covenant on Economic, Social and Cultural Rights. New
York: United Nations.
United Nations. 1966b. International Covenant on Civil and Political Rights. New York: United
Nations.
Uvin, P. 2002. On high moral ground: The incorporation of human rights by the development
enterprise. Praxis xvii: 1–11, http://fletcher.tufts.edu/praxis/archives/xvii/Uvin.pdf.
Waldron, J. 1993. Liberal Rights: Collected Papers. Cambridge: Cambridge University Press.
Jubilee Australia. 2007. www.jubileeaustralia.org.
Make Trade Fair. 2007. www.maketradefair.com/en/index.htm.
Medecins Sans Frontieres. 2007. Malnutrition Campaign: Breaking Open the Debate on Treating
Malnutrition. Geneva: MSF, www.accessmed-msf.org.
The Ethics of Taking Sides
Peter Ellis
1 Introduction
In this paper, I will argue that development NGOs have glossed over the political
nature of their work. This gloss might be due to a lack of awareness; a desire to
please donors; or a tactic to facilitate political activity; but the precise reason is not
material to its results. Those results include growing criticisms of NGOs as political
actors which could lead to reduction in NGOs’ capacity if widely accepted by the
public; and a danger of corruption of the public sphere due to lack of a clear basis
and integrity for NGO political behaviour.
My starting point is that for all countries at all stages of development there are
different groups that consciously or unconsciously struggle for power and hold dif-
ferent visions for their society. While other definitions provide useful insights, for
the purpose of this chapter I will define politics as the process of change, coalition-
building, compromise and threat or fear of force by which these struggles and
differences are resolved, for both elites and the masses. This conceptual lens is sug-
gested in response to some of the other chapters in this book examining questions of
donor NGOs’ basis for cultural, political or economic interventions in others’ soci-
eties. Envisaging those societies as I suggest makes the basis of NGOs’ responses
to such conundrums clearer—they are simply taking sides in an existing political
struggle.
Which side to take and how to do so most effectively becomes the burning
question. I am particularly interested in some of the ethical issues that arise from
government funding.
The growing popularity since the 1980s of NGOs as a means for promoting polit-
ical change (especially the onset and consolidation of democracy) has led to a new
literature on what, if anything makes NGOs legitimate actors in such processes and
whether their own accountability is adequate to justify and control this participation.
P. Ellis (B)
Independent Monitoring and Results Advisor, NZAID
e-mail: ellisp@netspeed.com.au


66 P. Ellis
(Burnell and Calvert 2004, Jordan and Tuijl 2006, Slim 2002). Any concerns in these
areas become particularly acute when a relatively small number of donors gain some
degree of dominance over sources of funds. Due to their size, official donors are the
obvious source of such concerns.
Whatever one’s views on the general rights of funders with regard NGOs, when
the source of funds is a government and the activities being funded are political in a
particularly obvious fashion, the questions change in important ways. The dangers
for organisational integrity and strategic sustainability are obvious and well known
(Ahmed and Potter 2006). It is the unique position of states in development and
their relationship to NGOs that will be of most interest to me, and the bulk of the
paper will explore aspects of this relationship. In particular, I will highlight two sets
of attacks on NGOs that exploit vulnerabilities inherent in this relationship: from
those in recipient countries who argue NGOs are tools of imperialist governments
or international capital; and from those in donor countries who argue NGOs are
illegitimate political advocates.
2 Taking Sides in a Recipient Country
The following points attempt to illustrate several aspects of how NGO development
work is inherently political in the definition of this chapter. The starting point is a
(non-comprehensive) selection of criticisms that have been made of NGOs. The key
issue is the way that these criticisms highlight the fact that development work of any
sort is implicitly ‘taking sides’ in a political struggle.
2.1 Challenging the Political Equilibrium

A government or individual in the developing world might take as its starting point
the track record of the west in interfering in other countries through civil society
organisations. Historical examples such as CIA destabilisation of Chile in the early
1970s loom large in the minds of many. In that particular example, as a small part
of the overall scheme to foment a military coup the USA passed more than $1.5
million to civil society groups in opposition to Allende’s administration (Kinzer
2006). Today NGOs are hearing the message that there is no such thing as a neutral
policy in the so-called War on Terror (Stoddard 2003); and those in developing
countries know that this message is being heard.
As just the most prominent current example, governments and some relatively
mainstream social groups in countries such as Russia, Turkey and various nations
in central Asia are clearly concerned that the centuries-old Great Game of espi-
onage and covert extension of imperialist power continues unabated. Governments
in the former Soviet Union have initiated crack-downs on NGOs, particularly
foreign-funded NGOs, in response to fears:
The Ethics of Taking Sides 67
that foreign-funded NGOs spearheaded popular protests that overthrew governments in

Georgia, Ukraine, and Kyrgyzstan. Those observers say the government is worried that
NGOs could be planning a similar revolution in Russia. Russian President Vladimir Putin in
July said he would not allow foreign-funded NGOs to carry out what he said were political
activities (Bigg 2005).
Similarly, in Turkey, NGOs have come under attack for accepting funding from
groups such as the World Bank, European Union, George Soros and the CIA aimed
at covertly engineering cultural and political change (Geray 2006). Other examples
abound.
2.2 Substituting for the State and Promoting Neo-Liberalism
Recipient criticism of NGOs is not confined to fears that a human rights agenda can
make them the advance guard for a regime change operation. Between the decline
of the Cold War and the beginning of the War on Terror, neo-liberalism and struc-
tural adjustment held sway, aid and government budgets were both in decline, and
NGOs became increasingly fashionable delivery mechanisms for official develop-
ment assistance and promoting good governance. In some quarters, donor-country
NGOs became objects of suspicion. The coincidence of pressure to reduce the size
of the state and renewed interest in non-state actors as legitimate sources of service
delivery1 led to distrust on the part of those who saw nationalist states—democratic
or not—as the path to development.
As the state abdicated its traditional role, NGOs moved in to work in these areas (rural
development, agriculture, energy, transport and public health). But their available funds
are a minute fraction of the cut in public spending. Most wealthy NGOs are financed and
patronised by aid and development agencies, funded by western governments, the World
Bank and the United Nations and multinational corporations . . . they are certainly part of
the same political formation that oversees the neoliberal project and demands the slash in
government spending (Roy 2004); see also a discussion of similar criticisms from academia
in (Cleary 1997).
Similarly, Mudingu writes in the New Times in Rwanda:

The NGOs are financed and directed by the various imperialist agencies, the imperialist
governments and the comprador regimes. They act as the liaison between the people and
the governments. They are the vehicles through which the exploiters seek to influence the
opinions of civil society. They are the servants of imperialist capital. Almost all the NGOs
are directed by the invisible hand of the imperialists who set them up or fund them in accor-
dance with their strategic goals. Huge funds are thus poured into the coffers of the NGOs in
the name of development, social justice, human rights, grassroots democracy, etc . . . With
such huge funds at their disposal the NGOs act as elitist organisations completely divorced
from the masses . . . NGOs try to instil the false belief among the oppressed that there is no
alternative to capitalism . . . the state is thus absolved of all its social responsibilities towards
the people (Mudingu 2007).
Korten (1992) has argued against NGOs accepting any official funds because
doing so implicates them in advancing the interests of international capital and ‘a
development theology based on the worship of money’. Doing otherwise is to ‘join
68 P. Ellis
the line of people and institutions eager to exchange their values for money at the
World Bank teller’s cage’.
Arundhati Roy warns of ‘the NGO-isation of resistance’ that will blunt real pres-
sure for reform in poor countries; that while NGOs give the impression ‘they are
filling a vacuum created by a retreating state. . .their real contribution is that they
defuse political anger and dole out as aid or benevolence what people ought to have
by right’. Similar views are explored at more length in the academic literature by
writers who demonstrate the impossibility of genuine autonomy for NGOs reliant
on external funding and the stifling nature of foreign support for domestic NGOs
(Kapoor 2005, Vincent 2006). Fowler (2000) argues that the growth in use of NGOs
since the 1980s has been in placatory socio-economic services and that ‘NGDOs
have not really succeeded in co-opting the official aid system; more the other way
around’ (Fowler 2000).
2.3 NGOs as Differently Accountable to State Entities
The distrust between recipient governments (referred to earlier) and donor NGOs
is in at least one respect mutual in a way that interestingly highlights concerns
about a neo-liberal anti-state agenda from donor NGOs. Publics in donor coun-
tries regularly identify corruption in recipient governments as the key constraint to
effective development or aid delivery. NGOs’ good reputation with donor publics
is dependent in large part on the perception that they are a participatory, pro-
poor, effective alternative to developing country governments (ChildFund Australia
2007).
However, even if an NGO is only digging wells, this is a political act in the terms
of the definition of this chapter. The choice of where to dig them, what method
to use, which ethnic or political groups to favour or treat equitably, inevitably
mean that even the most innocuous development activity is taking sides with one
view of what should happen in a society. NGOs typically engage in activities that
governments do not, because of either ineffectiveness or deliberate political deci-
sion or simply because they can deliver benefits to people who would otherwise
lose out.
There are obvious difficulties for NGOs seeking to be more democratic, account-
able, and pro-poor—not to mention less corrupt—than recipient governments. At
least such governments often have the advantage (sometimes dubious, but better
than nothing) of complex institutions such as Parliaments, ombudsmen, auditors and
elections. While NGOs have their own forms of accountability and are undertaking
strenuous efforts to improve accountability to beneficiaries, they quite rightly do
not aspire to a complete substitute of such processes (Jordan and Tuijl 2006, Kilby
2006, Lawday 2006).
Empirical work has shown that while development NGOs may believe they are
participatory and empowering and doing what the poor want and need, the reality
can be much more organisationally driven. NGO programs can end up as a pale
reflection of those of the state; but possibly without the checks and balances (such
as they are) of even a recipient country government’s systems:
NGOs behave like the state in attending to public projects, like the market in generating
finances, and like civil society in promoting and protecting civic values and participatory
development . . . When the interface between development NGOs and rural communities is
based on state or market agendas, rather than those of the people themselves, they are likely
to facilitate new forms of imperialism (Tembo 2003).
3 The Attack on Advocacy NGOs in Australia
I now move to another aspect of attacks on NGOs which involves critics of their
political relation to donor states. Whereas the criticisms outlined above have tended
to come from the left (or from nationalist movements in poor countries, which often
have the same political opponents as the left in rich countries), this set of criti-
cisms has tended to come from the right. I will focus on the Australian case study
where the issue has become particularly prominent since the Howard Government
came to power in 1996, but the phenomenon in question is in no way limited
to here.
Two important parts of the relationship between NGOs and the state have been
highlighted in debates over this period: participation in public debate; and gov-
ernment funding. Inevitably, these two elements become entwined. As we shall
see below, the most common criticism of NGOs in this context is a view that the
more NGOs receive public funding, the more they have an obligation to withdraw
from public debate (or to support Government positions). NGOs themselves express
this more directly—from their perspective, ‘you don’t bite the hand that feeds
you’.
Since the 1970s, NGOs and a variety of quasi-independent bodies (such as
the Human Rights and Equal Opportunity Commission) have gained a foothold
in Australia that consisted of both funding and an increasingly formalised role
in debate on public policy. This role was enhanced through newly strengthened
committee processes in Parliament, new Freedom of Information legislation, and
improved consultation between the executive arm of government and ‘stakeholders’.
Both the funding and public debate roles were, overall, strengthened throughout the
Fraser, Hawke and Keating Governments.
Then in 1996, the winds of electoral change began to blow strongly from the
right. As Joan Staples writes in the 1996 Menzies lecture that has set the tone for
his successive governments, ‘Howard referred to the NGO sector as “single-issue
groups”, “special interests” and “elites” and he promised that his government would
be “owned by no special interests, defending no special privileges and accountable
only to the Australian people”.’ (Staples 2006) Elsewhere at that time, he claimed
that there is a ‘frustrated mainstream in Australia today which sees government deci-
sions increasingly driven by the noisy, self-interested clamour of powerful vested
interests.’ It is important to note parenthetically that in this context ‘self-interested
70 P. Ellis
. . . powerful vested interests’ and ‘elites’ refers to groups seen as hanging parasit-
ically off the state and dependent on it for funding and political power, not groups
(such as business lobbies) that are powerful in their own right before taking a hand
in policy creation (Maddison et al. 2004).
Howard’s apparent dislike of NGOs reflects public choice theory (Maddison and
Dennis 2005, Staples 2006). Public choice theory is suspicious of any politically
or economically organised subset of society. It sees such groups as self-interested
rent seekers, aiming to exercise influence on the state disproportionate with their
appropriate democratic level of influence. Unlike democratically elected govern-
ments, so-called special interest groups have by definition a limited set of issues
over which they wish to exercise influence. Allowing them to do so is predicted to
result in piecemeal policy-making, to favour interest groups that have political clout
for often irrational reasons (e.g. media-friendliness of their imagery), and lead to
a range of ‘squeaky wheel’ problems—problems where a small but vocal minor-
ity can influence the state to support them at the cost of welfare for a much larger
group of society, none of whom are impacted severely enough to take up political
arms. Economically irrational (where total costs outweigh total benefits) protection-
ism and industry subsidies are the paradigmatic examples of such ‘squeaky wheel’
problems in public choice theory—the costs of subsidies are spread over the whole
population who hardly notice them, while the benefits accrue to a small but high
profile group with media and political clout even if the total benefits are less than
total costs.
Bearing in mind this critique, we can identify a continuum of four views, already
articulated in the public sphere, on the role of ‘special interest’ groups including
NGOs. The categorisation is my own but is influenced by other authors (Phillips
2006, Dalton and Lyons 2005). These four views and a brief explanation of their
philosophical rationale within a fundamentally liberal perspective on the state and
society are as follows:
A. NGOs can criticise the government, even if using government funding.

Rationale: Effective democracy requires continued community involvement and
consultation, not just periodic elections. Marginalised and powerless groups
cannot afford their own lobby groups and hence it is appropriate for the state
to identify such groups by a fair, open and democratic process and to subsidise
the engagement of NGOs that represent them in public debate and the political
process. It is important only that the NGO has open and accountable ways of
ensuring it represents the interests of its constituency. The government should
not intervene even when a subsidised NGO enters the political fray against the
government’s own perceived interests.
B. NGOs can criticise the government, so long as they carefully quarantine the
resources to do so from any government funding.
Rationale: It is appropriate for NGOs to engage in policy debate to represent
their constituency, but it would be dangerous for them to use government funds
to do so. In contrast to View A, Government funds should be strictly allocated
to service delivery to avoid inappropriate use of state resources for propaganda
purposes; otherwise we would see hijacking of the political debate by NGOs

that are in effect arms of the state with an unfair advantage over other political
players.
C. NGOs that receive government funding shouldn’t criticise (or, in some versions,
praise) the government even if they do so with other resources.
Rationale: NGOs that accept money from the state are giving up their indepen-
dence and become de facto an arm of the public service even if they receive
some funding from other sources. The risk of state funding being mixed up with
the independent funding (in reality or perception) is too great to allow such a
body to engage in the public debate. All resources are fungible. Also, the pres-
tige and political positioning of the NGO will come in part from its receipt of
government funds. The only acceptable form of pluralistic political engagement
is by groups that represent the views and interests of their constituencies without
any state subsidy. The corollary of this argument is that NGOs that do receive
government funding, even for quarantined programs, should be distrusted as
having possibly ‘sold out’ and any political statements they make—particularly
if positive towards the government—treated with skepticism.
D. NGOs shouldn’t criticise the government, regardless of their funding source.
Rationale: The only acceptable political engagement is by individuals and politi-
cal parties. Once a government has won a mandate at an election, it should be left
alone to govern; perhaps even Parliament should not have a right to frustrate the
Government’s will, and certainly special interest groups should not be consulted
or brought in to the policy process between elections. Even privately funded
groups such as Greenpeace warp the public debate through emotive single-issue
behaviour and should not be allowed. The only role for NGOs is in service
delivery.
These views can be classified as ‘level the playing field and encourage political
pluralism’ (for A); ‘allow political pluralism but be careful to avoid mixing service
delivery with advocacy in the political process’ (B and C) and ‘keep the special inter-
ests out of politics and let the elected government govern’ (D). View D is sometimes
envisaged as the logical outcome of public choice theory; although interestingly, the
public choice paradigm (interpreted loosely as application of economics-style cal-
culations and individualist liberal political theory) can be, in my opinion, readily
adapted to any of the four positions. View A, for example, could be envisaged as a
straightforward and justified state response to a predictable failure in the market of
ideas and political debate; without invoking any socialistic, community-oriented or
altruistic critiques of neo-liberal foundation beliefs.
All four of these views are heard in the mainstream of political debate in Australia
today; in Parliament, newspaper columns, talkback radio and blogs to name just a
few of the main media. It would be fair to say that NGOs, in Australia at least, have
in public largely abandoned A and are struggling to defend the trenches of B; while
critics such as the Institute of Public Affairs (IPA) and the Government move freely
between B, C and D. Of course, there are interesting differences of philosophy;
some NGOs such as Greenpeace and Human Rights Watch have long refused any
72 P. Ellis
government funding, allowing them to reside more or less impregnably (so long as
they can attract a paying support base) within view C.
I will not overly linger over what the specifics of the alleged attack by the
Howard Government and its political allies (particularly the IPA) on NGOs. These
are well documented elsewhere (Hamilton and Maddison 2007, Maddison et al.
2004, Maddison and Dennis 2005, Maddison 2007, Maddison and Hamilton 2007,
Staples 2006). It suffices to say that the Australian NGO community has come under
sustained criticism and pressure on the basis of breaches of their role in the terms
of views B, C or D according to the circumstance. Accusations range from misuse
of government funds for political purposes (a breach of view B) through to lack of
representativeness and accountability to the public (an attack informed by view D,
which denies the legitimacy of any non-electoral political participation).
The weapons used have included freezing out from consultation processes and
debate; public abuse; defunding; micromanagement of relationships; tactical use of
evaluations and audits; and bullying threats that use of any of these tools might
be imminent if NGO criticisms are not withdrawn or changed. 90 percent of NGO
respondents to a large and credible survey by the Australia Institute believe that dis-
senting organisations risk having their funding cut; 74 percent believe that NGOs
are being pressured to make their public statements conform with government pol-
icy; and 92 percent disagree with the view that dissenting voices are valued by the
government as part of a robust democracy (42 percent strongly disagree). 70 per-
cent of NGOs receiving government support report restrictions on their ability to
comment on government policy; many comment on implicit constraints.
4 The Consequences of Taking Sides at Home
The literature is relatively silent on the impact on development and humanitarian

NGOs of the Howard Government’s push to ‘silence dissent’. One study that looked
at the issue internationally found no relationship between the proportion of income
from government sources and advocacy expenditure (Anderson 2000). However,
this measure could not take into account any changes in message or strength of
advocacy as a result of direct or indirect government pressure. In fact, there is no
reason to think that the development NGO community has been in any way exempt
from the government push to control its actual and potential critics. In my own
experience in the Australian public service, I came across micro-examples of the
government’s side of the issue including:
• An NGO-organised conference with a prominent international speaker had

AusAID funding and sponsorship signs withdrawn at the last moment when the
speaker was found to have criticised USA foreign policy and Australia for its
dependence on it
• An Australian NGO that was subjected to non-scheduled performance audit,
widely believed within AusAID to be because the CEO repeatedly criticised the
Government
• Australian NGOs entering cooperation agreements with AusAID were told by the
Canberra-based director ‘we now consider you part of the family, and don’t want
you making any public comment on this country without clearing it with us’
• A university-organised conference that expected AusAID funding had it denied
at the last minute by Minister for Foreign Affairs Alexander Downer for fear that
some speakers would use the opportunity to criticise Australian policy towards
the subject country
• An East Timorese human rights NGO (Forum Tau Matan) had its contract bro-
ken (on the decision of Minister Alexander Downer; Senate 2005) when it was
found that it had earlier signed joint media releases criticising the Australian
Government over Timor Sea boundary negotiations (La’o Hamutuk 2005, East
Timor NGO Forum 2004a, b). Other NGOs that had signed the same media
release were later denied funding for the same reason (Haburas 2005).
The East Timorese example brought into the open some of the arguments nor-
mally held behind closed doors. Don D’Cruz, at that time a Research Fellow for
IPA and Director of the NGO Watch project, wrote of the withdrawal of funding
from Forum Tau Matan that it was
about time . . .. Since when did Australian taxpayer funding become a “human right?” I think
I missed this one. They can still talk, but it is just that the long-suffering Australian taxpayer
doesn’t have to foot the bill. Australian taxpayers do not owe East Timorese activists a
living! (D’Cruz 2005).
D’Cruz uses some View B rhetoric (‘Australian taxpayer doesn’t have to foot the
bill’, implying concern that Australian funds should not be used for criticism of the
Australian Government), but I think that the View C basis for the original Ministerial
decision to break the contract is undeniable. In fact, the case of the Timorese NGOs
is unusual exactly because it is a clear case of a donor government’s assertive use
of View C, which is only rarely given such a public airing. A senior AusAID offi-
cial advised the Senate ‘It was reviewed, and it was considered, in the end, that
it was not appropriate to continue funding an organisation that was openly criti-
cal of Australia’ (Senate 2005). There was never any claim by the Government its
own funds had been used for advocacy against Australia or might be diverted to
such purposes. Interestingly, as the original grant was for something intrinsically
political—advocacy and training on human rights issues in the justice sector in East
Timor—there could be no argument that Australia objected in principle to the NGO
engaging in political and advocacy activities. The only complaint in this instance
was the particular content and target.
It is interesting to contrast these actions against Timorese NGOs, informed by
View C, with the less assertive response to politically active Australian NGOs.
Several Australian NGOs—most prominently, Oxfam Australia—had engaged in
at least as much political activity on the maritime boundaries issue as had Timorese
NGOs such as Forum Tau Matan. Yet funding to those NGOs—for example, through
the AusAID NGO Cooperation Program—was not (to my knowledge) cut, even
when Oxfam released a key report on the issue and Downer appeared on the BBC
74 P. Ellis
lamenting ‘hysterical and emotional and irrational claims’ and ‘a whole lot of emo-
tional claptrap which is being pumped up through sort of left wing NGOs’ (BBC
World Tonight 2004).
For Australian development NGOs engaged in advocacy then, View B still pre-
vails. We can compare the View B line in the ANCP guidelines (for Australian
NGOs) that AusAID will not fund activities which ‘are determined reasonably to
be contrary to the interests of the Commonwealth of Australia’ (AusAID 2007)
with the View C clause in the guidelines of the East Timor Community Assistance
Scheme (under which two East Timorese NGOs were refused grants) that rules out
support to whole organisations but on a similar criterion. The most obvious explana-
tion is that Australian NGOs have domestic political clout in Australia. This would
make a general move against them—based on View C—damaging politically in a
way that such a move against Timorese NGOs (which have no brand recognition or
significant support base in Australia) was not.
The hypothesis that only Oxfam’s domestic political clout protected it from an
attack by the Government on the basis of View C is supported by the withdrawal
of tax deductibility status from the small Australian watchdog NGO Aid/Watch in
May 2007, apparently because it has been engaging in political activities defined as
‘seeking changes to government decisions and propagations of a particular point
of view’ (Wade 2007). It should be noted that campaigning organisations such
as Greenpeace operating in other segments of the NGO market have retained the
tax-deductible status of their donations. Oxfam’s campaigning funds (unusually for
Australian NGOs) are deliberately kept separate from the tax-deductible ‘develop-
ment’ funds and so it is technically not as vulnerable to this exact counter-measure
as Aid/Watch (although it is vulnerable in other areas including withdrawal of direct
funding). However, perhaps more germane is that it is hard to imagine any sig-
nificant number of Aid/Watch supporters voting for the Coalition, whereas Oxfam
donors would include a proportion of actual or potential Coalition voters. It seems
probable that, had Aid/Watch a wider public following in Australia including in the
crucial swinging voter category, a move against it would have been much harder
politically for the Government to sustain.
When we look internationally for more evidence of the sort of direct or indirect
pressure that NGOs may come under as a result of accepting government funding,
two issues have had particular prominence over the past 5 years or so. These are
involvement in the humanitarian and reconstruction efforts in Iraq; and limitations
placed on the use of HIV/AIDS funding by the Bush Administration.
In the Iraq instance, the ethical issues are complex and have changed over
time. In the build-up to the invasion—opposed by many but not all NGOs as
illegal—belligerent nations sought to involve NGOs in planning for the inevitable
humanitarian disaster. A number of NGOs refused to accept such funds—at least
two stated that they did so to avoid being silenced—but some did participate even
from this stage. At a later stage, there is evidence that at least two international
NGOs came under pressure from their USA affiliates to tone down any criticism of
the invasion in response to political realities in the USA (the extent to which any
such toning down occurred is disputed).
After the invasion, as the security situation deteriorated and NGOs began to shut
down operations for security reasons, the invading governments sought to play down
the breakdown in society, law and order. Statements were made by Ministers deny-
ing Iraq was descending into the chaos of sectarian violence and civil war. Pressure
was exerted on NGOs amongst others to maintain operations in Iraq to back up these
claims. This phase reached a tragic conclusion when the Country Director of Care
International—one of the NGOs most heavily dependent on government funding
and regarded by many commentators as a ‘public service contractor’ in Fowler’s
terminology—was kidnapped and killed by insurgents. Recriminations and ethical
soul searching followed in the NGO community (Brown 2004, Cater 2003, Maguire
2003, Read 2003, Save the Children UK 2003, Slim 2003).
The HIV/AIDS funding issue is fairly straightforward. President Bush’s $15
billion commitment to address the issue internationally came with an emphasis
(some would argue over-emphasis) on pharmaceutical treatment as well as some
public-comment conditions on implementing bodies. Amongst other requirements,
organisations had to emphasise abstinence as the primary strategy for prevention
and to pledge to condemn prostitution despite the additional stigma and hence
outreach problems such an approach involves (Sealey 2005, Boonstra 2003, and
numerous other similar references). Court challenges to the requirement for a
pledge (which was thought to potentially breach US constitutionally mandated
rights for freedom of speech) were unsuccessful (SIECUS 2006). As a result
organisations—and even countries, such as Brazil—have had to make hard choices
between turning down badly needed funds, or refocusing their programs and taking
a public advocacy position that is either at odds with their technical knowledge
and experience or simply an issue on which they have never sought a mandate
to speak.
The HIV/AIDS issue shows the US Administration using View C to exercise
leverage with its funds. By accepting responsibility as the largest funder of programs
in a particular sector, the Administration has bought itself the right not only to direct
the use of those funds but to censor the public debate even when that debate is
paid from other sources. Organisations receiving US government assistance now no
longer can debate the issue freely even on their own time and funding.
What is the purpose of documenting these instances of pressure on development
NGOs’ public comment as a result of funding? Primarily, I intend to bring into the
open the undeniable existence of the phenomenon. I don’t intend to lament it, or ask
the governments of the world to refrain from such pressure. Instead, I argue that the
pressure to silence dissent by donor governments is just another, particularly visible,
consequence of the political nature of NGOs’ activities in general. This aspect of
NGO politics is different but complementary to the criticisms from recipient coun-
tries described earlier. It is only because of the particular content of these political
activities and because advocacy close to home quickly attracts attention that donor
governments are exerting pressure in these areas rather than the political actions of
more traditional development work. The case study of Forum Tau Matan illustrates
this perfectly—donor governments tolerate or encourage political action in recipient
countries but have more concerns when it comes home to roost.
76 P. Ellis
If NGOs’ programs in recipient countries are inevitably political, then just as

much are their advocacy efforts that impact on donor governments. NGOs should
expect attacks and attempts to control their comment such as those above—as much
as if they were an acknowledged political player in their countries. I am not arguing
that the solution is necessarily to reject government funding (although I believe
there is a stronger basis for doing so because of this political pressure than the
‘imperialism’ arguments). Instead, the solution is to identify the basis of both attacks
and NGOs’ defence—Views A, B, C or D—and for NGOs to adopt more solid
ground for their political activities.
5 The Structure and Political Legitimacy of NGOs

The critiques of NGOs influence in recipient countries and the range of criticisms
of their advocacy role in donor countries have in common a basis in the ambiguous
and contested political status of NGOs, particularly when in receipt of government
funding. I am suggesting that in their response NGOs need to acknowledge the
fundamentally political nature of their work.
5.1 Structure of NGOs
Few NGOs—in either donor or recipient nations—have broad membership struc-

tures with a strong voice in choosing policy and the approach to advocacy. More
common is a board that is effectively self-perpetuating (by choosing its own mem-
bers) and an executive staff that is highly influential in policy direction for both
advocacy and development work. Funders (private or government) typically have
an ill-defined and informal but significant voice in policy formulation through
the implicit threat of withdrawal if they disapprove of organisational directions.2
Beneficiaries’ structural role is typically informal, albeit the subject of much
soul-searching on the part of NGOs.
The relation between members or board and the executive is a familiar principal-
agent challenge from management theory. The relationship between the executive
and beneficiaries, mediated by partner organisations, is the primary subject of
development NGO programming expertise. And the relationship between the exec-
utive and funders is the domain of professional fundraisers and often of senior
management within the NGO.
Within the NGO community, discussion of both advocacy and development often
emphasises the importance of speaking or working on behalf of beneficiaries (not
‘representing’ them—although the distinction can be a fine one) and the questions
of donor NGOs’ legitimacy to do so.
Within this structure, what gives an NGO political legitimacy in either a donor
or recipient country?
5.2 Political Parties’ Political Legitimacy

It is instructive to start the consideration of this question with a counter example—
the groups in donor society that, however imperfectly, are agreed by all to possess
legitimacy for political action. Donor countries have well established systems to
regulate the more formal and acknowledged political actors. Political parties are
often governed by laws and regulations mandating internal democracy; funders of
political parties need to be listed publicly; and there are rules about who is allowed
to provide such funding (often preventing, for example, foreign funders). There are
usually strict rules about political advertising (including from third parties) requir-
ing, for example, an authorising individual’s name and address to be published with
all election campaign material. Party members’ control over and ultimate responsi-
bility for policy is jealously guarded; and funders’ relationship to policy is strictly
regulated by ethical or legal rules. Finally, for a political party, the ‘feedback’ from
the intended beneficiaries is well defined—it comes in the form of electoral support,
primarily expressed at the ballot box or through opinion polling.
Of course, it is equally true that imperfections and manipulation in all these areas
by political parties are widely known and growing. Grassroots members of political
parties have always complained that they don’t have enough say over policy, but
those complaints are growing and are reflected in chronic decline in party mem-
bership. Just as worryingly, many critics increasingly fear that funders of political
parties have an inappropriate disproportionate say on policy formulation. However,
the point for now is not so much the degree to which the norms are followed as that
those norms exist; and it is their existence that gives political parties the ‘right’ to
engage in political debate. It is precisely because such norms are so important that
some commentators are so concerned at possible breakdown of these two points.
The motives of those—individuals or corporations—who fund political parties
are of course complex, and they certainly include the hope to buy policy influence.
But formally, it is an accepted democratic norm that no funder, no matter how impor-
tant, should be able to buy policy changes, public comment, questions in parliament,
or silence on an issue, at least without ‘legitimate’ political legitimation through the
membership in the party’s grassroots and/or from beneficiaries via the ballot box.
The fact that nearly all parties seek to make political capital by accusing their oppo-
nents of allowing their policies and advocacy to be purchased3 merely demonstrates
that the norm still holds traction.
5.3 NGOs’ Political Legitimacy

Comparing NGOs to the obvious benchmark, political parties in a pluralist democ-
racy, could be seen as adopting a very narrow view of legitimacy for political
activity. This would implicitly derive from what I have categorised as View C or
even View D; a world view that assumes the only legitimate political players are
political parties and that all elements the political contest should be based on some
form of representative democracy. This comparison has been strongly critiqued:
78 P. Ellis
. . . representation is only one of many routes to legitimacy, and for most NGOs not the
most relevant one (unless, of course, they claim it for themselves). It is high time that this
particular bugbear was laid to rest. NGOs do not have to be representative to be legitimate,
but they do have to be accountable for their actions, whatever they are, if their claims to
legitimacy are to be sustained. This conclusion places the focus of the debate back where it
belongs—on the costs and benefits of different, concrete approaches to accountability—and
not abstract criticisms about NGOs that supposedly compete with governments as represen-
tatives of the electorate, a goal that no NGO, to my knowledge, has subscribed to (Edwards
2006)
Martha Schweitz has put forward three reasons for NGO existence with no basis
in representation: ‘first, being sources of information and expertise; second, deliv-
ering services to people; and third, standing up for a core value’ (Charnovitz 2006).
Charnovitz writes, ‘representation is simply a red herring’ (p. 36). For Peruzzotti
(2006), ‘it would be erroneous to extrapolate mindlessly from the conceptual frame-
work employed to analyse representative relations to civil society . . . civil society
is not a representative instance, but a constituent one’. Accordingly, he argues that
NGOs must be free to participate politically, no matter how little support there is
for their values in the society outside the freely associating citizens that make up
the NGO.
I agree that freely associating citizens should be able to participate in public
political debate. However, there is something slightly disingenuous about the claims
from the above authors that representation is irrelevant. There is more similarity
between NGOs and formal political actors than is sometimes admitted; in their work
in both recipient and donor nations.
When leaders of NGOs with tens or hundreds of thousands of largely passive
donors (such as child sponsors or other regular contributors) meet with a Minister
to talk about aid and foreign policy, are those individuals really only providing
information and expertise and standing up for core values? Or are they seen as repre-
sentatives of an important political constituency? Do donor NGO leaders never point
out how many of their donors may live in marginal electorates? To what degree are
these individual donors knowingly participating in a political free association? Do
NGO leaders not claim (or receive without claiming) public legitimacy from their
support bases in donor countries and presumed grass-roots networks in recipient
countries? If not, then why do their pronouncements receive more respect and atten-
tion in the media than those of individual (but expert) academics or critics? Even a
cursory examination of the nature of the political sphere would make it clear that,
intentionally or not, NGOs cash in on their perceived representativeness in public
debate.
However, the matter gets murkier still when we consider the wild card of gov-
ernment funding or tax deductibility. Once an organisation receives state support
from its own government or another, it is not accurate to describe it as a collection
of freely associating individuals that can put forward its views like any other con-
stituent group. The relationship of that funding to any political activity needs to be
very carefully defined indeed.
Building on Schweitz’s list, I suggest the following as sources of legitimacy for
NGOs engaging in political activity, be it advocacy or development work:
i. technical expertise, including both development knowledge and standing up for

core values such as rights presented in a technical-legal discourse
ii. representation of citizens or groups in a recipient country, including facilitating
bottom-up development
iii. representation of citizens in a donor country, whether they be members, donors
or a vaguely defined set of like-minded supporters
iv. delegated from another source, such as a donor or aid-recipient government
(perhaps when undertaking a public service contractor role)
6 Ethics for Political NGOs
6.1 Ethical Challenges

I now suggest that NGOs not only need to acknowledge the political nature of their
work in both developing and donor countries. They also need to clearly distinguish
to themselves and to the outside world which of these four sources of legitimacy
they are drawing on for doing so, in any particular circumstance.
There are practical reasons for doing so. Those reasons amount to recognition
of the danger of critiques growing in volume and power until they severely impede
NGOs’ capacity. But there are also powerful ethical reasons, including the pursuit of
integrity; honesty and good faith when acting as an agent or representative of others;
and obligations for collective nurturing of a peaceful and productive political sphere.
Publicly advocating or undertaking development activity on the pretence that an
NGO is doing so as representative of some group of citizens (funders, supporters or
beneficiaries) when in fact the motivation is the professional views of NGOs’ man-
agers is particularly dangerous. The danger of a ‘not in my name’ lashback from
those claiming to be represented—whether they are donors or beneficiaries—is a
serious practical one; and the misrepresentation would clearly violate commonly
held norms of honesty and integrity. Comparable practical ethical issues would
arise when any NGO political action that is genuinely legitimised by one of the
four reasons above is blurred as taken place on another basis. Hypocrisy of any
sort would cause similar problems. The use of funds from donors (government or
not) for ends that are perceived to have been to some degree hidden is particularly
explosive in its potential ethical and practical problems. Further, the failure to be
honest and to uphold models of organisational integrity could have the cumulative
impact of a corrosion of the public sphere. Such corrosion would take the form of
declining trust in political institutions and in peaceful mechanisms of conflict res-
olution, increasing the threats to democracy in donor countries as well as in aid
recipients.
It might be argued that the acknowledgement of the political nature of devel-
opment and advocacy work would bring about a constriction in NGO activity. If
funders were aware of the political nature of NGOs’ development work in recip-
ient countries they might be less inclined to donate. If donor governments were
80 P. Ellis
confronted with a collective of NGOs demanding autonomy for their domestic

advocacy, they might simply cut off the funding tap.
This appears to be a weak argument for not addressing the problem. Most NGOs
espouse a strong concern with fairness, transparency and due process. A deonto-
logical ethic would certainly denounce hypocrisy of any sort for such agents; or the
wilful (albeit passive) deception of funders or corruption of the political sphere even
if they were means to an end of increased NGO activity.
Yet a utilitarian ethic would seem to provide a similar answer. A no-hypocrisy
principle is important on practical grounds to avoid a potentially devastating
counter-campaign accusing NGOs of espousing ‘do as I say, not as I do’ which
would certainly lead to problems for NGOs’ efforts to scale-up and build community
support for their objectives. More broadly, few commentators argue that NGOs can
solve the problems of poverty and global injustice simply by increasing the level of
fund-raising by whatever means. In as far as there is a global vision held by NGOs,
it is principle-based and such visions require adherence to principles from their pro-
ponents if they are to gain support. The principles commonly include empowerment
of the poor, bottom-up participation and addressing social injustices. While short
term and localised gains might be made in some of these areas by pretending to be
apolitical and raising funds whatever the cost, the long term and global costs of this
compromise of principle would surely be much greater, even if they came only in
the inevitable exposure of hypocrisy and reduced capacity for NGOs in the future.
One bottom line on perhaps the most important aspect is: if some funders are
unwilling to support what the NGO sees as a necessary political change process (in
a donor or recipient country) it would be unethical to accept their funds to support
that process.
6.2 Ethical Responses

While I acknowledge the comparison can be taken only so far, I argue that NGOs
have something to learn from the norms and regulations society relies on (or tries
to rely on!) with regard to political parties. One crucial step would be protection of
NGOs’ mission, values, programs and policies from being purchased by funders. A
second would be firm control of the executive arm of the NGO to ensure that what-
ever the accountability mechanisms of the NGO (be it democratic mass-membership
or ownership by a small board vested with responsibility for ensuring certain con-
stitutional values are set) they function as claimed. Third, given the absence of
electoral and polling feedback, it is necessary to tighten the feedback loop from ben-
eficiaries to the NGO on policy and advocacy questions; a topic beyond the scope
of this paper but considered by Roche’s chapter in the same volume.
On the important funding question, NGOs perhaps need tighter self-regulation
of their relations to funders, firmly setting out their status. Such regulation needs
to firmly specify that funding an NGO makes a group or individual a stakeholder
and gives rights of financial accountability and feedback, but it does not give the
power to control the NGO. Whether a funder is an individual or a government,
when there is politics on the line they should have the right to sign up to a program
they like, but not (at least as far as can be regulated) to buy influence. Alternatively,
NGOs rejecting this model (perhaps because they want individual donors to have a
democratic say over policy direction) should establish transparent and accountable
mechanisms that do give funders clearly-defined influence.
An important step on clarifying the most urgent funding question would be if
agencies in any particular donor country collectively stated on what conditions they
planned to accept government funding. A possible model can be found in New
Zealand, where the Strategic Framework for Relations between NZAID and New
Zealand NGOs provides inter alia ‘recognition of and support for the independence
of the NGO sector, including its right within law to comment on government pol-
icy and work for change in that policy, irrespective of any funding relationship that
might exist.’ (NZAID and New Zealand NGOs 2003). One can imagine the leading
Australian NGOs approaching the Australian government collectively and clearly
setting out a similar bottom-line—that a condition for them accepting government
funding is a similar recognition of their advocacy autonomy.
The ACFID Code of Conduct for Australian NGOs is recognised internationally
as a world leader. Yet it is silent on these issues. It has strong language preventing
the support of political parties or candidates, but no delineation of what might be
acceptable advocacy. It states that funds raised for ‘aid and development’ are only to
be used for those purposes. There is a section on communication with the public but
it focuses on basic organisational communication (e.g. Annual Report) and honest
and truthful fundraising. Advocacy and campaigning seem to have been placed in
the too hard basket.
It is possible to conceptualise a Code of Conduct with broad applicability to a
range of organisations that covers some of the ethical issues arising from the polit-
ical nature of NGO work in recipient and donor countries. While a full debate in
the community is necessary before any such Code were introduced, some possible
precepts such a Code might cover include:
• Don’t downplay the political side of NGO activity. It is commonly accepted in the
NGO community that aid is political yet this aspect seems to be systematically
under-emphasised in many public communications, particularly fund-raising but
also some development education efforts.
• Be transparent about political positions including in both ‘development’ and
advocacy work. Potential funders (private or government) and members (if there
is a membership structure) should have ready access to material information
about any political activity that may be of interest.
• Clearly define the process of policy development and exactly on what basis the
NGOs executive participates in political activity.
• Don’t claim representation, even implicitly, without a basis.
• Be clear to funders—especially including governments—on the basis on which
their support is accepted. Either funding buys influence over an NGOs policy or
it doesn’t. If it does, the means for doing so and parameters around that influence
should be completely transparent.
82 P. Ellis
• Don’t give in to funder pressure—particularly from governments—on advocacy

positions. This could be seen as a corollary to the previous point. It should be
adopted as an ethical stance by NGOs that pressure from funders is not a basis for
changing policy—unless, of course, that is the transparent accountable process
agreed to by the NGO.
• No hypocrisy in advocacy. An important element of ethical political activity is
that NGOs should live by the ideals they advocate for. As one of the most straight-
forward examples, an NGO campaigning for freedom of information for citizens
and tax-payers should not itself secrete information from its members and
donors.
7 Concluding Remarks
Many funders might find the political action of supporting a aid-recipient state-run
well-digging program acceptable, but be worried by a modified proposal to work
deliberately with those the state excludes for its own good or bad reasons. Other
funders might have exactly the opposite political concerns; or prefer a solution that
explicitly aims at rebalancing power in the recipient country or campaigning in the
donor’s. The point of acknowledging the political nature of all development activity
is not to scare donors away but allow informed and transparent choices about exactly
what political activity is acceptable.
This chapter set out initially to document one particular problem—some of
the critiques of NGOs’ political activities, amplified by the fact of state fund-
ing. The proposed solution is straightforward in conception—to acknowledge the
political nature of development and adopt a set of basic ethical standards to gov-
ern NGO activity with this insight in mind. This acknowledgement would be for
all development NGOs—not excluding, for example, church-based organisations
and so-called public service contractors.4 An adequate response might entail addi-
tions to codes of conduct to clarify minimums of transparency and organisation
integrity.
However, there is no move towards a codified recognition of political ethics at
the moment. Instead, with the continued growth of managerialism and profession-
alism of donor NGOs and the sudden recent growth in private donations (on top of
continued steady growth of government funding), NGOs are if anything heading in
the direction of reduced acknowledgement of their political role.
It is up to NGOs to accept and even embrace the political nature of their activ-
ity and put in place self-regulation accordingly. Failure to do so will only see an
increase in the sort of criticisms outlined in this paper and many more, and increas-
ing attacks on the reputation, honesty and integrity of NGOs. On the other hand, a
proper acknowledgement of the politics of development, including in NGOs’ deal-
ings with donors, would be a powerful statement of what constitutes acceptable
behaviour, and could even contribute an empowering vision to the broader social
change objectives of so many in the NGO movement.
Notes
1. By ‘service delivery’ I mean any state-funded program aimed at a population’s welfare; so any
official development assistance needs service delivery of a sort. I do not intend the distinction
between development work (investment) and service delivery (recurrent costs) sometimes used
in the development field.
2. This is often an unusual form of voice in that it is largely exercised unknowingly. While endeav-
ouring to stick to a set of core values and technical expertise, it is well known that NGOs
self-restrain themselves from undertaking advocacy or development work that is likely to alarm
too many of their donors.
3. In Australia for example, the Liberals accuse the ALP of being run by the unions through
their funding and internal political arrangements; the ALP accuse the Liberals of being owned
by their big business funders; and the Greens accuse all the major parties of being overly
influenced by their funders.
4. I do not intend to excuse government or multilateral aid agencies from making explicit the
political nature of their aid as much as NGOs’; but a discussion of those issues would exceed
the reference point of this chapter.
.
References
Ahmed, S. and D.M. Potter. 2006. NGOs in International Politics. Bloomfield: Kumarian Press.
Anderson, I. 2000. Northern NGO advocacy: Perceptions, reality, and the challenge. Development
in Practice 10(3&4): 445–452.
AusAID. 2007. AusAID-NGO Cooperation Program (ANCP) Guidelines. Canberra: Australian
Agency for International Development.
BBC World Tonight. 2004. ‘‘I’ve been talking to the Australian Foreign Affairs Minister Alexander
Downer. . .’’ BBC, London 19 May 2004, http://www.foreignminister.gov.au/transcripts/2004/
040519_bbc_world_tonight.html.
Bigg, C. 2005. Russia: NGOs Say New Bill Threatens Civil Freedom Radio Free
Europe, 23 Nov 2005.http://www.rferl.org/featuresarticle/2005/11/45a60888-5deb-4700-80cb-
e4dc3f68f277.html.
Boonstra, H. 2003. US Aids policy: Priority on treatment, conservatives’ approach to prevention.
The Guttmacher Report on Public Policy 6(3): 1–3.
Brown, I. 2004. This fatal compromise. The Guardian, Manchester, 19 Nov 2004.
Burnell, P. and P. Calvert, eds. 2004. Civil Society in Democratization. London: Frank Cass.
Cater, N. 2003. UK Offers Iraq Crisis Cash to NGOs Alertnet 12 Mar 2003, reproduced at:
http://www.globalpolicy.org/ngos/fund/2003/0312uk.htm.
Charnovitz, S. 2006. Accountability of non-governmental organizations in global governance. In
NGO Accountability: Politics, Principles and Innovations, eds. L. Jordan and P. van Tuijl, 21–
42. London: Earthscan.
ChildFund Australia. 2007. Perceptions of Children’s Needs and Aid Effectiveness in Developing
Countries 2007. Surry Hills: ChildFund Australia.
Cleary, S. 1997. The Role of NGOs Under Authoritarian Political Systems. London: Macmillan
Press Ltd.
Dalton, B. and M. Lyons. 2005. Representing the Disadvantaged in Australian Politics: The Role
of Advocacy Organisations. Canberra: Australian National University.
D’Cruz, D. 2005. Why it’s time to re-examine NGO funding. Crikey, 22 Nov 2005; also available
at http://friendsofsame.org/pipermail/fos-announce/2005-November/000240.html.
East Timor NGO Forum. 2004a. ET NGO Statement on ET-Australia Boundary Talks, media
release, Dili 29 Sep 2004, http://www.etan.org/et2004/october/01-03/29etngo.htm.
East Timor NGO Forum. 2004b. Timor-Leste Civil Society Demands a Fair Maritime Boundary’,
media release, Dili 27 Oct 2004, http://members.pcug.org.au/∼wildwood/04oct27etngo.html.
84 P. Ellis
Edwards, M. 2006. Foreword. In NGO Accountability: Politics, Principles and Innovations,

eds. L. Jordan and P. van Tuijl. London: Earthscan.
Fowler, A. 2000. NGO futures: Beyond aid: NGDO values and the fourth position. Third World
Quarterly 21(4): 589–603.
Geray, I. 2006. Does foreign funding make NGOs into puppets? New Anatolian, 11 Oct, 2006
reproduced at http://www.globalpolicy.org/ngos/fund/2006/1011puppets.htm.
Haburas. 2005. ‘AusAID Blacklists’ East Timor NGOs over Timor Sea Comments, media release,
14 Nov 2005, http://www.etan.org/et2005/november/14/14ausaid.htm.
Hamilton, C. and S. Maddison. 2007. Dissent in Australia. In Silencing Dissent, eds. C. Hamilton
and S. Maddison. Sydney: Allen &Unwin.
Jordan, L. and P. van Tuijl, eds. 2006. NGO Accountability: Politics, Principles & Innovations.
London: Earthscan.
Kapoor, D. 2005. NGO partnerships and the taming of the grassroots in rural India. Development
in Practice 15(2): 210–216.
Kilby, P. 2006. Accountability for empowerment: dilemmas facing non-governmental organiza-
tions. World Development vol. 34(6): 951–963.
Kinzer, S. 2006. Overthrow: America’s Century of Regime Change from Hawaii to Iraq. New York:
Times Books.
Korten, D. 1992. Open letter to an NGO working with the World Bank, People-Centred develop-
ment Forum, 7 Mar 1992 http://www.pcdf.org/1992/NGOSWB.htm.
Lawday, A. 2006. Accountability: A report card. Humanitarian Exchange 36(39–41).
La’o Hamutuk. 2005. Chronology of AusAID cancellation of grant to Forum Tau Matan
with links to relevant documents, 6 Oct 2005 http://www.laohamutuk.org/reports/AusAID/
FTMchronology.html.
Maddison, S. 2007. Redefining democracy. In Silencing Dissent, eds. C. Hamilton and
S. Maddison. Sydney: Allen &Unwin.
Maddison, S. and R. Dennis. 2005. Democratic constraint and embrace: Implications for pro-
gressive non-government advocacy organisations in Australia. Australian Journal of Political
Science 40(3): 373–389.
Maddison, S., R. Dennis, and C. Hamilton. 2004. Silencing Dissent. Canberra: The Australia
Institute.
Maddison, S. and C. Hamilton. 2007. Non-government organisations. In Silencing Dissent,
eds. C. Hamilton and S. Maddison. Sydney: Allen &Unwin.
Maguire, K. 2003. How British charity was silenced on Iraq. The Guardian, Manchester, 28 Nov
2003, reproduced at http://www.globalpolicy.org/ngos/fund/2003/1128silence.htm.
Mudingu, J. 2007. How genuine are NGOs? New Times, 7 Aug 2006.
NZAID and New Zealand NGOs. 2003. Strategic Policy Framework for Relations Between NZAID
and New Zealand NGOs. Wellington: NZAID, http://www.nzaid.govt.nz/library/docs/nzaid-
strategic-policy-framework.pdf.
Peruzzotti, E. 2006. Civil society, representation and accountability: restating current debates
on the representativeness and accountability of civic associations. In NGO Accountability:
Politics, Principles and Innovations, eds. L. Jordan and P. van Tuijl, 21–42. London:
Earthscan.
Phillips, R. 2006. The Role of nonprofit advocacy organizations in Australian democracy and
policy governance. Voluntas: International Journal of Voluntary and Nonprofit Organizations
17(1): 59–75.
Read, R. 2003. Aid agencies reject money due to strings. Oregonian, 6 June 2003, reproduced at
http://www.globalpolicy.org/ngos/fund/2003/0610aid.htm.
Roy, A. 2004. Help that hinders. Le Monde Diplomatique, Nov 2004 reproduced at http://www.
globalpolicy.org/ngos/credib/2004/1104hinders.htm.
Save the Children UK. 2003. Save the Children UK not silenced on Iraq. 28 Nov 2003.
Sealey, G. 2005. An epidemic failure. Rolling Stone, 2 June 2005, http://www.rollingstone.com/
politics/story/7371950/an_epidemic_failure/.
Senate. 2005. Supplementary budget estimates. In Foreign Affairs, Defence and Trade Legislation
Committee. Canberra: Commonwealth of Australia Senate, 3 Nov 2005.
SIECUS. 2006. U.S. HIV/AIDS funding dependent on anti-prostitution pledge that jeopardizes
HIV-prevention success. Making the Connection 4(2).
Slim, H. 2002. By What Authority? The Legitimacy and Accountability of Non-governmental
Organisations. The International Council on Human Rights Policy, International Meeting
on Global Trends and Human Rights—Before and after Sep 11, http://www.jha.ac/articles/
a082.htm.
Slim, H. 2003. Belligerent Funding, BOND, Apr 2003, reproduced at http://www. globalpol-
icy.org/ngos/fund/2003/0401bell.htm.
Staples, J. 2006. NGOs Out In the Cold: The Howard Government Policy Towards NGOs.
Australian National University, Canberra, June 2006.
Stoddard, A. 2003. With Us or Against Us? NGO Neutrality on the Line. Humanitarian
Practice Network, December 2003, reproduced at http://www.globalpolicy.org/ngos/fund/
2003/1200against.htm.
Tembo, F. 2003. The multi-image development NGO: An agent of the new imperialism?
Development in Practice, 13(5): 527–533.
Vincent, F. 2006. NGOs, social movements, external funding and dependency. Development 49(2):
22–28.
Wade, M. 2007. Canberra’s gagging us, say charities’. The Sydney Morning Herald, 30 May 2007,
Sydney.
The Epistemic Problem: Potential Solutions
Keith Horton
1 Introduction
Those of us who live in developed countries frequently receive appeals for funds
and other forms of support from aid agencies such as Oxfam, World Vision, and
CARE.1 How should we respond to such appeals? Should we give to such agencies?
Presumably, the answer to this question depends on how good or bad the effects of
their work are.2 If those effects are as positive as the fundraising literature of such
agencies tends to suggest, then there would be a very strong case for saying that
we should give to them. If such agencies do less good, and more harm, than they
like to imply though, the case for giving to them would presumably be weaker.
And if the effects of their work were bad enough, that case might break down
altogether.
If this is right, then it is an urgent matter to try to find out good or bad the effects
are;3 for until we do so, we will not know whether or not we should give to such
agencies. And this is morally troubling, especially as this is one of those cases in
which we must respond in one way or another (counting not giving as itself a type of
response). Of course, it would be unrealistic to expect any very precise judgement
about this matter. International aid is a complex business, and any judgements about
the effects are therefore likely to have to be very rough and probabilistic. And for the
same reason it would also be unrealistic to expect anything approaching certainty
that any such estimate is correct. If we literally had no idea at all about how good
or bad the effects were, though, then we would also have no idea about whether we
should give to such agencies, given that the answer to the second question depends
K. Horton (B)
University of Wollongong, Wollongong, NSW, Australia
e-mail: khorton@uow.edu.au
An earlier draft of this paper was discussed at the workshop which led to this book, Ethical
Questions for International NGOs. I would like to thank the participants at that workshop for their
comments on that draft, especially Chris Roche, who also gave me extensive written comments.
Thanks also to Emma Rooksby and Leif Wenar for their comments on that draft. All remaining
errors and shortcomings are of course entirely my own responsibility.


88 K. Horton
on the answer to the first. In order to make a minimally well-informed decision

about whether or not to give to aid agencies then, we do need to form at least some
sort of rough estimate about the effects that we have at least some good reason to
believe accurate.4
Unfortunately, though, it is very difficult for those of us who are not experts on
international aid to know what to think about this matter. I call this the ‘Epistemic
Problem’,5 and argue in Section 3 that it has certain unfortunate consequences.
Accordingly, in Section 4, I consider what might be done to reduce the size of that
problem—to make it easier, that is, for those of us who are not experts on aid to
form judgements about the effects that we have more reason to be confident about.
First, though, I will say something to support the claim that it is very difficult for
us to form such judgements.6
2 The Epistemic Problem
2.1 The Lack of Reliable Data and Expert Guidance

What might enable us to reach judgements about the effects that we have good rea-
son to believe accurate?7 Two sources of information would appear to be particularly
helpful: first, evaluations or assessments of the activities aid agencies perform that
attempt to measure at least their more significant effects and second, judgements
or estimates concerning the effects by independent experts in the field. Best of all
would be a combination of these two sources of information: an impartial assess-
ment of the typical effects of the different kinds of work that aid agencies do, by a set
of experts, making use of all the available evaluations, and filling in with informed
guesses where such data is lacking.
If such an assessment existed, based on sufficiently robust data, then it would not
be difficult for us to reach judgements about the effects that we had good reason to
believe accurate, for we could do so simply by accepting that assessment. Of course,
there would be no guarantee that assessment would be correct. In the real world,
there is always the possibility of error, particularly with regard to messy questions
like those concerning the effects of aid. But we could not do better than relying on
such an assessment. And as long as it were based on sufficiently good evidence, we
would have all the reason for confidence that it is reasonable to expect.
Unfortunately, though, no such assessment is available. Indeed, there is surpris-
ingly little of both of the sources of information just mentioned. To take evaluations
first, aid agencies themselves tend not to publish evaluations of their own activities,
or indeed any other reliable data on their own work.8 And there is no independent
body charged with evaluating the activities of aid agencies and reporting back to the
public on their findings. As a result, there is surprisingly little data on the effects of
the work aid agencies do available to the public.9
Furthermore, many of the evaluations of the work that aid agencies do that
have been done—whether publicly available or not—are by all accounts of very
The Epistemic Problem: Potential Solutions 89
poor quality. To quote from the most comprehensive review of evaluations of NGO
projects that has been conducted to date (Kruse et al. 1997, Section 9.4):10
a repeated and consistent conclusion drawn across countries and in relation to all clusters
of studies is that the data are exceptionally poor. There is a paucity of data and information
from which to draw conclusions about the impact of projects, about efficiency and effec-
tiveness, about sustainability, the gender and environmental impact of projects and their
contribution to strengthening democratic forces, institutions and organizations and build-
ing civil society. There is even less firm data with which to assess the impact of NGO
development interventions beyond discrete projects . . .
The data available, then, is disappointing in terms of both quantity and quality.
In this context, the role of experts becomes especially important. For given the lack
of reliable data, the best hope of reliable estimates about the effects would appear
to come from such experts—academics, consultants, and others who spend their
working lives studying NGOs and their activities. Unfortunately, though, the experts
seem unwilling to offer such estimates. Some do address the question of how good
or bad the effects are. Those who do so tend to focus mainly on emphasising the
lack of reliable data, though. They don’t then go on to ask the next question: ‘Given
this lack of reliable data, what is the most reasonable assumption to make about the
effects?’11
More broadly, specialists on NGO aid have provided surprisingly little guidance
to the general public about aid agencies and the work they do. The oddity of this sit-
uation is sometimes remarked on by such experts themselves. In 1996, for example,
Alan Fowler and Kees Biekart wrote that ‘a consumer’s guide to agencies, pro-
duced by an independent entity, is long overdue’ (Fowler and Biekart 1996, p. 131).
Absent such a guide, they continued, one is left with ‘a bizarre situation of agencies
competing in a donor market-place where buyers cannot reasonably compare the
products on offer or their relative value for money; they must still choose by an act
of faith’ (ibid.). In the intervening years, however, the situation has not substantially
changed.12
2.2 Does One Really Need Such Data or Guidance?

Those seeking to find out how good or bad the effects are cannot do so, then, simply
by turning to a reliable supply of data or to the judgements of experts in the field.
One might question whether such data or guidance is necessary, though. For there
are a number of strong reasons for thinking the effects are easily good enough to
imply that we should give which do not depend on such data or guidance.13 Given
that many of the people aid agencies work with lack even basic goods and services,
for one thing, it seems clear that such agencies are in a position to do a huge amount
of good for very little money. And this impression is reinforced by the information
such agencies supply, which often states or implies that even very small sums of
money can produce great benefits, such as saving lives. Such familiar considera-
tions, one might suggest, give us all the reason we need to believe that the effects
are good enough, despite the lack of supporting data and expert assurance.14 And
90 K. Horton
if that is correct, then it would not be difficult to arrive at a judgement about the
effects that one had sufficient reason to believe accurate after all. All that would be
necessary is to think through the implications of the kinds of familiar considerations
just cited.
Are such considerations strong enough to give us sufficient reason to assume
that the effects are good enough, though? They would be, I think, if there were no
other factors which provided significant reasons to doubt that assumption. If one
turns to the literature on aid agencies that has emerged in the last 20 years or so,
however, one finds a number of concerns about aid agencies and their work that
do, in my view, provide such reasons.15 And consequently, it seems to me that one
cannot simply assume that the effects are good enough on the basis of the kinds of
everyday considerations just sketched.
This part of my argument presents me with a major expository problem. To get
an adequate sense of the kinds of concerns I have in mind one has to engage with the
literature in question; any short summary is bound to be insufficient. On the other
hand though, it does not seem satisfactory merely to assert that there are concerns
significant enough to put the presumption that the effects are easily good enough in
serious doubt without giving any indication of what those concerns are. In response,
I do offer a relatively short summary of such concerns, while emphasising that it is
sketchy, incomplete, and much simplified, and providing references to the literature
for those who want to find out more.16
I should perhaps also emphasise that what follows is a sketch of some of the
concerns about aid agencies and their work that are raised most often in the
literature, rather than a balanced summary of both positive and negative com-
ments. I focus on such concerns because, as I have just said, I think that there
are very strong reasons to assume that the effects are easily good enough, and
that if one could simply make that assumption there would not be any significant
Epistemic Problem. In order to support my claim that there is such a problem,
then, I need to show that there are significant reasons to doubt that assumption.
And in order to do this, I need to show that there are sufficiently serious con-
cerns about aid agencies and the work they do. This, then, is why I focus in what
follows on concerns about aid agencies, rather than seeking to give a balanced
summary.17
2.3 Concerns About the Effects of the Work Aid Agencies Do
Aid agencies do a variety of different kinds of work. The most basic distinction
commonly made is between humanitarian aid, which is given in response to disas-
ters and emergencies, and development aid, which is given in other circumstances.18
The greater part of NGO aid is development aid, and the greater part of NGO devel-
opment aid, in turn, involves ‘the support or implementation of specific projects and
programs for particular groups of (poor) people and (poor) communities’ (Riddell
2007, p. 262).
Such projects and programs fall into several categories. One is ‘service
delivery’—the provision of basic goods and services to meet immediate needs for
food, health, housing, water, education, and so on. It was projects of this kind that
aid agencies first focused on when they began working in developing countries on a
significant scale in the 1950s.19 Initially, the hope was that such support would only
be necessary for a while, after which the recipients would be able to meet their own
needs. In many cases though, that was not how things turned out. As a result, the
concern grew that communities that received aid would be locked into an unending
cycle of dependency on foreign agencies.
Worse, many began to suspect that such forms of aid were actually undermin-
ing the forces necessary to put the poor in a position to meet their own needs.
Such concerns manifested themselves at a number of different levels. At the eco-
nomic level, supplying certain cheap or free foreign goods can depress the prices
that local producers of such goods are able to charge, and thus undermine local
production.20 At the institutional level, the intervention of relatively well-resourced
foreign NGOs can impede the development of local self-help organisations. And
at the political level, the provision of basic services by foreign NGOs can enable
government authorities to divert funds that they might have used on such services to
other, less salubrious purposes. It can also appease the poor, and thus reduce their
motivation to pressurise authorities for change (Lappe and Collins 1988, pp. 93–94,
Fowler and Biekart 1996, pp. 126–128, Biekart 1999, pp. 121–122).
In part in response to such concerns, some agencies began to supplement or
replace service delivery with other kinds of aid. One category consists of activities
aimed at enhancing the income-generating capacities of individuals and communi-
ties, often by teaching new skills or introducing new technology. This is the kind of
aid associated with the famous Chinese proverb, ‘Give a man a fish and you feed
him for a day; teach him to fish and you feed him for life’. The logic of such a devel-
opment is clear, but the results appear to have been disappointing. According to the
authors of the DAC Study (see Section 2.1), ‘INGO interventions are more success-
ful when focusing on social service provision and least successful when attempting
to promote economic and more productive-type projects’ (Kruse et al. 1997, Section
4.2). Indeed, they report that ‘there would appear to be unanimity among all the
reports which address this issue that whatever good individual projects do, they are
insufficient to enable the beneficiaries to escape from poverty’ (3.3.3).
Why such disappointing results? A variety of factors are often cited, many of
which concern features of the local social and political context. As David Korten
(1990, p. 120) puts it, with reference to the Chinese proverb:
The NGOs are prone to point out that villagers who live near the water already know a great
deal more about fishing than do the city kids that NGOs send out as community organizers.
The more substantial need is to insure the access of the poor to the fishing grounds and
markets that local elites control.21
Accordingly, some NGOs began to develop strategies aimed at increasing the

capacity of poor and marginalised people to fight for their rights, such as supporting
‘Southern’ NGOs and community based organisations in developing countries. The
92 K. Horton
lack of lack of reliable data on the effects of such ‘capacity-building’ work, however,
is especially acute. According to Riddell (2007, p. 283), ‘of all the different ways
that NGO aid is allocated and spent, least is known about the overall impact of
capacity-building initiatives’. Indeed, he continues, it ‘is extremely rare for NGOs
separately to assess even the direct impact of such activities, never mind the wider
effects on the lives of beneficiaries’ (ibid.).
It is not clear, then, how often such capacity-building initiatives achieve posi-
tive effects. We do know, however, that such work can do harm as well as good.
Funding local organisations can induce leaders to orient themselves more to those
who supply the funds than to their members, for example, and provide opportuni-
ties for corruption.22 And the risks involved in more confrontational political work
will be obvious enough. According to Ian Smillie (1995, p. 232), ‘Some NGOs have
taken terrible risks in the name of democracy and human rights. Workers have died
or been jailed, organizations have been crushed, lives ruined’.23
The record on of development projects, then, is unclear. The pervasive sense in
the literature, though, tends to be one of disappointment and frustration at best.24
The following is, I think, a fairly representative summing up of the general mood
(Black 2002, p. 37):
Anyone who has spent time on aid-financed projects knows that the process of interven-
ing usefully in how people conduct their lives is fraught with difficulty. It is hard enough
in one’s own community; how much harder in societies with quite different political, eco-
nomic, social and cultural dynamics. Neat constructs of success and failure—which aid
organizations dependent on support in donor countries are guilty of perpetrating—rarely
apply.
Indeed, some have come to believe that, as Michael Edwards and David Hulme
(1992, p. 14) put it, ‘small-scale NGO projects by themselves will never be enough to
secure lasting improvements in the lives of poor people’ (emphasis in original). This
is because the context in which such projects operate are shaped by unsupportive
policies and institutions at the national, regional, and global scale. This kind of
diagnosis has led some NGOs to develop activities aimed at changing such policies
and institutions, such as building up networks of local organisations in developing
countries, and advocacy, lobbying, and campaigning on issues such as trade and
debt in developed countries.
Aid agencies are more often praised than criticised for such ‘beyond-the-project’
work, though once again it is unclear how much impact it has.25 In any case, though,
few agencies apparently devote a large proportion of their budget to such activities at
the more structural end of the scale. According to Alan Fowler, for example (2000,
Chapters 19 and 18):
There is scant evidence to suggest that the bulk of NGDOs have substantially shifted their
operations towards redressing the structural or root causes of poverty and insecurity . . . .
Delivery of technical, social, and, increasingly, micro-financial services . . . . still forms the
primary weight of NGDO activity.26
Finally, a word on humanitarian aid: again, NGOs are often praised for their con-
tribution to such work, and there is no doubt that it saves many lives. Again, though,
experience has also shown that such work can lead to significant negative effects,
especially in conditions of conflict.27 Thus one problem concerns the leakage of
supplies. Agencies are sometimes forced to pay protection ‘taxes’ to get supplies
through to those who need it, for example. Or supplies may simply be looted. If so,
those supplies may then be sold or exchanged for arms. And in this way, attempts to
provide relief can help to fuel the very conflict they are a response to.28 Even when
supplies do get through, moreover, there can be similar unintended negative effects.
Resources that warring parties might have spent looking after their civilian popula-
tion will be freed up for military use, for example, if the international community
takes care of that population. And armed factions can use refugee camps as bases
from which to prepare future attacks, as happened around Cambodia in 1979 and
around Rwanda in 1994.29
Many commentators have argued that relief aid also tends to foster the kinds
of more long term and systemic negative effects associated with service delivery,
such as dependency, disincentives to local production, and the undermining of local
self-help organisations. More broadly, some critics have argued that the provision of
relief tends to undermine the development of the kinds of local social and political
arrangements that are necessary to provide secure, long-term solutions (De Waal
1997).
2.4 Concerns About Aid Agencies as Organisations
The concerns just sketched are discouraging, but shouldn’t of course be blown out
of proportion. It is one thing to say that NGO aid can have serious negative effects,
for one thing, and another to say that it often or typically does. And the lack of
evidence of success is not in itself evidence of failure. The kinds of concerns just
sketched do nevertheless provide reasons to think that aid agencies may need to be
highly capable, well-organised, and well-focused to find out which activities have
the best effects and ensure that those activities are executed effectively.
Do NGOs have these qualities? Again, it is hard to judge, for even less attention
has been given to the evaluation of NGOs as organisations than to the evaluation
of their activities,30 and to my knowledge none of the evaluations that have been
conducted are available publicly.31 There has been quite a lot of discussion of such
issues by aid specialists, however. Rather than providing reassurance, though, this
literature highlights a number of further concerns.
Thus one common complaint concerns the lack of attention that NGOs tend to
give to evaluating their own activities, and more broadly to learning and research
(Smillie 1999, pp. 21–31, Riddell 1999, Davies 2001, and Riddell 2007, pp. 265f).
According to Ian Smillie (1999, pp. 30–31), for example:
. . . many NGOs themselves believe that training, evaluation, research and publications are
luxuries . . . Too busy ‘getting the job done’, many have sacrificed corporate memory and
learning to the exigencies of time and place.
94 K. Horton
This raises some awkward questions. If NGOs do so little rigorous evaluation,32

how do they know what the effects of their activities are? And if they do not know
what the effects of their activities are, how can they ensure that they focus on the
activities that have the best effects?
Of course, evaluations and other forms of research cost money, and it will always
be difficult to spend money on research that could be spent directly on aid activities.
Given the risks of aid, however, the failure to conduct sufficient research may be a
false economy. And in any case, many commentators claim that such tendencies are
not due solely to considerations of cost. As Michael Edwards (1999, p. 211) puts
it, for example, ‘Development agencies often have an activist culture which sees
learning as a luxury when compared with the ‘real work’ of operations’.
Another common complaint concerns organisational and managerial problems.
As Chris Roche puts it (1999, p. 274):
Development agencies, including large NGOs, are not immune from the problems con-
fronting other bureaucracies in terms of complacency, hierarchy, inertia, and poor informa-
tion flow, which can lead to loops of self-deception as feedback from activities is distorted
or manipulated as individuals seek to protect themselves. Senior managers who subscribe
to the view that ‘if it ain’t broke, don’t fix it’ will tend to go along with positive news and
carry on without checks until something ‘breaks’. As Robert Chambers has suggested, this
kind of institutional blindness has led to the most remarkable feature of development efforts
over the last few decades: that ‘we’ were so wrong when ‘we’ thought that we were so right
(Chambers 1994).
Given the lack of probative data available on such agencies, it is hard to say
how widespread these tendencies are. One of the tendencies for which NGOs are
most often criticised is the failure to learn from experience, and the consequent
tendency to repeat the same errors again and again. In a set of recent evaluations of
the response to the 2004 Indian Ocean tsunami, for example, there is, to quote from
Riddell (2007, p. 346), ‘some extremely strong criticism, not only of inefficiencies
and mistakes, but of the repetition of mistakes made in earlier disasters, the lessons
of which do not appear to have been learnt, or acted upon effectively’ (see also
Edwards 1999 pp. 82–85; Minear 2002, pp. 46f, and Terry 2002, pp. 224 f.).
Given such problems—the neglect of rigorous evaluations, a culture that down-
grades learning, ‘institutional blindness’, the failure to learn from experience and
correct mistakes, all on top of the fact that aid is a difficult business and much can
go wrong—it no longer seems quite so straightforward to assume that NGOs will
find out which activities have the best effects and execute them effectively. As an
outsider, it is of course hard to know what to think about such matters, but it has to
be said that some apparently fair-minded commentators issue some pretty damning
verdicts. Here is one example (Fowler and Biekart 1996, p. 122):
Development, again, is not easy—why should it be? But professional, competent agencies
will surely have got this right after so many years of trying without evaluators having to tell
them what development criteria are? The truth is that, despite strong personal commitment
of staff, most agencies are not that professional or competent. They are just not found out.
Many commentators also accuse NGOs of a growing tendency to give more

attention to institutional ends (such as increasing ‘market share’) than to develop-
mental ends (such as alleviating poverty) (Edwards 1996, Biekart 1999, pp. 77–92,
Dichter 2003, Chapter 6; cf. De Waal 1997, p. 66). Given the lack of reliable infor-
mation on the work aid agencies do, an agency’s ability to raise funds is likely
to depend less on how effective it is than on such factors as maintaining a good
reputation, keeping up its public profile, and putting across a message that is attrac-
tive to potential contributors. And this can lead to several negative tendencies. For
one thing, such pressures can distort what activities NGOs focus on. Agencies may
focus on activities that attract the most publicity, for example, or that are espe-
cially appealing to contributors, even if those are not the activities that would
do the most good. And indeed, many commentators complain of a tendency for
NGOs to make sure they have a presence in places where they are most likely to
get media attention, such as well-publicised disasters, whether or not that is the
best use of their resources (Rieff 1995: p. 7, De Waal 1997, p. 80, Vaux 2001,
pp. 29–30).
Again, it should be acknowledged that aid agencies are in a difficult situation.
They need to appeal to a fickle public for funds and other forms of support,33 and so
it will be tempting to make certain compromises. Further, any losses resulting from
such self-publicising measures might be more than made up if those measures lead
to increased donations. To the extent that different agencies are merely competing
with one another for the same funds, however, the result will merely be that less
scrupulous agencies outcompete more scrupulous ones.34 And the competitive ethos
that results appears to reinforce a number of other negative tendencies that are often
discussed in the literature, such as the failure of different agencies to cooperate with
one another or coordinate their activities, even when doing so would promote their
shared ends (Clark 1991, p. 65, Smillie 1995, p. 233, Edwards 1999, p. 105, Biekart
1999, p. 299, Lindenberg and Bryant 2001, pp. 75–76, Minear 2002, Chapter 2, and
Riddell 2007, pp. 338–339).
Such institutional and motivational problems are reinforced, moreover, by the
fact that many agencies have switched from devising, running, and staffing their own
activities to working with and through local organisations in developing countries.
As a consequence, there is often a whole chain of different organisations involved in
any particular project. And this provides further opportunities for institutional and
personal pressures to create mischief. As Stuart Carr (1998, p. 49) and his colleagues
put it:
the explicit agenda at each point may be that of poverty alleviation, but . . . more ‘personal’
agendas such as the organization’s survival or an individual’s desire for promotion can take
precedence at any particular point in the chain.
Again, it is hard for outsiders to judge how widespread and severe such institu-
tional problems are. But again, some apparently reliable commentators make some
worrying claims. David Hulme and Michael Edwards (1997, p. 282), for example,
write that ‘the focus of attention has shifted away from empowering others, for inde-
pendent action, towards empowering NGO leaderships and serving the organisation
and its own ends’.
96 K. Horton
2.5 Taking Stock

The material summarised in Sections 2.3 and 2.4 makes pretty grim reading. With
that it mind, it is important to emphasise the context. My main aim in this section
is to support the claim that it is very difficult for those of us who are not experts on
international aid to arrive at a judgement about the effects of the work aid agencies
do that we have good reason to believe correct. In Section 2.1, I suggested that
this claim would be false if an impartial assessment of the effects were available,
made by independent experts and based on robust data. For in that case one could
arrive at a suitable judgement about the effects simply by accepting that assessment.
Unfortunately, though, no such assessment is available.
In Section 2.2, I turned to another factor that might make the claim that it is very
difficult to arrive at such a judgement false. I pointed out that there are a number
of familiar, everyday considerations that give one strong reasons to assume that the
effects are easily good enough—so strong, I suggested, that it would be reasonable
simply to make that assumption if there were no other factors which provided sig-
nificant reasons to doubt it. And if that were so, then again it would not be difficult
to arrive at a judgement about the effects that one had sufficient reason to believe
accurate, for it is not difficult to follow such an argument.
I then claimed, however, that the literature on NGOs reveals concerns that do
provide significant reasons to doubt that assumption. The material in Sections 2.3
and 2.4 constitutes my summary of those concerns. If such concerns do provide
significant reasons to doubt that the effects are good enough, then one cannot after
all simply make that assumption on the basis of the kinds of everyday considera-
tions sketched in Section 2.2. It doesn’t follow, of course, that assumption is false.
In order to assess any such substantive claim, one would need to consider what
might be said in response to those concerns. Given that here I have focused exclu-
sively on the negative side, moreover, I would like to emphasise that there are a
number of forceful responses that one might appeal to.35 I do not know of any
such response that provides a quick, decisive argument that the effects are good
enough, however. Instead, those who seek to find out how good or bad the effects
have to try to weigh up a complex variety of positive and negative considerations.36
This is a very difficult task, especially given the lack of reliable data and expert
guidance. And that, to repeat, is the claim that I was seeking to support in this
section.
3 Why the Epistemic Problem Is a Problem
I shall take it, then, that it is indeed very difficult for those of us who are not experts
on international aid to find out if the effects are good enough to imply that we should
give to them. And as I said in Section 1, I am calling the fact that it is so difficult for
us to do so the ‘Epistemic Problem’. Why call this fact a problem? In part, because
it follows that it is also difficult for us to determine whether or not we should give
to such agencies. For if it is difficult for us to find out whether the effects are good
enough to imply that we should give to aid agencies, then of course it will also be
difficult for us to determine whether or not we should give to such agencies.
Might one question the assumption that we need to find out how good or bad the
effects are in order to determine whether we should give? Not really. As I said in
Section 1, whether or not we should give to aid agencies clearly depends on how
good or bad the effects of their work are. This is easiest to see if one focuses on
the worst-case scenario—that aid agencies do more harm than good. For surely it
would not be the case that we should give to them if this were so. The effects must
have to come up to some standard, then, in order for it to be the case that we should
give. There are likely to be different opinions about what that standard is, and as I
said above I will leave that question open here. Whatever that standard is, though, if
we unable to find out whether the effects meet it, we will also be unable to find out
whether we should give.37
The fact that it is difficult for us to find out whether the effects are good enough,
then, means that it is also difficult for us to determine whether we should give to
such agencies. And this is awkward, given that none of us can avoid the choice of
whether to give to such agencies or not. Of course, to say that it is difficult to find
out whether the effects are good enough is not to say that it is impossible to do so.
The more difficult it is to do so, though, the more likely it is that even those who
make a serious attempt to do so will fail, and that others will be discouraged from
making such an attempt.
As long as it is seriously unclear to us whether the effects are good enough,
moreover, we are unlikely to give to such agencies, or at least to give substantially.
For in view of the sacrifices that doing so would impose, and the variety of other
morally important initiatives that compete with aid agencies for our money, we are
likely to hold back unless we are fairly confident that the effects are good enough.
Whether this is a good or a bad thing depends, of course, on whether the effects are
in fact good enough. If they are not, then it may be no disaster if the difficulty of
finding out whether the effects are good enough leads us not to give. If the effects
are good enough, though, matters would be much more serious. For in that case, the
difficulty of finding out how good the effects are would result in our failing to do
what we should do. This would be bad from our point of view, at least if we care
about whether or not we do what we should. And it would be bad for the global
poor, for it would mean that they would be getting less support than they should be
getting.
Of course, some people would not give no matter how easy it was to find out
that the effects were good enough.38 As long as there are some people who would
give, or give more, if it were easier to find out that the effects were good enough,
though, aid agencies would receive more money if it were easier to find out that the
effects were good enough. And it seems hard to deny that there are some people
of this sort. The most common reasons people give for not giving to aid agencies,
in my experience, tend to be worries of one sort or another about the effects.39 In
some cases, no doubt, such reasons function merely as an excuse, but it would seem
sceptical to the point of cynicism to suggest that is so in all cases. If it were easier to
98 K. Horton
find out that the effects were good enough, then, it would be easier to see that such
worries do not undermine the case for giving. And if it were easier to see this, then
some people who currently do not give because of such worries would start to give,
or give more.
At the workshop this paper was written for, I should acknowledge, there was
quite a lot of scepticism about this claim (that donations to aid agencies would go
up if it were easier to find out that the effects were good enough) from people with
NGO experience. The general view seemed to be that those who give to NGOs
do so mostly for self-regarding reasons—in order to feel good, for example, or to
assuage guilt—and that therefore they tend not to be much interested in the effects
of the work aid agencies do.40 I do not think that such suspicions undermine the
claim in question, though. For one thing, giving to aid agencies would presumably
not make one feel good (or be effective in assuaging guilt) unless one believed
that their work had generally good effects. If it were easier to see that this claim
were true, then, more people would believe it, and thus more people would be in
a position to feel good (or assuage guilt) by giving to such agencies. If it were
easier to find out that the effects were good enough, then, it seems likely that more
people who are responsive to such motives would give. Even if such motives are
very common, moreover, it seems reasonable to suppose that there are nonetheless
significant numbers of other potential contributors who are more concerned about
the effects, and who would therefore be more likely to give if it were easier to find
out that the effects were good enough.41
It seems likely, then, that donations would go up if it were easier to find out that
the effects were good enough. And this would benefit the global poor, given the
assumption that the effects are good enough, for that assumption would presumably
hold only if the effects of the work aid agencies do were at least very good on
the whole. Given that assumption, then, the Epistemic Problem is a very bad thing
indeed, for it would be preventing the global poor from getting the support that they
would be getting if it were not so difficult to find out that the effects were good
enough.
Is that assumption true, though? Are the effects good enough? As a non-expert
it is of course difficult for me to say—that is just one instantiation of the Epistemic
Problem. Though I will not argue the point here, though, I think they probably are.42
And as I have just said, given this assumption, and the assumption that donations
would increase if it were easier to find out that the effects are good enough, the
Epistemic Problem is a very bad thing. Those who share those assumptions, then,
will agree that it is an urgent matter to consider what could be done to make it easier
for potential contributors to find out how good or bad the effects are.
What if one does not share one or both of those assumptions, though? Would
there then be no strong reason to seek out ways to make it easier to find out how good
or bad the effects are? Not necessarily. Whether those assumptions are true or false,
doing so would still make it easier for people to find out what they should do, which
is a worthwhile goal. Furthermore, doing so may also have good consequences for
the global poor, even if the effects are not good enough. Consider, in particular, the
possibility that the effects are very bad—that such agencies (or a subset of such
agencies) do more harm than good, for example. Making it easier for people to find
that out would also be good for the global poor. For those who currently give to
aid agencies would presumably not do so if they knew that such agencies did more
harm than good.43 If it were easier to find out how bad the effects were, then, at
least some of these people would stop giving, and this would again be good for the
global poor, for with less money those agencies would presumably do less harm.
On a number of different assumptions, then, there are strong reasons to see if
there are any measures that could be taken that would make it easier for ordinary
people to find out if the effects are good enough. In the remainder of this paper,
consequently, I discuss a number of potential measures of this kind.
4 Potential Solutions
4.1 NGOs Could Make More Data on Their Own Activities

Available
One thing NGOs could do is to make more evaluations of their own work pub-
licly available, for example by putting them on their websites. Doing so would give
potential contributors more information about the activities of such NGOs than they
have now, and might also be taken as an encouraging sign of a willingness to be
more open about their work. These are important merits. At the same time though,
it is not clear whether doing so would help much with the Epistemic Problem. For
one thing, such evaluations may not give potential contributors much of an idea
about the real impact of the activities evaluated, for they tend to be written for insid-
ers in a style full of technical jargon, and many do not even contain estimates of the
effects of the activities evaluated.
One response to these problems would be for NGOs to publish not (only) eval-
uations themselves, but summaries or interpretations of those evaluations, written
in a style accessible to non-experts. Ideally, such summaries would also contain
estimates of at least the more significant effects of the activities in question.44
This suggestion highlights another obvious problem, however, concerning whether
potential contributors would have sufficient reason to trust any such material NGOs
made available, whether evaluations themselves, or interpretations of those evalu-
ations, for NGOs may of course be tempted to select and interpret the available
evaluations in ways that give a rosy view of their activities. And it will be hard for
outsiders to tell whether or not this has happened if there is no independent body
advising them about such matters.
Even if these worries were put aside, moreover, making such material available
may still be insufficient to deal with the Epistemic Problem. For what the potential
contributor presumably requires, in order to reach a well-informed decision about
whether or not to give to a certain NGO, is some idea of the effects of that NGOs
activities as a whole, or at least of the typical or average effects of such activi-
ties. And it is far from clear whether reading interpretations or summaries of stray
evaluations of this or that activity would enable one to form such an idea.
100 K. Horton
The obvious response to this problem would be for the agency, or some other
body, to put together a synthesis or meta-evaluation of all of the agency’s evalu-
ations, or of a representative sample of those evaluations. Might this be enough
to give one a tolerably clear idea of how good or bad the effects are? That would
depend on a number of factors. One would be whether the NGO in question conducts
or commissions enough evaluations of sufficient quality to enable any such conclu-
sion to be drawn with any confidence. Another would concern the type of activities
the NGO in question did. It might be relatively straightforward for agencies that
focus on just one type of activity to produce a meaningful synthesis, especially if
the activity in question were of a kind that lends itself relatively easily to measure-
ment. It would be much harder for agencies that do a range of different activities,
especially if some of those activities are of a kind that is hard to evaluate.45 And a
third factor would concern the objectivity of such reports, even if they are conducted
by independent consultants (Do critical consultants get invited back?).
For all of these reasons, it is not clear how much impact taking such measures—
publishing more evaluations, or summaries and syntheses of evaluations—would
have on the Epistemic Problem. There are also a number of reasons why NGOs may
have reservations about publishing such information on a significant scale. Doing
so would use resources that could be used on aid activities, for one thing. This
would be unfortunate in itself, and may also put off those potential contributors
who disapprove of spending on ‘overheads’. Of course, if it were merely a matter
of publishing or interpreting or synthesising evaluations that the agency already
had available, then the resources used would presumably be fairly small. If new
evaluations had to be commissioned, however, then the exercise could become much
more expensive.
Even if the exercise were relatively cheap, moreover, it would nonetheless con-
stitute a salient example of resources not being spent on aid activities, which would
not go down well with some contributors.46 Unless they were heavily massaged, for
example, such evaluations would be likely to reveal more about the negative aspects
of aid than many agencies have been ready to admit to publicly up to now,47 and
NGOs might fear that those negative aspects would loom larger in people’s minds
than any positive findings.48
As a result of these factors, agencies that took such measures might lose out
in relation to agencies that did not. The obvious way to prevent such unfortunate
consequences is to introduce forms of regulation that give all agencies an effec-
tive motive to follow the same measures.49 Though this may prevent opportunistic
agencies from profiting at the expense of their more scrupulous fellows, however,
there remains the danger that all agencies, or aid agencies taken as a whole, might
receive less funds as a result of taking such measures. For all agencies would then
be spending more money on such measures that could be spent directly on aid
activities, and revealing more about the problems of aid, and for the reasons given
above this might be off-putting to some potential contributors. There is also the
danger that the media would tend to focus on any negative findings, rather than
seeking to give a balanced view, which would help to foster an unfairly negative
view of the work NGOs do.50 And similarly, organisations ideologically opposed to
NGOs might also pounce on any negative findings and use them in their campaigns
against NGOs.
Of course, if publishing more evaluations and related material made a significant
impact on the Epistemic Problem, and if in particular doing so provided strong evi-
dence that the effects were good enough, then for the reasons given in Section 3
any reductions in donations resulting from such factors might be more than com-
pensated for by extra donations given by people willing to give, or to give more, as
a result of this new evidence being available. NGOs that are unsure whether these
conditions are met, though, or that doubt whether making it easier for people to find
out that the effects are good enough would lead them to give more, may believe that
the net effects on donations of a policy of publishing more evaluations and related
material would be negative. And if so, this may lead them to resist such a policy,
whatever its merits.
A policy of publishing evaluations might also have other negative consequences.
If NGOs know that their evaluations will not be published, they need not be inhibited
from evaluating their own work critically, or from focusing on problem activities.51
If they know that their evaluations will be published, by contrast, they might be
tempted to try to ensure that the only projects that are evaluated are the ones most
likely to be judged successful, and that all projects are evaluated in a ‘friendly’
manner.
A policy of publishing more evaluations, then, might have negative conse-
quences. It is difficult to assess how significant these consequences would be, but in
any case the risk of such consequences may lead NGOs to think twice about adopt-
ing such a policy. Given that it is also unclear how much impact adopting such a
policy would have on the Epistemic Problem, it seems a good idea to turn to other
responses to that problem.
4.2 An Independent Body Could Be Set Up to Evaluate the Work

of NGOs and Report Back to Contributors
As I have said, one reason why the kinds of measures discussed in Section 4.1 might
be insufficient to solve the Epistemic Problem is that contributors may be unsure
whether to trust data supplied by NGOs. The obvious solution to this problem is to
set up a body of experts to advise contributors, a body that is genuinely independent
of NGOs, and (as far as possible) free from any other pressures that might distract
it from the task of providing clear and honest advice to the general public. If such a
body were able to access or collect sufficient data to support well-based assessments
about the effects of the work aid agencies do, it could then convey those assessments
to contributors, and non-experts could simply trust those assessments. And thus, as
I said in Section 2.1, the Epistemic Problem would be solved.
One could imagine more or less ambitious versions of such a body. The most
ambitious would be one with sufficient resources to do its own evaluations of the
activities of NGOs. Despite calls from aid specialists from time to time for such
102 K. Horton
a body,52 however, few steps in this direction have been made.53 Are significant
further steps in this direction a realistic prospect? One major issue would be funding.
It would not be cheap to set up such a body and finance its evaluations, and who
would pay? Contributors may generally be unwilling to fund such a body, given the
widespread preference that all their money is spent on projects. NGOs themselves
might have mixed attitudes, in part for the reasons given below. The best hope, then,
may be UN or other intergovernmental bodies, or private foundations.
If funding could be provided, then the other major issue would be that of get-
ting permission from NGOs to evaluate their activities and report back to the public
on their findings, for there are a number of reasons why NGOs might feel uncom-
fortable about such a process. One obvious reason would be the fear of a negative
report. As far as this consideration goes, one would expect NGOs’ decisions about
whether to sign up to such a scheme to be determined by the degree of confidence
they have about their work. If that is right, then a two-tier system might develop,
with some NGOs taking part in the process and others not. And on the face of it,
such a two-tier system would not be a bad thing, as long as the NGOs that signed
up tended to be the ones whose activities had the best effects. It would be good for
contributors, keen to give in ways that would do the most good, for they could direct
their funds to those NGOs that were confident enough to take part in the process and
which received positive reports. And it would be good for the system as a whole,
as the more effective NGOs would presumably receive more funds while the less
effective would receive less.
Matters may not be quite as simple as that, though, for there are also a number of
reasons why even the more confident NGOs might have reservations about taking
part in such a process. Even if the overall findings of such a review were positive,
for one thing, it would again inevitably reveal more about the negative aspects of
aid than most NGOs have been willing to be open about publicly so far. And as I
said in Section 4.1, this might have a number of consequences that NGOs would not
welcome. Despite the competitive tendencies mentioned in Section 2.2, moreover,
it appears that NGOs will go to great lengths to present a united front in public, and
this too may discourage them from a process that would lead to a two-tier system.
Again, if the findings of such a review were sufficient (in combination with other
sources of evidence) to show that the effects were good enough, then any such deter-
rent effects on donations might be more than compensated for by new donations. I
discussed this issue in Sections 3 and 4.1 and will not do so again here. For now, I
shall just say that if NGOs suspect that the kind of process I am discussing would
lead to fewer donations they might refuse to sign up to it, which would undermine
its feasibility.
Setting up a body which conducted evaluations of NGO activities and reported
back to the general public on the findings would in effect give a greater impor-
tance to evaluations, and this might provide further reasons why some NGOs would
resist it., for there are a number of concerns about giving evaluations such a signif-
icant role (Riddell 1999, pp. 223–237; chapter “Whose Impact, and Is It All About
Impact?” by Isbister, this volume). One such concern is that evaluations may lead
to a temptation to focus more on achieving the kinds of indicators that will ensure
a good evaluation than on what would really do most good. Relatedly, a focus on
evaluations might affect the kind of work that gets funded, and consequently the
kind of work that aid agencies tend to concentrate on, in negative ways. There may
be a temptation to focus on activities that lend themselves better to evidence of
concrete results, for example, rather than less tangible work that may in fact be
more important.54 And a further concern that is often mentioned is that focusing on
evaluation may engender moves away from innovation and experiment.
This issue is further complicated by the fact, noted in Section 2.2, that rather
than implementing projects themselves, many Northern NGOs now work largely
or wholly through local ‘partners’, such as Southern NGOs and Community Based
Organisations. In order to evaluate the activities of such a Northern NGO, then,
one would presumably need to evaluate the work of its local partners. This, how-
ever, raises thorny issues concerning surveillance and lack of trust. More broadly,
there is what Alan Fowler (2000, pp. 23–24) calls the ‘sustainability-accountability
paradox’:
On the one hand, sustainability requires progressive integration of the products and effects
of interventions into the ongoing processes of economic, social, political and cultural life
that surround them . . . . On the other hand, to demonstrate and be accountable for perfor-
mance, the effects of NGDO interventions must be visible and attributable. In attempting
to make this possible—to satisfy donors—NGDOs often deliberately or unconsciously
“ring-fence” projects to show differences resulting from specific inputs. Where this phe-
nomenon occurs—irrespective of the “integration” that project documents call for—it acts
as a constraint to linking, embedding and sustainability.
Such factors, then, provide another set of reasons why some NGOs might resist
having their work evaluated in the kind of way I am considering here. Are they
sufficiently strong reasons to justify refusing to do so? As a non-expert, it is again
difficult for me to judge, but I have to say that I am somewhat sceptical. For one
thing, it seems reasonable to hope that ways of conducting evaluations could be
developed which would reduce the kinds of risks just sketched to manageable levels.
If so, then those risks would not present very strong objections to giving evaluations
a greater role. If not, then as far as I can see the issue would depend largely on
whether it is possible to gain sufficient knowledge of the effects of aid activities
without conducting such evaluations. To the extent that it is possible to do so, it may
be better not to do more evaluations, if the disadvantages of conducting such evalu-
ations are sufficiently serious. To the extent that it is not possible to do so, however,
it would seem hard to justify not conducting more evaluations, even if doing so had
the kinds of unfortunate consequences just sketched, for it seems highly irresponsi-
ble to conduct activities with such vulnerable people if one really has little reliable
evidence about the effects of those activities.55 This is mainly due to the risk of sig-
nificant negative effects. Good intentions would provide no excuse for conducting
activities that had such effects, if all reasonable steps had not been taken to find
out how great the risks of such effects were. And if (more) evaluations would have
helped, but have been resisted, then this condition would not be met.56
104 K. Horton
In any case, whatever the rights and wrongs are, concerns about the potential
negative consequences of giving a greater role to evaluations may provide fur-
ther reasons why some NGOs would resist taking part in the kind of process I am
considering here, and thus undermine its feasibility.57
4.3 Less Ambitious Bodies
If it would not be feasible to set up the kind of body described in Section 4.2—and
at least in the meantime, while such a body does not exist—one might consider less
ambitious alternatives. One such alternative would be a body with a similar remit
of reporting to the general public about the work of NGOs, but which restricted
itself to data that was already available, rather than conducting its own evaluations.
Such a body would of course be a great deal cheaper than one that also conducted
evaluations. Could it still make a significant impact on the Epistemic Problem? That
would depend on the degree to which there is already enough reliable data on the
work aid agencies do (whether publicly available or not) to enable such a body
to come to well-based judgements about the effects of their work. To the degree to
which there is enough data, we would already be tantalisingly close to a solution to
the Epistemic Problem. All that would be necessary is for a suitable group of experts
to get access to that data, draw relevant judgements based on it, and then pass on
those judgements to the general public.
Is there already enough data of this kind? As an outsider it is again difficult to
judge, and the issue is complicated both by the fact that it concerns a matter of
degree, rather than an all or nothing matter, and that the answer might be different
for different forms of aid. At the time the DAC Study was produced (in 1997), the
authors evidently didn’t feel that they had sufficient reliable data to make any judge-
ment about the overall effects of the development work NGOs do. And though more
attention has been given to evaluation over the last 10 years, it still seems unlikely
that there is currently sufficient data for such a body to make such a judgement
with any confidence (Riddell 2007, Chapters 16–19). Even if this is right, though,
there might be sufficient data for such a body to offer at least certain kinds of esti-
mates about certain kinds of aid. There might be some NGOs which already ensure
that their own activities are evaluated with sufficient rigour and objectivity to allow
those who know how to interpret such data to make reasonably reliable estimates of
their effects, for example. In such cases, an independent body might merely have to
play the role of an honest broker between these NGOs and the public, interpreting
the data in question and reassuring the public of its reliability. And similarly, there
might already be enough data about certain sectors or types of activity for such a
body to offer reasonably confident judgements concerning them.58
Even such partial information might be of decisive import to potential contribu-
tors, if such a body were able to report with sufficient confidence that at least certain
forms of aid had effects that were judged to be good enough. For as I said in Section
3, it might follow that individuals should fund that particular form, whether or not
they should give to aid agencies in general. And if so, this would solve the key prac-
tical question for potential contributors: ‘Is there something in this area that it is
relatively clear I ought to do?’59
Publishing just such information might also have other positive consequences.
If the body in question were able to identify certain forms of aid had relatively
good effects, then the general public would presumably respond by giving more
to those forms, which on the face of it at least would be a good thing.60 It would
also reward those agencies that have conducted or commissioned evaluations of their
own activities rigorous and objective enough to enable an independent body to make
reasonably confident judgements about the effects of their work. And it would also
provide an incentive to other NGOs to follow their example.
What about negative consequences? A number of the considerations from
Section 4.2 would be relevant again. The effect of setting up such a body would be
to give evaluations more importance, and this may have negative effects as well as
positive ones. It is also possible that publishing more information about the effects
would deter some contributors from giving, even if that information showed that at
least certain forms of aid had effects that were good enough. For the reasons given
in Section 4.2, though, such concerns do not seem to me to give sufficient reason not
to go ahead, at least absent any alternative serious attempt to tackle the Epistemic
Problem. Even if this is correct, though, such factors might again provide some
reasons why some NGOs would think twice about signing up.61
The body I have had in mind so far in this section, like that considered in Section
4.2, would aim to provide as comprehensive information about NGOs and the effects
of their work as possible, though of course it might have to be much more selec-
tive in practice, if few NGOs signed up, or if there were insufficient data to enable
such a body to form judgements about certain forms of aid. An even less ambitious
alternative would be a body that did not even aim to provide such comprehensive
information, but focused instead on sifting through the available data in order to
find forms of aid that it was able to say with some confidence have good enough
effects, in order to alert the general public to their existence. As long as the reports
of such a body were sufficiently reliable, they too would be able to resolve the
key practical question for potential contributors, and could therefore make a big
difference.
Interestingly, certain potential models for such a body have recently emerged in
the world of philanthropy, in response to a more uptown version of the Epistemic
Problem. The Economist (2006) quotes Gavyn Davies, ex-Goldman Sachs banker,
as saying that when he and a colleague decided they wanted to do some philan-
thropy, ‘We found there wasn’t enough information produced in a hard-headed,
independent, high quality way, made available to all’. In response, they created
an organisation called New Philanthropy Capital, with the remit of producing
such information, and in particular information about effective organisations and
projects/alerting philanthropists about effective organisations and projects. Their
hope was the same as the one that informs this paper: that providing such infor-
mation might lead to more donations, both in total, and especially to the best
organisations. As Davies puts it (ibid.):
106 K. Horton
This was an investment designed to have a levered effects on other people’s giving. We
wanted to increase giving by enabling donors to be more confident that they were having an
impact on people’s lives.
Perhaps other philanthropists, or foundations, or other agents or agencies with

sufficient resources might consider setting up similar organisations for people who
lack the resources to be philanthropists on any significant scale.62,63
4.4 Initiatives to Improve Accountability and Standards
In recent years, a number of initiatives have been set up by groups of NGOs

to improve accountability and standards in the field of humanitarian aid. These
include Sphere, which has developed a set of ‘Minimum Standards in Disaster
Response’ that ‘sets out for the first time what people affected by disasters have
a right to expect from humanitarian assistance’; the Active Learning Network
for Accountability and Performance in Humanitarian Action (ALNAP), which is
‘dedicated to improving humanitarian performance through increased learning and
accountability’; and the Humanitarian Accountability Partnership (HAP), which
aims ‘to make humanitarian action accountable to its intended beneficiaries through
self-regulation, compliance verification and quality assurance certification’.64
In principle at least, such initiatives might be able to have a significant impact on
the Epistemic Problem. One way in which could do so simply is by increasing the
quality or quantity of evaluations available in some form to the general public.65 But
the standards developed by certain of these initiatives might also be able to provide
a useful or alternative or supplement to such evaluations. If, in particular, following
such standards were reliably correlated with achieving good enough effects, then by
checking whether the relevant activities have met those standards it may be possi-
ble to judge whether those activities have achieved such effects without having to
conduct the kind of evaluations that attempt to measure the effects more directly.66
Given the current undersupply of rigorous evaluations,67 and the apparent obstacles
to conducting more of them, such a substitute or supplement may prove very useful
indeed.
In practice, though, it is less clear how much of an impact such initiatives might
have on the Epistemic Problem. By its own account, ALNAP seems to have had little
success so far in improving the quality of evaluations so far.68 The development of
auditable standards is currently in its infancy even in the field of humanitarian aid,
moreover, and there is to my knowledge nothing comparable in relation to other
forms of aid.69 Furthermore, it is unclear whether following such standards is in
fact sufficiently strongly correlated with achieving good enough effects as to be able
constitute a substitute for more standard evaluations. Those who are not experts on
aid will not of course be in a position to judge this matter, and there appears to be
little consensus among the experts about such issues concerning accountability and
evaluation.70
It is unclear at present, then, how much impact such initiatives might have on the
Epistemic Problem. Nevertheless, they are surely positive development. As long as
the relevant standards are suitably gauged, the fact that an NGO had met those stan-
dards would certainly give some reason for reassurance to contributors, whether not
doing so is able on its own to provide sufficient evidence that the effects are good
enough. The fact that these initiatives have been set up by NGOs themselves is,
moreover, an encouraging sign that (at least some) NGOs are serious about improv-
ing their practice, and also an indication that such initiatives may encounter fewer
feasibility problems than some of the other measures considered in this section.71
Such initiatives, then, seem to offer some hope for the future whatever their impact
on the Epistemic Problem now.
4.5 Experts Can Simply Offer Their Personal Judgements

The final way I will consider here to make it easier for non-experts to find out how
good or bad the effects are is simply for those with relevant expertise to offer their
judgements in print about this matter. Potential contributors may not be able to rely
on such judgements as confidently as they could rely on the findings of the kinds
of bodies considered in Sections 4.2 and 4.3, for while it would presumably be
clear that such findings were based on a thorough, impartial review of the available
evidence, it would be harder to determine how this or that expert had arrived at their
judgements. In the absence of any such body, however, or any other way of making
it easier for non-experts to find out how good or bad the effects are, such informal
judgements made by experts might be a lot better than nothing. Indeed, if such
judgements were sufficiently positive, and were taken to be sufficiently reliable,
then they might again be decisive for certain potential contributors, willing to give
only if they receive sufficient assurance about the effects.
On the conception of expertise that I have in mind here, one is an expert in a
certain field to the degree that one is well-placed to make reliable judgements about
that field. The relevant judgements here are generalisations about the effects of NGO
aid, and so the experts would be those well-placed to make such generalisations.
And that would typically reflect such factors as relevant training and experience,
familiarity with the relevant literature and data, as well more general traits such as
intelligence, insight, and so on.72
Chris Roche asked me why I focus particularly on the judgements of experts
here. Can’t one also draw on the views of other people who may not qualify as
experts, such as the recipients of aid, and aid workers with limited experience and
knowledge? In reply, I would say that such views may indeed be helpful, but much
less helpful for current purposes than the judgements of experts. Recipients and aid
workers would be able to offer their views about the particular activity or activities
they had experience of. It would be rash, though, to base a judgement about the
effects on such accounts. It would be hard to tell whether such views are represen-
tative, for one thing. And in addition, such people may not know much about the
108 K. Horton
long-term or systemic effects of the relevant activities, or about their effects on those
apart from recipients.
In order to arrive at an overall view that one has much reason for confidence
in, then, one would at the least need to take account of a wide range of such per-
sonal accounts, and also of whatever data is available on non-immediate effects.
And this would be a time-consuming and demanding task. What I am seeking here
is precisely a way in which non-experts can form judgements about the effects
that they have some reason for confidence in without having to take on such a
task. Even those who are willing to do so, moreover, are presumably less likely
to reach a correct judgement than those who have years of experience and train-
ing in the relevant field, as well as a thorough grip on the relevant data—that is,
the experts.
If experts were willing to offer their judgements, then, that could save non-
experts a lot of labour, as well as providing judgements that would probably be
more reliable than anything non-experts could come up with on their own. As I said
in Section 2.1, though, experts seem very reluctant to do so.73 Why might this be?
A variety of factors may have some influence. Like experts in many other fields,
aid experts may simply tend to prefer communicating with other experts than with
the general public. Some may not realise the full extent of the Epistemic Problem,
or its unfortunate consequences, and thus not appreciate how helpful to non-experts
such judgements would be. And some may think that it is more important to high-
light problems with aid, in an attempt to prompt reform, than to attempt estimates
concerning the effects.74
None of these factors constitutes reasons of principle not to offer such judge-
ments, and so none of them would prevent experts who did appreciate the impor-
tance of offering such judgements from doing so. Are there any other factors that
do constitute such reasons? One obvious candidate would simply be the fact that
experts too find it very difficult to make such judgements with much reason for con-
fidence, given the complexity of the issues, the variability of the effects in different
circumstances, and the lack of reliable data. The decisive response to this, however,
is to point out that however difficult experts might find it to make such judgements,
it would still be easier for them to do so than for those of us who are not experts.
Given that we need such judgements in order to determine whether we should give,
then, it would still be very helpful if they told us—tentatively, perhaps, with what-
ever qualifications are necessary—what they think.75 Even if the experts felt unable
to offer a judgement about the effects of the work aid agencies do as a whole, more-
over, it seems wildly implausible that none of them is able to offer any informed
guesstimate about any form of aid.
The collective failure by aid specialists to offer any such help to potential
contributors—ideally, by putting together some kind of review or handbook, but
failing that by simply offering their own judgements—seems to me a serious fail-
ing. Since they are the people best-placed to make the relevant estimates, such
experts are uniquely able to play the key intermediary role of helping non-experts to
form judgements about the effects that are more reliable than any they could form
on their own. Given that non-experts need to make such judgements in order to
make well-informed decisions about whether to give to aid agencies, and given the
potential consequences of making the wrong decision (as set out in Section 3 above),
this role is an absolutely crucial one. So far, though, the experts have failed to engage
with it.
Given certain not entirely implausible premises, the consequences of such neglect
would be momentous indeed. If, in particular, we (non-experts) should not give to
aid agencies unless we have sufficient reason to believe that the effects are good
enough, it is impossible for us to arrive at sufficient reason to believe this on our
own, those effects (or at least the effects of certain forms of aid) are in fact good
enough, it is possible for aid experts to find this out (to the appropriate degree of
confidence), and we would have sufficient reason to believe this too if those experts
told us so in a suitable fashion, then those experts would be in possession of a kind of
awesome normative power—the power to make it the case that we should or should
not give to aid agencies. For given those premises, if such experts did tell us that
the effects were good enough, that would make it the case that we should give. And
if they did not, that would make it the case (given that we shouldn’t give unless we
have sufficient reason to believe that the effects are good enough, and that we can’t
arrive at such reason on our own) that we should not.
If this argument is sound, then, the relevant experts have made it the case that
we should not give, since they have not in fact told us whether the effects are good
enough. What a tragedy that would be if the effects are good enough! Now in fact,
I do not think that all of the premises in the little argument sketched above are quite
correct.76 And so I do not think that experts have quite this much normative power.
I do think, nonetheless, that those premises are close enough to the truth to bring
out how crucial the position experts occupy is, and how important it therefore is for
them to provide the general public more assistance in making judgements about the
effects.77
5 Review
In Section 2, I explained why I think it is very difficult for non-experts to find out
how good or bad the effects of the work aid agencies do are. In Section 3, I showed
that this state of affairs may have certain very bad consequences. And accordingly, in
Section 4, I articulated and discussed several measures that might be taken to make
it easier to find out how good or bad the effects are.78 This exercise turned out to
be even more complicated than I had anticipated, and I am aware that I have barely
scratched the surface of many of the issues. I am also aware that the fact that I am not
an expert on aid means that I am hardly the person best-placed to write about these
issues. I hope, though, that this attempt will at least be enough to stimulate others—
including others with more knowledge of NGOs and their work than I have—to do
so. For the potential reward of doing so is very great. If, in particular, what I said in
Section 3 is correct, then reforms that made it easier for non-experts to find out how
good or bad the effects are might lead to major benefits for the global poor.
110 K. Horton
Notes
1. In this paper, ‘aid agencies’ refers to non-governmental organisations (NGOs) working in the
fields of international relief and development. I shall also sometimes refer to them simply as
‘NGOs’.
2. It does so unless there is some conclusive reason why it would not be the case that we should
give to aid agencies even if the effects of their work were as positive as they could be. This
chapter is addressed to those who do not think there is such a reason.
3. ‘The effects’, henceforth, is my shorthand for ‘the effects of the work aid agencies do’.
4. For ease of exposition, I shall continue sometimes to write about ‘finding out good or bad the
effects are’, but that phrase should be interpreted with the points just made in mind; that is, as
shorthand for ‘arriving at an estimate about the effects that one has at least some good reason
to believe correct’.
5. Compare Wenar (2007, pp. 270–271).
6. Those who are already convinced of this claim may skip Section 2 and go straight to
Section 3.
7. Henceforth, when I use the terms ‘us’ or ‘we’ I will mean people living in developed countries
who have no special expertise in international aid.
8. As Michael Edwards and David Hulme (1996, p. 6) put it, ‘Internal evaluations are rarely
released, and what is released comes closer to propaganda than rigorous assessment’. Cf.
Fowler (2000, pp. 13–14), Riddell (2007, pp. 267–268).
9. Much of the data on the work NGOs do that is publicly available has emerged indirectly,
as a result of the fact that around the 1980s many ‘official’ (that is, governmental) develop-
ment agencies started to channel more of their aid through NGOs. These official agencies
conducted evaluations of some of this work, and some then went on to compile and publish
omnibus reports based on these evaluations. This constituted almost the first data of any kind
on the work NGOs do available to the general public. See, for example, Riddell et al. (1994),
Kershaw (1995), and Tvedt (1995). See also OECD (1992); Riddell et al. (1995), ODI (1996),
and Oakley (1999).
10. This was a study commissioned by the Development Assistance Committee (DAC) of the
Organisation for Economic Cooperation and Development, which drew on much of the data
on NGO development work then available. This work—the ‘DAC Study’—was published in
1997, but there has been no further attempt at such a wide-ranging review since then.
11. See for example Fowler and Biekart (1996), Biekart (1999), Smillie (1999), and Fowler
(2000). See also Section 4.5 below.
12. Perhaps the most significant addition to the literature is a recent book by Roger Riddell
(2007, Chapters 16–19), which includes an extensive review of available information about
the work NGOs do. This is the most authoritative work of its kind yet produced, and conse-
quently I draw on it extensively in what follows. Nevertheless, it too conforms to the tendency
described at the end of the previous paragraph in the main text.
13. This way of putting it leaves open a major normative question, concerning how good those
effects have to be in order to imply that we should give. This question is controversial, and I
won’t try to settle it here. To say that the effects are ‘easily’ good enough, though, I shall take
it, is to say that they are so good that they will meet any reasonable standard concerning how
good those effects need to be in order to imply that we should give.
14. When I write about whether the effects of the work aid agencies do are ‘good enough’,
henceforth, I mean whether those effects are good enough to imply that we should give to
them.
15. There has always been a significant critical literature on ‘official’ aid (aid given by govern-
ments), but a literature on NGOs that includes frank discussion of problems with their work
has emerged more recently. Influential early pieces of this kind include Tendler (1982), Korten
(1987, 1990), and Clark (1991). For more references, see Sections 2.3 and 2.4.
16. Many of the pieces I cite are surveys of one sort or another; as such, most contain more
detailed references.
17. Of course, if such a summary would make it easy to see that the effects were good enough,
then again there would be no Epistemic Problem. Unfortunately, though, I do not think such
a summary would do so; see Section 2.5.
18. Development aid is often described as aid that aims at long-term solutions to the problems
of poverty and deprivation, rather than short-term relief. Given that development aid is also
standardly taken to include all aid that is not given in response to emergencies, the appar-
ent implication is that all aid that is not given in response to emergencies aims at long-term
solutions. By all accounts, however, that implication would be false: some aid that is given
in non-emergency situations aims primarily to meet immediate needs, rather than to pro-
vide long-term solutions. Therefore, it seems less misleading to characterise development aid
simply as aid given in non-emergency situations.
19. The following combined historical sketch, taxonomy, and summary of some of the problems
with development aid draws especially on Korten (1990), Clark (1991), Smillie (1995, 1999),
Biekart (1999), Fowler (2000), and Riddell (2007, Chapters 16–17).
20. The best-known example is food aid, though it should be said that official aid agencies (par-
ticularly USAID), rather than NGOs, have been the main force behind such aid. Similar
problems have arisen in connection with the transfer of other products by NGOs, however,
such as the provision of second-hand clothing, which tends to undermine local producers and
tailors (Fowler and Biekart 1996, p. 123).
21. Compare Black (2002, p. 112), and Kremer and Miguel (2004, p. 3).
22. Burnell (1991, p. 107) quotes from the 4th edition of the Oxfam Field Directors’ Handbook:
‘many small groups have been destroyed by excessive funds’. See also Burnell (1991, p. 175),
Smillie (1995, Chapters 10), Biekart (1999, pp. 298–299), and Fowler (2000, pp. 26–28).
23. On the other hand, International NGOs are also sometimes accused of emasculating local
organisations, turning them from political movements into mere providers of services; see
e.g. Arellano-López and Petras (1994), Sogge (2002, p. 180).
24. ‘At best’, because there is a vein in the literature that is much more negative about aid (includ-
ing aid given by NGOs) than most of the authors I cite here. Examples include Maren (1997)
and much of the ‘post-development’ school (see e.g. Sachs 1991, Escobar 1994, and Rahnema
and Bawtree 1997). These authors take such a one-sided, polemical line that it seems highly
questionable whether they are reliable sources, however, and consequently I do not rely on
their work here. Most of the authors I do draw on (such as Alan Fowler, Ian Smillie, Kees
Biekhart, Michael Edwards, and Roger Riddell), by contrast, are sympathetic to NGOs and
their aims, which of course makes the concerns they raise harder to discount.
25. See Riddell (2007, pp. 288–298 for a review).
26. NGDO stands for ‘non-governmental development organisation’. Similarly, Riddell writes
that ‘only a small proportion of aid (usually less than 20 percent for large NGOs) is used to
address human suffering and to tackle extreme poverty indirectly. The overwhelming majority
of aid funds donated is channeled directly to particular groups of poor people, mostly on the
basis of need’ (Riddell 2007, pp. 158).
27. For brief reviews of this area, see Collins (1998), Edwards (1999, Chapter 5), Weiss (1999),
Lindenberg and Bryant (2001, Chapters 3), Addison (2000), and Riddell (2007, Chapters
18–19). For longer studies, see De Waal (1997) and Terry (2002).
28. ‘In 1996,’ writes Tony Addison (2000, p. 394), for example, ‘Liberia’s warlords looted over
$8 million worth of relief supplies, much of which found military use’.
29. According to Alex De Waal (1997, p. 78), aid delivered to refugees and fugitives in Thailand
in 1979 ‘provisioned the defeated army of the Khmer Rouge and allowed it to remobilize’.
For the Rwandan case, see Millwood (1996), De Waal (1997, Chapter 9) and Terry (2002,
Chapter 5).
30. See Fowler (2000, p. 14). NGOs that wish to join the relevant NGO umbrella group in their
country (such as ACFID in Australia, BOND in the UK, and InterAction in the USA) have
to go through an accreditation process in order to do so, as do NGOs that wish to be eli-
gible for money from governmental sources. One can deduce, then, that NGOs that are
members of such groups or receive money from governments have met certain standards.
112 K. Horton
How much reassurance one can take from this depends of course on how stringent those
standards are, and as usual it is not easy for non-experts to know what to think about
this (though I am told that the AusAID accreditation process is pretty onerous—see
http://www.ausaid.gov.au/ngos/accreditation.cfm).
31. There are a number of organisations (such as Guidestar) that aim to provide information to
those who are thinking of funding NGOs, but the information they provide tends to be limited
to the kind of thing one would find in such NGOs’ annual reports. None, to my knowledge,
conducts rigorous evaluations of NGOs or their work.
32. It is important to emphasise rigorous evaluation because the term ‘evaluation’ is used to refer
to a number of different kinds of activity, only some of which provide reliable information
about the effects. On these issues, see for example Cracknell (2000, esp. Chapter 3).
33. In recent years, many NGOs have received an increasing share of their income from govern-
ments. Doing so avoids some of the problems involved in raising funds from the general
public, but gives rise to new dangers of its own. Competing for government funds can
lead agencies to censor any criticisms they have of government policies, for example,
and working for governments can lead them to lose their distinctive ethos and become
more like government contractors. For discussion of these issues, see Hulme and Edwards
(1997).
34. See De Waal (1997, pp. 138–139) and Pogge (2007, pp. 241–245). See also chapter
“Compromised Humanitarianism” by Cullity and the Afterword, this volume for more
discussion of such collective action problems NGOs face.
35. I discuss such responses in Horton (2009).
36. In Horton, Keith (Unpublished ms, Aid Agencies: The Epistemic Problem), I discuss in some
detail the various measures non-experts may take in an effort to find out how good or bad the
effects are.
37. As I also said above, it would be unrealistic either to expect any very precise judgement about
this matter, or to expect anything approaching certainty that any such judgement is correct.
This shows only that any such judgement may have to be rough and far from certain, though,
not that we can do without such a judgement at all.
38. And conversely, some people do give how despite the fact that it is difficult to find out whether
the effects are good enough (though few give a substantial proportion of their income). But
this is not of course inconsistent with the claim that some people who do not currently give
would start to do so if it were easier to find out that the effects were good enough.
39. David Crocker (1995) expresses a similar view: ‘what challenges aid to distant peoples is not
so much skepticism concerning moral foundations as pessimism about practical results’.
40. This possibility is also put forward and discussed briefly in Riddell (2007, pp. 156–157).
41. Opinion polls are regularly conducted on public attitudes to government aid, but the data they
provide appears to be somewhat ambiguous. Many polls provide evidence for what I take to
be the natural assumption that support for aid would go up if there were more evidence that
it was effective. Other polls, however, suggest that many people support aid even though they
believe it to be ineffective, which might be taken to imply that people’s support for aid is
relatively unaffected by how effective they think such aid is. For references and discussion,
see Riddell (2007, pp. 114–117).
42. Why not argue the point here? In part, because I don’t think that I, as a non-expert, should be
the one to take on this task. For obvious reasons, experts in the relevant field would be in a
better position to do so, and so it is really for them to do so. Now as I noted in Section 2.1,
such experts don’t in fact seem willing to take on this task, and in that context it might be
necessary for non-experts to step in as best we may. As I said in Section 2.5, though, I do not
think there is any quick, decisive argument that the effects are good enough, and I don’t have
space for a long and detailed argument here. (I do provide such an argument, however, in
Horton 2009.)
43. That includes even those who currently give in order to feel good (etc.) for presumably giving
would not make one feel good if it were easy to see that aid agencies did more harm than good.
44. Some NGOs have started to publish summaries of evaluations on the websites, though the
ones I have seen do not tend to be written in a form that enables one to form a very clear idea
about the effects of the relevant activities.
45. This is because of the difficulties involved in aggregating information derived from evalua-
tions of different activities into meaningful overall measures. One common response to this
problem is to treat whether activities have met their objectives as a kind of common factor.
Unfortunately, though, it is often difficult to know how to interpret such assessments. This
problem is illustrated in the nearest thing to such publicly available meta-evaluations that I
am aware of, published by CARE on its website (see http://care.ca/libraries/dme/). The most
recent is ‘The Mega 2002 Evaluation’, based on 65 evaluation reports (Goldenberg 2003).
This document says that 82 percent of those evaluations ‘reported that projects had achieved
most of their intermediate objectives’ (p. 5). It is difficult, though, for an outsider to know
what to make of this, in part because the report does not explain what is meant by ‘interme-
diate objectives’. It does contrast ‘intermediate objectives’ with ‘final goals’, though, which
presumably are matters of real importance. Unfortunately, though, the report does not give
any data (so far as I can see) concerning what proportion of projects achieved their final
goals.
46. Evidence for this claim may be found in Oxfam GB’s ‘Stakeholder Surveys Reports’, based
on responses to questionnaires concerning Oxfam’s performance by different groups such
as donors, staff, volunteers, and ‘beneficiaries’. These reports were set up precisely in an
effort to respond to concerns about accountability. But there appears to be little enthusiasm
for such initiatives among Oxfam’s supporters. In the latest report (2003), for example, only
12 percent of financial supporters bothered to return the questionnaire, and many who did so
expressed the view that the money spent on the survey should instead have been spent directly
on projects (Oxfam 2003, pp. 5, 35).
47. According to Riddell (2007, p. 268), ‘There are only a handful of NGOs today which put
information into the public domain which draws attention to failures and problems as well as
successes’.
48. It is a further question how well-grounded such fears would be. If the figures cited by Riddell
apparently showing that many people are happy to fund aid even though they believe it to be
ineffective are reliable (see Note 42 above), it might be that NGOs would have little to fear
from revealing more about the negative aspects of aid.
49. This is also the natural solution to danger that less scrupulous agencies might outcompete
more scrupulous ones (see Section 2.3 above). NGOs have been slow, however, to set up such
forms of regulation. For discussion of some of the initiatives that have been set up, see chapter
“The Seeming Simplicity of Measurement” by Roche, this volume and Section 4.4 below.
50. This danger is illustrated by the experience of the Disasters Emergency Committee (DEC)
in the UK, an umbrella organisation representing 13 UK aid agencies. For several years,
DEC made publicly available independent evaluations of the work of their agencies. The
media tended to focus exclusively on the negative elements of such reports, however, and in
response DEC now commission two reports, one for public consumption and the other for
insiders only.
51. There is some evidence that this happens at present, for those who have access to the relevant
material often say that evaluations conducted by NGOs themselves tend to be more critical
than evaluations conducted by other parties (see e.g. Kruse et al. 1997, Sections 2.3 and 4.1).
And it seems plausible that one reason for this is that evaluations conducted by NGOs tend
not to be made publicly available.
52. Michael Maren (1997, p. 269) writes, for example, ‘What is really required is a truly indepen-
dent agency . . . to look after the interests of the targets of development and relief, a.k.a. the
needy. The organization should be staffed by professionals who have the time and resources
to produce detailed analyses of what these organizations [NGOs] are doing for the poor of the
Third World. Those that do effective work should be singled out so that “customers” know
where to spend their money’.
114 K. Horton
53. I briefly discuss some of most significant ventures of this kind—all set up by NGOs
themselves, rather than by an external body—in Section 4.4.
54. It is often claimed, in particular, that development is complex, non-linear and hard to
measure, and that therefore an insistence on measurement leads to activities that are
anti-developmental. Rondinelli (1993) is often cited in this context.
55. A lot depends here on what forms of evaluation are able to provide sufficiently reliable infor-
mation about the effects. This is a complex and somewhat technical matter, though, and
aid experts themselves seem to be divided about it. For relevant discussion, see Cracknell
(2000), Banerjee (2007), and Roche (chapter “The Seeming Simplicity of Measurement”,
this volume).
56. As Savedoff et al. (2006, Chapters 3) write, ‘No responsible physician would consider pre-
scribing medications without properly evaluating their impact or potential side-effects. Yet
in social development programs, where large sums of money are spent to modify population
behaviours, change economic livelihoods, and potentially alter cultures or family structure,
no such standard has been adopted’.
57. For more on why NGOs (and other organisations) tend to be so resistant to evaluation, see
Biekart (1999, pp. 104f) and Pritchett (2002).
58. The authors of the DAC Study reported that both the quality of the data and the substantive
findings were better in relation to certain of the activities evaluated, in particular certain child
survival and microenterprise programs (Kruse et al. 1997, Chapter 5).
59. Providing such information would also provide a strong sincerity test for people who say, ‘I
would give to an aid agency if I had sufficient reason to believe that the effects of its work
were good enough’. (If, on the other hand, it turns out that no kind of aid has effects that are
good enough, then it would of course also be helpful to be told that.)
60. ‘On the face of it’, because such a tendency may also have certain negative consequences.
If a lot of people supported the forms of aid in question, those forms may become swamped
with more resources than they can profitably use, while other equally good or better forms of
aid are neglected. Such problems would be compounded, moreover, if there was a mismatch
between the forms of aid that it is easiest to confirm have good enough effects, on the one
hand, and the forms of aid that have the best effects, on the other. For if so, the forms of aid
that people would tend to gravitate to when they follow this strategy would not be those that
have the best effects.
61. Again, I focus on the more ‘respectable’ reasons. Other more questionable reasons may of
course also play a role, such as the fear of being ‘found out’, or of losing ‘market share’.
62. As this paper goes to press, I read in Singer 2009 that such an organisation has in fact been
created, called ‘GiveWell’ (see the relevant website and Singer 2009, Chapter 6).
63. A further possibility, suggested to me by Chris Roche, is that AusAID and the Committee for
Development Cooperation could make public the accreditation studies (or elements thereof)
that they undertake for every accredited agency every 5 years. (Naturally, similar bodies in
other countries could do the same.) Indeed, one could argue that it would be very much in
line with AusAID’s accountability to the taxpayer to do this.
64. All the quotations are from the relevant websites. For more discussion of these developments,
see chapter “The Seeming Simplicity of Measurement” by Roche, this volume.
65. ALNAP in particular focuses on issues concerning evaluation—see http://www.alnap.org/
themes/evaluation.htm.
66. The set of standards developed so far that would appear most likely to be able to play
that role is that of the Humanitarian Accountability Partnership, which employs a certifi-
cation system that measures performance against a set of apparently tough-minded auditable
benchmarks; see http://www.hapinternational.org/projects/certification.aspx. For discussion,
see again Roche (chapter “The Seeming Simplicity of Measurement”, this volume).
67. This apparently applies to official aid as well as NGO aid. To quote from a World Bank study
of evaluation, for example, ‘Despite the billions of dollars spent on development assistance
each year, there is still very little known about the actual impact of projects on poor people’
(Baker 2000, cited in Easterly 2006, p. 194). See also Savedoff et al. (2006).
68. See its annual meta-evaluations, available at http://www.alnap.org/themes/evaluation.htm.

69. See chapter “The Seeming Simplicity of Measurement” by Roche, this volume, though, for a
sketch of how some of the initiatives developed in the humanitarian field might be extended
to other forms of aid.
70. See the references given in Note 56 above.
71. At least in part, this may because such measures are focused mainly on improving practice
and increasing accountability to recipients, for generally NGOs seem to be more motivated
by those goals than by the goal of providing more reliable information to contributors.
72. I fear that the use of term ‘experts’ may set off alarm bells among certain readers; if so, one
may substitute the term ‘aid specialists’ instead. The important point is of course not the label,
but what the label denotes: those who are relatively well-placed to make generalisations about
the effects of NGO aid.
73. There are of course a number of books and articles on NGO aid by experts, such as those I
cited in Section 2, many of which offer helpful information, and some of which are aimed at
least in part at the general reader. None of those I am aware of offer the kind of estimates of
the effects potential contributors need to determine whether they should give to aid agencies,
though.
74. On an optimistic reading, this factor would help to explain why so much of the literature on
NGOs is so negative.
75. Riddell (2007, p. 444, No. 23) writes that ‘it is neither possible, nor particularly helpful, to try
to draw simple and sweeping across-the-board conclusions about the impact of NGO develop-
ment interventions’. He does not address the obvious reply: that even if ‘simple and sweeping
across-the-board conclusions’ are not possible, complex conclusions, or estimates concerning
particular forms of aid, might be. Similarly, Chris Roche has put it to me that because they
of the very fact that they recognise the complexity of the relevant issues, experts will not be
willing to provide a ‘simple homogenising answer which would contradict what they know
to be true—i.e. that success and failure are very context-specific’ (private communication).
But again, let me emphasise that I am not assuming that the relevant judgements need to take
this form. Such a judgement might, for example, be so qualified and nuanced that it takes
thousands of words even to state. For an example of such a complex judgement—though not,
unfortunately, from an expert—see Horton (2009).
76. For one thing, I think that it is probably difficult rather than impossible for non-experts to
come to a judgement about the effects that they have sufficient reason to feel confident about
on their own.
77. In response to this argument, one expert at the workshop where a draft of this paper was dis-
cussed offered his view that the effects were indeed good enough. This individual has a lot
of professional experience with such NGOs, including evaluation of their activities, but has
never worked for an NGO himself, and so is presumably relatively unlikely to feel institu-
tional or personal pressures to give a positive verdict. (Indeed, he emphasised how much it
pained him to give this verdict, which I take to be a reflection of how focused on the deficien-
cies of aid such experts tend to be.) For these reasons, among others, his judgement seems
likely to me to be a relatively reliable one (though of course one does not know how good
this expert believes the effects need to be in order to be ‘good enough’ (see Note 14 above)).
Another expert expressed the same view to me privately after the workshop, and added that
in his view 8 out of 10 ‘key informants’ would agree. Though this is again encouraging, it
should be said that this individual has spent his life working for an NGO, which might mean
that the risk of bias is higher in his case.
78. There are other measures which I lack space to discuss here. One would be for NGOs
to change their activities in ways which would make it easier for potential contributors
to find out more about the effects, for example by selecting activities that lend them-
selves relatively easily to measurement, and then providing such measurements. Obviously,
such a suggestion would raise several of the concerns mentioned towards the end of
Section 4.2.
116 K. Horton
References
Addison, T. 2000. Aid and conflict. In Foreign Aid and Development: Lessons Learnt and
Directions for the Future, eds. F. Tarp and , P. Hjertholm. London: Routledge.
Arellano-López, S. and J.F. Petras. 1994. Non-governmental organizations and poverty alleviation
in Bolivia. Development and Change 25: 555–568.
Baker, J.L. 2000. Evaluating the Impact of Development Projects on Poverty. Washington, DC:
World Bank.
Banerjee, A.V. 2007. Making Aid Work. Cambridge: The MIT Press.
Biekart, K. 1999. The Politics of Civil Society Building. Utrecht/Amsterdam: International Books
and Transnational Institute.
Black, M. 2002. No Nonsense Guide to Development. London: Verso Press.
Burnell, P. 1991. Charity, Politics, and the Third World, Harvester Wheatsheaf Brighton, in the
UK. New York: St Martin’s Press.
Carr, S.C., E. Mcauliffe, and M. Maclachlan. 1998. Psychology of Aid. London: Routledge.
Chambers, R. 1994. Participatory rural appraisal (PRA): Analysis of experience. World
Development 22(9): 1253–1268.
Clark, J. 1991. Democratizing Development: The Role of Voluntary Organizations. London:
Earthscan.
Cracknell, B. 2000. Evaluating Development Aid. New Delhi: Sage Publications.
Collins, C. 1998. Critiques of humanitarianism and humanitarian action. In Humanitarian
Coordination: Lessons Learned, 12–26. New York: Office for the coordination of Humanitarian
Affairs.
Crocker, D. 1995. Hunger, capability, and development. In Morality and World Hunger, eds.
W. Aiken and H. La Follette (1996). Upper Saddle River, NJ: Prentice-Hall.
Davies, R. 2001. Monitoring and Evaluating NGO Achievements, at http://www.mande.co.uk/
docs/arnold.htm.
De Waal, A. 1997. Famine Crimes: Politics and the Disaster Relief Industry in Africa. London:
African Rights and the International African Institute.
Dichter, T. 2003. Despite Good Intentions. Amherst and Boston, MA: University of Massachusetts
Press.
Easterly, W. 2006. The White Man’s Burden. New York: The Penguin Press.
Edwards, M. 1996. International Development NGOs: Legitimacy. Accountability, Regulation and
Roles’, in Meeting the Challenge of Change: The Report of the Commission on the Future of
the Voluntary Sector, Commission on the Future of the Voluntary Sector, London.
Edwards, M. 1999. Future Positive. London: Earthscan.
Edwards, M. and D. Hulme, eds. 1992. Making a Difference: NGOs and Development in a
Changing World. London: Earthscan.
Edwards, M. and D. Hulme. 1996. Introduction: NGO performance and accountability. In Beyond
the Magic Bullet: NGO Performance and Accountability in the Post-Cold War World, eds.
M. Edwards and D. Hulme. West Hartford, CT: Kumarian Press.
Escobar, A. 1994. Encountering Development. Princeton, NJ: Princeton University Press.
Fowler, A. 2000. Civil Society, NGDOs and Social Development: Changing the Rules of the Game.
Occasional Paper, UNRISD, Geneva, athttp://www.unrisd.org/unrisd/website/document.
nsf/0/F553495F06F98DCE80256B5E005C9DDC?OpenDocument.
Fowler, A. and K. Biekart. 1996. Do private aid agencies really make a difference. In Compassion
and Calculation: The Business of Private Aid Agencies, ed. D. Sogge. London: Pluto Press.
Goldenberg, D.A. 2003. The Mega 2002 Evaluation (Meta-Evaluation of Goal Achievement in
CARE projects), CARE USA Program Division, Feb 2003, at http://care.ca/libraries/dme/.
Horton, K. 2009. Should We Give to Aid Agencies? Edinburgh: Edinburgh University Press.
Hulme, D. and M. Edwards, eds. 1997. NGOs, States and Donors: Too Close for Comfort? London:
Palgrave Macmillan.
Kershaw, I. 1995. Review of the Effectiveness of NGO Programs. Canberra: AusAID.
Kremer, M. and E. Miguel 2004. The Illusion of Sustainability, Working Paper no.
W10324, National Bureau of Economic Research at http://papers.ssrn.com/sol3/papers.
cfm?abstract_id=509855.
Kruse, S.-E., T. Kyllönen, S. Ojanperä, R. Riddell, and J. Vielajus. 1997. Searching for Impact and
Methods: NGO Evaluation Synthesis Study. Helsinki: Finnish Ministry of Foreign Affairs, at
http://www.valt.helsinki.fi/ids/ngo/.
Korten, D. 1987. Third generation NGO strategies: A key to people-centred development. World
Development 15(Supplement): 145–159.
Korten, D. 1990. Getting to the 21st Century: Voluntary Action and the Global Agenda. West
Hartford, CT: Kumarian Press.
Lappe, F.M. and J. Collins. 1988. World Hunger: Twelve Myths. London: Earthscan.
Lindenberg, M. and C. Bryant. 2001. Going Global: Transforming Relief and Development NGOs.
West Hartford, CT: Kumarian Press.
Millwood, D. ed. 1996. The International Response to Conflict and Genocide: Lessons from the
Rwanda Experience, Steering Committee of the Joint Evaluation of Emergency Assistance to
Rwanda, at http://www.reliefweb.int/library/nordic/index.html.
Minear, L. 2002. The Humanitarian Enterprise: Dilemmas and Discoveries. West Hartford, CT:
Kumarian Press.
Oakley, P. 1999. A Review of Danish NGO activities in Developing Countries. Oxford: Intrac.
ODI (Overseas Development Institute). 1996. The Impact of NGO Development Projects, Briefing
Paper, Overseas Development Institute, London.
OECD (Organisation for Economic Cooperation and Development). 1992. The Evaluation of
Non-Governmental Organisations’ Activities, Organisation, Methodology and Results. Paris:
OECD.
Oxfam. 2003. Stakeholder Survey 2003/4, Oxfam GB, London at http://www.oxfam.org.uk/
about_us/index.htm.
Pogge, T. 2007. Moral priorities for international human rights NGOs. In Ethics in Action:
The Ethical Challenges of International Human Rights Nongovernmental Organizations, eds.
D.A. Bell and J.-M. Coicard. Cambridge: Cambridge University Press.
Pritchett, L. 2002. It pays to be ignorant: A simple political economy of rigorous program
evaluation. The Journal of Policy Reform 5(4): 251–269.
Rahnema, M. and V. Bawtree. 1997. The Post-Development Reader. London: Zed Books.
Riddell, R. 1999. Evaluating NGO Development Interventions. In NGOs and Voluntary
Organisations: Learning from Each Other, ed. D. Lewis. London: Earthscan.
Riddell, R. 2007. Does Aid Really Work? Oxford: Oxford University Press.
Riddell, R., A. Bebbington, and L. Peck. 1994. Promoting Development by Proxy:
The Development Impact of the Swedish Government’s Support to NGOs. Stockholm:
SIDA.
Riddell, R., M. Robinson, J. De Coninck, A. Muir, and S. White. 1995. Non-Governmental
Organizations and Rural Poverty Alleviation. London: Clarendon Press.
Rieff, D. 1995. The humanitarian trap. World Policy Journal 12(4 (Winter)): 1–11.
Rondinelli, D. (1993). Projects as Policy Experiments: An Adaptive Approach to Development
Administration. London: Routledge.
Roche, C. 1999. Impact Assessment for Development Agencies. Oxford: Oxfam GB.
Sachs, W. 1991. The Development Dictionary: A Guide to Knowledge as Power. London: Zed
Books.
Savedoff, W.D., R. Levine, and N. Birdsall 2006. When Will we ever Learn? Improving
Lives Through Impact Evaluation, Center for Global Development, at http://www.cgdev.org/
content/publications/detail/7973.
Singer, P. 2009. The Life you can Save. New York: Random House.
Smillie, I. 1995. The Alms Bazaar: Altruism Under Fire—Non-Profit Organizations and
International Development. London: IT Publications.
118 K. Horton
Smillie, I. 1999. At sea in a sieve? In Stakeholders: Government-NGO Partnerships for

International Development, eds. I. Smillie and H. Helmich. Paris: OECD.
Sogge, D., ed. 2002. Give and Take: What’s the Matter with Foreign Aid? London: Zed Books.
Tendler, J. 1982. Turning Private Voluntary Organizations into Development Agencies: Questions
for Evaluation, Discussion Paper No. 12, USAID Programme Evaluation, Washington, DC.
The Economist. 2006. Feb 25, 16
Terry, F. 2002. Condemned to Repeat? The Paradox of Humanitarian Action. Ithaca and London:
Cornell University Press.
Tvedt, T. 1995. Non-Governmental Organisations as a Channel for Development Aid: The
Norwegian System, Evaluation Report. Norway: Ministry of Foreign Affairs.
Vaux, T. 2001. The Selfish Altruist: Relief Work in Famine and War. London: Earthscan.
Weiss, T.G. 1999. Principles, politics, and humanitarian action. Ethics and International Affairs
13: 1–22.
Wenar, L. 2007. The basic structure as object: Institutions and humanitarian concern. Canadian
Journal of Philosophy in Global justice, global institutions Supp. 31: 253–278.
The Seeming Simplicity of Measurement∗
Chris Roche
1 Introduction
The key issue this chapter tries to address relates to a challenge raised by Keith
Horton’s chapter “The Epistemic Problem: Potential Solutions”, this volume in
response to Peter Singer’s proposition (1999) that the rich have a moral obliga-
tion to assist the world’s poor and therefore should give a reasonable proportion
of their income to those agencies whose aim it is to alleviate poverty and suffer-
ing. Horton’s challenge is that surely this moral obligation only applies if those in a
position to give some of their income in this way have some ability to satisfy them-
selves that the agency or agencies to which they might give, are able to demonstrate
the net effect of their work is good enough to imply that we should give to them.
He goes on to argue that it is in fact very difficult for those who are not experts
on aid to find out what the effects of aid actually are; this he calls the ‘Epistemic
Problem’.
The paper therefore does not address the ethical or moral questions that are the
primary subject of Peter Singer’s work i.e. whether there is a moral obligation to
assist the poor—other chapters in this volume do—but rather one of the secondary
questions that stems from this. In particular it explores what the role of one set of
agencies—International NGOs (INGOs) such as Oxfam Australia that are one of the
potential beneficiaries—might be in seeking to increase the general public’s ability
to assess if their donations are liable to make a positive difference to the lives of
people living in poverty or affected by disasters.
This is an important question not least if, as Peter Singer has suggested (1999),
that it can indeed be the practical uncertainties about whether aid will really reach
the people who need it, that can act as reason not to do anything. Or to put it another
way ‘what challenges aid to distant peoples is not so much skepticism concerning
C. Roche (B)
Oxfam Australia, Carlton, VIC, Australia
e-mail: croche@oxfam.org.au
∗ The trouble with measurement is its seeming simplicity’.∼ Author Unknown

120 C. Roche
moral foundations as pessimism about practical results’ (Crocker 1996). And as

Singer notes, and Horton confirms, ‘Nobody who knows the world of overseas aid
can doubt that such uncertainties exist’.
What should International NGOs approach to this challenge be? Should we be
simply saying trust us—if you give to us there is a high probability of us making a
real difference to people’s lives? Is it being clearer about our principles and values
and how we try to live up to them? Or do we need to provide better information,
analysis and evidence that we make a difference? If so how do we gather this?
What constitutes ‘evidence’, and to what degree of rigour? How much of donors’
money should we spend on monitoring, evaluation and research, which arguably
could be otherwise spent on directly relieving poverty? This chapter tries to provide
some answers to these questions based on NGOs’ experience, as well as trying to
explore what else the sector might do to better inform its donors and in so doing
lessen the potential excuses that undermine the moral obligation that Singer, and
others, suggest exists. In particular the paper looks at whether some of the initiatives
recently developed in the humanitarian sector can be usefully applied to other forms
of aid.
2 Potential Biases for INGOs
In the first instance, and as someone employed by an INGO, looking at the above
questions necessitates being aware of the likely biases that myself or others from
INGOs are likely to bring to the discussion. These are biases that are above and
beyond the more obvious likelihood that NGOs—particularly in their fundraising
and public relations material—are liable to be largely positive about their work.
Horton (2004) offers a useful framework for looking at other potential biases that
can distort thinking. I feel these offer a valuable framework for starting to address
the above questions from a position that is relatively less subjective than it otherwise
might be.
Horton suggests some of these mutually reinforcing biases might be:
• belief perseverance: the tendency to maintain previously held beliefs and neglect
potentially disconfirming evidence which might conflict with those beliefs e.g.
that aid might not work,
• conformity: the tendency for our own beliefs to be unduly influenced by what we
take others to believe e.g. that aid is a moral imperative,
• the vivid/pallid dimension; the tendency to be much more influenced by vivid,
concrete data than by the same data, or even much more rigorous or revealing
data, presented in a pallid or abstract way e.g. the potential bias to respond to
a more ‘newsworthy’ emergency such as the Tsunami, than ones that are less
‘vivid’ in the public imagination like that in the Democratic Republic of Congo
(Howard 2007),
• wishful thinking: the tendency to be inclined to come to believe what we want
to be true e.g. that information and stories that tend to fit with the ‘good news’
The Seeming Simplicity of Measurement 121
one wants to believe is privileged over that which undermines what we wish to
be true,
Each of these biases speaks to a number of concerns that need to be addressed if

the practical uncertainties that Singer refers to are to be lessened. These include the
recognition that:
(a) aid sometimes does not work and indeed may make things worse for some,
(b) there are dangers of group-think in and around the sector are real and
dangerous,
(c) Developmental and Organisational or Institutional Development Objectives can
clash (Edwards 1996); this can lead to institutional priorities, profile and sur-
vival becoming the driver of decision-making at the expense of development
outcomes or working effectively with others,
(d) our own assessments of our effectiveness need to be cross-checked and veri-
fied in order for potential belief perseverance and wishful thinking biases to be
adequately balanced.
3 INGOs, Ethics, Principles and Evidence
In the first instance as a number of observers have suggested, value based organ-
isations need to be clear about what their values and principles are, particularly
when they are confronted with the potential ‘dark side’ of what they do. As Hugo
Slim (1997) noted ‘for many people, it is almost counter-intuitive to have to con-
sider that humanitarian action may also have a dark side which compromises as
well as helps the people whose suffering it seeks to assuage. However, the many
difficult situations in which relief agencies have found themselves in recent years
have inevitably led members of the relief community to examine this shadow side
of humanitarianism’. Thus leading to what Didier Cherpitel argued in his fore-
word to the 2003 World Disasters Report ‘There are no simple answers. We can
only develop this essential art of humanitarian judgement through openly declar-
ing the ethical principles we believe in, trying our best to put them into practice
and being prepared to measure the effects of our decisions on a continual basis.’
(IFRC 2003, p.7).
So what might those ethical principles look like? INGOs have been looking at
such principles for some time, often associated with particular moral dilemmas they
have faced for example in Biafra, Somalia, the great Lakes, Bosnia etc. This search
is not new and has been especially raised in the above humanitarian situations pre-
cisely because they tend to bring these issues to the fore most forcefully. This has
resulted in: efforts to define agreed and collective standards and principles (the Red
Cross Code, SPHERE); debates on moral dilemmas and how to deal with them (e.g.
Hugo Slim, Do No Harm etc); and initiatives to enhance collective learning and
evaluation (ALNAP) and develop initiatives on joint accountability (HAP).
122 C. Roche
3.1 The Red Cross Code

One of the most widespread attempts to define key principles for humanitarian
agencies was the Red Cross Code of Conduct (IFRC 1994) which was developed
by a wide range of NGOs and has now been signed onto by more than 300 agencies.
This framework contained an implicit hierarchy of principles with the humani-
tarian imperative at the top and a number of subsidiary principles related to non
discrimination, independence from foreign policy interests, accountability to both
beneficiaries and donors, and a desire to both build capacity and reduce vulnerabil-
ity in the longer term. Whilst there have been a number of critiques of the Code,
notably that it is too vague and there is no enforcement mechanism (Hilhorst 2005),
there are also a number of robust defenders of the code (Vaux, T. 2002, The DEC
and the Red Cross Code—A Policy Proposal unpublished memo). They see the code
as a set of principles that allow for more context specific indicators and principles to
be defined, rather than being overly prescriptive, and a structured means of assess-
ing trade offs between principles—particularly in relation to debates around the
emergencies-development continuum—rather than such discussions being confined
to either or debates around humanitarian minimalism (prioritising saving lives over
anything else) vs. humanitarian maximalism (who also seek to build capacity and
local peace dividends at the same time).
It is also possible to argue that the principles, with some tweaking, can also serve
as a useful basis for asking some pointed questions of long term development or
advocacy work that are equally relevant. Table 1 below illustrates this for example
indicating how some of the key issues that using these principles in the evaluation of
humanitarian work have raised (column entitled Issues raised by DEC evaluations)
can also be applied to:
• advocacy agendas—which can be as publicity driven as opposed to ‘needs’ driven
as humanitarian work;
• developmental approaches—which can ignore short term needs, in the same
way that humanitarian agendas can ignore longer term capacity development by
displacing local organisations during a response;
• work which runs the risk of being done to support organisational interests
and imperatives as opposed to being driven by humanitarian, development or
advocacy imperatives;
• common issues across all areas which might relate to: creating risks for others
because of the work that is done; being too close to donors or lobbying targets;
the challenges of participation, representation, transparency and accountability.
Whilst the nuances between approaches are certainly different, some of these
questions are fundamentally the same for all of them. In this sense it is perhaps sur-
prising that NGOs have done little to explore how approaches adopted and trialled
in the humanitarian arena—one that often exposes and magnifies an organisation’s
weaknesses—might be relevant to other areas of work. Arguably such an approach
would recognise that, as Horton rightly argues in Section 2.2 of his chapter, there
are different strengths and weaknesses to the diverse work that NGOs do and a more
holistic way of assessing this is needed.
Table 1 Extension of Red Cross Code of principles to development and advocacy work
Issues raised by DEC

evaluations of Similar development Similar advocacy
humanitarian work challenges challenges
Humanitarian or • Self interest of • Self interest of • Self interest of

developmental or agencies vs. needs, agencies vs. agencies vs. policy
advocacy imperative • Publicity driven development needs, needs,
agendas, • Income or supply • Publicity driven
• Individual vs. driven agendas, agendas,
Collective • Individual vs. • Individual vs.
functioning Collective Collective
functioning functioning
Non discrimination • Targeting, gender, • Understanding and • Understanding and
caste, class, poor vs. analysis of analysis of
rich differential differential
impacts on gender, impacts on gender,
class, poverty, caste class, poverty,
etc countries
Religion and politics • Impartiality, • Creating a backlash, • Creating a backlash,
neutrality, • Bias in religious • Closing down
• Bias in religious based agencies, political space,
based agencies • Ignoring spirituality • Bias in religious
and culture, based agencies,
• Working with • Political bias
government
Foreign policy • Identify strategic • Identify strategic • Identify strategic

interests of donors, interests of donors, interests of lobby
• Neutrality in conflict, • Assess if funding targets,
• Assess if funding from government • Assess if funding
from government donors undermines from government
donors limits development goals donors limits
humanitarian goals advocacy agenda
Build on local capacity • Work with local • Building capacity of • Building capacity of
organisations, local organisations, local organisations to
• Training and local • Training and local speak out,
staff development, staff development, • Training and local
• Awareness of • Awareness of staff development,
political implications unequal partnerships • Awareness of
and dangers of undermining local
undermining local organisations by
organisations speaking out
Involve beneficiaries • Formal consultation, • Formal consultation, • Formal consultation,
• Representation, • Participation, • Participation,
• Transparency, • Representation, • Legitimacy of
complaints and • Transparency, agenda,
redress complaints and • Representation,
redress • Transparency,
complaints and
redress
124 C. Roche
Table 1 (continued)
Issues raised by DEC

evaluations of Similar development Similar advocacy
humanitarian work challenges challenges
Reduce Future • Long term • Build reducing • Providing linkage

vulnerability engagement, vulnerability into and protection to
• Advocacy, long term work, those speaking out,
• Disaster Risk • Disaster Risk • Push for systemic
Reduction Reduction change in legal
systems which
protect
Accountable to donors • Efficiency, • Efficiency, • Efficiency,
• Accountability, effectiveness, effectiveness,
• Evaluation • Accountability, • Accountability,
• Learning and • Learning and
evaluation evaluation,
Dignity in images • Positive images, • Positive images, • Positive images,
• Increased public • Increased public • Increased public
understanding of understanding of understanding of
causes and what limits to development limits to advocacy
people do for
themselves
3.2 Slim’s Ethical Framework

One of the clearest ethical frameworks for humanitarian work has been developed
by Hugo Slim (1997). Slim sets out a number of questions which suggest how NGOs
might incorporate an ethical dimension into their decision-making,
Firstly Slim reminds us that it should not be forgotten agencies working in con-
flict situations are always responding to the violence of others and that the difficult
moral choices they face are often due to the immoral choices of others. It is possi-
ble, I believe, to extend this to what others have called ‘structural violence’ (Galtung
1969) and thus see many of the dilemmas facing those involved in long-term devel-
opment and advocacy work as also resulting from the unjust policies and practices
of other actors. This does not excuse us of the need to promote ethical decision-
making but helps us to look at where appropriate boundaries to our deliberations
might be.
The first step Slim suggests is to determine what basic code of ethics the organ-
isation adheres to and whether it is essentially ‘duty-bound’, like most of the
medical profession or the ICRC, i.e. certain actions are determined as intrinsi-
cally ‘good’ or ‘right’ and responsibility for broader outcomes is more dispersed
(one does not refuse to treat a patient because of what they may do in the
future), or more ‘goal-bound’ like many development or advocacy NGOs: i.e. it
involves a more complicated and uncertain process of anticipating wider outcomes
and holding oneself, at least partially, responsible for events beyond the present
time. From this understanding Slim then suggests three sets of questions relat-
ing to what drives an agency, what informs an agency, and what empowers an
agency.
In Table 2 I indicate again how the questions asked by Slim have an important
relevance to development and advocacy work as well as to humanitarian response.
In particular there are major challenges in:
Table 2 How Slim’s framework for ethical assessment of humanitarian work might apply to
development and advocacy
Relevance to development and

Main areas Key questions advocacy work
What drives the agency? What code of ethics? What are the limits to a
Is the organisation secure in the consequentialist approach?
belief that its actions are How good is our crystal ball?
always good in themselves?
Or does it believe it needs to
have a sure grasp of the
wider consequences of its
work to be certain of the
goodness of its work?
How are different Intentions When does getting close to the

and Motivations balanced? powerful start to distort
Is the agency acting out of the advocates thinking?
best motives and intentions How do we prevent the need
or doing the right thing for for public profile and ‘the art
the wrong reasons? How of the possible’ make us
does it balance different complicit in legitimizing
organisational motives such agendas of organizations that
as compassion, income, are the cause of the wrong
publicity and influence? we seek to right?
How do we ensure that a
legitimate desire for
organizational survival and
growth does not start to drive
our responses?
What Informs the Agency What Knowledge is sought? For global advocacy, in
To what degree does/has the particular, how do we
agency made every effort to know—and who do we
gather all information ask—about the differential
relevant to its effects it might have? How
decision-making? If the do advocacy agendas get
agency does/did not have all created and who determines
knowledge at its disposal them?
could it have reasonably
been expected to collect it?
126 C. Roche
Table 2 (continued)
Relevance to development and

Main areas Key questions advocacy work
What empowers the agency? What capacity is available? Is speaking out going to make a
Does/did the agency have the difference? Would feeding
capacity (resources, information to others be
influence, power) to do more effective? When do we
anything about the dilemmas accept—and who do we
it faces? An agency can only ask—when more direct
he held responsible for not support should take priority
doing something it could over development or
have, but chose not to advocacy?
Is deliberation valued? Has adequate deliberation

Does/did the agency make occurred about: negative
every effort to seek counsel, impacts; risks to staff or
consult, debate and weigh others; or whether speaking
carefully in mind the various out is done for the right
aspects of an ethical reasons?
problem?
Does/did the agency seek to If differential impacts and risks

mitigate any negative are identified e.g. if
impacts? Does/did the empowering people living in
agency seek to minimise any poverty risks a backlash from
likelihood of negative the more powerful; or if
impacts? In a choice between changes in trade policy are
‘two evils’ did it seek to limit likely to affect the rural and
the damage caused? urban poor differently, are
mitigation strategies possible
and if not should advocacy
go ahead on the basis of the
greater good?
Is there an organisational Is past development or
conscience? Do senior advocacy work evaluated and
managers promote a moral unexpected and unintended
ethos and ensure staff have effects identified? Is there a
the ethical skills and good understanding amongst
knowledge of past dilemmas staff of these lessons learnt
to increase the likelihood of and do they inform ongoing
staff making sound decisions work and future planning?
in difficult moments?
• ensuring adequate analysis and consultation has occurred to inform decision-

making without being prone to the ‘analysis paralysis’ that might lead to
unethical outcomes by not doing something one could. This problem becomes
potentially greater the more global, or general, the work being done not least
because working on generic global or international policy change, for example,
is more likely to have differential impacts, and secondly it becomes harder and
harder to ensure adequate participation and/or representation of those affected in

determining that agenda;
• creating organisational environments and cultures that value deliberation and
learning when many of the staff of NGOs are activist by nature. This is
compounded by the fact that the environment in which NGOs work is becom-
ing increasingly competitive and arguably dominated by narrow approaches to
results-based management which can create perverse incentives to not ‘embrace’
and admit error, as well as to downplay the role of others in achieving results
so that changes can be attributed to individual agencies so that they can be held
accountable.
Arguably this framework provides a useful tool for undertaking an element of
institutional and policy analysis that can be lacking in INGO evaluative practices.
When combined with other tools and approaches for looking at organisations and
impact assessment, such as the Red Cross Code of Conduct, and the Humanitarian
Accountability Partnership (see below), it can help in making the link between sig-
nificant changes in individuals lives and the organisational context in which this
happens. An organisation’s values and ethics (or lack of them) are an important
indication not only to how decisions are taken, but also which projects or policies
will, or will not, be delivered and the value that is placed upon the impacts that are
achieved. In many ways a more conscious and deliberate use of this sort of ethical
framework would address several of the concerns summarized by Horton in his sec-
tion on ‘Aid agencies as institutions’ as it focuses on some of the specific areas such
as organizational learning and the danger of agencies own organizational growth
and survival becoming more dominant than its own development or humanitarian
goals.
3.3 Quality and Accountability Initiatives—SPHERE, HAP

and ALNAP
SPHERE was launched in 1997 by a group of humanitarian NGOs and the Red
Cross/Red Crescent Movement. A handbook has been developed which includes
a Humanitarian Charter, Standards for four sectors (Water/Sanitation and Hygiene
Promotion; Food Security; Nutrition and Food Aid; Settlement and Non- Food Items
and Health Services) plus Standards common to all sectors (SPHERE 2004). The
Charter and Standards are seen to be contributions to an operational framework for
accountability in disaster assistance. The handbook is revised in consultation with
users. The most recent revision was published in 2004 and the next is due in 2009.
The Sphere Humanitarian Charter and Minimum Standards is intended to provide
a ‘common language’ for humanitarian organisations. Not all humanitarian agen-
cies however agree with the Sphere approach, for example critics like the Groupe
URD in France, believe the development of universal standards is highly contro-
versial. They underline the need for diversity rather than universality. They stress
128 C. Roche
the importance of participation and an understanding of context in order to assess

impact. This, they argue, is not consistent with ‘pre-set formulae’ and universal
standards.
ALNAP (Active Learning Network for Accountability and Performance in
humanitarian action) was established in 1997 and is dedicated to improving the
quality and accountability of humanitarian action by sharing lessons, identify-
ing common problems and, where appropriate, building consensus on approaches
(ALNAP date). ALNAPmembers include organisations and experts from across
the humanitarian sector, including donor, NGO, Red Cross/Red Crescent, UN, and
independent academic organisations.
ALNAP was instrumental in establishing the Tsunami Evaluation Coalition
(TEC), established eight weeks after the earthquake, by a group of predominantly
humanitarian agencies met in Geneva. The TEC’s intention was to focus on examin-
ing systemic problems in the humanitarian sector, concentrating at the policy rather
than program level. According to the TEC ‘this independent learning and account-
ability initiative represents the most intensive study of a humanitarian response since
the Rwanda multi-donor evaluation in the mid-1990s. It is also the first time since
then that the sector has sought to scrutinise itself as a whole’. The TEC is managed
by a Core Management Group (CMG) of agencies and the ALNAP Secretariat is
the facilitating platform for the
The TEC evaluation (Telford et al. 2006) of the first 2 years of the Tsunami
response concluded:
several ‘quality initiatives’ have emerged in the last decade . . . [d]espite important steps,
the lack of quality enforcement mechanisms means that the same problems keep reappear-
ing in emergency responses . . . [t]he recurrence of many of the problems . . . suggest that
the various quality initiatives are not having a sufficient impact. The quality delivered by
a normal business is driven by its customers. The same model of quality control does not
operate in the aid sector. The biggest potential driver for quality should be feedback to
the donor public on the quality of an agency’s operations. Public knowledge is often lim-
ited, however, to the materials produced by organisations’ communications departments
and/or media that concentrate either on these agency sources or on single dramatic issues
rather than presenting a comprehensive analysis of the situation . . . A regulatory system is
needed to oblige agencies to put the affected population at the centre of measures of agency
effectiveness, and to provide detailed and accurate information to the donor public and
taxpayers on the outcomes of assistance, including the affected populations’ views of that
assistance.
These findings have focused further efforts on exploring how accountability
to affected populations can be improved, in particular through the Humanitarian
Accountability Partnership.
The Humanitarian Accountability Project was set up in 2001, and like the Sphere
Project was the result of a stated commitment by humanitarian NGOs to address
concerns regarding their lack of accountability. It followed the Humanitarian
Ombudsman Project which, in response to a recommendation in the multi-donor
evaluation of the Great lakes response, explored the feasibility of establishing a
regulatory mechanism and complaints procedure. Field testing however concluded
that given the diversity of actors involved and the contexts in which they worked,
as well as poor regulatory environments that an Ombudsman would not deliver
improvements in accountability (Davis 2007).
In the first instance HAP explored ways of improving accountability by field
testing different approaches (HAP date). HAP was succeeded by the Humanitarian
Accountability Partnership in 2003 which focuses on developing practice through
a small network of members. HAP membership requires a formal commitment to
uphold HAP’s Principles of Accountability developed through five years of action
research and field trials. HAP felt however that ‘the principles of accountability
did not provide a sufficient basis to enable a consistent and coherent approach to
compliance monitoring and validation of an agency’s humanitarian quality man-
agement system’. It has therefore developed the HAP Standard in Humanitarian
Accountability and Quality Management. This standard comprises a set of
auditable benchmarks that seek to assure accountability to beneficiaries (see text
box below).
Benchmarks for the HAP 2007 Standard
1. The agency shall establish a humanitarian quality management

system
2. The agency shall make the following information publicly available to
intended beneficiaries, disaster-affected communities, agency staff and
other specified stakeholders: (a) organisational background; (b) human-
itarian accountability framework; (c) humanitarian plan; (d) progress
reports; and (e) complaints handling procedures
3. The agency shall enable beneficiaries and their representatives to partici-
pate in programme decisions and seek their informed consent
4. The agency shall determine the competencies, attitudes and development
needs of staff required to implement its humanitarian quality management
system
5. The agency shall establish and implement complaints-handling proce-
dures that are effective, accessible and safe for intended beneficiaries,
disaster-affected communities, agency staff, humanitarian partners and
other specified bodies
6. The agency shall establish a process of continual improvement for its
humanitarian accountability framework and humanitarian quality manage-
ment system
Sphere and HAP are seen as mutually supportive: HAP providing the account-
ability framework that strengthens the practice framework provided by Sphere and
vice versa. HAP also complements, but differs from other international initiatives
including People in Aid, the Quality Compass, and the Active Learning Network
130 C. Roche
(ALNAP). Where People in Aid focuses on improving human resources in emer-

gency response, and has developed a code of good practice, the Quality Compass is
an initiative that seeks to implement quality assurance during project management
and evaluation stages.
HAP now also has a system of accreditation for their members or affiliates for
assessing their compliance with the HAP Accountability Principles and Standards.
Furthermore it is now exploring the development of a decentralised accreditation
system in collaboration with suitable NGO networks and associations, such as
ACFID in Australia.
The accreditation process assesses organisations on the above benchmarks
for which a number of requirements have been defined with specific means
of verification for each. Below are the requirements and benchmarks for
Benchmark 5.
Again it is interesting to reflect on whether if the word ‘humanitarian’ was remo-
ved from the above benchmarks whether they would be equal applicable and
relevant to INGOs in general or development or advocacy agencies more
specifically.
Arguably this sort of approach to quality might contribute to responding to the
critique raised by Horton (chapter “The Epistemic Problem: Potential Solutions”,
this volume) and others. This is because (a) it helps to ensure that agencies adhere
to appropriate standards; and (b) doing so constitutes a proxy for more difficult to
collect impact data. Clearly this assumes that following such standards will ulti-
mately increase the probability that positive impact is achieved. HAP’s experience
and research suggests that meeting such standards is reliably correlated with achiev-
ing certain positive effects. However it is clearly the case that this hypothesis needs
to be regularly tested.
3.4 The Evidence Base for Development and Advocacy work
In the development field NGO performance has also been subject to a number of
studies and reviews most of which find that drawing definitive conclusions is diffi-
cult either because of the paucity of the data and/or the complexity of the evaluative
task.1 Nevertheless many of these studies (Kruse et al 1997, Riddell 1990, Riddell
and Robinson 1992) draw similar conclusions to a recent UK National Audit Office
review (2006):
Notwithstanding limitations in assessing impacts, the evidence we collected from project
visits and from documentation review indicated that engaging with CSOs [Civil Society
Organisations] is leading to benefits, in line with development objectives. Monitoring
reports found that 80 percent of these projects were largely but not fully success-
ful in achieving their intended outputs. Project evaluation reports commissioned by
DFID showed that working with CSOs had been beneficial, through strengthening
the institutional capacity of CSOs and others, improving advocacy and delivering
services.
Requirements and Means of Verification for HAP Benchmark 5
No. Requirement Means of verification
5.1 The agency shall ask intended Demonstrate that findings from
beneficiaries and the host consultations have been incorporated
community about appropriate ways into complaints-handling procedures.
to handle complaints
5.2 The agency shall establish and • Review the documented
document complaints handling procedures;
procedures which clearly state: • Review samples of complaints
• The right of beneficiaries and to verify that complainants have
other specified stakeholders to been able to understand and use
file a complaint; the procedure;
• The purpose, parameters and • Review budget, contracts and
limitations of the procedure; support provided to humanitarian
• The procedure for submitting partners to implement this
complaints; requirement;
• The steps taken in processing • Interview field staff, affected
complaints; community members and/or
• Confidentiality and intended beneficiaries about their
non-retaliation policy for perceptions and the adequacy of the
complainants; procedures
• The process for safe referral of
complaints that the agency is not
equipped to handle;
• The right to receive a response.
5.3 The agency shall ensure that • Review strategy and activities for
intended beneficiaries, affected raising awareness among these
communities and its staff groups of their right to file a
understand the complaint and the procedures to use
complaints-handling procedures • Review documents about the
complaints-handling procedures
made available to intended
beneficiaries
• Interview staff, affected community
members and/or intended
beneficiaries to verify awareness of
and adequacy of the procedures
and confidence in its integrity
5.4 The agency shall verify that all • Review a sample of both pending
complaints received are handled and processed complaints, to check
according to the stated procedures the integrity of the system
• Review reports on the integrity of
the complaints-handling process
132 C. Roche
(continued)
No. Requirement Means of verification
5.5 The agency shall establish and • Review the procedure and samples
implement an effective and safe of complaints made
complaints handling mechanism • Interview staff to verify awareness
for its staff, consistent with the of procedure and confidence in its
requirements set out in 5.2 integrity
In general however many of these studies also suggest that there is (even) less evi-
dence to demonstrate that NGOs were achieving their objectives efficiently, or that
the benefits of the projects they support were sustainable. Indeed in one of the most
comprehensive and recent overviews of Foreign Aid, Roger Riddell (2007) suggests
that ‘the gaps in data and information which limit our ability to form judgments
about the impact of official aid are even more severe in the case of NGO aid-funded
development initiatives’ However a recent macro-economic study by the IMF sug-
gested that ‘[o]ur results show that NGO aid reduces infant mortality and does so
more effectively than official bilateral aid. The impact on illiteracy is less signif-
icant’ (Masud and Yontcheva 2005). They go on to suggest that ‘[f]rom a policy
perspective, if we take infant mortality as a “flash indicator” of the living conditions
of the poor . . . in our study, NGO aid appears more effective in reaching out to the
poor and vulnerable populations, and therefore donors who have chosen to chan-
nel aid through NGOs have made the right choice. This chapter is the first empirical
study on the effectiveness of NGO aid at the macro level and confirms the legitimacy
of NGO actions. Whether NGO aid can be easily scaled up is a different question
that should be answered in future research.’
It is also increasingly recognized that the nature of NGO development work is
changing with a growing involvement in both north and south in seeking to influ-
ence policy processes as well as engagement in direct service delivery (even if for
some agencies this has always been the case). However as some note ‘[s]urprisingly,
there is very little systematic research on how successful NGOs are in influencing
policy processes, especially from the point of view of those actually involved in
the policymaking process in the South’(Chowdhury et al. 2006, p. 8). Whilst it is
argued that there are particular methodological challenges with assessing the impact
of advocacy work, and in particular the attribution of a given policy outcome to a
specific agencies lobbying or campaigning, observers suggest that the success of
NGOs in their advocacy work is ‘far less than has often been claimed by many,
but, almost certainly far more than NGOs’ harshest critics maintain’ (Riddell 2007
p. 291). Riddell suggests that there are three reasons for reaching this conclusion:
(e) that the information and publicity generated on issues where NGOs have cam-
paigned for a long time such as debt and aid levels has almost certainly
influenced decision making;
(f) that there is hard evidence in specific policy areas and environments such as
trade policy in the UK and aid levels in Luxemburg to indicate that NGOs
specifically have influenced policy makers;
(g) that there is evidence of the targets of NGO lobbying, such as the World
Bank and the UN, openly admitting the effects of NGO campaigning on their
decisions.
However he goes on to say that there is ‘much we still do not know. There remains
a significant gap between the growing number of advocacy, lobbying and campaign-
ing initiatives undertaken by NGOs and the number of rigorous assessments of their
results.’(Riddell 2007, p. 298).
What emerges then is a picture that is mixed. Firstly the evidence base is poor,
even if specific studies can point to a number of positive achievements as well
as weak performance, particularly in large scale emergency response situations.
Secondly, that an increasingly important part of NGO work (i.e. advocacy and
supporting civil society organizations to influence others) is not subject to sys-
tematic review.2 Thirdly that what evidence that does exists suggests that NGOs’
ability to promote systemic change in the face of other powerful forces is limited3
and that whilst they may promote positive change in people’s lives this may well
not be sustained. In many ways this mirrors the picture for aid in general which
also is recognized as: being difficult to evaluate; having a mixed record; and
being but one part of a much bigger policy picture which includes trade, for-
eign policy and relations and more recently an increasingly important security
agenda.
So why is the evidence so weak? And why is change seemingly so hard to
achieve? I will now explore some of the challenges and approaches to learning and
accountability in INGOs which may also offer some clues to why existing quality
initiatives may have not delivered what some expected they might.
4 INGOs, Learning and Accountability
Many of the issues that emerge from looking at the Red Cross Code, Slim’s ethical
questions and the quality initiatives summarised above relate to how INGOs know
either before, during, or after, the event whether they are doing, or have done, a good
job, or will do a good job in the future, and how this is then communicated. And
this is also the fundamental ethical question that Keith Horton challenges us with
i.e. there is no moral or ethical obligation to give to charities (a la Peter Singer) if
there is little or no evidence that they do a good job. Now a lot has been written on
the problems of assessing project, program and organisational performance which
I don’t want to repeat here. However there are a number of key problems that it is
worth recalling.
134 C. Roche
4.1 The Measurement Problem

There is a long and complex debate on the merits, possibility and desirability
of what can or cannot be measured in emergency, development and advocacy
work. Sometimes a (useful) distinction is made between measurement (which often
demands agreed baselines, metrics, and measuring tools) and verification (the ability
of different observers to verify if an event or process has or has not occurred). There
are also important debates about, amongst other things, what constitutes rigour in
qualitative and quantitative research, the degree to which objectivity and indepen-
dence is possible, and who is best placed to make judgements about achievements
and progress made.4
This debate also takes place within a sector that is becoming clearer that local
ownership, participation, experimentation, action-learning and adaptation to local
circumstances are critical success factors. As many agencies now note, there are
no magic bullets or single approaches; contexts are too varied, and ‘one size fits
all’ proscriptions should be avoided. Much of what needs to be done therefore in
situations like this cannot be routinised but requires the skill and judgement of
imperfect information by skilled decision makers. In addition many of the outcomes
from such processes have low ‘specificity’ i.e. awareness-raising, capacity build-
ing, appropriate consultation, participatory evaluation and research etc. In these
cases of low specificity which are highly discretionary, monitoring performance
through centralised bureaucratic process using pre-determined indicators and stan-
dards can be highly problematic and lead to perverse incentives.5 As Hirschman
(1970) and others have argued in these situations there are important alternatives,
namely peer review and what he calls ‘voice’ options, which attempt to give
those who are ultimately meant to benefit from a particular service or process
much better access to information and decision-makers in order to hold providers
responsible.
Finally there are different costs associated with different levels of rigour
and research. Producing rigorous research that is liable to convince others who
may be skeptical of the findings is expensive. NGOs generally need to adopt
pragmatic approaches which combine: measurement where feasible or necessary
and verification where more appropriate; independent or external evaluation with
internal self-assessment; the triangulation of information from different sources in
order to make judgements about progress made.
4.2 The Attribution Problem
There is also a long and complex debate about issues of attribution. In recent years it
has been recognized that in many cases it might be liberating—given the complexity
of the real world—to think more about reducing our uncertainty about the effects of
what we do, as opposed to be able to prove causality. This is particularly the case
where we are talking about more about assessing a project, or program, or agency’s
contribution towards broader processes, and where contribution analysis may be

appropriate (Hirschman 1970). This approach essentially seeks to make a plausible
argument for the contribution that an agency might make to a given process—
and exploring alternative explanations for the observed results—and then through
greater processes of transparency and triangulation to invite others to contest or
enrich that argument.
4.3 The Aggregation Problem

All aid organizations, and many others in the not for profit sector, struggle with the
challenge of aggregating information. This is in part because if something cannot
be turned into a comparable quantitative dimension then it is difficult if not impos-
sible to add it up, and also because even it is possible to quantify things—through
for example rating or scoring processes—that adding them up may simply not tell
one very much (Roche and Kelly 2003). This reality has led a number of agencies to
focus their quantitative aggregation on the input and output levels i.e. funds invested,
numbers of people reached, numbers of organizations supported, whilst looking for
other means of making sense of outcomes and impacts i.e. through qualitative pro-
cess such as the Most Significant Change Approach (Davies 1996), through more
in depth impact evaluation or research in a limited sample of project or programs,
or by collaborative ventures across agencies i.e. working together on more holistic
approaches to monitoring and evaluation.
Of course in some aid organizations, such as AusAID or the World Bank, and
some NGOs, the search to aggregate outcomes and impact into overall organiza-
tional performance metrics or results databases continues unabated often through
the use of performance ratings e.g. percentage of projects achieving their goals.
4.4 The Ownership Problem
Monitoring and Evaluation is often seen as a dry technical exercise that is sepa-
rate from the ‘real work’. It can also be seen as punitive and threatening. In these
circumstances it is unsurprising that it is also therefore not seen as an opportunity
for real learning—particularly in relation to mistakes made—as this might lead to
funding being stopped or staff being laid off. At an organizational level the growing
tendency to be ‘transparent’ has led to more evaluations being published on web-
sites for external audiences. In competitive aid environments, and ones in which
there may be political interest in undermining the reputation of development agen-
cies or NGOs, this may lead to either more bland evaluations and/or more shadow
reviews and informal evaluations that remain internal. Evaluation, except amongst
some evaluators, is rarely seen by NGO staff and partners as a means of ‘speaking
truth to power’.
136 C. Roche
4.5 The Responses

There seem to be three broad and very crude caricatures of responses to these
problems that can be described.
The first might be called the ‘Scientific’ or’ ‘Measurement’ solution. This
solution starts from a premise that whilst there are challenges with measurement,
attribution and aggregation that they can be overcome by improved planning
and design, greater standardization, better monitoring and evaluation and greater
alignment of individual and organisational objectives. In this sense what is required
is: being clearer about our objectives; having more clearly defined and standardized
indicators and targets; having more rigorous processes of establishing baselines
and measuring progress; establishing control groups or quasi-control experiments
to resolve the attribution problems; and being able to ‘roll-up’ the data collected in
standardised formats in corporate Management Information Systems which can gen-
erate information about overall organisational performance. In doing this agencies
will be able to collect objectively verifiable ‘evidence’ about what works and what
does not, and thus be more publicly accountable. Staff and organisational ownership
is largely generated by aligning individual and organisational performance through
a hierarchy of Key Performance Indicators and the creation of ‘incentives’ to
achieve them.
The second might be called the ‘Adaptive Systems’ solution. This solution
starts from a premise that ‘the course of events is complex beyond our capacity
to readily understand it, and well beyond our capacity—whether as individuals or
collectives—to predict or control’ (Monk 2007). However it is also recognised that
‘the intractable dilemma we face is that we must, nevertheless try to understand,
anticipate and be able to adapt to the course of events—but anticipation is not pre-
cisely the same as prediction, and adaptation is very different from control’(Monk
2007). This approach recognises the ‘open-system’ of which aid agencies are a part,
that there are multiple, unpredictable, linear and non-linear processes which help
determine the outcomes of what we do, and that context is important and that adap-
tation to local circumstances is key (Monk 2007). The emphasis in this approach is
to: emphasise organisational values and principles; to invest in staff skills and com-
petencies but to leave front line staff in particular the freedom to make decisions
(including ‘wrong’ decisions) and learn from that; to value relationships and dia-
logue with others; to encourage action-learning and feedback from others; to look
for patterns and archetypes to make learning and exchange easier. In this model
accountability is more about being able to test adherence to values, to seek real time
feedback, and to engage in multi-stakeholder dialogues and peer to peer reviews
about performance which are transparently reported (Burgis and Zadek 2006).
Ownership in this model is encouraged through rewarding adaptation, experimenta-
tion and personal and organisational learning as well as meeting more ‘democratic’
standards.
The third solution might best be described as the ‘Political’ or ‘Power-based’
solution. This solution starts from the premise that a major part of the humanitarian,
development or advocacy agenda is essentially about ‘creating the checks and bal-
ances that ensure that the less privileged and powerful can challenge and re-shape
the dynamic of social power’ (Bonbright 2007). It therefore follows that one of the
key processes that constitutes effective development is empowering communities to
hold us and others to account. Oxfam’s experience (Roche et al. 2005) suggest that
this requires (a) communities to have the information and the capacity to do this, (b)
that appropriate venues and means exist for these views to be freely expressed, and
(c) for there to be clear and transparent mechanisms of grievance and redress. This
in turn requires aid agencies to be clear and transparent about: who we are, what we
have done and what we plan to do; how we seek views, engagement and feedback
on our plans, strategies and performance; a commitment to learning and adapta-
tion in the light of the feedback received; and how complaints or grievances will be
genuinely addressed. All key elements of accountability as defined by organizations
such as One Word Trust (2006) and similar to many of the elements stressed in the
HAP principles and standards. In this approach monitoring and evaluation becomes
‘owned’ by staff because it becomes central to their developmental work rather than
being seen as a technical add-on or the outcome of a performance framework driven
by organizational targets and indicators (Table 3).
Ultimately these different perspectives—and others that one could elaborate
through different metaphors—originate from different disciplinary traditions and
paradigms related to different ways of knowing and understanding. Indeed some
have referred to ‘paradigm wars’ within the evaluation literature as in other fields
(Crawford 2005). However there is a growing recognition both within the evalua-
tion field (Patton 1997), in the humanitarian arena (Slim 2003), and amongst NGO
observers that a methodological pluralism is necessary because:
• different parts of an organisation, or different processes within an organisation,

may need different approaches
• balancing different stakeholders needs will also mean balancing the different
ways in which they seek to know and understand things and
• different approaches can also provide a cross-checking mechanism to verify or
refute findings and thus enhance the robustness of conclusions. However it is
clear to most observers that despite this there is a tendency for measurement
solutions to remain the predominant approach in the sector in general, and the
challenges of operationalising this pluralism remain (Crawford 2005).
For many INGOs, I would argue, that whilst there have been some attempt to
balance measurement and open systems approaches, through a mix of more rigor-
ous program design and evaluations as well as innovations using more participatory
and multi-stakeholder methods, genuine attempts to really experiment and bring
more power based processes into the mix are few and far between, particularly at an
organisational level. This gap is particularly problematic as a measurement based
approach that ignores power dynamics is liable to privilege the powerful,6 who
have a greater ability to manipulate the tools of measurement. Equally an adaptive-
systems approach can also fall prey to the voices of the privileged to dominate if it
is insufficiently grounded in a rigorous power analysis.
138 C. Roche
Table 3 Advantages and disadvantages of different solutions
Adaptive systems
Measurement solution solution Power-based solution
Advantages • Allows comparison • Places agency in a • Makes power relations

and aggregation across broader context and central to the
agencies work, ‘eco-system’, approach,
• Seen as more rigorous • Builds flexibility and • Seeks to make those
and objective by many, learning into an who agencies seek to
• Encourages strong organisation, benefit privileged
planning and design, • Encourages stakeholders and
• Makes management context-specific equips them to hold
control easier, solutions, others to account,
• Assists in upward action-learning and • Strengthens and gives
accountability and ‘let-go’ management, priority to bottom-up
reporting • Allows for multiple accountability,
accountabilities to be • Links M&E to the
balanced, development process
Dis- • Can create perverse • Can create perverse • Can create perverse
advantages incentives, and incentives, and incentives,
undermine downward undermine proactive • Makes comparison
accountability, work, and aggregation
• Tends to emphasise • Makes comparison harder,
planning and and aggregation • Can lead to
standardisation over harder, manipulation and
adaptation and • Can ignore power vexatious complaints,
context-specific relations, • Deliberately cedes
solutions, • Can underestimate some management
• Can be inflexible and importance of good control,
rigid, planning and design, • Can undermine
• Tendency to ignore • Makes management potential for
power relations as they control harder, collaboration with
are difficult to • Can lead to ‘multiple multiple stakeholders,
measure, accountability • Can institutionalise
• Tends to disorder’ and endless conflict and zero-sum
overemphasise the negotiation processes, thinking
agencies place in the • Can be seen as
scheme of things and ‘wishy-washy’ and too
under-emphasise role post-modern
of others,
• Can be seen as a
‘western’, positivist
approach by others
However I also recognise that to the ‘non-expert’ as Horton calls them, who
are increasingly seeing calls for more measurement in their lives (for example at
work) or in the lives of their children (for example at school), alternative approaches
may seem overly complex or simply a means of obfuscation by the profession.
There remains therefore an important communication challenge, as much as a
methodological challenge that needs to be addressed.
5 So What?
So what does all this mean when it comes to the question of providing potential
supporters of INGOs with the information they need to satisfy themselves that the
agency or agencies to which they might give, are actually doing a good (enough)
job?
Firstly and most obviously it means that INGOs cannot assume that all potential
supporters will want or need the same thing. Just as there are different intellec-
tual traditions, which will inform how people think about their decision-making, so
there are different personal approaches to what makes for effective accountability,
relationships and trust.
Secondly I think it means being more up front about the challenges INGOs face
and more humble about what we can deliver. If this is not done, as Isbister suggests
in his chapter “Whose Impact, and Is It All About Impact?”, (this volume), then
the ever increasing rhetorical claims for what aid can achieve will return to haunt
us if and when they are not achieved. As has been suggested for many years if we
continue to suggest that aid always works or that it is more important than other
things such as trade policy, foreign policy or national government policies, then we
not only leave ourselves open to any polemic attack on aid as a result of its failure
(Riddell 1984), but also fail to properly portray the reality of what we experience on
the ground. This is something that will be hard for individual agencies to do alone;
it is something that the sector as a whole needs to debate more fully.
Thirdly this needs to be done in a way that whilst honest does not undermine
the political momentum created by Make Poverty History, or the commitment to
MDGs which remain an important means of holding the rich countries to account.
Furthermore it needs to be done in such a way that it does not naively play into the
hands of those, such as the Institute of Public Affairs in Australia7 or the American
Enterprise Institute in the USA, who would use any additional information from
INGOs to further their attack on progressive agendas. As a number of authors have
recently reminded us the debate about accountability is not neutral, it is located
within a political arena in which different interests are represented and conflict
(Jordan and van Tuijl 2006, Newell and Wheeler 2006).
Fourthly it means engaging with supporters in ways that reinforces account-
ability to other stakeholders rather than distorting it. Isbister in chapter “Whose
Impact, and Is It All About Impact?”, this volume suggests that the pressure to
demonstrate impact is contributing to an ever increasing bureaucratization of NGOs
based on a linear cause and effect model. There is also growing evidence that
the power of INGO donors, particularly institutional donors, is creating incentives
for field staff to pass on and impose onerous and sometimes inappropriate plan-
ning, reporting and evaluation processes to their partners and allies (Wallace and
Chapman 2006). This is not dissimilar to what has been reported for bi and multilat-
eral agencies and their relationship with recipient governments (World Bank 2003),
and the way this can in turn reduce these governments’ accountability to their own
citizens.
140 C. Roche
Finally it means coming up with solutions that provide the right incentives for
what Bill Easterly (2006) calls ‘searchers’ in our organizations to keep exploring
how best to make a difference, rather than incentives for the ‘planners’ who can
deflect organizations into unproductive processes.
So what might be some practical initiatives to make this happen?
5.1 Making the Link Between Donor and the Front-Line

Much Closer
Whilst first hand experience and direct exchange between donors and communities
is usually hugely effective in bridging a divide which is often filled by INGO pro-
cesses which can distort and obscure the reality, it is not realistic or cost-effective
to think that this is a major solution. However there are perhaps means of at least
lessening the distance these might include:
• Using staff or partner blogs and stories more systematically, as some agencies
are now doing (see MSF date, Oxfam GB 2007), which can give a grittier, more
realistic less marketing oriented flavour of work on the ground,
• Using supporters who do get to engage directly with staff and communities to act
as more independent guarantors of organisational quality with other supporters,
• Equipping partners and communities with the ability to self represent their achi-
evements, challenges and assessment of the support they receive. Experiments
in Oxfam with participatory video (Braden 1998), and a recent collaboration
between Global Voices on-line and Witness (Global Voices 2007) who have
established a Video Hub Pilot using the model of You-Tube whereby groups
and communities can post evidence of human rights abuses on to the web, are
examples of this,
• Developing Storytelling. A number of people in different sectors are looking
at storytelling e.g. Steve Denning (2007) in the private sector and the World
Bank, Rick Davies’ ‘Most Significant Change’ approach (Davies and Dart 2005);
and David Cooperrider’s Appreciative Inquiry method (2007). Storytelling brings
immediacy and can make connections quickly, and is increasingly being explored
as an impact assessment method.
5.2 Providing a Better and More Honest Evidence Base

The debate about methodology tends to camouflage the fact that good evidence—
whether based on a positivist, open systems or power based criteria—about INGO
performance is hard to find, partially because it is dispersed and hidden, and partly
because much may not exist. Dealing with the lack of INGO transparency will help
here—see below. However a number of other things will need to happen if this
evidence is to be constructed, these might include:
• Experimenting and adapting existing social accountability mechanisms such as

the Uganda Participatory Poverty Assessment Program, the Right to Information
Campaign and Citizen Report cards in India, Participatory Budgeting in
Brazil, the Publish What You Pay initiative with Mining Companies,—which
INGOs often help support to hold governments and private sector companies
accountable—to empower communities to hold us to account.
• Developing more effective relationships with Australian and local research orga-
nizations who can provide methodological support as well as ongoing third party
feedback.
• Developing more collaborative, multi-stakeholder processes whereby commonly
sought outcomes as well as individual contributions can be jointly assessed.
• Ensuring that impact studies explore not only the immediate changes that have
been brought about by a given INGO initiative, but the degree to which those
changes are liable to be sustained in the light of larger and more powerful forces.
Collinson et al. (2002) provide a useful review of how this might be done more
effectively done in the humanitarian situations. This is important in being able
to illustrate that whilst NGO work may succeed on its own terms it is situated
in broader contexts and effected by processes (often the policies, practices and
behaviours of government, multi-lateral agencies or private sector companies)
that may undermine it. This can enable INGOs who are prepared to be more
honest about their achievements, to explain to their supporters both the limits
of what they can achieve whilst at the same time galvanizing some of them at
least to mobilize against those more malevolent forces which may be amenable
to broader advocacy and campaigning.
5.3 Promoting Greater Transparency, Learning and Accountability

INGO transparency has been assessed by the One World Trust (2006) as worse than
many private sector companies and multilateral organizations. It is not routine for
INGOs to place evaluations on their websites, to have clear and public grievance and
complaints mechanisms—particularly for the communities and groups they seek to
support. Whilst in some countries there have been some advances—in Australia
NGOs through their umbrella organization ACFID have one of the most effective
Code of Conducts (BOND 2006) and means of assessing compliance through the
Code of Conduct Committee—such examples are few and far between. Steps that
could be taken include:
• Explore the simplification and bringing together of a number of the principles,

codes and standard which INGOs are currently signatories to, in order to provide
a clearer benchmark of quality which is more accessible to public consideration,
as well as to staff and partners. This needs to be done in ways that avoid the risks
of ‘introducing deeply unrealistic measures and systems of quality, improvement
142 C. Roche
and accountability into agencies and contexts which simply cannot absorb them,
handle them or benefit from them’ (Slim 2005).
• Developing a clearer understanding of what the barriers to Organisational
Learning are in INGOs8 . Recent research amongst British INGOs by the INGO
umbrella group BOND found that over 50 percent of NGOs and donors surveyed
felt that a radical departure was needed to overcome the ‘stuckness’ of the sector
on the issues of learning (Goold 2006).
• Agreement amongst the larger INGOs to develop a common policy for the treat-
ment and public disclosure of evaluation findings, which is realistic about the
risks, but bold in its thinking,
• Re-opening the debate about the establishment of an independent Aid
Ombudsman which allowed for legitimate complaints to be effectively addressed,
and vexatious complaints rejected. Whilst there has been a debate and research in
the humanitarian arena, the development and advocacy areas need to also explore
the issue,
• Debating more rigorously the merits of greater regulation. The TEC evaluation
of the Tsunami recommends that this is the only way to ensure greater account-
ability. Is that the case, or will the result be greater uniformity, less innovation,
and a likelihood that the sector will be further dominated by a few larger players?
6 Conclusion
This paper has explored how some International NGOs have attempted to assess
their performance and communicate that to others. It has reviewed in particular how
this has been done in the humanitarian field and suggests that this has relevance for
INGOs long-term development and advocacy efforts.9 This has included looking
at some of the principles, codes of conduct and ethical frameworks that have been
adopted as well as different approaches to learning and accountability that have
emerged in the sector more broadly.
I have argued that this experience suggests that no single approach or method is
going to provide INGOs with a simple ’off the shelf’ solution to the problem of how
to assess their performance. Furthermore meaningfully communicating about their
performance to their stakeholders in ways that can satisfactorily indicate that they
are actually making a real contribution to alleviating poverty or suffering is by no
means straightforward.
However what the experience of the sector, and of other sectors, suggests is that
whilst any solution will need to be relatively pragmatic, and aligned to the circum-
stances of individual agencies, there are perhaps some common dimensions that
need to be addressed. These include:
• Being clear about the ‘shadow’ or dark side of the work that aid agencies are
necessarily involved in, and also being clear about the potential biases we bring to
our work.
• Defining the core values and principles that guide the agency in all its work
and how its decision-making processes deal with situations when they might
be compromised, or when there are potential clashes or trade-offs between
them.10
• Establishing processes to put those values and principles into practice and
engaging different stakeholders in assessing how well this is done.
• Developing Organisational Development strategies that create environments and
cultures that value ethical deliberation and learning based on past experience, and
that provide the right incentives for staff to be open to feedback from key stake-
holders, and in particular those the agency seeks to benefit, as well as providing
then with the skills to do so.
• Exploring some more radical ways of lessening the distance between donors
and the communities we seek to benefit, developing a more honest evidence
of what we do and do not achieve, and developing greater transparency and
accountability.
I would argue that if INGOs could address these areas then much of the
concern expressed by Horton would diminish. This is for a number of reasons.
Firstly it would ensure that the average donor to NGOs would have an enhanced
ability to make judgments about their performance. Secondly it would give a
greater voice to those who are ultimately meant to benefit from the work of
NGOs. Thirdly it would lead to the kinds of institutional changes11 that would
increase the probability that NGO staff have the skills, capacities and incen-
tives to not only perform well but communicate this in a clear and honest
manner.
Finally if INGOs are to pursue these types of approaches then this means link-
ing this work to efforts to promote change at individual, organisational and sector
levels.12 Being open to feedback on one’s performance, and adjusting to it, needs
a level of individual and organisational maturity that demands certain skill sets and
competencies. This demands managers and leaders in the sector to focus on this, as
well as on creating an environment that is porous and transparent enough to accept
feedback, even when it is unpalatable, and which provides positive incentives for
staff to go out of their way to elicit that feedback and adjust their work as a result.
Such changes in organisational culture will also need efforts to persuade peers across
the sector to think about changing in a similar way. Being a sole tall poppy can be
very hard.
Notes
1. See the last section of Ronald’s paper for more on the aid effectiveness debate.
2. See however NAO attempts in the report cited above to develop tools to explore this, as well
as Chapman et al. (2001), Roche (1999), California Endowment (2005).
3. See for example Kruse et al. (1997) who conclude for example that ‘what is clearly prov-
ing most difficult is to introduce processes which have a positive and more systemic
144 C. Roche
impact on the status of women in societies and cultures perceived to be discriminatory.’

And that ‘even the best projects are insufficient to enable the beneficiaries to escape from
poverty’.
4. See for example Pretty (1994).
5. See Fukuyama (2004, pp. 74–91) for a longer discussion on this.
6. See Isbister’s chapter “Whose Impact, and Is It All About Impact?”, this volume.
7. See Ellis’ chapter “The Ethics of Taking Sides”, this volume for more on ‘the attack on
advocacy NGOs in Australia.’
8. See Kelly’s chapter “Ethical Behaviour in Non-government Organisations”, this volume on
the importance of organizational learning for NGOs.
9. This is not to suggest that there are not also important distinctions between humanitarian
and development work—see Harmer and Mcrae (2004) But it also acknowledges that the
‘securitisation’ of the aid agenda is increasingly making these distinctions blurred.
10. See Isbister’s chapter for more on the importance of adherence to values.
11. See the conclusion to Kelly’s chapter “Ethical Behaviour in Non-government Organisations”,
this volume on why this is important
12. See the chapter “Ethical Behaviour in Non-government Organisations”, this volume by Kelly
on some of the organisational and structural challenges faced by the sector.
References
ALNAP. www.alnap.org accessed 1 July 2007.
Bonbright, D. 2007. The Changing Face of Accountability. Talk at the International Seminar on
Civil Society and Accountability, Montevideo, 16 Apr.
BOND. 2006. A BOND Approach to Quality in Non-Governmental Organisations: Putting
Beneficiaries First. Report by Keystone and Account Ability for the British Overseas NGOs
for Development (BOND).
Braden, S.. 1998. Video for Development a Casebook from Vietnam. Oxford: Oxfam.
Burgis, T. and S. Zadek. 2006. Reinventing Accountability for the 21st Century, Account Ability,
London.
California Endowment. 2005. The Challenge of Assessing Policy and Advocacy Activities:
Strategies for a Prospective Evaluation Approach. Los Angeles, CA: The California
Endowment.
Chapman, J. and A. Wameyo. 2001. Monitoring and Evaluating Advocacy: A Scoping Study. Lima:
Preval, http://www.preval.org/documentos/00545.pdf .
Chowdhury, N., C. Finlay-Notman, and I. Hovland. 2006. CSO Capacity for Policy Engagement:
Lessons Learned from the CSPP Consultations in Africa, Asia and Latin America, Working
Paper 272, Overseas Development Institute, London.
Collinson, S., M. Bhatia, M. Evans, R. Fanthorpe, J. Goodhand, and J. Stephen. 2002. Politically
Informed Humanitarian Programming: Using a Political Economy Approach, HPN Paper 41.
London: Overseas Development Institute.
Cooperrider, D. http://appreciativeinquiry.case.edu /. Accessed 1 July 2007.
Crawford, P. 2005. Aiding Aid: a Monitoring and Evaluation Framework to Enhance International
Aid Effectiveness. PhD thesis, University of Sydney.
Crocker, D.. 1996. Hunger, capability and development. In World Hunger and Moral Obligation,
eds. A. William and L. Hugh. Englewood Cliffs: Prentice-Hall, NJ, place, reprinted in Hugh
Lafollette (ed.) Ethics in Practice, Routledge, 2nd Ed.
Davies, R.. 1996. An Evolutionary Approach to Facilitating Organisational Learning: An
Experiment by the Christian Commission for Development in Bangladesh. Swansea: Centre
for Development Studies.
Davies, R. and J. Dart. 2005. The Most Significant Change’ (MSC) Technique: A Guide to Its Use,
MandE News, Cambridge. http://www.mande.co.uk/docs/MSCGuide.htm .
Davis, A.. 2007. Concerning Accountability of Humanitarian Action, HPN Paper 58. London:
Overseas Development Institute.
Denning, S. http://www.stevedenning.com/LeadersGuide.html . Accessed 1 July 2007.
Easterly, W.. 2006. The White Man’s Burden. London: Penguin Press.
Edwards, M.. 1996. International Development NGOs: Legitimacy, Accountability, Regulation and
Roles. London: Commission on the Future of the Voluntary Sector.
Fukuyama, F.. 2004. State Building: Governance and World Order in the 21st Century. Ithaca, NY:
Cornell University Press.
Galtung, J.. 1969. Violence, peace and peace research. Journal of Peace Research 6(3):
167–191.
Global Voices. http://globalvoicesonline.org/-/human-rights-video /. Accessed 1 July 2007.
Goold, E.. 2006. Working with Barriers to Organisational Learning. London: BOND.
HAP. www.hapinternational.org . Accessed I July 2007.
Harmer, A. and J. Mcrae. 2004. Beyond the Continuum: The Changing Role of Aid Policy in
Protracted Crises, HPG Report 18. London: Overseas Development Institute.
Hilhorst, D.. 2005. Dead letter or living document? Ten years of the code of conduct for disaster
relief. Disasters 29(4): 351–369.
Hirschman, A.. 1970. Exit, Voice, Loyalty: Responses to the Decline in Firms, Organisations and
States. Harvard: Harvard University Press.
Horton, K.. 2004. Aid and bias. Inquiry 47(6): 545–561.
Howard, R.. 2007. Where anti-Arab prejudice and oil make the difference. Guardian Wednesday
16 May.
IFRC. 1994. Code of Conduct for the International Red Cross and Red Crescent Movement and
NGOs in Disaster Relief. Geneva: International Federation of Red Cross and Red Crescent
Societies. http://www.ifrc.org/publicat/conduct/code.asp .
IFRC. 2003. World Disaster Report 2003. Bloomfield, NJ: Kumarian Press.
Jordan, L., van Tuijl, P. eds. 2006. NGO Accountability: Politics, Principles & Innovations.
London: Earthscan.
Kruse, S.-E., T. Kyllonen, S. Ojanpera R. Riddell (1997), Searching for Impact and Methods: NGO
Evaluation Synthesis Study. Report prepared for the OECD/DAC Expert Group on Evaluation.
Helsinki: Department for International Development Cooperation, Ministry of Foreign Affairs.
Masud, N. and B. Yontcheva. 2005. Does Foreign Aid Reduce Poverty? Empirical Evidence from
Nongovernmental and Bilateral Aid, IMF working paper WP/05/100, 20. Washington, DC:
IMF.
Medecin Sans Frontieres. http://www.msf.ca/blogs /. Accessed 1 July 2007.
Monk, P. 2007. Fantasy, Paranoia, Enthusiasm and Reality. Griffith Review Ed 16:Unintended
Consequences.
National Audit Office. 2006. Working with Non-Governmental and other Civil Society
Organisations to Promote Development. London: Department for International Development.
Newell, P. and J. Wheeler. 2006. Rights, Resources and the Politics of Accountability. London: Zed
Books.
One World Trust. 2006. Global Accountability Report. London: One World Trust. http://www.
oneworldtrust.org/?display=index_2006 .
Oxfam GB. 2007. Oxfam Blogs. Oxford: Oxfam. http://www.oxfam.org.uk/applications/blogs /.
Accessed 1 July 2007.
Patton, M.. 1997. Utilization-Focused Evaluation: The New Century Text, 3rd edn. London: Sage
Publications.
Pretty, J.. 1994. Alternative systems of inquiry for a sustainable agriculture. IDS Bulletin 25(2):
37–49.
Riddell, R.. 1984. Foreign Aid Reconsidered. London: James Currey/Overseas Development
Institute.
Riddell, R. 1990. Judging Success: Evaluating NGO Approaches to Alleviating Poverty in
Developing Countries, Working Paper No. 37, Overseas Development Institute, London
146 C. Roche
Riddell, R.. 2007. Does Foreign Aid Really Work? Oxford: OUP.
Riddell, R.C. and M. Robinson 1992. The Impact of NGO Poverty Alleviating Projects: Results of
the Case Studies. Working Paper No.68, Overseas Development Institute, London.
Roche, C.. 1999. Impact Assessment for Development Agencies. Oxford: Oxfam GB.
Roche, C., N. Kasynathan, and P. Gowthaman. 2005. Bottom-up Accountability and the Tsunami.
Paper prepared for the International Conference on Engaging Communities, Brisbane, 14–17
Aug.
Roche, C. and L. Kelly. 2003. Evaluating the Performance of Development Agencies: Perspectives
from NGO Experience. Paper for World Bank Conference on Evaluating Effectiveness;
Challenges and the Way Forward
SPHERE. 2004. Humanitarian Charter and Minimum Standards in Disaster Response.
http://www.sphereproject.org/content/view/27/84/lang,English /. Accessed 1 July 2007.
Singer, P. 1999. The Singer Solution to World Poverty, http://people.brandeis.edu/˜teuber/
singermag.html . Accessed 13 June 2007.
Slim, H.. 1997. Doing the right thing. Disasters 21(3): 244–257, Sep.
Slim, H.. 2003. Chapter 1. In World Disasters Report 2003. Bloomfield, NJ: IFRC, Kumarian
Press.
Slim, H. 2005. Idealism and Realism in Humanitarian Action. Talks given at the ACFID
Humanitarian Forum, Canberra, 5 Oct.
Telford, J., J. Cosgrave, and R. Houghton. 2006. Joint Evaluation of the International
Response to the Indian Ocean Tsunami: Synthesis Report. London: Tsunami Evaluation
Coalition, http://www.tsunami-evaluation.org/NR/rdonlyres/0B60502D-167D-478C-82DF-
1961FCB48A8A/0/ExSum.pdf . Accessed 13 June 2007.
Vaux, T. 2002. The Selfish Altruist: Relief Work in War Famine and War. London: Earthscan.
Wallace, T. and J. Chapman. 2006. An Investigation into the Reality Behind NGO
Rhetoric of Downward Accountability. Oxford: INTRAC, http://www.ngopractice.org/docs/
INTRACpaper.pdf . Accessed 13 June 2007.
World Bank. 2003. Toward Country-Led Development: A Multi-Partner Evaluation of the
Comprehensive Development Framework, Washington, DC.
Whose Impact, and Is It All About Impact?
Jamie Isbister
The present catch phrase around the community development

sector in Australia and overseas is: ‘what impact are you really
having?’ Which is quickly followed by: ‘And how do you know?’
Both questions are clearly important, for the obvious reasons that if you are not
having a positive impact then why are you doing what you are doing? And secondly
if you don’t know then you open yourself up to being seen as misleading the public
through the assertions made in your public material.
However, this paper proposes that this intense focus on demonstrating ‘our
impact’ is beginning to have negative effects on how international NGOs approach
development, who they recruit and more concerning their values. The problems and
issues I raise come from a practitioner’s perspective of managing the challenges and
balancing the expectations placed on an international humanitarian and development
agency.1
The title and argument I propose is in itself controversial. In a world of increasing
competition, drive for ‘performance’ and profit, and increasing demand for measure-
ment and certainty, questioning whether it is all about impact seems heretical. The
title also opens me, and to a more limited extent the sector, to criticism and asser-
tions of ‘see the NGOs are neither interested in impact or professional enough to
understand what it is’
Acknowledging this risk, I believe it is important we are open about the chal-
lenges in our work, because in the past, and possibly still now, we have over
simplified what it is that development agencies do and at times overstated our
achievements. This has been primarily motivated by the pressure to increase
J. Isbister (B)
Caritas Australia, Sydney, NSW, Australia
e-mail: jamie.isbister@ausaid.gov.au
Jamie is the International Programs Director for Caritas Australia and has worked in the interna-
tional human rights and development field in Australia, Asia and Latin America for the past 15
years.

148 J. Isbister
fundraising but also NGOs have tried to distinguish themselves from government
or large UN development agencies that can be seen as bureaucratic and ineffective.
As a result, we have painted ourselves into a corner. We have presented the case
that alleviating poverty is as simple as giving money and trusting us, and people
have understandably taken us at our word. As we have grown in influence and size
we have increasingly been asked to back up our claims and demonstrate our impact.
This is fair and understandable.
The difficulty this brings is that in having presented a rather unrealistic picture of
what it is we do we are now left attempting to demonstrate the very tangible impact
we have on the basis of this artificial picture. I argue that this leads to agencies focus-
ing on the wrong things. Time is spent developing increasingly bureaucratic systems
and processes that are based on assumptions that development is achieved through
a simple cause and effect relationship which we can control. You give money to an
‘aid agency’ who then gives it either directly or indirectly to the poor.
Paul Ronalds in his article tackles the ethical and moral reasons why we should
do something in response to injustice and massive inequities within and between
countries. Peter Singer (2002) has also put an argument that citizens in developed
countries who are materially wealthy have a moral responsibility to do something
to alleviate abject poverty. I accept much of Paul and Peter Singer’s argument that
there is a moral responsibility for action; the debate is what form and motivation
should shape such action? The issues raised in this paper focus primarily on these
questions.
1 Defining Impact
The quickest way to put someone on the back foot, particularly somebody working
in the area of social or human development, is to ask them to demonstrate the impact
they are having. In some areas/professions there are relatively easy and quite tangi-
ble measures of your impact or performance: fund managers may be measured by
the quarterly performance of the stock portfolio they manage against the stock mar-
ket average; for a car company it could be based on the number of cars manufactured
and sold in a year.
In other areas particularly social programs, while the question is appropriate
and important it is often difficult to succinctly answer. For example, social workers
working with drug dependent clients the impact could be how many people’s lives
have been turned around in the past year? In the area of community development
the question could be how many communities or people have been empowered or
been able to gain access to basic health care? All very difficult questions to quickly
answer and demonstrate in a tangible and relatively short time frame.
Measuring corporate performance or impact such as the number of cars manu-
factured in a year can be done in a relatively controlled and stable environment of
a factory. Assessing the impact of a social workers program assisting drug depen-
dent clients can be impacted by a wide range of external factors such as the client
group whether they are first-time users or long term drug users, government policy,
availability of drugs on the street or access to drug rehabilitation programs.
Whose Impact, and Is It All About Impact? 149
The debate in Australia at the moment about performance based incentives for
teachers in state schools reflects some of the same difficulties of defining the impact
we want from our schools and what can be attributed to the teacher. Systems can be
developed that measure basic literacy or numeracy skills, but does this translate to a
good education? Are these the right indicators for measuring the impact of a good
teacher? (Invarson et al. 2007)
Similarly in the world of art and literature the question could be asked what
defines good literature or a good writer, is it what sells the most? If so, Dan Brown
would be one of the greatest writers the world has seen, or is it the critics who define
what good literature is and its contribution to society?2
I would argue that great art or writing is not determined on its release or
even at a certain snap shot in time, but rather emerges from the collective reflec-
tions, debates and judgments of readers and observers over time. Similarly the
impact of social programs cannot always be determined in a certain snap shot of
time, but often emerges following and reflection of how a program has impacted
on a range of other factors that influenced, and were influenced by, such an
intervention.
The danger in placing such a heavy emphasis on being able to demonstrate
the impact of a social program at the beginning is in how it shapes the thinking
and approach individuals and organizations have in planning and implementing
programs. The design is overly skewed towards outcomes or issues that can be
most easily measured and which we feel most confident will occur. An example
could be a program that is aimed at protecting old growth forests and biodiver-
sity in the Pacific. Excessive concern about demonstrating the impact of a program
from the beginning can result in identifying activities we can in the main control
and measure such as a reforestation program that increases the number of trees
in a certain area over the next year. This is tangible, relatively easy to imple-
ment, and can be measured. However, if the government regulation of the timber
industry and ownership of traditional community landholders is not changed or
increased then in the longer term any re-forestation program could be simply feeding
the ongoing demands of the timber industry, resulting in the further marginaliza-
tion and loss of land to communities and the ongoing destruction of the regions
biodiversity.
Developing a program that aims to reform government forestry policy and
strengthen community ownership and entitlement to land and forests is difficult
to achieve. There are a wider range of factors outside the control of an orga-
nization or program, the change is also not so tangible and hence difficult to
measure.
2 Whose Impact?
Having outlined some of the difficulties that can arise in being able to define what
the impact being sought by a development program is, there is an additional real
challenge that comes in assessing the impact of social development programs which
relates to who defines the impact being sought.
150 J. Isbister
The impact of Australia’s policy of integrating Indigenous Australians into

‘Australian’ society during the late 1800s and first part of the 1900s would today
be seen by most people as tragic. However, the impact was seen quite differently
at the time it was being implemented and also quite differently depending on who
was consulted. The number of kids who could be provided with a white education
and be employed into white jobs defined the impact being sought and achieved by
the government of the time.3 The impact from the perspective of Aboriginal people
affected by the program was described as one of alienation of family, marginaliza-
tion and exclusion from society, and in many cases having to deal with the life long
trauma of separation from family and culture (Human Rights and Equal Opportunity
Commission report 1997 p. 32).4
This terrible disjuncture between the planned impact of a program and the actual
impact of program highlights that defining impact is both a difficult concept, par-
ticularly in relation to social programs, but can also be defined quite differently
depending from where you are defining it. Therefore, it is critical when arguing that
programs must be able to demonstrate impact, to be clear who defines that impact.
I believe that the disjuncture in defining the impact of programs between donors
and communities affected, or benefiting from, programs is increased the greater
the separation between them. This separation occurs at three main levels: geo-
graphic, cultural and bureaucratic. For international development agencies we are
constrained at all three levels: we implement programs around the world, across
very different cultures and through increasingly bureaucratic systems. This creates
a number of problems. Firstly, that impact is being defined in isolation from the
communities we support, and secondly that the impact is defined in a way that is
most easy for donors (NGOs) to control and measure.
The social programs implemented by government and church agencies in the
early twentieth century that separated aboriginal children from their families were in
the main, done out of goodwill. As international development agencies also acting
with good will we need to be always asking the question as to how are we are
ensuring that we are not supporting or implementing programs based on what we
believe is positive impact, but is in fact resulting in the further marginalization and
poverty of the communities we support.
The response by many international humanitarian agencies to the Asia Tsunami
is a more recent example, where the pressure to have ‘impact’ and to ‘be seen to be
doing something’ resulted in responses being developed and implemented with lim-
ited community involvement or participation (Disaster Emergency Committee Dec
2005 p. 3). Large scale humanitarian crises, by their very nature, require quick deci-
sions and effective response. However, history has shown that unless programs have
strong involvement of affected communities the sustainability and positive impact
is seriously undermined. This obviously creates major problems for international
development agencies that distinguish themselves on the basis of having the capacity
to promote strong community participation and involvement in programs.
The other side of who defines impact is: who claims it? As outlined earlier, social
programs are inherently complex with a wide range of factors influencing the impact
on people’s lives and communities. There is a real danger that in agencies focusing
their energies on proving their impact, leads to the problem of error of attribution,
where agencies claim impact or achievement that they are either only a part of, or
worse, that has come about independently of their involvement. This is the inherent
danger that you tell me what you want and I will find it.
3 Have We Got the Right Accountability Mechanisms?

Too Much Counting, Not Enough Accountability
If impact is one side of the coin, accountability is presented as the other. Who are
you accountable to? Are you accountable enough? And are you promoting vertical
and horizontal accountability? Accountability, like impact, is a self-evident principle
for development agencies: if you’re not accountable to your donor over time they
will leave; and if we are not accountable to the communities we say we support we
contradict our mission statement of empowering the poor and alleviating poverty.
Accountability is rarely a linear process, there are often multiple relationships
and accountabilities that international development agencies have: donors, media,
government, boards, staff, local partners and the communities that programs are
supporting. It is important that an agency acknowledges each of these relationships
and develops appropriate ways for being accountable to them.
What is often hidden, however, when accountability is discussed is how it is heav-
ily overlaid by power and control.5 Accountability mechanisms are often skewed to
those who hold the power and money. If agencies aren’t clear about their values
they are in danger of falling into the temptation of delivering more of what they
believe donors, government or individuals, want, , rather than the local communities
supposedly benefiting from a program (Eyben 2005).
If one looks at the range of accountabilities facing International NGOs today,
nearly all of them are oriented at three levels:
• donors, governments and individuals;

• the wider international development sector;
• or the NGO itself.
There are very few genuine accountability mechanisms to communities directly

affected or benefiting from international development programs. The audits, accred-
itations and even evaluations are processes primarily used to report to their
constituents and funders, not communities they are aiming to support.
For a long time NGOs in general have been looked upon as ‘do-gooders’ who
lack genuine professionalism in their work. I believe the sector has changed, grown
and matured significantly over the last ten years. People are recruited for their
specific skills; it is a competitive sector and salaries increasingly reflect the expe-
rience and skills agencies seek. However, the hangover remains and has resulted
in an inferiority complex where the sector attempts to prove its professionalism
152 J. Isbister
by establishing codes of conduct, accreditation processes, internal ‘accountabil-

ity’ mechanisms, results based management and complex program management
systems,6 all which require lengthy documentation of its work, to prove to itself
and the wider community that it is ‘professional’ in what it does.7
Many of these measures are in themselves important; however, when all placed
together they result in the development of complex and increasingly bureaucratic
systems to meet a range of expectations and requirements. It is arguable whether
such systems provide the accountability donors want. From my experience, they
don’t promote accountability to the communities that programs are aimed at bene-
fiting, and I believe they risk taking an agency further away from being accountable
to such communities.
Having an overly heavy focus on codes, standards, accreditation and in-depth
logical planning and implementing processes can result in the belief that meeting
these standards is the same as good development. Agencies then begin recruit-
ing staff who bring these skills, and performance is based on the compliance
with these systems above equally, if not more important, skills required to sup-
port good development initiatives, including managing across different cultures, the
capacity to adapt to changing contexts, working autonomously, managing complex
relationships and being prepared to take risks.
Rosalind David and Antonella Mancini (2004) highlight how ActionAid iden-
tified through an independent evaluation that there were major contradictions
between the agency’s goal of fighting poverty together and the systems and
procedures through which the organization was run. This review of the agency’s
systems and processes led to the development of the Accountability, Learning and
Planning System (ALPS).8 The evaluation highlighted that there were significant
resources spent on documenting accountability upwards, but little accountability to
communities.
There are some important initiatives in the sector that are attempting to over-
come this lack of accountability to communities we support. A number of these
have been outlined in Chris Roche’s paper, but two that are worthy of specific men-
tion are the Listening Project which focuses on understanding the real concerns and
needs of partners and providing them an independent structure to provide feedback
on programs and the Humanitarian Accountability Project (2007) aimed at holding
organizations accountable to the communities they support. Again there are already
signs that agencies are taking some of these initiatives on board but also quickly
turning accountability to communities into bureaucratic check lists to be signed off
as part of approving the funding of a program.
4 What Does this Mean for How We Work?
In outlining the challenges and dangers in trying to define or assess impact is not to
say that assessing the impact of development programs is not relevant. Rather, it is
an attempt to push impact assessment away from the increasingly heavy emphasis on
demonstrating what a specific program or agency has achieved, towards a direction
focused more on understanding the wider context we are part of and how we may
be positively or negatively influencing this.
An example of how measuring the impact of my efforts on a development chal-
lenge can be futile, is in the response to the very real problem of climate change.
It is impossible for me to know what impact I alone may be having in response to
climate change by riding my bike to work. Efforts to determine this are going to be
hugely time-consuming and almost definitely futile. However, I can examine what
impact we as a society are having in response to the climate change fight, and try
to better understand what is working and whether my actions seem to be making a
positive contribution or not and then shape my future actions accordingly.
Agencies should therefore focus impact assessment on achieving four critical
goals in descending priority.
• Ensuring a shared understanding of the impact being sought;

• Understanding the wider context a program is being implemented in and how this
should shape the program an organization is carrying out;
• Assessing jointly with the community how the program is or isn’t contributing to
the shared goal/impact;
• Determining in participation with the local community or partners how well an
agency is aligning its work with it values.
5 Increasing Transparency
If this wider focus on impact is to be realized agencies need to be more upfront
about explaining their work and the complexities of humanitarian response and
international development.
Having argued that most organizational accountabilities are orientated upwards
towards donors, I think it is true that there is a substantial gap between what agencies
are providing donors and what the informed donor or public actually wants. Energy
is often focused on providing fundraising materials or overly complex reports on
projects. I believe donors want a better understanding of the development issues
they are responding to with partners and why they are responding in the way they
are and then providing a clear summary of the achievements and difficulties arising
in the program.
Agencies need to stop presenting messages that we are the primary agents of
change and can largely control the community development process. At best we are
catalysts of change that often occurs incrementally and in unexpected ways.
Agencies spend increasing time carrying out evaluations of their work and doc-
umenting future plans; sharing these on our web-sites would be a starting point in
giving a clearer and fairer picture of our work.
Increasing transparency probably doesn’t mean doing more or providing more
information but being smarter and clearer about why we are doing the things we
presently do.
154 J. Isbister
6 Clear About Values, Approach and the Change We Promote

The way we often determine who we support politically or who we want to work
with is determined by listening to their values. Are they values that resonate with
me? Would I feel comfortable having this person represent me? As international
development agencies we are all guided by certain values that may be shaped by
religious, philosophical, and/or international human rights principles.
There is the need for us to be able to communicate these values and principles
clearly and outline how we work to achieve them and are assessed against them.9
We need to have opportunities through governance but also through membership
or supporter structures where we are required to explain our values and how we
promote and work to put these into practice in our work. An example of this could
be around the principle of promoting community participation in our work. How
well is this realized in deciding where we work, what we will focus on and how a
program is designed and implemented?
The values that guide agencies’ work are one of the greatest assets that we as
agencies have. They arise from the experience and tested wisdom of work with
communities and our predecessors over time. However, values are not worth a great
deal if they are not put into practice. The move towards greater measurement has
taken us away from a focus on how we realize these values and are accountable to
them. As agencies I believe that evaluations and accountability frameworks need
to place a greater emphasis on demonstrating alignment with our values. Donors
and communities should be aware of the values that guide an agency’s work and be
able to gain access to evaluations and reviews that demonstrate how agencies have
or haven’t been able to promote such values. Agency evaluations should also allow
communities to report on how well a program or agency has aligned itself to the
community’s own values. This could simply be a section in program evaluations
that is dedicated to agency values. Individuals or communities could directly speak
to or present on the agency’s commitment to these values in their work. Examples
could be how effectively the community has been involved in planning and imple-
menting a program or how effectively women and men feel they have been included
or benefited from a program.
7 Promoting More Dynamic Models of Planning

and Implementation
As mentioned at the beginning of the paper my concerns about the increasing
emphasis on measuring impact and ‘countability’ over and above values and learn-
ing is how it starts shaping the agency’s plan and work. The most concerning aspect
is the move towards increasingly bureaucratic and linear approaches to community
development. Increasing time and emphasis is being placed by agencies and govern-
ment donors on the belief that if enough effort and planning go into understanding
the problem and developing a clear and logical plan for dealing with it, this will
ensure good impact. This is based on a belief that societies are predictable.
This creates a false frame for understanding development, and consequently
organizations create the belief and expectation that they will have the capacity to
change others and their situations in unrealistic ways. The political scientist Aaron
Wildavsky (1988) highlighted how societies were not rational and that the most
effective ways to deal with crises was to enhance the capacity of communities to
adapt and change rather than try to predict and prevent crises. The emphasis is on
incremental change and a capacity for ongoing adaptation.
8 Conclusion
I have argued in this paper that the increasing emphasis on agencies proving ‘their
impact’ is resulting in NGOs moving away from the more community participatory
values that have traditionally shaped their work. As NGOs are challenged to prove
their impact they have begun to frame systems and approaches that result in designs
that will ensure greater control as to what is being supported, and consequently
ensure we can measure the ‘impact’ that is being sort.
We live in an environment that increasingly looks for certainty and it is important
that we do not confuse assessing impact with the temptation of seeking certainty.
Development means change and we must encourage communities, individuals and
organizations to take the risks required for important change to occur.
Development agencies must be encouraged and pushed to see themselves as part
of a bigger whole rather than the main actor. It is critical that we acknowledge
that we don’t have the answers and that adequate time needs to be given to our
relationships with local partners and communities if real learning and change is to
effectively happen.
Notes
1. I state this in acknowledging the natural bias this brings to my argument.
2. Dan Brown is the author of The Da Vinci Code
3. The 1937 Commonwealth-State Native Welfare Conference stated that efforts of all State
authorities should be directed towards the education of children of mixed aboriginal blood at
white standards, and their subsequent employment under the same conditions as whites with a
view to their taking their place in the white community on an equal footing with whites.
4. There is a wide range of stories about the impact and consequences of the government’s policy
of assimilation in the Stolen Generations report.
5. This could more crudely be said as he who pays the piper chooses the tune.
6. Chris Roche outlines in his chapter a number of the Australian and international codes,
standards etc that have been developed.
7. This is a generalization about the sector and agencies there are undoubtedly agencies that have
not attempted to comply with codes of conduct.
8. This is referred to by Chris Roche in his chapter.
9. Chris Roche in his chapter has highlighted a range of accountability frameworks based on
values.
156 J. Isbister
References
David, R. and A. Mancini. 2004. Going Against the Flow. London: Institute of Development
Studies.
Eyben, R. 2005. Donors’ learning difficulties: Results, relationships and responsibilities. IDS
Bulletin 36(3): 98–107, September.
Human Rights and Equal Opportunity Commission. 1997. Bringing Them Home. Canberra:
Commonwealth of Australia.
Humanitarian Accountability Project. 2007. www.hapinternational.org/Humanitarian
Accountability project.
Invarson, L., E. Kleinhenz, and J. Wilkinson. 2007. Research on Performance Pay for Teachers
Australian Council for Education Research, http://www.dest.gov.au/NR/rdonlyres/D477C6A5-
C8EF-4074-8619-FF43059445F8/16287/ACERPerformancePaypaperMar07.pdf.
Singer, P. 2002. One World: The Ethics of Globalization. New Haven, CT : Yale University Press.
The Listening Project. 2007. www.listeningproject.org/.
Vaux, T. Independent Evaluation of the DEC Tsunami Crisis Response, December 2005.
http://www.reliefweb.int/rw/RWFiles2006.nsf/FilesByRWDocUNIDFileName/KHII-
6KZ9HA-dec-sas-31dec.pdf/$File/dec-sas-31dec.pdf.
Wildavsky, A. 1988. Searching for Safety. Edison, NJ: Transaction Books.
Compromised Humanitarianism
Garrett Cullity
The circumstances that create the need for humanitarian action are rarely morally
neutral. The extremes of deprivation and want that demand a humanitarian response
are often themselves directly caused by acts of war, persecution or misgovernment.
And even when the direct causes lie elsewhere—when suffering and loss are caused
by natural disaster, endemic disease or poverty of natural resources—the explana-
tions of why some people are afflicted, and not others, are not morally neutral. It is
those without economic or political power who starve in famines, those who inhabit
the most marginal areas who are killed by floods and landslides, those without
access to basic education and health care who die from easily preventable diseases.
The societies in which these things can happen are unjust, and the injustice of disem-
powerment is both itself one of the universal afflictions of poverty and a condition
of its many other vulnerabilities.
This is now coming to be widely realized by those members of the rich world who
are troubled by the scandal of world poverty. Our understanding has advanced: it
was common 30 years ago to depict severe poverty as a natural phenomenon, caused
by scarcity of resources independently of facts about the justice of social structures.
However, within the community of aid workers, there has surely always been a very
pressing realization of these facts—of the force of the social structures that keep
the poor. For generations of aid workers, the injustice of poverty-creating societies
has been a fact that is not just of theoretical importance in explaining the need that
they try to address. It is also a fact that poses a confronting practical challenge: how
can humanitarianism operate in such conditions without itself becoming morally
compromised?
The problem is that working within an unjust structure can mean reaching accom-
modations with it—accommodations that must be uncomfortable for anyone who is
both clear-sighted and aims to do good. The most visible examples of this concern
the taxes and outright bribes that are often paid to the oppressors of the poor in
order to get relief supplies through; the security that is hired, the equipment that
is rented. However, arguably the more important effects are much larger, and less
G. Cullity (B)
University of Adelaide, Adelaide, SA, Australia
e-mail: garrett.cullity@adelaide.edu.au

158 G. Cullity
visible because they are so large. The humanitarian industry now operates on such a
large scale in comparison with the economies of the poorest countries that, accord-
ing to some critics, it tends to support and entrench the political structures that are
themselves the causes of poverty. I turn shortly to some of the specific respects in
which commentators claim this to be true.
According to the most pessimistic critics, it is too late to ask how these effects
can be resisted. The most despairing analyses of this kind come from those disillu-
sioned former aid workers like Michael Maren who now recoil from the aid business
in deep distaste. The picture Maren paints is one of an industry that has become
corrupted by its own unregulated business imperatives into collaborating with the
oppressors of the poor to deepen the problems they claim to be addressing (Maren
1997, Chapters 6, 8, 9). Western aid agencies compete with each other for donating
customers, selling the illusion of help in order to increase their own income and
thereby their power in the poor countries they dominate. They create artificial aid
economies that draw them into close commercial allegiances with the oppressors
of the people they claim to be helping, and operate without any serious scrutiny of
the inflated claims they make about their own effectiveness. According to Maren,
humanitarianism’s clarity of vision, and even its good intentions, have been lost.
Most commentators and critics are not so bleak. Few are willing to abandon
the enterprise of bringing succour to those who urgently need it; but there are now
intense debates over the principles that should govern this enterprise, and the meth-
ods by which it should be pursued. In Section 1, I start by examining the differences
of opinion concerning the proper role of a ‘humanitarian imperative’ in govern-
ing the work of relief agencies. I then aim in the rest of this essay to show that
moral philosophy can make a helpful contribution to clear thinking about these
issues. It can help us to think about what kinds of humanitarian activity are morally
compromising, and what kinds of moral compromises it could be right to make.
1 The Humanitarian Imperative
First formulated in 1994, the Code of Conduct for the International Red Cross and
Red Crescent Movement and NGOs in Disaster Relief now has more than 400
relief agency signatories.1 The first of its ten principles is that ‘the humanitarian
imperative comes first’. Part of the Code’s explanation of this principle runs as
follows:
The prime motivation of our response to disaster is to alleviate human suffering amongst
those least able to withstand the stress caused by disaster. When we give humanitarian aid
it is not a partisan or political act and should not be viewed as such (IFRC 2007).2
This principle is an object of contention. The different practices and policies of

the agencies that are signatories to the Code show that they interpret it differently.
The International Committee of the Red Cross itself adheres to the policy of ‘neu-
trality’ adopted at its foundation in 1863 as a means of ensuring access to the victims
of war—including, most controversially, during World War II when it refrained from
Compromised Humanitarianism 159
speaking out against Nazism. In contrast to this is the more activist policy of organi-
zations such as Médecins Sans Frontières (MSF), which includes protesting against
human rights violations as part of its mission:
It is part of MSF’s work to address any violations of basic human rights encountered by
field teams, violations perpetrated or sustained by political actors. It does so by confronting
the responsible actors themselves, by putting pressure on them through mobilisation of the
international community and by issuing information publicly (MSF 2007).
– or the more strongly activist organization Médecins du Monde, set up by MSF

founder (and now French foreign minister) Bernard Kouchner, in part because of
disagreements within MSF over Kouchner’s advocacy of military intervention on
humanitarian grounds.
More strongly, the critics of compromised humanitarianism object that when
relief agencies give priority to the alleviation of suffering, this can be naïve, and
worse than naïve. These simple motives can be exploited and the resulting actions
manipulated by others. And it can be plain enough that this is happening to mean
that the aid agencies become complicit in the manipulation. The ‘humanitarian
imperative’ is a trap.
One apparently strong argument of this kind comes from Fiona Terry’s (2002)
account of the manipulation of humanitarian aid by military groups.3 Her cen-
tral example is the control over refugee camps in Tanzania and Zaire by those
responsible for perpetrating the genocidal violence in Rwanda in the mid-1990s;
but her concern is to demonstrate that this was only one instance of a persistent
and predictable pattern seen in many other conflicts in various countries. Terry
writes of the debate within MSF about whether to withdraw from the Rwandan
refugee camps. When aid agencies’ recognition of the ‘humanitarian imperative’ is
being exploited to provide succour and power to the oppressors of a civilian pop-
ulation, she argues, the only politically responsible action—the only way of doing
no harm—may be to withdraw the aid that is being exploited.4 The French section
of MSF in Tanzania, which she headed at the time, did accordingly withdraw from
the camps.
A broader and more far-reaching objection to an apolitical ‘humanitarian
imperative’ is developed by Alex de Waal. According to de Waal (1997, p. 6),
‘a humanitarianism that sets itself against or above politics is futile. Rather we
should seek a form of politics that transforms humanitarianism.’ Drawing on the
analysis of Amartya Sen (1997, 1999), De Waal argues that problems of extreme
poverty are fundamentally problems of disempowerment—of the absence of struc-
tures of political accountability. The intrusion of economically powerful agencies
who are answerable only to their wealthy donors can actually impede the all-
important process of establishing the structures of political accountability that are
needed in order to establish protections against want. Such ‘external involvement,
however well-intentioned, almost inevitably damages the search for local political
solutions.’ (de Waal, 1997, p. xvi) It does so by further reinforcing the structures
that have disenfranchised the poor in the first place. ‘Aid empowers the powerful,
and also creates new political or economic groupings around it. Any established
160 G. Cullity
aid programme creates a local constituency to defend it. As programmes become

entrenched, they tend to become more conservative, and as the donor agency invests
more, it becomes less willing to change the programme or withdraw’ (De Waal,
1997, p. 137).
In apparent opposition to these objections, other commentators see the defence
of the ‘humanitarian imperative’ as itself crucial to aid agencies’ ability to protect
themselves from political exploitation. According to David Rieff (1995, 2002),
the real humanitarian trap is that humanitarians have started to see themselves as
activists.5 But humanitarian NGOs lack the competence and the authority to con-
tribute to political reform. Instead, humanitarianism should confine itself to its
proper role, which is honourable, but limited. Humanitarianism is what we turn
to, inadequately, when those who do have the capacity to protect human rights have
failed (Rieff 1995, p. 8; 2002, p. 281). The dream of a humanitarianism that can
right wrongs is only a dream, and it can be a pernicious one (Rieff 2002, p. 154). It
can add to the vulnerability of the victims of persecution by creating the illusion of
humanitarian protection, as it did in Bosnia (Rieff 2002, p. 153). And it can provide
an alibi for the non-intervention of other powers, by creating the illusion of action,
as it did in Rwanda (Rieff 2002, pp. 161, 170).
So: should the humanitarian imperative be upheld, or not? If so, in what form?
When is it naïve to provide aid, and when is it unconscionable to withdraw it? It
will help us to think about these questions if we start by distinguishing three parts
of morality.
2 Concern, Respect and Cooperation
Most of us think of morality as including three things. The first of them I call ‘the
morality of concern’.6 This covers the various ways in which we ought to respond to
the welfare of others. By others’ ‘welfare’, I mean what is good or bad for them—
what benefits or harms them. The most obvious parts of the morality of concern,
then, are helping others to get what is good for them and avoid what is bad, and
not harming them or preventing them from getting goods—that is, beneficence and
non-maleficence. These are the most obvious parts of the morality of concern, but
not the only ones. When you cannot help a needy person, there can be other things
you ought to do, like expressing your solidarity with them or at least sympathizing
with them; these are part of the morality of concern too.
Secondly, ‘the morality of respect’ comprises the ways in which we ought to
take proper account of the rights and dignity of others. There are various ways in
which we ought to respect others’ autonomous agency, and at least at first glance
they do not coincide with a concern for their welfare. They include respecting peo-
ple’s entitlements to decline needed help, to pursue projects that are misguided or
involve personal sacrifice, and to refuse to be used in the service of others’ welfare.
The morality of respect thus prominently includes various negative obligations of
non-interference with others’ autonomous projects. Respecting others also includes
discharging those positive obligations to respect the rights conferred through volun-
tary acts of commitment. And, more debatably, it may be thought to include a range
of positive obligations to secure for others the conditions in which autonomous
deliberation and agency can be fully exercised.
The third group of requirements and expectations that most of us think of as part
of morality is ‘the morality of cooperation’. This comprises the ways in which we
ought to treat each other as partners in a common project. As a member of a group
that is cooperating to further a goal, I ought to play my part in exerting myself
alongside the other members of the group, and I ought not to take advantage of
the cooperative-spiritedness of others by exploiting them. Again, first appearances
suggest that these requirements go beyond the morality of concern and respect. What
is required of me as a partner in a joint project with you is not that I further your
welfare, nor that I respect your entitlement to conduct independent projects of your
own. It is that I work together with you to a common purpose.
The moralities of concern, respect and cooperation encompass the reasons that
govern three different kinds of relationship we bear to others. The morality of
respect governs our relationship to others as agents who are our equals: it calls
upon us to respond to them as bearers of rights. The morality of concern, by con-
trast, governs our relationship to others not as agents but as patients, who stand to be
affected for better or worse by us: it calls upon us to respond to them as bearers of
welfare. The morality of cooperation governs our relationship to others as partners,
combining our agency together: it calls upon us to respond to them as co-agents in
the performance of a common project.
Moral philosophy offers us a range of different views about the relationship of
these three parts of morality to each other. Many moral theories can be understood
as attempting to show that one of the three is more fundamental than the others,
which can be derived from it. My own sympathies lie with the view that morality
has at least these three, mutually irreducible foundations. However, this is not the
place to pursue that theoretical debate. For present purposes, I shall assume only
that each of the three groups of norms I have just referred to is indeed a part of
morality.
Each of these three parts of morality is relevant to thinking about the moral char-
acter of humanitarian action. Clearly, the core of humanitarianism is a recognition
of the moral imperative of concern—of responding to others’ needs. However, it
is now often pointed out—but the importance of the point bears repetition—that
needy people must be treated not just with concern but with respect. The humani-
tarian action it makes most sense to admire treats people with compassion but not
condescension, and respects them as authors of their own lives, rather than sim-
ply assuming an entitlement to take responsibility for their welfare. The core of de
Waal’s criticism of humanitarian activity, for example, is that it stands in the way
of securing the self-determination that is the most fundamental of poverty’s many
deprivations.
The point that is less often noticed is the bearing that the morality of cooperation
has on the moral character of humanitarian action. My main concern in what follows
will be to explain this.
162 G. Cullity
We can begin by noticing how this is true of each of us as individual contributors

to international aid agencies. My action in donating money to an aid agency is an
action of contributing to a wider collective action of beneficence. It is not so obvious
that it is itself a beneficent action. For it is not obvious that the consequence of
my donating $100 to an aid agency will be that this itself makes any significant
difference to anyone’s welfare: aid agencies do not seem to work that way (Cullity
2004, Chapter 4). I hasten to add that does not undermine the case for thinking that
I morally ought to contribute. But it does affect how that case should be stated, if
we are going to be accurate. We ought to help; and I ought to join in. (I shouldn’t
leave it to everyone else to do what we ought to be doing.) What I ought to do is to
contribute to our collective action of beneficence.
The further important point is that our collective action of beneficence—what
we do together when the aid agency to which we are contributing delivers food
or medical treatment to people who need it—can itself be part of wider col-
lective actions performed by larger groups. There is what all the aid agencies
are doing together. And there is what outside agencies and local governments
may be agreeing to do together. It is on collective actions of these various kinds
and their moral implications that my main focus will lie in the rest of this
discussion.
3 Cooperative Concern
We cooperate when we work together to a common purpose. Collective actions—

that is, actions of which we together are the subject—are actions that there can be
various different reasons for joining in. Sometimes, there are reasons of self-interest:
there may be heavy penalties for non-cooperation, and joining in may be the best
way for me to avoid them. Sometimes, there are other reasons: reasons of concern
or respect, perhaps. However, it is striking that many of us often recognize another,
more direct reason for joining in collective actions. When I see that we ought as
a group to be doing something together, I can treat this as itself a good reason for
joining in. This way of thinking—thinking about what we collectively ought to be
doing, and taking that to settle what I ought to do—is what we are admiring when
we praise people as ‘cooperative-spirited’.
This kind of motivation for contributing to joint action is worth marking out for
special attention, because it is of special and fundamental importance. Someone
who thinks in this way relates to others in a way that goes beyond concern and
respect for them. She deliberates directly as a member of the group, treating reasons
for us as reasons for her. This is not only theoretically but practically important.
There appear to be some things that we collectively ought to do, but there are not
clear reasons of self-interest, concern or respect directing individuals to contribute
to doing them. In those cases, it is practically very important that there are people
who take the fact that we ought to be doing something as a reason for doing it—for
otherwise, it would not be done at all.
The practical problems that cooperative-spirited thinking can help us to solve

are especially prominent in the situations that Derek Parfit calls ‘each-we dilem-
mas’. Parfit describes these slightly differently in different places (Parfit 1984, p. 91;
unpublished). Here, I shall understand them as situations in which:
if each of us does what (given everyone else’s behaviour) would make things better in a
certain way, we together make things worse in that same way.
One much-discussed kind of each-we dilemma is the so-called ‘Prisoner’s

Dilemma’: a situation in which, whatever others do, each person serves his own
interests best by non-cooperation; and yet, if everyone acts uncooperatively, every-
one is worse off than they would have been had they cooperated. Prisoner’s
Dilemmas are each-we dilemmas for self-interest—situations in which we do worse
in terms of self-interest if each does what is self-interestedly best. Situations with
this structure are common, and potentially disastrous. It is in the interests of each
of us to pollute (given what others are doing), but worse for everyone if we all do
so. Faced with such situations, cooperative-spirited people treat the fact that the
group needs individuals to pull their weight in contributing to a worthwhile collec-
tive action as itself a sufficient reason to do so. If enough people think like that, we
can avoid the potential disasters.
There are also each-we dilemmas for altruism, and we need cooperative-spirited
thinking to avoid these too. When other people need help or support and we are
committed to providing it, the following can be true. Even if each of us does the
thing which, out of the alternatives available to him, makes the situation better—
where this means, better for the people who need help, not better for him—we might
collectively be making the situation worse. It might seem hard to see how that could
be true. But again, situations with this structure are common. Given the volume of
traffic on the roads, my children may be safer if I drive them to school rather than
allowing them to walk. That may also be true of every other parent. But our children
may all be less safe than they would be if they all walked to school rather than
being driven. The same structure can affect actions of a more purely altruistic kind.
Suppose a building has collapsed in an earthquake. It may be that the best I can do to
help, given what everyone else is doing, is to dig at the rubble myself to try to rescue
some of the trapped people. That may also be true of every other bystander. But at
the same time, it might be true that we would all do better to coordinate our efforts
and respond in a more organized way. Indeed, in extreme circumstances, it could be
true that our uncoordinated rescue efforts, taken as a whole, are making the situation
worse: we are making the structure more unstable, and no one can hear those who
are trapped. Against that chaotic background, it might still be the case that the best I
can do is to continue digging. Each of us might be making the situation better, even
though we are making it worse.
We need cooperation in order to avoid this kind of problem: cooperation
in collective actions of helping others. And what is needed to sustain this is
cooperative-spirited thinking of the same kind as before—accepting that what we
collectively ought to be doing settles what I ought to do in playing my part.
164 G. Cullity
Let me emphasize this point, because it is easily overlooked. We need

cooperative-spirited thinking not only to create collective actions of cooperation
between altruistically-motivated people, but to sustain them. For each-we dilemmas
can threaten to derail collective altruism even once it has been established. To
appreciate this, return to the example of the earthquake. Suppose now that there
is a coordinated rescue effort. This changes what I ought to do: I should do what
I can to support that collective effort. If our joint effort requires individuals not
to try to shift rubble themselves, I should not do that. And notice that this could
be right even if it remains true that, of all the actions open to me, I would pro-
duce the most good by ‘going it alone’ and shifting rubble by myself. Given the
small scale of my own action, it will not itself undermine the cooperative activ-
ity of the others: it would simply mean that some extra rubble was shifted by me.
However, if enough of us reasoned that way, we would not have any coordinated col-
lective action. The collective action relies on our thinking and acting as members of
the group. The group’s action requires individuals to contribute even when each of
us could do more good by acting unilaterally—and it requires this on the grounds
that we do more good by cooperating with each other than we would if each acted
unilaterally.
Notice the difference between cases in which effective cooperation has been
established, and those in which it has not. The cooperative-spiritedness it makes
sense to admire is a willingness to join in the worthwhile collective actions that
are being performed, and to encourage those actions. It requires me to support our
cooperation. But it does not require me to do, in the absence of cooperation, what
would have made sense had the cooperation existed. I exhibit no failure of coopera-
tive spirit in not unilaterally doing what we all ought to be doing, when no one else
is doing it. If my city is failing to increase its water reservoir capacity as it ought,
I am not morally required to start digging a hole in the ground by myself. Where
there is a collective action that ought to be supported, I should join in and support
it; where there is not, I may need to do my best on my own, while supporting efforts
to establish cooperation.
4 What We Cooperate In
Before we can apply these ideas to the problems faced by humanitarian agencies, a
further point needs to be recognized. Different levels of collective action can coexist.
When we talk about the action that ‘we’ are performing, we need to be clear about
the group to which ‘we’ refers.
When I do something, it can be a contribution to different collective actions—
different things that we do together (for some ‘we’ or other). This can be true in
various different ways. The same group of people might be doing different things,
and a single action of mine might make a contribution to them. My department
might be trying to reach out to the broader community, and also trying to impress the
university administration with its standing in the profession: when I help to organize
a public lecture, that might make a contribution to both of these things. A second
possibility is that a single action of mine can contribute to the collective actions
of groups that only partly overlap. Thus, writing this paper may be a contribution
to the collective scholarly effort of my department, and part of the exchange of
ideas at a conference—in this case, the two groups to whose collective actions I am
contributing are very different: I am their only common member.
There is a third possibility, which is the relevant one here. This is that the collec-
tive actions to which I contribute can be ‘nested’ within each other, in the following
sense. There may be two groups, one of which is wholly contained within the other.
In contributing to one collective action performed by the smaller, contained group,
I may also be contributing to another, different collective action performed by the
larger, containing group.
There are many examples of this kind of nesting of collective action. The work
of aid agencies is one of them. The actions performed by any aid agency are collec-
tive actions, performed by the members of the agency working together. But there
is also what the aid agencies do together: the broader collective actions to which the
actions of each individual agency contribute. The action of the individual agency is
nested within what the agencies together do. We need to be alert to the possibility
that the collective actions at the two different levels—contained and containing—
differ morally. What each agency does individually might be all-things-considered
good, even if what they do together is all-things-considered bad. There is the possi-
bility that each-we dilemmas apply to humanitarian aid. Given the context in which
our agency is working, it may be doing good, by bringing relief to people who des-
perately need and would not otherwise receive it. But even if that is true of what
each agency is doing, it could also be true that the agencies are collectively making
the situation worse than it would be if they acted differently.
Why should we believe that is true? Three different reasons for thinking so were
discussed in Section 1.7 The different problems raised by Terry, Rieff and de Waal
for humanitarianism are each best understood in this way. Let me return to them
now, and show how the general points I have been making about the morality of
cooperation help us to think about the proper response to these problems.
5 Humanitarian Cooperation
Terry shows how privileging ‘the humanitarian imperative’ can render relief agen-
cies vulnerable to exploitation by military groups, and describes the way in which
soul-searching over the question:
To what extent were we responsible for the manipulation of humanitarian aid in the camps?
(Terry 2002, p. 3)
led her section of MSF to withdraw from the Rwandan refugee camps.
However, when this question is asked, who is ‘we’? (Whose soul is being
searched?) The collective action of each individual agency needs to be distin-
guished from the action of all of the agencies together. Separating these two levels
of collective action makes it clear where effective action to resist this manipula-
tion must come from. It requires concerted and politically purposeful action at the
166 G. Cullity
inter-agency level. Indeed, this was one of the lessons of MSF’s withdrawal from
the Rwandan camps. As Terry (2002, p. 4) acknowledges, this action itself attracted
little notice, and the abuse of aid in the camps continued.
This makes it hard to see how it could be right to think that Terry’s own section
of MSF was (at all) responsible for the manipulation of aid in the camps. At most,
we might say that the agencies collectively were responsible for failing to cooperate
to resist the manipulation of their aid.8 We have to distinguish the different levels
of collective action that are relevant here, and this should lead us to distinguish
between three different questions:
1. How ought humanitarian agencies to be cooperating with each other to resist the
manipulation of aid by military groups?
2. What can each agency contribute towards achieving this cooperation?
3. Meanwhile, given that inter-agency cooperation has not yet been achieved, what
is the best thing each agency can do to help the needy?
Could withdrawing aid be the answer to (1)? Even if so, that would not settle what
any individual agency should do. Section 3 pointed out that the morality of coopera-
tion requires us to contribute to collective actions that ought to be performed, when
they are being performed. It does not require now unilaterally doing what would
make sense as a contribution to a collective action that is not being performed. And
that was the situation in the Rwandan refugee camps, where inter-agency cooper-
ation was clearly lacking. In the absence of such cooperation, each agency had to
assess what was the most constructive contribution it could make towards helping.
That question has two parts: (2) and (3) above.
In a situation of this kind, there are significant barriers to thinking that with-
drawing aid altogether could be the right answer to either of these questions. Those
barriers are not insurmountable. One agency’s withdrawal might give the others
enough of a jolt to stimulate inter-agency cooperation now or in the future. Or per-
haps so much of our own agency’s aid is being misappropriated and diverted to
bad ends that the net impact of our own involvement on the needy is negative.9
However, those are the questions that need to be thought about and distinguished.
Failing to do so runs the risk of overlooking the following possibility. Maybe the
humanitarian agencies have made a mistake in allowing a situation to arise in which
a civilian population has been made dependent on the aid coming into this camp.
Maybe even now, it is not too late for them to cooperate in standing up to the
military oppressors of that population. But the right answer to (2) and (3) might
be that each agency ought to stay, do what it can to look after the refugees, and
meanwhile try to support attempts to coordinate politically effective inter-agency
cooperation.
Rieff raises a different kind of concern about the exploitation of humanitarian
work: exploitation not by military groups but by donors—in particular, as an alibi
for the absence of real political engagement. Again, resisting this requires effec-
tive inter-agency cooperation. In the absence of such cooperation, one agency’s
refusal to accept politically compromising donations is simply an opportunity for
other agencies to increase ‘market share’ and influence at its expense. Again, the
three important questions are versions of (1)–(3) above. What can the agencies do
together to resist this? How can our agency contribute to that? And what can we best
do to help the needy meanwhile?
De Waal raises a third, deeper problem. According to him, Western aid agency
involvement in poor countries has the overall effect of reinforcing the structures of
disempowerment that create and sustain destitution. Although de Waal does not dis-
tinguish between the assessment of the actions of individual agencies and the overall
effects that the agencies have collectively, it is plausible to do so. The problem he is
pointing to is actually an ‘each-we dilemma’. Given the unsatisfactory political con-
text in which they operate, the effects of each agency’s actions are good in helping
people who need it; but aid agencies together form an important part of that polit-
ical context. Even if each agency makes bad circumstances better than they would
otherwise be, they might still together be helping to make the circumstances bad.
If this problem—the problem of impeding the establishment of local political
accountability—arises through the effects of the agencies’ actions collectively, then
its solution again requires inter-agency cooperation. This cannot be a ground for
withholding relief from those who need it. As de Waal (1997, p. 220) acknowledges,
‘It is morally unacceptable to allow people to suffer and die on the grounds that
relieving their suffering will support an obnoxious government or army.’ But it is a
ground for raising three questions: again, versions of (1)–(3). What can the agencies
do together to stop their work from impeding the establishment of local political
accountability?10 How can our agency contribute to that?11 And what can we best
do to help the needy meanwhile?
As first described in Section 1, the views of these three writers seemed to be fun-
damentally opposed: where Rieff defends the humanitarian imperative, Terry and De
Waal attack it. However, this discussion should help us to see that the directions in
which they are calling on humanitarian agencies to move are not after all contradic-
tory. Terry and Rieff are drawing attention to different ways in which the practice
of humanitarian assistance must be actively asserted and protected—and that this
requires a sensitivity to the larger political context in which aid flows are seen as
an opportunity to be managed for the benefit of other groups. And De Waal is not
arguing, against Rieff, that humanitarian agencies should act as agents of political
change, but only that they should not act to impede it. ‘The humanitarian imperative’
is only indefensible if it is taken as a licence for political blindness. These writers
draw attention to three different forms that this blindness can take: a blindness to the
exploitation of aid, a blindness to the political limitations of humanitarianism, and
a blindness to the overall contribution that Western aid intervention makes to the
political context in poor countries. Avoiding each of these things requires effective
inter-agency cooperation.
6 Imperfect Cooperation
Cooperation is almost always imperfect, in one of four ways. Sometimes, it does

not exist at all: although we ought collectively to be doing something, we are not.
When it does exist, it is usually partial: some of us are joining in, but others are
168 G. Cullity
not. Amongst those of us who are cooperating together, there will be limitations to
our cooperation: we will be cooperating in some ways but not others. And fourthly,
when you and I are cooperating in a joint activity, we might have different views
about exactly what the aims of that activity should be. Since these imperfections are
all importantly relevant to the situation of humanitarian agencies, I should explain
how the approach I am advocating deals with them.
No one can join in a collective action that does not exist. But when cooperation
that is desirable does not exist, this raises two questions for aid agencies: versions
of (2) and (3) above. The answers to those questions can be in tension with each
other, and that tension is usually strongest for the larger agencies. These are the
agencies with the greatest potential to influence others to cooperate with them, but
they also have the largest capacity to help those in need of humanitarian aid now.
Resources devoted to one of these ends have to be diverted away from the other.
In this situation, the question that must be answered is clear. The expected benefit
of spending resources on securing future cooperation must be weighed against the
expected benefit of using them to help people now. Of course, to say that the ques-
tion is clear does not mean that the answer to it is often obvious.12 But it is at least
clear what we should be asking.
Today, however, the situation of humanitarian NGOs is not one of total non-
cooperation. Following the manifest problems produced by inter-agency compe-
tition in the 1990s—above all, those of the Rwandan refugee camps—significant
initiatives have been taken to formalize cooperation between humanitarian agencies.
To be sure, these fall short of establishing the mechanisms for collective decision-
making that are required to solve the problems discussed above—problems of the
manipulability of aid, and its tendency to promote political inertia.13 Humanitarian
agencies need to operate in this context: one that is changing. It is a context of par-
tial, limited and contested cooperation. How does that affect the application of the
views advocated here?
Whenever a group ought collectively to be doing something, we can divide its
members into two sub-groups: those who are joining in, and those who are not. Of
the first sub-group, we can then ask: ought this group to be doing what it is? If so,
then the action is one that the morality of cooperation tells us to join. Thus, when
cooperation is partial, that does not affect the moral question that should be asked.
What should be asked is whether we should be collectively acting as we are: if so, I
should join in. This can be true when ‘we’ are only a subset of all those who ought
to be cooperating. For a simple illustration of the point, return to the earthquake
example. Suppose 50 of us are able to search for survivors and we ought to cooperate
in doing so. But suppose 20 are unwilling to help. The only sensible question for
the remaining 30 is whether we should cooperate in searching for survivors. If the
answer to this is ‘Yes’, then I should join in.
Similarly, situations in which cooperation is limited can be understood by decom-
posing them into their parts. When we say that cooperation between aid agencies is
limited, what we mean is this. There are some collective inter-agency actions that
are being performed as they ought, and others that are not being performed although
they ought to be. Actions of the first kind give rise to requirements to join in; the
others are instances of desirable cooperation that does not exist, and give rise to
the questions mentioned above. Aid agencies should support the forms of desirable
cooperation that have been established, and support efforts to establish the more
substantial forms of cooperation that are needed. But this does not mean that they
should now do what would be a way of joining in forms of cooperation that do not
yet exist. That could only make sense as a way of securing such cooperation, and
only to the extent that it tends to do that.
The fourth limitation concerns differences of opinion about the nature of our
cooperative action, and what kinds of contributions to it are appropriate. We should
join in forms of desirable cooperation; but what if we disagree about what is desir-
able? A prominent example of this is the divergence amongst signatories to the
ICRC Code concerning whether its ‘humanitarian imperative’ should be understood
as enjoining political neutrality. When such differences of opinion are expressed as
definite divergences of practice, we face a situation in which the cooperative activ-
ity is limited, in the manner just discussed. However, differences of opinion raise a
further important question: the question of who directs a cooperative activity, and
how the parties to such activity decide what to do. The answer is usually (when
this is feasible) that we ought to implement a fair, inclusive process for collective
decision-making, and carry out together the decisions that are reached through this
process. But notice that this invites a further application of what I have been saying
about the morality of cooperation. If we ought together to be deciding in such a way,
then each of us ought to join in such a decision-making process when it exists, and
support efforts to establish it when it does not.
A further point is worth making. When it is not obvious what will best achieve
the goals that we share, pursuing a diversity of approaches often makes good sense.
Many practitioners and commentators claim that is importantly true of the aid
sector.14 This might look like an objection to what I have been saying. However,
to think so would be a mistake. The cooperation that morally ought to be supported
is cooperation in doing the things we collectively ought to do. The case for encour-
aging diversity is that we (collectively) ought to try different approaches and share
the knowledge that this gives us about what does and does not work. When this is
true, each of us should cooperate in supporting that.
7 Two Moral Divisions of Labour
When David Rieff mounts his defence of the ‘humanitarian imperative’, he is mak-
ing a case for a certain moral division of labour—a distinction between the work of
protecting people’s rights from violation, and alleviating the suffering that is caused
when they are violated. He argues that humanitarian action becomes confused, and
loses sight of its own important if limited objectives, if it strays from the second of
these things and attempts the first.
However, as Terry and de Waal point out, there are ways of invoking this division
of labour that lead to a trap. If each humanitarian agency adheres single-mindedly
170 G. Cullity
to serving ‘the humanitarian imperative’, taking the political context as fixed, their
collective activity can encourage bad overall effects. It can be manipulated for the
benefit of those who are violating rights. And more generally, it can stand in the way
of establishing the kind of political accountability that is needed to fight poverty. The
objection is not just that humanitarian action does not itself succeed in protecting
rights. It is that, taken as a whole, it can actually support their violation.
Terry (2002, p. 2) describes this as a ‘paradox’ of humanitarian action: ‘it can
contradict its fundamental purpose by prolonging the suffering it intends to alle-
viate’. I have been arguing that the real ‘paradox’ of humanitarian action is an
‘each-we dilemma’: if each agency does produce the best humanitarian results avail-
able to it in the context, they can still collectively contribute to making things worse
than they otherwise would be. Inter-agency cooperation is needed to address the
political effects that Terry and De Waal alert us to. Seeing this is consistent with
taking Rieff’s warnings seriously. There is a real question how much humanitarian
agencies should properly aspire to by way of rights-protection. But, at the very least,
those agencies need to coordinate their actions in order to minimize the support they
collectively give to rights-violation.
This has revealed a second important moral division of labour: a division between
my work and our work—between the tasks that one agent or group should itself take
on, given what others are doing, and what should be pursued together, through col-
lective action. There are certain morally important goals that we can only pursue
through inter-agency cooperation. And there are certain goals that are appropriate
in the absence of such cooperation, which may include doing what one can to foster
desirable cooperation, but which also include doing one’s best to address people’s
needs in the imperfect situation where desirable cooperation does not yet exist. This
has led me to two main conclusions. First, it can be both consistent and defensible
for an agency to continue to take the actions of pragmatic accommodation that are
necessary to help people in a bad context, while also aiming towards the achieve-
ment of higher-level cooperation that makes those actions unnecessary. Secondly,
where such cooperation does exist, that can make a moral difference to whether my
agency should be aiming to produce the best results available to it. ‘The morality
of cooperation’ can require me to play my part in important collective actions. And
the success of those actions can rely on the cooperative contributions of those who
could each actually produce better results by not cooperating.
In this way, moral philosophy can help to bring some clarity to our thinking about
the issues surrounding the moral character of humanitarianism. However, while
moral philosophy can help to provide us with tools for moral thinking, it requires
practical wisdom and experience to apply them well. Four further important ques-
tions are ones that it requires practical, not theoretical, competence to answer.15
How can aid interventions be designed so that they do not impede the establish-
ment of local structures of political accountability? How can the insertion of large
volumes of aid into the poorest countries avoid the creation of self-perpetuating aid
economies? For any given aid intervention, how far will it serve the purposes of
other political agents? And what kinds of coalitions between aid agencies are both
worthwhile and feasible?
I have been urging that when we answer these questions, we should bear the
following point in mind. Some of the most important compromises anyone has to
make concern the constraint of individual goals for the sake of contributing to what
we can do together. This applies to humanitarian action just as strongly as anywhere
else.16
Notes
1. For two overviews of the Code and the first 10 years of its use, see Walker (2005) and Hilhorst
(2005).
2. In full: ‘1: The Humanitarian imperative comes first. The right to receive humanitarian assis-
tance, and to offer it, is a fundamental humanitarian principle which should be enjoyed by
all citizens of all countries. As members of the international community, we recognise our
obligation to provide humanitarian assistance wherever it is needed. Hence the need for
unimpeded access to affected populations, is of fundamental importance in exercising that
responsibility. The prime motivation of our response to disaster is to alleviate human suf-
fering amongst those least able to withstand the stress caused by disaster. When we give
humanitarian aid it is not a partisan or political act and should not be viewed as such.’ (IFRC
2007).
3. As Terry explains in Chapter 1, this can be used to secure benefits of four main
kinds for military groups: protection from attack; economic benefits through appropri-
ating or taxing aid, or working as contractors to aid agencies; acquiring legitimacy
through aid agencies’ recognition of their authority; and controlling civilian populations.
Compare Barber (1997), Rieff (1995) and, for the perspective of a senior ICRC official,
Schweizer (2004).
4. See also Anderson (1999), esp. Chapters 4 and 5, for an acknowledgement of this pattern, but
also an account, illustrated by case studies, of the strategies that can be effective in resisting it.
5. See also Fox (2001).
6. The label is convenient, but potentially misleading. I do not mean to imply that the knowledge
of others’ welfare should always prompt feelings of sympathetic concern. That seems both
unrealistic and undesirable (given how debilitating it would be).
7. There are others. If humanitarian NGOs compete aggressively against each other to raise
funds for an emergency, each agency might raise more money than it would have done had it
been less competitive (given what the others are doing), but they might collectively raise less
money than if they coordinated their fund-raising efforts through a mechanism such as the
UK Disasters Emergency Committee.
8. It still sounds too strong, though, to say that the agencies collectively were responsible for
the manipulation of their aid. The groups doing the manipulating were responsible for that.
Compare Slim (1997, pp. 246–247).
9. The ‘net impact’ of the continued delivery of aid by an agency should be seen as including not
only the material effects of the supplies provided to the needy population and their oppres-
sors, but also the less easily quantifiable badness of cooperating with those oppressors. The
difficulty in assessing this was a major part of the moral complexity in the situation faced by
Terry and others in the Rwandan refugee camps.
10. De Waal’s (1997, p. 220) own answer includes implementing schemes of regulation and inves-
tigation of NGOs, eschewing claims to have long-term solutions, and refusing to seek the
media limelight for themselves.
11. For an instructive recent case study, see Roche et al. (2005). See also Stephenson (2005).
12. What is obvious, given the importance of inter-agency cooperation, is that devoting no
resources to pursuing it would be indefensible.
172 G. Cullity
13. The Sphere Project, launched in 1997, establishes technical standards for the delivery of
humanitarian relief (www.sphereproject.org). The Humanitarian Accountability Partnership
was initiated in 2001 as a mechanism for self-regulation of the humanitarian sec-
tor (www.hapinternational.org). For commentary on their limitations, see Terry (2002,
pp. 51–4) and Smillie and Minear (2003). For an overview of the UN’s ‘cluster approach’
to coordinating disaster relief, see ActionAid (2006).
14. See for example the materials disseminated by Voluntary Organisations in Cooperation in
Emergencies (VOICE), which lobbies the European Union on behalf of 90 European NGOs—
at http://www.ngovoice.org/.
15. For informed discussions of these questions, see (in addition to the authors cited above)
Minear (2002) and Edwards and Hulme (1996).
16. I am grateful to Chris Roche for his helpful comments on an earlier draft of this essay.
References
ActionAid International. 2006. The Evolving UN Cluster Approach in the Aftermath of the Pakistan
Earthquake: An NGO Perspective. Geneva: International Council of Voluntary Agencies.
http://www.icva.ch/doc00001756.html.
Anderson, M.B. 1999. Do No Harm: How Aid Can Support Peace—Or War. Boulder: Lynne
Rienner.
Barber, B. 1997. Feeding refugees, or war? The dilemma of humanitarian aid. Foreign Affairs 76:
8–14.
Cullity, G. 2004. The Moral Demands of Affluence. Oxford: Clarendon Press.
De Waal, A. 1997. Famine Crimes: Politics and the Disaster Relief Industry in Africa. London:
African Rights & The International African Institute.
Edwards, M. and D. Hulme eds. 1996. Beyond the Magic Bullet: NGO Performance and
Accountability. West Hartford, CT: Kumarian Press.
Fox, F. 2001. New Humanitarianism: does it provide a moral banner for the 21st century? Disasters
25(4): 275–289.
Hilhorst, D. 2005. Dead letter or living document? Ten years of the code of conduct for disaster
IFRC. 2007. Code of Conduct for The International Red Cross and Red Crescent Movement and
NGOs in Disaster Relief. International Federation of red Cross and Red Crescent Societies,
Geneva. http://www.ifrc.org/publicat/conduct/code.asp. Accessed 7 Sep 2007.
New York: The Free Press.
Medecins Sans Frontieres. 2007. Geneva. http://www.msf.org/msfinternational/. Accessed 7 Sep
2007.
Minear, L. 2002. The Humanitarian Enterprise: Dilemmas and Discoveries. Bloomfield, CT:
Kumarian Press.
Parfit, D. 1984. Reasons and Persons. Oxford: Clarendon Press.
Parfit, D. unpublished. Climbing the mountain. http://individual.utoronto.ca/stafforini.parfit/parfit_
-_climbing_the_mountain.pdf. Accessed 7 Sep 2007.
Rieff, D. 1995. The humanitarian trap. World Policy Journal 12: 1–11.
Rieff, D. 2002. A Bed for the Night: Humanitarianism in Crisis. New York: Simon & Schuster.
Roche, C., N. Kasynathan, and P. Gowthaman. 2005. Bottom-Up Accountability and the Tsunami.
http://www.engagingcommunities2005.org/abstracts/Roche-Chris-final.pdf.
Schweizer, B. 2004. Moral dilemmas for humanitarianism in the era of “humanitarian” military
interventions. International Review of the Red Cross 86: 547–563.
Sen, A. 1997. On Economic Inequality, 2nd Ed. Oxford: Clarendon Press.
Sen, A. 1999. Development as Freedom. New York: Alfred A. Knopf.
Slim, H. 1997. Doing the right thing: relief agencies, moral dilemmas, and moral responsibility in
political emergencies and war. Disasters 21: 244–257.
Smillie, I. and L. Minear. 2003. The Quality of Money: Donor Behavior in Humanitarian
Financing, Humanitarianism and War Project. Boston, MA: Tufts University.
Stephenson, M., Jr. 2005. Making humanitarian relief networks more effective: Operational
coordination, trust and sense making. Disasters 29(4): 337–350.
Terry, F. 2002. Condemned to Repeat? The Paradox of Humanitarian Action. Ithaca, NY: Cornell
University Press.
VOICE. www.ngovoice.org/. Accessed 7 Sep 2007.
Walker, P. 2005. Cracking the code: The genesis, use and future of the code of conduct. Disasters
29(4): 323–336.
Aid Agencies, States, and Collective Harm
Ramon Das
The best rule of all for Western helpers is, first, do no harm.
William Easterly, The White Man’s Burden (2005)
In other words, the NGOs should stop being NGOs and
convert themselves into members of socio-political movements.
James Petras and Henry Veltmeyer, Globalization Unmasked
(2001)
1 Introduction
Many critics of aid agencies1 have observed that the main challenge facing such
organizations takes the form of a collective action problem. The most familiar
versions of such problems are those in which each individual is tempted, out of
self-interest, to ‘defect’ from cooperating toward the realization of some collective
project or goal. Prisoner’s Dilemma is the best-known of this sort of problem, and,
if some well-informed critics are correct, this framework is not irrelevant to think-
ing about the morality of aid agencies, particularly when it comes to issues such
as competing for financial donors (Maren 1997, Sogge 2002). However, there is
another, less widely-discussed, class of collective action problem that arises not out
of self-interest but rather out of altruistic motivation. For it is a surprising fact that
collective action problems can arise when everyone is motivated to further the inter-
ests of others no less than when they are motivated to further the interests of only
themselves. The general feature of altruistic collective action problems is that there
is a tension between what it is best (in some overall sense) for us to do together and
what it is best for each of us to do individually.2 This sort of problem is particularly
relevant to NGOs that focus on humanitarianism and development, as opposed to
human rights.
Indeed, the problem of altruistic collective action is related to the now familiar
question whether aid agencies should orient themselves more toward human rights,
R. Das (B)
Victoria University of Wellington, Wellington, New Zealand
e-mail: ramon.das@vuw.ac.nz

176 R. Das
and, perhaps in turn, toward a politically activist agenda.3 As many have noted,
the major humanitarian NGOs have moved closer to a rights-based approach over
the last decade or so. According to Abby Stoddard (2006), this approach grew out
of the widespread concern in the 1990s that humanitarian NGOs were allowing
their activities to be manipulated by armed belligerents, further contributing to war
and perpetuating violence. Many aid workers began to feel that addressing these
problems required a stronger human rights orientation, and that adopting such an
orientation, to be effective, required a collective effort amongst different agencies in
a way that purely humanitarian efforts perhaps did not.4 However, the move toward
a concern with human rights has sparked a ‘back-to-basics’ response from human-
itarian practitioners and observers alike (Stoddard 2006, pp. 22–23). Advocates
of this position, such as David Rieff (2002), have argued that when humanitar-
ian NGOs advocate for human rights or social justice they not only step beyond
their expertise, they also risk making it impossible to continue their humanitarian
activities.
Although this chapter does not directly take a stand on the debate between the
rights-based approach and the back-to-basics movement, its argument is aligned
with the former position. In particular, I believe it supports the view that aid agencies
have a special obligation—one that goes beyond the norm—to ensure that the basic
rights of those they are trying to help are not violated; moreover, that they have an
obligation to work cooperatively with others to try to realize this goal. However, my
specific departure point is the framework adopted by Garrett Cullity in his chapter
‘Compromised Humanitarianism’, this volume. Cullity’s topic is the problem of
promoting humanitarian ends in morally compromised environments—the sort in
which aid agencies routinely operate. He argues, I think rightly, that in thinking
about the morality of aid agencies working in such environments, it is essential to
distinguish two questions:
1. What (morally) ought such agencies to do as individual agencies?

2. What (morally) ought such agencies to do as members of a larger collective (e.g.
the collective comprising all aid agencies)?
The importance of distinguishing these questions derives from the fact that aid
agencies face a collective action problem for altruism. They often find themselves
in situations where they are faced with the prospect of doing less good, acting in the
capacity of an individual agency, than they could do if they were to act in concert
with others. This naturally raises the question to what extent each agency should be
trying to act as a part of a larger collective. I shall refer to this approach to thinking
about the morality of aid agencies a ‘cooperative approach’.
How should the two questions that constitute a cooperative approach be
answered? In this chapter I focus on the apparently special case in which aid agen-
cies are conceived as members of a larger group that is collectively harming the
very individuals the agencies are trying to help or protect. It is not my purpose
to try to define the concept of harm precisely, but two remarks are in order. First,
harming should be distinguished from ‘failing to help’. (It should also probably be
Aid Agencies, States, and Collective Harm 177
distinguished from ‘making worse off,’ although the harms I am concerned with in
this chapter almost invariably make the persons harmed worse off).5 Second, harm-
ing usually implies injustice, which is arguably the fundamental reason why one
should not harm others. If I harm you I have usually done something more than
make you worse off: I have wronged you in some way. Aid agencies, like the rest of
us, should be especially concerned to avoid harming others where doing so involves
committing a clear injustice.
Now, from one natural perspective—that of a single aid-agency engaged in moral
reasoning about how it should act—cases in which such an agency needs to be
worried about harming others may seem to be comparatively rare. Cullity cites the
example of Fiona Terry’s experience as a member of Medicins Sans Frontiers, which
withdrew from refugee camps in Rwanda during 1994 out of the concern that the
group’s humanitarian activities were being exploited in ways that were harming the
very refugees they were trying to protect (Terry 2002).6
However, from another perspective that I shall adopt here—one that takes a wider
view—the sort of case I have in mind is not rare at all. If the relevant ‘larger group’
(i.e. collective) is large enough, and if one employs a wider notion of harm such
as that used by Thomas Pogge, an uncomfortable case can be made that most aid
agencies are involved in collectively harming the very persons they are trying to
help, just as, arguably, most adult Westerners are involved in harming such persons.
This wider perspective can be captured by extending Cullity’s reasoning about the
representative case raised by Fiona Terry, mentioned above. Terry (2002) asks, of
MSF’s activities in Rwanda,
To what extent were we responsible for the manipulation of aid in the camps?
Cullity rightly points out that before this question can be answered one must
know: who is ‘we’? In particular, does ‘we’ refer to MSF, or does it, for instance,
refer to the collective formed by MSF in conjunction with other aid agencies? I
propose to take this question a step further, and ask how it affects what aid agen-
cies morally should do if the referent of ‘we’ is enlarged even more, perhaps so
that it includes Western states, or even Western-dominated global institutions more
generally—both of which collectives most aid agencies are in some sense members.
To be sure, there is a cost in widening the scope of moral inquiry in this way.
Aid agencies after all, are hardly the only, or even the most important, members of,
for instance, their respective home states. And there is a risk that in taking such a
wide view we shall lose sight of what is special or distinctive about their obligations.
A further concern is that by using the term ‘cooperation’ within the context of such
large groups we risk stretching the concept beyond all recognition. One can sensibly
talk of cooperation between different aid agencies working in Rwanda, because they
share fundamental goals even if they disagree about how to realize them. But in what
sense can an aid agency such as Oxfam America ‘cooperate’ with, say, the Bush
administration on the conduct of US foreign policy –an issue about which (one
presumes) the two are in profound disagreement? Isn’t it the case that any actions
Oxfam might take along these lines are best described as instances of advocacy and
conflict, not cooperation?
178 R. Das
I shall take these two concerns in reverse order. With respect to the issue of coop-
eration, I think the best response is to begin by partly conceding the point. Doubtless
it is true that changing the actions of any large and diverse group will inevitably
involve advocacy and conflict. However, this is not unique to large groups: it is
true (admittedly to a lesser extent) of small groups as well. As long as groups are
composed of individuals, disagreements will arise and to that extent changing the
group’s behaviour will involve some level of conflict. On the other hand, even grant-
ing the inevitably of conflict in large groups there is nonetheless plenty of scope
for cooperation, perhaps even more so than in small groups. The key, of course,
is to find others with whom one can cooperate. For an American aid agency that
opposes the current war in Iraq, there is no shortage of potential cooperative part-
ners outside the NGO sector, even if the Bush administration cannot be counted
amongst them.
Turning to the first concern about the distinctiveness of aid agencies’ obligations,
it is probably true that the difference between the obligations of aid agencies and
those of other members of society is more a matter of degree than one of kind.
However, this difference in degree is significant, deriving primarily from the fact
that aid agencies have set themselves up for the very purpose of making better the
lives of others. It may seem paradoxical that acting on the intention to help others
can generate moral demands that exceed the norm. Without arguing the point here,
I believe it is nonetheless true.7
In any case, the idea that aid agencies should be especially concerned not to harm
others is hardly original, or unintuitive. As noted, the importance of cooperation in
achieving this goal appears to have become widely recognized over the last decade
or so, in the wake of the Rwandan genocide in 1994. One noteworthy effort along
these lines is the ‘Do No Harm’ project, associated with Mary Anderson and others,
which is a collaborative effort aimed at understanding the impact of humanitarian
assistance in conflict settings (Anderson et al. 2004). We shall examine some of its
basic tenets in the next section, along with those of other collaborative efforts such as
the Sphere project. At this point, however, I wish to guard against a bias—common,
I believe, to many of these efforts—toward a concern with harms perpetrated by
(non-Western) others. As I read the relevant literature, the harms that aid agen-
cies are mainly concerned not to be a part of are perpetrated by armed groups in
the non-Western countries in which they operate—Rwanda being a prime example.
Understandably this is a major area of concern. However, the sorts of harms I have in
mind in this chapter are perpetrated by Western groups, and directed almost exclu-
sively against the citizens of non-Western countries. From a moral point of view
there are obvious reasons for Westerners to be especially concerned with harms
perpetrated by our own groups and governments, and I believe this point applies a
fortiori to Western aid agencies.
In short, the central question I wish to address is this:
Assuming that aid agencies have an obligation to cooperate with each other to avoid col-
lectively harming those they are trying to help, how worried should they be that they may
belong to very large groups (i.e. those not limited to other aid agencies) that are responsible
for such harms? In particular, are the harmful actions of an aid agency’s government beyond
its reasonable moral concern?
The question is hardly academic. A good case can be made that the largest
humanitarian NGOs, based primarily in the US and Britain, are members of, and
are substantially funded by, governments that are directly and indirectly responsi-
ble for a tremendous amount of harm and misery in the world. If this is right, these
NGOs operate in many countries where they try to alleviate suffering caused by very
large groups of which they are (fairly prominent) members. At the very least, this
(apparent) fact sharpens the question raised by Cullity, regarding how aid agencies
can best promote humanitarian ends in morally compromised environments.
I shall address the above question in part by examining some key codes of
conduct that many aid-agencies already subscribe to, particularly as these have
implications for cases where harms are being perpetrated, both by non-Western oth-
ers and by Western governments and groups. I shall argue that if aid agencies are
complicit in harming others, they have much greater reason to act cooperatively—
specifically, to try to end such harms—than they would have if they were not so
complicit. Basically, harming makes a difference to the question of how aid agen-
cies should act in morally compromised environments, and this leads to a slightly
different conclusion than the one reached by Cullity in his chapter. I don’t think
the conclusion is strong enough to suggest that aid agencies should turn away from
humanitarianism immediately. However, it does seem to support the view that in the
medium term most Western aid agencies are morally required to work cooperatively
to end basic human rights violations (i.e. serious harms) against those they are trying
to help. Arguably—given the underlying causes of most human rights violations—
this entails that they are morally required to take on a much more politically activist
role than they have to this point.
As noted, this last conclusion is in line with a controversial position that many aid
agencies appear to have been moving toward for some time. It clearly has important
and far-reaching implications for how they are funded and who they are accountable
to. It could also be said that the conclusion is supported by the simple recognition
that providing foreign aid is an inherently political activity (Chapter ‘The Ethics of
Taking Sides’ by Ellis, this volume).8 If this is right, then aid agencies should not
expect to be able to advocate for human rights in a politically neutral way. (For that
matter, neither should human rights organizations such as Amnesty International).
Such advocacy inevitably involves taking sides. This does not mean that the political
issues are cut and dried (though in many cases they are clear enough), nor that
practical realities on the ground should not temper when to advocate and how best
to do it. It does mean that humanitarianism cannot in the end be separated from co-
operative political action, and that from a moral perspective the latter should form a
central component of any aid agency’s activities.
The remainder of the chapter is structured as follows. In the next section I con-
sider and ultimately reject the possibility that the issue of harm does not make any
difference to how we should think about a cooperative approach to the morality of
aid agencies. In Section 2, I take up the question of harms committed by very large
groups of which one is a member. In particular, I argue that for aid agencies their
status as members (and beneficiaries) of powerful states is the most important factor
to consider when assessing their potential role in committing harm. I shall illustrate
this claim by considering the case of the ongoing war in Iraq.
180 R. Das
2 Harm and the Cooperative Approach to Morality

Suppose we think of an aid agency as part of a larger group that is implicated in
harming the very persons the agency is trying to help. Suppose further that in the
short term the agency could promote the most good or welfare simply by going
about its usual business and not bothering to pursue cooperative efforts that could
curtail or even end the harms in the longer term. Then our question in this section
is as follows: what effect does the first supposition, concerning doing harm, have
on an aid agency’s obligations to cooperate with others in ways that can curtail or
end harm in the longer term? In particular, does the issue of harm make a moral
difference to whether an aid agency should cooperate to try to end the harm, even
given our second supposition that not cooperating will promote the most good in the
short term?
As a preliminary it will be useful to consider the reasons for focusing on the
amount of good or welfare an aid agency can promote, rather than the possible harms
it might try to prevent. As Garrett Cullity argues, these reasons often point toward
not cooperating. Cullity notes that in many cases, substantial forms of cooperation
with other aid agencies either do not exist or only partially exist, rendering attempts
at cooperation of dubious value: they are incapable of achieving the benefits that
could be achieved if others were cooperating. In contrast, the urgent humanitarian
ends that can be realized by an individual agency are achievable in the immediate
future. Given such circumstances, it seems to be mistaken to think that aid agencies
should sacrifice the humanitarian ends that they could promote individually, merely
for the sake of pursuing some as-yet-unrealized ideal of cooperation. The human-
itarian ends are simply too important, and the cooperative ideal too unformed, to
make such a strategy sensible.
This line of reasoning is undeniably appealing. There is indeed something futile
in trying to act cooperatively when no one else is cooperating, and doing so can
seem morally offensive when the opportunity cost is so great, as it often seems to
be in the contexts in which aid agencies operate. And in cases where no questions
of harm arise the argument for non-cooperation when others are not cooperating is
much stronger than it would otherwise be—certainly in the short term. This said, it
is worth spelling out a couple of points that support the alternative view, given their
relevance to cases where questions of harm do arise.
First, and most important, there is the worry that continuing to act in a purely
individual capacity tends to ensure that the status quo will not change. Given that
aid agencies could do more good cooperatively than they could do individually, this
potentially represents a significant medium-to-long term cost of failing to cooper-
ate. It thus seems to support, at a minimum, that aid agencies devote an increasing
portion of their time and energy to building cooperative links with others in ways
that will increase their collective ability to promote good and relieve suffering. If
the middle of a humanitarian crisis isn’t the best time to try to begin this sort of
cooperation, then perhaps a start can be made once the current crisis subsides.
Second, there are other important values besides human welfare, notably includ-
ing human dignity. Respecting these values might be sufficiently important that an
agency morally should do so even when other agencies, which should also respect
them, do not. An example of this is the view that aid agencies should not use
emotional/manipulative images in their advertising to increase donations. On the
assumption that such images are morally problematic, agencies ideally would all
agree not to make use of them. Yet many aid workers and observers evidently believe
that a single agency should not use them even if other agencies continue to do so,
and even if it means that the lone agency misses out on donations and hence is less
able to promote human welfare.
Both of these points are reflected in various recent NGO protocols, and, to vary-
ing extents, in current practice. Taking the point about dignity first, Article 10 of the
International Red Cross Code of Conduct (ICRCC, 1994; henceforth ‘Code’) states:
In all our information, publicity and advertising activities, we shall recognize disaster
victims as dignified humans, not hopeless objects.
Likewise, Article 2.1 of the Code of Conduct for the Australian Committee for
International Development (ACFID) states that in all its communications to the pub-
lic it must ‘accord due respect to the dignity . . . of the people with whom it works’
(ACFID 1997). And the Humanitarian Charter of the Sphere Project (1997) affirms,
in its first principle, the ‘right to life with dignity.’ In general, the best known NGO
codes of conduct reflect the awareness that aid agencies have a responsibility to
respect human dignity as well as to promote human welfare.
Now consider how the first point, concerning the importance of cooperation to
changing and improving the status quo, is reflected in aid agency codes of conduct.
Most simply, its importance is reflected in the very existence of such codes, insofar
as agreeing to follow universal standards of conduct represents a basic form of coop-
eration. More substantively, Article 6 of the Code of Conduct for the International
Red Cross refers to ‘build[ing] disaster response on local capacities.’ Similarly,
ACFID (1997) speaks of ‘aim[ing] to build creative and trusting relationships with
people of developing countries’. As these examples suggest, a common theme in
the various codes is the importance of working in partnership with people and
organizations from the countries in which aid agencies are based.
Two other initiatives, The Sphere Project (1997) and ALNAP (Active Learning
Network for Accountability and Performance, 1997), represent efforts to improve
the quality of humanitarian assistance through a variety of cooperative means. These
include developing operationally useful standards of conduct, sharing experiences
from the field, and improving transparency and accountability mechanisms for aid
agencies.9 In short, there is clearly good evidence that aid agencies recognize the
importance of cooperation with one another, as a means, presumably, to promoting
the greater overall good that they are able to achieve.
Still, it would be misleading to say that the major humanitarian NGO codes of
conduct make cooperation with others a high priority. Indeed, a good deal of what
they say points in a rather different direction. Most obviously, the first article of
the Red Cross Code reads: ‘The humanitarian imperative comes first.’ Although
this directive does not mention cooperation, the most natural way to interpret
it in the context of altruistic collective action problems is in line with Cullity’s
182 R. Das
anti-cooperative analysis sketched above. The Sphere Project reaffirms the primary
commitment to humanitarianism, and the other major codes tend to align with the
Red Cross Code on this point. Second, Article 3 of the Code is generally interpreted
as advocating a kind of political neutrality. Although it affirms the right of NGOs
to: ‘espouse particular political or religious opinions,’ it states that: ‘Aid will not
be used to further a particular political or religious standpoint.’ In a lucid commen-
tary on the Code, Dorothea Hilhorst notes that this directive seems to imply that aid
agencies should remain neutral and not take sides in a conflict: ‘Neutrality is impli-
cated merely because is hard to see how taking sides in hostilities can be separated
from using aid to further political standpoints.’ She goes on to comment that the
word ‘neutrality’ was left out of Article 3 because so many humanitarian NGOs
are indeed committed to principles of justice and human rights (Hilhorst 2005,
p. 356).
I now wish to examine the above two considerations—concerning dignity and
increasing welfare over the long term—in light of the problem of harm. In cases
where harms are being perpetrated I think these considerations weigh even more
heavily in favour of cooperation, even if this means promoting less welfare in the
short term. For doing harm to others not only represents an assault upon their dig-
nity; it also typically reduces their level of well-being. So let me spell out these
considerations in a bit more detail in the context of harming.
For many laypersons as well as philosophers, harm represents a qualitatively dif-
ferent moral category from failing to help, or even from ‘making worse’. There is
something wrong about harming others that goes beyond the (impersonally mea-
sured) badness that such harm also almost inevitably involves. This is the main
reason that the category of harm resists easy assimilation into a wholly impersonal
moral calculus such as utilitarianism employs. It is also connected to one popular
view of rights—which are meant, perhaps in the first instance, to protect against
harm—which sees them as ‘trumps’. Respecting others’ rights is a moral impera-
tive that trumps or pre-empts other moral considerations, in the sense that one must
always take care to respect others’ rights whatever else one does. More cautiously,
the moral importance of respecting rights outstrips the value, in welfare terms (of
goodness/badness), that respecting rights tends to promote. Arguably, one cannot
‘make up’ a harm that one has done to another simply by promoting other aspects
of his or her good.
These points about harm may be expressed too strongly. They are not uncon-
troversial amongst philosophers, but nor are they unfamiliar. Those working within
a broadly Kantian framework tend to be sympathetic toward them; others, less so.
Just as important, they appear to have a secure place within commonsense morality.
Most non-philosophers do seem to think that harm represents a special moral cat-
egory in something like the sense articulated above. Another point: what has been
said above holds most strongly for the most serious sorts of harms—those against
which human rights, in particular, are intended to protect. I have in mind threats
to bodily integrity and safety; the denial of freedom of movement and settlement;
avoidable severe deprivation of food, water, shelter, minimal health care, education,
and so forth. It is these most egregious sorts of harm, against which human rights
are meant to protect, that any of us should be most concerned not to commit—either
individually or as members of a group.10
The preceding remarks about harm are fairly abstract and theoretical. It is instruc-
tive to compare and contrast them with how aid agency practitioners and observers
perceive the issue. A natural place to look is the ‘Do No Harm Handbook,’ which
is meant to be useful for ‘understanding the impacts of assistance programs on the
socio/political schisms that cause, or have the potential to cause, destruction or vio-
lence between groups.’ Reading through the handbook it is easy to see how it would
indeed be useful to practitioners. It asks them to consider, amongst other things, how
the provision of aid affects the local economy, whether it provides incentives to war-
makers or peace-makers, and whether it serves to divide or unite peoples in conflict.
However, the philosophical points about harm expressed above –particularly the
idea that harm is a qualitatively different moral category such that agents should
always take special care to avoid doing it—are hard to find in the handbook. Partly,
no doubt, this reflects a difference between academic philosophy and a hands-on
manual for practitioners that must take into account messy realities on the ground.
Mostly, however, I think it reflects the fact that the handbook is primarily concerned
with how to deal with harms committed by others. As the quoted sentence above
indicates, the handbook sees aid agencies essentially as third parties whose work
happens to bring them into conflict situations where people are being harmed, not
as members of large groups that may be perpetrating harms directly themselves. This
focus on harms committed by others may also explain the handbook’s disclaimer—
surprising, given its title—that it is not meant to be ‘prescriptive’: it does not mean
to tell aid agencies what to do or not do.
Similar points, I believe, apply to Hugo Slim’s analysis of ethical dilemma’s
facing NGOs (Slim 1997). Slim’s focus again is on aid agencies trying to provide
aid in situations of conflict. One of his case studies, which he titles ‘The Ethics of
Humanitarian Disengagement,’ involves the provision of aid to Burundi in the mid-
1990s. Slim argues against what he calls the ‘catharsis position’ of withdrawing
the provision of aid from conflict situations in which that aid is being exploited,
but again he basically assumes that aid agencies are third parties to a conflict—not
members of a group that may be one of the chief protagonists and thus directly
involved in doing harm. This assumption may be what leads him, understandably
but mistakenly, to characterize an aid agency’s choice to disengage from Burundi as
reflecting a consequentialist ethical position, and the choice to remain in the country
to render assistance a duty-based or deontological one. In fact, if aid agencies are
members of a large group that is directly harming others, the most straightforward
duty-based imperative is to (work to) stop the harming. Crucially, this is not the same
as alleviating or ameliorating such harms, which in this case would be precisely to
provide humanitarian assistance to the victims. To use a simply analogy, if I go about
breaking other persons’ limbs, my primary duty is to stop (and to make reparations).
I do not discharge this duty by arranging for my victims to be taken by ambulance
to hospital at my expense.
So it would appear that aid agencies are thinking about the issue of harm within
a rather different framework from the one contained in the analysis offered above,
184 R. Das
a difference stemming principally from the fact that they see themselves mainly
as third parties involved only indirectly in harmful conflicts, rather than members
of large groups that are directly involved in doing harm. In the following section I
explore some of the consequences of this difference. Amongst other things, I suggest
that being directly involved in doing harm may tend to resolve the moral dilemma
that aid agencies evidently sometimes feel, underwriting a stronger requirement in
favour of cooperative political action.
However, for the purposes of this section the difference alluded to may not terri-
bly important. My claim in this section, recall, was that if aid agencies are implicated
in doing harm, this strengthens their obligation to cooperate with others to end the
harms, even when not cooperating will produce the most good in the short term. And
it seems fair to say—based, e.g., on the Do No Harm Handbook—that aid agencies
would not disagree with this claim, even if they are approaching the issue of harm
from a different standpoint.
3 Group Size and Sub-group Responsibility

Let us now assume that being part of a group that is committing harm underwrites
an obligation for aid agencies to work cooperatively to end such harms, even if they
can do more good in the short term by not cooperating. Our focus in this section is
on the size of the group implicated in harming. In particular, should aid agencies be
morally concerned that they may be involved more directly in perpetrating harms,
albeit as members of very large groups such as states? It is not my primary purpose
in this chapter to argue that aid agencies are, indeed, members of large groups that
are directly engaged in doing serious harm to others. For the most part, I shall simply
assume that they are. What I shall argue is that if this is so, then aid agencies have
a very strong obligation to engage cooperatively with others, ideally in the form
of political advocacy, to end such harms. I realize that most aid agencies already
pursue a mixture of humanitarian and political/advocacy activities (Hilhorst 2005,
p. 358). The practical upshot of what I shall argue is that their activities should be
tilted much more heavily toward political action, a claim I shall illustrate using the
example of Iraq.
I turn first to the question of the size of the harmful groups of which aid agencies
may be members. These groups can be very large indeed. At the limit, one made
familiar by Thomas Pogge, such groups can comprise virtually all adult Westerners.
At least this is so if one can sensibly speak of the harm done by the imposition of
an institutional order for which just about any adult may, in principle, be partially
responsible. Pogge puts his account of harming as follows: ‘On my view, you harm
others insofar as you make an uncompensated contribution to imposing on them an
institutional order that foreseeably produces avoidable human rights deficits’ (Pogge
2005, p. 61). Pogge’s ‘institutional’ approach to human rights and harm is well
known amongst political philosophers. The main virtue of such an approach is that it
is fairly easy to show that institutional factors do, indeed, produce avoidable human
rights deficits amongst the world’s poorest people. Pogge himself has shown this
convincingly, and I will mention only two aspects of the compelling case he has
built in his published work.
First, what he calls the ‘international resource privilege’ refers to the interna-
tional legal convention according to which any government, no matter how dubious
its legitimacy, has the right to sell its country’s natural resources to foreigners (the
main buyers are multinational corporations). Pogge emphasizes the perverse incen-
tives that this convention provides. Within poor countries, it encourages the rise of
authoritarian and exploitative regimes. Just as important, it provides affluent coun-
tries no incentive to change the status quo—and much incentive to keep it—as long
as their own economies continue to benefit from poor countries’ resources. Second,
the ‘international borrowing privilege’ refers to a similar convention whereby the
same authoritarian governments are granted the privilege to borrow freely in their
country’s name (from, e.g., the IMF and World Bank). Not only does this ill-
deserved line of credit encourage coup attempts and civil wars within poor countries,
it invariably saddles the population with massive debts even if they do somehow
succeed in overthrowing the authoritarian regime (Pogge 2002, pp. 112–116).
Again, I think Pogge’s case for holding that we harm the poor through the imposi-
tion of a global institutional order is very strong, and I think any discussion of global
poverty must highlight the role this institutional order plays. Nevertheless, I think
there is a danger in putting too much emphasis on the global institutional order, par-
ticularly when considering questions of moral responsibility. First, although agents
create institutions, institutions themselves are of course not agents. So any account
that focuses on an institutional order as the immediate cause of harms done to the
global poor also owes an account of who is responsible for creating/imposing that
institutional order. Second, however, it is not always clear that there is a reasonably
small group of agents of whom it can be said that that group, by itself, imposes
a human-rights depriving institutional order on the poor. In any case, focusing
on the importance of institutions tends to take the spotlight away from the actual
agents who bear responsibility for creating and imposing such institutions. This
may explain Pogge’s reference to those of us who ‘contribute’ to the imposition of
such an order. The upshot, I think, is that we mustn’t lose sight of the actual agents
who bear primary responsibility for imposing the relevant institutional order.
Although from the broadest perspective, again, these agents include virtually all
adult Westerners, I think that from the point of view of NGOs there are good reasons
for focusing on powerful states as the most salient agents responsible for impos-
ing the global order. First, such states clearly play a significant direct role in the
imposition of that order. Second, virtually all such states are democracies,11 making
it a relatively straightforward matter (at least in principle) of distributing respon-
sibility to their individual members: democratic citizens (and groups thereof) are
ultimately responsible for the actions of their elected officials. Finally, the most pow-
erful democratic states also happen to be the largest donors of foreign aid, including
the largest donors to humanitarian NGOs. In combination, these three factors under-
write a very strong moral requirement for aid agencies to engage in cooperative
political action to end harms done by powerful states.
186 R. Das
I shall illustrate this claim by looking at a clear example of powerful states doing
harm, namely, the ongoing case of the war in Iraq. My focus, in particular, is on US,
British, and Australian forces in perpetrating that harm. Much has written about this
case, of course. What I shall do here is state what I take to be the most salient moral
facts about the case, and then critically compare the presentation of the same facts
in a recent report of a major humanitarian NGO. I shall focus on the time period
since the US-led invasion in early-mid 2003.
For Western aid agencies, the first morally salient fact about Iraq is that for the
past several years it has ranked amongst the worst humanitarian crises anywhere
in the world. For instance, it is significantly worse in humanitarian terms than the
well-publicized crisis in the Sudan. In Iraq, over 1 million persons have died of vio-
lent causes since 2003, according to the best estimates available (Burnham et al.
2006, Chomsky 2008).12 Some 2 million persons have fled the country, and another
2 million or so are internally displaced. By any humanitarian standard the situa-
tion in the country is dire. The second morally salient fact for Western aid agencies
is that the US-led coalition is primarily responsible for what has happened there,
and what continues to happen there. Accurate reports are hard to come by simply
because the country is so dangerous, but the most reliable information underscores
perhaps the most important fact about the occupation, virtually never reported in
the mainstream: the US-led occupation is itself the primary cause of death and dis-
placement in Iraq, and the withdrawal of its forces the single most important step
toward ending, or at least mitigating, the suffering of the Iraqi people (D3 Systems
KA Research Ltd. 2008).13 If the Bush administration has its way, withdrawal of US
troops will not occur before the passage of the so-called ‘Iraqi Oil Law,’ which it
has made one of its top priorities. At the time of writing, none of the major US pres-
idential candidates are advocating an immediate end to the war, and certainly none
have mentioned making reparations to the Iraqi people for the devastation caused to
their country.
Against this background, I now shall consider a recent Briefing Paper on Iraq
published by Oxfam International (Oxfam International 2007). I focus on Oxfam,
which is highly regarded particularly in the academic community, for a few reasons.
First, of the major humanitarian NGOs it is perhaps the most heavily oriented toward
political advocacy. Second, again amongst major NGOs it receives a relatively small
percentage of its funding from government sources.14 Finally, it is clear in reading
Oxfam literature that the organization is well aware of the political context in which
global poverty occurs. As an Oxfam supporter, my guess is that the organization sees
itself as highlighting that context in as much critical detail as it considers politically
feasible.
It is, then, instructive to review Oxfam’s presentation of the above information
in the recent Iraq briefing paper. I believe it falls short of the level of advocacy that
is morally required—particularly in what it leaves out—and points up the difficulty
of promoting humanitarianism without taking a stand on key political questions.
Consider, first, the relationship between armed violence and the humanitarian crisis
in Iraq. On the cover of the briefing report, the first sentence begins by acknowledg-
ing that ‘Armed violence is the greatest threat facing Iraqis . . .’ It also recognizes
that such violence is the key factor in people fleeing their homes and becoming
refugees. However, the report contains very little by way of trying to determine who
is responsible for the violence that it criticizes; specifically, there is no real attempt
to hold Western governments or groups responsible for the violence. For instance,
the report does not mention the fact that the US-led coalition launched an aggressive
and illegal invasion of Iraq in 2003.
To be fair, Oxfam notes that it opposed the war ‘on humanitarian grounds.’
Yet, although these were undoubtedly good grounds to oppose the war they were
hardly the most fundamental ones. Rather, Oxfam could have opposed the war on
the grounds that it violated Article 3 of the UN Charter, which forbids the use of
force in international affairs except in cases of self-defence or when specifically
sanctioned the Security Council. Not only is this principle arguably the cornerstone
of international law, but (from the point of view of an organization such as Oxfam)
there are excellent humanitarian reasons for upholding it and for calling to account
those who violate it. As a moral or legal principle it is certainly no less important
than the principle that civilians must not be targeted in war.
Consider, next, the fact that the US-led coalition (the ‘Multi-National Force in
Iraq’ or MNF-I) is effectively the occupying power in Iraq. The Oxfam report
asserts that ‘The primary duty-bearer for the provision of basic services remains the
national government.’ (Oxfam International 2007, p. 4). Yet the national govern-
ment in Iraq is a puppet government, entirely dependent on the force of the MNF-I
to avoid being overthrown.15 But then the primary duty-bearer for providing basic
services, and more generally the party responsible for taking care of the population,
is the occupying power. Granted, the briefing paper does note that ‘The Iraqi gov-
ernment cannot [meet basic humanitarian needs] without support from international
governments with capacity and influence in Iraq . . . including the USA and the UK.’
(ibid. pp. 18–19). As criticism, however, this is exceedingly tame. The problem is
not simply that the ‘international governments’ mentioned are failing to meet basic
humanitarian needs in Iraq. Rather, the problem is that those very governments are
themselves attacking, killing, imprisoning, and displacing innocent Iraqis on a huge
scale, just as they have been doing for the past 5 years.
Another quote from the report basically acknowledges the role of the invasion
and occupation in the humanitarian catastrophe—again, however, without assigning
responsibility to the coalition forces in any direct way: ‘Basic indicators of humani-
tarian need in Iraq show that the slide into poverty and deprivation since the coalition
forces entered [sic] the country in 2003 has been dramatic, and a deep trauma for
the Iraqi people. The number of refugees and displaced persons is now massive by
any modern standards’. (ibid. p. 8)
A related omission in the report is the absence of any serious attempt to deter-
mine who or what is the cause of the huge death toll in Iraq. Note that this is not
merely a question of retroactively assigning blame. Rather, the question is a crucial
first step toward ending or at least curtailing what is, after all, an ongoing problem.
In particular, although the report does mention the criminal gangs, death squads,
and suicide bombings that dominate news coverage, there is no recognition that the
deaths caused by these sources do not begin to add up to the Lancet study’s estimated
188 R. Das
death toll of some 650,000 by Nov 2006 (extrapolated to over 1 million by early
2008). The best explanation I know of for the discrepancy is that huge numbers of
Iraqis are being killed by the operations of the coalition forces themselves. In partic-
ular, as American academic Michael Schwartz (2007) has showed, violent episodes
that follow from standard operating procedures of (predominantly) US forces in
door-to-door searches—of which there are literally thousands each day—easily add
up to the hundreds of thousands of deaths that are otherwise unaccounted for.
Granted, this analysis is not uncontroversial; certainly it is unfamiliar. But con-
sider the moral implications for groups like Oxfam America or UK if anything like
it is true. The armed forces of one’s own (democratic) country launch an aggressive,
illegal invasion of another country and then proceed to occupy it for years, building
14 apparently permanent military bases in the process. Those armed forces routinely
engage in military operations, through air and over land, that kill hundreds of thou-
sands of innocents and drive four million more from their homes in the space of a
few years. They create a humanitarian catastrophe that leads to 43 percent of the
country’s population living in ‘absolute poverty,’ with 70 percent without access to
clean water or sanitation. (Oxfam International 2007) Last but not least (in the case
of Oxfam UK) the government responsible for perpetrating these harms is a major
financial contributor to one’s own humanitarian organization.
This sort of scenario, I submit, creates tremendous moral pressure for an organi-
zation such as Oxfam to make political advocacy a very high priority, certainly in
the medium term. First, if anything like the analysis just provided is correct, from
the perspective of Oxfam the case of Iraq presents a clear case of being a member of
a group that is perpetrating egregious and widespread harms. Second, from a more
consequentialist viewpoint, if the actions of the MNF-I forces are themselves the
primary cause of the humanitarian crisis affecting ordinary Iraqis, then engaging in
political advocacy directed toward ending those actions may be the most important
thing a (highly respected) humanitarian organization such as Oxfam can do. Again,
and as the above quotes make clear, I am well aware that Oxfam already does this
to a relatively high degree. What I am suggesting is that from a moral point of view
the balance of reasons suggests that its advocacy level should be much higher and
the critical content of that advocacy much sharper.
Of course, moral considerations are not the only considerations, and inevitably
they must be balanced against practical ones. These latter include the possible loss
of financial support from states that come in for sharp criticism, and, perhaps even
more important, the possible loss of non-profit status due to the perception that one’s
organization is too overtly political. Yet, for an organization whose very reason for
existing is broadly moral, even such practical reasons must ultimately have humani-
tarian import. Practical considerations having to do with funding, in particular, must
ultimately be justified on the grounds that greater funding allows the organization
to do more good than it otherwise would.
Now, undoubtedly Oxfam would defend itself against the present critique by
making just this sort of claim. It would object that what I am calling practical rea-
sons are in fact moral reasons about promoting the greater good. What is really to be
gained, it might ask, by having Oxfam take on a markedly more politically activist
role? Does anyone really think that the political activism of one NGO could alter
the long odds against bringing about fundamental political change and its undoubted
attendant benefits? In the meantime Oxfam would have lost considerable revenue,
and with it much of its ability to realize achievable and worthwhile humanitarian
goals in the here and now. Why on earth should it do that?
By this point I hope it is clear that the last paragraph effectively restates the
fact that Oxfam, like other aid agencies, faces a collective problem for altruism.
In this chapter I have been concerned to address this sort of problem in a special
case, namely, that in which such agencies are conceived as members of a larger
group that is collectively harming the very individuals they are trying to help. I have
not challenged the idea that aid agencies are morally permitted to continue, non-
cooperatively, their humanitarian aims in the immediate term—at least when others
are failing to cooperate in ways that would produce greater good overall. However,
I have argued that in the medium to longer term, aid agencies are morally required
to act cooperatively, in particular to engage in political advocacy toward ending
the harms perpetrated by those larger groups to which they belong, even if other
agencies are not. If this is correct, the medium to long term responsibilities of aid
agencies may diverge sharply from their immediate responsibilities. Indeed, to the
extent that they take these longer term responsibilities seriously, it may well mean
transforming themselves into quite different sorts of organizations.
Notes
1. My focus here is on international non-governmental organizations. Further, I am mainly inter-
ested in those international NGOs that focus on humanitarian and development activities, as
opposed to human rights.
2. To take a very simple case, consider someone who has fallen through thin ice into freezing
water. Many people are just off shore; if they all try to help at once the ice will give way and
everyone will fall in. In this case, acting together or collectively means agreeing that only one
person in the group should venture out onto the ice, while others go for help, etc.
3. As Kieran Donaghue notes in his chapter ‘Human Rights, Development INGOs and Priorities
for Action’, this volume, international NGOs are typically divided into three broad categories:
humanitarian/relief; development; and human rights. I am supposing, perhaps tendentiously,
that of these three the sort of work done by human rights NGOs lends itself most naturally to
politically activism.
4. In his chapter Donaghue sympathetically explores this sort of view through the work of Henry
Shue on basic rights.
5. As I read the relevant NGO literature, this is the sense in which NGOs seem to be using the
term.
6. Terry’s analysis can be challenged. For instance, Hugo Slim has suggested that the type of
case Terry describes may not count as a moral dilemma if the NGO involved has reflectively
formulated its policy in such a way that recognizes (and accepts) that there may be adverse
effects to its furthering humanitarian ends. (Slim 1997). However, for present purposes I shall
assume that the sort of case Terry describes presents a genuine moral dilemma for any aid
organization that sees its mission as one of minimizing suffering.
7. It is fairly standard in moral philosophy to distinguish ‘special’ obligations from generic
ones. The former obligations, which are typically more stringent, derive from one’s status as
a friend or family member, or—most relevantly in this case—one’s professional role.
190 R. Das
8. Ellis, however, reaches a different practical conclusion from the one suggested in this chapter.
9. Also see the discussion in chapter ‘The seeming simplicity of measurement’ by Roche, this
volume.
10. Cf. Donaghue’s (chapter ‘Human Rights, Development INGOs and Priorities for Action’, this
volume) useful discussion of Henry Shue’s typology of rights.
11. The most powerful non-democratic states—China and perhaps Russia—have not played a
significant role in imposing the current global economic order, although this is starting to
change.
12. Chomsky (2008) cites the Oxford Polling Bureau, which has updated its death estimate to 1.3
million.
13. For a clear-headed account of the number of Iraqis killed by the invading and occupying
forces, see Schwartz (2007).
14. As far as I am aware, Oxfam America is the only major INGO that does not receive any fund-
ing from government sources. According to Abby Stoddard (2006), Oxfam UK receives about
25 percent of its funding the UK; MSF is the only other major humanitarian that receives a
comparable amount of government funding from the government.
15. As I write (early April 2008) this has been demonstrated once again with the government
assault on the Mehdi Army in Basra and elsewhere. According to observers in the country,
without strong US backing (it appears, in fact, that US troops did much of the fighting) the
Iraqi government forces would have been completely overrun by the Mehdi Army militia. See
Rosen (2008).
Acknowledgement I would like to thank Chris Roche for very helpful suggestions on an earlier
draft of this paper.
References
ACFID . 1997. Code of Conduct. Canberra: Australian Committee for International Development.
http://www.acfid.asn.au/code-of-conduct.
ALNAP. 1997. www.alnap.org. Accessed 1 July 2007.
Anderson, M., H. Wolfgang, S. Jackson, and M. Wallace. 2004. The “Do No Harm” Framework for
Analyzing the Impact of Assistance on Conflict: A Handbook. Cambridge, MA: Collaborative
Learning Projects. http://www.cdainc.com.
Burnham, G., R. Lafta, S. Doocy, and L. Roberts. 2006. Mortality after the 2003 invasion of Iraq:
A cross-sectional cluster sample survey. Lancet Online, Oct 11, http://www.thelancet.com.
Chomsky, N. 2008. Good news: Iraq and beyond. Z Magazine Apr 2, http://www.zcommunications.
org/zmaq/viewArticle/17048.
D3 Systems and KA Research, Ltd. 2008. Iraq Opinion Poll. http://www.globalpolicy.
org/iraq/invasion-and-war/iraqi-public-opinion-and-polls.html. Accessed Mar 2008.
Easterly, W. 2005. The White Man’s Burden: Why the West’s Efforts to Aid the Rest Have Done So
Much Ill and So Little Good. New York: Penguin Books.
Hilhorst, D. 2005. Dead letter or living document? Ten years of the code of conduct for disaster
ICRC. 1994. Code of Conduct for the International Red Cross and Red Crescent Movement and
Non-Governmental Organizations (NGOs) in Disaster Relief. Geneva: International Committee
of the Red Cross.
Oxfam International. 2007. Rising to the Humanitarian Challenge in Iraq, (Briefing Paper No.
105), July 30, published in conjunction with the NGO Coordination Committee in Iraq.
Petras, J. and H. Veltmeyer. 2001. Globalization Unmasked: Imperialism in the 21st Century.
London: Zed Books.
Pogge, T. 2002. World Poverty and Human Rights. New York: Polity Press.
Pogge, T. 2005. Severe poverty as a violation of negative duties: Reply to the critics. Ethics and
International Affairs 19(1): 55–83.
Rieff, D. 2002. A Bed for the Night: Humanitarianism in Crisis. New York: Simon & Schuster.
Rosen, N. 2008. Iraq Becomes Somalia: A journalist returns from a broken Iraq. Interview on
Democracy Now! Apr 3, http://www.zcommunications.org/znet/viewArticle/17058.
Schwartz, M. 2007. Is the US killing 10,000 Iraqis Every Month? Or is it More? July 5,
http://www.zmag.org/znet/viewArticle/15028.
Slim, H. 1997. Doing the right thing. Disasters 21(3): 244–257.
Sogge, D. 2002. Give and Take: What’s the Matter with Foreign Aid? London: Zed Books.
Stoddard, A. 2006. Humanitarian Alert: NGO Information and Its Impact on US Foreign Policy.
San Francisco, CA: Kumarian Press.
Terry, F. 2002. Condemned to Repeat? The Paradox of Humanitarian Action. Ithaca, NY: Cornell
University Press.
The Sphere Project. 1997. Humanitarian Charter and Minimum Standards in Response.
http://www.sphereproject.org/.
To Respect or Not to Respect . . . Ethical
Dilemmas of INGO Development Practitioners
Conny Lenneberg
“In all its activities . . . the organization will accord due respect
to the dignity, values, history, religion and culture of the people
with whom it works consistent with principles of basic human
rights.” Clause 2.1. under Organisational Integrity. ACFID1
Code of Conduct
Most development practitioners would agree respect for local people and their
culture is a paramount value that underpins our work. In one form or another
it is embedded in INGO codes of conduct and effectiveness standards, as well
as emerging from program evaluations as a critical element in the success of
any development interventions. It is a truism to state that development programs
must genuinely engage communities in determining their own priorities and needs
in order to establish the minimum conditions for relevance and sustainability.2
More recently debate has centred on how to strengthen NGO accountability to
communities not only to make participation meaningful, but for the achieve-
ment of agreed outcomes. Yet development practitioners (as opposed to charity
workers) also set out with a radical social change agenda—to change the power
dynamics within communities that lead to systematic marginalization and per-
petuate conditions of enduring poverty and exploitation for its most vulnerable.
Reconciling these two imperatives is clearly problematic. The question I want
to explore in this paper, from a practitioner perspective, is ‘what is the ethi-
cal basis for INGOs challenging and seeking to change local cultures, belief and
practices’?
C. Lenneberg (B)
World Vision Australia, Melbourne, VIC, Australia
e-mail: conny.lenneberg@worldvision.com.au
Revised version of paper presented to the Workshop: Ethical Questions for Non-Governmental
Development Organisations. (Centre for Applied Philosophy and Public Ethics, Charles Sturt
University and Oxfam Australia).

194 C. Lenneberg
1 The Dilemma
At its simplest the question is by what standards do we judge others’ cultural beliefs
and practices? For cultural relativists, the answer is simple: there is no universal
ethical standard and no objective moral truth. Therefore, we have no right to judge,
challenge or seek to change traditional practices and beliefs—the community must
determine its own change agenda.3 At the other end of the spectrum, for funda-
mentalist faith-based organisations this question is equally simple: universal ethical
values and objective moral truth are to be found in the literal precepts of their
faith.4 Therefore, traditional practices and beliefs which are contrary with these
standards, not only can, but should be challenged and changed regardless of the view
of local communities. These positions represent theoretical extremes, with twenty-
first century INGOs placed along a continuum between them. They seek to balance
community primacy in determining their own priorities and development objec-
tives, within their agency’s theory of development and the practitioners’ broader
understanding of context, opportunities and risks of change. Achieving this balance
cannot, and should not, be left to individual decisions within different contexts, and
the continuing development of standards of practice is one way in which the sector
as a whole is attempting to provide better guidance and practice frameworks.5 But
the question is more than simply one of program frameworks—it requires a clearly
articulated ethical basis for decision-making which acknowledges more transpar-
ently the intrinsic values which development practitioners inevitably bring to their
assessments.
It is important to first consider the question of what it means to ‘respect’. Is it
possible to respect a person but not necessarily their beliefs? A great deal is written
by INGOs about the need to ‘respect’ culture and cultural differences, but it is not
always clear what ‘respect’ means in this context. Are some beliefs so constituent
of communal identities that trying to change them demonstrates a fundamental lack
of respect for their ‘dignity, values, history, religion and culture’? This is at the heart
of this paper—how can social development practitioners with an agenda of social
change respect a culture and yet still wish to challenge and change some attitudes
and practices, which may be held to be fundamental to a particular culture?
‘Respect’ can include a broad range of concepts including acknowledgement,
deference, esteem, valuing, judgement and behaviour (Stanford Encyclopaedia of
Philosophy 2003). All of these elements seem to be part of what INGOs mean when
they wish to ‘respect’ a culture, although maybe with differing weights attached to
the various parts. INGOs use of ‘respect’ also seems to be tied up with the Kantian
notion of the respect that is owed to individuals, simply because they are human
and free rational beings. This respect should be theirs, regardless of their individual
character, achievements or social position. Fundamentally, the notion of respect
revolves around the concept of the giving of proper and appropriate attention com-
bined with trying to see clearly—to see something as it really is and without the filter
of one’s own likes, dislikes, prejudices, culture and so on. With this in mind, it is
therefore possible, though perhaps difficult, given our individual idiosyncrasies and
cultural background, to both respect a culture and also work to change aspects of it.
To Respect or Not to Respect . . . Ethical Dilemmas of INGO Development Practitioners 195
2 The Literature
In reviewing the literature I have been surprised by how little has specifically been
written on this particular issue. There is great work on recognizing the role of values
within international development, on the nature of global rights and obligations, and
the ethical dilemmas confronting practitioners, particularly in humanitarian emer-
gency responses, but relatively little which deals explicitly with identifying the
ethical basis for INGOs seeking to change local cultures, beliefs and practices.
Bell and Carens’ paper (2004, p. 301), which emerged out of a similar dia-
logue between theorists and practitioners as this workshop, is the only paper I
have found which specifically addresses aspects of this issue. They identify ‘four
kinds of ethical dilemmas frequently face by INGOs: (1) conflicts between human
rights principles and local cultural norms; (2) the tension between expanding and
restricting the organization’s mandate; (3) whether and how to collaborate with
less than democratic governments; and (4) the ethical limits of fund-raising.’ They
stop, however, at analysis of the dilemmas and different approaches which INGOs
have taken to resolving them, dealing which each case as a pragmatic choice, rather
than seeking to derive more fundamental principles to guide these ethical choices in
future.
To be more specific, the cases they explore under category 1 provide stark
choices. In many contexts there are a range of critical human rights issues at stake.
A case involving an agency working in Nigeria is considered, where the pragmatic
decision was taken to tolerate the clash in beliefs in relation to gay and lesbian rights,
in order to focus on other rights issues. This decision was based on the assessment
that the cultural context for challenging discrimination against gays and lesbians was
hostile. This case obviously begged the question of who determines which rights can
be ‘sacrificed’ on the altar of other values,6 leading one practitioner to pose the ques-
tion ‘what if an INGO cannot work on women’s rights or religious minorities issues,
would we find that acceptable?’ (Bell and Carens 2004, footnote 12, p. 305) This
is a question I will explore later in the paper explicitly in relation to Afghanistan
under the Talibaan. The other relevant case they consider in this category involved
Habitat for Humanity requiring 30 percent women’s representation on their local
boards. This was clearly contrary to local norms, but in Habitat’s experience it was
an effective way to begin the slow process of changing these norms.
The other relevant category of common dilemma that Bell and Carens explore
is the issue of whether to collaborate with governments that are known abusers of
human rights. They focus on different INGO approaches to the question of working
with China. Again these dilemmas are analysed as a pragmatic issues—can you
achieve your objectives—rather than how this question may be addressed, and what
the challenges are when local communities dispute your conclusions around the
ethics of working in these contexts. A more challenging case to consider may have
been the debate around working with the junta in Burma, given Aung San Suu Kyi’s
continuing opposition to INGO work in Burma.
Hugo Slim’s (2005) work exploring the gap between humanitarian ideals and
the reality of humanitarian response provides useful insights into this question of
196 C. Lenneberg
development practice. Slim posits that ‘ideals of humanitarian practice are partly
apolitical, partly moral common sense and partly deeply political and liberal’ (Slim
2005, p. 16). The apolitical aspects relate to tactical effectiveness: neutrality and
independence in order to have access to those in need in a conflict. Moral com-
mon sense is the basis for those practice standards that address the need to, not just
respond to suffering, but prevent future suffering: participation, gender sensitivity,
cooperation with local authorities and capacity building generally. But underpinning
these operational ideals, Slim argues, is a profound influence of human rights and
liberal political thinking. He further suggests that humanitarian ideals will always
be in some tension with reality, but this is no reason to be discouraged about our
capacity to realise our ideals. It is through recognising the challenge, Slim argues,
that we can be guided by ‘very practical virtues like: reasonable expectation, com-
passion, fine judgment, and a little bit of cunning.’ (Slim 2005, p. 7) This is useful
in thinking through how an ethical framework to guide reflection on the reasonable-
ness of a social change agenda in different contexts may work. Any framework for
guiding ethical choices is also going to rely on these practical virtues, though for
this task I would add to the list both humility and a capacity for reflection on one’s
own values.
Amartya Sen’s work (1999) on development as expansion of freedom and Martha
Nussbaum’s (1999) writing on social justice and women’s rights provide sound
starting points for developing a more clearly articulated ethical basis for INGO
practitioners’ dialogue with communities about culture, beliefs, practices and social
change objectives. For Sen, social justice is the critical factor in analysing the qual-
ity of life of individuals, as well as both determining and evaluating social policy
responses. He critiques utilitarian and simple economic analyses of poverty, argu-
ing that poverty is best understood as the lack of substantive freedoms which enable
individuals to determine their own destiny. These ‘freedoms are not only the pri-
mary ends of development, they are also among its principal means’ (Sen 1999,
p. 10). Using the lens of individual capabilities to exercise substantive freedoms as
basis for assessing different public policy options, inevitably requires value judge-
ments. Sen acknowledges that in ‘any choice of criteria for evaluative purposes,
there would not only be use of value judgements, but also, quite often, use of some
judgements on which full agreement would not exist.’ (Sen 1999, p. 80). His argu-
ment is directly relevant to this question about the ethical basis for social change
programs, acknowledging that value judgements can never be completely objective,
but that it is important to be transparent about the lens through which judgements
are made. It is the basis for accountability, enabling dialogue and challenge about
the judgements while not resiling from the need for judgement.
Martha Nussbaum has developed Sen’s ideas of ‘development as freedom’ and
‘substantial freedoms’ to come up with a list of ten ‘Basic Capabilities’ that she
sees as central to a life that is truly human. ‘Capabilities’ and rights overlap but are
not the same; a capability is something that a person is actually able to do, whereas
a right might only exist in theory. The example often used is that a woman might
have the legal right to vote, but not the capability, if, for instance, she is threatened
with violence on leaving the home. Her list of ten fundamental capabilities are life
(not dying prematurely); bodily health and integrity (health, shelter, food); bodily
integrity (including freedom to move, security from violence); senses, imagination
and thought (including education and freedom of thought and expression); emo-
tions; practical reason (including protection for liberty of conscience); affiliation
(being able to be treated as a dignified being whose worth is equal to that of others);
other species (concern for animals, plants and nature); play; control over environ-
ment (political and material including the ability to seek employment on an equal
basis) (Garrett 2006). Nussbaum sees these as general political and social principles
and any society that does not guarantee these at a threshold level is not a fully just
society.
For INGOs these provide a useful framework for several reasons. First, they cir-
cumvent the language of ‘rights’ and their perceived Western-centric cultural basis
(although arguably this is a perception rather than a reality, it can be a real fac-
tor). Second, they are not culturally specific: individuals and societies can decide
how they play out in practice provided the basic capabilities are protected. Third,
by stressing capabilities, it focuses on what people are actually able to do, and thus
brings the inequalities between people to the foreground. Nussbaum sees this as a
particular advantage for women within families, but the same principle applies to
other inequalities. As such, they are a goal for a social change agenda and a starting
point for an individual development practitioner’s assessment of a cultural belief or
practice.
3 In Development Practice
There is a fundamental tension between the values of respect for local cultures
and seeking to be catalysts of change within these cultures to reduce poverty and
marginalization. How do we resolve that tension? Does reference to ‘basic human
rights’ articulate a commonly understood set of ethical standards which can con-
sistently guide practice in ways that are transparent to communities? Or do we
acknowledge that our own cultural beliefs and practices influence our development
practice and be increasingly transparent about these? Related questions include:
‘who do you listen to in communities’ as often there are local dissenting voices,
and ‘what expectations do local people have of change—where does it begin and
end?’ I propose to examine these questions through considering specific scenar-
ios in development practice, focusing on the areas of sexual beliefs and practices,
the status of women, status of children, conflict resolution, and working with local
institutions and governments.
In any culture, sexual beliefs and practices are among the most sensitive, deeply
held and change resistant—for ourselves as outsiders as much as for the communi-
ties with which we work. For development practitioners, this is nonetheless a highly
relevant area of engagement as it intersects with issues of public health as well as
human rights concerns. For many years now, and despite the strong cultural endorse-
ment of female circumcision (FGM)7 in parts of Africa and the Middle East, INGOs
198 C. Lenneberg
have unequivocally fought against the practice. I would suggest INGO opposition
to FGM is not only a response to the harm caused by the immediate pain and suf-
fering to little girls, and the increased risks of infant and maternal mortality. Nor
is it based on anything explicit in the Human Rights Charter. Rather, it is equally
based on our own beliefs about what this practice implies about women’s status.
Even if this practice were performed in hospital, alleviating pain and risk associated
with traditional methods, and guided by good medical advice which mitigates risks
associated with childbirth, would that satisfy our objection to the practice? I do not
believe it would. Rather, what concerns development practitioners are also the social
motivations underpinning the practice, rooted in beliefs about women’s sexuality as
dangerous to community morals and a potential source of shame for their fathers
and husbands.
Similarly, in HIV prevention programs, practitioners are regularly brought up
against the hard clash of values around sexual practice. Addressing issues of vul-
nerability to infection because of exploitation is straightforward—i.e. girls from
child-headed households trading sex for food. It becomes more difficult where tradi-
tionally sanctioned practices themselves create vulnerabilities. For example, in parts
of rural Malawi, teenage girls (13–14 years old) are sexually initiated by an adult
male, identified only as ‘the hyena’ (Chindongo 2006). The village elders bestow
this responsibility on a single male for a defined period of time. After initiation the
girls are expected to remain chaste until marriage, which is generally arranged very
quickly. One could argue that this practice needs to stop because of the obvious risk
of HIV infection. However, this risk could also be simply mitigated by promoting
consistent condom use. Again, I would suggest our own beliefs around the sanctity
of childhood, consent and rape, and the appropriate age of sexual debut underlie our
view that there is something intrinsically unethical about this practice. In many ways
the Child Rights Convention provides the ethical basis for challenging this practice.
However, culture still mediates how people interpret the application of those rights
and INGO opposition to these kinds of practices is also linked to views about dig-
nity of children, exploitation and consent, which are just as relevant as concerns
about public health or anything explicit in the Child Rights Convention (CRC) in
driving our program response. Taking Nussbaum’s capability framework, one could
argue that justification for challenging the practice would be based, not around our
concepts of childhood, nor necessarily the CRC, but rather on how it compromises
children’s capability for achieving bodily integrity.
Beliefs and values about women’s status underpin many of those values more
directly about sexual practices and lie at the heart of INGO commitment to change.
I want to raise two issues here which demonstrate the fundamentally unresolved
tension between respect for local cultures and seeking to change culture and belief.
The OECD DAC has recently put out a series of Gender Tipsheets as a guide to
practitioners which provides the following advice:
Cultural and religious beliefs are often cited as one of the most intransigent obstacles to
implementing gender sensitive programmes promoting gender equality. Women and men
themselves must determine their strategic needs in this area, and strategies to address such
constraints must be devised and tested with great caution. Nevertheless, a human rights
approach is a direct challenge to fundamentalist or narrow definitions of women’s rightful

place in the world of work and labour, which seek social cohesion at women’s expense by
proscribing them to marginal economic and domestic roles (OECD 2002).
While acknowledging the need for communities to determine their own gender
specific strategic priorities, clearly the challenge as articulated in these guidelines
is beyond what can reasonably be expected to emerge from many local communi-
ties. For example, for most rural Afghan communities, the strict gender division of
labour meant that education beyond basic literacy and numeracy was not considered
necessary for girls. For these communities, the experience of rural life and long term
conflict further confirmed their belief that it was dangerous for women and girls to
go beyond the confines of the village. I am not suggesting that we should not find
ways to challenge these beliefs and practices, but rather that we need to acknowl-
edge that development practitioners are bringing their own values and judgement
about what is required to achieve sustainable development—not simply respecting
local cultures, beliefs and values and seeking only incremental change. This point
is further reinforced by recent research on the universality of humanitarian values,
undertaken by the Feinstein International Centre. The specific report on Afghanistan
concluded that:
The humanitarian values propounded by aid agencies are generally not a problem in
Afghanistan. With the rather large exception of women’s rights and women’s employment,
which remain points of friction, there is, broadly speaking, an easy fit between the values
of the outsiders and the traditional beliefs and mores concerning, for example, the rights of
civilians in armed conflict. (Feinstein International Centre 2006, p. 9)
And the ‘rather large exception . . .’ is precisely the point. Our belief in the equal
rights of men and women as being crucial to genuine development is based not
only on evidence of gender equity contributing to sustainable change in poverty
and marginalisation, but also on our values around the equal dignity of women and
men, together with a particular conception of what that means.8 Okin (1994; 1998)
describes how the definition and implementation of human rights have been chal-
lenged, developed and expanded in the years around the 1993 UN World Conference
on Human Rights as feminists in Western and developing nations focussed specif-
ically on the rights of women. Early concepts of the ‘rights of man’ (the precursor
of later statements of ‘human rights’) were just this—the rights of (assumed) male
heads of households against others and the government, with a separate ‘private
sphere’ which was self-governing and free from outside interference. However,
many of the violations of women’s human rights occur within the private sphere
and have thus tended to be less visible and neglected. In addition, power differen-
tials within households were not taken into account, leading to a male-oriented view
of human rights and their protection and implementation.
A clearer focus on women’s rights and the disaggregation of data, arguments and
practices by gender, has also brought a focus on cultural differences—whose cul-
ture, whose rights, who defines the culture?—and whether and how they should be
‘respected’. In Okin’s view ‘the term ‘respecting cultural differences’ has increas-
ingly become a euphemism for restricting or denying women’s human rights . . ..
200 C. Lenneberg
the relevance and even the sanctity of ‘cultural practices’ is most often claimed
when issues of sexuality, marriage, reproduction, inheritance and power over chil-
dren are concerned’ (Okin 1998, p. 34). Among development practitioners too there
is a diversity of views on gender equity. One of the starkest examples of this was the
preparedness of some to ‘sacrifice’ women’s rights on the altar of other values in
Afghanistan. When the Talibaan took over Kabul, and imposed their edicts denying
girls education9 and women work outside of the health sector, many INGO and UN
leaders were quick to comply with these requirements, arguing they were simply
culturally determined choices or that to do otherwise would cause greater harm.
At the same time, from about 1997 till the collapse of the Talibaan in 2002, other
INGOs began covertly supporting ‘hidden’ schools. These schools were primarily
established by women in their homes, in order to teach local girls and boys and allow
women to support their families. With the withdrawal of women teachers, not only
were girls denied access to education, but also for boys the quality plummeted as
class sizes tripled or more. These ‘hidden’ schools were established at great personal
risk to the teachers and parents involved, revealing the value a significant minority
placed on education in deliberate defiance of regime sanctioned values. This would
fit with the argument that solidarity with local groups provides the ethical basis for
an INGO social change agenda. The problem with this argument is that it begs the
question of with whom ones stands in solidarity. Without an underlying notion of
social justice, the solidarity argument doesn’t work. The other particular dilemma
this case study poses is around working with governments. In Bell and Carens’
example, the rationale proffered by some human rights organisations for working in
China was that, while unable to directly challenge the regimes human rights abuses,
they were able to subtly progress the human rights agenda through their programs.
In Afghanistan there was no reasonable expectation that INGOs through their sup-
port of hidden schools would shift regime values and practice around education
policy. Support was provided, despite the significant attendant risks and in response
to the views of only a portion of the community, precisely because it accorded with
strong INGO values and beliefs around justice, supported in part by Human Rights
Conventions.
Some interesting issues highlighting the dilemmas around divergent values of
local communities and development practitioners are also beginning to emerge
from recent research by the Collaborative for Development Action.10 The Listening
Project aims, through listening to local people (both recipients and non-recipients
of international assistance) to gain insight into their views of the value and out-
comes of that assistance. One of the preliminary reports, in addition to the expected
issues of conflicting values, views and expectations of assistance and how it is deliv-
ered, points to local views on the ‘cascading effects of external agendas’ (Woodrow
2007). They reported that
[m]any people with whom LP Teams talks described problems arising from the influences
of outsiders’ priorities, trends and agendas in determining the types of interventions and
assistance they received. For example, in Bosnia, virtually all of the donors agreed on and
funded efforts to support the return of refugees, and most of the international and local
NGOs implemented these projects even though they were not what the people on the ground
wanted at the time.
Unfortunately further details were not provided, but one can surmise that, as
commonly in the case of conflict, local beliefs and practices around the rights of
certain groups to land explicitly excludes the claims of others to the same land.
INGOs are often at the front line of trying to mediate/ reconcile the claims of differ-
ent groups—whether in areas of ethnic conflict, or competing claims of pastoralists
and agriculturists or even in state-sanctioned community relocations. Action clearly
cannot respect the competing traditional beliefs and practices, but rather must
impose external standards of justice, together with a healthy dose of pragmatism,
to reconcile them.
In reflecting on the issue of the ‘cascading effects of external agendas’ I was
reminded of a recent field visit to a World Vision Area Development Program in
Jaipur, India.11 One of the traditional occupations of the Muslim community in this
region is gemstone cutting. Boys are apprenticed at an early age, often only 5 or 6
years old, to polish gemstones, 6 days a week for a pittance.12 However, once they
are skilled, after many years, they have the opportunity to make a decent living from
their trade. The WV program is seeking to convince parents to send their children to
school instead of work, but without much success. While poor, poverty is not the key
driver for the community’s decision but rather culture, tradition and an alternative
economic rationale. This community considers formal primary school education to
be a waste of time when their sons can be learning a useful trade in the same time-
frame. For World Vision the issue is important because of our agenda—our mission
which aims for ‘life in all its fullness for every child.’ This view encompasses our
commitment to the MDGs as well as our conviction that children need to have a
broader education to survive and thrive in a rapidly changing world. Our agenda
and values would appear to be simply inconsistent with of the majority of the com-
munity, as is our belief that education is essential for every child to equip them
with the basic capabilities they require as adults to achieve their full potential. So,
what is the legitimacy of our agenda in context? Okin (1994, p. 19) suggests that
‘committed outsiders can often be better analysts and critics of social injustice than
those who live within the relevant culture’ so the view of an INGO on a particular
cultural practice can have validity not despite it being an outside view but precisely
because it is a view from outside—which may be so, but is still insufficient in and of
itself, as well as not respectful, as the basis for seeking to change local customs and
practices. As Jamie Isbister’s analysis of the issues around social policy in Australia
with respect to indigenous children highlights, good people with good intentions
can reach conclusions about the kinds of social change required that later genera-
tions recognise as at best ill-advised and enormously damaging.13 A more rigorous
process of reflection and engagement with affected communities is required than
simply a well-intentioned, impartial external assessment.
4 Conclusions: A proposed Ethical Basis
To return to the key question that this paper began with—i.e. ‘what is the ethi-
cal basis for INGOs challenging and seeking to change local cultures, belief and
practices’? This analysis of the literature and development practice would suggest a
202 C. Lenneberg
number of tentative conclusions. These are proposed as the basis for further dialogue
and debate among practitioners.
Firstly, the ethical basis for development practitioners challenging and seeking
to change local cultures, beliefs and practices is a fundamental concern with social
justice, firmly grounded in the Human Rights Conventions. These Conventions
recognise the equal value and dignity of all human beings simply by virtue of their
shared humanity, and not their achievements or capabilities. However, the analy-
sis of development practice dilemmas highlights that these conventions are not by
themselves sufficient articulation of the rationale for social change in all contexts. I
suggest that deeper values and interpretations of what equal value and dignity mean
in different contexts underpin much development work.
Secondly, Sen and Nussbaum’s ‘capabilities framework’ provides a more com-
prehensive basis for evaluating social justice issues, as well as the development
policies and interventions seeking to address them. Within this framework, it is
possible to reconcile the practitioner’s expressed commitment to respect ‘the dig-
nity, values history, religion and culture of the people with which they work’, with
actively seeking to change those aspects of culture which are constrain the realisa-
tion of fundamental freedoms. Respect for people does not require acceptance of
any pernicious beliefs and values which deny the value of individuals or groups of
individuals.
Thirdly, professional judgement and impartial assessments provide further under-
pinning for the legitimacy of the development practitioner’s social change agenda.
The perspective of the concerned but independent observer is both valid and
influential. However, it needs to be grounded in clearer articulation about values
and transparent dialogue with affected communities about the proposed objectives
of any social development program. This is critical to any notion of downward
accountability.
Finally, a conscious process of critical reflection is required to ensure that
social development programs are not simply based on individual practitioner or
agency likes, prejudices and preferences, but rather on a thoughtful analysis of the
social justice issues and context for change. The following questions are proposed
to guide critical reflection about the legitimacy of proposed social development
interventions:
• What freedom or capabilities are being denied, constrained or not realised by

particular local values, traditions or behaviours?
• What is the practitioner’s own reaction to these values, traditions or behaviours?
• How would the proposed intervention contribute or detract from realisation of
greater freedom/capabilities?
• How realistic is the proposed change and at what social cost? That is, the issue
of grounding idealism in practical realities.
In cross-cultural settings, the capacity to critically reflect on the values we bring

to any engagement is vital for the legitimacy and effectiveness of any development
programs.
INGOs explicitly define themselves as values-based organisations and acknowl-

edge that their values underpin their development practice. While the sector
encompasses significant diversity, fundamentally shared values include a commit-
ment to not just alleviate, but to addresses the causes of, poverty and marginalisation
as well as to enable communities to meet their basic needs and participate in
determining their own futures in meaningful ways.14 This paper is written from
a practitioner perspective to prompt thought and discussion about a key aspect of
greater accountability to the communities we work with—that is, to acknowledge
the often enormous gulf that exists between our conceptions and theirs of human
rights, equal dignity, basic needs and meaningful participation. This is a funda-
mental requirement for any genuine, respectful dialogue about social change and
downward accountability. While faith-based agencies are sometimes critiqued for
the role their values play in their work, we need to recognise that the influence of
practitioner values is universal and inevitably determines where respect for local
culture, beliefs and practices begins and ends. This does not negate the critical need
for practitioners to respect ‘the dignity, values, history, religion and culture of the
people with whom’ we work. Rather it requires us to be more reflective and trans-
parent about how far our own values influence our assessment of what is, and is not,
‘consistent with principles of basic human rights.’
Notes
1. Australian Council for International Development, which is the Australian International
Development Non-Government Organisations peak body.
2. Despite the fact that many agencies continue to fall short of this goal in their programming.
3. Academic position rather than that of any specific INGO.
4. This position is best exemplified by traditional missionary organisations of the nineteenth
century who tried to eradicate practices from idol worship to polygamy, child marriages and
foot binding.
5. For example SPHERE, HAP, ALNAP, Red Cross Code of Conduct, ACFID, Interaction.
6. Which was discussed but not resolved in the workshop that led to this paper.
7. Female circumcision is more commonly referred to by INGOs as Female Genital Mutilation,
which articulates the implicit value judgement of this practice.
8. Similarly in the UNDP Human Development Report 2004, in the essay on ‘Cultural Liberty’
the claim is made that one of the sources of global ethics is ‘the belief in the basic moral equal-
ity of all human beings’ (p. 90) But how that translates into practical rights and opportunities
is the issue which is vehemently contested.
9. Girls could be educated up to grade 4, as long as separate school facilities and teachers were
available, which in effect denied the majority any access to school.
10. Established by Mary Anderson, who wrote the Do No Harm: Building Local Capacities for
Peace, series. The Listening Project sits within this Not For Profit Agency.
11. Jaipur Area Development Program, author’s field visits in October 2006 and January 2007.
12. Some boys receive as little as $1 a week while they are learning.
13. Jamie Isbister, discussion of his paper presented at the Workshop: Ethical Questions for Non-
Governmental Development Organisations. (Centre for Applied Philosophy and Public Ethics,
Charles Sturt University and Oxfam Australia).
204 C. Lenneberg
14. The ubiquity of PRA/PLA as a key approach in the sector provides expression of this shared
value as ‘it seeks and embodies participatory ways to empower local and subordinate peo-
ple, enabling them to express and enhance their knowledge and take action.’ (Chambers
2007, p. 3).
Acknowledgement I would like to acknowledge Anna Roche, who provided valuable research
and editorial support to the revision of this paper.
References
Bell, D.A. and J.H. Carens. 2004. The ethical dilemmas of international human rights and human-
itarian NGOs: Reflections on a dialogue between practitioners and theorists. Human Rights
Quarterly 26: 300–329.
Bell, D.A. and J.-M. Coicaud. 2007. Ethics in Action: The Ethical Challenges of International
Human Rights Nongovernmental Organisations. New York: Cambridge University Press.
Chambers, R. 2007. From PRA to PLA and Pluralism: Practice and Theory. IDS Working Paper
286, July.
Chindongo, M. 2006. WVMalawi, Operations Director Discussing WVA-funded HIV & AIDS
Programs.
Crocker, D.A. 2000. International Development Ethics, Paideia, http://www.bu.edu/wcp/Papers/
OApp/OAppCroc.htm.
Crocker, D.A. (2007). Development ethics, globalization, and stiglitz. In Globalization,
Development, and Democracy: Philosophical Perspectives, eds. M. Krausz and D. Chatterjee.
Rowman & Littlefield: Lanham, MD.
Feinstein International Centre. (2006). Humanitarian Agenda 2015: Principles, Power and
Perceptions, Preliminary Report. Medford: Friedman School of Nutrition and Science Policy,
Tufts University. http://fic.tufts.edu/downloads/HA2015PreliminaryReport.pdf.
Garrett, J. (2006). Martha Nussbaum on Capabilities and Human Rights. Kentucky: Western
Kentucky University. http://www.wku.edu/˜jan.garrett/ethics/nussbaum.htm.
Gasper, D. 2000. Anecdotes, situations, histories—Varieties and uses of cases in think-
ing about ethics and development practice. Development and Change 31: 1055–
1083.
Mathiesen, K. 1999. Review of Nussbaum, Martha. Sex and Social Justice. Oxford University
Press: New York.
Nussbaum, M. 1999. Sex and Social Justice. Oxford University Press: New York.
Nussbaum, M. 2003. Capabilities as fundamental entitlements: Sen and social justice. Feminist
Economics, special issue on Amartya Sen 9(2&3): 33–59, July.
Nussbaum, M. 2003. Beyond the Social Contract: Capabilities and Global Justice.
http://www.unipv.it/deontica/sen/papers/Nussbaum.pdf.
OECD. (2002). Gender Equality and Human Rights Tipsheets. Paris: Development Cooperation
Directorate, OECD. http://www.oecd.org/document/34/0,3343,en_2649_34541_1896290_1_
1_1_1,00.html.
Okin, S.M. 1994. Gender inequality and cultural differences. Political Theory 2(1): 5–24.
Okin, S.M. 1998. Feminism, women’s human rights, and cultural differences. Hypatia 3(2): 32–52,
Spring.
Rawls, J. 2002. A Theory of Justice. (selections). In Ethics: History, Theory, and Contemporary
Issues, eds. S.M.Cahn and M. Peter, 2nd edn. New York and Oxford: Oxford University Press.
Sen, A. 1999. Development as Freedom. Oxford: Oxford University Press.
Sen, A. 2000. Social Exclusion: Concept, Application and Scrutiny, Social Development
Paper No. 1, Office of Environment and Social Development, Asian Development Bank,
June.
Slim, H. 2005. Idealism and Realism in Humanitarian Action. Two talks given at the
ACFID Humanitarian Forum, Canberra, 5 Oct. Available at http://www.acfid.asn.au/resources/
speeches/?searchterm=ACFID%20Humanitarian%20Forum%202005.
Stanford Encyclopaedia of Philosophy 2003. Respect, The Metaphysics Research Lab,
Centre for the Study of Language and Information, Standford University, Standford,
http://plato.standford.edu/entries/respect.
UNDP. 2004. Human Development Report 2004: Cultural Liberty in Today’s Diverse World. New
York: UNDP.
Woodrow, P. 2007. Presentation to the Australian Council for International Development,
Humanitarian Reference Group Meeting, Canberra, June.
Ethical Behaviour in Non-government
Organisations
Linda Kelly
1 Introduction
What does it mean for NGOs to behave ethically? Most NGOs operate from a value
base and seek to promote those values more widely in the world. As organisations
they do good work, they seek to help poor people and others who are marginalised
in their society and generally try to make the world a more just and reasonable place.
What does the consideration of ethics add to this value based work?
In considering these questions I want to draw from case studies of different types
of NGOs and examine what might be ethical about their behaviour. The case studies
appear to illustrate two types of ethical considerations for NGOs. First that NGOs,
which seek support from others on the basis of their good work ought to behave with
integrity, that is they should do what they claim to do. Second and perhaps more
importantly NGOs might be considered to be behaving ethically if they choose to
undertake their interventions with people as effectively as possible.
These propositions have implications for the way NGOs might be understood
and classified. They also have implications for the way in which others might be
able to assess the merit of development organisations and the degree to which they
deserve public and other support.
2 Case studies
The three NGOs described below are all Australian based, they all have an interna-
tional focus and are committed to creating change which will benefit others. Apart
from these similarities, they are all very different sorts of organisations, reflecting
some of the wide diversity and difference that exists across the sector.1
L. Kelly (B)
Praxis Consultants Pty Ltd, Sutton, NSW, Australia
e-mail: linda.kelly@praxisconsultants.com.au

208 L. Kelly
2.1 Caritas Australia

Caritas Australia is the development arm of the Catholic Church in Australia. As
a middle size international NGO2 it works across countries in Africa, Asia and the
Pacific.
The mandate of Caritas Australia includes the following statement.
We work in the field through our local partners. The partnership approach is central to our
philosophy that international partners are best placed to identify their own community’s
needs and problems. Working through local partners is central to ensuring sustainability
as the local communities design and manage their own development programs in manner,
which is culturally appropriate and owned by the community.
This statement is quite typical of what a number of NGOs would claim about
their work. That is, that they work through local partners and therefore that their
development interventions are determined and controlled by local people and that
the relevance, sustainability and legitimacy of their work is maintained by this local
control and direction.
An enormous amount has been written and said about the nature of partnership
between rich country NGOs and the poor country local agencies that receive the
funds and undertake much of the actual work known as development. Most of it is
not complimentary (Adong 2005). Recent research has suggested that local NGOs
and community based organisations (CBOs) in poor countries tend to be driven by
the current fads and trends of their donor agencies (Kilby 2006); rarely is the power
distributed evenly. While the intention may be to characterise the relationship as
one of equal partners, the reality is that the money, power and control largely reside
with the rich country NGOs. There have been repeated calls to shift the power and
control of the relationship between donor agencies and recipients (Adong 2005) and
to focus on processes that will build a trusting and honest relationship between the
two (Eyben 2005).
The reasons for such a focus on the nature of relationships between donor and
recipient agencies are important. The research on NGOs and aid effectiveness sug-
gest very strongly that more effective outcomes are achieved when donor agencies
cede control and certainty of the aid interventions to those closer to the intervention
itself (Roche and Kelly 2005). Better development comes from less control by the
donor and more direction and participation by the partner agencies.
Over the past few year Caritas has been exploring the basis of the relationships
they have with their partners. For them, as for most international NGOs, the main
relationship they have with their partner agencies in poor countries is constructed
around the transfer of funds for projects. That transfer depends upon those agen-
cies presenting a funding proposal that meets the basic requirements of Caritas’
accountability systems.
One of the difficulties in this type of relationship has been the notion of project
proposal ‘design’. Similar to other NGOs and donors, Caritas has largely relied upon
a ‘management by objectives’ approach to project design, using a well known tool
called the ‘logical framework analysis’ to shape their interpretation of an adequate
Ethical Behaviour in Non-government Organisations 209
and comprehensive design. Much has been written about the inadequacy of this
approach to aid management and design (Hailey and Sorgenfrei 2003, Easterly
2006) but the approach persists, in part because it presents a simple and linear
approach to aid intervention that appears to give control and logic to the whole
process. It’s easy to communicate to public and official donors (indeed almost
all official donors such as AusAID in Australia, use this design approach as their
default method). NGOs in Australia tend to use a simplified version of the approach
but still generally adhere to the key principles of the logical framework in their
conceptualisation of projects.
One of the key issues for the approach however is that is creates an artificial view
of how change happens and how development interventions will be managed (Eyben
2005). This in turn forces partner agencies and community based organisations to
present their work and requests in an artificial and inadequate framework, squeezing
their view of the world into a linear and cause and effect process that implies easy
movement from inputs to change. In fact the aid world is a messy and unpredictable
place where partner agencies and communities have to balance many competing
priorities and problems in order to try to effect change and where the actual process
of change is far more dynamic and challenging (Marsden 2003). Having to use an
approach that does not represent reality further undermines the power and control
of partner organisations.
For some time now there has been work developing on alternative approaches
to describing and delivering aid and development. Many of these approaches are
built on the more complex ways of knowing about the world and include models
which allow for change, adaptation and unexpected outcomes, given these are all
characteristic of the development world in which we work (Brinkerhoff and Coston
1999).
Partner organisations and people in poor countries are generally more able to
engage with such approaches. Partners have more opportunity to represent their
work and their proposals in ways which are congruent with their real experience.
They are able to present some of the messiness, unpredictability and uncertainty
which are the reality of most development work, but which is usually ignored in the
projects and proposals designed for marketing to either institutional donors or to the
public.
As a result partners are given some more control over the project funding arrange-
ment. The projects become more based in what exists and more attuned to what is
possible, and rather less driven by external views about what projects should do and
how they will proceed. Yet, despite this the logical framework approach to project
design continues as the dominate model. Partners continue to experience their rela-
tionship with most donors as disempowering and controlling. Aid continues to be
less effective than it might be able to be. In part this appears to be because changing
design approaches and systems is hard and is likely to lessen the control of NGOs.
Caritas is one NGO that has decided to tackle this dilemma. They have explored
approaches to design and have decided to adopt those approaches which best match
their partner’s experiences and approaches. They have redesigned their own systems
and processes to accommodate these new design approaches. They have given up
210 L. Kelly
some control in favour of better control and understanding for their partners. This
has not been without some cost. Already some questions have been asked about the
quality of the designs and proposals being developed under this new approach, yet
Caritas has chosen to move in this direction in order to meet their stated intention of
working in partnerships with organisations from poor countries.
So what is ethical about this example? First and perhaps most obviously, that
Caritas Australia is actually trying to behave with some integrity in being true to its
public statements. They have chosen to adjust their own organisational arrangements
to bring about the true partnership approach they claim to use.
The second aspect of the organisational behaviour is less clear but perhaps more
important. If enabling partners to have more control and participation in the way
aid and development is delivered is likely to lead to more effective outcomes then
changing their relationship with partners is also an attempt by Caritas to be more
effective as an agency. Caritas Australia has struggled with changing itself in order
to bring about the partnership process because it is also committed to effective aid
and development.
These changes have been undertaken in spite of the difficulties of so doing
(Caritas has had to change its own systems and educate and challenge its own staff);
and the costs of so doing (Caritas faces some risk that donors and supporters might
withdraw their resources from less predictable and assured aid interventions).
2.2 Oxfam International Youth Partnerships
The next example comes from a different type of agency, the Oxfam International
Youth Partnerships (OIYP). OIYP is an initiative of Oxfam International, managed
by Oxfam Australia. It focuses upon work with young people across the world, with
a particular focus on working with young people in Asia, Pacific, Africa and South
America. The aim of OIYP is to skill and equip those young people to become
change activists within their communities and their countries. OIYP shares the val-
ues and vision of the wider Oxfam organisation buts sees itself as a network of
young people.
OIYP started with the assumption that it is not possible for it to know everything
that will be required to achieve the impact it wants to see after several years of work.
The staff and members assume that they will have to learn as they go. So OIYP
has focused its growth and development around a process of critical action reflec-
tion (Schön 1983). That is, acting, learning from that action, and then undertaking
improved action. It has adopted a model where the staff and members are increas-
ingly clear about what impact they want to achieve and yet also increasingly clear
that they will have to learn as they go, improving their strategies and approaches in
the process. They expect to be different from year to year.
The immediate level outcomes of all this are not clear, but they are committed to
change and improvement in order that the longer term goal of the impact they have
collectively identified (which is in large part that shared by the wider parent body
of Oxfam Australia) will be achieved. OIYP intend to remain relevant and able to
act with the widest possible knowledge of the situation precisely by choosing not to
pretend they know everything before they start.
OIYP has not adopted this approach with total ease. Articulating the approach has
been challenging in itself, and the management and accounting processes required
by others does not always allow for the flexibility and agility OIYP are trying to
pursue
Why is it significant? One of the common requirements made of effective
development organisations is that they should become ‘learning organisations’
(Sorgenfrei and Wrigley 2005). What this actually means is often very vague, but
the idea essentially arises from the common view that the development world is
complex and ever-changing and that for organisations like international NGOs to
be effective they have to be constantly ready to receive and learn from new infor-
mation, knowledge and experience, and be able to improve their approaches and
interventions as a result of what they learn (Easterly 2006). This in turn challenges
their organisational structure, suggesting they have to remain flexible, able to adapt
and change according to learning and new information.
Key to a learning organisation is that it exists only in the sense of a complex set of prac-
tices, systems and relationships which link the organisation’s vision, mission, values and
behaviour to desired outcomes and results. An effective learning organisation will con-
stantly question its assumptions and review its objectives in light of its own experience and
changes in the external environment (Britton 1998).
So once again what is ethical about the development approach chosen by OIYP?
This network of young people has reviewed the ways in which other development
agencies operate and to chosen to use a process of internal development at odds
with the practice of many traditional international NGOs. However, according to
the research undertaken on development practice, it is an approach which is more
likely to lead to a more effective development organisation and thus higher quality
development outcomes.
In this case, the ethical stance appears to be related less to what OIYP present to
others and more about choosing, despite some difficulties and other costs, to be true
to a way of operating that appears to lead to more effective aid.
2.3 Australia Rwanda Network
The last example is an agency that sits even further outside the mainstream of
Australian NGOs. The Australia Rwanda Network has a goal
. . . to contribute to the nation building of Rwanda and Australia through the development
of personal networks and mutual understanding between the two countries.
The Network operates to develop personal links and networks between the peo-
ples of Rwanda and Australia at the community, civic and government levels, in
order to support a development process which builds on the existing strengths of
those communities, enabling them to participate in building their respective nations.
212 L. Kelly
The aim of the Network is to bring about change. But in this case the mandate is
clearly stated as change in both Rwanda and Australia. While not entirely defining
what that change might be, although it is clearly related to increased development
opportunities for people and increased quality of life, the intention is that in working
together on projects and strategies for improvement for people, people from both
countries will be changed. Largely that means the Rwandan people gaining some
material or other benefits. For the Australians it means challenges and changes to
their attitudes, lifestyles and maybe their actions.
The Network has developed as a formal organisation following several years of
relationships and contacts between the two countries. The Network does not seek
to control the relationships which are developed, although there is some control
on quality and intentions of these contacts. But mostly it is about facilitating and
allowing those relationships to flourish. To date the Network has facilitated several
relationships between schools, local governments and other local organisations in
both countries.
Rather than a one-way helping relationship, they are promoting a mutual change
arrangement where everyone is changed and challenged by their interaction with
people from another country. The process shows respect for people both from
Rwanda and Australia, recognising their different but relevant skills, knowledge and
experiences.
So why is this significant? In part it reflects a response to a growing trend in
the development sector. There are increasing examples where individual people and
communities are bypassing the formal development agencies in their quest to assist
people in poor counties. The response to the Tsunami disaster in 2004 and other
recent disasters has seen a growth of non-NGO assistance coming into countries
and situations. People hear about a problem, they want to help and they, perhaps
with assistance from friends and colleagues, fly into Sri Lanka, Indonesia, India or
wherever and set about providing their own form of assistance.
This sort of amateur approach has some drawbacks. There are many horror
stories told of the mistakes made by well meaning and yet inexperienced people
striding into the murky world of international aid. However the sentiment and the
drive to help and to do so personally, appear to be growing rather than diminish-
ing. It is a trend driven by the ease of communication and international travel. It
also appears driven by the need for personal experience and personal interaction in
bringing about change.
More importantly however, the international experience suggests quite strongly
that developing relationships between people in rich and poor countries, beyond the
funding process, is critical for effective development (Brehm 2004). That relation-
ships should develop to be more mutual and based on learning and sharing processes
(Chambers 2005). Also that people in the richer countries themselves have to change
as well as those in poorer countries if we are to move to more sustainable solutions
to poverty and alienation. Yet most international NGOs do not offer their support-
ers that sort of direct relationship and experience. Generally international NGOs
claim to embrace the notion of supporter engagement, yet too often the engage-
ment is limited to fundraising and perhaps some related advocacy work. The actual
development of relationships with people in poor countries is largely controlled and

administered by the NGO itself.
The Australia Rwanda Network has chosen to embrace an approach which is
based upon this direct contact between people. The organisation has taken seriously
the notion that people need to be respected and allowed to act for their own develop-
ment. They have accepted the need for change in both rich and poor countries as a
means to addressing poverty and have established a structure which will be focused
on such mutual change.
Once again this agency appears to be acting ethically because not only does it
clearly try to be and act in a way that corresponds to its stated intentions, it also is
acting in a way that promotes more effective outcomes. The Network has identified
what they consider needs to be changed in order to bring about the most effective
outcomes for poor people. In this situation it is the idea that change has to be mutual
that is it has to involve both those who are in need as well as those who want to
help. Despite some pressure to conform, especially in order to meet tax deductibility
requirements and to be accepted more generally into the NGO sector, the Network
has quietly pursued what to them seems the appropriate way to operate based on
their values.
3 Implications
So what do these three examples suggest about ethical behaviours in NGOs. First it
is important to say that there are other examples of NGOs which one could argue
do behave ethically. These three examples are useful because they represent very
different types of NGOs and also because the changes they have chosen or the struc-
tures they have adopted are very different to many of the more traditional NGOs in
Australia.
Second there seems to be different ways that ethical behaviour could be
interpreted from these examples. As discussed, the most obvious is that these organ-
isations or networks are trying to operate in ways which match their stated intentions
or descriptions of themselves. They do this in different ways but each has struggled,
with varying degrees of opposition and cost, to try to be what they claim to be.
To act with integrity. Caritas has tried to ensure its does work in real partnership
with agencies in poor countries. OIYP has tried to operate using a model of criti-
cal action reflection so that it can be the learning organisation it claims to be. The
Rwanda Australia Network has resisted attempts to become project focused in order
to promote the relationship base it believes should be the way in which people in
rich and poor counties work for change.
The other way in which these agencies could be considered to be behaving eth-
ically is in their commitment to behaviour or organisational structures which they
consider will contribute to more effective aid and development.
There have been various standards and frameworks developed to measure NGO
effectiveness. Recent research in Australia (Kelly and Chapman 2003; ACFOA
214 L. Kelly
20023 ) and elsewhere seems to confirm the view that aid effectiveness is the result of
a complex set of processes and interrelationships (Smillie and Hailey 2001; Moore
et al. 2001). There now seems to be some agreement among donors and NGOs that
assessment of effectiveness has to go beyond the outcomes (however well measured)
of single activities, projects or programs (Lehtinen 2002, UNDP 2003, Hofmann
et al. 2004, Interaction 2006). Increasingly the assessments being undertaken of the
work of donors and multilateral organisations are pointing to the need to examine
their field based achievement and also the synergy and calibre of internal systems
and the values and approaches underlying their work (Fowler 1997, Scott 2004,
Stewart 2003, DFID 2003).
So there is now more agreement about what makes for effective aid and in par-
ticular what causes NGOs to be more effective. It is not simply about what they do,
it is also about who they are as organisations, their staff, leadership and their sys-
tems and the consistency between their organisational values and how they operate
(Fowler 1997, Hailey and Sorgenfrei 2003).
It is not easy to assess and measure this notion of effectiveness. There is limited
evidence about NGO effectiveness in part for this reason. It obviously means that
we need to know more about NGOs than what they do in projects and programs
but it is very hard to capture and convey the complicated inter-relationship of dif-
ferent standards, organisational systems, staff quality and overall agency coherence
with its own values, which all need to be considered in judging the quality of an
organisation and its work (Radelet 2006).
It is also hard to ‘do’ this type of aid and development work. The literature on
effective aid presents a number of organisational challenges which are not neces-
sarily in line with the way traditional development NGOs organise themselves or
present themselves to the wider world. The three examples given here suggest that
in order to promote effective aid NGOs need to consider changing their systems,
their organisational forms and their ways of relating to supporters, partners and oth-
ers in order to meet some of the emerging views on what makes for more effective
aid. NGOs in Australia, as elsewhere, have been slow to embrace a notion of change
and development in their organisational form. Despite the classifications of NGOs
that suggest a pattern of growth and change, the reality is that most NGOs still
appear to value organisational growth as the main indicator of increased effective-
ness. In Australia, NGOs do not vary too much. They certainly incorporate different
ideas and trends and program approaches, but they rarely allow these to actually
challenge the way they operate or the form they take.
However if NGOs are committed to their stated intentions of helping poor and
marginalised people as effectively as possible, as they become aware of what would
make for more effective aid, one would expect that they would embrace that fea-
ture and make it part of how they operate. Despite how hard it may be, if NGOs
were acting principally to meet their intentions then one could expect to see them
struggle with organisational forms and different systems and processes which match
some of the emerging notions of how development can be more effective (Fleshman
2004). There would be a ‘narrowing of the gap between words and actions’
(Adong 2005).
In the three case study examples there appears to be evidence of different NGOs
struggling with this idea. As they understand what makes for more effective aid they
have worked to change themselves as organisations or to change their systems, staff
and their relationships with their supporters in order that they can be more effective.
They have all done so at some cost. NGOs would seem to be acting more ethically
as they choose this approach.
Finally, does it matter? In a world where there is so much to be done to assist poor
people, what does it matter if some organisations are operating more ethically than
others? This paper suggests that depending on how you define ethical behaviour it
may well be critical.
If ethical behaviour for NGOs means that they do what they say, that might be
important because it indicates their integrity as an organisation. And maybe we want
to support organisation which act with integrity. So these organisations may be more
worthy of support because of what they are.
However if ethical behaviour also means NGOs are acting as effectively as pos-
sible to fulfil their mandate of helping the poor and marginalised, even to the extent
of changing themselves and their systems, then knowing this would provide impor-
tant information for potential donors who want their resources and support to be
as effectively utilised as possible (Wenar 2006). In this case assessing the ethical
approach of an NGO becomes one way for outsiders to judge the value of what that
organisation does. They might be considered more worthy of support because they
are more likely to be able to achieve the very outcomes NGOs exist to fulfil.
Notes
1. The Australian NGO sector is diverse and has developed form a different history and set of
influences to those comparative sectors in North America or Europe (Crooke 1996). While
this is changing with the more recent globalisation of many of the larger Australian based
NGOs (Oxfam, CARE, World Vision, Plan, Save The Children and so on), it is still impor-
tant not to generalise from the experience of international NGOs based in other parts of
the world.
2. Caritas Australia belongs to an international network called Caritas Internationalis, one of the
largest aid and development agencies in the world comprising a network of 162 Catholic relief
aid, development and social service organisations working, in over 200 countries and territories.
In many ways it is similar to other middle size and larger Australian NGOs which operate within
networks or confederations around the world.
3. ACFOA is now known as ACFID (Australian Council for International Development)
References
ACFOA. 2002. Research Report for the ACFOA Strategy on Quality, NGO Effectiveness
Conference, July, Australian Council for Overseas Aid, Canberra.
Adong, R. 2005. The current procedures and policies dominating the disbursement of aid: Are they
building strong relationships and enabling NGOs to meet their stated aims? Research report
funded by ESCOR and DIFD, UK.
216 L. Kelly
Brehm, V. 2004. Autonomy or dependence? North-South NGO partnerships, INTRAC Briefing

Paper No. 6, Oxford.
Brinkerhoff, D.W. and J.M. Coston. 1999. International development management in a globalised
world. Public Administration Review 59(4): 346–361, July/Aug.
Britton, B. 1998. The learning NGO, Occasional Papers Series, No. 17, INTRAC, Oxford.
Chambers, R. 2005. Aid and administrative capacity. In Ideas for Development, ed. R. Chambers
Oxford: Earthscan.
Crooke, M. 1996. NGOs and official development assistance. In Australia’s Aid Program: Mixed
Messages and Conflicting Agendas, ed. P. Kilby. Melbourne: Monash Asia Institute and
Community Aid Abroad.
DFID. 2003. Better Government for Poverty Reduction: More Effective Partnership for Change.
Consultation document. London: Department for International Development.
Easterly, W. 2006. Planners vs. Searchers in Foreign Aid. Asia Development Bank Distinguished
Lecture Series. Manila: ADB.
Eyben, R. 2005. Donors’ learning difficulties: Results, relationships and responsibilities. IDS
Bulletin 36(3): 98–107, Sep.
Fleshman, M. 2004. Africa pushes for better aid quality. Africa Recovery 17(4): 18, Jan.
Fowler, A. 1997. Striking a Balance. London: Earthscan/INTRAC.
Hailey, J. and M. Sorgenfrei. 2003. Measuring success: Issues in performance management.
Keynote paper at the INTRAC 5th International Evaluation Conference, Netherlands.
Hoffmann, C., L. Roberts, J. Shoham, and P. Harvey. 2004. Measuring the Impact of Humanitarian
Aid. Humanitarian Policy Group. London: ODI.
Interaction 2006. NGO impact initiative, synthesis report, Oct, USA.
Kelly, L. and R. Chapman. 2003.Australian NGOs and AusAID—Partnerships for quality assur-
ance and improving effectiveness. Presentation to the joint French and DAC working party
on evaluation, workshop on partners in development evaluation—learning and accountability,
25—26 Mar, OECD, Paris.
Kilby, P. 2006. Accountability for empowerment: Dilemmas facing Non-governmental organiza-
tions. World Development 34(6): 951–963.
Lehtinen, T. 2002. Measuring the Performance of EC Development Cooperation: Lessons from the
Experiences of International Development Agencies. London: DFID.
Marsden, D. 2003. Rights, culture and contested modernities. Keynote paper for INTRAC, 5th
International Evaluation Conference, Netherlands.
Moore, M.H., L.D. Brown, and J.P. Honan. 2001. Towards a public value framework for account-
ability and performance management for international Non-Governmental Organizations. Paper
presented at the Hauser Center/ Keio University Workshop on ‘Accountability for International
Nongovernmental Organizations’, Nov, Harvard University, Harvard.
Radelet, S. 2006. A primer on foreign aid, Centre for Global Development Working Paper 92,
Centre for Global Development, Washington DC.
Roche, C. and L. Kelly. 2005. Evaluating the performance of development agencies. In Evaluating
Development Effectiveness, World Bank series on Evaluation and Development, eds. G. Pitman,
O. Feistein and G. Ingram , vol. 7. Edison, NJ: Transaction Publishers.
Schön, D.A. 1983. The Reflective Practitioner. Aldershot: Avebury.
Scott, A. 2004. Assessing Multilateral Effectiveness. London: International Division, Department
of International Development.
Smillie, I. and J. Hailey. 2001. Managing for Change: Leadership, Strategy and Management in
Asian NGOs. London: Earthscan.
Sorgenfrei, M. and R. Wrigley. 2005. Building Analytical and Adaptive Capacities for
Organizational Effectiveness. Praxis Paper No.7. Oxford: INTRAC.
Stewart, F. 2003. Evaluating evaluation in a world of multiple goals, interests and models. paper
presented at the fifth Biennial World Bank Conference on Evaluation and Development,
‘Evaluating Development Effectiveness: Challenge and Ways Forward’ July, Washington, DC.
UNDP. 2003. Development Effectiveness Report. New York: UNDP.
Wenar, L. 2006. Accountability in international development aid. Ethics and International Affairs
20(1): 1–23.
Afterword
In the final session at the workshop participants discussed some of the issues that
they felt had consistently stood out during the course of the previous two days, and
focused in particular on what ideas they felt most challenge INGOs’ organisation
and practice. Some of these issues corresponded quite closely to the questions iden-
tified before the workshop and which are summarised in the introduction. Others,
however, emerged during the course of the workshop. We have grouped issues from
both these categories into five thematic areas:
• Approaches to ethics
• Expectations, roles and obligations of different actors and stakeholders
• Collective challenges and moral divisions of labour
• Accountability, knowing and communicating what works and what does not
• Purpose, direction and ethical challenges for INGOs
We will finish by summarising some thoughts about these themes. We should,

emphasise that these ideas are not put forward as conclusions reached by the work-
shop participants—it was never our intention to try to reach such conclusions. They
are, rather, a summary of some thoughts raised in the final session of the workshop,
together with some reflections by the editors of this book.
1 Approaches to Ethics
One fundamental question that naturally came up at the workshop concerns what
ethical behaviour fundamentally consists of: is it a matter of achieving certain
results, such as maximising happiness or minimising suffering; or of sticking to
certain principles; or of acting from certain character traits, such as generosity or
honesty or courage; or some combination of these things? One traditional role of
philosophers consists in articulating and evaluating moral theories, some of which
correspond roughly to the answers just given. Thus consequentialist, teleological,
or goal-based theories put the emphasis on achieving certain results, deontological
or duty-based theories emphasise acting on certain principles, and virtue ethics puts
Ethics and Applied Philosophy 23, DOI 10.1007/978-90-481-8592-4,

218 Afterword
the focus on certain character traits. It would be interesting to see whether INGOs
grappling with the kinds of ethical questions discussed in this book would find it
useful to engage with these concepts and theories, and with philosophers who are
familiar with them.
There was also some discussion at the workshop of a related distinction between
two kinds of reasons why something might be worth doing. Consider, for example,
the claim that INGOs ought to become more accountable. Why might one think
that would be a good idea? One kind of reason why one might think this is because
one believes that certain positive consequences would be likely to result. If INGOs
become more accountable, one might think, they would be more likely to fulfill their
responsibilities diligently, and contributors would be reassured that their money is
being spent wisely. A different kind of reason one might think that INGOs should
become more accountable is because one believes that accountability is important
in itself (intrinsically valuable, rather than merely instrumentally valuable, as the
point is sometimes put), regardless of any consequences it might have. One might
think, for example, that if INGOs appeal for and take money from people they have
a duty to account to those people for how they have spent that money, whether or
not anyone benefits from their fulfilling that duty.
Answers to these questions might make a big difference to how INGOs should
respond to calls to become more accountable, especially if the consequences of
doing so wouldn’t be as positive as was suggested in the last paragraph. Suppose
that far from helping to ensure that they fulfill their responsibilities diligently, and
reassuring contributors, becoming more accountable would, in fact, distract INGOs
from fulfilling their responsibilities and/or put off contributors. If so, and if account-
ability is not intrinsically valuable, then presumably it would not be appropriate
for INGOs to become more accountable after all. If accountability is intrinsically
valuable though, it might still be the case that INGOs should become more account-
able, even if the consequences of doing so would be largely negative. For the
duty to be accountable might take priority over the wish to avoid those negative
consequences.
The same kind of point applies to many other similar claims often made about
INGOs: that they should become more open, honest, or transparent; that they should
practice what they preach, in the sense that their organisational procedures should
reflect their values; and so on. In each case one can ask whether such changes are
supposed to be worth pursuing in virtue of their consequences, or because they
are valuable in themselves, or both. (These two answers are not of course mutu-
ally exclusive—one might think that INGOs should become more accountable, for
example, both because one thinks that doing so would have positive consequences,
and because one thinks that accountability is important in itself.)
Another standard philosophical distinction that came up at the workshop is
between negative and positive duties. Negative duties are duties not to do certain
things, such as harming others. Positive duties are duties to do certain things, such
as helping others. Traditionally, it has been thought that negative duties are more
stringent than positive duties, other things being equal. If this claim is correct,1
it would be interesting to consider what implications it might have for INGOs.
Afterword 219
Might it imply that they should not do work that would do any harm, for example,
even if that work would also do a great deal of good? What if it would not be the
INGO that actually does the harm, but some other agent which exploits the INGOs
actions? How does one factor risks of harm into the equation? Furthermore, how
should one handle disagreements—between INGOs and those they seek to benefit,
for example—concerning what is harmful or not?
Another way in which this distinction might be relevant that came up at the
workshop concerns the type of obligations that those of us who live in developed
countries might owe to the global poor. Most commonly, those who believe that we
have such obligations at all think of them as positive obligations (whether of justice
or of beneficence), obligations to help others in need or to help them claim their
rights. Some, though, have argued that those of us who live in developed countries
are harming the global poor, by imposing an unjust global order upon them. And
if that is correct, we also have negative obligations to the global poor, obligations
not to harm them in such ways. Such claims are controversial.2 If they are correct,
though, this would potentially have significant implications for INGOs (see chapter
‘Aid Agencies, States, and Collective Harm’ by Das, this volume).
As well as making such distinctions, articulating different approaches to ethics,
and discussing their merits, philosophers also seek to identify forms of reasoning
that are tempting, but fallacious. One example of this that came up in discussion
at the workshop is Peter Unger’s concept of ‘futility thinking’. Unger (1996, esp.
pp. 75–77) describes it like this. Take a small group of people in need, group A,
who are part of a much, much larger group who are also in need, group B. If one
focuses on group A in isolation, disregarding the fact that they are part of group
B, Unger suggests, it is likely to be clear enough that there is a very strong moral
reason to help them. If one focuses on the members of group A primarily as a part
of group B, though, the sense that there is a strong moral reason to help them may
diminish. For even if one were to succeed in helping those in group A, doing so
would have little or no discernible impact on group B as a whole, and therefore may
seem ‘futile’—a mere ‘drop in the ocean’.
Unger was concerned that futility thinking might affect those potential contrib-
utors to INGOs who are discouraged by the sheer scale of world poverty. ‘Even if
my contribution helps a few people’, one might think, ‘there will still be millions
of other people in need, and so my contribution wouldn’t make any real difference’.
It is clear, at least on reflection, that such thinking is mistaken. If that contribution
would improve the lives of a few people, then it would indeed make an important
difference, even if that is not the type of difference that would show up in statistics
about world poverty.
Some of the philosophers were concerned that something like this fallacy might
also affect thinking about aid and development in other ways, though. It is common
now for those interested in aid and development to focus on large and ambitious
aims such as social and political transformation or the Millennium Development
Goals. There are of course good reasons for doing so, but there is also the danger
that it might lead one to regard measures that ‘merely’ improve conditions for a few
people for a while as not worth doing, if they have no discernible effect on those
220 Afterword
larger aims. And if so, that would again be a mistake. If a certain activity would
improve the conditions of a few people for a while, then there may be a very strong
moral reason to do it, especially if the conditions those people face are very harsh.
Of course, there might be even stronger reasons to do something else, such as some
activity that aims at social and political transformation—but only if the chances of
success of that activity are high enough. If the chances of success are very low, it
might be morally better ‘merely’ to focus on more immediate needs, rather than
seeking to ‘change the world’.
2 Expectations, Roles and Obligations of Different Actors

and Stakeholders
Another issue that stimulated a lot of discussion at the workshop concerns the expec-
tations, roles and obligations of different actors and stakeholders. One aspect of this
concerns what kind of voice different stakeholders should have in determining what
INGOs do: where they work, what kinds of activities they perform, the details of
their programs and projects, who judges their impact, and so on. The answer to this
question will of course be constrained to some degree by INGOs’ mission and policy
statements, but typically such statements leave room for a good deal of discretion.
And in any case, the same kind of question arises in relation to those statements
themselves: who should decide what goes into them?
One view might be that it is entirely up to each individual INGO to determine
its own priorities and practices. INGOs are voluntary associations, after all, and are
focused on doing good of one sort or another. Such a view can be challenged, how-
ever, from at least two different perspectives. First, though donations to INGOs are
typically regarded as not being obligatory, many philosophers have argued that those
of us who live in developed countries are in fact morally obligated to do something
to assist those living in extreme poverty in developing countries (Singer, 1972, 1993,
Chapter 10, and 2009, Unger 1996, Pogge 2002, Cullity 2004). Given that we can’t
discharge those obligations by ourselves as individuals (or at least that we can’t do
so efficiently), we turn to INGOs to discharge those obligations on our behalf. From
this perspective, then, INGOs would be seen as executors of the obligations of their
contributors and supporters.
If this sort of view is correct, both contributors to INGOs and those INGOs
themselves would be constrained by the content of the obligations in question.
Contributors would be obligated to give to INGOs that they have reason to believe
will achieve the aims set by those obligations, rather than being free to support
whatever cause they choose (or none at all if they prefer). And INGOs would be
required to use those funds as efficiently as possible to realise those aims, rather
than being at liberty to determine their own priorities and practices as they see fit.
If, for example, the aim were that of improving the conditions faced by those who
are worst off, globally speaking, then INGOs would be obligated to use whatever
funds in whatever way would be likely to be most effective in realising that goal.3
Afterword 221
A different reason why one might think that it is not entirely up to each individual
INGO to determine its own priorities and practices stems from the perspective of
those that aid seeks to benefit. It is of course commonplace to argue that men and
women in this position should participate in determining how any aid intended to
benefit them is used, how that aid is evaluated, and so on. One might think that
they should have a greater role than this, however, especially if one thinks that aid
is owed to them as a matter of justice. If I owe you $100, then normally at least I
should simply pay you back if I can, and leave it to you to decide what to do with
the money. I shouldn’t offer to pay you back only on condition that you agree to
use the money in some way that I consider appropriate, or suggest that we sit round
together to decide how you will use it. If, then, there is some relatively robust sense
in which we in the North can be said to owe aid to the global poor, this might suggest
that INGOs should simply hand over any resources they have at their disposal, or at
most just offer advice and support to help those they seek to benefit achieve what
they want to achieve with those resources.
There are of course several real-world complications, though. In some cases,
there may be persistent disagreement among the ‘recipients’4 concerning how the
money should be spent. If so, the INGO wouldn’t simply be able to do what ‘they’
say. In other cases, there might be broad agreement amongst them, but the INGO
may believe that they are making a mistake. In some of these cases, the disagreement
may concern only the means to ends that are accepted by all. Even in such cases, the
INGO may have to tread very carefully. They may have good reasons to believe that
their experience and knowledge puts them in a better position to determine the best
means than the ‘recipients’. There may be a constant temptation to over-estimate
the extent to which this is so, however, and so this is an area which requires great
sensitivity.
The harder cases, though, are those in which the INGO disagrees with (at least
some of) the community they are working with about the ends that should be aimed
at. What if the INGO thinks that any resources should be used to help men and
women gain secure access to certain basic rights, for example, but they would prefer
simply to divide any available resources amongst them to use as they choose? Who
gets to say what happens then? Does it make a difference if the INGO takes itself
to be acting on an obligation to help secure people’s rights? What if the ‘recipients’
want to use those resources in ways the INGO considers positively harmful, because
of environmental reasons, for example?
There seemed to be general agreement at the workshop that such issues concern-
ing how to balance and reconcile the expectations, roles and obligations of different
actors and stakeholders are among the most difficult facing INGOs, and that they
would therefore provide a very important area for further research. It is perhaps also
worth noting here that the recognition of the need to balance different accountabili-
ties has led to the rise of ‘Collaborative Initiatives or Multi-sector partnerships’ that
according to some ‘are emerging as a twenty-first century institutional innovation
to address development’ (Litovsky and MacGillivray 2007, p. 21). These partner-
ships may help to develop new forms of mutual accountability, and may therefore
offer hope for the future. There is also the danger, however, that they ‘mirror the
222 Afterword
relative bargaining power of donors and recipients’ unless participants are sensi-
tive to unequal power relations and take specific steps to address them (ibid.). This
challenge, then, may be a particularly important focus for further work.
Whilst these issues are not new to INGOs, the workshop indicated how ethical
thinking is useful in providing different perspectives on dealing with these tensions
and dilemmas.
3 Collective Challenges and Moral Divisions of Labour
A number of papers presented at the workshop raised the issue of collective prob-
lems and responsibilities faced by INGOs. This led to a number of discussions
about different forms of collaboration, as well as what a moral division of labour
might look like in different circumstances (recognising that collaboration doesn’t
necessarily mean all doing the same thing).
In the first instance, it was argued that INGOs need to be clear about their aims.
As noted in Cullity’s contribution to this volume, by acting in certain ways an INGO
on its own might manage to do some good itself, but several INGOs working in
such ways might do less good overall than they could do by acting in some other
way, or even do harm. For example, a certain competitive form of fundraising might
increase funds for an individual INGO, which would enable that INGO to do more
good. Doing so might however reinforce a view amongst the general public that
INGOs are wasting money in competing with one another for dollars, and if so
the net result might be that less funds are raised in total, and so less good is done
overall. In cases like these, the action that would lead to INGOs individually doing
most good (practicing the form of fundraising in question) would not lead to most
collective good being done overall. In view of such possibilities, each INGO needs
to think about whether its most fundamental aim is that of doing as much good as it
can as an agency, or acting in the way that leads to most good being done overall.5
This in turn will shape the priority given to processes of collaboration and collective
action.
A number of different collective challenges were debated. These included:
• Cases where individual actions may be ‘good’, but collective action could pro-
duce even greater benefits. One example might be for agencies to pool funding to
conduct collective (rather than individual) base-line research or monitoring—for
example of mortality and morbidity rates in a refugee camp—which would allow
for a consistent assessment of the combined impact of all the agencies involved.
• Cases where individual actions may in themselves be ‘good’ but the combina-
tion of those individual actions ’bad’. One example might be the situation in the
Great Lakes region of Africa in the 1990s, where some argue that the collective
presence and action of INGOs and others allowed the perpetrators of genocide to
regroup and manipulate relief supplies and other refugees.
• Cases where if everyone acts purely on an individual basis, with no reference to
what consequences might result if all or most INGOs act similarly, all are worse
Afterword 223
off. For example, if there is no consistency across agencies about willingness to

pay bribes, the result may be that all agencies face the demand to pay bribes.
In connection with these circumstances philosophers also pointed out the dis-
tinction between cooperation for prudence (in order to lessen collective risks) and
cooperation for altruism (in order to achieve a greater good). These distinctions are
important in that they derive from different motivations and are liable to break down
for different reasons, and thus need to be encouraged in different ways.
From the discussion a number of useful questions emerged which might help in
addressing these challenges, namely:
• What collective are we a part of? Is it just the INGO community or is it other
actors? Are we part of a wider collective that also is causing harm? For example,
as an Australian agency with many staff who are Australian citizens, are we part
of an Australian community that has elected a government whose trade, aid or
foreign policies might be detrimental to particular groups overseas? And if so, do
we have different obligations as a result?
• What would be a more effective moral division of labour or collaboration?
For example, if speaking out on human rights abuses was critical to afford-
ing protection to particular groups in Darfur, and the circumstances made this
impossible for an agency providing relief on the ground, would collaborating
with and passing on information to a specialist Human Rights organisation, such
as Amnesty International, be even more important than collaborating with other
relief agencies in the refugee camps on the ground?
• What might individual agencies do to promote collaboration and collective action
where that is not yet happening? In certain circumstances INGOs might have
the opportunity to play a leadership role in creating opportunities for collective
action, which would then make it more likely that others can see the bene-
fits of cooperation. Too often the incentives of individual organisations and the
difficulties and challenges of cooperation make the non-collaboration option
easier.
• Under what circumstances do the costs of collective action outweigh the ben-
efits? This was not really a question addressed at the workshop. However, the
experience of INGOs suggests that there are times when the inordinately complex
processes of inter-agency collaboration can result in time-consuming, inconclu-
sive processes that stymie creativity, delay responses and lead nowhere. Having
clear criteria and decision-making processes for when the costs of collective
action outweigh the benefits, then, would be very useful.
4 Evaluation and Accountability

The importance of being able to assess and communicate the impact of INGO
work was a recurrent theme in several of the papers and in the ensuing discus-
sion. Several of the INGO participants, however, stressed that the pressure to
224 Afterword
demonstrate impact (particularly to institutional donors, or within their own hier-

archies, often using inappropriate tools and methods based on linear cause/effect
thinking) is having a detrimental effect. It is doing so, in particular, because it
increases bureaucracy, reduces flexibility and can alienate partners. A number of
the philosophers felt that clear measures to assess and communicate achievements
were important, however, in order (a) for INGO themselves to arrive at effective and
ethical decisions about how to use their resources, (b) to allow other stakeholders—
especially private supporters—to make informed decisions about whether or not to
support INGOs in general, or a particular INGO, and (c) to be transparent about their
performance.
In the debate there was some recognition from all parties that the question of
assessing impact and performance is complex and needs a variety of approaches,
tools and methods. It was also recognised that there may be a number of proxies
of impact that may provide some reassurance that impact is liable to be achieved
as these proxies correlate closely with effective programs. In this sense there may
be methods of improving accountability without necessarily being able to demon-
strate impact. These might include adherence to certain codes or standards derived
from research and evaluative work, such as those developed by the Humanitarian
Accountability Partnership (HAP) or by the Australian Council for International
Development (ACFID) through its work on quality and effectiveness (see chapter
‘The Seeming Simplicity of Measurement’ by Roche, this volume). Similarly, a
number of participants felt that the most important challenge in this area was to
improve the accountability of INGOs to those they ultimately seek to benefit, and to
incorporate their perspectives and views much more centrally into consideration of
impact. (Indeed, the HAP standards are designed precisely to assess the degree to
which organisations are really doing this.)
However, there remained a certain reluctance amongst INGO participants to fully
commit themselves to agreeing that clear metrics, adherence to codes or standards
(particularly related to processes), and greater accountability to communities they
seek to benefit could resolve the Epistemic Problem raised in Horton’s paper (that
is, the difficulty of arriving at an estimate about the effects of the work aid agen-
cies do that one has at least some good reason to believe accurate). There seems
to be a number of possible reasons for this. Firstly, as Dennis Rondinelli suggested
(1993, p. vii) some time ago, there are very good reasons why practitioners are
wary about rationalistic techniques of planning, management and evaluation. They
often see these as constraining their ability to practice the ‘flexibility, experimenta-
tion and social learning that are crucial to successfully implementing complex and
uncertain development activities’. Secondly, many INGO staff fear that the nature of
the power relations which are inherent in the donor/INGO/community nexus means
that the voices of the least powerful in this relationship will always remain marginal.
There is also a tendency to believe that those with the most power will turn off the
funding tap if poor performance or negative results are found. These beliefs can
lead to INGOs either being ‘economical with the truth’ about mistakes, or simply
working with those groups, or on those activities, where the risks of failure are less.
This is liable to mean increasingly working with the less poor or vulnerable and
Afterword 225
in more secure environments. And thirdly, there is the view that the plethora of
codes, standards and standardised planning and evaluation methods are (a) increas-
ing bureaucracy and reducing the time for the ‘real work’ that needs to be done;
(b) reducing the role for staff’s professional judgement and acting as a disincentive
for learning and being honest about mistakes; (c) increasingly leading to formulaic
responses independent of context; and (d) undermining the development of effec-
tive relationships between INGOs, their partners and communities which are key
to effectiveness (chapter ‘Whose Impact, and Is It All About Impact?’ by Isbister
and chapter ‘Ethical Behaviour in Non-government Organisations’ by Kelly, this
volume).
For some of the philosophers this position was problematic for a number of rea-
sons. First, it seems to mean that decision-making and judgements about who to
work with and where are liable to be haphazard. One result of this might be that
questions concerning whether INGOs could have been more successful somewhere
else are simply not asked. Second, if there is a level of inter-subjective agreement
amongst INGOs about some of the processes and proxies that lead to effective
work—e.g. the ACFID quality and effectiveness work—then why not be more sys-
tematic about ensuring that these lessons are properly learnt and implemented?
And third, if INGOs don’t provide a more convincing, honest and independent
assessment of their performance, then private individuals have no means of making
well-informed judgements about whether to support INGOs. This is clearly prob-
lematic in itself, but may also lead in turn to a number of further ethical problems,
notably that people may be being asked to contribute to activities that, unknown to
them, do people harm.
Despite this difference of opinion there were a number of areas were most
participants agreed progress could and should be made. These include:
• Being clearer about negligence standards—i.e. those things that INGOs would
definitely seek to avoid doing and for which there could be mechanisms for
redress and complaint. Such an approach is less limiting than codes or standards
that try and prescribe what INGOs must or should do, and thus it potentially
allows for more flexibility and adaptation to local context.
• Being clearer about burdens and standards of proof of effectiveness, and the
admissibility of evidence. Part of the problem with debates about evaluation and
impact assessment is that different stakeholders have different criteria for evalu-
ating the methodology and findings. Being clearer about this would help reduce
this problem as well as potentially help in developing a more appropriate set
of criteria for judging effectiveness than those found in criminal courts (beyond
reasonable doubt) or in scientific enquiry (statistical significance).6
• Developing more transparent processes for the dissemination of evaluation
findings—for example by developing clearer policies for INGOs about what they
do or do not put on their websites.
• Exploring new institutions or mechanisms for feedback and accountability, such
as Ethics Committees or Councils related to individual INGOs or INGO umbrella
groups, Aid Ombudsmen, and complaints mechanisms.
226 Afterword
5 Values, Purposes, and Institutional Issues for INGOs

Some of the discussion that provided the greatest interest and challenge to INGO
participants at the workshop concerned INGOs as institutions. This is perhaps
unsurprising in that much of ACFID’s quality and effectiveness research indi-
cates that institutional values and culture play a key role in determining how, in
practice, INGOs field operations are implemented (chapter ‘Ethical Behaviour in
Non-government Organisations’ by Kelly, this volume).
There was an important debate stimulated by Lenneberg’s chapter about how
to manage the ethics of cross-cultural relationships and dealing with cultural dif-
ferences, for example. This discussion raised the ‘big’ issue concerning the degree
to which INGOs do, or should, respect the culture of those they seek to benefit—
sometimes seen as a core value of these organisations. Indeed, this is explicit in
some of the codes of conduct they have signed up to, such as the Red Cross Code of
Conduct.7
Some of the philosophers suggested that one could usefully distinguish between
respecting a person and respecting their beliefs, and also between respecting some-
one’s beliefs and agreeing with those beliefs. One can respect a person but want
to change their beliefs, for example, even if it is often hard to separate the two
things. If that is right, then trying to support change in society may not be incon-
sistent with proper forms of respect. Furthermore, some cultural practices that we
don’t share nevertheless have a justification that we can accept while others do not.
Others argued that one can also overstate the case for respect of local culture or
customs, for (a) there are sometimes situations where communities or groups have
been traumatised through exploitation, colonisation and so on, which can lead to
dysfunctional behaviour that is clearly causing harm to some, even if it is defended
in the name of culture; (b) culture is not completely monolithic or static—certain
groups and interests will be seeking to change practices and can be supported with-
out that support necessarily taking the form of promoting ‘alien’ values; and (c)
there are also situations where the powerful in a given group or community will
use the banner of culture to maintain their privilege and position at the expense of
others.
As such, the question of whose cultural values should be respected is usually one
of balance, judgement and context, with either of the extremes—total deference, on
the one hand, or total rejection of any of ‘their’ values that conflict with ‘ours’, on the
other—being inappropriate. In the end it was suggested that this matter is inevitably
framed for INGOs by their own values, and that therefore these need to be clear.
Some of the participants felt that Human Rights approaches provide a useful means
of helping to make judgements in this area, whereas others found them insufficient.
This debate is linked to another recurrent theme across the discussions which
relates to INGOs’ adherence to their values, whether they ‘walk the talk’ in terms
of their rhetoric, how independent they actually are, and the degree to which their
growth and ‘corporatisation’ is undermining their purpose and direction. This was
interestingly as much if not more of a concern for INGO participants as it was for
the philosophers present.
Afterword 227
Some of the suggestions that emerged from this discussion were:
• To develop more effective mutual learning across the sector on ethical questions,
and in particular to explore opportunities for professional development of staff
that includes ethical components.
• To be careful to distinguish between instrumental values and intrinsic values. At
the workshop, innovation, accountability and transparency were all mentioned as
key values, for example. As noted in Section 1 above, though, some philosophers
asked whether such values were largely instrumental. If that is right, care needs
to be taken in promoting these values above those that might be more intrinsic
such as the imperative to do no harm.
• To be much clearer about political legitimacy, agendas and independence. It
was suggested (chapter ‘The Ethics of Taking Sides’ by, Ellis, this volume) that
a codified recognition of INGOs’ political role—including their dealings with,
and independence from, donors—would be a powerful statement of what con-
stitutes acceptable behaviour in relation to these issues, and could contribute
to an empowering vision of the broader social change objectives of the INGO
movement.
• For INGOs to experiment with more radical changes to their form and the way
they work, changes that are consistent with the rhetoric of partnership, organi-
sational learning and building effective relationships. It was argued, by Kelly in
particular, that we know what effective INGO work looks like, and that achieving
it is difficult but doable, but that it needs real commitment and leadership within
senior levels of INGOs to make it happen. And for some INGOs this might mean
becoming ‘brokers of social change’ whereby they see themselves more as cata-
lysts for the exchange of ideas, experience and strategies between activists than
as purveyors of development projects, and in so doing help to enhance social
movements and transnational ‘monitory democracy’ (Keane 2009).
6 Final Thoughts
The bringing together of a number of practitioners with academics was perhaps a
somewhat risky exercise. Participants were generally very positive, however, about
the experience. Why was this?
Firstly, the process of the preparation for the workshop—in which nearly all par-
ticipants prepared papers and these were circulated well in advance so that everyone
had the chance to read them beforehand—was important. This meant that time at
the workshop was not wasted by people presenting or reading out their papers, and
that INGO participants had to reflect on their experience, and that of others, before
the workshop rather than coming to it ‘cold’ (which is often the case).
Secondly, INGO staff are generally concerned about the degree to which they
are behaving ethically, not least because they know that much of what they do raises
difficult ethical questions. However, there is rarely the time, nor the expertise, within
228 Afterword
INGOs, for discussions to be as useful and as profound as INGO staff wish. In this
sense having the space and supportive expertise in the room was key.
Thirdly, the exchange between participants was respectful and not overly aca-
demic. The philosophers were sensitive to the pragmatic and practical concerns of
INGO participants, and prepared to patiently explain things that were not under-
stood, as well as introduce ethical or moral concepts when appropriate. The INGO
participants, for their part, were very willing to engage in philosophical argument,
were generally not defensive about their work (perhaps in part because many of
them knew each other relatively well prior to the workshop and they tended to keep
each other ‘honest’ in the discussions), and were mostly open to fresh thinking and
ideas, even when this challenged some firmly held assumptions.
All of which suggests that it may be worthwhile organising similar exchanges or
other forms of collaboration between philosophers and INGOs in the future. Ideas
that emerged from the workshop, other than this publication, included running an
ethics workshop at an INGO conference on Measuring Development Effectiveness
in Melbourne in 2007,8 which was indeed held, and which proved very popular
with INGO staff, and future plans include involving philosophers on forthcoming
evaluations of INGO programs. It would seem that there is a hunger for this sort of
dialogue and we hope this publication can contribute in some small way to it.
Notes
1. Not everyone accepts it. For consequentialists, for example, all that matters are the con-
sequences, not whether those consequences result from doing something or failing to do
something.
2. For discussion of them, see Pogge (2002, 2005), and chapter ‘Ethical Obligations to the Poor
in a World of Nation States’ by Ronalds, this volume.
3. For a strong statement of this kind of position, see Pogge (2007, pp. 218–222).
4. We have tried to avoid using the word ‘beneficiary’ and wherever possible have tried to use a
term that does not label people as objects of charity or development. Where this has proved
difficult or misleading we have used term ‘recipient’ in quotes to denote our unease with the
term.
5. For more discussion of such cases, see Pogge (2007: 241–245).
6. For discussion of such issues in a different but related context, see Barry (2005).
7. The 5th of the 10 elements of the Red Cross Code of Conduct states, ‘We shall respect culture
and custom’, and more specifically, ‘We will endeavour to respect the culture, structures and
customs of the communities and countries we are working in.’ See http://www.icrc.org/web/
eng/siteeng0.nsf/htmlall/57JMNB.
8. See http://www.worldvision.com.au/learn/conferences/me/index.asp.
References
Barry, C. 2005. Applying the contribution principle. Metaphilosophy 36(1/2): 210–227.
Bell, D.A. and J.-M. Coicard. 2007. Ethics in Action: The Ethical Challenges of International
Human Rights Nongovernmental Organizations. Cambridge: Cambridge University Press.
Cullity, G. 2004. The Demands of Morality. Oxford: Oxford University Press.
Afterword 229
Keane, J. 2009. The Life and Death of Democracy. New York: Simon and Schuster.
Litovsky, A. and A. MacGillivray. 2007. Development as accountability. AccountAbility 21.
Pogge, T. 2002. World Poverty and Human Rights. Cambridge: Polity Press.
Pogge, T. 2005. Real world justice. The Journal of Ethics 9(1–2): 29–53.
Pogge, T. 2007. Moral Priorities for International Human Rights NGOs. In Bell and Coicard.
Rondinelli, D. 1993. Projects as Policy Experiments: an Adaptive Approach to Development
Administration. London: Routledge.
Singer, P. 1972. Famine, affluence, and morality. Philosophy and Public Affairs 1: pp. 229–243.
Singer, P. 1993. Practical Ethics, 2nd edn. Cambridge: Cambridge University Press.
Singer, P. 2009. The Life you can Save. New Haven, CT: Yale University Press.
Unger, P. 1996. Living High and Letting Die. Oxford: Oxford University Press.
Index
Note: The letter ‘t’ following the locators refer to tables
A three levels, 151

Accountability rights language, 42
ALPS, 152 standards, solutions to improve, 106–107
ethical questions/principles/evidence, see state entities, NGOs, 68–69
Ethics transparency/learning, steps promoting,
HAP, 106, 127, 129, 224 141–142
initiatives and quality, 127–130 See also International Non-Government
ACFID in Australia, 130 Organisations (INGOs)
ALNAP, 127 Accountability, Learning and Planning System
Groupe URD in France, 127 (ALPS), 152
HAP 2007 Standard, 129 ACFID, see Australian Committee for
Humanitarian Accountability International Development
Partnership/Project, 128 (ACFID)
Humanitarian Ombudsman Project, 128 ActionAid, 152
international initiatives, 129 Actions, single to collective, 164
Slim’s framework for ethical assessment Active Learning Network for Accountability
of humanitarian work, 125t–126t and Performance in Humanitarian
Sphere Humanitarian Charter and Action (ALNAP), 106, 121,
Minimum Standards, 127 127–130, 181
TEC evaluation, 128 ‘Adaptive systems’ solution, 136
and learning Advocacy, 3, 11, 14, 54, 59
advantages and disadvantages of agendas, Red Cross Code, 122
different solutions, 138t Code of Conduct, 81–82
aggregation problem, 135 consequences of taking sides at home,
attribution problem, 134–135 72–76
measurement problem, 134 See also Taking sides, consequences of
ownership problem, 135 ethical questions, see Ethics
responses, 136–138 evidence base for development and,
See also Learning and accountability, 130–133
INGOs See also Ethics/principles/evidence,
limited accountability of BRAC, 57 INGOs
mechanisms, 151–152 extension of Red Cross Code of principles
ActionAid, 152 to development and, 123t–124t
ALPS, 152 Kouchner’s advocacy of military
Humanitarian Accountability Project intervention, 159
(2007), 152 NGOs in Australia, attack on, 69–72
Listening Project, 152 political, 11, 73, 184, 186, 188–189
231
232 Index
role of NGOs in donor countries, 76–79 priorities among rights and rights-based
Slim’s ethical framework, relevance to duties, 49–52
development and, 124–127 division of labour between organisa-
See also Slim, Hugo tions, 51
suggestions by Riddell on NGOs and, 132 extent and nature of need, 50–51
Aid agencies, see International Non- two factors, 49
Government Organisations Basic Rights (Shue, Henry), 44, 50, 58
(INGOs) ‘Beyond-the-project’ work by aid agencies, 92
ALNAP, see Active Learning Network for ‘Black Hawk Down’ incident, 22
Accountability and Performance in Bosnian Serb targets in 1995, 22
Humanitarian Action (ALNAP) BRAC, see Bangladesh Rural Advancement
ALPS, see Accountability, Learning and Committee (BRAC)
Planning System (ALPS) Briefing Paper on Iraq by Oxfam International,
Altruism, 163–164, 176, 189, 223 186–187
Altruistic collective action, 175, 181 Bush administration on conduct of US foreign
American public support for humanitarian policy, 177–178
intervention, 22
Amnesty International (AI), 39, 47, 179, 223 C
Appreciative Inquiry method, 140 Capacity-building initiatives, 92
AusAID (Australian Government’s aid Caritas Australia, 208–210
agency), 27–28, 72–74, 135, 209 CBOs, see Community based organisations
Australia (CBOs)
attack on advocacy NGOs, 69–72 Ceteris paribus duty, 44
See also Advocacy Child Rights Convention (CRC), 198
performance based incentives for teachers, ‘Clinical economics’ (Sachs, Jeffrey), 24
149 Code of Conduct for International Red Cross
policy of integrating indigenous and Red Crescent Movement
Australians, 150 (1994), 158
-Rwanda Network, 211–213 “Coercive regimes,” 20
World Vision Australia (2006), 25 Collaborative for Development Action, 200
Australian Committee for International Collective action, 162–167, 168, 170, 222–223
Development (ACFID), 181 problems, 8, 10, 181–182
Community based organisations (CBOs), 91,
B 103, 208–209
Bangladesh Rural Advancement Committee Compromised humanitarianism, 157–172, 176
(BRAC), 57 See also Humanitarianism
‘Basic Capabilities,’ 196–197 Cooperation
Basic rights and correlative duties AusAID NGO Cooperation Program, 73
development INGOs, duties and moral cooperative concern, 162–164
priorities ‘cooperative-spirited’ thinking/people,
cost effectiveness, 54 162–163
development assistance and rights ‘each-we dilemmas’ (Parfit, Derek), 163
protection, 56–59 examples of nesting of collective
enjoyment of rights, lack of action, 165
development, 55–56 ‘prisoner’s dilemma,’ 163
naming/shaming methodology, single to collective actions, 164–165
HRW, 53 See also Humanitarianism
protecting against international threats, effective inter-agency cooperation,
54–55 166–167
protecting against violation of humanitarian, 165–167
economic/social rights, 53–54 exploitation of work (Rieff, David),
human rights INGOs, addressing 166–167
economic/civil/political rights, MSF’s withdrawal from Rwandan
47–49 camps, 165–166
Index 233
imperfect cooperation, 167–169 female circumcision (FGM), against,

differences of opinion, 169 197–198
sub-groups, 168 ‘four kinds of ethical dilemmas, 195
inter-agency, 166–167, 170 further dialogue and debate among
‘morality of concern,’ 160 practitioners, 202
morality of cooperation, 10, 161, 165–166, gay and lesbian rights, 195
168–170 Habitat for Humanity, 195
‘morality of respect,’ 160 ‘hidden’ schools, 200
relevance with moral character of humanitarian ideals and response (Slim,
humanitarian action, 161 Hugo), 195–196
‘Cooperative-spirited’ thinking/people, Human Rights as feminists in
161–164 Western/developing nations, 199
Corporate performance, measuring, 148 Kantian notion of the respect, 194
Correlative duties, threefold structure, see Listening Project, 200
Duties ‘respect,’ meaning, 194
CRC, see Child Rights Convention (CRC) social justice and women’s rights
(Nussbaum, Martha), 196
D World Vision Area Development
Danish Institute for Human Rights, 48 Program in Jaipur, India, 201
Development of NGOs (Slim, Hugo), 183–184
ACFID, see Australian Committee for Disaster Relief, NGOs in, 158
International Development (ACFID) ‘Do No Harm Handbook,’ 183–184
and advocacy work, Slim’s ethical ‘Do No Harm’ project, 19–20, 121, 178, 227
framework, 124–127 Donors and communities, impact of programs,
aid, 90 140, 150, 154
assistance and rights protection, 56–59 Duties
DAC study, 91, 104 human rights and correlative duties-
as expansion of freedom (Sen, Amartya), conceptual issues, 42–45
196 duties allocation to particular agents, 44
INGOs, duties and moral priorities rights to aid (positive rights), 43
arms trade, 54 rights to forbearance (negative rights),
development assistance and rights 43
protection, 56–59 social or institutional dimension of
lack of development, 55–56 human rights, 44
protecting against international threats, threefold structure (type I, II, III), 43
54–55 rights-based duties, 49–52
protecting against violation of economic division of labour between organisa-
and social rights, 53–54 tions, 51
in terms of rights, 41–42 extent and nature of need, 50–51
See also Human Rights two factors, 49
Development Assistance Committee (DAC) structure of moral right and complexity of,
study, 91, 104 45–47
Dilemmas ‘indirection’ thesis (Pogge, Thomas),
humanitarianism, cooperative concern 47
‘each-we dilemmas’ (Parfit, Derek), ‘successive waves of duties’ (Waldron,
163–165 Jeremy), 47
‘Prisoner’s Dilemma,’ 163, 175 See also Basic rights and correlative duties;
INGO development practitioners, 193–194 Rights
Collaborative for Development Action,
200 E
CRC, 198 ‘Each-we dilemmas’ (Parfit, Derek), 163–165
development as expansion of freedom East Timor Community Assistance Scheme, 74
(Sen, Amartya), 196 Economic, social and cultural rights (ESC), 40
234 Index
The Economist, 105 Iraq war, ethical issues, 74

Effective inter-agency cooperation, 166–167 Journal of Medical Ethics, Business Ethics
Epistemic problem Quarterly, and Environmental
aid agencies as organisations, 93–95 Ethics, 2
aid agencies, concerns about effects of lifeboat ethics (Hardin, Garrett), 25
work, 90–93 literature, ethical dilemmas, 195–197
lack of reliable data and expert guidance, See also Dilemmas
88–89 politics INGOs and, see Political
need of data or guidance, 89–90 legitimacy; Political NGOs, ethics
qualities, 93 for
taking stock, 96 principles/evidence, INGOs, see
Equal Opportunity Commission, 69, 150 International Non-Government
ESC, see Economic, social and cultural rights Organisations (INGOs)
(ESC) questions, see Ethical questions
Ethical basis and extent of obligations, 18–26 Slim’s ethical framework, 124–127,
American public support for humanitarian 125t–126t
intervention, 22 of taking sides, see Taking sides,
‘Black Hawk Down’ incident, 22 consequences of
Bosnian Serb targets in 1995, 22 ‘The Ethics of Humanitarian Disengagement,’
“coercive regimes,” 20 183
globalisation impact, 14 Ethics of taking sides
lifeboat ethics (Hardin, Garrett), 25 attack on advocacy NGOs in Australia,
Perpetual Peace (1891), 18 69–72
proposals (Shiva, Vandana), 25 consequences of taking sides at home,
RTP, 23 72–76
utilitarian scholars, Singer’s approach, ethics for political NGOs, 79–82
20–21 NGOs and political change, 67
Washington Consensus, 24 recipient/donor countries, 67
‘well ordered’/‘burdened societies,’ 23 in recipient country, 66–69
World Vision Australia (2006), 25 structure and political legitimacy of NGOs,
See also Poor in world of nation states 76–79
Ethical questions Ethics/principles/evidence, INGOs
raised by specific kinds of activity INGOs, ability to promote systemic change, 133
2–3 evidence base for development and
relationship between INGOs and advocacy work, 130–133
contributors/public in northern IFRC, 121–122
countries, 5–6 non-systematic review of work, 133
relationship between principles in evaluation of humanitarian
INGOs/‘partners’/those aim work, 122
to benefit, 4–5 quality and accountability initiatives,
selection of activities and areas of work, 127–130
3–4 requirements and means of verification for
Ethics HAP Benchmark 5, 131–132
accountability and quality initiatives, Slim’s ethical framework, 124–127,
127–130 125t–126t
basis and extent of obligations, 18–26 weak performance, 133
See also Poor in world of nation states See also International Non-Government
behaviour in NGOs, case studies, 207–213 Organisations (INGOs)
cooperative approach to morality
‘ethics of humanitarian disengagement,’ F
183 Feinstein International Centre, 199
international relations, role of, 15–17 Female circumcision (FGM), 197–198
See also Poor in world of nation states FGM, see Female circumcision (FGM)
Index 235
Forced sterilisation, 41 MSF’s withdrawal from Rwandan camps,

Fund managers, assessing, 148 165–166
Humanitarian ideals and response (Slim,
G Hugo), 195–196
Gay and lesbian rights, 195 Humanitarian imperative, 158–160
Gender analysis of Sen (Amartya), 159
discrimination in inheritance laws in de Waal, views of, 159
Kenya, 53 genocidal violence in Rwanda, 159
division of labour, 199 manipulation of humanitarian aid by
equity, 199–200 military groups, 159
Red Cross Code of principles to See also Humanitarianism
development and advocacy, Humanitarianism
123t–124t concern/respect/cooperation, 160–162
sensitivity, 196 cooperative concern, 162–164
specific strategic priorities, 199 humanitarian cooperation, 165–167
Tipsheets (OECD DAC), 198–199 humanitarian imperative, 158–160
See also Women imperfect cooperation, 167–169
General Agreement on Trade and Tariffs Maren’s views, 158
(GATT), 19 moral divisions of labour, 169–171
Genocide(al), 1, 17, 222 Humanitarian Ombudsman Project,
violence in Rwanda, 1, 159, 178 128
‘Global alliance for global values,’ 16 Humanitarian/relief organisations, 39
Globalisation, 14–17, 24–25, 27 Humanitarian work
Global poverty, 14, 18–19, 23–25, 55, 185–186 principles in evaluation of, 122
Global Voices on-line and Witness (Global Human rights
Voices date), 140 conceptual and ideological divides, 40
Greenpeace, 71, 74 and development, 40–42
Groupe URD in France, 127 development in terms of rights, 41–42
Group size and sub-group responsibility,
constitutive relevance, 41
184–189
instrumental relevance, 41
See also International Non-Government
rights language, 42
Organisations (INGOs)
as feminists in Western/developing
nations, 199
H
INGOs, addressing economic/civil and
Habitat for Humanity, 195
political rights, 47–49
HAP, see Humanitarian Accountability
Partnership (HAP) HRW, 47–49
HAP 2007 Standard, benchmarks for, 129 right to development, 41
‘Hidden’ schools, 200 ‘new international economic order,’ 41
Humanitarian Accountability Partnership and states, 42
(HAP), 106, 127, 129, 224 Human Rights Charter, 198
Humanitarian Accountability Project, 128, 152 Human Rights Watch (HRW), 39, 47, 48,
Humanitarian aid, 1, 90, 92–93, 106, 158, 165, 53, 71
168
and development aid, 90 I
manipulation by military groups, account ICRCC, see International Red Cross Code of
on, 159 Conduct (ICRCC)
manipulation of humanitarian aid in camps, IFRC, see Red Cross Code of Conduct (IFRC)
165–167 Impact
Humanitarian Charter of Sphere Project, 181 accountability mechanisms, 151–152
Humanitarian cooperation, 165–167 ActionAid, 152
effective inter-agency cooperation, 167 ALPS, 152
exploitation of humanitarian work, Humanitarian Accountability Project
166–167 (2007), 152
236 Index
Listening Project, 152 protecting against violation of economic

three levels, 151 and social rights, 53–54
Asia Tsunami, 150 development in terms of rights, 41–42
assessment on achieving four critical goals, See also Human Rights
153 economic/civil and political rights, 47–49
clarity about values/approach/change, 154 See also Rights
definition, 148–149 ethical questions, see Ethical questions
increasing transparency, 153 human rights, see Rights
policy of integrating indigenous learning and accountability, see
Australians, 150 Accountability
of programs between donors and NGOs
communities, 150 ethics for political, 79–82
promoting dynamic models of planning political legitimacy, 77–79
and implementation, 154–155 structure of NGOs, 76
Wildavsky on rational societies, 155 and politics, 6–7
Imperfect cooperation, 167–169 potential biases for, 120–121
differences of opinion, 169 suggested by Horton, 120–121
sub-groups, 168 uncertainties, 121
India, Right to Information Campaign and principles and evidence, see
Citizen Report cards, 141 Ethics/principles/evidence,
Institute of Public Affairs (IPA), Australia, 71 INGOs
‘Institutional’ approach (Pogge, Thomas), 184 rights and duties, development INGOs
Institutional ends and developmental ends, 95 cost effectiveness, 54
Inter-agency cooperation, 166–167, 170 development assistance and rights
International borrowing privilege, 185 protection, 56–59
International Covenant on Civil and Political enjoyment of rights, lack of
Rights, 40 development, 55–56
‘International moral scepticism,’ 15 human rights INGOs, addressing
International Non-Government Organisations economic/civil/political rights,
(INGOs) 47–49
aid agencies naming/shaming methodology, HRW,
altruistic collective action, 175 53
Bush administration on the conduct of protecting against international threats,
US foreign policy, 177 54–55
‘compromised humanitarianism,’ 176 protecting against violation of
‘cooperative approach,’ 176 economic/social rights, 53–54
‘Do No Harm’ project, 178 and rights-based duties, priorities,
group size and sub-group responsibility, 49–52
184–189 supporters, satisfaction of, 139–142
harm and cooperative approach to See also Advocacy
morality, 180–184 International Red Cross Code of Conduct
human rights or social justice, 176 (ICRCC), 181
manipulation of aid in camps, 177 International relations, role of ethics in, 15–17
as organisations, 93–95 collective ‘responsibility to help to protect
Rwandan genocide in 1994, 178 populations, 17
development INGOs, duties and moral ‘global alliance for global values,’ 16
priorities ‘international analogue of nineteenth-
arms trade, 54 century liberalism,’ 17
development assistance and rights ‘international moral scepticism,’ 15
protection, 56–59 neo-liberalism, 15
lack of development, 55–56 protests against the WTO in Seattle, 16
protecting against international threats, realism’s dominance, 16
54–55 See also Poor in world of nation states
Index 237
‘International resource privilege,’ 185 Morality

Iraqi Oil Law, 186 ALNAP, 181
Iraq war, 178–179, 186 of concern, 160
complex, ethical issues, 74 of cooperation, 10, 161, 165–166, 168–170
humanitarian and reconstruction efforts, 74 cooperative approach to, 179–184
MNF-I, 187 of respect, 160–161
US-led coalition, aggressive and illegal Most Significant Change Approach, 135, 140
invasion, 187 MSF’s withdrawal from Rwandan camps, 166
worst humanitarian crises, 186 Multi-National Force in Iraq (MNF-I), 187
J N
Journal of Medical Ethics, Business Ethics Neo-liberal economic theory, 24
Quarterly, and Environmental Neo-liberalism, 15, 24, 67–68
Ethics, 2 ‘New international economic order,’ 41
New Philanthropy Capital, 105
K ‘NGO-isation of resistance’ (Roy, Arundhati),
Kyoto protocol, 28 68
NGOs, see International Non-Government
L Organisations (INGOs)
Labour, moral divisions of, 169–171 NZAID and New Zealand NGOs, Strategic
inter-agency cooperation, 170 Framework for Relations between,
morality of cooperation, 170 81
See also Humanitarianism
Learning and accountability, INGOs O
‘adaptive systems’ solution, 136 Official development assistance (ODA), 13, 67
advantages/disadvantages of different OIYP, see Oxfam International Youth
solutions, 138t Partnerships (OIYP)
aggregation problem, 135 OIYP, development approach, 211
attribution problem, 134–135 One Word Trust, 137
measurement problem, 134 Organisational and managerial problems, 94
methodological pluralism, necessity of, 137 Oxfam America or UK, 177, 188
ownership problem, 135 Oxfam International, Briefing Paper on Iraq by,
Oxfam’s experience, suggestions, 137 186–187
‘political’ or ‘power-based’ solution, Oxfam International Youth Partnerships
136–137 (OIYP), 210–211, 213
responses, 136–138
‘scientific’ or ‘measurement’ solution, 136 P
See also International Non-Government Pacific, protecting old growth forests and
Organisations (INGOs) biodiversity (social program), 149
Lifeboat ethics (Hardin, Garrett), 25 ‘Paradigm wars,’ 137
Listening Project, 152, 200 Participatory Budgeting in Brazil, 141
Literature, ethical dilemmas, 195–197 People in Aid, 129–130
See also Dilemmas Performance based incentives for teachers,
‘Logical framework analysis,’ 208–209 Australia, 149
Perpetual Peace (1891), 18
M Pogge, Thomas, 25, 55, 177, 220
Manipulation approach for ethical basis and extent of
of aid in camps, 177 obligations, 18–19
of humanitarian aid by military groups, 159 See also Poor in world of nation states
by political parties, 77 ‘indirection’ thesis, 47
Medicins Sans Frontiers, 177 See also Rights
Methodological pluralism, necessity of, 137 ‘institutional’ approach, 184
MNF-I, see Multi-National Force in Iraq Policy of integrating indigenous Australians,
(MNF-I) 150
238 Index
Political legitimacy lifeboat ethics, 25

NGOs’, 77–79 Perpetual Peace (1891), 18
sources of legitimacy for engaging in Pogge’s approach, 18–19
political activity, 78–79 proposals (Shiva, Vandana), 25
political parties’, 77 RTP, 23
motives, individuals or corporations, 77 utilitarian scholars, Singer’s approach,
say over policy, 77 20–21
Political NGOs, ethics for, 79–82 Washington Consensus, 24
ACFID Code of Conduct for Australian ‘well ordered’/‘burdened societies,’ 23
NGOs, 81 World Vision Australia (2006), 25
Code of Conduct, conceptualisation, 81–82 political support for obligations to, 27–28
ethical challenges, 79–80 AusAID (Australian Government’s aid
ethical responses, 80–82 agency), 27
funding, 80–81 elimination of extreme poverty, 28
no-hypocrisy principle, 80 role of ethics in international relations,
NZAID and New Zealand NGOs, 81 15–17
strategic framework for relations between collective ‘responsibility to help to
NZAID and New Zealand NGOs, protect populations, 17
81 ‘global alliance for these global values,’
See also International Non-Government 16
Organisations (INGOs); Political ‘international analogue of nineteenth-
legitimacy century liberalism,’ 17
‘Political’ or ‘power-based’ solution, 136–137
‘international moral scepticism,’ 15
Political support for obligations to world’s
neo-liberalism, 15
poor, 27–28
protests against WTO in Seattle, 16
AusAID (Australian Government’s aid
realism’s dominance, 16
agency), 27
Potential biases for INGOs, 120–121
elimination of extreme poverty, 28
See also Poor in world of nation states suggested by Horton, 120
Politics uncertainties, 121
definition, 65 See also International Non-Government
INGOs and Organisations (INGOs)
acting illegally, 7 Poverty, global, 14, 18–19, 23–25, 55, 185–186
being political, 6 ‘Prisoner’s Dilemma,’ 163, 175
ethics for, see Political NGOs, ethics Private sector and the World Bank (Denning,
for Steve), 140
funding from governments of developed Problems/solutions
countries, 7 aid agencies
internal politics, 7 concerns about work of, 90–93
legitimacy, 6 as organisations, 93–95
relations with governments of lack of reliable data and expert guidance,
developing countries, 7 88–89
See also Political legitimacy; Political need of data or guidance, 89–90
NGOs, ethics for potential solutions
Poor in world of nation states independent body to be set up to
ethical basis and extent of obligations, evaluate work of NGOs, 101–104
18–26 initiatives to improve accountability
American public support for and standards, 106–107
humanitarian intervention, 22 less ambitious bodies, 104–106
‘Black Hawk Down’ incident, 22 to make more data on their activities,
Bosnian Serb targets (1995), 22 99–101
“coercive regimes,” 20 personal judgements by experts,
globalisation impact, 14 107–109
Index 239
See also Solutions for epistemic development INGOs, duties and moral
problem priorities
taking stock, 96 cost effectiveness, 54
See also Epistemic problem development assistance and rights
Proposition (Singer, Peter), 119 protection, 56–59
Public choice theory, 70 enjoyment of rights, lack of
See also Advocacy development, 55–56
Public service contractors, 82 human rights INGOs, addressing
‘Publish What You Pay initiative with Mining economic/civil/political rights,
Companies,’ 138 47–49
naming/shaming methodology, HRW,
Q 53
Quality Compass, 129–130 protecting against international threats,
54–55
R protecting against violation of
Rational societies (Wildavsky, Aaron), 155 economic/social rights, 53–54
Realism, 15–16, 27 and rights-based duties, 49–52
Recipient country See also International Non-Government
challenging political equilibrium, 66–67 Organisations (INGOs)
CIA destabilisation of Chile (1970s), 66 and duties, see Duties
Turkey, NGOs in, 67 ESC, 40
NGOs, accountability to state entities, gay and lesbian rights, 195
68–69 HRW, 39, 47, 48, 53, 71
difficulties for NGOs, 68 human rights
substituting for state and promoting conceptual and ideological divides, 40
neo-liberalism, 67–68 and correlative duties-conceptual
Mudingu’s write up, 67 issues, see Duties
‘NGO-isation of resistance’(Roy, development in terms of rights, 41–42
Arundhati), 68 as feminists in Western/developing
against NGOs accepting any official nations, see Dilemmas
funds (Korten, David), 67–68 INGOs, addressing economic/civil and
suspicion on donor-country NGOs, 67 political rights, 47–49
taking sides of, see Ethics of taking sides ‘new international economic order,’ 41
Red Cross Code of Conduct (IFRC) right to development, 41
extension of code for development and and states, 42
advocacy work, 123t–124t Human Rights Charter, 198
Red Cross/Red Crescent Movement, 127 India, Right to Information Campaign and
Relief aid, 93 Citizen Report cards, 141
Respect International Covenant on Civil and
Kantian notion of, 194 Political Rights, 40
meaning, 194 ‘Right-grounded duties and the Institutional
Responsibility To Protect (RTP), 16, 23 Turn,’ 44
‘Right-grounded duties and the Institutional rights language, 42
Turn,’ 44 rights to aid (positive rights), 43
Rights rights to forbearance (negative rights), 43
-based duties, 49–52 social justice and women’s rights
division of labour between organisa- (Nussbaum, Martha), 196
tions, 51 structure of moral right and complexity of
extent and nature of need, 50–51 duties, 45–47
two factors, 49 ‘indirection’ thesis (Pogge, Thomas),
Basic Rights (Shue, Henry), 44, 50, 58 47
CRC, 198 ‘successive waves of duties’ (Waldron,
Danish Institute for Human Rights, 48 Jeremy), 47
240 Index
Universal Declaration of Human rights, 42 voice in policy formulation, 76

1993 UN World Conference on Human Supporters of INGOs, satisfaction of
Rights, 199 Aid Ombudsman, 142
women’s rights, focus on, 199 link between donor and front-line, 140
Rights to aid (positive rights), 43 promoting transparency/learning/
Rights to forbearance (negative rights), 43 accountability, steps, 141–142
RTP, see Responsibility To Protect (RTP) See also International Non-Government
Rwandan genocide in 1994, 178 Organisations (INGOs)
S T
‘Scientific’ or ‘measurement’ solution, 136 Taking sides, consequences of, 72–76
‘Service delivery,’ provision of basic goods AusAID NGO Cooperation Program, 73
and services, 91 Care International (NGO), 75
Singer, Peter, 23, 26, 119–121, 133, 148, 220 D’Cruz, views of, 73
proposition (1999), 119 East Timor Community Assistance
utilitarian scholars, Singer’s approach, Scheme, 74
20–21 government’s side of the issue, examples
See also Poor in world of nation states of, 72–73
Single-issue groups (Howard), 69 HIV/AIDS funding, Bush administration,
Slim, Hugo, 39, 66, 75, 121, 133, 137, 142 74–75
analysis of ethical dilemma’s facing NGOs, Iraq instance, 74–75
183 Oxfam Australia, 73
ethical framework assessment of Oxfam’s campaigning funds, 74
humanitarian work, 125t–126t TEC, see Tsunami Evaluation Coalition (TEC)
relevance to development and advocacy TEC evaluation, 128
work, 125–126 Traditional liberals, 17
See also Accountability; Advocacy Tsunami, 94, 120, 128, 142, 150, 212
HAP, 106, 127, 129, 224 Tsunami Evaluation Coalition (TEC), 128, 142
humanitarian ideals and reality of
humanitarian response, 195–196 U
structural violence, 124 Uganda Participatory Poverty Assessment
Social justice and women’s rights (Nussbaum, Program, 141
Martha), 196 Universal Declaration of Human rights, 42
Solutions for epistemic problem UNOSOM II mission, 22
advantages and disadvantages of different UN Security Council resolution 814, 22
solutions, 138t 1993 UN World Conference on Human Rights,
experts to give personal judgements, 199
107–109 Urbanisation, 28
improving accountability and standards, Utilitarianism, 182
106–107
independent body for work evaluation,
101–104 V
less ambitious bodies, 104–106 Video Hub Pilot using the model of You-Tube,
making data on own activities, 99–101 140
Special interest groups, 70
Sphere Humanitarian Charter and Minimum W
Standards, 127 War on Terror, 66
Sphere Project, 128, 178, 181–182 Washington Consensus, 24
Status quo/human dignity, 180–181 ‘Well ordered’/‘burdened societies,’ 23
‘Structural violence,’ 124 Westphalian system, 13
Structure of NGOs, 76 Women
relationship between the executive and beliefs and values about status of, 198–199
beneficiaries, 76 in development practice, 197–201
Index 241
rights, 195–196, 199 World Summit in September 2005, 17

See also Rights World Trade Organization (WTO), 16, 19
social justice and women’s rights
World Vision Area Development Program in
(Nussbaum, Martha), 196
Jaipur, India, 201
See also Dilemmas
See also Gender World Vision Australia (2006), 25
2003 World Disasters Report, 121 WTO, see World Trade Organization (WTO)

Ethical Questions and International NGOs - An Exchange Between Philosophers and NGOs (PDFDrive)

Uploaded by

Document Information

Original Title

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

Ethical Questions and International NGOs - An Exchange Between Philosophers and NGOs (PDFDrive)

Uploaded by

Copyright:

Available Formats

LIBRARY OF ETHICS AND APPLIED PHILOSOPHY

Marcus Düwell, Utrecht University, Utrecht, NL

Deryck Beyleveld, Durham University, Durham, U.K.

For further volumes:

With a Foreword by Peter Singer

© Springer Science+Business Media B.V. 2010

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

We are very grateful to the Australian Research Council Governance Research

September 2009 Keith Horton and Chris Roche

Garrett Cullity University of Adelaide, Adelaide, SA, Australia,

Keith Horton and Chris Roche

K. Horton, C. Roche (eds.), Ethical Questions and International NGOs, Library of 1

2 Ethical Questions and INGOs: A Classification

2.1 Ethical Questions Raised by Specific Kinds of Activity INGOs

2.1.1 Emergency Relief

2.1.2 Service Delivery

2.2 Ethical Questions Concerning the Selection of Activities

2.2.1 Strategic Choice

2.2.2 Specific Choices

2.2.3 The Risk of Negative Effect

2.3 Ethical Questions Concerning the Relationship Between

2.3.3 Cultural Differences

2.3.5 Systems and Procedures

2.4 Ethical Questions Concerning the Relationship Between

2.4.4 Active Engagement

2.4.5 Institutional Imperatives

2.5 INGOs and Politics

2.5.1 Being Political

2.5.3 Internal Politics

2.5.4 Funding from Governments of Developed Countries

2.5.5 Relations with Governments of Developing Countries

2.5.6 Acting Illegally

2.6 Ethical Questions Related to Broader Questions of Individual

2.6.3 Collective Action Problems

2.6.4 Institutionalising Changes

3 The Papers in this Volume

• Suggestions about how asserting the importance of organisational values, adjust-

We finish here by briefly summarising each of the papers.

and argues that it has a number of unfortunate consequences. Accordingly, he then

Can we find in the human ‘thou’, a fragment of the divine

K. Horton, C. Roche (eds.), Ethical Questions and International NGOs, Library of 13

2 The Role of Ethics in International Relations

view.12 However, as realism’s dominance began to be challenged after the end of

and support armed humanitarian intervention to protect the delivery of international

3 Ethical Basis and Extent of Obligations to the World’s Poor

to as the Washington Consensus, requires national and global institutions to be

investments proved to be a key to their strong economic growth (Diamond 2005,

4 Political Support for Obligations to the World’s Poor

Secondly, the elimination of extreme poverty which would significantly con-

Contemporary globalisation has changed the world. Subsistence farmers in Laos

as globalisation continues, pressure for such co-ordinated global action in relation

International NGOs (INGOs) dedicated to securing the vital interests of vulnera-

K. Horton, C. Roche (eds.), Ethical Questions and International NGOs, Library of 39

2 Human Rights and Development

2.1 Human Rights: Conceptual and Ideological Divides

2.2 A Right to Development

2.3 Understanding Development in Terms of Rights

Alongside arguments over whether development is properly conceived of as the

2.4 Human Rights and States

3 Henry Shue on Basic Rights and Duties

3.1 Human Rights and Correlative Duties—Conceptual Issues