International Law, Politics, and Inhumane Weapons: The Effectiveness of Global Landmine Regimes 9780415622059, 0415622050

This book contributes to contemporary debates on the effectiveness of international humanitarian law (IHL) in regulating

192 97 2MB

English Pages 173 [193] Year 2012

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

International Law, Politics, and Inhumane Weapons: The Effectiveness of Global Landmine Regimes
 9780415622059, 0415622050

Table of contents :
International Law, Politics and Inhumane Weapons The effectiveness of global landmine regimes
Copyright
Contents
List of tables
Preface
Acknowledgements
List of abbreviations
1 The design, implementation and effectiveness of global landmine regimes
2 Regimes prohibiting the use in war of poison gas and dum dum bullets
3 The emergence of the landmine regimes
4 Implementing the landmine regimes
5 Humanitarian demining
6 Stockpile destruction
7 The effectiveness of global landmine regimes
Notes
Bibliography
Index

Citation preview

LAW, CONFLICT AND INTERNATIONAL RELATIONS

International Law, Politics and Inhumane Weapons The effectiveness of global landmine regimes Alan Bryden

International Law, Politics and Inhumane Weapons

This book contributes to contemporary debates on the effectiveness of international humanitarian law (IHL) in regulating or prohibiting inhumane weapons, such as landmines. Two treaties have emerged under IHL in response to the humanitarian scourge of landmines. However, despite a considerable body of related literature, clear understandings have not been established on the effectiveness of these international legal frameworks in meeting the challenges that prompted their creation. This book seeks to address this lacuna. An analytical framework grounded in regime theory helps move beyond the limitations in the current literature through a structured focus on principles, norms, rules, procedures, actors and issue areas. On the one hand, this clarifies how political considerations determine opportunities and constraints in designing and implementing IHL regimes. On the other, it enables us to explore how, and why, ‘ideal’ policy prescriptions are threatened when faced with complex challenges in post-­conflict contexts. This book will be of much interest to students of international humanitarian law, global governance, human security and International Relations in general. Alan Bryden is Deputy Head of Research, Geneva Centre for the Democratic Control of Armed Forces (DCAF ), and has a Phd in International Relations from the University of Bradford, UK.

Law, Conflict and International Relations Series Editors: Chandra Lekha Sriram SOAS

Julie Mertus

American University

Karin Aggestam University of Lund

and Lorraine Elliott

Australian National University This series will bring together cutting-­edge, interdisciplinary scholarship on law, conflict, and international politics, encompassing the fields of international criminal law, international human rights law, and international humanitarian law, law relating to the use of force, and conflict prevention, resolution, peacemaking, and peacebuilding, and resort to the use of force. International Law, Politics and Inhumane Weapons The effectiveness of global landmine regimes Alan Bryden

International Law, Politics and Inhumane Weapons The effectiveness of global landmine regimes Alan Bryden

First published 2013 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2013 Alan Bryden The right of Alan Bryden to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-­in-Publication Data Bryden, Alan. International law, politics, and inhumane weapons: the effectiveness of global landmine regimes/Alan Bryden. p. cm. Includes bibliographical references and index. 1. Land mines (International law) 2. Mines (Military explosives) (International law). 3. Humanitarian law. I. Title. KZ5645.B79 2013 341.7′3–dc23 2012010832 ISBN: 978-0-415-62205-9 (hbk) ISBN: 978-0-203-09680-2 (ebk) Typeset in Times by Wearset Ltd, Boldon, Tyne and Wear

Contents



List of tables Preface Acknowledgements List of abbreviations

1 The design, implementation and effectiveness of global landmine regimes

vi vii x xi

1

2 Regimes prohibiting the use in war of poison gas and dum dum bullets

17

3 The emergence of the landmine regimes

41

4 Implementing the landmine regimes

64

5 Humanitarian demining

87

6 Stockpile destruction

106

7 The effectiveness of global landmine regimes

123



Notes Bibliography Index

139 158 170

Tables

3.1 3.2 3.3 4.1

Alignment of states within the CDDH US negotiating red lines in Oslo Key features of APII and the mine ban treaty Landmine regime implementation mechanisms

42 46 47 67

Preface

‘Bloody NGOs’ was a fairly common expression when I was desk officer for international humanitarian law (IHL) at the Ministry of Defence. It was 1998 and my personal gripe was that the targeted letter-­writing campaign to the Defence Secretary had not abated with the UK signature of the mine ban treaty. As a consequence, on top of providing policy advice on IHL, I was required to draft Ministerial responses to literally hundreds of letters from concerned members of the public. Frustratingly, ‘cut and paste’ was not an option since they were all slightly differently worded and asked hard (i.e. politically sensitive) questions on the scope and meaning of UK commitments relating to landmines. The more difficult ones somehow found their way into Parliamentary Questions, which had to be even more carefully answered. In short, from a very humble position I was on the receiving end of an extremely well-­organized and well-­informed advocacy campaign. Beyond my own travails, there was wider unease around Whitehall. Within defence circles there was a feeling that having been involved in lengthy negotiations about regulation, the UK had suddenly been railroaded through pressure from non-­governmental organizations (NGOs) into banning anti-­personnel mines. Among concerned diplomats there was a less vocal, but no less real, sense that this was not business as usual. Most uncomfortably, in supporting the landmine ban as well as the creation of an International Criminal Court, the UK had within a relatively short period taken the highly unusual step of going against the Americans on two major security policy issues. Again, a large part of the blame seemed to reside with an NGO coalition that did not play by the accepted rules of the game – co-­opting states and naming and shaming those out of step with the movement were two particularly unsporting practices adopted by the International Campaign to Ban Landmines (ICBL). This sense of ‘them and us’ was only reinforced by the substance, tone (and effectiveness) of interventions by ICBL representatives at any number of conferences and workshops. In 2000 I was seconded to the Department for International Development to help manage the portfolio of UK-­supported projects conducted by the recently created Geneva International Centre for Humanitarian Demining. This was my first real exposure to the mine action community. During two years at the centre,

viii   Preface I worked with a wide range of professionals involved in establishing and running mine action programmes, or on specific activities from landmine awareness training to the use of dogs or technology to detect landmines. At the same time, Geneva was becoming the hub for implementation efforts under the mine ban treaty. Yet there did not seem to be much of a connection between treaty implementers and mine action experts. Cooperation was not pursued by either side. One reason expressed by some within mine action was that the proliferation of treaty-­related meetings offered little more than a talking shop for advocacy groups and well-­meaning diplomats rather than contributing meaningfully to their work. These very different work experiences in London and Geneva drove me to consider, in more depth, efforts under IHL to regulate or prohibit landmines. The stereotypes that different stakeholders assigned to each other felt unsatisfactory. Beginning in late 2000, I went beyond my own experience to examine what had been said and written about these treaties. This brought me to the conclusion that simplistic views had resulted in simplistic assumptions: the Ottawa Process is unique and unprecedented; a ‘ban’ renders regulation meaningless; actors central to regime creation can make an equally important contribution to implementation. But the most striking assumption was the following: widespread adherence to and observance of treaty provisions will generate mine action payoffs that improve the lives of people living with landmines. The issues involved, and the people with a stake in them, are too important not to ask hard questions of the landmine treaties. This book certainly does not claim to have all the answers, but it does try to move the debate forward in three ways. First, it provides some historical context and perspective to IHL treaties that remain relatively young and untried. Second, it looks beyond partisan lines that seem particularly sharply drawn in this field: an analytical approach grounded in regime theory unpacks treaty design, implementation and effectiveness through a structured focus on principles, norms, rules, procedures and issues. Finally, this book represents an attempt to build bridges between knowledge communities. It seeks to demonstrate that for the landmine regimes to be effective, they need to engage with mine action expertise in the implementation phase where ideal treaty provisions come face to face with complex realities in challenging environments. At the same time, the mine action community needs to open up and actively develop synergies with related fields. Since the two landmine treaties were agreed, governments and civil society actors have learned to partner in ways that advance common humanitarian, security and development goals. In this respect, the Ottawa Process is rightly regarded as having provided a truly catalyzing effect. Success in developing new IHL obligations on cluster munitions is just one more recent example of the efficacy of civil society-­committed state coalitions. Yet lessons from the landmines experience should also provide a cautionary note moving forward. Once we look beyond quantitative measures of success – numbers of states parties, weapons destroyed etc. – the hard questions still need to be asked. Is the conduct of weak or authoritarian governments being influenced? Are the roles of armed non-­state

Preface   ix groups acknowledged and addressed? Can we demonstrate that international treaty regimes tangibly improve the security of individuals and communities? Answers to this kind of question make all the difference between implementation and effectiveness. Alan Bryden

Acknowledgements

This book took shape over a long period. While I am fully responsible for the outcome, a number of people should be acknowledged that have – in different ways – strongly influenced my approach and thinking on the subject. While on the IHL desk at the Ministry of Defence, Andrew Wood allowed me a great deal of flexibility but also provided top cover when it mattered. Over the same period, Jonathan Passmore was the first of a number of outstanding Royal Engineers officers that I have been privileged to work with. My line managers at the Geneva International Centre for Humanitarian Demining – Alastair McCaslan and Paddy Blagden – taught me a great deal through a combination of dedicated and inspired leadership. Then GICHD Director, Martin Dahinden, showed me that strategic direction and a genuine commitment to the field can go hand in hand. Owen Greene, in the Department of Peace Studies at the University of Bradford, accompanied me throughout and provided the initial idea to analyse the landmine treaties as international regimes. I spent a month as a trainee with the HALO Trust in Cambodia. During this period I was involved in surveying, worked with explosives, trained as a deminer and spent three days clearing a lane in a minefield near Anlong Veng. Most of my time was spent with a team of dedicated, generous, friendly, humble and everyday brave Khmer deminers working for HALO. I can honestly say that this experience profoundly changed my own outlook and approach. For this reason, sincere thanks go to HALO Trust Director Guy Willoughby and all the dedicated staff at HALO who made my being there possible. Finally, I could not have completed this work without my family. My parents Andrew and Mary have always been unstinting in their love and support. My beloved wife and children – Evelyne, Gordon and Sam – provided the motivation to finish what I started. They also bore the brunt of my many absences. I love them dearly and promise to do better in the future.

Abbreviations

AHD AICMA ANSA AP II APEC APM AVM BWC CARICOM CBW CCM CCW CD CDDH CMAC CPA CWC DDA DDR DfID DRC E-­MINE ERW EU FFM HALO ICBL ICC ICJ ICRC IHL IMAS

Anti-­handling device Action against anti-­personnel mines Armed non-­state actor Amended Protocol II Asia-­Pacific Economic Cooperation Anti-­personnel mine Anti-­vehicle mine Biological Weapons Convention Caribbean Community Chemical & biological weapons Convention on Cluster Munitions Convention on Certain Conventional Weapons Conference on Disarmament Conference on the Reaffirmation of IHL Applicable in Armed Conflicts Cambodian Mine Action Centre Comprehensive Peace Agreement Chemical Weapons Convention Department of Disarmament Affairs Disarmament, demobilization and reintegration UK Department for International Development Democratic Republic of the Congo Electronic Mine Information Network Explosive remnants of war European Union Fact-­finding mission Hazardous Areas Life Support Organization International Campaign to Ban Landmines International Criminal Court International Court of Justice International Committee of the Red Cross International Humanitarian Law International Mine Action Standard

xii   Abbreviations IMSMA IRC ISU KLA KPC LSN MAC MAG MASG MINURSO MRE NATO NGO NPA OAS OAU ODA OECD DAC OPCW PfP PKK POF SADC SALW SPLA/M SSR UEFA UN UNDP UNGA UNMAS UNMIK WMD

Information Management System for Mine Action International Red Cross Implementation Support Unit Kosovo Liberation Army Kosovo Protection Corps Landmine Survivors Network Mine Action Centre Mines Advisory Group Mine Action Support Group United Nations Mission for the Referendum in Western Sahara Mine risk education North Atlantic Treaty Organization Non-­governmental organization Norwegian Peoples Aid Organization of American States Organization of African Unity Official Development Assistance Organization for Economic Cooperation and Development, Development Assistance Committee Organization for the Prohibition of Chemical Weapons Partnership for Peace Kurdistan Workers Party Pakistani Ordnance Factory Southern African Development Community Small arms and light weapons Sudanese Peoples Liberation Army/Movement Security sector reform European Union of Football Associations United Nations United Nations Development Programme United Nations General Assembly United Nations Mine Action Service United Nations Mission in Kosovo Weapons of mass destruction

1 The design, implementation and effectiveness of global landmine regimes

Introduction Two treaties have emerged under international humanitarian law (IHL) in response to the humanitarian scourge of landmines. Situating the landmine treaties within the wider body of IHL is essential in order to develop a contextualized understanding of regimes that have emerged only relatively recently. However, despite a considerable body of related literature, clear understandings have not been established on the effectiveness of these international legal frameworks in meeting the challenges that prompted their creation. This book seeks to address this lacuna. In particular, it applies a regime theory framework in order to explore the hypothesis that design factors, and their relationship to subsequent implementation, are critical to the effectiveness of these regimes. IHL is based on the premise that the legitimate means of warfare are not unlimited and that a balance must be struck between humanitarian concerns and the onus of military necessity. The body of IHL treaty law and custom attempts to reconcile conflicting interests from the security concerns of authoritarian governments to the aspirations of peace advocates, humanitarian and development workers. The range of actors involved is thus extremely diverse. Primary stakeholders are states, service personnel and civilians caught up in armed conflict and its aftermath. However, other interested parties may include international organizations, commercial companies, the media, international lawyers, academics and civil society more broadly. Even within national governments, contradictory responsibilities and interests that touch on IHL are sometimes found within foreign and defence ministries, development agencies and trade and industry departments. This multiplicity of issues, interests and actors provides a complex basis from which to analyse IHL effectiveness. The complexities of IHL are particularly apparent in the international treaties that address landmines. The existence of two treaties with the common goal of alleviating the human suffering caused by these weapons1 suggests that the international community recognises the importance of this issue. According to the prevalent narrative on the mine ban treaty2 it represents, in both process and substance, a novel and effective approach to IHL that has created an international

2   Regime design, implementation and effectiveness norm against the use of anti-­personnel mines (APMs). The corollary to this argument is that the ‘other’ landmine treaty, Amended Protocol II (APII) to the Convention on Certain Conventional Weapons (CCW),3 is the ineffective outcome of a slow and inadequate process. In reality, certain mine action practitioners are sceptical about the effectiveness of both regimes. Concerns over the utility of the landmine regimes can only be addressed through a rigorous focus on their effectiveness. Regime theory provides a useful framework to understand why, during the shift from policy goals to their application, theoretically sound principles often come unstuck when faced with challenging real world situations. This approach unpacks the principles, norms, rules, procedures, actors and issue areas that have shaped the design, implementation and effectiveness of multilateral regimes. It seems particularly appropriate to the study of landmine regimes, because effectiveness not only involves addressing complex technical problems, but also coming to terms with dilemmas and ­obstacles of an inherently political nature.

The landmines issue in context IHL, also known as the ‘laws of war’, is intended to minimize the suffering caused in armed conflict. It refers to rules between states governing armed conflict (jus in bello) but not the resort to armed conflict (jus ad bellum). The origins of IHL can be traced back almost as far as warfare itself: the Greeks and the Romans customarily prohibited the use of poison or poisoned weapons in combat.4 There is an extensive body of literature on IHL in general, as well as on particular treaty regimes. This includes negotiating histories and other commentaries,5 as well as numerous works looking at the impact of IHL within the broader framework of international relations. IHL can be divided into two branches. ‘Geneva law’ deals with the treatment of combatants, non-­combatants and civilians caught up in armed conflict, while ‘Hague law’ regulates the means and methods of warfare. Geneva law includes Conventions drawn up in 1864, 1906, 1929 and 1949 with the Geneva Conventions of 1949 replacing the previous ones. The two Additional Protocols of 1977 to the 1949 Geneva Conventions, are considered to combine both Geneva law and Hague law in that they govern the treatment of individuals caught up in war, but also prohibit weapons that cause ‘unnecessary suffering’ or ‘superfluous injury’.6 The restrictions and regulations on the means and methods of warfare found in Hague law are most applicable to the landmines issue. An important distinction should be made between binding international treaties and customary rules of warfare. Custom complements treaty law, because any treaty provision which embodies customary law is binding on all states whether or not they are parties to the treaty in question. From a regime perspective this adds the complicating factor that obligations are not simply laid down in a specific treaty, but also derive from a much wider corpus of international law and practice. There are multiple links between rules governing the conduct of armed conflict and controls and prohibitions on specific weapons systems. The 1925 Gas

Regime design, implementation and effectiveness   3 Protocol, like the CCW and mine ban treaty, falls into both categories: containing strong elements of IHL and arms control. However, the differences between these two branches of international law are equally important. In particular, there is a clear shift in discourse between arms control frameworks that take into account the military utility of a given weapon, and the humanitarian perspective of IHL, premised squarely on alleviating the human suffering caused by certain means and methods of warfare. The weight that the respective regimes accord to these two approaches can be a major influencing factor on how rules, norms and actors influence treaty design and implementation. The quality of ‘nesting’ within a wider normative framework is an important distinguishing feature between IHL and ‘hard’ security regimes. It is thus important to take into account how far the landmine regimes, individually and collectively, engage with the broader framework of treaty law and IHL norms that all states are bound to observe, regardless of their respective membership choices. Landmines and IHL Given the relatively short period since APII and the mine ban treaty entered into force, examining relevant historical case studies can permit a more mature critique of regime implementation from a longer historical perspective. Thinking about earlier IHL treaties as regimes can also help to situate landmines within this broader framework of efforts to ban or restrict the use of weapons. Are regime characteristics – constellations of actors, interests and norms – specific to this issue area or can they be linked to dynamics found in other IHL regimes? This approach also tests the frequent claim that the Ottawa Process represents a new and unprecedented departure in the field of IHL. More generally, case studies can inform our understanding of how IHL regimes develop over time. This enables us to identify trends, processes and influences that, if not visible to primary stakeholders, may nevertheless be highly influential in shaping treaty design, implementation and effectiveness. The 1899 Hague Declaration 3 Concerning Expanding Bullets (Hague Declaration 3) and the 1925 Geneva Protocol for the Prohibition of the Use in War of Asphyxiating, Poisonous or Other Gases, and of Bacteriological Methods of Warfare (the 1925 Gas Protocol) offer particularly relevant cases in relation to the landmine regimes. Hague Declaration 3 is the only IHL treaty prior to the mine ban treaty that prohibits a conventional weapon in widespread use. Like APMs, expanding bullets – commonly known as dum dum rounds – were condemned by medical practitioners who had witnessed at first hand the effects of these weapons.7 The international advocacy campaign that generated widespread public support for Hague Declaration 3 bears comparison with the Ottawa Process. While the latter is often highlighted by supporters as a unique example of international civil society and media mobilization, these same factors would seem to have proved influential in building support to pressure governments to adopt a ban in 1899. A gap of nearly one hundred years between the two treaties

4   Regime design, implementation and effectiveness represents a lengthy period over which to trace implementation dynamics and assess regime effectiveness. Hague Declaration 3 therefore provides an important comparative case study in order to understand how far the two landmine regimes represent continuity or change within the corpus of IHL. The 1925 Gas Protocol is one of the earliest binding legal restraints on a specific weapon. As with Hague Declaration 3, this case provides a number of significant parallels to the processes surrounding the design and implementation of the landmine regimes. The regime formation process tapped into strong public concern over the horrifying human costs of trench warfare in the First World War. Like landmines, a major impression was made by evidence of victims left alive to suffer from terrible injuries after exposure to these weapons. The 1925 Gas Protocol offers an early example of civil society exerting pressure to outlaw a weapon, resulting in an international norm against their use. Significantly, the negotiating process that led to its agreement was marked by the exclusion of states from the developing world. Considering Hague Declaration 3 and the 1925 Gas Protocol through the lens of regime theory, offers an opportunity to develop new insights into their interplay and the factors that have contributed to their effectiveness over time. Analysis of the shifting clusters of actors involved in design and implementation allows us to identify how these processes resonate with APII and the mine ban treaty. The weapons addressed by these regimes share the quality of having been stigmatized in the international public consciousness. Consequently, important issues for the landmine regimes such as norm-­building, spillover effects and regime nesting can only be understood through situating this analysis within its broader historical context. Origins and emergence of the landmine treaties The use of landmines in armed conflict can be traced back as far as the nineteenth century, although the practice only became widespread during the Second World War.8 The specific regulation of these weapons under IHL began in 1980 with the agreement of Protocol II to the CCW. The protocol not only covers APMs, but also anti-­vehicle mines (AVMs) as well as booby traps and ‘other devices’. Growing international recognition in the early 1990s of the effects of APMs in contemporary conflicts and the inadequacies of the existing regulatory framework to protect both combatants and, in particular, civilians, led to the convening of a review conference. This involved eight months of negotiations between 1995–1996. The rules of procedure for the conference limited participation to government representatives. This prevented many other interested stakeholders from playing a direct role in the shaping of the treaty, and contributed to a lack of public interest in the negotiations except through criticism of the process from a vociferous pro-­ban lobby. The resulting ‘amended’ Protocol II provides for a number of more stringent restrictions on the design of landmines than is contained in the original protocol. A significant element of those excluded from the CCW negotiations comprised the non-­governmental organizations (NGOs) and other civil society actors

Regime design, implementation and effectiveness   5 that between 1992–1993 began to coordinate a campaign to ban APMs. This group formed the nucleus of an International Campaign to Ban Landmines (ICBL) which, by 1995, incorporated some 350 different NGOs. Support for a ban also came from the International Committee of the Red Cross (ICRC) as well as the United Nations, with UN Secretary-­General Boutros Boutros-­Ghali highlighting the issue in his 1992 Agenda for Peace.9 At a conference held in Ottawa during October 1996, Canadian Foreign Minister Lloyd Axworthy took the initiative to set a date for formal negotiations with the goal of agreeing a complete ban on APMs. Following a year-­long, two-­track process that combined state-­level discussions with an extensive civil society-­driven advocacy campaign, the negotiating conference took place in Oslo over a three-­week period in September 1997. A wide range of states from both the developed and developing world participated while numerous NGOs, international organizations and mine action practitioners were given full access. Decision-­making was by two-­thirds majority and lengthy position statements were prohibited in order to achieve an agreed treaty text within as short a timeframe as possible. The resulting mine ban treaty provides for a complete ban on the use, production, transfer and stockpiling of APMs. APII pursues ways to minimize the effects of landmines through regulating their use while balancing these restrictions against concerns of military utility. A number of criticisms are levelled at this approach. In particular, it is claimed that through applying a logic that draws heavily on arms control antecedents, the humanitarian impact of these weapons is not directly acknowledged. A temporal consideration is also highlighted. APII does not reflect the reality that landmines may pose a danger to both communities and individuals decades after conflicts have ended. A further criticism is that APII essentially proposes technical fixes that presuppose a technological and resource base out of reach for many developing countries. In other words, the regime may have limited relevance in the very countries that suffer the most from landmines. The counterpoint to criticism of APII is that a restrictions-­based approach, developed through consensual negotiations, draws landmine producers and users that would not consider an outright ban into a constructive arms control process. While the mine ban treaty has a considerably wider membership than APII,10 many militarily-­significant states including China, India, Pakistan, Russia and the United States (that are members of APII) have not signed up to the ban. The mine ban treaty has generated a great deal of analysis and commentary in a relatively short period. In contrast, APII has attracted little attention outside of governmental circles. Yet much existing work on the evolution of the former focuses narrowly on the achievements of the coalition of like-­minded states and civil society organizations. This narrative, encapsulated in Cameron, Lawson and Tomlin’s seminal volume on the subject, To Walk Without Fear, the Global Movement to Ban Landmines,11 emphasises the unique nature of this process. It starts from the position that the Ottawa Process is fundamentally ‘a good thing’ that has re-­written the rules for the design and implementation of IHL treaties. This perspective sees international civil society, in conjunction with like-­minded

6   Regime design, implementation and effectiveness states, successfully pressuring sometimes reluctant states to be bound to new humanitarian treaty obligations through the medium of coordinated, well-­ targeted advocacy.12 The flipside of this narrative casts APII as an ineffectual response to the global landmine problem from states unwilling to ban these weapons. The mine ban treaty literature emphasizes the innovative nature of both the diverse global coalition and the tightly organized core group as pivotal to successful regime formation. These characteristics are commonly juxtaposed with a narrower (and therefore less effective) constituency contributing to the agreement of APII. In practice, little attention has been given to the influence of mine action practitioners and representatives from mine-­affected states across regime design and implementation. This raises important questions on the role, influence and interactions of different stakeholder groups. Drawing on insights from the wider regime theory discourse, one hypothesis proposes that how actors cluster at different stages of regime evolution has an important bearing on regime effectiveness. The relationship between agency and legitimacy is also significant. In particular, how far claims linked to the engagement of the global South actually extend beyond representation to ownership of an implementation process that almost always plays out within developing states, has evident consequences for regime effectiveness.

A regime approach to landmines Although regime theory emerged as a field of study in the 1970s, the issue of rules and their influence on the behaviour of states has much longer antecedents. Richard Little suggests that the study of regimes should therefore be set within a tradition that can be traced back to Hugo Grotius (1583–1645), the ‘father of international law’.13 Thus, regime theory falls within a wider framework of international rule-­making and governance that embraces both international relations and international law traditions. Different definitions of ‘regime’ have been proposed.14 Levy et al. (1995) offer a comprehensive classification, applied in this book, that defines international regimes as ‘social institutions consisting of agreed upon principles, norms, rules, procedures and programs that govern the interactions of actors in specific issue areas’.15 If there are widely contrasting views on the role played by international regimes, different schools agree that while the international system is characterized by anarchy, it is governed by rules. The regime theory discourse, though it has long historical antecedents, reflects the fact that ‘it is only during the course of the twentieth century that regimes can be regarded as a global phenomenon, with states becoming enmeshed in increasingly complex sets of rules and institutions which regulate international relations around the world.’16 Regime theory provides an analytical framework that has developed new insights into issue areas including the environment, human rights, security and trade. The relative absence of extant research that systematically considers IHL in general, or the landmines treaties in particular, from this perspective offers an opportunity to better understand what contributes to success or failure within this field.17

Regime design, implementation and effectiveness   7 Regime theory identifies a number of clear foci in relation to the development of international regimes: regime formation, implementation, effectiveness and consequences. Deconstructing APII and the mine ban treaty through examining these different elements and their interrelationships, enables us to move beyond a perceptions-­based assessment and address important gaps in knowledge. Levy et al. (1995) highlight the importance of shared (nested) principles and norms as a contributing factor to strong regimes. Agency dynamics are also influential: ‘where networks of regimes link the same set of participants, actors’ perceptions of each other’s behavior are more likely to be affected than where regimes are isolated’.18 The concept of regime nesting bridges questions on the role of norms in relation to the two landmine treaties with a more mature IHL discourse. From a theoretical perspective, this wider normative framework allows us to explore important issues of nesting and interplay and thus to delineate ‘deep’ linkages that need to be unpacked in order to determine real effects across regimes. The relationship between the design phase and regime implementation is a key determinant of regime effectiveness. For the landmines issue, focusing on both political and technical challenges connects analysis of the emergence of the treaty frameworks to a practitioner-­driven mine action literature. Drawing together these related, if disconnected, approaches is necessary in order to understand how the theory of the landmine regimes is applied in practice. The significance of political will in regime formation is well documented, but much less so in implementation. The ability of the regimes to learn and adapt in order to better meet implementation challenges therefore represent important considerations. If the mine ban treaty has been held up as an unprecedented, successful example of how to drive forward the IHL agenda, part of the success of the Ottawa Process has been in promoting its brand and juxtaposing this with the CCW. Without downplaying the significance of adroit marketing, a regime approach permits us to move beyond facile differentiations, enabling a critical assessment of claims to the uniqueness of particular regime characteristics. The three themes outlined below focus on if and why regime rules are obeyed, and how this behaviour contributes to the realization of core objectives. The interplay between design and implementation Regimes invariably require action from their members, so a clear understanding of what is needed to comply with their rules must underpin the design of specific provisions. It is essential that actors are able to comply with regime provisions. Richard Putnam describes as ‘involuntary defection’19 the situation where states fail to meet commitments because the demands placed on them are unrealistic. Implementation challenges also argue for effective compliance verification of obligations. The absence of a monitoring capability can mean that regimes lack the self-­awareness to understand implementation challenges and thus recognize when changes are needed to improve performance. A lack of clarity on these issues can also undermine confidence, leading some regime

8   Regime design, implementation and effectiveness members to question whether others are observing obligations to which all have agreed to be bound. Political will is vital to ensuring wide ratification. Yet adherence to an international treaty is only the first step. It is equally important that treaty requirements are fully integrated into domestic legal structures and regulations if behaviour is to be influenced at the national level. Thus, Antonio Cassese notes that ‘international law cannot work without the constant help, cooperation, and support of national legal systems’. This integration of international rules in national frameworks is particularly important for what Andrew Hurrell refers to as the ‘hard cases’ where national security interests form part of the calculus.20 Decisions made by regime designers can be highly influential in supporting or constraining effectiveness. Unpacking common or distinct principles, norms, rules, procedures, issues and actors within and across the landmine regimes can shed new light on the design–implementation relationship. When regimes shift to the implementation phase, obligations are tested against practical realities and political constraints. These constraints can be particularly onerous for landmine regimes given that the problem is most acute in developing countries that face a range of political, economic and security challenges. This is important because implementation presents a double test. On the one hand, this relates to the ability of members to meet commitments and avoid involuntary defection. On the other hand, the implementation phase provides the litmus test of whether states’ willingness to cut a positive international figure by joining a regime with evident humanitarian credentials, is matched by the political will to accept the costs associated with their new obligations. The ability to recognize and address implementation issues through adjusting to meet challenges that may not have been considered in regime design is therefore essential. Effectiveness is founded not just on observing rules, but on the contribution of compliant behaviour to achieving regime goals. This raises the central question of how to define landmine regime effectiveness. A useful definition of effectiveness proposed by Arild Underdal acknowledges that even if widely adhered to and fully implemented, a regime may not alleviate the problems it was set up to address. For a regime to be effective, the relevance and appropriateness of specific provisions must therefore go hand in hand with the political will and practical ability to implement them. Thus, effectiveness is a function of the interplay of design and implementation factors. Underdal’s dual approach to regime effectiveness – how far members abide by rules and the extent to which objectives are fulfilled21 – suggests that a regime can be extremely effective in terms of compliance without fulfilling its overarching objectives. A comprehensive definition of effectiveness, therefore, goes beyond the implementation of legal provisions or policy adjustments by members to include ‘changes in the behaviour of actors and in patterns of interactions among them in ways that contribute to management of targeted problems’.22 Oran Young cites two widely used criteria for regime performance: An economic criterion stresses efficiency and asks whether the same results could have been achieved at a lower cost or, alternatively, better results

Regime design, implementation and effectiveness   9 achieved at the same cost . . . a political criterion directs attention to equity and raises questions about the fairness both of the results of institutional arrangements and of their procedures or processes. 23 The nesting of the landmine regimes within the broader context of IHL points to another consideration: the humanitarian imperative to reduce the suffering caused by these weapons. A humanitarian criterion therefore represents a third dimension of landmine regime effectiveness. Effective regimes in different issue areas are often evolutionary, adapting to better address shortcomings in regime design or to meet new challenges that were not apparent to treaty drafters.24 How regimes learn is thus important. ­Demonstrating flexibility through undertaking course corrections in order to better achieve regime goals requires responsiveness to the conduct of members. Overtly, the two regimes apply very different approaches in this area. APII provides for flexibility in implementation through permitting optional deferral periods for certain obligations. In contrast, the mine ban treaty seeks to preserve the integrity of the core prohibition on APMs by categorically ruling out reservations or deferrals. While both approaches may be seen to have merits, detailed analysis of implementation processes in relation to specific mine action priorities is necessary to understand their practical implications. International regimes can generate significant consequences beyond the issue areas they are intended to address. These can include altering relations between members (as well as non-­members) and changes to broader policy positions as a result of learning through regime-­driven interactions. Regime nesting can also prompt spillover effects. For example, in the case of landmine regimes there is a need to understand why some participants actively seek to extend the regimes’ coverage into new issue areas beyond their original scope, or to apply lessons from the process to the development of new regimes. Key actors Actors influence regime formation in qualitatively different ways. In this context, Oran Young describes three kinds of leadership:25 Structural leaders who represent states and seek ways of bringing material power to bear on regime formation processes; Intellectual leaders who shape the way ideas are framed and energized; and Entrepreneurial leaders who craft options for consensus and broker contractual deals. Finnemore and Sikkink link individual leadership and norm emergence in regime formation. They identify ‘norm entrepreneurs’ as playing a significant role in convincing states to form a regime around a given issue area.26 The different types of leadership available in the regime formation phase may have important consequences for subsequent implementation and effectiveness. Implementation often requires the contribution of sets of actors beyond those involved in the regime formation phase. This points to the need to understand the contributions of different stakeholder groups as regimes evolve. In particular,

10   Regime design, implementation and effectiveness mapping how actors cluster and determining how these groupings shift from design to implementation can provide important insights. One perspective juxtaposes the mine ban treaty implementation process with the humanitarian imperatives that led to its creation. According to this narrative, while the experts are back in the field, implementation is being driven by Western diplomats and lobbyists. The perception is of an over-­emphasis on bureaucracy, at the expense of practical work, to solve the global landmine problem. This perception raises important questions because effective regimes in many issue areas require practitioner expertise in both design and implementation phases. Moreover, if political will is essential to move from consensus around an issue of concern to the negotiation of an international treaty, sustained political commitment is at least as important to ensure compliance with provisions that are costly and sensitive at the national level. While states are central players, international regimes are also social institutions in which a range of actors network. Joseph Camilleri and Jim Falk take an extreme position by citing transnational movements as sources of ‘support for a new system of multiple allegiances and jurisdictions, held together not by supreme authority but by an emerging world culture and a dynamic network of communities, movements and organizations’.27 Although this perspective may be regarded as exaggerated in its downplaying of the role of states, it demonstrates the complex interdependencies that need to be unpacked in order to understand international regimes. Mine action effectively illustrates this complexity, uniting technical demands with socio-­economic challenges that require a close understanding of impact on communities and individuals. Consequently, a wide range of non-­state actors involved in advocacy, policy and programming have a stake in this issue. The mine ban treaty engages with these diverse constituencies. APII is more state-­centric, has a lower profile and is viewed with markedly less enthusiasm by the NGO community. The problems posed by landmines are most acute for the developing countries least equipped to deal with them. This points to a clear distinction between APII and the mine ban treaty, given the lack of engagement of the former with developing world concerns, and the centrality of the global South to the latter. These agency dynamics need to be examined in detail. In particular, it is necessary to address the extent to which regime frameworks distinguish between ‘political support’ and ‘political will’ to assume the costs of regime membership. Normative considerations Norms are clearly important to regime formation.28 Richard Putnam demonstrates that international negotiations are by their very nature a two-­level process, with domestic issues and actors bringing influence to bear on the international level. Governments seek to maximize their ability to meet domestic pressures, while minimizing adverse consequences of foreign developments. However, he emphasizes that:

Regime design, implementation and effectiveness   11 the political complexities for the players in this two-­level game are staggering. Any key player at the international table who is dissatisfied with the outcome may upset the game board, and conversely, any leader who fails to satisfy his fellow players at the domestic table risks being evicted from his seat.29 While considering the specific factors that influence regime formation, it is therefore also important to recognize the importance of context and the influence of external pressures. Norms would seem to be central to how the two landmine regimes are positioned in relation to IHL. Individual regimes can be embedded in wider normative frameworks as with Ruggie’s well-­known example of embedded liberalism and post Second World War international trade institutions. Oran Young stresses the importance of these cognitive constructs because ‘international institutions cannot remain effective for long after the collapse of their intellectual underpinnings’.30 Overlapping normative or ideational frameworks can also generate spillover effects between linked issue areas. A norms-­based perspective distinguishes the mine ban treaty, which emerged rapidly through capitalizing on a norm bandwagon effect from the more drawn out military– technical approach of APII. The emergence of an anti-­APM norm seems to fit Finnemore and Sikkink’s description of a norm cascade with the support of states coalescing swiftly around the position that a ban was the only appropriate response to the suffering they cause.31 In considering the choices made by states balancing domestic and international interests, norms are clearly influential to how these two-­level games play out. External events may also be relevant. Although not directly linked to the regime formation process, the death of Princess Diana, just prior to the Oslo negotiations, further increased public pressure, at least within the UK, to support a humanitarian cause that was known to be close to her heart. The historical case studies on dum dum bullets and chemical gas share with APMs the feature that their use became widely acknowledged across the international community as unacceptable. Understanding how the concept of stigmatization influences regime effectiveness therefore links the analysis of APII and the mine ban treaty to the broader normative framework of IHL. Situating the landmine issue within the IHL tradition also points to potential effects of the landmine regimes beyond their stated objectives. In particular, it is important to determine what (if any) spillover effects, over time and in other issue areas, can be attributed to the existence of the regimes. Research into some regimes highlights a quality of resilience that derives from nesting within broader normative frameworks. On one level there may be a lack of apparent overlap between the two treaties. Yet both form part of a long history of international efforts under IHL to regulate or ban certain types of weapon. Normative considerations may thus provide a bridge between APII and the mine ban treaty even if such overlap and interplay dynamics are not apparent (or are denied by regime stakeholders).

12   Regime design, implementation and effectiveness Compliance with regime provisions may generate a number of ‘outputs’ from members that are intended to contribute to the specific ‘outcome’ the regime is established to achieve. Effects may also be felt in other issue areas. Drawing from experience in environmental regimes, Oran Young notes that: some regimes produce spillover effects by influencing relations among members in functional areas beyond their nominal scope. Regimes, especially those widely regarded as successful, can also generate demonstration effects by creating precedents that affect the thinking of both their own members and others as they confront new problems.32 Inevitably, the direct effects of regimes are easier to trace than indirect effects. Equally, it can be a question of perspective whether consequences are negative or positive. While recognizing the inherent complexities involved in tracing the effects of a specific regime, a distinction can be drawn between two types of regime consequence: those that affect state and inter-­state relations; and those that affect international society and transnational relations.33 Whether at the level of individuals or institutions, regimes can contribute new knowledge and alter perceptions. Through facilitating greater understanding of a particular subject area, policy changes can be effected that have long term consequences beyond the issue of the day. Fostering transnational links and encouraging common understandings among actors can affect how interests are calculated. This reinforces the importance of regime learning. Interactions may prompt new ways of thinking about specific problems as well as the ways members view each other.

Bridging implementation and effectiveness The literature on the landmines issue is partial, self-­interested and lacks systematic analytical critique. Additional conceptual clarity is essential in order to understand these complex, multi actor and multi-­level institutions. Related to this shortcoming is the need to apply a more process-­oriented perspective. Distinguishing the different phases of regime development – formation, implementation, effectiveness and consequences – provides a comprehensive framework within which to situate this analysis. Deconstructing APII and the mine ban treaty in this way thus enables us to move beyond a perceptions-­based approach and better understand these treaty regimes. An important caveat applies. Decisions made in one phase impact on others. If it is useful to distinguish these categories, their characteristics need to be understood as intrinsically linked and mutually reinforcing (in positive or negative terms). Many more articles and books have been written on the mine ban treaty in comparison to APII. Yet assessments tend to share an uncritically positive evaluation of the former at the expense of the latter.34 This reflects strong self-­interest, since the majority of these works are written by participant observers with a stake in the mine ban treaty. The debate is constrained by focusing primarily on  the political success of creating a new and ‘unprecedented’ international

Regime design, implementation and effectiveness   13 instrument banning APMs. Linking observance of regime rules to the realization of regime objectives is necessary in order to move beyond such limitations. This brings us to the relationship of the regimes to mine action. Since the underpinning objective of both treaties is to alleviate humanitarian suffering, the achievement of mine action-­related implementation goals must be a pre-­condition for landmine regime effectiveness. Through a regime theory framework, this book attempts to bridge related, but hitherto disconnected, discourses. One of these discourses focuses on the political evolution of APII and the mine ban treaty, while the other is heavily based on field experience and addresses the practice of mine action. Implementation histories that draw on civil society-­based monitoring are particularly useful in providing an alternative source of information on issues such as the interpretation of obligations, changing patterns of behaviour over time and evidence of regime defection. Since mine action as an activity pre-­dates the two regimes and many non-­regime members are also important contributors to this work, it is necessary to address counter-­factual questions on the relationship between landmine regimes and mine action. The subject explored in this book is divided into two sets of issues. A first theme relates to how the effectiveness of the landmine regimes is shaped by the relationship between design and implementation. This leads to a number of secondary research questions: • • • • • •

In what ways does regime design influence commitment to regime formation processes? How important is the involvement of implementation actors in regime design, and what can be learned from analysing stakeholder clustering dynamics in different phases of regime development? What consequences can be discerned from gaps between international commitments and domestic implementation processes under the regimes? What are the implications for effectiveness of how the regimes acknowledge and address technical and political dimensions of implementation? How significant are different approaches to rules in shaping the ability of the regimes to learn and adapt? Can disjunctions between regime and mine action priorities be discerned, and what are the resulting costs for regime effectiveness?

A second theme explores ways in which the design, implementation and effectiveness of international efforts to address landmines depend on the interplay between, and nesting of, treaty regimes within the broader IHL discourse. Related questions include: • •

How significant is the role of norms in determining regime effectiveness across design and implementation phases? In what ways does landmine regime evolution draw on longer historical processes of IHL regime development (in particular Hague Declaration 3

14   Regime design, implementation and effectiveness

• •

and the 1925 Gas Protocol) and what are the consequences for their effectiveness? Are normative considerations significant in generating ‘deep’ linkages between the two regimes (regardless of distinct frameworks and membership)? How does the concept of regime nesting within a broader IHL framework help us to understand regime dynamics?

Approach and structure There are a number of important gaps in our understanding of the design, implementation and effectiveness of APII and the mine ban treaty. In part, these gaps can be attributed to the complexity of the issues and the wide range of actors involved. An absence of systematic analysis in this field is problematic. But it also provides an opportunity to develop new insights into IHL more broadly and landmine treaties in particular, from a regime perspective. Subsequent chapters build on the framework and insights introduced in this chapter in order to develop a deeper analysis of the principles, norms, rules, procedures, issues and actors that have shaped the design and implementation of the landmine regimes. This research fills an evident gap by enabling a more nuanced understanding of how the interplay of design and implementation influences regime effectiveness. Chapter 2 considers two historical IHL regimes in order to better understand strengths and weaknesses in regime design and how this relates to regime effectiveness over time. The extent to which Hague Declaration 3 and the 1925 Gas Protocol have been effective in achieving specific and underlying objectives, provides a point of reference for the relatively recent processes of landmine regime implementation. An initial analysis of both Hague Declaration 3 and the 1925 Gas Protocol highlights the need to consider how factors such as the visibility of an issue area and rules relating to participation in processes of design and implementation determine the extent of political will or practical support for regimes. On one level, insights from a regime perspective on these historical treaties enable us to measure claims to uniqueness made by supporters of the landmine regimes against historical antecedents. On a more theoretical level, given the recent pedigree of the landmine regimes, analysis of mature regimes enables us to recognize normative dimensions that contribute to IHL regime effectiveness. Insights that point to the significance of nesting within wider normative constructs provide a nuanced set of issues that are further examined in subsequent chapters. In particular, despite the (deliberate) lack of interaction between the two landmine regimes, shared norms provide a means to consider how APII and the mine ban treaty interact with each other within a broader regime framework delineated by a norm against inhumane weapons. Chapter 3 focuses on the formation of APII and the mine ban treaty. Considering the different processes that led to the creation of the regimes sheds light on the objectives and motivations that underpin regime dynamics. Understanding

Regime design, implementation and effectiveness   15 processes of landmine regime formation forms the basis for exploring the ­relationship between design factors and the quality of implementation. Evaluating commonalities and distinctions between an established framework (APII as part of the CCW) in comparison with the ‘one off ’ mine ban treaty is one focus. The significance of rules is examined in determining the actors able to contribute directly to regime formation and also the flexibility given to shape regime frameworks. Specific sets of actors are associated with each regime, and the roles of different stakeholder groups – from advocacy to technical expertise – are considered in order to distinguish the engendering of political will from contributions to regime design. Distinguishing interest groups from epistemic communities with contextual or technical knowledge provides an important means to clarify how these design processes are shaped. Chapter 4 considers the implementation and effectiveness of the two landmine regimes. Analysis relating to design, agency and norms in regime formation is recalled in order to determine how these factors play out in implementation. The influence of practitioner communities and mine-­affected states in regime design can therefore be tested in relation to how complex, context-­specific implementation issues are addressed. Unpacking these challenges in their political and technical dimensions, through reconciling regime analysis with mine action best practice, is essential. ‘Carrots’ and ‘sticks’ to support or enforce compliance and avoid defection are therefore particularly important. Related to this point, experience from the wider IHL discourse is applied to examine the significance for treaty compliance of embedding international obligations in domestic law and practice. Chapters 5 and 6 consider the ways in which design and implementation dynamics play out in the areas of humanitarian demining and stockpile destruction. These activities represent central means for the regimes to deliver against their humanitarian objectives. Enhanced resource provision and technical support are two ways to link the influence of the regimes to these activities. Given that the core obligations in these areas rest with mine-­affected regime members, political will at the national level (and the ability of the regimes to reinforce this quality) are also important considerations. If key stakeholder groups do not have mine action expertise, this can lead to misperceptions over performance and an inability to adapt to new or evolving challenges. Moreover, the onus of demonstrating success may contribute to a reluctance to highlight problems. Analysis therefore touches on the ability and willingness of the regimes to set out and enforce realistic, appropriate obligations. In this respect, the absence of strong verification and monitoring provisions within either landmine regime raises questions over their ability to evolve, with evident consequences for effectiveness. These chapters also measure regime contributions to specific elements of the mine action agenda, against overarching regime goals, to address the human suffering caused by landmines. Landmine regimes represent an area of research that cannot be de-­linked from the policy implications of a priority issue on international security, humanitarian and development agendas. Given the stakes at play, the practical implications of

16   Regime design, implementation and effectiveness greater conceptual clarity may be substantial. The concluding chapter therefore draws together insights from across the previous chapters, that relate specifically to regime effectiveness, and considers them under the themes set out in this chapter: the interplay between design and implementation, key actors and normative considerations.

2 Regimes prohibiting the use in war of poison gas and dum dum bullets

Introduction If the landmine regimes are to be analysed as nested regimes, then the 1925 Gas Protocol and Hague Declaration 3 offer important parallels as part of wider efforts under IHL to regulate the conduct of armed conflict. The ban on dum dum bullets is the first codification of the customary international law principle prohibiting weapons causing unnecessary suffering,1 and the only treaty-­based prohibition on a conventional weapon in widespread use until the mine ban treaty nearly 100 years later. The 1925 Gas Protocol can trace its roots to Hague Declaration 2 from the 1899 Peace Conference, which prohibits the use of projectiles whose sole object is the diffusion of asphyxiating or deleterious gases.2 It provides a very early binding legal restraint on a weapon that had been widely condemned around the world.3 Considering commonalities and distinctions between the relatively ‘new’ landmine regimes and IHL antecedents enables us to better understand IHL regime effectiveness. A link between dum dum bullets and APMs may be found in the arguments deployed by those who debated the issues from military utility and humanitarian perspectives, and how the weight given to these different approaches fed into regime provisions. Like APII, Hague Declaration 3 and the 1925 Gas Protocol were considered in relation to military targets. The mine ban treaty is therefore unique in focusing directly on the impact of landmines on affected civilian populations during hostilities, whilst also addressing the post-­ conflict challenge. If normative considerations lie at the heart of IHL, the arms control expert Jozef Goldblat observes that: all laws of war suffer from one common weakness: the rules of conduct established for belligerents in time of peace may not resist the pressure of military expedience generated in the course of hostilities, and the attempts to ‘humanize’ war may sometimes prove futile.4 A corollary to this point is the argument that states only ban weapons which they have no intention of using.5 Assessing the effectiveness of Hague Declaration

18   Prohibition of poison gas and dum dum bullets 3 and the 1925 Gas Protocol, by considering how far outcomes are consistent with the prevailing political will and national security requirements of the states involved, informs two important sets of questions. First, it enables us to better understand the influence (and limitations) of stigmatization as a tool for behaviour change. Second, it further clarifies the two-­level dynamics of IHL regimes; international commitments need to be followed by national level implementation.

The development of Hague Declaration 3 and the 1925 Gas Protocol Origins and negotiation of the 1899 Hague Peace Conference Geoffrey Best points out that between 1750 and 1850 there had been little change in the science and technology of warfare. However, the latter part of the nineteenth century saw large standing armies become the norm as other European states followed the German example of introducing conscription. These armies were equipped with a number of new inventions with revolutionary implications for warfare. Metal-­hulled, screw-­propelled warships, smokeless powder and machine guns, in different ways, meant that ‘both the means and the measure of destruction and killing became far greater than they had ever been before’.6 Between 1885 and 1895, the rate of fire of a rifleman had increased from three to sixteen shots per minute, while the range of artillery had grown from one to 6,000 yards.7 Technological developments therefore provide an important contextual dynamic in order to understand the roots of regime development. The 1899 Hague Peace Conference was called by the Tsar of Russia with the overt goal of slowing the mounting cycle of military expenditure among the Great Powers. The objective of the Hague Peace Conference was not disarmament, but as described by Russian Foreign Minister Count Mouravieff in his circular of 11 January 1899: an understanding stipulating the non-­increase, for a definite period, of the present effective military and naval forces, and also of the military budgets pertaining to them; and a preliminary investigation of the means by which even a reduction of these forces may be secured in the future.8 The initiative was described as a response to the ‘armed peace’9 that characterized contemporary Europe with states investing ever-­increasing amounts in new and improved armaments. The Tsar’s manifesto condemned the system of ‘armaments a l’outrance transforming the armed peace into a crushing burden that weighs on all nations and if prolonged will lead inevitably to the very cataclysm which it is desired to avert’.10 The point of departure for the conference thus reflected economic and ‘hard’ security interests as well as humanitarian considerations.

Prohibition of poison gas and dum dum bullets   19 The Hague – the capital of a small, neutral country – was carefully selected as the site of the conference. Opening on 18 May 1899, twenty-­six predominantly European countries attended. None of the participating states had high expectations going into the conference. Indeed, many were cynical as to Russia’s motives; well aware that, humanitarian goals notwithstanding, the Tsar’s initiative had to be set next to the fact that Russia was lagging behind in the contemporary arms race. The event was therefore characterized by what Inis Claude terms ‘multilateral insincerity’.11 States’ participation in the Russian initiative reflected the requirements of a two-­level game: to do otherwise would be diplomatically embarrassing, and run contrary to domestic and international public opinion. However, the underlying security, political and economic interests of participating states ran counter to the overt goals of the conference. The conference was divided into three Commissions: on Armaments, on the Laws of War, and on Arbitration, which in turn divided into sub-­committees. The use of sub-­committees and chairmen, as well as majority voting rather than unanimity, was innovative for the time. This approach was to be adopted and further developed in future institutional procedures for international negotiations.12 However, the results of the 1899 Hague Conference reflected the minimalist expectations of the participants. The conference produced three Conventions: on Arbitration, Laws and Customs of War on Land, and Extension of the Geneva Rules to Maritime Warfare; three Declarations on Projectiles from Balloons, Asphyxiating Gases and Expanding Bullets; six ‘Wishes’ for future endeavours, and a Resolution recalling the desirability of limiting expenditures for armaments and new types of weapon. In concrete terms this was a realistic, if much more limited, return than aspired to by the Russians. Delegates were aware that although a failure to agree reductions in expenditures on armaments was always the likely result; nevertheless there would be significant public disappointment at such an outcome. Baron de Bildt of Sweden and Norway emphasized to delegates that: when the results of our deliberations shall become known, there will arise, notwithstanding all that has been done for arbitration, the Red Cross etc., a great cry: ‘it is not enough!’ And this cry, ‘it is not enough’, most of us must conscientiously acknowledge to be just.13 Although the proposal put forward by Switzerland to prohibit ‘the use of projectiles which aggravate wounds and increase suffering, such, for example, as “dum dum” bullets’14 was not on the original conference agenda, it was readily seized upon as an achievable objective after the failure of states to agree concrete limitations on the headline conference goals. Despite the underpinning realpolitik interests of many participating states, the need to demonstrate progress in humanitarian terms proved compelling in terms of agenda setting and thus provided the opening for a focus on exploding bullets. Bullets that expand and flatten in the human body were first mass produced at a British Indian arsenal in Dum Dum, near Calcutta,15 thus providing the more

20   Prohibition of poison gas and dum dum bullets colloquial generic title of ‘dum dums’ for a shell which ‘had the capability of spreading out on entering the target body and of inflicting a much bigger wound than did the normal hard-­nosed high-­velocity rifle bullet’.16 Discussion in the relevant sub-­committee began by focusing on the design of dum dum bullets. General Poortugael, representing the Netherlands, supported the Swiss initiative, making public his Government’s instruction that he should pursue a formal prohibition on these rounds. He characterized them as inhuman projectiles which, by design, cause incurable wounds that go beyond what is necessary to make soldiers incapable of fighting. General Sir John Ardagh, for Great Britain, disputed this description, comparing dum dum bullets to any other kind of projectile. A number of more or less specific formulations were put forward by delegates with the sub-­committee finally adopting the following wording: ‘The contracting Powers prohibit the use of bullets which expand or flatten easily in the human body, such as bullets with hard jackets, whose jacket does not entirely cover the core or has incisions in it.’17 When the proposition was put to the vote, nineteen out of twenty delegations present voted in favour and only Great Britain voted against. General Ardagh, for Great Britain, in justifying his negative vote, put forward the unedifying argument that bullets with greater stopping power – such as dum dums – were required in battle against ‘savages’ who continued to fight when wounded by smaller calibre rounds. This distinction was rejected by the sub-­committee and the original proposition was accepted by twenty-­two delegations with only the US and Great Britain voting against, while Portugal abstained.18 The majority of states present were thus happy to form a coalition on the ‘right side’ of this issue. In practice, the British magnified the norm bandwagon effect by combining a national policy position that was on the wrong side of a humanitarian issue with an accompanying message that was insensitive in the extreme. Origins and negotiation of the 1925 Gas Protocol Custom and usage stretching as far back as ancient times forbids the use of poison in battle. Hague Declaration 2 was the first codification of this principle, prohibiting projectiles designed to diffuse asphyxiating or deleterious gases. However, gas warfare rose to the top of public and political agendas due to the widespread use of these weapons during the First World War. Throughout the course of the war, 6,000 tonnes of lachrymators and 7,000 tonnes of respiratory irritant gases were used.19 John Keegan describes how ‘gas in a variety of forms, the more deadly asphyxiant phosgene, and the blistering “mustard”, would continue in use throughout the war, and chlorine would kill thousands of Russian troops in German offensives west of Warsaw’.20 Gas warfare therefore fused a normative imperative that can be traced back to the earliest accounts of warfare, with high levels of contemporary concern. This provided a potent mix, combining the same fears over military–technological progress, that provided impetus to the Hague negotiations, with the psychological and physical scars of ‘the war to end all wars’.

Prohibition of poison gas and dum dum bullets   21 The Treaty of Versailles, and the other treaties that delineated the end of the First World War, codified a ban on possession as well as use of these weapons by the defeated powers.21 These measures constitute neither a multilateral arms control nor an IHL regime, but a series of conditions applied to the defeated by the victors. The 1925 Gas Protocol, negotiated over a period of less than eight weeks in 1925, was another significant development. The protocol, which provides for a general prohibition on ‘the use in war of asphyxiating, poisonous or other gases, and of all analogous liquids, materials or devices’,22 marks a clear break from victor’s justice and a return to negotiations between states under international law. As an attempt to regulate the arms trade, the 1925 International Conference for the Control of the International Trade in Arms, Munitions, and Implements of War, convened by the Council of the League of Nations from 4 May to 17 June 1925 was, like the first Hague Peace Conference, guided by economic and hard security imperatives. International regulation was no more feasible then, than it had been in 1899. In another parallel, the inclusion of poison gas on the conference agenda had not been planned, but was added at the behest of one of the delegations – the United States – in order that the conference achieve (and be seen to achieve) a meaningful outcome in such a high-­profile area of concern. The need to respond to domestic and international humanitarian constituencies was keenly felt. The document initially considered by delegations in Geneva dealt solely with the trade in conventional weapons. The US representative proposed an amendment prohibiting the trade in chemical weapons. This initial emphasis represents the balance sought by the US who wanted to be ‘the moral leader in the worldwide quest for security through disarmament’ while also seeking ‘a modicum of insurance’23 in case deterrence failed. During negotiations, although the US amendment was widely supported, delegations acknowledged the challenges of implementing a ban on trade when it was almost impossible to distinguish between chemicals used for civilian and military purposes. The difficulty of war readiness for non-­gas producers was also recognized. These discussions highlight the underpinning security dynamics that had to be balanced with economic and humanitarian concerns in shaping the regime design process. Multiple interests presented participants with a version of the Prisoners’ Dilemma. Support for the common good of achieving a complete ban was conditioned by concern over the potential inability to react preventatively to the development of chemical weapons by another state. The US proposed that these issues be devolved for discussion in technical committees. However, two central problems emerged when they were considered in detail: 1

The difficulties posed by dual-­use technologies became swiftly apparent. Expert opinion at the conference was unanimous, that nearly all materials used in chemical weapons were to be found in non-­military industrial products and processes. As Zanders notes, ‘the inability to distinguish unambiguously

22   Prohibition of poison gas and dum dum bullets

2

between chemicals used as warfare agents and those that have peaceful industrial purposes rendered any ban on their trade or transfer impractical because of the impossibility of verifying the end use of the recipient state’.24 The principle of equal treatment among states, in the spirit of the Covenant of the League of Nations, was highlighted by many delegations as a guiding principle for their deliberations. There was a recognition that a ban on trade was discriminatory, freezing and legitimizing the superiority of those states that already had an advanced chemical industry and would therefore be unaffected by the ban.

Unable to find a compromise between prohibition and unrestricted use, the US withdrew the reference to trade in their text. Instead, a complete prohibition was proposed on the use of poison gas in war. The prohibition was further enlarged due to the acceptance of a proposal by the Polish delegation such that ‘any decisions taken by the conference concerning the materials used for chemical warfare should apply equally to the materials employed for bacteriological warfare’.25 As in The Hague some twenty-­five years before, technological development was an influential factor. The humanitarian effects of recently developed weapons provided a backdrop to discussions that left no doubt as to the right side of the gas warfare issue. Moreover, the challenge of distinguishing civilian from military use of chemicals, clearly demonstrated by technical expert communities, proved influential in moving negotiations beyond the initial focus on trade.26

The interplay between design and implementation Relating regime design to implementation and effectiveness During negotiations, British and US delegates argued that the definition of expanding bullets provided in the declaration contained too much technical detail. However, all the other delegations present recognized that clarity was essential. A prohibition based on design characteristics was necessary to prevent states finding loopholes in their obligations. Linking the ban to more general principles of IHL would have given ample opportunity for applying different interpretations. Instead, Hague Declaration 3 is a prohibition based on a particular technical specification: the construction of bullets. The appropriateness of this design-­based prohibition has been demonstrated over time. The declaration has proved very successful in eliminating both the manufacture and use by states of this type of ammunition for military purposes. Rapid acceptance on a political level, and the absence of any known violations of the prohibition by regime members are evidence of the declaration’s effectiveness.27 However, the wording of the declaration reflects the nature of firearms and ammunition at the end of the nineteenth century, and the understanding of wound ballistics at that time. It has been argued that this wording is no longer adequate to achieve the regime’s objectives:

Prohibition of poison gas and dum dum bullets   23 given the variety of ways in which bullets are now constructed, a modern understanding of wound ballistics and recognition that other factors such as bullet velocity are also responsible for the degree of injury and suffering from rifles and handguns.28 Drawing on wound ballistics research, Robin Coupland and Dominique Loye argue that certain full metal jacket bullets (standard military issue) can cause wounds similar to dum dum rounds. Other ‘field’ factors such as ricochet, range and the condition of the weapon can also generate a variety of wound effects: ‘in brief, bullet construction is only one of the factors which lead to large wounds’.29 By basing the prohibition on a technical characteristic, the regime has lacked the flexibility to adapt to developments in firearms and ammunition since 1899. Coupland and Loye note that ‘it has become evident that adhering to the strict wording of the declaration does not always achieve its apparent objective and purpose, that is, to eliminate the unnecessary injury and suffering associated with very large bullet wounds’.30 A related consideration links technological development to changing practice in the conduct of armed conflict. Evolving weapons technology has had profound effects on military strategy and tactics. The arguments put forward by Ardagh against the wording of the declaration relate to the use of single shot rifles, at short range, to stop a charging enemy. But the development of automatic weapons, and the integration of infantry with armour and artillery on the battlefield have made both these arguments and the regime itself less relevant in practice than in 1899. A technical definition has ‘frozen’ the regime in binding it to the technology and military practice of the time. This does not limit the regime’s initial effectiveness but does constrain its ability to evolve, explaining why regime membership has not grown significantly over time beyond its initial adherents.31 The 1925 Gas Protocol also faces a number of challenges as a result of temporal considerations. The protocol restricts its non-­use obligation to states parties that are at war. This means that its provisions do not apply in the case of the internal armed conflicts that have become increasingly prevalent in the years since the protocol entered into force. In fact, as Jozef Goldblat suggests, ‘it can also be argued that the Protocol does not cover those international conflicts, in which the belligerents do not consider themselves to be formally at war’.32 Moreover, it does not prevent research, manufacture, stockpiling, transfer or training in the use of these weapons.33 There are thus a number of loopholes that members may exploit that are consistent with regime rules, but which undermine regime goals. Ambiguous wording leaves considerable room for diverging interpretations. For example, states have taken very different positions on whether tear gas and other non-­lethal gases fall within the protocol’s scope. In 1930 the British clarified a ‘serious ambiguity’ between the English and French language versions of the treaty. In legal terms, both versions were equally valid but the English version referred to ‘asphyxiating, poisonous or other gases’, while the French version had ‘similaires’ in place of ‘other’. The UK, although later

24   Prohibition of poison gas and dum dum bullets changing its position, stated its understanding that lachrymatory (tear) gases were included and the French subsequently confirmed this.34 Regime compliance has been mixed. Allegations of voluntary defection can be found in several cases. Egyptian forces were accused of using poison gas in the Yemen during the 1960s. This issue was discussed in the United Nations General Assembly (UNGA) on 5 December 1966, and led to a resolution calling for observance of the ‘principles and objectives’ of the 1925 Gas Protocol.35 In 1982, the US government alleged the use of chemical weapons in Laos, Cambodia and Afghanistan by the Soviet Union and its allies. Both sides also used chemical weapons during the Iran–Iraq war, while Iraq gained worldwide notoriety for its use of chemical weapons against its own Kurdish minority population with the attack on Halabja, in Northern Iraq, in March 1988.36 Examples of regime defection demonstrate the importance of compliance verification. While Hague Declaration 3 was agreed without verification provisions it may be argued that this is unnecessary since, in terms of manufactured production, the declaration has been almost universally implemented.37 However, in contrast to chemical weapons, which require significant technical and scientific skill to develop and deploy, expanding bullets represent a much simpler technology. Dum dum rounds can be easily improvised prior to battle through cutting an ‘X’ into the tip of a regular small calibre round38 which then expands on impact, causing similarly large wounds to rounds designed for that purpose. This would seem to reinforce Goldblat’s point about rules made in peacetime governing the conduct of conflict, a view also expressed in 1899 by delegates who feared that the protocol’s provisions might not withstand the strain of hostilities.39 Similarly, the 1925 Gas Protocol does not contain provisions for monitoring or verification of treaty compliance. In order to address this lacuna, UNGA Resolution 37/98 of 13 December 1982 requested the Secretary-­General ‘to investigate, with the assistance of qualified experts, information that may be brought to his attention by any Member State concerning activities that may constitute a violation of the protocol or the relevant rules of customary international law’.40 The resulting report on chemical and bacteriological (biological) weapons, including provisional procedures, was presented to the General Assembly in 1984.41 A UN Mission, investigating allegations of the use of chemical weapons in the Iran–Iraq war, issued seven reports between March 1984 and August 1988. On 21 March 1986, a United Nations Security Council Statement criticized Iraq for the ‘use of chemical weapons, in clear violation of the Geneva Protocol of 1925’, while on 26 August of the same year, the Security Council unanimously adopted Resolution 620 condemning ‘the use of chemical weapons in the war between Iran and Iraq’.42 Further UN Missions in 1992 investigated alleged use of chemical weapons in Azerbaijan and Mozambique. The UNGA initiative thus went some way to addressing the absence of any mechanism to verify compliance among states parties. The various UN missions and resulting Security Council Resolutions demonstrate the mutually reinforcing relationship between the 1925 Gas Protocol

Prohibition of poison gas and dum dum bullets   25 regime and the wider interests of international organizations. The UN is fulfilling its mission to safeguard international peace and security by providing a de facto verification mechanism. But it could not have stepped into this role without the regime’s existence, membership and credibility. These measures have in turn been superseded by the much more detailed verification system established under the CWC with the creation of the Organization for the Prohibition of Chemical Weapons (OPCW). Overlapping and interlocking regimes have gained strength through being able to evolve over time. Despite its lack of flexibility, the 1925 Gas Protocol has continued to attract new members. The large number of states bound by the protocol,43 various statements of the UNGA and the Security Council in support of its provisions, and the many references found in other IHL treaties, explain why the 1925 Gas Protocol is widely regarded as forming part of customary international law. If this issue has been clouded by interpretations and reservations applied by some states, Guelff and Roberts acknowledge that ‘at least the first use of lethal chemical and biological weapons is prohibited by customary international law. Less consensus exists on the status under customary international law of non-­lethal chemical weapons.’44 Spillover effects The continued reference to both Hague Declaration 3 and the 1925 Gas Protocol in later treaties is a significant example of IHL gaining strength from the mutually reinforcing nature of its component treaties. Reference to both regimes in the Statute of the International Criminal Court (ICC), defining the use of expanding bullets and chemical weapons as war crimes, demonstrates that the two regimes remain relevant. These linkages are particularly important in the case of chemical weapons which, as weapons of mass destruction (WMD), have been recognized by the international community as providing a greater threat to peace and stability than any conventional weapon. The contemporary significance may be less evident in the case of Hague Declaration 3, which has been to an extent superseded by developments in weapons technology and military tactics. This suggests that technological progress and changing military practice may be important considerations for the landmine regimes. There is continued reference to Hague Declaration 3, in relation to weapons, in which there is a fine line between military necessity and humanitarian consequences. So, while on one level the regime may seem to have diminished currency, it remains relevant through its nesting within a broader IHL framework. The final declaration of the January 1989 Paris Conference on the Prohibition of Chemical Weapons,45 convened following concerns about the use of chemical weapons during the Iran–Iraq war, acknowledged the importance and continued validity of the 1925 Gas Protocol. The declaration also stressed the need to conclude an international legal instrument on the prohibition of the development, production, stockpiling and use of chemical weapons.46 The resulting CWC was signed in 1993 and entered into force in 1997.47 It provides for a complete

26   Prohibition of poison gas and dum dum bullets prohibition on chemical weapons without the possibility of ‘no first use’ reservations. It also applies to internal armed conflicts. The 1925 Gas Protocol has therefore been augmented by the broader prohibition on production and possession found in the chemical and biological weapons conventions. The mutually reinforcing nature of these treaties is demonstrated by the fact that several states withdrew reservations to the 1925 Gas Protocol following adherence to the Biological and Chemical Weapons Conventions. Moreover, the CWC contains preambular clauses and articles reaffirming the principles, objectives and obligations of the protocol. It is thus an important antecedent to these broader and deeper measures, providing for the international de-­legitimization of chemical weapons. This was an essential precursor to both practical disarmament measures and the consolidation of a norm against their use. In this sense, the broader regime focusing on chemical weapons has adapted and evolved, demonstrating an ability to learn over time and exploit increased political will in order to address more comprehensively the dangers posed by these weapons. Equally, Hague Declaration 3 can be situated within a broader regime framework aimed at restricting or prohibiting the use of weapons causing superfluous injury or unnecessary suffering. This is evident through demonstration effects on later weapons. Arguments have been put forward that high velocity ‘tumbling’ rounds, which turn end over end when striking a soft target, have a similar effect to dum dums. Although this ammunition is not specifically banned, concerns have been justified by analogy to Hague Declaration 3. A draft protocol on small calibre weapon systems was submitted to the United Nations conference that led to the adoption of the CCW in 1980. In order to conform to the letter and spirit of Hague Declaration 3, a resolution was adopted encouraging states to conduct further research into new types of bullets, particularly those which turn once entering the human body.48 Subsequently, another draft protocol was presented at the 1995 CCW Review Conference calling for additional testing on the effects of these weapons. Although this proposal, as with the earlier initiative, has not led to new IHL, the issue retains a level of currency through its association with Hague Declaration 3. One demonstration effect common to both regimes is that they have been infused with meaning beyond their specific provisions. Although the draft protocol on small calibre weapons was ultimately rejected, considerable research has since taken place on energy transferral and bullet construction, with subsequent steps to improve bullet design. In particular, Eric Prokosch claims that the Belgian-­designed NATO standard 5.56 mm round was designed and selected on the basis of criteria which sought to reduce unnecessary suffering.49 Debates on this question have generated familiar arguments as to whether the design or effects of a weapon should be the telling factor in determining its legality.50 On a broader level, the ICRC has launched the Superfluous Injury or Unnecessary Suffering (SIrUS) Project. The project seeks to develop an objective definition of these terms – by applying the epidemiology of the effects of weapons to IHL – and therefore bridge a gap between current IHL and advances in weapons technology.51

Prohibition of poison gas and dum dum bullets   27 Spillover effects relating to nuclear weapons can be attributed to the 1925 Gas Protocol. It has been argued that the effects of nuclear weapons imply that their use is prohibited by the terms of the regime as well as by customary principles prohibiting weapons causing unnecessary suffering. This position was ultimately rejected in the 1996 Advisory Opinion of the International Court of Justice (ICJ) on the Legality of the Threat or Use of Nuclear Weapons. The intentional design of a weapon was determined to be the defining factor in whether a weapon should be prohibited, rejecting the effects argument that the secondary asphyxiating and poisoning caused by nuclear weapons rendered them illegal. Thus, ‘if the asphyxiating or poisoning effect is merely a side-­effect of a physical mechanism intended principally to cause totally different results (as e.g. the use of nuclear weapons), then the relevant munition does not constitute a “poisonous gas.” ’52 Regardless of the outcome, it is significant that the regime contributed to a debate on nuclear weapons and IHL that needed to be resolved through a landmark ICJ judgement. The nexus between de-­legitimization and political will is particularly important in relation to the broader questions posed in this book. Analysis of the 1925 Gas Protocol demonstrates how powerful effects can be exerted if a weapon becomes stigmatized, including beyond the regime’s membership. Furthermore, the association of specific regimes with customary principles of IHL creates a mutually reinforcing relationship where the existence of a regime strengthens the underpinning norm, while encouraging states to adopt compliant behaviour. This relationship also works against regimes becoming irrelevant, since long term goals can be re-­considered again and again in different fora.

Key actors The Great Powers and the 1899 Hague Peace Conference The states attending the 1899 Hague Peace Conference were conscious of the need to play to the concerns of the Peace Movement without sacrificing the requirements of military necessity. Not long before the conference, China and Japan, Turkey and Greece, Spain and the US, had all been at war. Tensions were further raised by the Dutch hosts who, as supporters of the Boers, had demanded invitations for the Transvaal and Orange Free State despite a looming conflict with Great Britain. Barely masked realpolitik concerns under ostensible humanitarian imperatives are highlighted in a private communication in which the US stressed the need for an international court of arbitration to be seen to be created, while reassuring the Germans that ‘the purely voluntary character of the jurisdiction proposed . . . should be so clearly emphasized that the very last trace of any compulsion, moral or otherwise, upon any nation, be it great or small, should disappear.’53 Thus, while normative considerations and the public profile of the issue generated a significant convening power, an effective IHL regime was not the goal for many participating states. Because initiatives from the Tsars led to the 1868, 1874, 1899 and 1907 peace conferences, Russia stood out, at least in the public eye, as the state most

28   Prohibition of poison gas and dum dum bullets c­ oncerned with peace, disarmament and the laws of war. For the other Powers, this was a case of ‘the voice being so often the voice of justice and peace, the Realpolitik being forever suspect as that of unregenerate imperialist and militarist bear’.54 Whatever the motives, Russia was alone among the Powers to push for reductions in armaments. This is significant because despite the high levels of multilateral insincerity, providing a framework for negotiations exposed states to a high profile, two-­level game. The intertwining of domestic and international concerns left open the possibility of becoming locked into a process leading to new international obligations. Germany, the ‘modern Sparta’55 built by Chancellor Bismarck, was the driving force behind the contemporary arms race, and it is perhaps no coincidence that the Tsar’s call for a peace conference first came less than a month after Bismarck’s death on 30 July 1898. However, Kaiser Wilhelm II (who had dismissed Bismarck in 1890) greeted the Tsar’s initiative with disdain: ‘what will Krupp pay his workers with?’56 Taking a cue from the Kaiser, German rejection of the notion that Europe was beset by the malaise of armed peace, was clearly stated by the German military representative Colonel von Schwarzhoff at the outset of the conference: The German people are not crushed beneath the weight of expenditures and taxes; they are not hanging on the edge of a precipice; they are not hastening towards exhaustion and ruin. Quite the contrary: public and private wealth is increasing; the general welfare and standard of life are rising from year to year. As for compulsory military service, which is intimately associated with these questions, the German does not regard it as a heavy burden, but as a sacred and patriotic duty, to the performance of which he owes his existence, his prosperity, his future.57 This statement was to be typical of German interventions during the conference. Whereas all the Powers viewed the Russian proposals in more or less the same light, the Germans were less concerned about being cast in the role of conference spoilers. Unlike the Germans, the British Government recognized the balancing act required at The Hague. Mindful of the 750 favourable resolutions received by the Foreign Office from public groups in the four months following the publication of the Tsar’s manifesto,58 they did not want to be seen to be undermining the conference and its humanitarian goals. The War Minister, Lord Landsdowne, noted in his instructions to Ardagh that: You have a very difficult hand to play. It is clear that the conference has availed itself with the avidity of at once achieving something in the interests of humanity and gibbeting us as the inhuman power of the age.59 Ardagh attempted to justify the British position as the sole opponent to the declaration on dum dum bullets in the sub-­committee. His main argument was that ‘the use of these words describing technical details of construction will

Prohibition of poison gas and dum dum bullets   29 result in making the prohibition a little too general and absolute’.60 Ardagh argued that recent conflicts had shown that the fully jacketed bullet of the British Lee-­Metford rifle did not have the stopping power to put an enemy hors de combat. The wounds caused by the dum dum bullet have this necessary effect ‘but their result is by no means designed with the aim of inflicting useless suffering’.61 He therefore argued for: a phraseology which shall leave aside technical details of construction and affirm the principles on which we are all agreed . . . that is to say, the prohibition of the use of bullets whose effect is to aggravate uselessly the sufferings of men placed hors de combat, or to render their death inevitable.62 The US, while also opposed to the Russian proposals on armaments, was keen to distance itself from being seen to meddle in European affairs. Captain Mahan, their naval representative, noted that the US position: is not meant to indicate mere indifference to a difficult problem, because it does not affect the United States immediately, but expresses a determination to refrain from enunciating opinions on matters which, as concerning Europe alone, the United States has no claim to enter.63 Captain Mahan went on to argue during the conference that: the military and naval armaments of the US are at present so small, relatively to the extent of territory and to the number of the population, as well as in comparison with those of other nations, that their size can entail no additional burden of expense upon the latter, nor can even form a subject for profitable mutual discussion.64 Despite this rhetoric, the US was about to use dum dum rounds in the Philippines. They therefore supported the British position in arguing for a more general prohibition on bullets that, in effect, exceeded the limits needed for putting a combatant hors de combat. Ultimately, British and US proposals were rejected by the other delegations. The rationale was that a non-­definition-based prohibition would be too vague to be effective. While the British–US position was consistent in seeking to balance military necessity with humanitarian concerns, it was recognized among other participating states that such language would not be acceptable to the Peace Movement. This demonstrates the power of binding states within a process of IHL regime formation. While a prohibition was not on the table at the beginning of the negotiating conference, the convening of the different states and pressure to demonstrate progress meant that this solution was acceptable for the majority of participants. The humanitarian imperative was further strengthened by the reality that only Great Britain and the US had an immediate military use for these rounds. It was therefore cost free for others to (be seen to) take the moral high ground.

30   Prohibition of poison gas and dum dum bullets Civil society and the 1899 Hague Peace Conference At the end of the nineteenth century, a recognized Peace Movement had emerged, popularly associated with support from Quakers, Mennonites and other radical protestant groups. The base of the movement was very broad, attracting many who made the link between industrial progress and peace. As Barbara Tuchman describes, ‘science made all phenomena seem subject to certitudes and laws, and if man’s physical world could be understood and controlled, why not his social relations also?’65 Captains of industry provided significant patronage, the best-­known examples being Nobel’s creation of the Peace Prize in 1897 and the Carnegie Endowment for International Peace, which published the Proceedings of the 1899 Hague Peace Conference ‘to seek the most effective means of ensuring to the peoples a lasting peace, and of limiting the progressive developments of military armaments’.66 State and non-­state actors clustered together, motivated by common interests. This mutually reinforcing relationship offered a powerful, tripartite combination of (state) convening power, the financial leverage of the business community and compelling civil society advocacy. Geoffrey Best situates the Peace Movement as related to, but distinct from, a more diverse Internationalist Movement that: included every kind of informal organization and arrangement, formal institution, agreement and treaty which could promise to bring different nations, and people within different nations, together more easily, and to promote whatever interests they might have in common; peace not least, and the peaceful resolution of international disputes.67 Multilateralism lay at the heart of this movement, with the late nineteenth century witnessing a growth in international regimes covering, for example, the mail, telephones and public health. The Young Men’s Christian Association (1844), Red Cross (1863), Institute of International Law (1873) and the Inter-­ Parliamentary Union (1888) were all founded during this period. Indeed, it is suggested that one of the inspirations behind the Hague Conference was the reporting of the Russian Consul at Budapest, who described an 1897 meeting of the Inter-­Parliamentary Union (IPU) in that city which discussed the possibility of reductions in armaments by international agreement.68 For the Peace Movement, the Tsar’s initiative was seized on with great enthusiasm. The journalist William T. Stead, founder and editor of the Review of Reviews, campaigned around Europe in favour of the Tsar’s proposals and established a new weekly, War on War, to strengthen public demand for the peace conference.69 Sessions of the conference were closed, excluding the press and other observers not linked to official delegations. Stead, through exploiting his numerous personal connections, nevertheless published a daily chronicle of the conference in Dagblad, The Hague’s leading newspaper. It was read by the majority of delegates, heavily quoted by other correspondents and relayed to home societies by supporters. The media was therefore a self-­conscious force

Prohibition of poison gas and dum dum bullets   31 multiplier for the humanitarian lobby. The conference belatedly accepted transparency as a fait accompli and opened up to the press, acknowledging the power of civil society and, as a consequence, further increasing public scrutiny of delegates’ endeavours. A range of observers influenced delegates’ behaviour. Correspondents and representatives of the Peace Movement lobbied in favour of specific objectives, raising petitions around Europe and beyond. Despite the bellicosity of the official German position, a Committee of Reichstag Deputies, professors and writers descended on The Hague to support the conference.70 Public attention increasing the heat on delegates to achieve meaningful results was particularly evident in the pressure applied through the press for the establishment of a Permanent Court of Arbitration. Germany was the last of the Powers to resist the idea of arbitration, but even they were eventually forced to agree to the proposal, in spite of the strong reservations of the Kaiser. Their agreement to the prohibition of expanding bullets must be seen in this same light. The de facto opening up of the regime design process demonstrates that strong public interest, coupled with an effective advocacy campaign, can overrule formal rules of procedure. The potent combination of an international movement and committed domestic constituencies is particularly significant. To take the most conspicuous example, Germany would never have voluntarily agreed to a ban on dum dums. The negotiating environment, in particular the effect of channelling domestic and international advocacy, fundamentally changed the political calculus of costs and benefits in its favour. The Great Powers and the 1925 Gas Protocol The issue of poison gas presented a policy dilemma for US decision makers. In public the US delegation to the Geneva conference clearly responded to the humanitarian imperative, prefacing their proposals with the goal of ‘lessening the horrors of war and of ameliorating the sufferings of humanity incident thereto’.71 Behind the public statements there were major splits within the delegation – as within the Administration – over policy on chemical warfare. The War Department was opposed to any prohibition other than on its use against cities and non-­combatants. Although official statements did not reflect this position, the War and Navy Department representatives were actively attempting to undermine the proposals of their own delegation.72 While not evident in Geneva, where the US signed the protocol at the end of the conference, splits in the US position were to prove decisive when the focus switched to implementation, and Senate consent for ratification was sought. Resistance from the War Department was compounded by a lack of coordination by pro-­regime factions when dealing with the Senate or the chemical industry. Influential opponents in the military argued against adherence on the grounds of military preparedness. With the support of sympathetic senators, resistance was reinforced by the fact that the War Department had not been consulted until after the US had committed itself to the regime. In contrast, supporters in the US were

32   Prohibition of poison gas and dum dum bullets conspicuously quiet. The protocol seemed to represent a broader restatement of the terms of the Washington Treaty, which had been ratified without dissent. Lulled by this, as well as by favourable reporting from the Foreign Relations Committee, ‘the peace groups forfeited the opportunity to influence the most significant debate on chemical warfare during the interwar period’.73 Ratification was rejected by the Senate. Two arguments can be identified as particularly persuasive in undermining the ratification process: 1 2

The US should not sign up to international obligations with a ‘mental reservation’74 that would be ignored in time of overriding military necessity; Ratification would preclude readiness for gas warfare since resources could not be allocated for this purpose when the use of gas in war had been prohibited.75

The Senate considered the 1925 Geneva Protocol from a national security (rather than humanitarian) perspective. Its rejection was not through any sense that gas warfare was desirable, but because the view was taken that gas weapons would probably be used in future conflicts and the US had to be prepared for that eventuality. As Brown points out, rejection of the protocol (which the US eventually ratified only in 1975) marked an important shift in US policy with preparedness for gas warfare taking priority over international efforts towards prohibition.76 This unilateralist turn reflected the views of only one strand of the US negotiating team. However, the absence of a bridge between international policy formulation and national policy adoption was crucial in shaping the long term US policy stance. Many states, including France and Great Britain, became parties to the 1925 Gas Protocol, subject to the reservation that it is binding only as long as other states are bound by it. The protocol is therefore viewed by some adherents not as an absolute prohibition but as an obligation of ‘no first use’. These reservations clearly weaken the protocol. Yet it could be argued that they represent a small price to pay if they permit states to adhere to the regime. The primary motive behind the Soviet Union’s reservation was that two major potential enemies – Japan and the United States – had not ratified the protocol.77 The Soviet Union also reserved the right to use gas against non-­signatories (hence alleged Soviet use of gas in Afghanistan during the 1980s would not have constituted non-­ compliance). Reservations to regime membership therefore provided a means to stabilize relations between states, but also came with associated humanitarian costs. In the years preceding the development of the regime, European powers had adopted ambivalent approaches to chemical weapons. France, in 1921, had allegedly provided the Spanish in Morocco with a filling capacity for chemical weapons to put down the Riffian Berber rebellion. State of the art chemical agents were subsequently also provided by Germany, in violation of the Versailles Treaty.78 Great Britain and France turned a blind eye to this collaboration, because of their own interests in shaping the outcome with its implications for

Prohibition of poison gas and dum dum bullets   33 control of the Straits of Gibraltar. The Geneva conference took place in the midst of the uprising at a time when the Spanish made widespread use of chemical bombs.79 Resistance to a ban on trade in chemical weapons by certain European powers would therefore seem to be a logical consequence of national security interests, rather than reflecting the practical difficulties of implementing such a prohibition.80 The 1925 Gas Protocol failed to bind the US because it was not sufficiently embedded within a broader IHL framework. The lack of normative ‘push’ factors meant that the US policy position did not come with costs at domestic or international levels. The reservations tabled by some regime members also display an at best conditional commitment to prohibition – born out by the mixed implementation history of some states parties. However, the policies of the European Powers were to evolve very differently to those of the United States. In part, this can be attributed to factors exogenous to the regime. The rise of totalitarian governments, coupled with technological advances in air power, presented a threat within Europe that, for geopolitical reasons, the US did not face.81 Thus, while the US position hardened in favour of chemical warfare readiness, the European Powers increasingly supported a complete prohibition on chemical weapons, including control of domestic production and trade as well as verification. In contrast to the US, regime membership provided a mechanism in which humanitarian concerns and broader security policy priorities ultimately aligned themselves. Civil society and the 1925 Gas Protocol Opponents of chemical warfare realized that gas weapons possessed ‘the twin appeals of being new and unique in its effects on man’.82 The evident horrors of its use in the First World War were compounded by predictions of even more terrible effects in future conflicts as a result of scientific and technological advances. Campaigners, linking chemical weapons to strategic air power, magnified these effects to those of an ‘absolute weapon’. Dr G. Woker, writing for the Inter-­Parliamentary Union, described how ‘100 aeroplanes, each carrying a ton of gas, could cover Paris with a gas cloud 20 meters thick. This could be done in an hour, and if there were no wind, Paris would be annihilated.’83 For the International Red Cross (IRC) poison gas represented a vivid symbol of the horrors of war. The Final Resolution of the 1925 IRC Conference declared the objective of the IRC and its national societies as: not then merely a question of making a scientific and practical study of the best means of assisting the victims of gas; it is above all a question of undertaking moral propaganda against the use of gas, and thereby also a moral propaganda against war itself.84 These claims were not limited to the IRC and other advocates from the Peace Movement. The catastrophic effects of air-­deployed chemical weapons were

34   Prohibition of poison gas and dum dum bullets d­ iscussed in the British House of Lords, while a Chinese delegate to the 1925 Geneva Conference, echoing Woker, was quoted asserting that twenty large gas bombs could destroy a city the size of Berlin or Chicago.85 Civil society advocacy in support of the 1925 Gas Protocol is consistent with the linking of peace to advances in science and technology that characterize the earlier Hague conferences. The regime is also notable as a case where domestic civil society actors mobilized to militate against treaty ratification. A one-­year gap between the signature of the treaty and its reporting out of Congress on 26 June 1926, coupled with a lack of supporting activity by the State Department, provided space for opponents within civil society. In particular, veterans associations in the United States, notably the American Legion, exerted their considerable influence. Significantly, the absence of convincing counter-­arguments enabled this well-­organized body to be effectively mobilized by the War Department and other proponents of chemical warfare preparedness. Public opinion was not unified in its opposition to chemical weapons and national security interests offered persuasive counter-­arguments. Yet the development of international regimes as a recognized mechanism to promote humanitarian issues between 1899–1925, represents an important development. This period saw the emergence of the IRC as a proponent of ‘moral propaganda’. And if the influence of the humanitarian lobby was still too diffuse to radically change the cost–benefit calculus in relation to chemical gas, the emergence of early IHL regimes at least provided a means to force states to consider the humanitarian implications of their choices.

Normative considerations Humanitarian push or national security pull? In the case of expanding bullets, states were giving up nothing that was central to their national security interests. It is also arguable that, over time, chemical weapons have been of diminishing strategic value in the face of improved defensive and protective equipment as well as the emergence of other weapons less dependent on climatic conditions. Brown acknowledges that ‘the immense logistical and training burden unique to gas warfare’ seemed wasteful given ‘it could not be proven that the use of gas would provide any quantum jump in probability of battlefield success’.86 Moreover, the threat posed by their use as a weapon of terror has spurred regime development.87 The paradox of the US position in relation to the 1925 Gas Protocol is that the demand for gas warfare readiness was not matched by a viable operational capability. No new gas masks had been made since 1921, no toxic shells had been filled since 1922, chemical weapon war reserves were in perpetual deficit and, by 1924, chemical warfare training was not even included in the War Department Training Regulations.88 Although the reasons behind US rejection of the protocol were complex, interest groups expended significant political capital – both at home and abroad – to safeguard a weapon that could not be used.

Prohibition of poison gas and dum dum bullets   35 The helpful role of the UK in clarifying linguistic differences between the English and French language versions of the 1925 Gas Protocol text makes all the more conspicuous their subsequent volte-­face. In 1970, the British Foreign Secretary announced that the UK considered CS89 and other such tear gases to be outside the scope of the protocol, arguing that CS is not harmful to man and therefore quite different from those lachrymogens existing in 1930. This contrasted with repeated statements following ratification. The reason for the British policy shift lies in a proposal from UN Secretary-­General U Thant, submitted in 1969 as a draft UNGA Resolution,90 that the 1925 Gas Protocol ‘applies to the use in war of all chemical, bacteriological and biological agents (including tear gas and other harassing agents) which now exist or which may be developed in the future’.91 UK troops had recently used CS in Northern Ireland for the first time, influencing Britain’s abstention from the draft Resolution and the Foreign Secretary’s subsequent ‘clarification’ in the House of Commons. The British action was doubly harmful as a unilateral reinterpretation of regime obligations, and a blow to national credibility as a proponent of arms control and disarmament. Lord Chalfont, then Minister for Disarmament, was damning in his assessment that ‘the military pragmatists and the realists had won a victory. The options so dear to the defence planners had been kept open’.92 The US, both as a non-­party and on their eventual ratification on 22 January 1975, made clear their understanding that the scope of the 1925 Gas Protocol did not extend to control agents and chemical herbicides. When the Nixon Administration asked the Senate to consent to US ratification on 25 November 1969, efforts by Secretary of State William P. Rogers to permit retaliatory use of chemical weapons were resisted, thus moving away from the policy that had prevented ratification fifty years previously. But the exclusion of non-­lethal chemical weapons remained, and these weapons were used extensively in Vietnam. Progress was therefore partial. Regime membership had become more acceptable in line with developing humanitarian norms against the use of chemical weapons, but it was subject to a ‘network of reservations’.93 The reservations by the US – compliant in 1925 if not a signatory until fifty years later – as well as Russia, cast a fundamental doubt on whether the 1925 Gas Protocol had actually banned the use of chemical weapons. James Hammond notes that: these reservations of the two most powerful military nations at that time regarding the use of gas warfare set the stage for doubts about whether gas warfare had really been outlawed by international law . . . signers of the pact wishing to initiate its use will ensure that they have an excuse to justify themselves.94 It is therefore relevant to recognize the importance of which states include reservations. That these positions are held by Great Powers gives them a much greater weight than would be the case for smaller states. Multiple reservations and interpretations suggest that, at least in the case of chemical weapons, national security ‘pull’ was the driving factor in states’

36   Prohibition of poison gas and dum dum bullets approaches to implementing their international obligations. Similarly, the UK signed up to Hague Declaration 3 only after concluding a colonial war in which these rounds were used, while the US has never signed the declaration. Again, national security trumped humanitarian considerations. The absence of international condemnation following Japanese use of chemical weapons in China before and during the Second World War also points to an erosion, in implementation, of the political will generated during the regime design phase. States applied a calculus that weighed humanitarian with national security concerns. There was no norm bandwagon to tip the balance in favour of regime compliant behaviour. On the other hand, without humanitarian ‘push’, it is questionable whether states would have even made it to the negotiating table. The emerging norm against these weapons actively promoted by civil society was a key contributing factor to the convening of the 1899 Hague Peace Conference. The IRC’s moral propaganda also proved critical to initiating the Geneva negotiations in 1925. Despite evident multilateral insincerity, states were still obliged to come together and found regimes in order to demonstrate their humanitarian credentials to domestic and international constituencies. The pressure applied by civil society, including through innovative use of print media, was evident in the buildup to the conferences, affecting negotiations in ‘real time’. Both these cases therefore reflect Richard Putnam’s analysis of national positions stemming from a two-­ level calculation that balances domestic with international considerations. Certainly, the contradictions in the US position on chemical weapons bear this out. While the United States led efforts to seek a ban on trade during negotiations, this translated into an outright ban that was supported in Geneva but could not be realized when taken back home for endorsement. The complexity of the issues involved is also apparent in the British case. Domestic pressure played a significant part in pushing the UK to sign the 1925 Gas Protocol, but it was domestic security concerns that lay behind the reinterpretation of those obligations to exclude CS gas forty-­five years later. Hague Declaration 3 and the 1925 Gas Protocol are agreements born of their times. While dum dum bullets had been designed for use in colonial wars, no African nations were represented in The Hague.95 However, the late decision to open up the conference sessions allowed for a degree of transparency in the decision-­making process and permitted a broader audience to voice their opinions. In terms of implementation, although Hague Declaration 3 has been widely observed, improvised expanding bullets remain in use, particularly in developing world conflicts. By rejecting initial proposals outlawing the trade in chemical weapons, delegates in Geneva recognized the implementation challenges such proposals posed for non-­gas-producing (i.e. less developed) states. This demonstrates progress from 1899; by invoking the principles of the League of Nations, Southern countries had a voice in the negotiations. However, breaches of the protocol paint a different picture. From Africa to Southeast Asia the developing world has born the brunt of these weapons, while the response of the developed world has been muted.

Prohibition of poison gas and dum dum bullets   37 Norms and regime effectiveness The legal restraint of the 1925 Gas Protocol focused public and political attention on the issue, thereby reinforcing the stigmatization of chemical weapons. As an illustration, official State Department documents of the period characterized chemical weapons as ‘inhuman’.96 Brown, comparing the First and Second World Wars, points out that ‘for the first time since the advent of the nation at arms a major weapon employed in one conflict was not carried forward to be used in a subsequent conflict’.97 He argues that codification had the effect of reinforcing other restraints, strengthening public and military fear of chemical warfare and providing an excuse for a lack of military preparedness in this area. As a consequence, in the Second World War, ‘President Roosevelt would not even consider the possibilities of American initiation or preparation beyond the minimum amount required for retaliation.’98 If the concerns of powerful constituencies ensured that the US did not join the regime, the norm against the use of poison gas in war – magnified through the regime – ensured that such a capability would not be used. For different reasons, the 1925 Gas Protocol played a significant role in Ger­ many’s non-­use of chemical weapons during the Second World War. The prohibition, in reinforcing the specific arms control terms of the Versailles Treaty, effectively cost the Germans ten years of research and development. Despite German scientists making a major breakthrough in producing nerve agents, a lack of confidence engendered by the prohibition contributed to their forfeiting the use of this weapon.99 This shows that stigmatization is not the only positive regime effect being exerted. The regime also reinforced prescriptive measures found in other agreements to inhibit technological progress in this area. The conduct of both the US and Germany shows how the development of a norm against the use of chemical weapons was instrumental in their non-­use during the Second World War. While it is evident that the 1925 Gas Protocol contributed to the emergence of an internationally recognized norm against the use of chemical weapons, its application has not been universally effective. Italy, as a state party, used poison gas in 1935–1936 during its invasion of Abyssinia.100 This was in spite of the strong censure of international public opinion when confronted by the romantic image of Princess Tsahsi of Ethiopia pleading in the international news media for assistance against naked aggression, exemplified by the use of gas by the Italians. During the conflict the ICRC received numerous complaints, confirmed by delegates on the ground, over the use of chemical weapons by Italian forces. The response of the president of the Italian Red Cross was that the 1925 Gas Protocol did not preclude the use of gas in reprisal for the ill-­treatment of Italian prisoners of war. The ICRC categorically rejected this position, re-­emphasizing the absolute prohibition on the use of chemical weapons in the protocol. If the argument seeking to justify the use of gas as a legitimate reprisal does not ring true,101 the conduct of the Italians suggests that neither the norm nor the legal prohibition were significantly robust when measured against national interests and a much weaker developing world enemy that could not retaliate in kind.

38   Prohibition of poison gas and dum dum bullets As a threat only to individual combatants, the normative dimension of the ban on expanding bullets is less apparent than in the case of gas warfare. However, the humanitarian ‘push’ provided by the Peace Movement was an important factor in bringing states to the table. Public attention on the conference increased the pressure on delegates to achieve tangible results. Ken Rutherford argues that peer pressure to emulate other states was also a powerful mechanism.102 In this respect, the norm against expanding bullets proved important in the negotiation of a ban which, once agreed, was widely implemented. The Italian example above recalls that IHL regimes face their most significant tests during implementation. However, the rejection by other delegations of the British position as ‘contrary to the humanitarian spirit’103 represents an important statement of solidarity in a forum where the developing world was not well represented. Indeed, despite their rejection of the ban, the regime’s moral force clearly influenced British behaviour. It was decided not to use these rounds during the Boer War and dum dums sent in error to South Africa were swiftly recalled.104 Disagreement over the criteria for the unacceptability of dum dum bullets is unsurprising because technological progress had outpaced the evolution of IHL. As T.J. Lawrence notes in his 1895 Principles of International Law: men could not make up their minds whether means of destruction were to be deemed unlawful because of their newness, or their unfairness, or their secrecy, or their cruelty, and they generally solved the difficulty by objecting to what they disliked, and regarding as unobjectionable what suited their tastes or worked to their advantage.105 The 1868 St Petersburg Declaration and subsequent codifications that prohibit weapons that cause superfluous injury or unnecessary suffering, did not establish objective criteria to measure this term or its counterbalance, military necessity. This highlights an important wider issue for IHL regimes. General principles can be interpreted very differently in their application to specific weapons. The inherent non-­negotiability of stigmatization has the effect of cutting through this Gordian knot, and preventing states from exploiting ambiguities.

Implications for landmine regimes What does the design, implementation and effectiveness of Hague Declaration 3 and the 1925 Gas Protocol mean for the two landmine regimes? First, they show that regime interplay and nesting dynamics are important. Over time, powerful effects are generated by situating individual regimes within a broader IHL framework addressing weapons causing unnecessary suffering or superfluous injury. In this broader context, these two historical regimes have proved effective in raising the political costs of continued use, providing ‘a solid moral and political basis for criticizing other states’ ownership or use of these weapons’.106 Yet implementation challenges also point to the difficulty of operationalizing general principles of IHL with respect to specific weapons. This reinforces the

Prohibition of poison gas and dum dum bullets   39 point that the relationship between design and implementation factors can be critical to regime effectiveness. Design factors are shown to be particularly significant. Weaknesses in the 1925 Gas Protocol regime framework stemmed from reservations as well as the absence of a formal verification mechanism. This mix of conditionalities and lacunae reflect the compromises, necessary at the time, to reach agreement. Conversely, the subsequent creation of the OPCW displays the flexibility of the regime in a wider sense. This was necessary to address problems that were either unresolved in the regime design process or only became apparent over time. Conditional obligations and verification challenges must therefore be considered in light of differing approaches adopted within the two landmine regimes. APII is characterized by a number of reservations while, in the absence of a formal verification mechanism, the mine ban treaty has been supported through the development of an unconventional civil society-­based approach to compliance monitoring. How landmine regime frameworks address these implementation challenges thus represent important considerations for regime effectiveness. Formal compliance has been almost complete in the case of Hague Declaration 3. For the 1925 Gas Protocol, if there has been voluntary defection, it has been detected and attracted international opprobrium. However, from a regime perspective, the lack of flexibility within the Hague Declaration 3 regime, the consequent inability to deal with improvised dum dums, and its non-­applicability to new munitions with similar effects are particularly important considerations. Given that contemporary use of landmines has evolved significantly from the static minefields found in earlier conflicts, there may be a significant risk of diminishing relevance if regimes do not adapt in response to changing behaviour or technological advances. Civil society is a prominent actor in IHL regimes. The IRC focused its moral propaganda in order to further the development of IHL. The peace and internationalist movements exerted influence to shape state behaviour at particular points in history: the armed peace of the 1890s and the ‘never again’ zeitgeist that followed the First World War. Even when barred from the negotiations, civil society actors developed novel approaches in order to influence proceedings. In particular, the focus of national and international media was carefully exploited to ensure that national delegations felt under strong pressure to achieve demonstrable results. At the same time, the role of civil society in undermining regimes also needs to be acknowledged. Domestic lobby groups played an important role in the US decision not to ratify the 1925 Gas Protocol, providing a counter-­ intuitive example of civil society influence on national decision-­making. It is important to recognize that multilateral insincerity does not preclude the behaviour of participating states being influenced in important ways. Once involved in a process of regime development, there is an obligation to show progress that is magnified by the humanitarian nature of the issue area. The embedding of specific issues within the larger concept of inhumane weapons, contributes to regime effectiveness even if countervailing national security pull factors can dilute the effectiveness of this norm. The normative weight of the

40   Prohibition of poison gas and dum dum bullets IHL subject matter was clearly influential in bringing the major powers of the day to the negotiating table as well as in fostering broad-­based support for the regime formation process. Finally, the impact of stigmatization creates powerful effects beyond the immediate orbit of the regime. The US did not ratify the 1925 Gas Protocol for many years, but as a result of the stigma associated with chemical weapons, their use was never a possibility in the intervening period. This dynamic needs to be qualified since stigmatization has not been universally effective in precluding the use of dum dums and poison gas. Despite the emerging norm against the use of these weapons, a number of states voluntarily defected from regime commitments on the use of chemical gas. Moreover, the UK and the US were confronted by compelling arguments to ban dum dum bullets on humanitarian grounds, but did not join the regime because of military interests. These considerations should provide important caveats for the landmine regimes.

3 The emergence of the landmine regimes

Introduction In the case of APII, the ‘closed’ nature of the process and its low public profile has resulted in a lack of knowledge on the emergence and design of this treaty. As was acknowledged by one US negotiator in 1980, while verbatim accounts exist for the plenary sessions, ‘the real negotiations took place in unrecorded private discussions and in the many sessions of the three working groups’.1 In contrast, the Ottawa Process is commonly characterized as highly participatory and has generated an extensive body of writing. However, there is very little critical analysis of the regime or its origins. Bringing together insights from IHL and mine action within a regime framework enables us to fill some of these gaps. It is particularly important to consider the extent to which obligations placed on members take into account both the technical and political challenges associated with this issue. From a design perspective, effectiveness is understood in terms of how landmine regime obligations relate to key mine action objectives. In order to capture this, it is necessary to address the extent to which the regimes, individually and collectively, confront the core problems they were set up to meet. A degree of consonance between regime and mine action priorities would seem to be necessary in order to bridge design and implementation concerns. Effectiveness may also be impacted – positively or negatively – through unforeseen consequences provoked by the political dynamics of regime formation. Given the apparent dislocations between different stakeholders, cognitive dissonance may be one factor affecting regime effectiveness. The emergence of both treaties marks important steps in the development of IHL. Protocol II sets an important standard as the first codification of IHL specifically regulating landmines,2 while the mine ban treaty constitutes the first formal ban on a conventional weapon to be adopted by the international community since the 1899 Hague Declaration on dum dum bullets.3 This underpinning humanitarian imperative that situates the landmine regimes within the broader normative framework of IHL is doubly important. First, it can shed light on how far support for regime formation derives from a quality of nesting that draws on deeper IHL trends, encouraging stakeholders to stay on the right side

42   Emergence of the landmine regimes of a compelling humanitarian issue. And second, it allows a critical examination of claims that they represent a new departure in the practice of IHL. In order to address these related issues, this chapter critically assesses the origins and processes leading to the formation of the two landmine regimes. It then analyses the roles of different actors in shaping regime design and considers the significance of norms to landmine regime emergence and formation.

Understanding landmine regime formation Origins of the CCW and mine ban treaty regimes The ICRC began preparing Draft Rules4 from the early 1950s for the protection of civilians from weapons deemed to have uncontrollable effects. Beginning in the mid 1960s, the use in Indochina of tear gases and other weapons perceived to be excessively injurious or indiscriminate led to a number of resolutions within the UNGA, as well as studies commissioned by the UN Secretary-­General, on the effects of various weapons.5 In 1971 and 1972, the ICRC convened Conferences of Government Experts to consider proposals for prohibitions or restrictions on certain conventional weapons. These conferences, informed by studies commissioned by the Swedish government in which military and medical experts examined recently developed weapons, produced a report that proposed language for several anti-­personnel weapon bans. Growing concern over the use of certain anti-­personnel weapons in contemporary conflicts led the ICRC in 1973 to establish an expert working group, which met in Lucerne in September 1974, then in Lugano in January 1976. As a result of these meetings, consensus emerged around proposals relating to undetectable fragments, incendiary weapons, restrictions and recording requirements for remotely-­deliverable mines.6 Discussions continued in an ad hoc committee on conventional weapons formed as part of a Diplomatic Conference on the Reaffirmation and Development of International Humanitarian Law Applicable in Armed Conflicts (CDDH), held in Geneva in 1976 and 1977. The CDDH was convened by the Swiss government as the depository of the Geneva Conventions, in order to prepare for the negotiation of the Additional Protocols to the Geneva Conventions. The CDDH could not reach agreement on the types of weapon to be included, whether prohibitions or restrictions should be pursued, or if battlefield use or the protection of civilians should be the focus. As shown in Table 3.1, positions were broadly divided between clusters of neutral and developing states on the one hand and ‘militarily-­significant’ states7 on the other. Table 3.1  Alignment of states within the CDDH Neutral/developing states

Militarily-significant states

Sweden, Norway, Switzerland, Yugoslavia, Egypt, Mexico

USA, larger NATO members, USSR, larger Warsaw Pact members

Emergence of the landmine regimes   43 The concluding session of the CDDH recommended that these issues be addressed in a UN framework. As a result, after preparatory meetings in 1978–1979, which used as their basis the texts prepared in the CDDH ad hoc Committee, the CCW conference was held in 1979–1980. These early initiatives firmly situate progress on the landmine issue within the broader framework of IHL. As a single issue, landmines had not gained sufficient momentum to catalyse a process of regime formation. However, a regime was able to form within the CCW because it was embedded in a broader set of contemporary IHL issues of concern. The transition from working group meetings to the CDDH then to the CCW, points to an important causal relationship between the crystallization of issues to be addressed and the nature of the discussion forum. In the decades preceding the emergence of the landmine regimes, a clustering of concerned states with expert communities was coalescing around this issue area. The mutually reinforcing relationship between like-­minded states, non-­state actors involved in IHL and expert communities, clearly demonstrates the positive effects of intertwining normative considerations with technical know-­how on the effects of weapons. Interlocking and overlapping regime formation dynamics The CCW regime was established in 1980 with three protocols: Protocol I on non-­detectable fragments; Protocol II relating to mines, booby-­traps and other devices; and Protocol III on incendiary weapons. During the 1980s, the regime gained few new adherents and had a low international profile. This changed near the end of the decade as the issue of blinding laser weapons was highlighted through a series of ICRC-­initiated experts’ meetings between 1989 and 1991. At the same time, the humanitarian cost of landmines was gaining increasing international prominence. Both these factors resulted in France, supported by a number of other states, requesting a CCW review conference. A period lasting twenty-­seven months led from expert group meetings to the final conference session in May 1996. Increasing engagement in the process was reflected by a rise in the number of states parties to the original Protocol II in the run up to the conference, including a number of militarily-­significant states such as the United States. Additional Protocol IV, on blinding laser weapons, was agreed relatively easily. This ban was particularly significant in demonstrating a link between the CCW and the broader framework of IHL. For the first time since the 1868 St. Petersburg Declaration on exploding bullets, a weapon had been banned before ever being deployed in battle. On the landmines issue, two distinct sets of interests (supporters of a total prohibition on APMs versus advocates of strengthening Protocol II’s restrictions) led to a more challenging negotiating process. A number of states vacillated between these two positions, supporting a ban in principle, but recognizing that such an outcome would not be achieved in consensus-­bound negotiations. Accordingly, tortuous discussions meant the deadline for concluding the review conference was twice pushed back, from

44   Emergence of the landmine regimes October 1995 to January 1996 and then from 22 April to 3 May, before agreement on APII was finally reached. The 1979–1980 negotiations had been broadly split between what Roach terms the ‘prohibitionists’ and the ‘realists.’8 In 1995–1996, the same basic positions applied with prohibitionists unable to meet their goal due to entrenched realist opposition within a consensus framework. The United States played a proactive role in the negotiations, keen to demonstrate leadership on the issue, but with the same underlying assumption that the forum could be managed in line with US policy objectives. While knowledge and concern over the issue of landmines had evolved significantly in the fifteen years between conferences, the same regime rules constrained the progress that could be achieved to ‘more of the same’ technical restrictions. If this reality was reassuring for US security interests, it did not placate those who wished to address the humanitarian costs of landmines. Chairman of the review conference Johan Molander, acknowledged the link between regime rules and the outcome of the negotiations when he stated that results reflected as much as could be achieved on the basis of consensus.9 If the APII outcome represented a tactical victory for those that sought to limit their liability, it should also be understood as a strategic error from the perspective of those same interests. It was now undeniable that the only way to achieve a ban was via a distinct regime framework. The measured CCW approach jarred with the recognition of a humanitarian crisis caused by APMs, and thus rendered transparent the inherent tension between arms control and IHL imperatives. The ICRC in particular made a direct linkage between the nine-­year period allowed for implementation of APII’s detectability and self-­destruction provisions, and the human cost of mines already in the ground or that would be laid during that period.10 This mirrors similar complaints raised in 1980 at the insistence of certain states, to set a high number of ratifications as the trigger for entry into force – effectively a delaying tactic to prevent early review of the convention.11 On the outcome of the review conference, United Nations Secretary-­ General Boutros Boutros-­Ghali echoed a widespread view that progress had fallen short of expectations and did not reflect the groundswell of international public opinion in favour of a ban on APMs. Unsurprisingly, the review conference was condemned as a failure by the ICBL12 and this message was picked up by the international news media in its reporting of the conference. The leitmotif of the Ottawa Process, coined by ICBL Coordinator Jody Williams, stemmed from these partial results and was unambiguous: ‘no exceptions, no reservations, no loopholes’.13 The need for greater progress was also felt by states. Canada convened a meeting that brought together pro-­ban states, international organizations and NGOs in Ottawa between 3–5 October 1996. Spurred by a perception that momentum was being lost, Canadian Foreign Minister Lloyd Axworthy made a surprise announcement during the conference’s closing address. Cutting through the tangle of proposal and counter-­proposal, he asked states to return to Ottawa by the end of 1997 to sign a treaty banning APMs. This drew support from the

Emergence of the landmine regimes   45 ICRC and ICBL, as well as the UN Secretary-­General, but caused consternation at the bilateral level as a radical breach of diplomatic practice. At this stage, only some 50 governments had publicly declared themselves in favour of a ban so Canada had taken a real political risk. Events moved quickly. The Austrian Government circulated a first draft treaty that drew heavily on disarmament law – notably the Chemical Weapons Convention14 – only a few weeks after the Ottawa conference. They then hosted a meeting in Vienna between 12–14 February 1997 to exchange views on the draft treaty. It was attended by representatives of 111 governments, with the ICRC providing a lead on substantive issues. A revised text, issued on 14 March, was to remain relatively unchanged throughout the process. A subsequent meeting in Bonn between 24–25 April focused specifically on the issue of verification. A total of 121 governments participated in this meeting which was followed, between 24–27 June, by a conference in Brussels attended by 156 states. This meeting adopted a declaration forwarding the third Austrian draft to the diplomatic conference scheduled for September in Oslo. The ninety-­seven states that signed the Brussels Declaration endorsed the central elements of the ban treaty and reaffirmed their commitment to sign a treaty in Ottawa before the end of the year. Such extensive support only nine months after the Ottawa conference, provides strong evidence of a norm bandwagon. This effect is perhaps best exemplified by the very undiplomatic behaviour of many states in showing their hand before the start of the negotiating conference. This allowed the core group to focus on the uncommitted. The momentum of the government-­focused process was reinforced by ICRC and ICBL-­led workshops, and related advocacy efforts across the globe. NGO-­ driven conferences and events in mine-­affected states created a powerful push factor for states to embrace the anti-­APM norm. In March 1997, the Tokyo Conference on Anti-­Personnel Landmines, hosted by the Japanese Government, was attended by twenty-­seven States, the EU, and ten international organizations. Other national and international conferences took place in East, Central and South Asia, Australia, New Zealand and throughout Europe. Press conferences were deliberately held jointly with pro-­ban states to underline the collaborative nature of the process.15 In the same way, comments were sought from the ICBL on the treaty drafts and the ICBL was represented in each of the conferences in the governmental strand of the process. Following the Oslo negotiating conference, a further round of demarches, lobbying at the 1997 UN General Assembly and NGO-­sponsored media activities, were geared to encouraging wavering states to sign the treaty in Oslo. From a regime perspective, the parallel state and civil society-­driven strands of this process demonstrate a number of important characteristics. The momentum provided by the short deadline until the negotiating conference is particularly significant. This meant that discussions were never bogged down in the details. That technical issues were considered secondary to the political imperative driving the process is evident in the rapidly growing list of states indicating their support just a few months after the Ottawa conference. Stakeholder clustering

46   Emergence of the landmine regimes dynamics are also crucial during this phase. Governmental and civil society initiatives were closely coordinated and created important synergies that swiftly built up political momentum behind the process. Many of the same states were also involved in the CCW process. However, a combination of different rules of the game with a partnership-­based approach reaching across stakeholder groups, led to a significantly different outcome. The Oslo Diplomatic Conference took place between 1–18 September, concluding with the adoption of the treaty. Ninety states were registered as full participants and thirty-­two attended as observers, as did representatives of the ICRC, various UN agencies and hundreds of NGOs. The main sticking point in agreeing the text proved to be a package of five ‘non-­negotiable’ changes required by the US delegation. When initial lobbying met with failure these were eventually whittled down to three (see Table 3.2). An exemption for anti-­handling devices was intended to save US ‘smart’ mine systems, in which the primary weapon targeting vehicles is protected by APMs to avoid tampering by enemy forces. However, it was widely recognized that this was an attempt at reclassification since these weapons had already been defined as APMs during the APII negotiations. This is significant. A move to undermine the treaty was foiled due to parallel negotiations in the other landmine regime framework. In the end, fierce lobbying by the US was resisted. The majority of states – backed up by the ICBL – refused to dilute a straightforward ban as the price for US signature. A twenty-­four-hour delay, requested by the US on the final day of negotiations, saw President Clinton and various senior officials lobby fruitlessly for support. The nature of the process had drawn out their red line issues and effectively isolated the US delegation from the rest of the negotiating conference. The proposals were eventually withdrawn, paving the way for the conclusion of the conference and the adoption of the treaty. Subsequently, 122 States signed the mine ban treaty in Ottawa between 2–4 December 1997.16 The shortcomings of APII in the eyes of major international actors provided a strong argument to advance the agenda by other means. This feeling was magnified by civil society actors, denied space within the CCW process, who networked in order to lobby effectively through the Ottawa Process. APII did exert

Table 3.2  US negotiating red lines in Oslo Initial US position

Fallback US position

1  Strengthened verification provisions 2  Exemption for the Korean peninsula 3  Unrestricted right of withdrawal 4  9-year period for entry into force 5 Exemption for anti-handling devices placed near anti-vehicle mines.

1  Eliminate Korean mines within 9 years 2  Right to withdraw during conflict 3 Exemption for anti-handling devices placed near anti-vehicle mines.

Emergence of the landmine regimes   47 a pull factor on states through pressure to adhere to any regime addressing landmines. However, compromise between military utility and humanitarian concerns was an insufficient position from which to demonstrate responsible international conduct. As Prokosch notes, the CCW approach ‘seemed to be intended to satisfy the needs of military forces, which may later have to occupy a mined area, rather than to protect civilians’.17 On this level, the CCW was in the shadow of the pro-­ban agenda from the outset. In tandem, the lack of political space for civil society within the CCW framework meant that there was no safety valve for their concerns. This gave both a legitimate platform and a raison d’être for the ICBL and the Canadians to promote an alternative track. Restrictions, prohibitions and landmine regime effectiveness While both regimes pursue the same headline objective, approaches, actors and rules are distinct. Similarities and distinctions are summarised in Table 3.3. The mine ban treaty negotiations were founded on principles of transparency and majority decision-­making. This was reflected in an open approach to participation and flexible rules of procedure with decisions requiring a two-­thirds majority. In contrast, the narrow state-­based eligibility criterion meant that, within a consensus-­bound framework, the CCW negotiations were shaped by national militaries and arms control experts. The point of departure for the negotiation of Protocol II was that landmines are necessary, defensive weapons whose use needs to be restricted in order to minimize collateral risk. This emphasis on technical criteria, as opposed to humanitarian impact, is a direct result of the background and expertise of the negotiators. In the absence of any evidence to the contrary from affected states or mine action practitioners, the longevity and nature of the post-­conflict landmine threat and its socio-­economic consequences were not discussed.18 Specifically, the restrictions and prohibitions found in Protocol II and APII fail to take into account the indefinite time period during which the weapons can remain ‘live’. The temporal nature of the hazard after the cessation of hostilities as a mine action issue and the resulting humanitarian cost was not acknowledged except obliquely through an unsuccessful,

Table 3.3  Key features of APII and the mine ban treaty APII

Mine ban treaty

•  Reduce human suffering •  Military–technical approach •  Lengthy regime formation •  State-centred, low-profile •  Military/officials negotiating •  Consensus voting •  Technical restrictions •  No developing world buy-in

•  Reduce human suffering •  Humanitarian imperative •  Rapid process •  Open to all, high-profile •  Mine action experts •  2/3 majority •  Ban on APMs •  North–South consensus

48   Emergence of the landmine regimes politically motivated attempt by Libya to include an obligation on the former belligerents to clear mines placed in previous conflicts such as the Second World War.19 Protocol II and APII, like any negotiated treaty, reflect a number of compromises found in the permissive language which qualifies various obligations. Under Protocol II, parties to a conflict must ‘endeavour to ensure’ that non-­ pre-planned minefields are recorded (Article 7(2)), while the seemingly stronger obligation requiring recording of all ‘pre-­planned’ minefields (Article 7(1a)) is rendered irrelevant by the lack of such planning in how the majority of the world’s landmines are laid. Indeed, the term minefield implies that the weapon is deployed to a recognized military pattern for specific tactical or strategic reasons. This does not conform to the ‘nuisance’ mine use employed by state and non-­state forces in many developing world conflicts. The terminology used is particularly revealing because it demonstrates that the regime and its stakeholders were neither informed by, nor focused on, the realities of contemporary conflict. The input of developing countries can be discerned: forbidding the use of animals or their carcasses as booby traps in Article 6 of the protocol stems from the concerns of Mongolia for their population of nomadic herders.20 However, this remains a point of detail rather than any robust attempt to promote responsibility for the post-­conflict consequences of military actions. Article 9 of Protocol II, requires that former conflict parties ‘endeavour to reach agreement’ on the provision of information to facilitate mine clearance following the cessation of hostilities, a provision referred to as ‘the most questionable variant of the rules of warfare, whose true purpose is not to “safeguard the minimum standard of civilization” but rather to “cover up the inability or unwillingness to achieve this object”’.21 There is no obligation to require parties to a conflict to clear the mines they laid, nor is specific guidance given for safeguarding civilians. Moreover, if states parties are obliged to disseminate these requirements to their armed forces, translation into field manuals and standard operating procedures is not mandated nor are penal sanctions provided for against individuals. Regime rules are not linked to implementation procedures. The widespread use of APMs in the 1980s was a visible indication of the ineffectiveness of Protocol II. Militarily significant states did not want to be bound by its provisions, while others felt it not worth ratifying.22 APII addressed many of the concerns surrounding the original protocol by extending its scope to non-­international armed conflicts, strengthening restrictions on use, banning undetectable APMs and prohibiting anti-­sensing devices, as well as placing further restrictions and obligations on mine laying. Implementation provisions were also tightened, with an obligation on states to prevent violations, impose penal sanctions on individual violators and distribute relevant military instructions to their armed forces. These improvements saw an increase in membership. While this was certainly influenced by the newfound international prominence of the APM issue, at the same time APII remained the only IHL specifically regulating anti-­vehicle mines and booby traps.

Emergence of the landmine regimes   49 The origins of the Ottawa Process in field-­based organizations with practical experience of landmines, provides a notable contrast with the CCW approach. From a mine action perspective, the problems posed by the restrictions-­based APII approach would seem to be addressed by the mine ban treaty: it unambiguously requires states parties to destroy or ensure the destruction of all APMs under their jurisdiction or control. However, in reality, mine action practice does not reflect such a maximalist approach. The guiding principle is to reduce the threat posed by mines so that individuals and communities can live in a secure environment. In other words, there is an inherent tension between regime goals and the more nuanced calculus of mine action priorities. The resulting friction between expert communities and advocacy NGOs will become apparent in regime implementation. The mine ban treaty lacks an effective verification mechanism because states were unwilling to sign up to stringent compliance monitoring provisions. The regime establishes annual reporting requirements in which technical data is provided by states parties. Fact-­finding missions (FFMs) initiated by the UN Secretary-­General – modelled along the lines of various arms control and disarmament regimes – are also provided for should any state party be suspected of a breach. Article 8, describing the process for FFMs, is the longest in the treaty and details the steps for initiating such a mission. However, no practical modalities are established, obscuring how this process might play out in practice. This ambiguity de-­links the political will evident in the regime formation phase from subsequent implementation. On the one hand, there is limited formal verification. On the other hand, members are obliged to take the diplomatically uncomfortable step of pointing the finger at other states suspected of breaches. The absence of any FFMs from March 1999 to date raises major questions about the ability of the regime to acknowledge and address non-­compliance.

The interplay of state and non-­state actors in landmine regime formation While recognizing that issues are interdependent, in order to understand the dynamics of regime formation it is equally important to acknowledge the interdependence of different actors. Identifying the clusters of state, sub-­state and international actors, with distinct or overlapping regime interests, is particularly relevant for APII and the mine ban treaty given their contrasting approaches to participation. States As international legal bodies, the formal membership of the landmine regimes is comprised of states. The provisions that delineate them are binding on states, so implementation is gauged on the level of their participation and adherence to regime rules. In order to understand interests, priorities and commitments across the two processes, these actors are considered under the categories of militarily significant states, middle powers and developing nations.

50   Emergence of the landmine regimes Militarily significant states, regime membership and effectiveness Militarily significant states gravitated to hard or soft positions. Hardliners such as Russia, China, India and Pakistan initially viewed all attempts to regulate conventional weapons with suspicion. The US, UK and other major NATO powers adopted a pragmatic stance that emphasized arguments of military utility at the expense of humanitarian considerations in order to put a break on more radical proposals. Over time, there was greater recognition within military circles that the non-­detectable, long-­lived APMs used by many developing states and non-­ state armed groups posed a threat to their own forces. In certain cases, this shift in perception was reinforced by political change at home. On 21 May 1997, the new Labour Government pledged the total destruction of British APMs by 2005 and gave full backing to the Ottawa Process. France expressed similar support and reversed its earlier insistence on an ‘exceptional use’ clause in the treaty. About turns by two major European powers are important in themselves, demonstrating a clear move from a military-­pragmatist stance to political acceptance of the anti-­APM norm. Spillover effects also influenced other European states that did not necessarily fit comfortably within the pro-­ban camp; Italy, for one, agreed to outlaw the use and production of APMs and immediately prohibited their operational use by Italian forces. Non-­supporters in Europe included states with compelling national security concerns such as Finland23 and Turkey, or Yugoslavia, recently involved in an armed conflict. Beyond Europe: China, Cuba, Egypt, India, Iran, Iraq, Israel, North Korea, South Korea, Pakistan, Russia and Syria are some of the states that did not participate in Oslo. In these cases, national security concerns magnified by authoritarian governments allergic to the unconventional civil society-­driven nature of the process won the argument. However, although reluctant to endorse a ban, Russia and China did intensify mine action activities and adopt an export moratorium on APMs to coincide with the agreement of the treaty. Even for these states, emulation effects drove them to support certain mine ban treaty commitments – including important mine action obligations – while remaining outside of the regime. The US played an important role in the genesis of the two landmine regimes. At the 1974 Lucerne Conference, the military pragmatists dominated, spearheaded by the US delegation, which comprised a weapons systems analyst, two military surgeons, military officers, and Pentagon and State Department officials.24 Technical data was used to downplay humanitarian impact, in some cases debunking uncomfortable statistics while refusing to provide ‘classified’ information to back up arguments.25 The 1976 Lugano Conference followed a similar pattern with evidence produced by the US and others to demonstrate that weapons such as napalm and flechettes did not cause superfluous injuries. Despite such disingenuous tactics, the conference concluded with a general agreement on the need for measures, which became the three initial protocols to the CCW. During the 1970s, the US was described by one of its own negotiators at the 1980 CCW Conference as ‘not particularly desirous of concluding a weapons

Emergence of the landmine regimes   51 agreement and neither promoted nor opposed the multilateral negotiating process’.26 Yet once the CCW process became concrete, the US ‘soft’ military pragmatist position shifted in order to influence the outcome of the negotiations. This agenda was reflected in the much-­criticized long entry into force and amendment provisions which represented a specific US negotiating goal. The US delegation’s report on the conference was candid on their objectives: ‘These provisions should give Western countries ample time to ratify before amendments can be considered, and should limit the ability of radical governments to press for an endless series of conferences to expand the current restrictions.’27 The US agreed an export moratorium on APMs in October 1992, subsequently extended by three years with the unanimous support of the US Senate. This was an important step – the moratorium applied equally to NATO allies. Moreover, in another significant move, it was the US that first raised this issue in the UNGA where Senator Patrick Leahy introduced resolutions in 1993 and 1994 calling, respectively, for a moratorium on exports and the eventual elimination of APMs. The US leading role in setting the agenda for the 1995–1996 CCW review conference needs to be viewed in light of these earlier commitments. The Clinton Administration’s announcement on 17 January 1997, that it would observe a permanent ban on the export and transfer of APMs, should be understood as a response to the Ottawa Process. Internal policy divisions present close parallels to domestic debates on the 1925 Gas Protocol. Despite support from both government and domestic civil society, the US resisted the Ottawa Process, influenced by the reluctance of the military and its political supporters to give up this capability. The US therefore pursued the competing strategy to initiate negotiations within the Conference on Disarmament (CD) on a treaty banning the use, production, transfer and stockpiling of APMs. This approach, although supported by a number of states, was widely criticized because the CD, like the CCW, operates on the basis of consensus. In particular, there was a perceived danger that hardline mine producers and users, such as China and Russia, would act as spoilers. These arguments proved moot as the CD, riven by wider institutional problems, failed to include the landmines issue on its 1997 agenda.28 At first glance, the CCW is the natural home of militarily significant states, while the mine ban treaty invites a more diverse membership. In reality, categories proved to be fluid with positions fluctuating during the regime formation processes. The ability to move entrenched ‘hard’ military pragmatist positions was an important part of the CCW approach. Michael Matheson, chief US negotiator at the conference, points to how China, India and Pakistan were initially opposed to extending the scope of the protocol to non-­international armed conflicts, considering this a challenge to national sovereignty and a threat to their own forces when confronted by insurgents not playing by the same rules.29 Ultimately, many hard line states shifted their positions through a combination of political pressure, acknowledgement of the benefits to civilians of the proposed amendments, and recognition that the protocol would be meaningless without expanding its scope to reflect the internal nature of contemporary conflict. And if

52   Emergence of the landmine regimes the CCW does not include mine action stakeholders, it does influence the conduct of producers and users whose behavior will determine the likelihood of new mined areas that would need to be cleared. Representation and reality in the roles of middle powers The role of small and medium sized states – middle powers – is portrayed as central to the success of the Ottawa Process.30 Canada went out on a precarious diplomatic limb in calling for the negotiation of a ban treaty at a time when declaratory support remained limited. Canada was also proactive in forging a partnership with the ICBL and other non-­state actors. The decision was made during the conference itself, based on Canadian officials’ assessment of the momentum that could be tapped by such a step. Subsequently, they had to expend a great deal of political capital in unruffling diplomatic feathers, persuading states of the feasibility of the target that had been set and forming the core group of like-­minded states. This initial core group was made up of eleven members: Austria, Belgium, Canada, Denmark, Germany, Ireland, Mexico, Norway, Switzerland, South Africa, and the Philippines. The group regularly strategized, shared information and coordinated activities. Like Canada, other members maintained strong links with national and international NGOs. South Africa, Mexico and the Philippines worked as regional champions, while the Netherlands and Ireland exploited their EU Presidencies to exert influence. Austria assumed responsibility for developing the draft treaty text, while Germany, Belgium and Switzerland hosted important preparatory meetings. Moreover, if Canada and the Europeans were reluctant to undermine the CD by questioning the inclusion of landmines on its agenda, Mexico could be more forthright, pointedly stating that ‘swiftness is not this Conference’s main virtue’.31 The core group was influential because, as MacFarlane points out, ‘states are themselves formidable advocates and may well be taken more seriously than non-­state actors by other states’.32 Sweden has been a historically significant player in the disarmament field. Swedish initiatives proved particularly influential in contributing both state-­ driven advocacy and empirical evidence to support the emergence of the CCW. Spurred on by strong domestic criticism of the Vietnam War, it was the first state to call publicly for measures to address the legal and humanitarian aspects of the use of different conventional weapons.33 The Swedish government convened military and medical experts to study the effects of certain weapons from the perspective of IHL. Their 1973 report offered language for prohibitions or restrictions on anti-­personnel weapons, including small caliber high velocity projectiles, fragmentation warheads, flechettes, landmines, booby traps and incendiaries.34 The interplay between the two regimes and their linkage to IHL more broadly is demonstrated by the ways in which the seeds of cooperation underpinning the Ottawa Process were laid within the CCW process. A proposal tabled at Lucerne by Sweden, Egypt, Mexico, Norway, Sudan, Switzerland and Yugoslavia that

Emergence of the landmine regimes   53 included text on landmines, was rejected by the militarily significant states. Three of these states were to join the initial core group that drove the Ottawa Process. In the margins of the January 1996 CCW session, eight pro-­ban states (Austria, Belgium, Canada, Denmark, Ireland, Mexico, Norway and Switzerland) met with the ICBL to discuss future strategy.35 Further meetings of this group led to the Canadian offer to host the first Ottawa conference. A combination of applied research and coalition building that began in the 1970s gained traction within the Ottawa Process more than twenty-­five years later. Developing nations, participation and ownership Given the focus of discussions on technical characteristics, little emphasis was placed in the Lucerne or Lugano meetings on the experience of mine-­affected states. Making a point that was to be taken up to great effect twenty years later in the Ottawa Process, Prokosch points out that ‘the Lucerne Conference might have felt itself under more pressure to move toward banning the new weapons if representatives of the countries where they had been used had said more about their terrible effects’.36 This point is at least slightly contentious. One line of reasoning among developing states supported the use of landmines for their (low cost) equalizing effects against well-­armed aggressors. A similar argument was deployed over the use of booby traps, resulting in a compromise definition in Protocol II which only prohibits ‘perfidious’ booby traps.37 It is important to recognize that compromises in treaty language are not only a result of the interests of militarily significant states, but are also a reflection of different developing world concerns. The extensive use of landmines against civilian populations in the 1980s and early 1990s during civil wars not covered by Protocol II in Angola, Mozambique and Afghanistan (while the conflict was international in character it was treated as internal by the Soviet Union) gave little incentive for ratification by developing states. Although APII addressed this lacuna, the problem, as expressed by the ICRC, was that ‘poorly trained or equipped forces may be unwilling or unable to abide by a complex set of rules or pay an increased price for self-­ destructing mines.’38 The absence of input from mine-­affected states or the developing world more broadly was thus a major criticism of the CCW process. Moreover, a key argument against using the CD as the venue for negotiations on landmines was that if the major producers and users of landmines were present, the majority of mine-­affected states were excluded. In contrast, the needs of developing states are at the heart of the humanitarian rationale underpinning the Ottawa Process. The process advocated an inclusive approach focused on building support in regions affected by landmines. The selection of the South African Jacob Selebi as Chairman of the Oslo negotiating conference was symbolically, as well as practically, significant. The media campaign led by the ICBL and ICRC used vivid depictions of mine victims to highlight the human face of the issue. Indeed, many survivors were brought to the various meetings to bear witness through their injuries. Ban campaigns in the

54   Emergence of the landmine regimes global South were an integral part of the Ottawa Process. NGO-­driven meetings in Mozambique and South Africa focused attention on Africa and garnered widespread state support within the region. The case of the Cambodia campaign is also instructive. Launched in 1994, the 3rd International ICBL Campaign Conference took place in Phnom Penh in June 1995. This was the first landmine conference to be held in a mine-­affected country. It was also the first major ICBL event to be organized primarily through the medium of email. Cambodian campaigners engaged in a national signature drive and raised public awareness of the ban movement. The conference was attended by over 450 participants from more than forty countries and provided a model for other national campaigns and signature drives.39 The ICBL promoted regional support for the ban through the creation of ‘mine free zones’. The states of Central America were followed by the Caribbean Community (CARICOM) in committing themselves to no further use, production, trade or stockpiling of APMs. Southern Africa proved a particularly coherent regional bloc, while the Organization of African Unity (OAU) and the Organization of American States (OAS) also played important roles in mobilizing support. This regional solidarity bore fruit at Oslo. Central American and African States offered cohesive support for a strong and effective legal instrument, avoiding the emergence of damaging North–South policy splits. The commitment of the global South – particularly mine-­affected states – lent credibility to the Ottawa Process that was absent from the CCW. However, a less prominent feature is the narrow control of the treaty drafting process by the core group of states and the ICBL steering committee prior to the Oslo negotiating conference. Given that the treaty text was not significantly amended during the negotiations, this offers an alternative narrative on developing world input to the process. The absence of a formal verification mechanism, coupled with a lack of substantive buy in to the specifics of regime design, raises important questions considered in subsequent chapters on levels of commitment to implementation engendered during the regime formation phase. Non-­state actors: the ICRC–ICBL nexus The space accorded to civil society varied widely across the two regime formation processes. It is clear that civil society faced challenges of both access and influence within the CCW framework. In contrast, the transnational NGO coalition that came together to form the ICBL is seen as the centrepiece of the Ottawa Process. Contrary to the ICRC, the ICBL was able to directly criticize governments through its advocacy. The neutral, impartial status of the ICRC meant that the organization transcended the limitations imposed on NGOs. The ICRC thus held a unique position, acting as honest broker by facilitating inter-­state negotiations across both processes. Price notes that: the most basic effect of civil society, then, has been the transnational dissemination of information about the scope of landmine use and its effects,

Emergence of the landmine regimes   55 thereby helping to define the use of AP landmines as not only a problem but as a global crisis.40 The strength of the ICBL lay in its structure, with a small international staff providing direction and coordinating the policy line on behalf of hundreds of local organizations around the world. Effective lobbying was central to this process, combining engagement with political leaders and government officials with public awareness campaigns. As Hubert points out, it is important to distinguish between the ICBL steering committee – which provided strategic direction – and the wider coalition that formed national campaigns and made the link to grass roots activists.41 The steering committee’s membership expanded over time and significant individuals were included on an ad hoc basis.42 The NGOs that made up the rump of the campaign – reaching over 1,200 NGOs in sixty countries43 – were cohesive, although disagreements were apparent between organizations preoccupied with lobbying governments and those more focused on mine action concerns. The ICRC’s role is commonly downplayed in favour of the ICBL and the Canadians. This can be attributed to a combination of the deliberately modest profile of the organization, and the near monopoly on the Ottawa Process literature held by the ICBL and Canada. However, while the ICBL liaised with the like-­minded group and engaged many developing states, the ICRC, through its credibility and expertise, could also engage effectively with militarily significant states. This was critical to building trust in an IHL regime formation process that lacked both precedents and safety valves that would have emerged through consensus decision-­making. At the first CCW negotiating conference the ICRC swam with the current, supporting proposals on restrictions. The accepted wisdom was that pro-­ban advocates were unrealistic.44 At the 1995–1996 review conference the ICRC was again invited to act as expert observer. This status meant the ICRC could speak, submit proposals (they provided two working papers) and distribute documentation.45 In this role, the ICRC could exert influence and deploy subject area expertise. This was critical in shifting perceptions. ICRC field surgeons were among the first to highlight the impact of APMs on civilians. Expert conferences and studies on the effects of APMs, initiated by the ICRC, provided a bridge between the mine action community, militaries, government officials and NGO advocates.46 Arguments relating to military necessity, which underpin the CCW process, were addressed head-­on. The ICRC commissioned an analysis by a retired senior British military officer with a background in combat engineering, weapons research and humanitarian demining.47 By highlighting the gap between questionable military gains and terrible humanitarian impact, the study challenged the military effectiveness of APMs, striking a chord among military and non-­military audiences alike. The ICRC, together with the ICBL, was responsible for the concerted international campaign to stigmatize APMs. Despite the ‘without precedent’ discourse underpinning much of the Ottawa Process literature, this was not the ICRC’s

56   Emergence of the landmine regimes first ever such campaign. Similar efforts successfully contributed to the stigmatization of chemical weapons following the First World War.48 However, the convergence of interests and the possibilities offered by modern communications technology ensured that the anti-­APM campaign had a global impact within a very short timeframe. The non-­adversarial approach of the ICRC offered an effective foil to the ICBL. This aspect of the process has particular resonance with the role of the IRC during earlier IHL processes in narrowing gaps between the positions of states and civil society. Moreover, as the acknowledged guardian of IHL, the ICRC made substantive contributions throughout the drafting process – for example on the need for unambiguous definitions – which proved particularly important. The ICBL was given a campaign headquarters and meeting rooms (provided through the good offices of ICBL member Norwegian Peoples Aid) located directly opposite the negotiating hall. This physical location facilitated the direct link between state and civil society actors that distinguishes the two regime formation processes. The ICBL held official observer status on the same terms as observer governments. They were present at all sessions and could make oral interventions at any point, although they could not vote or formally propose treaty text. Underlining a central plank of the Ottawa narrative, Williams and Goose claim that ‘this is the first occasion on which NGOs were given official status in international negotiations of a disarmament/arms control or humanitarian law treaty’.49 The regime framework certainly allowed for wider participation than the CCW. But this is better understood as a return to the roots of IHL with a prominent role assumed by civil society. While novelty is an important element of the mine ban treaty brand, this narrative ignores a long history of humanitarian advocacy by both state and non-­ state actors. In a survey of delegates commissioned by the Canadian government, NGO pressure, particularly at the negotiating table, was acknowledged as a key influencing factor.50 The same survey cited government officials as the most significant obstacle to signature of the treaty and middle powers as having a particularly positive role. However, although the survey results seem to reinforce the importance of civil society in the process – a central message of the prevailing Ottawa Process discourse – it is not mentioned in To Walk Without Fear, which draws on these figures: that only 45 per cent of survey respondents were from governments and of these, 70 per cent came from regions outside of Europe and North America.51 The survey group is therefore unrepresentative in terms of governmental participation. The two strands of the Ottawa Process were reinforced by a number of events, activities and opportunities seized by individuals and institutions. Support within the UN system for a ban on APMs came from the highest political level. UN Secretary-­General Boutros Boutros-­Ghali became increasingly explicit in his support for a ban, first in the 1992 Agenda for Peace and subsequently in his foreword to the Proceedings of a 1995 symposium on landmines.52 The legitimacy of the process was reinforced by the landmark December 1996 UNGA Resolution 51/45S, calling on states ‘to pursue vigorously an effective, legally

Emergence of the landmine regimes   57 binding international agreement to ban the use, stockpiling, production and transfer of anti-­personnel landmines with a view to completing the negotiations as soon as possible’.53 Other significant factors included the World Bank, in August 1997, agreeing to fund mine clearance activities for the first time in its history. The timing of these supportive noises emanating from different international organizations, contributed to the sense of momentum surrounding the Ottawa Process, while also lending credibility to an unconventional approach to IHL treaty formation. From a regime perspective, they reinforce the nesting of the mine ban treaty through associating it with wider humanitarian and developmental goals. Stakeholder clustering and regime effectiveness In contrast to earlier practice, at the insistence of the Chinese delegation, NGOs were not even granted observer status in the APII negotiations.54 Attendance was limited to states with the ICRC and UN agencies invited as special cases.55 Organizations with extensive practical experience of landmines were excluded. The narrow knowledge base of the experts involved in APII regime design resulted in little questioning of the logic that these weapons were militarily useful without being indiscriminate. This was decisive in delimiting the language and scope of the protocol. The absence of civil society from the CCW negotiations does not mean that they failed to influence the process. From the outset, the French government’s request to initiate a review conference had been the result of a lobbying campaign by the French NGO Handicap International.56 During the conference, over 100 experts from seventy NGOs monitored the negotiations and lobbied in the margins.57 NGO members also shaped national policy processes through being invited to form part of official delegations. The Canadian NGO, Mines Action Canada, had strong links with the division of the foreign ministry responsible for the negotiations. For both Australia and Canada, NGO members formed part of official delegations.58 Even with the restrictive CCW rules of procedure, recognition of the benefits of subject matter expertise allowed some organizations to influence proceedings. If there may have been overlaps in the makeup of national negotiating teams across the two regimes, the phenomenon of regime learning is a unique feature of the Ottawa Process. Although diplomats, officials, military officers and lawyers brought their own expertise – to a great extent these were the same people involved in the APII negotiations – exposure to the first hand evidence of mine victims, humanitarian assistance workers and mine action practitioners led to a new appreciation of the magnitude of the problem and the measures needed to address it. This knowledge proved hugely influential in shaping the thinking of actors responsible for policy at a national level. The link forged between the ICBL and like-­minded states proved an effective influencing mechanism. Pressure was applied both to western and developing nations. Many states were prepared – or pushed – only six months after agreeing

58   Emergence of the landmine regimes APII, to move much further in signing up to a ban on APMs. The highly visible campaign to raise awareness on the humanitarian cost of these weapons had generated sufficient traction within a more open process of regime formation to enable such an outcome. This link was reinforced through a number of states including ICBL members in their delegations. As the former director of the Arms Division within Human Rights Watch, Ken Anderson, points out: in some cases that meant that some countries essentially handed their policy and negotiating apparatus to activists; in other cases the government had its own line, usually sympathetic but not identical to the NGO position, but allowed the NGO inside access to the negotiations.59 The clustering of actors within the Ottawa Process was determined by self-­ selection: if you agreed with the goal of a complete ban on APMs you could join. This approach had the unintended consequence of inviting spoilers within the tent. As it became apparent during the course of 1997 that the CD track was failing to erode support for the Ottawa Process, the US on 18 August, announced its intention to participate fully at the Oslo negotiating conference. Although, in principle, embracing the Brussels Declaration indicated their support for a ban, in practice this move allowed the US to more freely apply pressure on participants. However, at the Brussels meeting, strong arm lobbying by the US delegation – inviting other delegations to their hotel to press their position – was effectively countered by the ICBL tactic of briefing delegations going in and de-­ briefing them coming out.60 Looking beyond the open model for participation, a clear division of labour can be discerned within the mine ban treaty approach. In the run up to the Oslo negotiating conference, NGO activities in states with a tradition of civil society activism combined with state-­driven efforts to prepare for the negotiating conference. These were interlinked and coordinated with organizations playing to their strengths. On a political level, the credibility of the actors within the Ottawa strand moved states beyond ‘the committed’ to join the regime. Practitioner experience proved particularly effective in countering more minimalist positions that sought a nuanced outcome rather than an outright ban on APMs. The absence of such diversity among its proponents meant that APII could not transcend the knowledge base of its existing membership. A related cost of the narrow APII constituency was that it became easy for the Ottawa cluster, with its strong advocacy base, to shape the political landscape through painting the other regime as a failure. Leadership Both regime formation processes are characterized by clusters of state and non-­state actors. Oran Young’s distinction between structural, intellectual and entrepreneurial leadership provides a useful way to cast these actors into relief and to identify similarities and distinctions between the two processes.61 If this perspective implies

Emergence of the landmine regimes   59 attributing influence to individual actors, it is also important to understand how these leadership roles are clustered within and across stakeholder groups. While the Canadians were supported by the like-­minded group, their structural leadership role within the Ottawa Process is undisputed given the political risks uniquely taken by Axworthy and his officials as well as their initiative to partner openly with the ICBL. Canada was also the only actor that could both draw other states into the like-­minded group and mitigate (in concert with other states) the efforts by spoilers to derail or divert momentum. Structural leadership in the APII framework was provided by the US. This role was pursued with the clear goal of preserving the status quo and promoting only incremental change. They were the spoiler in the Ottawa Process, while at the same time seeking to demonstrate leadership in the international community on the landmines issue. Despite US efforts to shape the outcome of the negotiating conference away from a complete ban, their intervention, somewhat counter-­intuitively, actually proved helpful in achieving this result. The major changes sought by the US and the heavy-­handed way they went about achieving them offered a stark ‘with us or against us’ choice to states. Without this dynamic the negotiating conference may well have become bogged down in points of detail. Instead, contrary to expectations, given the sensitive issues under discussion and the rapid nature of the process, the draft treaty actually became stronger rather than weaker through the course of the negotiations.62 A major distinction between the two regimes lies in the area of intellectual leadership. If the intellectual content of the Ottawa Process was drawn from a number of sources – notably mine action practitioners and mine-­affected states – intellectual leadership came from the ICRC. The organization’s structure and credibility was instrumental in persuading many states to commit to a highly unconventional process of regime formation. For the mine ban treaty, this leadership was grounded in both IHL and mine action expertise. In contrast, beyond the ICRC, the knowledge available to shape the APII regime was based on arms control and disarmament experience. This narrow vision, reinforced by constraining regime rules, did not permit a focus on the realities of contemporary landmine use, their impact, or measures to address them. Entrepreneurial leadership in the Ottawa Process was provided by the steering committee of the ICBL. This group moved beyond a classical advocacy position of criticizing the behaviour of governments, to coordinating a two-­level campaign that combined collaboration with like-­minded states and a massive mobilization of national and international civil society organizations. Different NGOs found reasons to support the campaign within their own core mandates. Whether driven by human rights or IHL, development, medical or public health concerns, the process offered something for everyone.63 National level ICBL efforts to raise public awareness and pressure governments would then shift as closer cooperation was established. By contrast, the formation of APII is characterized by an absence of entrepreneurial leadership. Within a CCW regime framework that had been established for fifteen years, no actor was willing or able to broker deals that would lead to transformational change. The strict rules

60   Emergence of the landmine regimes governing participation of non-­state actors prevented such a role, while encouraging groups to look for alternatives within the Ottawa Process. Agency dynamics across the two regimes highlight the significance of the committed state-­civil society cluster. APII regime rules that excluded civil society proved ineffective in practice, as was the case with the negotiation of the 1925 Gas Protocol. However, if self-­selection by the like-­minded was a defining characteristic of the Ottawa Process, decision-­making was in fact coordinated within a much narrower group of actors. This apparently open process, in reality underpinned by tightly controlled decision-­making, calls into question the practical ability of expert communities and mine-­affected states to influence regime development. Funnelling of authority to the tip of the ICBL pyramid has consequences that play out in implementation.

The influence of norms on regime formation Regime nesting Supporters of the respective landmine treaties tend to underline the distinctions between the two approaches. Yet common normative underpinnings may provide significant if less visible linkages between the regimes. The ‘umbrella’ CCW treaty explicitly situates the landmine issue within a broader normative framework regulating weapons that cause unnecessary suffering or superfluous injury. The APII regime is an example of a specific treaty based on wider principles such as responsibility for clearance, provision of technical information and protection of civilians. However, if the concepts of indiscriminacy, superfluous injury and unnecessary suffering which underpin the CCW provide an important link to the broader corpus of IHL, they are of limited utility in their application to specific weapons. IHL does not provide a clear definition of the scope and application of these terms; as Kalshoven points out, these principles ‘are not particularly suited to serve as yardsticks for the legality of a weapon’.64 Even if provisions are derived from important IHL norms, this does not mean they will be effective in practice. The downstream effectiveness of APII needs to look beyond implementation to encompass the fulfillment of regime goals. The mine ban treaty embodies an emerging norm stigmatizing the use of APMs. The regime is freighted with a specific acknowledgement that landmines are unacceptable as a weapon of war. As a consequence, a ban offers the only solution that falls within the standard of customary IHL. What Price terms the ‘grafting’ of the emerging anti-­APM norm onto the broader norm against inhumane weapons65 greatly strengthened the pull of the Ottawa Process. The emergence of the ban campaign and its support by national governments is intrinsically linked to the perceived failure of the CCW, from a humanitarian perspective, to adequately meet this standard. Once states internalized the fact that there was a humanitarian emergency – at the same time that the CCW had failed to adequately address it – the argument in favour of a complete ban was sealed.66 This demonstrates a mutually reinforcing dynamic between the perceived failure to address the humanitarian costs

Emergence of the landmine regimes   61 of these weapons within the APII process, and the increasingly compelling stigmatization narrative of the Ottawa Process. Indeed, the technical nature of the APII negotiations, and its widely acknowledged failure to address humanitarian concerns, conferred a morally ambiguous status on its supporters while serving to boost the legitimacy of the Ottawa track. The humanitarian deficit of APII had a tangible effect in increasing state support for the mine ban treaty. The UK and France regarded APMs as a useful military capability but, despite heavy pressure from the US, still signed up to the ban. Political pressure ramped up by the civil society-­driven advocacy campaign – exacerbated in the UK as a result of the political shockwaves from the death of Princess Diana experienced by the new Labour Government – was a key factor. These cases illustrate the highly political nature of the regime formation process. Following Putnam’s model of the two-­level game, internal and external pressures combined to tip the balance of interests for many states in favour of joining the regime. Common to domestic and international discourses was the need to be seen to be on the right side of a high profile humanitarian issue. Certainly, the rapid progress from identifying the need for a mine ban treaty to forming the regime seems to fit the description of a norm cascade,67 with many states feeling obliged to follow the example of others and do the right thing. The Ottawa Process isolated APMs as unacceptable and provided a means to do something about them. The campaign derived its strength from a single focus that would have been lost had the humanitarian imperative been diluted. The message of the Ottawa Process was therefore closely linked to its normative pull. Anderson notes that ‘this utter moral and political clarity was an integral part of the campaign in reaching various publics’.68 Efforts by the Landmine Survivors Network (LSN) to include wording to assist victims in the draft treaty text, was regarded by some ICBL members as a distraction from the overall goal of a ban. LSN was criticized for not going through the steering committee in pursuing its goal.69 While language was eventually included, this was largely due to the strenuous efforts of LSN, the ICRC and Handicap International rather than it having been an original ICBL campaign goal. This example illustrates an important distinction between humanitarianism and humanitarian norms. The cold-­eyed realism of the ICBL leadership was demonstrated by its reluctance to include victim assistance provisions in the draft treaty text, despite compelling humanitarian arguments, because this might draw opposition from certain states and therefore complicate the negotiating process. The norm entrepreneurship guiding the campaign was therefore not purely idealistic. It recognized both the utility of the anti-­APM norm in compelling as many states as possible to join a regime centred on a ban, as well as the countervailing costs of expanding the regime’s scope and blurring this key message. The timing of related initiatives helped reinforce the normative weight of the Ottawa Process. The landmark 1996 UNGA Resolution, as well as the evident support for the ban movement by both famous figures and ordinary people in their thousands from all corners of the world, generated momentum that positively influenced the orientation of states. As already stated, the tragic death of

62   Emergence of the landmine regimes Diana, Princess of Wales, in a car accident just before the Oslo conference, greatly increased media attention due to her high profile support for a ban.70 Finally, the awarding of the Nobel Peace Prize jointly to the ICBL and its coordinator Jody Williams in October 1997, gave a boost to the signature drive between the conclusion of the negotiating conference and the subsequent signing ceremony in Ottawa. Japanese Foreign Minister Obuchi acknowledged that the awarding of the Nobel prize was a significant factor in Japan reconsidering its policy and signing up to the mine ban treaty.71 For Price, the commitment of mine-­affected states to the ban process signifies that ‘unlike some international norms such as those embodied in human rights instruments, the treaty has support where it is needed most – in countries where the pernicious effects of the now-­deviant practice have been most prevalent.’72 One result was the forging of links between governmental and civil society actors in the global South. In the case of South Africa, Noel Stott, a member of the South African Campaign to Ban Landmines, describes the importance of ‘critical solidarity’, with the discourse of the South African government influenced by an internationalist concern to signal the country’s return to the international community through embracing this humanitarian norm.73 This provides a parallel with EU applicant states that were keen to be seen to embrace the anti-­APM norm even though, for some, landmine production remained a significant economic activity. In both these cases, support grew through tapping into a broader normative imperative to display responsible international behaviour. Analysis of norms in regime formation illustrates important ways in which the two landmine regimes are nested within a wider IHL regime. Moral clarity revised the calculus for mine ban treaty membership based on humanitarian rather than security or economic criteria. It also extended beyond IHL to encompass norms relating to state legitimacy by drawing on the need to demonstrate appropriate conduct in international relations. Normative considerations also highlight a distinction within the Ottawa core group. As made evident by different attitudes to the inclusion of victim assistance provisions, while the mission of the ICRC is to promote humanitarian goals, the ICBL built on this humanitarian imperative to achieve its policy goals. As discussed in subsequent chapters, this distinction is not without consequences.

Conclusions Complex, fluctuating transnational dynamics shaped the design of the landmine treaties. Regime nesting enabled the emergence of APII. The regime demonstrates the progress states were prepared to make in 1995–1996 on the restrictions they had previously agreed in 1980. It also represents a chain of argumentation within the field of IHL that was launched as early as the 1950s. In contrast, even if there are important, under-­emphasised parallels to the wider IHL discourse, the dynamic behind the Ottawa Process is distinct from conventional processes of IHL treaty development in several important ways. Speed is

Emergence of the landmine regimes   63 one such trait. The intensely political nature of the regime formation process is shown by the fluidity of positions for many states in the face of a combination of domestic and international pressure to join the regime. In contrast to APII, many states involved in the Ottawa Process were compelled to revisit preconceptions about the costs and benefits of membership according to a humanitarian yardstick. Normative considerations highlight a key distinction between the regimes. APII attracted a nucleus of participating states, while the Ottawa Process transcended such a fixed grouping by exploiting push and pull dynamics that drew on the anti-­APM norm to oblige support from a much broader constituency. Stakeholder clustering dynamics clearly play an influential role in shaping landmine regime formation processes. The active engagement of representatives from mine-­affected states and expert communities was particularly significant in terms of the process and substance of mine ban treaty regime formation. The former, mobilized both at state level and through national civil society organizations, gave the campaign its global character and added greatly to its legitimacy. Mine action practitioners contributed substantive inputs that effectively countered dissenting arguments. The absence of these stakeholder groups from the APII design phase points to an important disjunction between the development of the regime and its capacity to draw on relevant expertise. Contributions from practitioners and mine-­affected states thus constitute key variables in accounting for differences in regime design. Unlike many successful regimes in other issue areas, inflexibility in refusing to dilute the core ban has been decisive in maximizing political support for the mine ban treaty. Rather than seeking compromise solutions, the integrity of the  anti-­APM message was critical in shaping state behaviour. In contrast, the absence of flexibility in rules governing participation or the modalities for decision-­making within the CCW – while admittedly providing a level of confidence for militarily significant states to engage in the regime – also comes at a cost. Resulting gaps in intellectual and entreprenneurial leadership may constrain regime learning if carried into the implementation phase. This is very different from the Ottawa Process. A unique combination of structural, intellectual and entrepreneurial leadership strongly influenced the political dynamics of the regime formation process. The juxtaposition of the two regimes was a highly successful tactic in promoting the Ottawa brand. However, both the massive early support for the mine ban treaty and the lack of amendments to the draft text during the negotiating conference, suggest that states may not have paid sufficient attention to the implications of their obligations. The apparent contradictions within the US position – Ottawa spoiler but single largest global mine action donor – shows the complex relationship between regime and mine action objectives. Regime membership is not a prerequisite for playing a positive international role in mine action. The costs of membership and potential gaps between regime and mine action goals thus present key foci for an analysis of implementation and effectiveness.

4 Implementing the landmine regimes

Introduction APII entered into force on 3 December 1998 followed by the mine ban treaty on 1 March 1999. In both cases, sufficient time has passed since regime formation and entry into force to construct a meaningful analysis of implementation and effectiveness to date. Resource mobilization, donor coordination and concrete contributions to the various activities comprising mine action all, in qualitatively different ways, represent mine action related objectives of the two regimes. Linking implementation to the various functional sectors of mine action in this way enables a clearer understanding of the extent to which the regimes reflect mine action concerns and priorities. As in the regime formation phase, implementation is characterised by a lack of source material for APII and a plethora of often uncritical narratives on the mine ban treaty. At the same time, a significant parallel literature has been developed by practitioners on approaches, good practice and lessons learned in mine action. Because work on treaty implementation and mine action have not been drawn together, meaningful findings on the effectiveness of the regimes have been slow to emerge. As Filippino and Paterson note: The gap between social scientists and practitioners is unclear. For example, the human toll exacted by landmines and UXO was the principal impetus behind the international movement to ban landmines. But what do we know about the contributions made by clearance, marking and mine awareness to a reduction in the number of deaths and disabilities? We know very little, at least in quantitative terms.1 This chapter addresses the ways in which landmine regime implementation is making a positive contribution to mine action. The relationship between policy, agenda setting and programming roles, as well as the influence of mine action practitioners and mine-­affected states, are considered. Some actors, such as the ICBL in the context of the mine ban treaty, are wholly situated within the regime. Others such as the GICHD have been at least partially assimilated into the implementation processes. A much broader category of state and non-­state

Implementing the landmine regimes   65 actors contribute in different ways to implementation either across the two regimes or favouring one over the other. In order to take into account all actors that play a part in implementation, it is essential to acknowledge the roles played by actors outside the regimes senso stricto, including non-­members as well as groups not directly addressed by the treaties. The nature and utility of formal and informal implementation mechanisms is assessed, and challenges of voluntary and involuntary defection are considered. Regime effectiveness is then analysed under the headings of resources, political will and humanitarian impact.

Understanding landmine regime implementation APII places a series of technical restrictions on APMs, AVMs and booby traps, prohibits undetectable mines, applies conditions on their use in combat, and obliges states parties to clear, remove, destroy or maintain them in accordance with the protocol. The mine ban treaty stipulates a complete ban on the use, stockpiling, production and transfer of APMs, as well as providing for their destruction. In order to implement these provisions the regimes must provide support to the five ‘pillars’ of mine action:2 1 2 3 4 5

Mine risk education Humanitarian demining Assistance to mine victims Destruction of stockpiles Advocacy

This introductory section relates key provisions within APII and the mine ban treaty to the component activities of mine action, providing a point of departure to situate the subsequent analysis of implementation and effectiveness. Mine risk education Through public information and education, mine risk education (MRE), also known as mine awareness, aims to make civilian populations better aware of the dangers posed by landmines. Landmine Monitor notes expanded mine risk education programmes in many countries and closer integration with other elements of mine action. Some mine action practitioners argue that MRE work has been de-­linked from the realities of mine affected communities. The value of this activity therefore remains contested, and advances have been limited in terms of its professionalization.3 Activities to reduce the risk of death and injuries to civilian populations are emphasized in both regime frameworks. Mine ban treaty states parties are requested to provide assistance in this area and may themselves request help from the UN, regional organizations, states parties, and other relevant actors (Articles 6 (3) and 6 (7d)). Article 3 of APII requires the protection of civilians through ‘possible’ measures that include ‘fencing, signs, warning and monitoring’. Moreover, ‘unless circumstances do not permit’, effective

66   Implementing the landmine regimes advance warning should be given of the emplacement of landmines, or their air delivery. Humanitarian demining Since the predominant threat to individuals and communities comes from mines already in the ground, humanitarian demining represents the most direct means to realize the humanitarian goals underpinning the two regimes. The mechanics of humanitarian demining in its most basic form have not changed since the Second World War: a metal detector is passed over a given piece of land until a signal is emitted; once this happens the land around the signal is excavated until either a piece of waste metal is discovered or a mine is found and blown up in situ, using explosives, or removed for destruction elsewhere. Various aids to the demining process, such as mechanical equipment or mine-­detecting dogs, have been developed over time but only as a complement to the individual deminer. Humanitarian demining (including mine and unexploded ordnance survey, marking and clearance) is an obligation on parties to a conflict under APII and on regime members (i.e. mine-­affected states) in the mine ban treaty. While APII puts the onus on parties to a conflict to clear mined areas following the cessation of hostilities, no explicit link is made to humanitarian demining. In contrast, the mine ban treaty places responsibility on states parties, assisted by the international community, to clear mined areas within ten years of their ratification of the treaty. Victim assistance Assistance to mine victims remains the subject of controversy within the mine action field. There is an unresolved question: whether this activity is genuinely a mine action issue, or better addressed through broader humanitarian and development work to address public health concerns. The mine ban treaty is the first example within IHL of a treaty that contains specific measures to assist victims of the weapon in question: Article 6 (3) requires international cooperation and assistance for mine victims. However, ambiguity is inherent to the provision because the term ‘victim assistance’ is not defined, nor are guidelines laid down for its implementation. Assistance to landmine victims is not directly addressed in APII, although states parties provide information on this activity as part of their annual reporting requirements. Stockpile destruction Stockpile destruction is the most visible way to reduce the number of landmines around the world. Destruction of stockpiles within a set timeframe is stipulated under APII for prohibited weapons and for all APMs under the mine ban treaty. Tens of millions of stockpiled APMs, destroyed in accordance with the requirement to eradicate such stocks within four years of entry into force, represents a key achievement of the regime (although this must be set against the much

Implementing the landmine regimes   67 higher numbers still held by non-­states parties). The destruction of stockpiled mines that are not in conformity with the provisions of APII, also constitutes a significant obligation for regime members not party to the mine ban treaty. Advocacy Advocacy is recognized as essential to wider adherence and respect for the regimes, as well as to raise awareness of the threat from landmines. Advocacy has been at the heart of the Ottawa Process and remains an important feature in the implementation phase. Universalization is a distant objective, with militarily significant states such as China, Russia and the US, as well as much of Asia and the Middle East, remaining outside the regime. The ICBL was at the heart of the advocacy campaign supporting the mine ban treaty regime process, although concerns over whether the organization has adapted to the qualitatively different demands of treaty implementation have resulted in some restructuring. In the absence of a coherent civil society initiative to support the CCW process, the United Nations Mine Action Service (UNMAS) lobbies states to ensure that they adhere to APII. UNMAS has also supported mine ban treaty universalization efforts by working with government representatives, notably in the Middle East and Asia, where regime penetration is weak. Relating compliance to design: formal monitoring and verification Both regimes hold annual meetings and require reporting from members to demonstrate progress on implementation issues. The mine ban treaty has established a programme of regular intersessional meetings. Within the CCW an experts group has a comparable mandate, although on a much smaller scale. Outside of the formal regime frameworks, various state and non-­state actors have found ways to assist implementation. One significant example is the Implementation Support Unit (ISU) created by the GICHD to support the mine ban treaty, but which later provided certain support services within the framework of the CCW as well. Another is the Landmine Monitor, an annual publication coordinated by the ICBL, which represents the de facto compliance monitoring mechanism under the mine ban treaty. Landmine regime implementation mechanisms are summarized in Table 4.1. Table 4.1  Landmine regime implementation mechanisms Implementation mechanism

APII

Mine ban treaty

Meetings of regime members Work programme Review conference Reporting Secretariat Compliance monitoring

Annual Yes As agreed by majority Annual No No

Annual Yes Every 5 years Annual Yes Yes

68   Implementing the landmine regimes There is no designated compliance monitoring body within either landmine regime. This responsibility rests firmly with individual states parties. In both APII and the mine ban treaty, members are accountable through the provision of annual reports. Reporting is intended to provide a basis for action when implementation challenges become apparent. As part of this obligation, regime members must submit detailed information on progress in meeting mine action-­ related implementation goals. Article 13 of APII requires reporting on mine clearance, national implementing legislation, technical information exchange and other cooperation as well as details of landmine production, stockpiling and use. In the case of the mine ban treaty, self-­reporting by states in line with the requirements of Article 7 represents the main monitoring mechanism. Providing information on the location of mined areas as a means to support humanitarian demining activities is one element of this commitment. Annual meetings of mine ban treaty states parties provide a forum for discussion on implementation issues. These meetings are complemented by an intersessional work programme involving states (regime members and non-­members) the UN, ICRC, ICBL and others. The programme was established in 1999 with individual standing committees created on victim assistance and socio-­economic reintegration, mine clearance, MRE and mine action technologies, stockpile destruction, and the general status and operation of the treaty. Standing committees were subsequently rationalized with meetings reduced in frequency, in 2005, on the basis of a decision taken at the first treaty review conference.4 Despite the clear linkage between treaty implementation and the various pillars of mine action, some mine action practitioners regard meetings as more significant for awareness-­raising among the diplomatic community than for bringing tangible benefits to mine action. The intersessional meetings, in combination with the scrutiny coming from Landmine Monitor, have certainly put additional pressure on regime members. The level of Article 7 reporting has risen consistently in the years since the treaty entered into force: statistics show a year on year increase in reporting between 2001–2007.5 However, this trajectory reflects the rising numbers of states signing the treaty. A more meaningfulful indication of states parties’ commitment may be found in the proportion of regime members providing annual updates to these reports. The rate for the calendar year 2009 is 56 per cent, representing a consistent trend of decline that has continued every year since 2003. Equatorial Guinea is eleven years late with its initial Article 7 report!6 APII compliance reporting is not linked to any monitoring mechanism. In contrast, the mine ban treaty provides for fact-­finding missions to investigate allegations of non-­compliance. The modalities found in the treaty for these missions are lifted almost verbatim from the text of the CWC, reflecting the model used during the treaty drafting process. Verification provisions do not therefore focus on the specific characteristics of APMs. In particular, the regime fails to take into account that while large-­scale use of landmines may be easily identifi­ able, not least as a result of the nature of the injuries provoked by these weapons, landmine stockpiles can be relatively easily hidden. A significant challenge to

Implementing the landmine regimes   69 compliance verification therefore goes unaddressed. A lack of practitioner input in this aspect of regime design thus has direct consequences on its ability to verify compliance. The mine ban treaty has not put measures in place to give teeth to its verification mechanism. No standing secretariat has been established, nor has provision been made for the conduct of such missions within the UN system. Moreover, the regime does not have established procedures in place should an FFM be requested. Despite the extensive intersessional work programme, the Article 8 mechanism has been characterized as modest from a disarmament perspective.7 This lack of emphasis on verification may be explained to an extent by the evident reluctance among states to ‘break from the pack’ in political terms. An FFM would need to be initiated on the basis of an accusation of non-­compliance by one state party against another. Regime members have shown no willingness since entry into force of the regime to do this. The ICBL maintains that ‘a mechanism or body is needed to facilitate attempts to address compliance concerns short of formally invoking Article 8’.8 Many states, including core group members, are less keen. A legalistic response given to this inherently political question was that regime members should ‘be prepared to respond to all serious allegations of non-­compliance within the core provisions of Article 1’.9 The implementation process presents distinct challenges of political will to those evident in the regime formation phase. FFMs provide a specific example of a broader tension within the mine ban treaty regime, between member states seeking to maintain authority over the implementation process and the efforts of civil society to proactively enforce compliance. Despite implementation problems being highlighted as a result of compliance monitoring, the verification mechanism has not been engaged. In fact, the regime has raised, rather than lowered, the political transaction costs of addressing overtly the sensitive issue of defection. In the absence of compliance monitoring provisions, tensions between states parties and other regime stakeholders are less in evidence within APII. The idea of adopting procedures along the lines of Article 8 of the mine ban treaty has been raised as both practically useful and a means to facilitate national implementation.10 It could also build synergies and provide for potential economies of scale across the two regimes. Unfortunately, the political barriers in the way of developing robust compliance monitoring provisions are even higher in APII than in the mine ban treaty. Within a framework governed by consensus, whose membership includes states displaying high concern for their own national security, agreement on measures for more intrusive verification seems unrealistic. Filling the gaps: informal compliance monitoring Formal implementation mechanisms – review conferences, annual meetings of states parties and intersessional work programmes – are intended to facilitate knowledge transfer and highlight where assistance to members should be focused or redirected. The provision of accurate and timely information is a

70   Implementing the landmine regimes common regime requirement that can support mine action through providing a clearer picture of the challenges posed by landmines in different national contexts. While not envisaged in the design of either regime, informal civil societymonitoring and verification has become an integral part of the mine ban treaty implementation process, at the same time generating important effects for APII implementation. In particular, civil society-­driven efforts have shed light on instances of defection from obligations that the regimes themselves have chosen to ignore. Landmine Monitor describes its role as ‘an attempt by civil society to hold governments accountable to the obligations they have taken on with respect to anti-­personnel mines’.11 Since 1999, its annual reports have provided the major source of information on mine ban treaty compliance issues. The initiative comprises a global reporting network, a central database and the annual publication. Beyond national reports, the publication includes updates on new signatories, areas of special concern and recent developments. Analysis is drawn from in-­ country researchers and augmented by external experts. By its own admission, the quality of the data in Landmine Monitor has been variable with discrepancies from country to country depending on individual researchers, although there has been a marked improvement over time. Bilateral and multilateral donors provide funding that allows for the research, publication and dissemination of the report. This support permits wide penetration of the report’s findings: the publication is distributed at no cost to government officials and policy makers and is fully accessible online. This civil society-­driven initiative fills an evident gap. Governments recognize the need for such a mechanism and have been prepared to support it year on year. The funding base for Landmine Monitor has been constant, despite the fact that entries have been critical of aspects of national compliance by the donors themselves.12 This points to an important advantage; it is easier to highlight sensitive implementation issues through informal verification than by invoking the formal apparatus of the regime. Landmine Monitor offers a norms-­based approach to compliance monitoring. Naming and shaming has led to important clarifications and provided advance warning where regime members appear unlikely to meet obligations. If this informal mechanism has generated political pressure within the regime that has not been realized through formal mechanisms, an important conduit does exist between formal and informal processes. The ICBL provides oral and written input to standing committee and annual states parties meetings. Whereas states are reluctant to criticize other states, the ICBL draws on the analysis generated by Landmine Monitor as an empirical basis for targeted advocacy. The task of focusing attention on states parties that risk missing compliance deadlines is facilitated because this is not a sensitive state-­to-state transaction. Informal implementation support mechanisms have not been replicated in the APII regime. Civil society has been permitted little space to contribute to implementation so shortcomings in regime design, such as the absence of verification measures, are not rounded out by informal means. Landmine Monitor does exert

Implementing the landmine regimes   71 an influence on APII regime implementation. By highlighting compliance issues to a wider international audience, it has created a positive knock-­on effect encouraging greater transparency within APII. There would also be significant scope for civil society to promote APII membership as a viable alternative for mine ban treaty rejectionists. These actual and potential entry points for civil society engagement demonstrate that, as in regime formation, formal rules are insufficient to prevent civil society from playing a proactive role. Expertise, ownership and regime implementation The development of the mine ban treaty was strongly influenced by the input of mine action practitioners and various mine-­affected states. APII design relied on arms control and disarmament experts, and reflects the interests of landmine users and producers. It is important to acknowledge these very different influences in considering how, explicitly and implicitly, the regimes engage with and address mine action concerns. If broad participation brought legitimacy and expertise to the regime formation process, substantive commitment becomes even more essential in the implementation phase. In supporting mine action, the most important stakeholders are mine-­affected countries. While largely ignored in the context of APII, these actors are made directly responsible for mine ban treaty implementation. The mine ban treaty has proved flexible in incorporating new mechanisms to support implementation. Based on the argument that resources ‘should be committed to the field where they were most needed, a secretariat was not initially deemed necessary’.13 However, the flawed logic behind this assumption became evident as requests multiplied for support from regime members. At the 2001 Managua meeting of states parties, an offer from the GICHD to host a secretariat – the ISU – was accepted. The ISU became operational in June 2002. It works directly with the chairs and rapporteurs of the intersessional committees, facilitates meetings of states parties, contributes to strategic thinking and, along with the ‘sponsorship group’, promotes (at least in principle) full participation of mine-­affected states. Through housing the ISU, the GICHD has become partially integrated within the regime. Given that the majority of the organization’s staff are technical experts involved in mine action research and operations, co-­location offers potentially important synergies. At the 3rd CCW Review Conference, which was held in November 2006, the GICHD was given responsibility for a new CCW sponsorship programme, paralleling one of its roles under the mine ban treaty. While many actors within the respective regimes have shown resistance to cooperation, this combining of secretariat roles could represent a first step towards developing more cooperative working relations across the two regimes. The ICBL is fully integrated within the mine ban treaty. The NGO coalition has a unique place in the (semi-­autobiographical) genesis story of the regime. However, the shift from regime formation to implementation has not been without challenges, requiring a partial move away from the organization’s

72   Implementing the landmine regimes a­ dvocacy roots. Advocacy remains a priority, finding expression in the promotion of treaty universalization. Virulent opposition to APII has also remained a constant over the years.14 By contrast, responsibility for Landmine Monitor has shown that transitioning from activism to implementation support is far from straightforward. Change is difficult since ‘the political activism that drove the campaign from its early beginnings is not necessarily compatible with the research and documentation demanded by a monitoring effort’.15 The partial shift of the ICBL, from entreprenneurial to intellectual leadership, is significant given the highly influential role of the organization in mine ban treaty implementation. By giving a voice to local actors, the Landmine Monitor network is firmly grounded in domestic circumstances. This has required changes to the existing activist network in order to develop a network of in-­country researchers. This raises the wider issue of the type of expertise within the implementation cluster. In particular, a lack of practitioner input can become problematic in two ways. First, this can result in a lack of appreciation of implementation challenges related to mine action. And second, there is always the temptation to revert to type by criticizing non-­members rather than focusing on implementation challenges beyond universalizing regime membership. Stakeholder clustering in regime implementation Stakeholder clustering dynamics proved important in shaping the two processes of landmine regime formation. Network effects across state, sub-­state and international actors are particularly relevant in the context of mine action provisions, because decisions taken at an international level need to make an impact on communities and individuals. Alongside states, a wide range of international as well as sub-­state actors should play a part in implementing the regimes’ provisions. Stakeholder clustering effects seem a particularly relevant phenomenon for the Ottawa Process given its linkage of an international campaign bringing governments, advocates and mine action experts closer together, with the ability to give a voice to local mine action campaigners around the world. The co-­chairing and co-­rapporteuring approach to the mine ban treaty intersessional meetings provides a mechanism to cluster actors by jointly engaging donors and affected states in implementation. Implementation clusters can provide opportunities for heightened cooperation, coordination and critical reflection on challenges. In this respect, Brinkert acknowledges a potentially counter-­productive divergence between states parties’ focus on realising the underpinning humanitarian goals of the regime, and the ICBL’s priorities on compliance, the conduct of non-­signatories, and applicability of the treaty to other weapons.16 However, an intervention by the ICRC at the 11th meeting of mine ban treaty states parties, held in Phnom Penh between 28 November and 2 December 2011, provides a different insight to the dynamic of the intersessional process. The ICRC representative proposed that future meetings should focus less on achievements and more on future challenges.17 This remark (not least because it comes from the normally sotto voce ICRC) is suggestive of, and

Implementing the landmine regimes   73 i­ndicates a cleavage between, the activism of civil society and a more passive approach to implementation from states parties. A sponsorship programme has enabled representatives of states parties needing financial assistance to attend and participate.18 However, there is an important, if seldom acknowledged, distinction between giving a voice to different groups and offering a genuine decision making role. As Beier points out, ‘the Achilles heel of the more optimistic renderings of global civil society lies in the apparent assumption that the formal right to speak is one with the practical ability to raise a voice’.19 The distinction between representation and influence is conspicuous within the APII regime. Echoing the mine ban treaty approach, CCW states parties decided on the model of equitable geographical rotation among regional groups when selecting the presidency for review conferences.20 Although an important principle, limited developing world participation in the regime means that this is more a symbolic gesture than representative of a shift to a genuinely participative approach. Indeed, at the CCW review conference which took place after this measure was introduced, the president was French and of the ten vice-­presidents, none came from severely mine-­affected countries.21 This weakness was at least recognized in the final conference report, which emphasized the need to seek broader participation from mine-­affected states. In a nod to the Ottawa Process, a sponsorship programme was developed to allow representatives from mine affected countries to attend CCW meetings. However, because the relationship between regime provisions and work to reduce the threat to civilians posed by landmines is tenuous at best, the practical benefits of such a programme are questionable. This shows that the benefits of emulation effects generated during regime implementation are strongly conditioned by design factors. Despite the seeming importance of demonstrating broad ownership of the implementation process, mine-­affected states are not a key target group for APII, regardless of efforts to increase their presence. Moreover, mine action organizations have been refused full admission to successive meetings of states parties, denying expert knowledge to the regime. Specialist NGOs provided expertise in informal discussions and at committee meetings during the first review conference. Yet they were excluded from the negotiations and on the objection of one state party, were even excluded from meetings of the preparatory committee for the conference.22 The diverse range of actors involved in the Ottawa Process has resulted in linkages being drawn to the study of epistemic communities. Faulkner identifies the ICBL’s function as a ‘humanitarian epistemic community’ as being critical to the creation of the mine ban treaty.23 He also concurs with Price, who distinguishes between the transnational coalition coalescing around the issue of a landmine ban and a genuine epistemic community: Because the organizers are not experts to whom governments turn for knowledge in times of uncertainty but rather are better seen as moral entrepreneurs . . . their influence derives less from the independent effects of

74   Implementing the landmine regimes authoritative claims of scientific knowledge than from their ability to successfully engage the policy process and engage in moral proselytising through persuasion.24 This distinction between experts and entrepreneurs is important in order to understand the nature of the mine ban treaty implementation cluster. Achieving the humanitarian goals of the mine ban treaty requires a combination of the norm entrepreneurship which drove regime formation, with lessons drawn from an expert community grounded in mine action practice. There is a nascent networking of expertise in humanitarian demining that, if harnessed, would constitute a genuine epistemic community. One element is the review board established to oversee the international mine action standards (IMAS). Chaired by UNMAS, its membership includes donors, commercial demining companies, national representatives of mine-­affected states, national and international NGOs and individual specialists.25 The UN has also been responsible for bringing together mine action stakeholders within a ‘network of networks’. The Electronic Mine Information Network (E-­MINE) provides a clearing house for documents, policies, project details and other information related to global mine action. Unlike similar UN initiatives in other thematic areas, the site is regularly updated. In providing this service it links relevant UN bodies, academic institutions, commercial companies, NGOs and international organizations.26 Online discussion networks dedicated to technical aspects of mine action have also developed, where mine action professionals share knowledge and deal with problems faced in their day to day work.27 The concept of an epistemic community, as developed by Haas, bridges concerns of regime formation and implementation. The emergence of ‘a network of professionals with recognized expertise and competence in a particular domain and an authoritative claim to policy-­relevant knowledge’,28 is one element. But the impact of such a network is based on its ability to consolidate bureaucratic power and institutionalize its influence.29 In this sense, UNMAS has attempted to connect its mine action work with the mine ban treaty through creating the post of Treaty Implementation Officer. This has increased transparency and information-­sharing on UN mine action, leading to a clearer picture of national implementation measures and their impact. UNMAS has also been involved in advocacy efforts on behalf of the regime in Asia and the Middle East, and developed a template to support national planning processes. However, while there is value in outreach activities coupled with increased transparency, this is qualitatively different from mine action practitioners being tied into the regime implementation process itself.

Analysing landmine regime effectiveness Voluntary and involuntary defection Even where rules are unambiguous, implementation challenges can be acute for states that lack the human, material or technological resources to fulfil their

Implementing the landmine regimes   75 obligations. Regimes must therefore address the risk of involuntary defection. Mine action combines high costs and the use of advanced technologies, with the location of the work in predominantly developing states. Defection would seem to be likely. How regime members are assisted by other stakeholders in a position to do so, as required under both treaties, is thus particularly important. If unresponsive to the potential for involuntary defection, regime effectiveness will be undermined through a failure to recognize members’ limitations. Consequently, assistance will not be channelled where it is most needed. States joining regimes for political reasons may also voluntarily defect from obligations when, in practice, they are unwilling to implement commitments. Certainly, a norm cascade in the regime formation phase which provided a strong push factor in favour of mine ban treaty membership, also underplayed the costs of obligations. By failing to integrate obligations into domestic law, regimes lose the strength given by what Muller terms ‘legal and constitutional reflexivity’.30 If obligations are not translated to national level laws and government policies, then prospects for successful implementation are low. A study on the development of national mine action legislation by the GICHD which examined seventeen mine- affected states, highlights a number of benefits flowing from the development of such legislation.31 These include improved coordination within government and with international actors, better accreditation procedures and increased transparency and accountability. Conversely, failure to embed regime commitments in domestic laws and structures erodes donor confidence, undermines mine action programmes and has negative consequences at the national level as a result of an absence of public scrutiny. Colombia is the only mine ban treaty state party to date that has consolidated a single instrument for national mine action legislation, incorporating all aspects of regime implementation. South Africa represents an important example of a state party voluntarily exceeding its obligations, by providing in its national implementing legislation for the appointment of domestic inspectors to conduct verification activities and facilitate international FFMs.32 Yet on a wider scale, less than 40 per cent of states parties passed new domestic laws to implement the provisions of the mine ban treaty.33 This implies a willingness by certain states to join the regime without addressing the practical consequences of membership at the national level. An implicit regime conditionality is exerting an influence over some members. The GICHD study found that the main reason for states developing legislation was to fulfil the terms of mine ban treaty membership and that, in some cases, these measures were directly linked to implementation requirements under the treaty.34 However, the link has not been made between insights on the utility of domestic legislation for mine action effectiveness and the potential for regime support in this area. The number of states parties that have not integrated obligations into domestic law indicates that supporting domestic implementation frameworks should be acknowledged as a priority for the mine ban treaty.35 The mine ban treaty obliges its members to ‘make every effort’ (Article 5) to identify and clear mined areas. The regime is therefore consistent with the IMAS

76   Implementing the landmine regimes in placing responsibility on states for clearing mines within their own territory.36 However, states emerging from conflict commonly lack the capacity to manage their mine action activities. The risk of involuntary defection is therefore particularly elevated. The provision under Article 6 for states ‘in a position to do so’ to provide resources for mine clearance, is thus crucial. Given the multiple challenges faced by states affected by landmines, the assistance of donors is a constant feature of mine action. Support by the mine action community includes direct assistance to mine-­affected states, work to sustain donor commitment and capacity building among local actors. The Article 6 obligation is particularly significant due to the clustering effect of the regime. Bringing together donors and mine-­affected states offers opportunities to pool resources and target support in a ‘joined up’ manner. In many mine-­affected states, responsibility for mine action planning and operations is devolved to the UN. Local capacity building is recognized by the mine action community as essential if responsibility for mine action is to be handed back as soon as possible to the legitimate national authorities. According to the UN, capacity building in mine action is understood as: A state’s ability and willingness to develop and articulate mine action policy and direction. It is also about a state’s ability to plan, coordinate, manage and sustain a mine action programme that is accountable, cost-­effective and able to address the humanitarian and socio-­economic implications of landmine contamination, and to provide appropriate legislation.37 There is a potential contradiction between external support and the principle of national ownership. If mine action capacity building can provide a model for re-­ establishing good governance, the inverse is also true. The influx of foreign investment can create tensions, attracting the corrupt and self-­interested. Mine action programming, as with other externally supported peacebuilding and development efforts, suffers from the Samaritan’s Dilemma. As Maslen notes, ‘the generosity of donors can make it less likely that the recipients exert the necessary efforts to help themselves’.38 National capacity to govern mine action is one significant gap. This concern demonstrates the need for a critical assessment of the actual, rather than intended, capacity-­building effects of externally sponsored and implemented mine action and the role of the two regimes in these efforts. In particular, there is a simmering tension between the responsibility placed by the mine ban treaty on states parties themselves, and the reality that mine-­affected states may not be capable of setting policy or overseeing and managing mine action programmes at the national level. There is an emerging recognition that capacity building of mine action actors at the local level can only be optimized within the framework of an effective national mine action strategy.39 Per Nergaard, Head of the Mine Action Unit at NPA, highlighted as early as 2006 the absence of national plans that support the implementation of the core mine ban treaty obligation to clear mined areas within ten years as something that ‘needs the immediate attention of all States

Implementing the landmine regimes   77 Parties to the Convention to fundamentally correct these things now’.40 UNMAS coordinated an initiative to assist mine-­affected countries implement Article 5 requirements through supporting priority setting at the national level, developing a template for mine action planning, and reporting on progress at statutory regime meetings. UNMAS also assisted mine-­affected countries in developing national plans related to APM destruction. This focused on priority setting and budget requirements in more than twenty countries.41 Progress in these areas is reported regularly at intersessional meetings, and has been useful as a means to better understand the specific nature of mine action challenges in different contexts. The mine ban treaty has done little in practice to encourage external actors to embed capacity building in their assistance. One major impediment has been a preference by donors for home-­grown organizations and in-­kind contributions – such as staff and equipment – when these have not been the most appropriate solutions for local circumstances. This was tacitly recognized in the 2005 Zagreb Progress Report which, rather than identifying the achievements of the regime in this area, states that ‘the United Nations has assisted a number of States Parties in establishing national plans and in making these plans publicly available on its E-­MINE website’.42 In other words, the regime is providing a platform to showcase UN efforts in this area rather than making a substantive contribution to mine action capacity building. Multiple calls by states parties for greater emphasis on capacity building in mine action assistance at the 2011 Phnom Penh meeting, suggests that much remains to be done in building national capacity to support mine action.43 There is clearly a fine line between involuntary defection from regime obligations and willful failure to implement national commitments. This picture is further complicated by the political imperatives that may drive states to deny that they have failed to meet international obligations. Real or alleged treaty violations among mine ban treaty regime members have been relatively rare. Angola admitted continuing to use APMs while a signatory although not yet a full state party,44 while credible allegations have also been made that Turkish armed forces used APMs in border areas adjacent to Iraq in 2009.45 In the case of APII, Russia in Chechnya and Pakistan in its border dispute with India have used landmines without marking, fencing or monitoring them in accordance with the protocol.46 The journal of the Israeli Defence Force reported laying APMs in the Golan Heights on the Syrian border in August 2011.47 The US has also incorporated Soviet-­era minefields into its perimeter defences in Afghanistan. As in the case of India and Pakistan’s use of APMs in border areas, there has been no information provided in their annual reports on measures taken to mark and monitor these minefields or protect civilians. In a different kind of breach, the state owned Pakistani Ordnance Factories (POF ) allegedly offered APMs for sale to an undercover British television journalist via an attaché to their embassy in London. The offer, followed up with a faxed quotation from POF, contravenes Pakistani and British domestic legislation as well as the APII prohibition on transfers.48

78   Implementing the landmine regimes Dislocation between implementation processes and mine action challenges constitutes an important defect whose roots can be traced back to regime design. The contrast between emphasis on national ownership as the hallmark of the Ottawa Process, and a lack of focus on national capacity building to help operationalize commitments is particularly stark. Examples of voluntary defection by states parties from the obligation not to use landmines (or to do so only under certain conditions) may be the exception. However, the need to focus more directly on mine action challenges linked to resources, political will and humanitarian impact should be self-­evident. Resources An area where the raised profile of the landmine issue is recognized as making a demonstrable impact is on resource mobilization. Between 1997 and 2005, annual donor funding for mine action rose from $139 million to $376 million. For 2009, the annual figure had risen to $622 million.49 In a telling example, following adherence to the mine ban treaty, the UK’s Department for International Development (DfID) committed to double its funding for humanitarian demining from £5–10 million per year in the period 1998–2001.50 By 2005, this figure reached £11.8 million per annum. The link between increased funding and regime membership is made clear by an explicit conditionality, intended to promote universalization, stating that ‘DfID will not provide direct support for governments’ mine action programmes in countries that ignore the growing international condemnation of APMs, and continue to use these weapons’.51 Additional non-­monetary resources for mine action constitutes a significant, if less visible, regime effect. Although the project is not explicitly linked to the mine ban treaty, Switzerland’s willingness to commit several million Swiss francs to the development and implementation of the Information Management System for Mine Action (IMSMA) by the GICHD, currently operational in over thirty mine action programmes,52 reflects a policy decision to support mine action as a priority over other humanitarian activities. Indeed, the creation of the GICHD itself represents a major new commitment to mine action. The timing of these initiatives, launched in 1998 and 1999, links the decisions to the raised profile of mine action as a result of the Ottawa Process. Support from multilateral development donors can also be traced to the Ottawa Process. In August 1997 the World Bank agreed for the first time to fund mine action, emphasising that the centrepiece of its support was for humanitarian demining. Although the Bank stresses that it has no role in treaties and conventions that relate to its member states, there is implicit membership conditionality in relation to humanitarian demining: The Bank stands ready to support implementation of obligations imposed under the treaty by any member state that requests it. . . . Although the Bank does not refer specifically to the Ottawa Convention in its guidelines, it stresses that any legal agreement for a project involving landmine clearance

Implementing the landmine regimes   79 must include a covenant under which the government undertakes not to lay new mines anywhere in the country that would undermine the execution or development objectives of the project.53 There is a subtle difference between DfID’s reference to the need to respect international condemnation of APMs, and the Bank’s emphasis on a national covenant against landmine use that distinguishes treaty obligations from their domestic implementation. However, World Bank mine action funding requirements shares with DfID an important regime-­driven conditionality linking support to either regime membership or the fulfillment of regime goals. In November 1997, the European Commission earmarked €15 million for research into new technologies to identify and destroy APMs.54 As with the World Bank, DfID and Switzerland, the timing of this announcement, at the culmination of the mine ban treaty regime formation phase, illustrates the strong pressures exerted on the international community to demonstrate solidarity at the time of the regime’s creation. In the first five years since entry into force, the European Commission and EU Member States together contributed nearly €700 million to mine action, with total support to 2009 estimated at €1.8 billion.55 These resources, and the regime-­based conditionalities that regulate their dispersement, demonstrate an important regime effect – at least for this period – on the priorities of bilateral and multilateral mine action supporters. Resource allocation for mine action in the implementation phase displays significant interaction effects. Without being a target of regime advocacy efforts, the profile of the issue has drawn in new actors beyond the traditional development donor community. Support by the European Football Association (UEFA) for ICRC mine action work56 and manufacturing company Daewoo’s donation to the Slovenian-­based International Trust Fund for Mine Action,57 show the pull of the anti-­APM norm reaching out to sporting and commercial sectors with no previous engagement in mine action. The norm cascade evident in the regime formation phase that obliged states to be on the right side of the issue is thus significant for resource generation during implementation. The reach of the anti-­ APM norm has extended beyond states to the private sector. These unconventional new mine action donors offer a convincing response to counterfactual questions: given their complete lack of engagement with this issue area prior to the high profile Ottawa Process, this new funding can only be understood as an effect of the regime. It is important to acknowledge that increased mine action funding since the beginning of the Ottawa Process has not come solely from mine ban treaty members and supporters. Nine states that have not joined the regime contributed over $425 million between 1997 and 2002 to support mine action.58 In particular, the resource issue provides an important correction to criticisms of the US. The US has not only pursued the mine ban treaty’s disarmament objectives (the US has not used APMs in combat since Kuwait in 1991) but also its mine action objectives as the biggest single donor in this field.59 Between 1993 and 2005, the US contributed over $1 billion in mine action funding.60 Although this spiked

80   Implementing the landmine regimes after 2003 due to mine action in Iraq and Afghanistan, the US has been heavily involved in this field since the late 1980s. Significantly, from 1995–1997, US mine action funding remained constant, but sharp rises were seen in 1998, 1999 and 2000, coinciding with the entry into force and initial implementation phase of the mine ban treaty. The timing of these increases points to an important regime effect. The evident success of an initiative with strong and highly visible humanitarian credentials induced the US to demonstrate the requisite level of political commitment to the issue. Contributing more money for mine action provides an indirect means to address criticism that has come about as a result of declining to join the regime. The resource issue shows that APII remains the poor relation in terms of profile, marketing and outreach. While actors involved in the Ottawa Process make conspicuous efforts to promote their brand, little comparable effort can be discerned within the CCW framework. The generation of additional resources for mine action thus provides one measure of the effectiveness of the mine ban treaty. The regime has persuaded multi- and bilateral donors to invest heavily in a humanitarian issue area that otherwise would not have attracted such elevated levels of funding. That this additional support has come from non-­traditional mine action donors and even states that explicitly reject the APM ban, demonstrates the underlying normative strength of the regime and its related emulation effects. It is important to acknowledge that the impressive resource base generated by the mine ban treaty is just one dimension of effectiveness. This is evidenced by facile assumptions that any mine action is ‘a good thing’. As Brinkert notes, ‘the matter of resources to ensure the implementation of the convention is not as simplistic as a one-­way flow of money from relatively wealthy, unaffected countries to relatively poor, mine affected counterparts.’ Citing the ICBL, Brinkert acknowledges that much remains to be done within the regime on the qualitative dimension of resource generation: while the ICBL has been very effective in focusing attention on the supply side of the resource issue (i.e. how much is generated), it perhaps could give increasing attention to better understanding the demand side of the equation (i.e. how much is required).61 Political will Political will is a sine qua non for effective regimes in different issue areas. This is particularly important in hard cases where security concerns may form a significant part of the calculus.62 In the regime formation phase, targeted advocacy raised the profile of the landmine issue and facilitated the development of national interest groups in different regions. Political will in favour of regime membership was therefore generated at both international and national levels. In implementation, with compliance monitoring a devolved rather than formal

Implementing the landmine regimes   81 c­ ommitment, a division of labour has evolved at the national level with civil society providing an oversight function, and states supporting mine action both within national programmes and through assistance offered to other states. On one level, this provides the regime with different levers to support implementation at state and sub-­state levels. However, it also raises important questions about the effectiveness of informal oversight as a means to generate political will among states parties to ensure compliance. Political will is essential if mine-­affected states are to deliver on demanding compliance obligations. Without such a commitment, international support for mine action is highly vulnerable to being diverted from its intended purpose. Corruption therefore represents a significant mine action challenge that finds a clear nexus with regime effectiveness. Diversion of funds, self-­interested selection of clearance tasks and ‘land-­grabbing’, have long been associated with certain demining programmes, so these same issues should represent a test of regime members’ political will to deliver on their commitments. But the issue is rarely taken up by civil management and oversight bodies or civil society at national and local levels. Corruption may affect donor support as in Cambodia, where allegations of financial impropriety, corruption and mismanagement led to the withholding of significant donor contributions.63 Emerging research also highlights concerns over corrupt mine action practices in Afghanistan and Bosnia-­Herzegovina.64 In principle, the mine ban treaty should respond to these concerns. However, this major challenge to mine action effectiveness has not been taken up as an implementation issue within the treaty framework. To take the case of Cambodia, the country played an important role in the Ottawa Process and retains an influential role in the mine ban treaty intersessional work programme. Misconduct within Cambodia’s national mine action programme was a politically sensitive issue that the regime ignored. This aversion to picking up a political hot potato therefore qualifies the success of the wider implementation process. Knowledge of the specific nature and location of the landmine threat can be particularly difficult to obtain in intra-­state conflicts involving armed non-­state armed actors (ANSAs). Such groups, characterized by decentralization, poverty and unwillingness to compromise, often control mined territory and may be responsible for the manufacture, trade, sale and use of landmines. Unrecorded nuisance mining is common practice. In short, ANSAs represent a major category of current mine users,65 so should be an evident priority for engagement. Indeed, the extension of APII to non-­international armed conflicts was a significant, if belated, recognition of the realities of contemporary landmine warfare. However, the regimes have not focused on ANSAs. From a practical perspective, engagement poses a qualitatively different challenge to state actors. In political terms, national authorities are often reluctant to engage with ANSAs on their territory for fear of lending legitimacy to their cause. The issue therefore represents a further entry point for engagement by civil society. The NGO Geneva Call was established in 2000 with the specific mandate to engage armed groups in a ban on APMs. It does so through encouraging groups to adhere to

82   Implementing the landmine regimes Deeds of Commitment that mirror the requirements the mine ban treaty places on states.66 Geneva Call’s own research found around 60 ANSAs using landmines, with the majority in Asia (31) and Africa (15).67 Although 60 per cent of landmine use by ANSAs occurs in the territory of non-­states parties, this issue is relevant for regime members as in Colombia, where groups such as the Revolutionary Armed Forces of Colombia (FARC) and the National Liberation Army (ELN) have been frequent mine users. This informal role played by Geneva Call is in fact essential, since the mine ban treaty has no formal mechanism to engage with these actors. The functional role assumed by Geneva Call is necessary because mine ban treaty states parties have shown themselves to be unwilling or unable to move forward. At the same time, the organization’s work could not be carried out without the normative pull of the regime. Although the procedure of binding groups to a Deed of Commitment is outside the regime proper, there would be no foundation for such action without the legitimacy fostered by the mine ban treaty. When the Sudan People’s Liberation Army/Movement (SPLA/M) signed a Deed of Commitment in October 2001, this preceded, and most likely precipitated, treaty ratification by the Sudanese Government.68 In this way, the work of Geneva Call clarifies and even influences interactions between state and non-­ state actors. Perhaps for this reason, intervention by Geneva Call has led to criticism from some regime members – predictably governments in conflict with such groups – demanding that permission be sought before engagement. Turkey has been particularly vociferous in condemning Geneva Call’s work with the Kurdistan Workers Party (PKK).69 The organization has resisted such pressures, recognizing the direct threat to its neutrality and therefore to its effectiveness. The first mine ban treaty review conference, held in Nairobi in December 2004, was the biggest international gathering on the landmine issue since the treaty was opened for signature in December 1997. Estonia and Papua New Guinea signed the treaty in Nairobi, while Latvia, Poland, Sri Lanka and Ethi­ opia announced their full adherence. Commitments made to coincide with this meeting provide important evidence that the regime continues to generate political will to be on the right side of the issue. This includes spillover effects affecting the behaviour of states outside of the regime. Egypt announced a moratorium on production at the Nairobi meeting, while China endorsed the purpose and objectives of the treaty and expressed a desire to expand its cooperation in this area. Similarly, non-­states parties Belarus, Cameroon, Gambia, Lithuania and Ukraine all submitted voluntary Article 7 reports. Militarily significant states including Egypt, Iraq and Israel have all ceased production of APMs since the mine ban treaty entered into force. A de facto global ban on licit trade in APMs has been realized. Of course, from a regime perspective, these developments have to be qualified because complying with some treaty commitments does not mean signing up to all regime obligations. The conduct of POF in the UK is important not just as a case of voluntary defection. Exposure of the issue resulted in rapid denials from the Pakistani government, and the immediate recall of concerned embassy staff to Islamabad. The

Implementing the landmine regimes   83 breach was clearly recognized as a serious international incident. This demonstrates an important quality of normative interplay linking members and non-­ members despite formal boundaries delineated by the respective treaties and their supporters. It is particularly significant that the profile of the mine ban treaty raised the political costs of discovery in the POF case. For certain actors, regime effectiveness may seem to be a numbers game, with each new member representing a step towards the ultimate goal of universalization. Yet the political compunction to fulfill key regime obligations without formally joining the mine ban treaty is a major, unacknowledged element of the regime’s effectiveness. The work of Geneva Call provides a positive example of flexibility in implementation. It also helps to show the mine ban treaty’s limitations. Key obligations have been adapted and applied to ANSAs. Yet Geneva Call’s capacity is limited. As a major category of landmine users, ANSAs should be an obvious focus for the regime. However, the reluctance of some states to address ANSAs on their own territory, or of others to rock the boat by highlighting this sensitive issue has created a tension between political and humanitarian imperatives. The inability to address corruption is another facet of the same phenomenon. As demonstrated by the POF case, the regime raises the political stakes of not falling in line with the anti-­APM norm. However, this is qualitatively different from enforcing compliance with mine action-­related commitments. This contradiction provides an additional insight to the consequences for effectiveness of how the implementation cluster is composed. Despite the innovative collaborations that have characterized regime formation, the mine ban treaty remains a state-­based regime. When national security interests are at play, political will is lost. The regime displays the paradox of having been created through the strong commitment of non-­state actors, yet is reluctant in implementation to support these actors in fulfilling key regime goals. The lack of engagement with ANSAs is a missed opportunity to develop reciprocal arrangements that would mirror states’ adherence to treaty provisions. Humanitarian impact A humanitarian imperative provided the motor that led to the rapid formation of the mine ban treaty, while also underpinning the move to enhance the CCW regime. In demonstrating the continued vigour of humanitarian push factors in the implementation phase, Kjellman identifies Landmine Monitor as a normative watchdog, keeping the goals and progress of the treaty in the spotlight.70 However, if Landmine Monitor highlights important implementation issues, this does not necessarily advance mine action priorities if they are ignored by the regime. The relationship between the regimes and mine action should therefore be gauged by their effectiveness in addressing common humanitarian objectives. The effectiveness of mine action is increasingly judged on the basis of humanitarian or developmental impact. The humanitarian contribution of regime compliance can be measured by the relative absence of serious allegations of

84   Implementing the landmine regimes APM use by members, despite the fact that many states were users in the recent past. The 2010 Landmine Monitor identified 3,956 new landmine casualties in 2009, the lowest annual total since they began monitoring these figures in 1999.71 However, it does not follow that these statistics permit a judgement of regime effectiveness in humanitarian terms. In particular, an overriding concern with extending regime membership may be counter-­productive if the quality of implementation and the ability to address hard implementation issues are ignored. Mine action resonates with other complex challenges in fragile states. Mine action can be an enabling activity for reconstruction and development activities, while gains are lost in all these areas if states revert to conflict. It is therefore important to consider how far the two regimes support the integration of mine action with related fields. This is important because expert assessments of the first decade of mine action work have concluded that there is a danger of mine action being compartmentalized away from other security and development activities. This sets mine action apart from poverty reduction strategies, and leaves it vulnerable to accusations that costs are greater than benefits relative to other activities such as providing clean water, or addressing preventable diseases.72 Many experts argue that mine action should not be individualized, but requires ‘a more multidisciplinary and multi-­sectoral approach as mine action realises the need for expertise and experience from numerous disciplines to achieve its goals’.73 This finding reflects a broader drive towards greater coherence so that international efforts are tailored to local contexts and show flexibility in response to shifting priorities. Ensuring that development activities are aligned with national priorities is an obligation in line with international commitments under the Paris Declaration on Aid Effectiveness and the Accra Agenda for Action.74 The mine ban treaty is unique because of its strong developing state membership. However, if hard implementation issues such as inappropriate or counter-­ productive mine action work are ignored, this represents a lost opportunity to enhance the effectiveness of the regime in humanitarian terms. In contrast, the CCW process has always lacked support from the global South. There is no pool of local knowledge and expertise to engage on context-­specific challenges within the implementation cluster. The absence of support for national implementation is thus both a cause and effect of this narrow membership base. Mine action needs to be de-­mystified and mainstreamed if it is to move beyond what Horwood characterizes as a ‘parallel existence’. This parallelism has been accentuated by the high profile and elevated donor support created by the Ottawa Process.75 Given that members support mine action as only one of many development commitments, the regimes offer fora that could be used to bridge these gaps. This expands the question ‘do the regimes matter’ in humanitarian terms beyond the confines of the landmines issue to reflect the complexities of the security-­development nexus. It also points to a potential role for the landmine regimes in bringing mine action closer to wider development concerns.

Implementing the landmine regimes   85

Conclusions Resource, political, and humanitarian factors provide for a more nuanced picture of landmine regime effectiveness. Both regimes demonstrate high levels of compliance by members according to the letter of treaty requirements. In each case, membership has increased over time while (albeit to a far greater extent in the mine ban treaty) new resources that support the realization of mine action goals have been mobilized. However, according to Underdal’s two-­fold definition of regime effectiveness, important gaps become apparent in the political will and material commitment of states parties to meet their own obligations, and to support other members requiring assistance. The interplay between domestic and international commitments is particularly significant as a determinant of regime effectiveness. Reflecting a negative consequence of the norm cascade that impelled many states to join the mine ban treaty, domestic implementation of regime obligations has not been prioritized by mine affected states parties. Inadequate national, legal or bureaucratic frameworks and the absence of regime support in these areas present significant barriers. In many cases involuntary defection is a major risk, since compliance is beyond the means of mine-­affected states. Although recognized as a key mine action priority, the lack of space in implementation for expert communities means that addressing these gaps through capacity building has not been a priority. Given that funding and technical support for national mine action programmes is largely provided by bi- and multilateral donors, this disjunction between regime and mine action priorities poses a serious challenge to mine ban treaty effectiveness. The exclusive approach to participation within the CCW framework is costly. Activities within APII that mirror mine ban treaty initiatives, such as equitable geographical rotation of responsibilities and the sponsorship programme, are rendered largely redundant by the lack of interest in the regime from the global South. In contrast, according to the prevalent narrative, the Ottawa Process is characterized by transparency and inclusiveness. However, strong participation of mine-­affected states in regime formation has not translated into an influential role in implementation. Moreover, while expertise is increasingly networked within the mine action community, practitioners do not form a core element of the mine ban treaty implementation cluster. The under-­emphasized role of ANSAs within the mine ban treaty shows how political imperatives can negatively affect implementation. Resistance from some regime members to the work of Geneva Call is a telling demonstration of the regime’s response to a hard implementation issue. Geneva Call’s role is entirely consistent with the origins and values of the Ottawa Process, with civil society taking the lead where states are reluctant to act. This clearly poses difficulties for some regime stakeholders. Unwillingness to address ANSAs and other hard case issues like corruption shows that if issues are unpalatable for diplomats or ‘off message’ for the advocacy lobby, they risk being ignored regardless of the humanitarian costs.

86   Implementing the landmine regimes A protectionist approach within both regimes works against meaningful synergies. The co-­location of the secretariat function for the two regimes with the GICHD, points to common implementation requirements that may offer the chance of greater coordination and perhaps cooperation. Similarly, many of the implementation issues faced by the landmine regimes are not unique, but form one part of the security and development challenge faced by states emerging from conflict. The elevated profile of the issue area means that political will and resources have been furnished that would not otherwise be available. Yet the combination of a narrow implementation cluster and the ‘parallel existence’ phenomenon within mine action constrains effectiveness. The need for mainstreaming mine action with related security and development issues is acknowledged by mine action stakeholders, but particularistic interests and a lack of technical expertise continue to present significant barriers to regime engagement in this wider context.

5 Humanitarian demining

Introduction The objective of humanitarian demining is to make land safe for communities and individuals. In the immediate post-­conflict phase, reducing the contaminated area is the key priority. This means prioritizing clearance of those mines that pose the most urgent humanitarian threat. Later on in the process, priorities shift to addressing the socio-­economic impact of landmines. Ensuring that areas are free from landmines and safe for civilian use is the most tangible way of achieving the humanitarian goals of the two landmine treaties.1 Humanitarian demining is therefore of central importance in meeting core regime objectives set out in both APII and the mine ban treaty. However, the apparent cognitive dissonance between advocacy actors and practitioners, highlighted in earlier chapters, is particularly marked in this field. Mike Croll, a former deminer and author of The History of Landmines, characterizes the ban campaign as a double-­edged sword whose positives must be set against a diversion of resources away from humanitarian demining: ‘certainly the campaign brought a great deal of publicity to the issue, but it . . . distorted the size and shape of the problem and distracted attention from the crux of the issue’.2 This points to the need to understand the interplay between technical and political considerations for regime effectiveness in the area of humanitarian demining. Overall regime effectiveness can depend greatly on specific sub-­issues and commitments. Humanitarian demining is particularly important because it can enable a focus on the actual (rather than assumed) contributions of the landmine regimes. Examining this issue can help to answer the question that divides many practitioners and advocates: whether there is a causal link, or a degree of parallelism, between the regimes and humanitarian demining. Because of the role of the Ottawa Process in pushing the landmines issue up the international policy agenda, it is tempting to posit a clear, positive relationship between regime implementation and progress in humanitarian demining. However, such a link should not be assumed. The mine action community has advanced significantly in its own right over the past decade in its approach to addressing the scourge of landmines. Humanitarian demining displays both lengthy historical antecedents and a short history as a distinct discipline. Mine action began in Afghanistan from

88   Humanitarian demining 1988 with a UN-­assisted appeal for funds to support humanitarian demining. Programmes subsequently expanded to many other countries,3 particularly in Asia, Africa and the Balkans but also, though to a more limited extent, in Eastern Europe and the Americas. The state of the art in humanitarian demining emphasises a ‘toolkit’ approach that combines the appropriate use of manual deminers, mine detection dogs and mechanical demining equipment.4 In reality, there is often a chasm separating best practice from the realities faced on the ground. Many practitioners point out that financial and practical constraints mean that they work with what they have (or what they are given) in often difficult conditions, rather than achieving an ideal combination of capabilities.5 This implies that adaptation and learning are essential if the regimes are to contribute to meeting real humanitarian demining challenges and priorities.

Regime implementation and humanitarian demining Regime rules and field realities Knowing the extent and location of mined areas is a prerequisite for effective humanitarian demining. Thus, each state party to the mine ban treaty must make ‘every effort to identify all mined areas under its jurisdiction or control in which anti-­personnel mines are known or suspected to be emplaced’. Further, ‘as soon as possible’ these areas should be ‘perimeter marked, monitored and protected by fencing or other means, to ensure the effective exclusion of civilians, until all anti-­personnel mines contained therein have been destroyed’ (Mine Ban Treaty, Article 5 (2)). APII requires that ‘all reasonable precautions should be taken to protect civilians from the impact of mines, booby traps and other devices’, (Amended Protocol II, Article 3 (10)), while their locations should be recorded in accordance with the requirements of the protocol’s Technical Annex (Amended Protocol II, Article 9 (1)). Manually emplaced APMs that do not self-­ destruct and self-­deactivate can only be used if ‘placed within a perimeter-­ marked area which is monitored by military personnel and protected by fencing or other means, to ensure the effective exclusion of civilians from the area’ (Amended Protocol II, Article 5 (2)). Locating and identifying mined areas thus represents a common priority across both regimes. In reality, few mined areas are fenced and in many cases animals, weather or the needs of local people, result in the destruction or removal of such barriers that have existed. Even poles and wire have an intrinsic worth that may outweigh the protective value of the barrier. Contaminated land, viable for hunting or farming, is frequently used by locals in the full knowledge of the risks being run. Moreover, maps often do not exist as a consequence of the widespread practice of nuisance mine laying targeted deliberately at civilian populations. Even where maps are available, the location of mines shifts over time due to weather and soil conditions, limiting their value. Landmine Monitor acknowledges that one aspect of the Article 5 requirement almost never met is that of perimeter marking and monitoring. It is reported that only Denmark, France (in respect to a military

Humanitarian demining   89 base in Djibouti) and the UK (Falkland Islands) have taken such measures. These are clearly exceptional cases, involving developed countries that know to a high degree of accuracy where APMs have been emplaced. That no mine-­ affected state has been able to fulfil this obligation points to the fact that the requirement is unenforceable in developing countries with large, often unmapped swathes of contaminated land. APII reiterates the earlier Protocol II requirement for the recording of ‘pre-­ planned’ minefields. However, many mined areas laid by government or irregular forces have either served immediate tactical reasons – so the provision does not apply – or when targeted at civilians, have not been planned or mapped at all. Even in the Falklands, where the location of mined areas is well known, landmines have shifted over time as a result of weather and ground conditions, rendering maps largely redundant. Although subject to little discussion at the 1979–1980 negotiating conference, it was recognized even at the time that the concept of a ‘pre-­planned’ minefield was both vague and represented a degree of advance preparation sufficient to exclude nearly all traditional minefields, as well as the less clearly delineated ‘mined areas’ frequently encountered by deminers.6 India has subscribed to ‘a phased approach (that) will narrow the scope in which landmines can be used only for the defense of borders’.7 This position stands or falls on whether India can meet its APII obligations: are mined areas perimeter marked and monitored? Have accurate maps of mined areas been made and exchanged with opponents at the end of hostilities to aid subsequent demining efforts? And, fundamentally, if there has been compliance with these measures, has this ensured that civilians have not suffered through the use of landmines? Reports from the Kashmir region suggest that the difficulties of mapping and marking during conflict have made these provisions impossible to implement.8 Economic necessity and the effects of climate have further undermined efforts to accurately identify mined areas and warn civilians of potential hazards. National compliance with the letter of regime commitments does not contribute to the intent and purpose of the regime by reducing the danger posed by these weapons. In practice, neither regime acknowledges that the tactical use of landmines almost never conforms to the image of static, clearly defined minefields. Ana­ lysis of mine ban treaty Article 7 reports provides evidence of this gap between regime rules and field reality. Moldova reported full clearance by August 2000. Yet subsequent reports have shown that communities continue to avoid certain ‘cleared’ areas. Other regions not subject to clearance are also suspected of containing landmines.9 Similarly, the Democratic Republic of the Congo stated that ‘no mined area has yet been identified’ at the same time as United Nations sources confirmed the presence of landmines within the national territory.10 While the stipulation to mark mined areas may be important regardless of implementation challenges, it is questionable whether many affected states are able to accurately identify their mined areas. And where such information could be usefully provided, there is no obligation in the case of APII to record the types of mines used, the pattern of mine laying or the location of individual mines: key

90   Humanitarian demining information for the purpose of humanitarian demining. In sum, it is evident that the provisions on minefield marking and mapping contained in both treaties reflect a lack of appreciation of humanitarian demining concerns. While these challenges are not open to ready solutions, the regimes are weakened by their failure to address these practical issues. The mine ban treaty obliges member states to destroy or ensure the destruction of all APMs in mined areas under its jurisdiction or control ‘no later than ten years after the entry into force of this convention for that State Party’ (Mine Ban Treaty, Article 4). While placing the responsibility for clearance squarely on individual regime members, the term ‘ensure the destruction’ of APMs is used deliberately to denote the external assistance that many post-­conflict states require in fulfilling this obligation. Indeed, states parties ‘in a position to do so’ must assist in these efforts (Mine Ban Treaty, Article 6 (4)). In case this timeframe proves unrealistic, a request may be submitted seeking an extension of the destruction deadline for renewable periods of up to ten years (Mine Ban Treaty, Article 5 (3)). During the negotiations, the time period for destruction of mined areas was subject to considerable discussion. It was also proposed not to provide a specific deadline at all.11 This reflects a tension between ban advocates pushing for short deadlines, and others seeking greater flexibility either in acknowledgement of the scale of the humanitarian demining challenge or as a means to dilute this core obligation. A fixed deadline was ultimately retained in order to maintain the need for urgency. However, the renewal clause recognizes the difficulty of this task for heavily mine-­affected states. In effect, the combination of fixed deadline and renewal clause embeds an unresolved tension between different stakeholders within the regime framework. APII neither commits states parties to clear the mines they laid nor to assist in mine action. It only requests that after the cessation of hostilities, former conflict parties provide information to facilitate this work ‘wherever possible’ (Amended Protocol II, Article9 (2)). APII does include a provision that requires all APMs to be detectable by commonly available mine detection equipment (Amended Protocol II, Article 4). If this cannot be complied with in respect to existing stocks, a nine-­year deferral period is permitted to make weapons compliant or destroy them. The ICRC estimated at the time that this delay could result in some 200,000 new mine victims.12 Looking back to the original protocol, Prokosch argues that landmine contamination in Afghanistan would be much less severe today if the Soviet Union had observed its earlier Protocol II obligations following ratification in 1982. Instead, ‘through lack of control and lack of publicity, the Soviet army felt free to act without fear of effective sanctions or public censure, as have government and opposition forces in many other conflicts since 1980’.13 This example highlights a dual challenge to regime effectiveness. Lengthy deferral periods for the implementation of key technical provisions come with humanitarian costs. This is exacerbated by the absence of political will to impel implementation for humanitarian reasons in advance of the required timeframe. Compromise solutions relating to destruction deadlines were made in both treaties. This reflects, in regime design, the practical challenges of implementation. It

Humanitarian demining   91 also brings the interests of different stakeholders into the open and points to the flexibility necessary to reach agreement. If both regimes display a lack of awareness of certain field realities, a core difference in their orientation and knowledge base is also apparent. Within the APII regime design process, destruction of stocks was understood as a straightforward technical exercise. For the mine ban treaty practical difficulties of implementation, as well as the humanitarian costs of different options, were discussed at length. This is reflected in an emphasis on support for national implementation that is absent from APII. Contemporary use of landmines and the practical constraints within mineaffected states create significant barriers to implementation. Certain provisions thus represent a best-­case scenario rather than achievable commitments. It is evident that in both regimes, humanitarian demining expertise has not been exploited to ensure realistic, enforceable obligations. If the mine ban treaty does provide space (unlike APII) for developing world concerns, neither regime demonstrates the necessary flexibility to address implementation challenges faced by mine-­affected states. Involuntary and voluntary defection Angola, Ecuador and Ethiopia admitted using APMs as mine ban treaty signatories. Similar allegations have also been made against Burundi, Cambodia, Rwanda, Sudan, Turkey and Uganda.14 Overall, such conduct by states parties has been rare. The drastic reduction in the use of APMs by states generates a humanitarian demining pay-­off by lowering the number of future mined areas that would need to be cleared. The decline in mine use thus fulfils the humanitarian criterion for regime effectiveness. In terms of the strength of the anti-­APM norm, it is also significant that a central pillar of the regime – the prohibition – has proved resilient during the years since the treaty entered into force. Landmine Monitor characterizes the obligation to clear mined areas as ‘the greatest challenge to the integrity of the treaty’.15 The Nairobi Action Plan, agreed at the first mine ban treaty review conference, calls on states to ‘strive to ensure that few, if any, States Parties will feel compelled to request an extension’ to their ten-­year clearance deadline.16 In fact, the majority of mine-­affected states have missed this deadline. The difficulties faced by affected states are made evident in a 2003 study on mine action in the countries of the Southern African Development Community (SADC). Taking one case, Neuma Grobbelaar states that ‘despite Mozambique’s Article 5 commitments under the Ottawa Convention, the reality is that Mozambique will probably never be able to rid itself of all landmines and unexploded ordnance’.17 The meeting of mine ban treaty states parties held in Amman, Jordan, in November 2007, seems to confirm this trend. The focus of the sessions dedicated to mine clearance was not on implementation challenges, but modalities for requesting extensions to clearance deadlines.18 The tension identified in the design phase between regime obligations and humanitarian demining realities is now apparent as a concrete implementation challenge. As of September 2010, twenty-­two states have been granted extensions to their clearance deadlines.19

92   Humanitarian demining Perhaps the most significant issue discussed at the 2011 Phnom Penh meeting of mine ban treaty states parties was one completely unforeseen in regime design. States parties that had not reported contaminated areas nor sought extensions to their Article 5 deadlines, have now found themselves to have mines on the national territory so are in material breach of treaty commitments. The circumstances are eclectic. Germany found a suspected mined area on a Soviet era military training base in the East of the country. Hungary suspects that mines remain along the border with Croatia, harking back to the conflicts in the former Yugoslavia. Mali may have new contamination due to mine laying by an ANSA while Niger may have a mined area around a former French military base.20 The ICBL has responded proactively to this challenge by proposing in a food for thought paper that a ‘special Article 5 deadline procedure’ be created. The new procedure would provide for a specific, timebound deadline extension for states with previously undetected mined areas, or where clearance deadlines have already passed without an extension being requested.21 While this proposed mechanism may fill a hole in the treaty, to be effective it will require the leveraging of additional political will. If Germany was proactively transparent, Hungary did not report on the problem along its border in line with its transparency obligations. Ignorance is no defence, since the land had been surveyed for clearance and funding sought from the European Commission many months previously.22 Clarifications have not been possible from Mali or Niger, because neither state participated in the Phnom Penh meeting. A number of mine ban treaty members that have not cleared mined areas in line with treaty commitments lack the necessary domestic capacity to support this work. Thailand’s failure to meet its clearance deadline is illustrative. Landmine Monitor identifies three reasons: ‘mine action has not been a government priority, inadequate financial support, and the military structure of mine action in Thailand’.23 For a number of other states parties that have also not advanced in their clearance operations, adequate capacity is clearly an issue. In each case, involuntary defection results from a combination of the scale of the task and the lack of domestic capacity or implementation mechanisms, as well as insufficient political will. With no demining work during a decade of regime membership, the case of Venezuela has crossed the line from involuntary to voluntary defection.24 Clearance of APMs – used to protect naval facilities near the border with Colombia – was reportedly conditional on finding an alternative protection system.25 Military necessity therefore seems to be the key consideration influencing Venezuela’s non-­compliant behaviour. Significantly, this failure to implement a clear regime obligation was highlighted by Landmine Monitor as early as 2005. Yet this was not taken up as an issue of non-­compliance within the framework of the mine ban treaty, reflecting reluctance to address a sensitive issue head-­on.26 The narrative emanating from the Venezualan authorities shifted in 2008 to emphasize practical challenges – weather, safety, terrain – impeding clearance. Subsequently, some initial demining work was initiated in 2010.27 Through balancing humanitarian concerns with those of military necessity APII is, in principle, sensitive to concerns of national security. However, long-­standing

Humanitarian demining   93 accusations that Russia employed landmines in Chechnya without marking, fencing or monitoring would represent a clear breach of the country’s obligations under APII.28 In a parallel to the mine ban treaty, this issue has been repeatedly raised by civil society campaigners without being addressed within the regime. The political constraints that impede national implementation of programmes come as no surprise to the UN, NGOs and commercial companies at the sharp end of humanitarian demining. While some humanitarian demining commitments have been widely observed, important examples of voluntary and involuntary defection can also be identified. Voluntary defection is just one part of the challenge faced by mine action practitioners. From a regime perspective, the ­inability to acknowledge and address gaps in implementation is doubly harmful. In effect, a lack of commitment at the national level is compounded by a lack of political will within the regimes to address hard implementation issues. Verification Verification mechanisms play an important role in understanding and monitoring landmine regime implementation. The accuracy of some reporting is questionable given the lack of capacity in mine-­affected states to survey the national territory. In certain cases, this is coupled with an evident reluctance to advertise bad news. Consequently, the Philippines and Bangladesh, in their Article 7 reports, have rejected claims that there are known or suspected mined areas on their territory, despite reports to the contrary. In the case of Bangladesh, the army reported that border security forces – the ‘Na Sa Ka’ – had laid mines along the border with Myanmar. The border guard leadership subsequently denied this.29 These discrepancies undermine the value of regime reporting as a tool for planning and prioritization. The ICBL, in its de facto role as compliance watchdog, pays close attention through Landmine Monitor to the numbers of annual reports provided. Less notice is paid within the implementation process to the quality of the data provided. This may be explained by the fact that those responsible for the mine ban treaty meetings that consider these issues, do not have the necessary humanitarian demining background to question the reports. Shortly after the treaty entered into force, David Atwood, a prominent civil society figure within the Ottawa Process, acknowledged an important distinction between advocacy and implementation. He stressed that: there are multiple goals, most of which do not lend themselves to ‘fast track’ solutions . . . other changes away from the informal decentralized structure of the ICBL – so appropriate in the pre-­Convention phase – may be required for the ICBL to have similar effectiveness in this phase.30 This point, although made narrowly in relation to the ICBL, recalls the importance of stakeholder clustering dynamics for regime effectiveness as a whole. A lack of practitioner expertise in the implementation cluster weakens the regime’s ability to engage with humanitarian demining issues.

94   Humanitarian demining As already discussed, ANSAs represent a major category of mine users. If the legal prohibition is binding only on states, the reality is that rebel opposition groups, insurgents, militias and warlord factions are using landmines in numerous countries and regions.31 The ICBL, through Landmine Monitor, provides details of APM use by ANSAs, but these actors are not addressed directly within the framework of the regime. This is another example of dodging an issue that would cause disputes among regime members (notably those with armed groups on their territory) even though engagement would offer benefits for humanitarian demining. If annual APII reports provide significant detail on mine action activities, there is no link between this reporting and regime obligations. The protocol’s focus on the technical characteristics of landmines could add value for the ­demining community through clarifying information that would assist clearance. In Kosovo, following the NATO air campaign, details of bombing missions as well as demining activities in support of NATO operations, were only forthcoming after navigating lengthy institutional bottlenecks and problems linked to security classification. These obstacles were overcome through the persistence of mine action practitioners. This offers just one example of a missed opportunity for the regime to forge a dialogue between the mine action community and states parties that are NATO members, in order to support humanitarian demining. Stark contrasts mark the approaches of the two regimes to informal verification. Civil society has not mobilized behind APII. In part, this reflects the nature of the beast, with civil society actors excluded from all but the public plenary sessions of states parties’ meetings. It also reflects a conscious policy decision on the part of the ICBL, that engaging with a regime that advocates restrictions is inconsistent with the pursuit of a global ban. Regardless of these divergent interests and approaches, informal verification demonstrates a level of regime interplay. Civil society plays a de facto role in APII verification, since a section on CCW compliance is included in Landmine Monitor. Admittedly, this consciously emphasizes the humanitarian gap between the two regimes through highlighting non-­compliance by APII members. However, while advocacy does not extend to encouraging the implementation or universalization of APII, it does raise the stakes for non-­compliance. This may not change the behaviour of militarily significant states where national security concerns are at stake (e.g. Russia in Chechnya), but it has resulted in increased transparency among APII regime members. Verification of humanitarian demining-­related provisions is particularly vulnerable to an unwillingness to address defection. This is a vicious circle. In certain cases regime advocates are highly critical of non-­compliant behaviour. In others, problems tend to be de-­emphasized if they are off message in terms of the positive evolution of the regime. At the same time, non-­compliant states hide their conduct both to offset the political costs of exposure and also to avoid a potential loss of funding for humanitarian demining. These tensions became apparent at the 2004 Nairobi Summit, where there was clear friction between states and the ICBL over whether the draft action plan document should be

Humanitarian demining   95 amended during the meeting. Steve Goose, head of the ICBL delegation, commented that ‘I can’t help but feel that I am sitting in a CCW conference in 1994 and not an Ottawa conference in 2004.’32 By neglecting to follow through on gaps in annual reporting or ignoring cases of non-­compliance that are common knowledge, a major opportunity is lost to support humanitarian demining.

Regime effectiveness and humanitarian demining Humanitarian demining has witnessed a major shift in recent years from emphasis on numbers of landmines cleared, to a focus on their socio-­economic impact.33 The mine action community is not fully situated within this new paradigm; although a new way of understanding the issue has emerged, its operationalization, with the contingent benefits to mine affected communities, remains an ongoing process. Resources, political will and humanitarian impact are three perspectives from which to understand the value added of the two landmine regimes. Generating additional revenues for humanitarian demining provides a measure of effectiveness. However, a distinction needs to be made between resource generation as a result of the profile of the issue, and the effective targeting of these resources through the regimes. The potential misuse of funds also points to a second determinant of effectiveness: the role of the regimes in fostering political will to act. This notion is not fully expressed by membership choices but by states parties taking the difficult steps necessary to fulfil humanitarian demining obligations. Finally, the intersection between regime goals and the broader logic of IHL provides a third, humanitarian criterion for effectiveness. Resources An expert assessment published in 1999 on the first ten years of humanitarian demining programmes, recognized the need for ‘better coordination and cooperation in exchange for the large pledges put forward’.34 In September 2002, Norway, in advance of the 2004 mine ban treaty review conference, presented a non-­paper to ‘address all aspects of how to secure sufficient funding for reaching the aims of the Convention’.35 The non-­paper identified sustainability as the cornerstone of effective humanitarian demining programmes. This recommendation is frequently observed in the breach. To take one example, according to the UN-­commissioned review of the mine action programme in Kosovo, ‘mine action should not be a discretionary activity left to the charitable impulses of the donor community’.36 Yet the UN Mission’s budget for 1999–2000 was not made available to the programme in Kosovo until late 2000.37 This is by no means a unique case. Despite increasing levels of overall resources, Landmine Monitor records a lack of sustained funding as causing lay-­offs of personnel, or the halting of ­demining operations during 2006 alone in Afghanistan, Croatia, Guinea-­Bissau, Iraq, Mauritania and Tajikistan.38 While each case has its own particular circumstances, the issue of deminer lay-­offs highlights an important distinction between

96   Humanitarian demining global figures and local realities. The 2011 Landmine Monitor lists Myanmar (–95 per cent), Albania (–92 per cent) and Niger (–88 per cent) as the top three states in terms of reduced international mine action funding.39 While Albania’s position is explained by the successful conclusion of clearance activities, the other two cases display different local realities. Myanmar does not have a national mine action programme, and the extent of the problem (predominantly in border areas with Bangladesh and Thailand) is not known. A lack of engagement from national authorities has seen international assistance dwindle away. In 2010, the only reported funding was a modest grant from Norway to support advocacy activities by Geneva Call. Niger has only received funding from one donor (Switzerland),40 suggesting that the country’s landmine problem is not recognised as a priority within the mine ban treaty. This may change with the increased visibility generated by recently discovered mined areas. Resource constraints are not a new phenomenon. Humanitarian demining programmes frequently suffer from shortfalls or slowly disbursed funding. The Norwegian initiative is significant because, for the first time, the key issue of effective rather than adequate funding was put on the regime’s agenda. Until then, despite the prominence of the issue within mine action, no particular attention was paid to resource issues at the operational level. Significantly, in the intervening years since the non-­paper was tabled, there has been no substantive progress within the regime on bridging the gap between overall resource levels and their application in mine-­affected states. Commitments made by states with their own resource limitations provide a strong indication of political will. Obtaining this information is particularly challenging within developing states. Responses by mine-­affected states to a questionnaire issued by Norway point to a growth in mine action funding between 1997 and 2002 amounting to US$18 million from expanded state budgets. Beyond the traditional donor base, Brinkert identifies thirty-­six member states from all regions of the world defining themselves as ‘in a position’ to assist mine action in third countries.41 Beyond financial support, practical assistance generated by the regime from this category of states parties include: Argentine ­demining and explosive ordnance destruction in Kuwait; Brazilian support to mine action in Central America and Angola; Malaysian training for developing countries in demining and destruction techniques; and Paraguayan peacekeepers’ clearance work in the DRC.42 The allocation of funding for clearance by state-­ owned electrical utilities in Peru,43 and public companies in Croatia,44 also demonstrates that such efforts are not limited to governments. Although the scale of such contributions may be modest in comparison to those of Western mine action donors, they can offer a significant value added to humanitarian demining. Support from actors in the same region and with similar experiences, is particularly useful as it is more likely to be tailored to local needs. This kind of contribution is also immune to the criticism frequently levelled at mine action donors providing ‘Rolls Royce’ solutions to developing world problems. As with any development activity involving the transfer of resources from external actors to (typically weak) national bureaucracies, there have been

Humanitarian demining   97 various corruption scandals within mine action. Humanitarian demining is particularly vulnerable to this problem. In Cambodia, allegations in 1999 over falsified payroll records in the Cambodian Mine Action Centre (CMAC) 45 proved the tip of an iceberg that led to further revelations, including land cleared at the behest of military commanders, logging firms and a former Khmer Rouge leader. CMAC had insufficient management and oversight capacity, so in the absence of convincing data about its practices, the allegations could not be refuted. Donor confidence plummeted and funding was withdrawn.46 Because Cambodia is a high profile member, the regime would seem to offer an avenue to exert influence on their mine action conduct. In practice, neither capacity shortfalls nor corruption allegations have been addressed. Hard issues relating to inappropriate conduct have been sidestepped. A press release announcing the Phnom Penh meeting of mine ban treaty states parties is titled ‘Convention banning landmines to return to its roots.’47 Given that a national anti-­corruption law first proposed in 2003 has still not been enacted,48 it is questionable whether the hard lessons from Cambodia’s mine action history are actually being learned. If it is unsurprising that the mine ban treaty has made a positive contribution in increasing available resources for humanitarian demining, unprecedented support from non-­traditional donors, often with their own resource constraints, represents an important, unanticipated interaction effect. However, the lack of take up within the regime of Norwegian-­led efforts to apply a qualitative approach to resource issues, points to a damaging gap between treaty implementers and the humanitarian demining community. The de-­linkage of implementation stakeholders from field practitioners’ concerns over impact has clear implications for the regime’s ability to meet its humanitarian objectives. Political will By placing responsibility for humanitarian demining on states parties, the mine ban treaty recognizes the importance of national ownership (and thus political will) as a pre-­condition for achieving regime objectives. The mine action community has also identified the need for ‘a commitment to the development of skills and capacities which national and other authorities require in order to take effective command of mine action programmes’,49 as vital for the sustainability of programmes. This complementarity between the regime framework and humanitarian demining has found little echo in the management of mine action at the national level. In Kosovo, mine action was labelled from an early stage as an activity suit­ able to be handed over to the local authorities. The United Nations Mission in Kosovo (UNMIK) passed responsibility for humanitarian demining to the Kosovo Protection Corps (KPC), largely made up of former Albanian fighters and therefore closely identified with the Kosovo Liberation Army.50 Though viewed as a useful avenue through which to demobilize and reintegrate former combatants, this decision had several negative consequences. First, by taking the political decision to allot this role to the KPC, Kosovars with relevant demining

98   Humanitarian demining skills were sacked in order to make space. Second, poor work resulted since insufficient training was given to the KPC mine clearance teams.51 And third, it raised concerns over political and ethnic bias, particularly if deployed in Serb areas. Although senior management posts were still held by international staff, responsibility was passed to newly created government authorities on 15 December 2001. Unfortunately, in 2004, this role had to be moved back to the United Nations Special Representative, reflecting the need for greater oversight and control. In many ways this example is not unique, but it is significant because Kosovo represents one of the very few examples to date of a programme being handed back to ‘national’ ownership. In practice, the relationship between international assistance and national ownership is often fraught. The necessary political support from governments that is essential for humanitarian demining is often lacking. In Cambodia, international NGOs must wait months for new vehicles and other equipment that has already arrived in the country. A policy of never paying bribes means that the customs service does not release them.52 In the same vein, Grobbelaar cites an unnamed Southern African country where international demining staff need to reapply for a work visa every three months.53 Similar examples of self-­interested bureaucratic logjams could be cited in most mine-­affected countries. These problems are well known among donors, policy makers and practitioners. However, the mine ban treaty – which requires states parties to make every effort to assist humanitarian demining – plays no role in confronting such behaviour within its membership, or as a condition for humanitarian demining support. From the perspective of humanitarian demining, US support has remained strong throughout the last decade. The US contribution to humanitarian demining has not been reflected by any movement on a political level towards acceptance of the anti-­APM norm. Under the (second) Bush administration, there was a policy shift that de-­emphasized the eventual move to similar weapons systems without the associated humanitarian problems.54 This stance has been confirmed under the Obama administration, with mine policy remaining the subject of a lengthy review process. These ambiguities in the US position seem to reflect similar dynamics to those that determined the US orientation towards the 1925 Gas Protocol. Despite declaratory policy, mine action support demonstrates commitment to the mine ban treaty’s humanitarian goals. Longstanding non-­use of APMs also falls in line with a core provision of the mine ban treaty. If this position may reflect incoherence at the national level, its consequences contradict the pariah label often raised as a reaction to US non-­ membership of the mine ban treaty. Although it can be argued that US commitment to a ban would create emulation effects among other states, it is questionable whether a US policy change would influence key militarily significant rejectionists such as China or Russia. If a ban is off the table for the forseeable future, it is certainly a significant political step that these states are part of APII. The benefit is less clear from a humanitarian demining perspective. Regime obligations have no bearing on Russian and Chinese landmines already in the ground in post-­conflict and conflict-­ridden

Humanitarian demining   99 states around the world. Compliance with APII detectability requirements will have the potential to facilitate future humanitarian demining activities. However, unlike the US, the proliferation of Chinese and Russian weapons over past decades has not been offset by a countervailing commitment to humanitarian demining that has increased markedly since the entry into force of the regimes. In keeping with its leadership role within APII, the US has invested significantly in alternatives to APMs as well as ensuring compliance of existing systems with the regime’s detectability requirements. Measures to make landmines more readily detectable reduce the risks faced by deminers, so contribute to the mine ban treaty’s goal of reducing humanitarian suffering. The contrast between this commitment and a policy position that is no nearer a ban, provides further evidence that embracing the core political message of the Ottawa Process (a universal ban at all costs) is not a pre-­condition for substantive progress. Ken Anderson questions the value of the mine ban treaty without US membership.55 Yet the nuanced reality is that support for humanitarian demining makes a contribution to regime objectives that transcends partisan regime politics. The topic of ANSAs has been discussed at mine ban treaty meetings since 2005. As with verification and compliance monitoring, informal arrangements have emerged because the regime proper has not addressed a key mine action issue. The role played by Geneva Call is particularly relevant from a humanitarian demining perspective. The increasing importance attached to such obligations by the international community, is demonstrated by the fact that the European Parliament has called on ANSAs to sign the Geneva Call Deed of Commitment.56 Despite resistance from certain quarters within the regime, Geneva Call could not function without the normative pull of the mine ban treaty. Stigmatization also generates emulation effects below the level of the state. For ANSAs that control mined areas, acknowledgement of the illegitimacy of APMs is an essential first step to opening up that land for clearance. This is also evident on the ground, contributing to a common perception of demining as an ‘honourable’ profession. The regimes are linked through the political will that underpins support for humanitarian demining. On one level, deep linkages are founded on a recognition that this activity contributes to a shared humanitarian imperative. Stigmatization is another common thread. The de facto support of the US for the humanitarian demining objectives of the mine ban treaty, and the willingness of different ANSAs to give up these weapons represent qualititatively different acknowledgements and consequences of their illegitimacy. However, if ANSAs have been a topic of discussion within the mine ban treaty, it remains a subject that the regime has not grasped as an implementation issue. Humanitarian impact Humanitarian demining is a key focus for concerns of regime effectiveness because of its direct ability to reduce the suffering caused by landmines. It is also important to acknowledge that landmines represent only one part of a broader challenge to build peace and lay the foundations for sustainable development. An assessment of

100   Humanitarian demining the humanitarian impact of the regimes therefore needs to focus on two dimensions: the link between obligations and humanitarian pay-­offs; and, the role of the regimes in linking humanitarian demining to broader security and development goals. Article 6 of the mine ban treaty encourages cooperation to improve the use of technology in humanitarian demining. This chimes with a UN-­sponsored study into global operational needs for mine action which finds that ‘in some cases, donors have forced unsuitable and ineffective equipment on national programmes and local demining projects. This has harmed the relationship between donors, researchers, industry and the user community’.57 Rae McGrath, an experienced mine action practitioner, points out that a disconnect exists between donor priorities and the needs of affected communities: ‘the indicators of success set by donors often totally ignore the priorities of the indigenous authorities and may only consider overlapping interests in other sectors of development where those projects are funded by the same donor’.58 This reflects a common perception among practitioners: that resources are wasted on expensive, inappropriate equipment that breaks down once exposed to the rigours of climatic and other conditions. The unavailability of spare parts only adds to the problem.59 Although gaps between laboratory conditions and field realities are common knowledge, inappropriate donor support has not been an issue in the mine ban treaty work programme. While intersessional meetings bring together donors and representatives of mine-­affected states, practitioners from the humanitarian demining community that could speak to these problems are less in evidence. Two significant conclusions are reinforced. First, a lack of technical expertise in the implementation phase prevents problem-­solving. And second, despite efforts to ensure broad developing world representation, there remains an inherent asymmetry between donors and recipients, unaddressed within the regime, that stops the latter from criticising the support they receive for fear of receiving less in the future. This insight echoes Beier’s observation; that while the Ottawa Process ‘collapsed political time’ in the regime design phase, there is ‘unchanged political space’ for the developing world in the regime’s implementation.60 Although APII’s stipulated self-­destruct and self-­neutralize requirements would prevent death or injury after the end of a landmine’s active life, in practice this would provide no additional benefit to humanitarian demining. All objects emitting a metal signature must be excavated in the same way, and all landmines must be treated as ‘live’ for the purpose of clearance. As demining expert Tim Lardner points out, ‘it doesn’t matter to my deminers on the ground whether there are no mines or whether there are a thousand mines, in the five hectares of land they are told to clear’.61 Moreover, as demonstrated in the UN Global Needs Study, the requirement to clear surrounding vegetation and excavate every metal source in a contaminated area means that the numbers of landmines, dangerous or otherwise, has limited bearing on the speed of clearance.62 The humanitarian benefits resulting from the application of these provisions are therefore more than offset by the threat posed by landmines already in the ground. APII detectability requirements are considered within the regime as a major contribution to humanitarian demining. Michael Matheson, chief US negotiator

Humanitarian demining   101 at the 1995–1996 CCW Review Conference, correctly argues that greater detectability of APMs – instead of the difficult to detect and long-­lived APMs frequently used in developing world conflicts around the globe – saves lives among deminers and military personnel operating in a mine-­affected area.63 China has already met its detectability obligations, while Russia deferred compliance to 2014. These states and their militaries have recognized the hazards to their own forces of undetectable APMs, and phased them out on their own initiative. The regime thus provides a framework that unites self-­interest – from the point of view of military utility – with a humanitarian demining pay-­off. However, it should be self-­evident that civilians in affected states do not have detection equipment. It is therefore wrong to suggest (as Matheson does) that this has led to a significant reduction in civilian casualties. Increased detectability (as opposed to self-­destruct or self-­neutralize capabilities) even if adopted universally, would provide no significant humanitarian benefits. Moreover, the detectability provision in APII only relates to current and future stocks of APMs. Again, this fails to recognize the temporal dimension; from a humanitarian ­demining perspective, the vast majority of the post-­conflict hazards to be cleared pre-­date APII, so its restrictions do not apply. The rapid identification and clearance of high-­impact areas of mine infestation can lead to major benefits. However, for many countries humanitarian ­demining will be a matter of decades. In contrast, given the resources available and the painstaking nature of manual clearance, this work will result in commensurately diminishing returns when weighing the costs against humanitarian benefits. While this trade off was acknowledged during the mine ban treaty design phase, ‘100 per cent clearance’ was retained as the benchmark for compliance. The distinction between absolute goals and relative benefits shows how understandings of effectiveness differ. There is a conviction – built into regime design and carried through to implementation – that failing to define and meet unequivocal targets will undermine the credibility of the mine ban treaty. From a humanitarian demining perspective, the key question should be whether the investment of scarce resources in striving to achieve absolute targets provides a justifiable humanitarian or developmental pay-­off. If there is minimal risk then scarce resources are better invested elsewhere. In the implementation phase, this divergence is captured in the distinction between ‘mine free’ and ‘impact free’. This distinction is perhaps best examplified in the Falkland Islands. Argentine minefields have been perimeter fenced and monitored by British forces, with the result that the only casualties recorded in the twenty-­five years since the end of the conflict are British soldiers involved in mine clearance.64 The landmine and UXO contamination has no humanitarian impact, while the territory that is mined – beaches and peat areas65 – have little or no socio-­economic value. Moreover, the permanent UK military presence regularly conducts mine awareness training and ensures that perimeter fencing and marking remains in place. Although under Article 5 of the mine ban treaty the UK is obliged to clear the Falklands minefields, both the islanders and the UK government have suggested that efforts would be better invested in assisting

102   Humanitarian demining severely mine-­affected countries. Even the NGO Landmine Action, highly critical of government policy in the regime formation phase, proposed under the ‘Falklands Initiative’66 that a ten year clearance extension be granted to the UK in exchange for additional revenues to be provided for severely mine- affected states. While the UK was ultimately granted a deadline extension, the lack of humanitarian impact has had no bearing on the UK’s treaty obligation. Facing harsh criticism from the ICBL, ICRC and some member states on its lack of progress, Falklands clearance operations began in late 2009 under the threat of the UK being branded non-­compliant.67 Criticism resurfaced at the Phnom Penh meeting of states parties. The UK was strongly censured both for the slow pace of clearance, and also for an apparent reinterpretation of its obligations through focusing clearance on areas of ‘humanitarian impact’ as opposed to all landmines.68 At least one state party may have misinterpreted obligations through describing measures to emplace permanent minefield marking – thus implying that the land would not be cleared.69 This represents a qualitatively different case from those that seek to balance the merits of clearance against the level of impact. The example of Kosovo, where responsibility for mine action was prematurely handed to newly founded, provisional government authorities, provides further evidence of the complexity of this issue. There has been significant disagreement between UNMAS, who declared in 2002 that ‘the problems associated with landmines, cluster munitions and other items of unexploded ordnance in Kosovo have been virtually eliminated’,70 and the HALO Trust, who subsequently resumed clearance operations in response to the continuing high numbers of civilian casualties. Local mine action authorities detail relatively few dangerous areas, yet ‘HALO knows these figures to be incorrect and is planning a joint survey . . . to define the remaining threat from cluster munitions and landmines’.71 As of March 2011, the Kosovo Mine Action Center identified twelve confirmed mined areas and forty-­four suspected mined areas.72 The same problems of certitude apply whether the objective is to be mine or impact free. These examples highlight the inherent tension that may arise between efforts intended to safeguard the clarity of the anti-­APM norm and good practice in humanitarian demining. For practitioners, if the consequence of regime obligations is to push states to expend resources in the hunt for the ‘last mine’ that could better be used elsewhere, then this is irrationally wasteful. However, deficits in resources and political will that contribute to states parties inadvertently or deliberately failing to meet their obligations show the that the humanitarian imperative underpinning the regime must be nurtured, even if this results in ­efficiency costs. APII provides for a qualitatively different approach to these concerns. Humanitarian demining issues and expertise were marginal to the negotiation of the regime. This is apparent in provisions that ignore temporal considerations or challenges faced by mine-­affected countries. Emphasis on the technical characteristics of weapons leaves little room for consideration of the civilians affected by them. Thus, while the regime generates positive effects, it also shows how

Humanitarian demining   103 humanitarian demining-­related provisions can make minimal humanitarian impact. The humanitarian imperative underpinning both regimes points to the need to understand humanitarian demining in relation to wider security and development challenges. An emerging policy discourse attempts to situate humanitarian demining within broader peacebuilding and development fields.73 This perspective highlights the important enabling role that humanitarian demining can play in providing security, rebuilding economies, transport and other infrastructure, as well as enabling the creation of jobs. Two examples of unintended consequences demonstrate the need for this activity to be better situated in relation to these wider processes. In Mozambique, road clearance was prioritized as an immediate way to support the UN peacekeeping mission. Contracts worth hundreds of thousands of dollars were given out to clear 2,000 kilometres of roads. Yet after months of operations, only six mines were uncovered.74 Worse, Harpviken and Roberts describe how the demining and reopening of Highway 9 in Sri Lanka enabled the internally displaced to return home to settlements that had not themselves been cleared, leading to casualties among returnees.75 Avoiding waste, and the imperative to ‘do no harm’, suggests that mutually reinforcing links need to be carefully developed between mine action and post-­conflict peacebuilding activities such as disarmament, demobilization and reintegration (DDR) of ex-­ combatants or security sector reform (SSR) programmes. The mine ban treaty draws together states and multilateral institutions with humanitarian, security and development priorities of which humanitarian ­demining forms only one part. Although the APII provision on information exchange following the cessation of hostilities is weak, it does make a reference to the need to include mine clearance obligations in peace agreements.76 This points to an unrealized potential for the integration of mine action with other peacebuilding activities through the convening power of the regimes. Better integration with wider peacebuilding and development should be a realistic objective that can generate clear humanitarian pay-­offs. However, potential spillover effects have not been considered within the framework of the regimes. This echoes a conclusion drawn from a review of donor policy that ‘the link between mine action and peacebuilding is generally acknowledged, but poorly developed’.77 Canada has established a Contact Group on ‘Linking Mine Action and Development’, within the mine ban treaty implementation process, to address this issue.78 However, the promotion of synergies has been met with resistance. In part, this reflects concern that integration will lead to a reduction in overall funding levels or a dilution of the effectiveness of humanitarian demining programming.79 But, it is also an acknowledgement that the success of the Ottawa Process has stemmed from the way APMs have been isolated and stigmatized. The relative lack of integration of humanitarian demining is thus telling. It shows that there is actually a common interest among mine ban treaty partisans and mine action practitioners to remain apart. While rarely discussed, such a strategy has costs for regime effectiveness (albeit some benefits) from the perspectives of resources, political will and humanitarian impact.

104   Humanitarian demining

Conclusions An important distinction can be made between overtly beneficial ideal provisions and their real world effectiveness in supporting humanitarian demining. A number of stipulations in both regimes, through design flaws and compromises, are largely redundant in the face of a combination of climate, geography and the socio-­economic needs of local populations. Through failing to address field realities in regime design, international regime commitments have in some cases not been translated adequately into national implementation processes. If the regimes have mobilized new resources, they do not contribute to ensuring that funding results in targeted, sustained and effective support to humanitarian demining. Facile assumptions, that more money equates to greater effectiveness, reflect an important split between an evolving mine action discourse and implementation processes that are constrained in their ability to learn and adapt. The regimes are particularly weak in focusing the influence and conditionalities that could be applied when faced with hard cases. Thus, while misuse of donor resources is open knowledge in many mine-­affected states, the issue stays off the table. Emulation effects have swollen the ranks of the mine ban treaty. However, a different kind of political will is necessary to implement humanitarian demining-­ related provisions. For many mine-­affected states, the desire to cut a positive international figure has not been followed up with domestic implementation measures or, in some cases, has led to misleading compliance reporting. Moreover, the number of applications for extensions to clearance deadlines shows the scale of the challenge for mine-­affected states, regardless of their willingness to meet obligations. Domestic implementation is either not a priority, or is beyond the means of many states. If the absence of domestic implementing legislation clearly signals this reality, it has not provided a catalyst for action. The inability within the mine ban treaty to link political will for regime formation to a commensurate commitment to implementation, nuances how the influence of norms should be understood across different phases of regime development. The lack of focus on hard implementation challenges also further clarifies the costs of disjunctions between treaty and mine action priorities. This highlights an unacknowledged blind spot that ignores humanitarian demining-­related implementation challenges in practice. Neither regime sufficiently benefits from the increasingly effective networking of humanitarian demining expertise. The mine ban treaty implementation process is de-­linked from this practitioner discourse. The APII regime is ­similarly disengaged, since its focus is not on humanitarian demining but on technical issues of detectability and self-­destruct/self-­neutralize that underpin the regime’s restrictions-­based approach. The consequence is: that while regimes package and disseminate humanitarian demining-­related information, capacity is lacking to assess this in qualitative terms and identify necessary course corrections. In this respect, a lack of developing world buy in during the implementation phase,

Humanitarian demining   105 p­ rovides an important corrective to the accepted narrative of the mine ban treaty as a participatory regime. Design flaws have not been addressed as a result of a lack of depth to decision-­making structures within the implementation process. Tensions between the concepts of ‘mine free’ and ‘impact free’ highlight an important cleavage between the mine ban treaty and humanitarian demining priorities. There is a clear challenge to regime effectiveness if states parties declare themselves mine free for political reasons or through a lack of appreciation of the threat. On the one hand, if ‘mine free’ becomes a negotiated term, rather than an absolute requirement, then this would permit the better targeting of resources to humanitarian demining operations. On the other hand, this runs the risk of eroding the key normative foundation of the regime, and thus reducing the pressure on members (as well as non-­members) to fall in behind regime goals. Lowering the bar for compliance to ‘impact free’ could also increase the risk of voluntary or involuntary defection through states misunderstanding or deliberately misinterpreting their obligations. The dilemma posed by non-­alignment of regime and mine action goals represents a key challenge for the evolution of the mine ban treaty regime.

6 Stockpile destruction

Introduction Destruction is the only way to ensure that stockpiled landmines are never used. A growing awareness of the importance of stockpile destruction is reflected in the emphasis placed within the two treaty regimes on states meeting their destruction deadlines. Stockpile destruction also offers an inherent attraction for political and advocacy purposes since it is easily measured (in comparison to activities such as mine risk education or victim assistance) and thus can provide a clearly demonstrable success story. However, it is necessary to unpack the meaning of ‘success’ within the regimes. This can be done through analysing potential tensions between a focus on mines destroyed as a quantitative (and highly visible) measure of effectiveness, and emphasis within the mine action community on socio-­economic and humanitarian impact as key determinants of effective mine action. If states parties fail to conduct stockpile destruction this would undermine regime credibility. With different memberships and rules, the regimes adopt contrasting methods in addressing the resource, political, and humanitarian dimensions of stockpile destruction. Common to both, is that fulfilling stockpile destruction obligations requires a combination of technical capacity and political will. Because of the costs of stockpile destruction and the national security implications of giving up landmines, the ability to identify and deal with defection is particularly significant. Interplay dynamics that result in effects beyond regime membership may also be important given that states with huge stockpiles of APMs are not members of the mine ban treaty, while a number of APII member states have deferred compliance with certain key provisions. Disconnects between stakeholder clusters involved in design and implementation may be particularly important because of the central role played by states and commercial companies in stockpile destruction. Moreover, if stockpile destruction is addressed solely from the perspective of state actors and stockpiled landmines held and used by ANSAs are ignored, overall regime effectiveness will be undermined. Thus, while IHL in general, and the landmine regimes in particular, may display a state-­centric bias, how the regimes interact with international and sub-­ state actors will be particularly significant for effectiveness in this area.

Stockpile destruction   107 The mine ban treaty contains provisions dealing specifically with stockpile destruction, while this activity is a consequence of complying with a number of restrictions contained in APII. This chapter begins by critically analysing stockpile destruction implementation through the regimes. New insights into why defection occurs, are developed through relating obligations to technical and political obstacles to stockpile destruction. It then considers the effectiveness of measures to destroy stockpiled landmines through the two regimes under the themes of resources, political will and humanitarian impact. This chapter concludes by considering new insights drawn from an individual and comparative analysis of the role of APII and the mine ban treaty in this field.

Regime implementation and stockpile destruction Regime rules and field realities It is a central tenet of regime theory that rules should reflect a clear understanding of the measures required to implement them. While APII and the mine ban treaty are very specific in relation to the weapons they address, the official definition of stockpile destruction set out within the IMAS, and applied by the mine action community, does not clarify what should be destroyed: APMs, AVMs or other weapons and ammunition.1 This reflects the reality that landmines are frequently stockpiled with other munitions that may themselves pose a hazard. Consequently, the destruction of landmines is not distinguished from other explosive ordnance. Conceptual ‘fuzziness’ also avoids narrow alignment with the definitions and objectives of a single regime framework. The contexts within which the activity takes place suggest that a distinction should be made between three types of stockpile destruction: the disposal of competently stored state inventories; the eradication of arms caches found in post-­conflict environments (as undertaken in Iraq following the US-­led invasion); and, the destruction of weapons found and grouped for destruction as a result of humanitarian demining activities. In the first category, the two landmine regimes have made a clear, measurable, and with the mine ban treaty, dramatic impact in reducing global landmine stockpiles. Landmine Monitor calculates that in the mid-­1990s, 131 states possessed stockpiles of more than 260 million APMs. Estimates for 2010 suggest that thirty-­five countries now stockpile about 160 million APMs.2 Among mine ban treaty states parties, 151 of 156 states parties do not hold stockpiles. Together, this amounts to some forty-­five million APMs destroyed since the treaty’s entry into force.3 APII obligations have also resulted in the significant destruction of non-­compliant stocks. China, in November 2005, declared that it had destroyed some 500,000 non-­APII compliant landmines. The US completed the destruction of over 3.3 million dumb (non-­self destruct) landmines in June 1998, retaining only those needed for training and research as well as for use in the South Korean de-­militarized zone.4 These figures demonstrate that both regimes have had a measurable impact in reducing global landmine levels.

108   Stockpile destruction In the latter two categories (post-­conflict caches and weapons collected through humanitarian demining), destruction activities have always formed a part of mine action. It is therefore necessary to distinguish regime-­specific effects. One important set of implementation issues relates to how far the provisions for stockpile destruction are consistent with demining field realities and requirements. Implementation also throws up politically sensitive questions on defection that relate to how stockpile destruction obligations are interpreted by regime members. Article 4 of the mine ban treaty obliges members to ensure the destruction of all stockpiled APMs ‘it owns or possesses or that are under its jurisdiction or control, as soon as possible but not later than four years after the entry into force of this Convention for that State Party’.5 The timetable for the destruction of stockpiled APMs is clear. Unlike Article 5, which sets a deadline of ten years for the clearance of mined areas but allows the possibility of extension periods, the stockpile destruction deadline is non-­renewable. Similarly, although APII contains no specific provisions on the destruction of stockpiles, if a state party is unable or unwilling to make stocks compliant with provisions on detectability, self-­destruction or self-­neutralization, then destruction is the only feasible alternative. As Maslen notes, ‘since the Protocol prohibits the use or transfer of an unlawful weapon, it is likely, in practice, to destroy it’.6 A deferral period of nine years for these provisions therefore represents a de facto deadline for stockpile destruction (unless stocks are adapted for compliance). Stockpile destruction continues to be a significant challenge for mine ban treaty regime implementation. Remaining stocks held by regime members represent the hard cases in terms of national capacity and political will. As of November 2011, these include Belarus, Greece, Iraq, Turkey and Ukraine. All bar Iraq are now formally in breach of their Article 4 destruction commitment. Outside of the mine ban treaty, the major holders of stockpiles – China (110 million), Russia (24.5 million) the US (10.4 million), Pakistan (six million) and India (4–5 million) – are APII members, and therefore must ensure compliance with its provisions.7 The relevance of APII is further emphasized by the fact that these landmine stockpilers are highly unlikely to sign up to an APM ban in the foreseeable future. The 2004 Nairobi Action Plan, calls for two mutually reinforcing sets of actions geared towards achieving the treaty’s stockpile destruction objectives. First, states parties are requested to identify and report on all stockpiled APMs to establish the capacities necessary to meet their deadlines and to develop plans, identify priorities, problems and requirements for assistance in a timely manner. Second, states parties ‘in a position to do so’ are required to assist others in meeting stockpile destruction obligations.8 Recognizing the danger of losing track of whether landmines have been stockpiled, retained for training purposes or destroyed, Action 15 of the Nairobi Action Plan stipulates: When previously unknown stockpiles are discovered after stockpile destruction deadlines have passed, report such activities in accordance with their

Stockpile destruction   109 obligations under Article 7, take advantage of other informal means to share such information and destroy those mines as a matter of urgent priority.9 The disjointed relationship between these requirements and field realities is well illustrated by the experience of one mine-­affected state. Angola, with work hampered because landmine caches are frequently found in heavily mined and difficult to access areas, has suggested that it may require an extension to its deadline for destruction of APM stockpiles. The terse statement in Landmine Monitor to the effect that ‘the Mine Ban Treaty does not allow extensions for stockpile destruction’,10 fails to acknowledge or address the implementation challenges that caused Angola to miss the deadline in the first place. Though in breach of its treaty obligations since January 2007, Angola is a heavily mine-­affected state, which is in need of assistance. Ignoring this reality demonstrates a gap between the principle, and practice, of regime support. Involuntary and voluntary defection A state’s ability to meet their obligations is a key determinant of successful implementation. Given the cost and potential complexity of stockpile destruction, the danger of involuntary defection would seem to be significant for mine-­ affected states. This poses a dilemma, since many countries have neither the resources nor the technological base to develop appropriate destruction facilities. A state’s willingness to meet regime obligations is at least as important. The potential for voluntary defection therefore represents another important factor for regime effectiveness. In a statement to the Standing Committee on Stockpile Destruction, the DRC declared that it had fulfilled its stockpile destruction commitment by destroying all stockpiled APMs ‘that it had been able to identify’.11 The statement went on to acknowledge the likelihood of finding additional stockpiles in the future, in parts of the country under the control of ANSAs. The existence of such stocks has been confirmed through the small numbers of APMs handed over to the United Nations through its DDR programme.12 For the ICBL, this statement creates an unsatisfactory doubt as to whether the DRC is compliant with its Article 4 obligation.13 However, the ambiguity is not the consequence of a lack of clarity, but rather a reflection of the inability of the government to exert oversight and control over the national territory.14 A December 2005 meeting on treaty implementation in the DRC, demonstrated a further dimension to the challenge by highlighting the absence of a coherent structure within the government to oversee the country’s treaty obligations.15 This situation was exacerbated by instability during a period of political transition in the build up to national elections. Afghanistan declared in May 2006, that all ‘known’ stockpiles would be destroyed by its March 2007 deadline under the mine ban treaty. At the same meeting, Afghan representatives acknowledged the challenge of addressing ‘the need for local commanders and villagers to voluntarily give up their stockpiled

110   Stockpile destruction caches or to disclose the locations of them’.16 If Afghanistan’s stockpile destruction deadline was missed by six months, the formal notification of the completion in October 2007 also retained the proviso that this relates to ‘known’ stockpiles.17 Subsequently, further small quantities of APMs have been recovered from caches, while there has also been reporting of minelaying by Taliban forces. In September 2009, US forces in Kandahar province attributed extremely high casualty rates to the Taliban’s use of victim-­activated explosive devices.18 How far provisions provide latitude for differing interpretations and enable regime members to evade their obligations is an important consideration. This challenge has been exemplified in the mine ban treaty context by Turkmenistan, which inherited a large stockpile of APMs from the Soviet Union. Its first Article 7 report, submitted in October 2001, stated that 761,782 APMs remained from an initial stockpile of 1.7 million. Destruction was estimated to take eight years, and a deadline extension was requested.19 Interventions by the co-­chairs of the Stockpile Destruction Standing Committee saw Turkmenistan clarify that it would meet its deadline and that only 250,000 APMs remained.20 In subsequent reporting of its successful destruction programme, it was announced that some 70,000 APMs would be retained for training and testing as provided for under Article 3. This figure is well beyond the numbers understood by other states parties to be necessary for this purpose. The country was therefore criticized for breaching the spirit of the treaty. The Turkmen Embassy to NATO subsequently wrote to the European Commission, stating that in February 2004 some 60,000 further APMs would be destroyed.21 Then, in April 2005, all stocks including those originally reserved for training were reported destroyed. An addendum provided by Landmine Monitor points out that all the earlier figures were wildly inaccurate, since the authorities had confused the number of containers with the number of individual APMs. Rather than 70,000 APMs, the real figure at that point was 572,200!22 Clarity, support from regime members, coherence and links between primary obligations and enabling provisions are important considerations for the implementation of regime rules.23 Although formally joining the regime on 1 March 1999, Turkmenistan only attended an intersessional meeting for the first time in June 2004 and failed to attend subsequent meetings. Its first two Article 7 reports were incomplete and the third consisted of a four-­sentence statement. The short space between signature of the treaty (3 December 1997) and ratification (19 January 1998) reinforces the hypothesis that provision for domestic implementation modalities represents an important indicator of commitment levels at the national level. These cases allow the identification of missed opportunities, but also show the limitations of the regimes in preventing defection. Reporting by Afghanistan and the DRC highlights the practical difficulties certain members face in ascertaining the numbers and locations of APM stockpiles. Declarations by states are shown to be inadequate where ANSAs retain de facto control of parts of the national territory. It is untenable for the government to claim full compliance with its national stockpile destruction obligations when armed groups hold

Stockpile destruction   111 stockpiles, and in some cases have been accused of mine laying. Stockpile destruction challenges also demonstrate the potential for abuse as well as the lack of capacity, in some mine-­affected states, to internalize and implement international obligations through domestic bureaucratic structures. This lack of depth to the central state – understood as the inability to effectively implement, manage and oversee commitments at the national level – renders meaningless a quantitative approach to implementation. Instead, cases of involuntary defection show the extent to which a lack of emphasis on national capacity building can undermine regime goals. Turkmenistan’s stockpile destruction history provides its own insights. This case shows how capacity gaps can lead to regime defection. It also offers a cautionary example of the limitations of regime support. Given that Turkmenistan has few diplomatic ties to other countries and very limited engagement in the mine ban treaty intersessional work programme, there has been little scope for engagement. The difficulty of dealing with a state that has severe capacity deficits and does not want, or is not able, to engage in the implementation process, is apparent. Verification More than withholding use, destruction represents the ultimate step in giving up a viable military capability. In this regard, compliance monitoring and verification provide essential checks and balances on regime members. The mine ban treaty requires states parties to report on the size of APM stockpiles, the numbers and types of mines retained for training in humanitarian demining-­related activities and progress in stockpile destruction.24 The priority accorded to accessibility of information has resulted in reports bring made available on the website of the United Nations Department for Disarmament Affairs.25 APII reporting requirements do not specifically mention stockpile destruction, but do require reporting on ‘steps taken to meet technical requirements of this protocol’.26 To facilitate this task, states agreed to a common reporting format. VERTIC, the UK-­based organization specializing in verification of disarmament treaty obligations, developed a guide to assist states parties in fulfilling reporting requirements. E-­MINE is another mechanism that supports stockpile destruction, providing a reference point for technical papers, guidelines and lessons learned on stockpile destruction.27 An in-­kind donation from Canada to UNMAS enabled the system’s development. It represents an example of bilateral mine action support to the UN by a member of both regimes that contributes directly to common goals. Enhanced transparency aligns regime and mine action interests. This has led to increased confidence among members, while contributing to implementation through making available consolidated advice on stockpile destruction. In the absence of formal verification as part of either regime, civil society mobilized through the ICBL plays a central role in verification of mine ban treaty stockpile destruction provisions, while also monitoring CCW compliance.

112   Stockpile destruction As with humanitarian demining, Landmine Monitor represents the most important source for information and analysis on compliance with stockpile destruction obligations. Despite this common informal verification framework, stockpile destruction is different from other pillars of mine action because of the dominant role played by states and commercial companies. Landmine Monitor’s function is thus particularly significant as a mechanism that provides alternative views coming from within the countries in question. The state or commercially driven nature of stockpile destruction means that this may be the only available evidence base on this issue. Russia ratified APII on 2 March 2005 and, for the first time ever, disclosed publicly the size of its landmine stockpile.28 Implementation is particularly complex because Russia possesses between 7–10 million PFM-­1 type APMs that are highly sensitive and well beyond their shelf life.29 In order to meet its obligations, Russia developed a method of destruction, involving encasing in concrete, which has been widely acknowledged as crucial in destroying over thirteen million of these weapons to date.30 An alternative perspective is provided by Landmine Monitor’s contributor on Russia. According to R. Dogov, destruction plans and adequate resources are not in place, reflecting a lack of commitment by Russian authorities and an absence of engagement by international actors. He suggests that inappropriate storage conditions and ‘self-­switching’ of mines to combat mode pose a very high risk of self-­detonation. This forms part of a broader problem, acknowledged by Dogov, that ‘In Russia itself, demining activities lack standardization, verification, and control mechanisms. They cannot be regarded as humanitarian, and the scale on which they take place is inadequate.’31 The 2008 Landmine Monitor reports ANSA stockpiles of APMs on the territory of seven countries.32 In most cases, these numbers represent stockpiles seized by government forces. Verification therefore provides a very partial picture that is of limited utility because it fails to capture the size of the stockpiles actually available to armed groups. Once seized, the weapons are no longer a humanitarian threat unless states fail to destroy them in line with their obligations. Geneva Call acknowledges the difficulty of obtaining good quality data on stocks held by ANSAs, attributing this in part to reluctance by groups to show their hand. In line with defection challenges faced by national actors in some mine-­affected states, it also reflects a problem of oversight and control.33 Factory and homemade devices can be cached, buried under ground, or in some cases hidden in the homes of civilians. Verification of ANSA stockpiles thus represents a doubly challenging regime obligation for national authorities and international actors since both quantitative and qualitative data are difficult to obtain. The diverse actors within the mine ban treaty implementation cluster are mobilized in ways that encourages a level of problem-­solving. States parties are lobbied by civil society to address stockpile destruction issues, while common interests have fostered links to expert communities – both donors and technical specialists – that contribute to dealing with challenges. In contrast, APII states parties report on stockpile destruction in order to demonstrate compliance with

Stockpile destruction   113 regime rules. There is limited discussion on how regime commitments are to be achieved, and no mechanism permitting comparative experience to be shared or advice sought that could address implementation challenges. Analysing official versus civil society views of Russia’s stockpile destruction record is instructive. Providing information on APM stockpiles for the first time in the country’s history represents an important step. But the fact that this only occurred in 2005,34 demonstrates a response to a regime commitment as opposed to a desire to support humanitarian goals. Moreover, Russia’s stockpile destruction progress cannot be isolated from the reality of its continued use of APMs. The absence of active verification informed by mine action practitioners thus perpetuates the lack of confidence reflected in successive Landmine Monitor reports.

Regime effectiveness and stockpile destruction While stockpile destruction is now understood as a key element of mine action, it continues to receive limited attention on a policy level as a regime implementation issue. This reflects the widespread perception that stockpile destruction is purely a ‘technical’ activity. However, this masks a need for flexibility that is particularly relevant as implementation challenges become more apparent over time. Increased resources and political will in this area represent two key indicators of regime effectiveness.35 Furthermore, if stockpile destruction is carried out ineffectively, or provokes other negative consequences, then it fails to meet the humanitarian criterion that underpins both regimes. Increased support for stockpile destruction as a result of the influence of the regimes is one clear indicator of effectiveness. Yet achievements must also be set against the scale of the task and the likelihood of goals being met. Responding to new and changing demands is crucial. This means addressing stockpile destruction requirements or challenges that arise during the implementation process, including the consequences of divergent interpretations of regime obligations. To gauge effectiveness in this area, three criteria understood as essential to realising regime goals are applied. Resources made available for this activity are a key consideration given the costs associated with stockpile destruction and the resource deficits faced by many mine-­affected states. Political will refers to the ability to gain support for regime objectives among members and non-­members, as well as international and non-­state actors. Finally, this section considers the humanitarian criterion for effectiveness. How far has the regimes’ focus on stockpile destruction made an impact on the humanitarian suffering caused by these weapons? Resources While stockpile destruction has been folded into wider security assistance programmes in different regions, few donors have placed significant emphasis on this activity as a priority mine action commitment. One argument explored below, is that this may be attributed to its apparently technical nature in comparison to the

114   Stockpile destruction more overtly ‘humanitarian’ pillars of mine action. This makes sense in the context of the norm bandwagon effect associated with the evolution of the mine ban treaty. Moreover, as an activity mostly conducted by national militaries and commercial companies, stockpile destruction does not involve a diverse implementation cluster when compared to other mine action sectors. Very different levels of visibility are apparent from design to implementation. Regime formation was internationalized and therefore relatively transparent, thus exerting political pressure on states. Implementation takes place at the national level where both pressure and support – in the form of expertise and resources – are lacking. This comes with costs for regime effectiveness. Canada provides one of the few examples of a bilateral donor that contributes significant resources for stockpile destruction, providing technical expertise, material, equipment and financial support. Canada has supported mine action in Central America through the OAS (Organization of American States) Comprehensive Action against Anti-­personnel Mines (AICMA) programme. Building on the regional approach to regime formation, one element involved using the third meeting of mine ban treaty states parties in Managua as a target date for the destruction of all APM stockpiles by regime members in the region. This example demonstrates how regime goals can be achieved through fusing the technical and political dimensions of implementation. Using political deadlines to achieve technical regime objectives provided a powerful means of exerting positive conditionality on regime members. Multilateral institutions have been prominent supporters of stockpile destruction. Work through NATO’s Partnership for Peace Programme (PfP), with twenty-­two mine ban treaty states parties, has resulted in the destruction of more than 6.25 million APMs,36 including support to Ukraine in the destruction of its (non-­PFM-1) APM stocks. The Belarus ministry of defence accepted technical assistance from the European Commission to assist in the destruction of 3.37 million PFM-­1 APMs, while the European Commission also awarded a contract worth €3 million for the destruction of Ukraine’s 5.95 million PFM-­type mines. From a security perspective, the broader policy implications of PfP stockpile destruction work were underlined by then NATO Secretary-­General, Lord Robertson, who characterized it as ‘a very practical example of our concrete cooperation on defence reform and our joint efforts to eliminate the legacy of the Cold War’.37 This initiative has also been identified as reducing the risk of trafficking posed by stockpiled weapons in the former Soviet Union.38 Scratching beneath the surface of international support for stockpile destruction in Belarus and Ukraine reveals the crucial relationship between financial resources and national capacities. Despite these cooperation efforts, both countries have missed their destruction deadlines so are in violation of mine ban treaty commitments. The contract agreed between the European Commission and Ukraine in 2005 was cancelled in 2007. In Belarus, attempts to agree assistance projects collapsed in 2006 and 2008. In each case, national management and oversight structures have proven unable to live up to European Commission contracting requirements. A new European Commission tender for stockpile

Stockpile destruction   115 destruction in Belarus was awarded in December 2010. In 2010–2011, Ukraine received $1,000,000 from Norway to upgrade its destruction facilities. Internal bureaucratic procedures were also amended to facilitate cooperation with NATO.39 Less clarity is available on the projected end dates for stockpile destruction in either of these countries. Angola’s stockpile destruction programme is jointly funded by the European Commission (85 per cent), the Government of Angola (10 per cent) and the United Nations Development Programme (UNDP), (5 per cent). This positive example of external assistance in conjunction with national commitment, is particularly conspicuous in the absence of other cases of resources dedicated to stockpile destruction by mine-­affected states. This is an important lacuna given that alongside the less tangible quality of political will, resource commitment provides the clearest indication of regime members taking responsibility for their obligations. While some mine-­affected states do provide information on stockpiles and their destruction, knowledge gaps are particularly acute in the case of ANSAs. Among mine ban treaty member states, few include details in their Article 7 reporting of APMs seized from non-­state groups. Moreover, as the work of Geneva Call shows; while ANSAs may be prepared to commit resources and destroy stocks under their Deed of Commitment, activities are not conducted under the regimes’ verification mechanisms, so issues of transparency and accountability remain. It is important to acknowledge resource imperatives beyond direct financial assistance that affect stockpile destruction. For example, destruction of ordnance found through humanitarian demining is frequently undermined by the difficulty of obtaining significant quantities of high explosive for the task. A day to day implementation issue for stockpile destruction is invisible to the mine ban treaty. This highlights a gap between the regime’s political objective of comprehensive implementation, and its ability to understand contextually defined challenges and mobilize its membership to overcome them in practice. Stockpile destruction thus presents another dimension to the issue of stakeholder clustering. The absence of practitioner expertise in implementation impairs regime effectiveness through a lack of appreciation of challenges that need to be overcome in this area. The performance of APII and the mine ban treaty in terms of resource mobilization for stockpile destruction can be considered a qualified success. Donors remain reluctant to support an apparently technical dimension of a humanitarian issue. While less visible in regime implementation, this perception continues to affect states’ choices. This has costs, because the significant humanitarian pay-­ offs offered by stockpile destruction are not taken into account. The absence of practitioner input to regime implementation rules out potentially significant returns that could be generated on the basis of relatively modest material or financial investments.

116   Stockpile destruction Political will Only through engaging in the process of implementation can states fully understand the implications of regime obligations. While the mine ban treaty has led to substantial stockpile destruction in developed states, the learning curve is particularly steep where states lack the capacity to ensure that international obligations will be implemented at the national level. Failing to enact domestic implementing legislation, or to adopt other necessary measures as a result of the short time period between signature and ratification, are indications that the costs of regime membership have not been fully assessed. Both the cost and the sensitivity of permanently removing landmines from arsenals represent significant obstacles. In particular, treating stockpile destruction as a technical activity fails to take into account that political will is required to implement these programmes. The ability of the regimes to generate the requisite political will, and to adapt to meet challenges that only become apparent during the implementation process is therefore particularly important. Article 3 of the mine ban treaty allows ‘the minimum number absolutely necessary’ of APMs to be retained for ‘the development of and training in mine detection, mine clearance or mine destruction techniques’.40 Some 216,000 APMs are currently held for research and training purposes as permitted under this provision. More than half of states parties retain between 1,000–5,000 APMs, with twenty-­three states retaining less than 1,000 units. However, four states account for nearly one-­third of retained APMs.41 It is also significant that thirteen states parties that retain APMs for training and testing have not reported using any of these stocks since the treaty entered into force for them.42 Bulk holdings and non-­use of training stocks suggest an absence of political will to re-­ evaluate training and testing needs over time to ensure consistency with the absolute minimum requirement stipulated in the treaty. In Sudan, ‘mines retained’ has become a highly political issue that has been linked to the peace process. Following the principle agreed in the 2005 Comprehensive Peace Agreement (CPA), 5,000 landmines kept by the Sudanese armed forces have been matched with the same amount held by the Government of Southern Sudan People’s Army. This is the only example where the domestic implementation mechanism for an international obligation is found in the terms of a peace agreement. It highlights an important confidence-­building dimension to engaging in stockpile destruction. Through developing an innovative destruction programme for its PFM-­1 stocks, Russia overcame a major technical challenge in order to meet regime obligations. Full disclosure of the national landmine arsenal for the first time constitutes an important political step that goes beyond Russia’s overt regime commitments. The US has also been more transparent than required by APII, in announcing the political and operational caveats that apply to landmine stocks not yet destroyed or made regime compliant. The US has said that it reserves the right to use non-­self destruct APMs only in South Korea. Non-­self destruct anti-­ vehicle mines may be used only with presidential approval.43 Beyond these

Stockpile destruction   117 r­ eservations, all stocks have been destroyed or modified to comply with APII. These examples from militarily significant states point to broader regime effects. Demonstrating transparency beyond their obligations shows that states parties are not only constrained by the letter of their commitments, but also feel the need to be seen to be complying with the intent and purpose of the protocol. The Russian and US examples reflect the overlapping and interlocking relationship between technical requirements and the broader political context. This provides evidence of a positive emulation effect with non-­members observing mine ban treaty provisions in the area of stockpile destruction. In 2005, South Korea disclosed its stockpile of 407,800 APMs, significantly less than had previously been indicated by officials. Israel also reported for the first time on the destruction of its outdated stockpiled landmines.44 Additional transparency contributes to confidence building in even those militarily significant states that would appear to place the highest premium on these weapons. The normative push that creates this commitment to the humanitarian principles underpinning the regimes, also generates a strong – though subliminal – interplay between the two landmine regimes. As international legal instruments, the two landmine regimes bind states as opposed to non-­state actors. This calls into question their relevance given the number of internal armed conflicts in which stockpiles are held by ANSAs. While states are bound to address conduct by any actors that contravene treaty obligations as part of their national implementation responsibilities, many are unable or unwilling to do so. This can undermine effectiveness when ANSAs controlling territory hold APM stocks. Research does show that these actors may be willing to destroy stockpiles either as a confidence-­building measure or as a form of reparation for communities in areas under their control.45 The Polisario Front, in the Western Sahara, following signature of its Deed of Commitment, destroyed over 3,000 landmines according to international standards. This was witnessed by the United Nations Mission for the Referendum in Western Sahara (MINURSO) and UNMAS, as well as representatives of foreign governments, the media and various NGOs. In other cases, such as Iraqi Kurdistan and Somaliland,46 land has been cleared but destruction left to international NGOs operating in the area. The political dynamics of stockpile destruction are thus also applicable to ANSAs and provide an entry point for the regimes to influence their conduct. Stockpile destruction by the Polisario Front offers clear benefits for individuals and communities. It also highlights a mutually reinforcing dynamic with the mine ban treaty, civil society and international organizations (the UN) jointly supporting national implementation. However, this and other examples, also show certain limitations of the regime. The political capital gained through being seen to comply with international good practice meets a non-­state armed group’s fundamental need for recognition and legitimacy. This runs counter to the national interests of the state against which these groups are in conflict, explaining strong criticism by some mine ban treaty members directed towards the work of Geneva Call. This criticism demonstrates that the regime does not offer a

118   Stockpile destruction mechanism to renegotiate the cost–benefit calculus of political interests in favour of overarching treaty goals. In sum, complex regime dynamics result from the highly political nature of stockpile destruction. While in some areas (mines retained for training) members are not motivated to revisit earlier decisions, other aspects of stockpile destruction implementation demonstrate a strong willingness to go beyond formal commitments. Increased transparency by APII states parties, such as Russia and the US, contributes to confidence building and generates emulation effects seen in the behaviour of other states. Stockpile destruction also demonstrates an important quality of regime interplay. In particular, for many non-­members of the mine ban treaty, demonstrating that compliance obligations have been exceeded, albeit in a different regime, provides an important opportunity to be associated with the moral clarity of the anti-­APM norm. From this perspective, increased political will to support APII objectives is a helpful by-­product of staying on the right side of the landmine issue by association with the Ottawa Process. Humanitarian impact Membership of both regimes has resulted in the destruction of millions of stockpiled APMs. To assess the value of this work from a humanitarian perspective, it must be set against those landmines held and, in some cases, still used. Humanitarian impact should be understood in a broad sense. This means taking into account the extent to which humanitarian objectives are realized through linking stockpile destruction to wider security and development goals. Different interpretations of the Article 1 obligation not to ‘assist’ anyone to engage in a prohibited activity have been apparent. Thirty-­one states parties have explicitly prohibited foreign stockpiling of APMs on the territory of a state party by a non-­regime member. Tajikistan is the only case where a mine ban treaty member has reported holding the APMs of a non-­state party (Russia) on its territory. Norway, which had jurisdiction and control of US APMs stored on its territory on behalf of NATO, ruled that the landmines would have to be removed by 1 January 2003, the deadline for completion of its stockpile destruction. Maslen argues that to be in accordance with the provisions of Article 4, the APMs should have been destroyed and that their transfer was not permitted under the treaty, since it was not for the purpose of destruction.47 He also suggests that a states party’s military forces that win control of APM stockpiles, while on peacekeeping or other missions, would be obliged to destroy them, citing the example of French peacekeeping troops in Afghanistan destroying 70,000 APMs stored near Kabul Airport in February 2002.48 Germany, Japan, Qatar and the UK have taken the position that US APM stocks in their territories are not under their jurisdiction and control, and thus not part of their regime obligations. Landmine Monitor scrutinized the US stockpiling of military hardware (including APMs) on container ships off the British Indian Ocean Territory of Diego Garcia to which, along with other British Overseas Territories, the UK Landmines Act was extended in 2002. The UK position was that:

Stockpile destruction   119 We wish to affirm that US stocks do not fall under our national jurisdiction or control and we do not therefore have any obligations under Article 4 . . . in respect of them. We have fully complied with our obligations in respect of stocks that were under our jurisdiction or control.49 The argument that the US ships have state immunity and are therefore not a UK responsibility was later ‘clarified’, with the British government accepting that if APMs were unloaded this would constitute a breach of treaty obligations.50 The UK restatement of its position was in response to questions posed during mine ban treaty meetings followed up by Parliamentary Questions in the House of Commons, initiated by the ICBL. In this case, the two-­level dimension of IHL regime obligations is well illustrated. Concerns over regime compliance expressed in an international forum create effects through their re-­articulation at national level. The implementation process draws out sensitive issues such as foreign stockpiling, as well as ambiguities left unresolved during the regime design phase. The meaning of ‘assistance’ under the mine ban treaty has been addressed by states parties in different ways. National security concerns have influenced contrasting interpretations of obligations. Where there have been suggestions that conduct has not been in the humanitarian spirit of the regime, as with US stocks held off Diego Garcia, targeted advocacy at international and national levels has resulted in the UK conceding its ground. However, an inherent ambiguity remains in the latitude between international obligations and national implementation. Stockpile destruction is an obligation of regime membership. From a mine action perspective, this activity may be motivated by a range of security and development concerns. The developmental rationale for stockpile destruction was clarified by the Development Assistance Committee of the Organization for Economic Cooperation and Development (OECD DAC) through stating that donor support for stockpile destruction constitutes Official Development Assistance (ODA).51 This reinforces the regimes’ stockpile destruction obligations by nesting them within the broader objectives of development donors. It also reinforces the message that stockpile destruction is not just a technical activity, but makes a contribution to development (and thus helps donors achieve their development aid targets).52 The United Nations recognises multiple logics for stockpile destruction. The organization supports stockpile destruction through its mine action activities, as a by-­product of DDR programmes as well as through resource mobilization activities. UNDP has declared that stockpile destruction should form a part of any integrated mine action programme that it supports.53 In many post-­conflict environments, abandoned stockpiles represent a danger to civilians and security personnel, as well as those charged with destroying them. This is apparent in Iraq and Afghanistan, where the roadside bombs that cause many casualties among soldiers as well as civilians have often been pilfered from unguarded stockpiles. The concept of a broader security framework for stockpile destruction has been elaborated by the European Commission,

120   Stockpile destruction which characterizes this activity as an important element of its support for a human security agenda.54 This demonstrates a positive dynamic of cross-­ fertilization across regime implementation and the broader agendas of bi- and multilateral actors. The different rationales for support to stockpile destruction put forward by the UN and the European Commission, acknowledge the relevance of this activity in addressing both hard and soft security challenges. It is therefore no surprise that where significant support for stockpile destruction has been forthcoming through the regimes, this has been linked to broader policy frameworks and political interests. While stockpile destruction may not have the visibility of other pillars of mine action, it reflects a range of motivations: the OECD DAC supports stockpile destruction for developmental purposes; NATO, from a security angle and the European Commission, as well as the UN, for a combination of the two. While such support could occur without reference to the regimes, linking support for stockpile destruction to mine ban treaty membership has created a powerful conditionality. Destruction of stocks was facilitated by a combination of technical/ financial support and political obligations linked to membership. The European Commission’s (to date) unsuccessful stockpile destruction assistance was in parallel to, and conditional on, Ukraine and Belarus’ treaty ratification. Linking stockpile destruction assistance to their regime membership has resulted in an obligation to destroy millions of APMs that may otherwise have posed a humanitarian or security risk. From the entry point of stockpile destruction, it has also bound the two countries to all the other applicable regime obligations. Through association with the underpinning security and development rationales for stockpile destruction, the regimes raise awareness of, and support for, this activity. However, this is a passive effect. Active engagement would provide further opportunities to exploit this synergy and thus enhance their humanitarian impact. Normative considerations have been influential in encouraging behaviour change as well addressing flaws in regime design. The dual focus of the ICBL at international and domestic levels that successfully exerted pressure on the UK to clarify its position on foreign stockpiles, provides an important example of the organization’s influencing role. Stockpile destruction also demonstrates how different international regimes and organizations interact through the association of the landmines issue with wider security and development imperatives. Regime effectiveness in the area of stockpile destruction is reinforced because the nexus is recognized between humanitarian obligations, international assistance and associated political conditionalities for membership.

Conclusions In terms of numbers, stockpile destruction resulting from APII and the mine ban treaty represents an impressive achievement. However, its ostensibly technical (rather than overtly political or humanitarian) nature, coupled with these visible successes, contributes to a lack of emphasis on stockpile destruction within the regimes. Significantly, an emphasis on states meeting their destruction deadlines

Stockpile destruction   121 within the mine ban treaty masks a critical distinction between the high numbers of APMs destroyed by the majority, and a more modest achievement in eradicating stockpiles and caches in mine-­affected states. The relationship between international and domestic commitments is critical to the fulfilment of stockpile destruction objectives. In many cases, successful stockpile destruction depends on the support of international and sub-­state actors. The work of Geneva Call in engaging ANSAs in all aspects of the landmine ban is instructive. Despite informal compliance monitoring highlighting that significant stockpiles are controlled by ANSAs, the regime has not been able to deal with the issue given its sensitivity for some members. As in other areas of mine ban treaty support to mine action, failure to address a hard case represents a major barrier to regime effectiveness. In the area of stockpile destruction, a lack of emphasis on ANSAs may be less the result of knowledge gaps than of a lack of political will to address a sensitive topic. Deficits in political will are compounded by inadequate human and organizational capacity. The combination of rapid signature and ratification without domestic implementation measures or adquate national management and oversight capacities, can be directly linked to defection from stockpile destruction obligations. In the case of the mine ban treaty, emphasis on the integrity of the anti-­APM norm sees states parties ‘succeed’ or ‘fail’ in meeting obligations within a given timeframe. This ignores context-­specific implementation challenges, creating tensions between regime and mine action priorities. It is important that binding obligations are not diluted. However, regime effectiveness suffers if defection risks within mine-­affected states are not taken into account. The cases of Belarus, the DRC, Turkmenistan and Ukraine thus show in different ways how implementation is inhibited less by misunderstandings over regime commitments, and more by a lack of capacity and political will. While additional resources for stockpile destruction have been channelled through the mine ban treaty, underlying deficiencies on the national level remain unacknowledged. Analysis of stakeholder clustering dynamics is particularly relevant to understanding the contributions of the regimes in the area of stockpile destruction. Despite an apparently open framework for implementation, politico-­diplomatic and advocacy communities that dominate the mine ban treaty implementation cluster lack the knowledge to respond to the needs of mine-­affected states. The regime has raised awareness of the risks of involuntary defection. It has been less effective in addressing concrete cases of defection from regime obligations. As conventional stockpiles diminish as a result of commitments under both regimes, the remaining implementation challenges will increasingly be hard cases. Entrepreneurial leadership, at the expense of practitioner expertise in the implementation phase, prevents the learning that would come from tapping into the broader mine action community of practice. Stockpile destruction has not received particularly high levels of donor funding relative to the vastly increased profile of the landmine issue as a result of the Ottawa Process. However, the interrelationship between stockpile destruction and broader security and development concerns has generated both political

122   Stockpile destruction will and resources. Stockpile destruction demonstrates important network effects derived from common goals. In this way, stockpile destruction shares in the normative imperative, encouraging actors to stay on the right side of the landmines issue. Where such pressures cannot be brought to bear (as in the case of Turkmenistan) the regime has very few options to exert influence in support of implementation.

7 The effectiveness of global landmine regimes

Introduction Examining the relationship between design, implementation and effectiveness helps to shed light on the impact of the landmine treaties in supporting mine action objectives. In particular, the transition from regime formation to the qualitatively different process of implementation creates a number of challenges. Findings can be counter-­intuitive. An apparent strength in regime design may actually represent an obstacle when transferred to the implementation phase. The resulting insights provide the basis for a re-­evaluation of the accepted narrative that juxtaposes APII and the mine ban treaty, demonstrating hidden effects and deep linkages. This enables us to better understand the effectiveness of global landmine regimes. Analysing APII and the mine ban treaty as nested regimes properly situates these cases within the long history of efforts under international law to alleviate the suffering caused by inhumane weapons. Considering Hague Declaration 3 and the 1925 Gas Protocol as regimes, has enabled us to pose nuanced questions for the landmine treaties on the basis of more drawn out processes of regime development. In particular, this enhances our understanding of effectiveness through clarifying the impact over time of factors such as mismatches between obligations and objectives, normative considerations and stakeholder clustering in earlier IHL regimes. This concluding chapter considers the relationship between landmine regime design and implementation. It then addresses issues of interplay and nesting. The penultimate section focuses specifically on landmine regime effectiveness, with a particular emphasis on the costs of disjunctions between regime and mine action priorities. Finally, wider research priorities in this field are considered.

The interplay between design and implementation Rules are shown to be particularly influential in shaping implementation. Including safety valves within regime frameworks allows them to function in the face of implementation challenges. APII, with its permissive language, long entry into force periods and opportunities for deadline extensions, offers a number of

124   Effectiveness of global landmine regimes safety valves for states. In contrast, the mine ban treaty presents a rigid set of obligations. Yet this absence of compromise contributes to the effectiveness of the regime exactly because commitments are clear and unequivocal. Members are under no illusion as to the extent of their obligations. In practice, therefore, the regime design for APII that allows for flexibility in meeting obligations, has not been more effective when set against the mine ban treaty imperative to preserve key norms and principles through avoiding reservations or permissive language. The integrity of the mine ban treaty on a normative level, and the absence of a safety valve giving greater space to civil society concerns in APII, clearly play a significant role in these contrasting regime dynamics. For APII, long implementation and optional deferral periods means that there are no surprises for members. Clearly defined rules of engagement for participation, including consensus decision-­making and very limited accountability beyond the level of states, also builds confidence. Although within the APII implementation process there are regular experts meetings, these remain largely at a governmental level. This profile contributes to the stability of the regime. At the same time, it undermines the ability to problem-­solve that more broadly-­ based regime participation would offer. The limitations of regime design only become fully apparent during implementation. How regimes learn is important for their ability to overcome technical and political hurdles to fulfilling regime obligations. Yet even where learning mechanisms exist, without clustering the appropriate expertise, fostering political buy-­in and supporting national capacities to address hard issues, the right lessons cannot be drawn. Measures adopted within the mine ban treaty to provide for greater clarity on national implementation processes have had limited impact because greater transparency has not led to increased accountability. Concrete measures have not been developed to address underlying challenges of political will and capacity. The relationship between technical and political factors The interplay of technical and political considerations is an important consideration for landmine regime effectiveness. If treaties are negotiated at a political level, implementation is strongly influenced by a combination of both political and technical factors. The mine ban treaty regime formation process was highly effective in altering the political calculus for membership. However, if states assume obligations in the international arena, they subsequently have to be implemented at home. The costs associated with these obligations in certain cases pose problems that effective regimes should be able to recognize and address. Implementation of the mine ban treaty’s stockpile destruction provisions demonstrates a contradiction. The relative lack of donor support for this activity stems from its apparently technical (rather than humanitarian) profile. At the same time, an implementation cluster lacking practitioner expertise cannot identify or address technical challenges. The technical–political split means that the regime is blind to potential humanitarian pay-­offs that could be achieved through relatively small investments in stockpile destruction.

Effectiveness of global landmine regimes   125 A common feature to both regimes is the absence of formal verification. This contributes to non-­compliance being a sub rosa issue rather than featuring prominently within either regime. The absence of compliance monitoring and verification, compounds an unwillingness to openly address hard issues – such as corruption – that may pose problems for certain members. For the mine ban treaty, there is an evident unwillingness to undermine its positive image. This points to a key finding. The absence of verification does not mean that the regimes lack empirical data on implementation challenges. As the single, comprehensive information source on states parties’ compliance, Landmine Monitor is a powerful mechanism to facilitate learning. But the regimes neither harness the technical know-­how necessary to recognize challenges, nor bring to bear the requisite political will to address them. The application of technical expertise as well as the leveraging of political will, are critical to the effectiveness of the mine ban treaty moving forward. The unanticipated examples of defection from core mine clearance obligations that prompted the ICBL proposal for a ‘special Article 5 deadline procedure’, constitute a key test. The credibility and effectiveness of the treaty can only be maintained through embracing flexibility and accepting the need for adaptation to address qualitatively very different cases of non-­compliant behaviour. On a political level, meeting stockpile destruction targets is closely linked to regime credibility. However, implementation is considered narrowly in terms of whether mine ban treaty regime members complete destruction within the stipulated timeframe. In practice, this endeavour is highly political. For mineaffected states, in particular where national authorities do not have full control of their territory, meeting these commitments can be impossible (or at least impossible to confirm with certainty). Practical challenges may be compounded by an unwillingness to address the politically sensitive issue of how to engage with ANSAs that use or stockpile APMs. Capacity gaps are thus exacerbated by a double deficit of political will at both the national level and within the regime. The extent to which political and technical considerations bifurcate or reinforce each other is under-­estimated in relation to APII effectiveness. If the scope of the regime and the flexibility of its deferral periods reflect the political limitations of a process defined by the consensus requirement, its focus on new technologies and future use of these weapons is a reflection of the knowledge base of the designers. While there may be political attractions for developing countries to participate in the company of major powers, the regime offers few substantive rewards; restrictions offer an expensive, technically challenging solution that does not address the basic problem of mines already in the ground. Initiatives that seek to widen participation are significant as a means to foster learning. Yet if such measures do not make sense from the perspective of regime goals or members’ interests, they will have limited utility. Without such reflection it is difficult to identify the incentives for developing states to contribute meaningfully to APII.

126   Effectiveness of global landmine regimes Key actors and the significance of stakeholder clustering Implementers often play a central role in designing effective treaties. Given that many challenges only become apparent after rules are agreed, the involvement of expert communities should be at least as important for regime implementation. The absence of mine action expertise or voices from mine-­affected countries within APII, has produced a regime that does not address the concerns of these constituencies. Consequently, mines already in the ground, or the costs for developing countries stemming from technological re-­engineering of weapon stocks, were not factored into the design process. Similarly, the focus of APII provisions on traditional military operations does not reflect experience in contemporary conflicts, in which landmines are often utilized as a weapon of fear against civilian populations. For the mine ban treaty, there is an absence of communication and coordination between politico-­diplomatic and advocacy communities active within the implementation process and the wider mine action practitioner community. This clarifies an important distinction between a true epistemic community, and an implementation process still strongly influenced by norm entrepreneurs. Emphasis on brand loyalty over cooperation in the implementation phase has adverse implications for regime effectiveness. Learning within the framework of the mine ban treaty has been stymied by a lack of mine action expertise and a consequent inability to problem-­solve. These capacity gaps do not impair effectiveness in terms of fulfilling obligations, but they do reduce opportunities for the regime to evolve and adapt. How actors combine and interact is a critical determinant for regime effectiveness. The composition of these clusters at different points represents an important regime dynamic that strongly influences the ways in which cooperation and problem-­solving occur. The accepted back story to the Ottawa Process focuses on the civil society coalition and its relationship to a core group of committed states. A broad cluster of actors including states, civil society experts and mine-­affected states, bring different experiences to bear on the process of regime formation. This combination of policy makers, practitioners, advocates and ‘victims’, proved decisive in winning support for the regime and undermining countervailing (predominantly technical) arguments. However, the successful self-­selection dynamic was more effective in garnering support from the undecided than in providing input to treaty development. In reality, the global coalition has been driven by a tightly constructed leadership structure. Although running counter to the accepted narrative, wide participation was less crucial for the design of the treaty than for increasing political will in favour of membership. This points to entrepreneurship over structural or intellectual leadership as being the dominant force. The distinction between representation and influence becomes particularly significant in the implementation phase. The effectiveness of APII is constrained by the unchanged nature of the implementation cluster. In particular, there is a persistent lack of entreprenneurial leadership able to champion change. The

Effectiveness of global landmine regimes   127 closed nature of the CCW, reflected by a limited public record of negotiations and few post facto analyses, offers scant opportunities for learning. Indeed, innovations within the APII regime – such as the introduction of a sponsorship programme to encourage participation from mine-­affected states – reflect an emulation effect in response to measures adopted by the mine ban treaty, albeit within a very different regime framework. In contrast, the influential constellation of actors has shifted within the mine ban treaty. While regime formation revolved around a diverse group of advocacy-­focused actors, supportive states and expert communities, implementation has seen the regime governed by a narrower cluster of advocates in tandem with the politico–diplomatic community. There is a continued involvement of mine-­affected states but, counter-­intuitively, implementation places less, rather than more, emphasis on mine action expertise. A consensus position has emerged, that there is no need to rely on practitioners now that the mine ban treaty regime has achieved a broad and growing membership. One argument is displayed on the opening page of the HALO Trust website: ‘HALO is not distracted by involvement in campaigns and conferences. We have a simple mission statement – “getting mines out of the ground, now.” ’1 Thus, while leadership of the implementation process is tightly controlled by non-­practitioners, elements of the mine action community are – quite deliberately – detaching themselves from implementation because their interests lie in mine action and not the regime per se. The intersessional work programme provides the major mine ban treaty mechanism to support regime learning. Attendees include officials as well as representatives of international organizations and NGOs. The agenda is shaped by the ISU in collaboration with core group states and the ICBL. Representation from mine-­affected countries is encouraged through the sponsorship programme. However, the nature of the implementation cluster means that meetings are geared towards information sharing and awareness raising, rather than problem-­ solving. The relative absence of mine action input conditions regime effectiveness through constraining the ability of the programme to identify and address challenges. Important shifts take place across mine ban treaty design and implementation phases. If ignored, the advocacy-­driven approach, so critical to the process of regime formation, can be counter-­productive. The ICBL provided targeted advocacy that was central to the creation of a strong coalition in favour of a ban. Landmine Monitor represents a transition mechanism for the organisation from advocacy to monitoring and verification service provider. However, there is a gap between the crucial knowledge generated by this research and demonstrable regime responses. If Landmine Monitor helpfully identifies whether states are meeting their obligations, the problem lies in the extent to which these findings are operationalized. In effect, entrepreneurial and intellectual leadership can be thwarted by the passivity of structural leaders within the regime. For many member states, hard implementation issues remain off limits.

128   Effectiveness of global landmine regimes

IHL and mine action Norms and regime interplay The norm bandwagon phenomenon of the Ottawa Process is a decisive factor in accounting for the broader membership of the mine ban treaty in comparison to APII. Being seen to be a concerned international actor was a key dynamic, underlined from the outset of the Ottawa Process by international support that grew rapidly subsequent to the first Ottawa conference. In contrast, while APII has not exerted a pull factor beyond a core constituency drawn predominantly from the developed world, on a political level membership has proved useful in order to demonstrate solidarity with the anti-­APM norm without signing up to a ban on APMs. Beyond accepting the loss of a military capability, it is significant that part of the calculus for joining the mine ban treaty includes contradicting the US, a step not taken lightly for close allies such as the UK or the many states dependent on US security and development assistance. The behaviour of EU applicant states was strongly influenced by the need to be seen to be on the right side of the issue. Even where there was a cost in terms of the defence-­industrial base, the political opprobrium of non-­membership outweighed any potential downsides to joining the regime. In the case of South Africa, its prominent role in the Ottawa Process embodied an internationalist concern, post-­Apartheid, to be seen to re-­ enter the international community as a morally proactive actor. The policy orientation of the US merits close scrutiny. The desire to be proactive on the landmines issue has led to the US increasing its support for mine action as well as placing greater emphasis on active participation in the APII regime. From a starting point of the US treating the CCW as one among many of its arms control obligations, it has become the leading force in both promoting APII compliance and in developing new instruments such as Protocol V, that address wider categories of post-­conflict hazard. This stance combines a desire to demonstrate good international citizenship with the inherently political tactic of promoting measures that can be ‘managed’ in a forum with clear rules, as opposed to a less biddable Ottawa-­style framework. The concept of stigmatization links concerns over specific weapons to a wider IHL agenda. The de-­legitimization of chemical weapons through the 1925 Gas Protocol influenced the behaviour of non-­regime members and was a necessary precursor to the later CWC. The emergence of a norm against these weapons fostered political will to enable further, more far reaching measures. The historical perspective helps to understand interlocking and overlapping dynamics of landmine regime formation. A regime could not be formed on this single issue in 1980, but was feasible when embedded within the CCW. Additional political space subsequently allowed both the further restrictions in APII and the landmine ban. An important quality of regime interplay is found in the de facto conditionality imposed on a number of levels by the anti-­APM norm. The behaviour of actors

Effectiveness of global landmine regimes   129 has been influenced within and across the two landmine regimes. There is a de facto international ban on licit trade in APMs (Russia and China have ceased APM exports) while production has also greatly decreased. The mine ban treaty thus continues to generate norm bandwagon effects which resonate with a broad swathe of the international community. Bi- and multilateral mine action donors have explicitly linked funding to regime membership. This has led to broader pay-­offs, since a membership conditionality as a pre-­condition for stockpile destruction assistance has obliged concerned states to meet the full range of treaty obligations. Evidence of regime interplay during implementation, provides a persuasive counter-­argument to claims that the mine ban treaty has superseded APII. An array of non-­members – from the US to Russia and India – have increased their commitment to APII or aligned themselves with mine ban treaty priorities in order to share in the normative collateral of this process. This has also led to increased transparency – beyond stated APII obligations – by Russia and China, a significant move for actors that place such emphasis on national security concerns. Recalling Putnam’s two-­level game, certain states remain outside the regime in order to meet the requirements of one domestic constituency (the security establishment). At the same time, they adopt measures in order to demonstrate to different domestic constituencies (including civil society but also other groups within government) as well as to the international community, that they are committed to reducing the humanitarian suffering caused by these weapons. Looking beyond the hyperbole, the US, which has not used APMs in combat since 1991, is actually one of the strongest de facto supporters of the mine ban treaty and its humanitarian objectives. US mine action support shows that membership is not a pre-­condition for fulfilling regime goals. Continued ICBL criticism of the US as ‘Ottawa rejectionist’ is far removed from the actual relationship between regime and mine action effectiveness. Juxtaposing the Ottawa and APII processes in normative terms was a highly successful tactic in advocating for a new international instrument geared towards an unequivocal ban on APMs. Bridges between the regimes, unacknowledged when they were being formed, become more apparent in implementation. Both Russia and the US have gone well beyond their APII obligations in showing transparency on national policy decisions and technical information on stockpiles. Positive emulation effects that build on the need to commit to humanitarian principles demonstrate a strong if subliminal quality of regime interplay. Such conduct by major militarily significant states generates ripple effects among other non-­members also keen to stay on the right side of the issue. From this perspective, the continued advocacy tactic of demonizing the US seems to reflect a counter-­productive hangover from the regime formation phase, rather than a forward-­looking strategy to realize humanitarian treaty goals. Regime nesting The case studies on dum dum bullets and chemical gas highlight important qualities of resilience as a result of the mutually reinforcing relationship between

130   Effectiveness of global landmine regimes h­ istorical regimes and newer IHL treaties. For Hague Declaration 3, the impact on the subsequent use of these rounds provides only a partial picture of its effectiveness. Spillover effects into more recent efforts to prohibit certain weapons demand a more layered understanding of IHL regime effectiveness. Even if specific provisions have become less relevant over time, its association with the customary IHL principles of superfluous injury and unnecessary suffering strengthens the underpinning norm, while encouraging regime-­compliant behaviour. It is important to critically examine the prevalent narrative that the Ottawa Process is unique in terms of civil society’s central role, resulting in a victory for an emerging human security agenda. In reality, the processes that led to the ban on dum dum bullets and the 1925 Gas Protocol saw similarly influential roles played by a wide array of civil society bodies in conjunction with sympathetic states. The media also had a notable impact, pushing states to the negotiating table in both these early IHL regimes. The myth of ‘immaculate conception’ surrounding the mine ban treaty is not harmful in itself, but does have consequences if it reinforces artificial divisions between stakeholders and thus works against synergies in the implementation phase. The nexus between de-­legitimization and political will is particularly significant in the context of a regime nested within the broader framework of IHL. Stigmatization provokes effects that extend far beyond declaratory commitments. Both increased US mine action support and the non-­use of APMs since the entry into force of the treaty, stem from the humanitarian imperative underpinning the mine ban treaty. This dynamic displays strong parallels to the US relationship to the 1925 Gas Protocol. While the US remained outside the regime for many years, the stigma associated with chemical gas meant that the use of these weapons was never an option after its entry into force. Actions that seem to run counter to membership choices reflect this quality of normative interplay. If non-­members have fulfilled some regime obligations, accusations of non-­compliant behaviour levelled at Pakistan in the POF case are important for another reason. The rapid response of the Pakistani government in seeking to refute these allegations demonstrates that a breach of the UK’s national obligations was regarded as politically sensitive. An important interplay effect can be found in the high political costs of discovery for a breach of mine ban treaty obligations, regardless of whether or not the state in question is part of the regime.

Regime effectiveness Critically assessing IHL and mine action discourses within a regime theory framework connects related but previously disconnected discourses. This approach has proved useful in addressing disjunctions between treaty and mine action priorities. The following section considers new insights gained through analysing APII and the mine ban treaty from the perspective of resources, political will and humanitarian impact.

Effectiveness of global landmine regimes   131 Resources The mine ban treaty places the onus on states parties to fulfil obligations, yet in nearly all cases mine action is strongly assisted by the expertise and financial support of international actors. The convening role of the regime is therefore essential in order to bridge these resource and capacity gaps. The mine ban treaty is effective in generating additional resources for mine action. Counter-­factual questions are addressed by membership conditionalities placed on mine action funding, evidence of interaction effects through support from non-­traditional donors as well as the timing of new commitments by different donors that coincide with the emergence of the regime. Important emulation effects are also shown, with non-­members increasing mine action funding to demonstrate political commitment as a response to criticism. The funding issue does highlight an important distinction between the two regimes. While the mine ban treaty has generated resources through effective marketing of a humanitarian brand, APII has neither the entrepreneurial leadership nor the moral clarity to provoke such effects. The mine ban treaty channels expertise to a number of areas, such as the development of national mine action strategies, or assistance in designing domestic implementing legislation. Yet in other areas, the human resource needs of mine-­affected states are not aligned with the implementation process. There is a contradiction between the regime’s emphasis on national ownership and its lack of attention to capacity building as a necessary enabling activity. As shown by the continuing difficulties surrounding stockpile destruction in Belarus and Ukraine, human and financial resources need to go hand in hand. This constitutes a significant gap between regime and mine action priorities. The Cambodia case highlights specific challenges associated with the use of resources for mine action. Cambodia has not failed to meet any of its explicit mine ban treaty obligations. Yet Underdal’s distinction between effectiveness in fulfilling obligations as opposed to achieving regime goals is clear: if national authorities undermine mine action work through misappropriating or misallocating funds, then this clearly runs against the overarching humanitarian objectives of the treaty. However, real or alleged corruption is a nettle that the regime has proved reluctant to grasp. If additional financial resources are a demonstrable achievement, much less influence has been exerted on how those resources are targeted and used. The gap between resource generation and effective deployment is particularly evident in the area of humanitarian demining. Despite strong awareness of this challenge within the mine action community, evidenced through country-­based research from Landmine Monitor and efforts to focus on the question led by Norway, the mine ban treaty has not made progress on resource sustainability issues. Perception and knowledge gaps between regime implementers and expert communities inhibit the shift from counting resources as an output, to measuring their humanitarian impact as an outcome. Thus, while the mine action community is undergoing its own ‘gestalt shift’ from quantitative to humanitarian or

132   Effectiveness of global landmine regimes developmental criteria, the regime implementation cluster is unable to bridge the gap between adequate and effective resources. Political will While the two-­track regime formation process proved extremely effective in building political will in support of a ban, in other ways political will was lacking during this phase. In mine ban treaty negotiations the core group displayed a reluctance to include issues – such as assistance for mine victims – that might complicate the task of achieving a widely-­adopted treaty centred around a ban. The strategy was therefore to address hard issues that might be found in the details by avoiding them. Beyond the US, unable to countenance an unadulterated ban but still wanting to drive the process, red line issues were never brought to the fore during negotiations. By contrast, in the APII framework, the consensus requirement for decision-­making meant that the political and technical red lines were clear. With the ‘nuclear option’ of banning weapons off the table, negotiations inevitably focused on technical restrictions. The norm bandwagon surrounding the mine ban treaty has stored up hidden costs for implementation. If the anti-­APM norm created widespread support for the mine ban treaty, it does not follow that this led to commensurate political will to meet obligations. Early signature and ratification without consideration for the implications of regime obligations, makes the implementation of humanitarian demining and stockpile destruction provisions less likely for mine-­affected states. Inevitably, the price of membership is higher for these states given that resource and capacity deficits are more acute in the settings that need support from the regimes. A major challenge to effectiveness can be discerned when states lack the relevant capacities or lose enthusiasm in the more technically challenging, expensive and politically sensitive (in domestic terms) implementation phase. This points to an apparent paradox, with the anti-­APM norm encouraging adherence, yet at the same time blinding states to the real costs of membership. One hard implementation issue that has not been addressed is inappropriate donor support. Technology for mine action is an area where there is a nexus between a clear mandate within the mine ban treaty (Article 6) and evidence from mine action practitioners that what is provided is frequently expensive, inappropriate and ineffective. Despite this issue forming part of treaty obligations, it has gained little traction. Again, a serious challenge to regime effectiveness is posed by the lack of practitioner expertise in the implementation cluster. The regime has chosen not to address this sensitive issue, and neither expert communities nor the affected states that receive such support have the leverage to challenge this unfortunate practice. The absence of ownership and a related lack of attention to capacity building are a major shortcoming of existing mine action programmes. This problem has not been directly addressed within either regime, begging the question whether its significance is recognized or ignored. Consideration of humanitarian demining

Effectiveness of global landmine regimes   133 and stockpile destruction shows that mine-­affected states have a less well defined and influential role in mine ban treaty implementation than in the regime formation phase. Given that in many mine-­affected contexts the state itself cannot be considered as a unitary actor, a more nuanced understanding is required of voluntary and involuntary defection. Both regimes lack the technical/contextual awareness to support mine-­affected states in ways that reinforce political will and mitigate non-­compliance. While less relevant in APII, with its narrower membership base, within the mine ban treaty this reality demonstrates an important distinction between the principle of ownership that is central to the Ottawa discourse and how this is translated into practice. Humanitarian impact Support to mine action represents the most tangible way to deliver on the humanitarian objectives of the landmine treaties. As IHL regimes, the effectiveness of APII and the mine ban treaty ultimately rests on this ability to reduce the humanitarian costs of landmines. The evident decline in the use of APMs as a result of the mine ban treaty clearly constitutes a major success in terms of this humanitarian criterion. However, an approach to implementation resting on statistics such as regime membership or stockpiles destroyed, does not address the qualitative concerns of affected states or mine action practitioners. Relatedly, the lack of focus on capacity building contributes directly to involuntary defection and thus to reduced effectiveness. While implementation of stockpile destruction provisions is understood within the regimes as a technical endeavour, in fact this is a highly political process. Failure to recognize this interplay between technical and political concerns has costs when political imperatives that led states to join the mine ban treaty, are not followed by a commensurate focus on the requirements assumed by taking such a step. The political and technical challenges that result in defection from clearance or stockpile destruction targets present dilemmas that have not been fully recognized, let alone addressed, within the framework of the regime. Applying a humanitarian criterion for effectiveness allows us to critically question claims of effectiveness relating to certain APII provisions. In particular, this perspective provides a necessary corrective to arguments deployed on the humanitarian impact gained through increased detectability of landmines. In fact, compliance with this provision offers no humanitarian pay-­off for civilians since at risk groups do not have access to mine detection equipment. More broadly, all provisions that apply to current and future stocks of landmines ignore the reality that to a great extent, the humanitarian threat relates to mines already in the ground long before provisions were agreed. The additional protection for military personnel and deminers does constitute a humanitarian benefit. But it should be noted that compliance with these provisions by China and Russia unites self-­interest with a humanitarian demining pay-­off, since motivations stemmed at least partially from an awareness of the risks to their own forces posed by undetectable APMs.

134   Effectiveness of global landmine regimes Considering effectiveness in its humanitarian dimension demonstrates a consequential tension between regime and mine action priorities. A commitment to 100 per cent clearance as a regime benchmark is inconsistent with the mine action imperative to only conduct work that has a clear impact. The regime seeks to preserve the anti-­APM norm at the expense of more effective humanitarian demining. For the mine ban treaty, nurturing the normative imperative in implementation is key to promoting compliance, while at the same time generating emulation effects beyond the regime. Lowering the bar from ‘mine free’ may contribute to defection through states misunderstanding or deliberately misinterpreting their obligations. Although the UK, with respect to the Falkland Islands, might wish it otherwise, conducting certain ‘ineffective’ humanitarian demining activities is the necessary price to pay for maintaining the integrity of the regime and thus ensuring its overall effectiveness.

Learning from the landmine regimes This book has placed significant emphasis on seeking to understand the relationship between current processes of IHL development and their historical antecedents. It has framed this analysis in ways that draw together insights from IHL, policy and practitioner communities. In looking ahead, this concluding section focuses on two related considerations: gaps in research that could be usefully addressed; and insights that can be drawn from spillover effects from the landmine regimes to related IHL issue areas. An IHL regime research agenda Important gaps can be found within both regimes between ideal provisions and field realities. Even when met, certain obligations do not result in security, development or humanitarian pay-­offs. This implementation–effectiveness gap raises important issues for treaty designers and implementers in the wider IHL field. The issue is further complicated by a counter-­intuitive, but significant, contradiction between regime and mine action interests, whereby the most effective approach from a mine action perspective can undermine the integrity of the mine ban treaty. Further work on IHL effectiveness must seek to address complex questions in which design, agency and normative considerations are intermeshed. Stakeholder clustering dynamics across regime formation and implementation demonstrates the need to nurture learning practices. A lack of input from mine action practitioners in implementation has adverse implications for the ability of the regimes to adapt and evolve. On the one hand, despite claims to openness in the case of the mine ban treaty, allocation of influential roles has been based on entitlement rather than need. On the other hand, a lack of engagement in implementation is not simply a reflection of the preferences of controlling interests within the regimes, but is also a consequence of a self-­conscious (and self-­ interested) desire from within to treat mine action as a ‘black art’ set apart from

Effectiveness of global landmine regimes   135 broader policy, programming and budgetary considerations. Clustering and leadership dynamics need to be carefully considered in order to shape these divergent interests in support of common goals. The two landmine treaties are intrinsically connected even if this is not widely understood or acknowledged. Interlinkages between APII and the mine ban treaty, if remaining taboo within the regimes, impact significantly on their effectiveness. Although a mechanism of one regime, Landmine Monitor influences both through its reporting. More broadly, an important quality of regime interplay supports common goals regardless of membership choices. Despite these evident complementarities and latent synergies, little effort has been made to bring the two regimes closer together. Further work on interplay dynamics would be necessary to understand an issue that is neglected, but may offer important dividends. Exploring regime nesting dynamics highlights aspects of the relationship between APII and the mine ban treaty that only emerge when they are embedded within the broader framework of IHL. In particular, critically examining historical antecedents demonstrates the importance of stigmatization for IHL regimes. This finds its echo in the catalyzing effect on both state and non-­state actors of the anti-­APM norm, promoting the observance of humanitarian objectives as ‘the right thing to do’. A number of emulation effects can also be discerned in the broader IHL agenda as a result of the landmine regimes. These consequences and spillover effects point to an important focus for new research. Clarifying the relationship of mine action to related activities necessitates deeper understandings of the interactions between regimes, international organizations, states and non-­state actors. Drawing on analysis of the landmine treaties can help shed light on determinants and variables that influence these complex interdependencies. Mine ban treaty membership conditionalities link regime goals to the broader security and development agendas of actors such as the EU and the UN. The ability to create virtuous cycles through a combination of obligations and incentives associated with membership, seem to offer promising opportunities to better situate mine action within poverty reduction, development and peacebuilding frameworks. Spillover effects The momentum created by norm spillover effects is integral to the mine ban treaty model. New vigour to advocacy efforts as a result of the Ottawa Process can be discerned in other areas of the human security agenda. Indeed, almost before the ink was dry on the treaty, commentators and participants in the Ottawa Process sought to draw parallels and identify lessons for efforts to control small arms and light weapons (SALW), to address the problem of child soldiers and to establish an International Criminal Court (ICC).2 While analysing the efforts of the international community in these areas is beyond the scope of this book, it is possible to make some tentative observations from a regime perspective on how the structure, momentum and agency dynamics of the Ottawa Process relates to these issues.

136   Effectiveness of global landmine regimes If moral opprobrium associated with APMs was critical to the mine ban treaty regime formation process, the right side of the issue has been less clear in attempts to mobilize states and civil society behind efforts to address SALW proliferation. Their widespread availability, the combination of legal and illegal use, as well as the self-­evident military utility of these weapons work against the black and white policy choice presented in the Ottawa Process. Echoing the 1925 Gas Protocol, there is also a strong civil society lobby in the US working against gun control. In the absence of the categorical stigmatization of these weapons, SALW control efforts have not made the transition from a security perspective – focusing on illicit trade – to a humanitarian imperative demanding urgent action. Realistically, the normative push against landmines was greatly facilitated by their questionable military utility. This cannot be said of SALW, where both humanitarian costs and effectiveness are undisputable. While the complexity of the SALW issue has been reflected in a lack of strategic direction across governments, international organizations and civil society actors, momentum has built around the need for a global normative framework for arms transfers. A process that started in 2006 with a request by the United Nations General Assembly for states to provide comments on the issue, led to a 2009 decision to negotiate such a treaty.3 The signature and subsequent entry into force of the ICC Statute was a major step forward in promoting accountability for the most serious war crimes, crimes against humanity and genocide.4 The negotiations surrounding the creation of the ICC share certain features with the mine ban treaty. These include an influential, broad-­based group of like-­minded states, strong chairmanship of the negotiating conference, expert NGOs invited as observers and a flexible negotiating framework premised on two-­thirds majority voting. In contrast, the regime lacks the simple humanitarian message that underpinned the Ottawa Process. The issues raised by the ICC are inherently complex and are consequently a ‘hard sell’. It is therefore more difficult to generate pressure on hardline rejectionists to join the regime. The ICC regime formation process tapped into an effective coalition of like-­minded states twinned with ‘expert’ civil society contributions. This specialized cluster lacked the salesmanship of the advocacy community, although broad-­based popular support has developed over time. But the combination of inherent complexity and a lack of savvy marketing negated any possibility of a norm cascade comparable to the Ottawa Process. Moreover, moving the right to try the most heinous crimes from a national to an international level touches directly on national sovereignty concerns, so immediately generates political resistance (as reflected by continued US rejection of the ICC). The consensus-­based approach to the Optional Protocol to the UN Convention on the Rights of the Child5 shares common features with the structured CCW approach. While the humanitarian imperative is more apparent than in the case of the ICC, building consensus on child soldiers has been made difficult by the nuanced distinctions that need to be made between different aspects of the issue. To give one example, under-­eighteen recruitment in professional western armed forces is fundamentally different from the fate of minors press-­ganged

Effectiveness of global landmine regimes   137 into armed groups in conflict zones. This complex picture has been translated into the compromise language agreed in the Optional Protocol. Like APII, it has been criticized for not going far enough, but the regime does contain the flexibility necessary to reflect the needs and interests of different constituencies. Where there is a much finer balance between military utility and humanitarian cost, it is less easy to deploy what Price calls the ‘transnational Socratic method’6 of placing the burden on states to justify their behaviour. As a consequence, like-­ minded coalitions of actors have struggled to focus their efforts and build synergies across different interest groups on the issue of child soldiers.

Conclusion The CCW represented an important step forward in addressing the need for better protection of civilians in conflict. Protocol II responded to concerns that had become more pronounced with the development of scatterable weapon systems which allowed munitions to be deployed quickly over a wide area.7 More recently, the humanitarian impact of categories of explosive remnants of war (ERW) beyond landmines have also gained prominence. Collateral damage became the leitmotif of NATO’s 1999 air campaign in Kosovo. Graphic images of numerous civilian casualties caused by cluster bomb sub-­munitions, were burned into the public consciousness thanks to on the ground reporting from the international news media. And like APMs, these weapons continue to kill and maim long after conflict ends. The ICRC was again prominent, along with various NGOs, in calling for new IHL to deal with this issue. Following three years of expert meetings, a third CCW Review Conference in November 2003 led to the agreement of Protocol V on ERW,8 designed to minimize their risks and effects in the post-­conflict period. As with earlier protocols, it has been criticized for its non-­binding nature and conditional language. Echoing Johan Molander’s comments in 1996, review conference chairman Chris Sanders said that the text was the best that could be achieved at the time.9 History is now repeating itself with this CCW process overshadowed by a new freestanding treaty banning the use, stockpiling, production and transfer of cluster munitions. The ‘Oslo Process’ was launched in February 2007 and the Convention on Cluster Munitions (CCM) adopted in Dublin on 30 May 2008, followed by a signing ceremony held in Oslo on 3–4 December that year.10 The two-­track dynamic of Protocol V and the CCM offer evident parallels to the relationship between APII and the mine ban treaty. Both models have their place as approaches to advancing the IHL agenda. While the Ottawa and CCM ‘big bang’ approach steals much of the limelight, the common framework within which different weapons are addressed is a design factor intrinsic to the CCW that has enabled incremental, but nonetheless genuine, progress over time. In particular, the extension of the scope of APII to non-­international armed conflicts has generated important emulation effects. It is argued that the 1998 Statute of the ICC, the 1999 Second Protocol to the Hague Cultural Property Convention and the practice of the International Criminal Tribunal at The Hague, all

138   Effectiveness of global landmine regimes extended their applicability to internal conflicts on the basis of the APII precedent.11 APII therefore represents a significant acknowledgement of the changing nature of warfare that has catalyzed other treaty regimes to embrace the realities of contemporary conflict. This is a field that is in constant evolution. Regime theory offers only one approach to analysing the design, implementation and effectiveness of IHL. Pressing humanitarian, security and developmental challenges provide a compelling argument for new thinking on both the landmine regimes and IHL more broadly. It is to be hoped that shining a critical light on the landmine regimes can help us tread carefully in the future.

Notes

1  The design, implementation and effectiveness of global landmine regimes   1 The mine ban treaty is explicit about this humanitarian goal, made clear in the first preambular paragraph to the treaty which declares its determination to stop the killing and maiming of civilians, especially children, by APMs. Article 3(7) of APII prohibits the use of any weapons covered by the protocol against civilians or civilian objects.   2 Full title: Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-­Personnel Mines and on their Destruction.   3 Full title: Amended Protocol II to the Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons Which May be Deemed to be Excessively Injurious or to have Indiscriminate Effects.   4 A. Roberts and R. Guelff (eds) Documents on the Laws of War, Oxford: Clarendon Press, 2nd edition 1989, p. 29.   5 See: W.A. Self and J.A. Roach, Index of International Humanitarian Law, Geneva: ICRC, 1992; J.L. Brierly, The Law of Nations, Oxford: Oxford University Press, 6th edition 1991.   6 Gen. Sir H. Beach, ‘Qualitative Arms Control and International Humanitarian Law applicable to Armed Conflict’, ISIS Policy Paper on Qualitative Arms Control 1, July 2001.   7 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, Humanitarianism and War Project, Occasional Paper 42, 2000, p. 2.   8 GICHD, A Guide to Mine Action, Geneva: GICHD, 2003, pp. 10–11.   9 B. Boutros Ghali, ‘Agenda for Peace: Preventive Diplomacy, Peacemaking and Peacekeeping’, Report of the Secretary-­General pursuant to the Statement adopted by the Summit Meeting of the Security Council on 31 January 1992, A747/277-S/24111. 10 As of 15 November 2011 APII has 97 full members while the mine ban treaty has 158 states parties. 11 M.A. Cameron, R.J. Lawson and B.W. Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, Oxford: Oxford University Press, 1998. See also: R. Mathew, B. McDonald and K.A. Rutherford, Landmines and Human Security, New York: Suny Press, 2004. 12 For these arguments, see: M. Bleicher, ‘The Ottawa Process: Nine Day Wonder or a New Model for Disarmament Negotiations?’, Disarmament Forum 4, 1999; D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’; K. Rutherford, Disarming States: The International Movement to Ban Landmines, Santa Barbara CA: Praeger, 2010. 13 R. Little, ‘International Regimes’ in J. Baylis and S. Smith, (eds) The Globalization of World Politics, Oxford: Oxford University Press, 2001, p. 300.

140   Notes 14 Ernst Haas, writing in 1980, suggests that ‘regimes are norms, rules and procedures agreed in order to regulate an issue area’. Stephen Krasner in 1983, offered the definition of a regime as ‘implicit or explicit principles, norms, rules and decision-­making procedures around which actors’ expectations converge in a given area of international relations’. Another definition was offered in 1989 by Robert Keohane as ‘institutions with explicit rules, agreed upon by governments, which pertain to particular sets of issues in international relations’. The similarities and distinctions between these three early definitions are instructive. Keohane offers a ‘minimalist’ definition of regimes, as formalized constructs agreed and established by states. In this respect, he acknowledges that rules alone do not make a regime; there is also a need for validity and recognition for a regime to exist. However, in responding to the Krasner definition, Keohane is wary of defining regimes in terms of observed behaviour, due to the complexity in interpreting implicit rather than overt regime characteristics such as principles and norms. In introducing the concept of norms to their definitions of a regime, Krasner and Haas point out that participation in international regimes is more complex than a function of rules and procedures. Norms explicitly incorporate multi-­actor involvement in regimes beyond the states that make up their formal membership. In other words, regimes are an expression of interdependence in international relations, and this is as relevant as structural aspects in defining and explaining them. See: E.B. Haas, ‘Why Collaborate? Issue Linkage and International Regimes’, World Politics 32, 1980, p.  358; S.D. Krasner (ed.) International Regimes, Ithaca: Cornell University Press, 1983, p.  2; R.O. Keohane, International Organizations and State Power: Essays in International Relations Theory, Boulder, Colorado: Westview Press, 1989, p. 4; R.O. Keohane ‘The Analysis of International Regimes’, in V. Rittberger (ed.) Regime Theory and International Relations, Oxford: Clarendon Press, 1993, p. 28. 15 M.A. Levy, O.R. Young and M. Zurn, ‘The Study of International Regimes’, European Journal of International Relations 1 (3), 1995, p. 274. 16 R. Little, ‘International Regimes’, in Baylis and Smith, The Globalization of World Politics, p. 299. 17 Relatively few sources apply regime theory to the landmine regimes. In an important contribution, Frank Faulkner utilises regime theory to understand the role and impact of the International Campaign to Ban Landmines. However, Faulkner does not substantitively address other actors within the process, nor does he cover the parallel process within the CCW. It should also be noted that Faulkner approaches the issue from an arms control, rather than an IHL, perspective. In contrast, in a 1999 article for The Nonproliferation Review, Ken Rutherford provides an insightful comparison of the Hague and Ottawa Conventions as IHL ‘regimes’. As acknowledged by the author, the article focuses on the processes leading to regime formation and not on the contents or effectiveness of the regimes. See: F. Faulkner, Moral Entrepreneurs and the Campaign to Ban Landmines, Amsterdam: Rodopi, 2007; K. Rutherford, ‘The Hague and Ottawa Conventions: A Model for Future Weapon Ban Regimes?’, The Nonproliferation Review, Spring/Summer 1999. 18 M.A. Levy, O.R. Young and M. Zurn ‘The Study of International Regimes’, p. 279. 19 R.D. Putnam, ‘Diplomacy and Domestic Politics: The Logic of Two-­level Games’, International Organization 42, 1988, pp. 427–460. 20 A. Cassese, International Law, Oxford: Oxford University Press, 2nd edition 2005, p. 167. A. Hurrell, ‘International Society and Regimes’, in Rittberger, Regime Theory and International Relations, p. 71. 21 A. Underdal, ‘The Concept of Regime Effectiveness’; Cooperation and Conflict 27, 1992, 227–240. 22 M.A. Levy, O.R. Young and M. Zurn ‘The Study of International Regimes’, p. 292. 23 O.R. Young, (ed.) ‘Global Governance. Drawing Insights from the Environmental Experience’, An Occasional Paper from the Dickey Center, Dartmouth College, Hanover, New Hampshire, 1995, p. 19.

Notes   141 24 M.A. Levy, O.R. Young and M. Zurn ‘The Study of International Regimes’, p. 278. 25 O.R. Young, ‘Political Leadership and Regime Formation: On the Development of Institutions in International Society’, International Organization 45 (3), pp. 281–308. 26 M. Finnemore and K. Sikkink, ‘International Norm Dynamics and Political Change’, Paper presented at the Ideas, Culture and Political Analysis Workshop, Princeton University, May 15–16 1998. Available online at: www.ciaonet.org/conf/ssr01/ ssr01ak.html (accessed 30 April 2002), p. 7. 27 J.H. Camilleri and J. Falk, The End of Sovereignty. The Politics of a Shrinking and Fragmenting World, Cheltenham: Edward Elgar, 1992, p. 232. 28 J.G. Ruggie, ‘Multilateralism: The Anatomy of an International Institution’, International Organization 46 (3), Summer 1992, p. 592. 29 R.D. Putnam, ‘Diplomacy and Domestic Politics: The Logic of Two-­level Games’, p. 434. 30 O.R. Young, ‘The Effectiveness of International Institutions: Hard Cases and Critical Variables’, in J.N. Rosenau and E.O. Czempiel (eds) Governance Without Government. Order and Change in World Politics, Cambridge: Cambridge University Press, 1992, p. 192. 31 M. Finnemore and K. Sikkink, ‘International Norm Dynamics and Political Change’, p. 7. 32 O.R. Young, (ed.) ‘Global Governance. Drawing Insights from the Environmental Experience’, pp. 27–28. 33 M.A. Levy, O.R. Young and M. Zurn ‘The Study of International Regimes’, p. 309. 34 Ken Anderson’s assessment of the role of international NGOs in driving the Ottawa Process is a notable exception. Anderson’s analysis, which focuses on the regime formation phase, critiques the conflation of international NGOs with civil society and questions the democratic legitimacy of the process. See: K. Anderson, ‘The Ottawa Convention Banning Landmines, the Role of International Non Governmental Organisations and the Idea of International Civil Society’, European Journal of International Law II (1), 2002. 2  Regimes prohibiting the use in war of poison gas and dum dum bullets    1 A. Roberts and R. Guelff (eds), Documents on the Laws of War, Oxford: Oxford University Press, 3rd edition 2000, p. 63.    2 Roberts and Guelff, Documents on the Laws of War, p. 155.    3 J. Goldblat, Arms Control. A Guide to Negotiations and Agreements, Thousand Oaks, CA: Sage Publications Ltd, 1994, Chapter 7.    4 J. Goldblat, Agreements for Arms Control – A Critical Survey, Stockholm: SIPRI, 1982, p. 89.    5 S.D. Bailey, Prohibitions and Restrictions in War, Oxford: Oxford University Press, 1972, p. 140.    6 G. Best, Humanity in Warfare, London: Weidenfeld and Nicolson, 1980, p. 159.    7 B.W. Tuchman, The Proud Tower. A Portrait of the World Before the War: 1890–1914, New York: Macmillan, 1966, p. 235.    8 W.I. Hull, The Two Hague Conferences and their Contribution to International Law, New York: Garland, 1972, p. 54.    9 W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 52.   10 B.W. Tuchman, The Proud Tower. A Portrait of the World Before the War, p. 230.   11 I.L. Claude, Swords Into Ploughshares. The Problems and Progress of International Organization, New York: Random House, 3rd edition 1964, p. 26.   12 I.L. Claude, Swords Into Ploughshares, p. 27.

142   Notes   13 W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 63.   14 W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 181.   15 A. Roberts and R. Guelff (eds), Documents on the Laws of War, Oxford: Clarendon Press, 2nd edition 1989, p. 39.   16 G. Best, Humanity in Warfare, p. 162.   17 A. Roberts and R. Guelff (eds), Documents on the Laws of War, 2000, p. 64.   18 During the subsequent 1907 Hague Peace Conference both Great Britain and Portugal announced that they would sign Hague Declaration 3.   19 J. Corkson and J. Nottingham, The Control of Chemical and Biological Weapons, New York: Carnegie Endowment for International Peace, 1971, pp. 65–66.   20 J. Keegan, The First World War, London: Pimlico, 1999, p. 215.   21 Articles 170 and 171 of the Versailles Treaty forbade Germany from importing or exporting any arms, munitions or warlike materials, making specific reference to chemical weapons.   22 Protocol for the Prohibition of the Use of Asphyxiating, Poisonous or Other Gases, and of Bacteriological Methods of Warfare. Geneva, 17 June 1925.   23 F.J. Brown, Chemical Warfare; a Study in Restraints, Princeton: Princeton University Press, 1968, p. 97.   24 J.P. Zanders, ‘The Chemical Weapons Convention in the Context of the 1925 Geneva Debates’, The Nonproliferation Review, Spring/Summer 1996, p. 41.   25 League of Nations, ‘Proceedings of the Conference for the Supervision of the International Trade in Arms and Ammunition and in Implements of War’, League of Nations Document A 13, 1925, p. 161.   26 The Protocol was signed on 17 June 1925 and entered into force in 1928.   27 Use of these rounds, albeit not by states parties to the Declaration, continued in the 1899–1902 Boer War and the Russo-­Japanese War of 1904–1905.   28 R. Coupland and D. Loye, ‘The 1899 Hague Declaration Concerning Expanding Bullets. A Treaty Effective for more than 100 Years Faces Complex Contemporary Issues’, International Review of the Red Cross 85 (849), 2003, pp. 137–138.   29 R. Coupland and D. Loye, ‘The 1899 Hague Declaration Concerning Expanding Bullets’, p. 139.   30 R. Coupland and D. Loye, ‘The 1899 Hague Declaration Concerning Expanding Bullets’, p. 135.   31 Hague Declaration 3 has only thirty-­one members of whom twenty-­seven signed before 1910. A full list of States Parties can be found on the ICRC IHL database, www.icrc.org/ihl.   32 J. Goldblat, The 1925 Geneva Protocol: its Origin, Scope and Reservations. Avail­ able online at: www.delegfranc-­cd-geneve.org/declarations/declafrancaises/armes% 20bio/Jozef%20%Goldblat.pdf (accessed 15 February 2011).   33 S.D. Bailey, Prohibitions and Restrictions in War, p. 127.   34 A. Roberts and R. Guelff (eds), Documents on the Laws of War, 2000, p. 156.   35 United Nations General Assembly Resolution 2162 B xxl.   36 A. Roberts and R. Guelff (eds), Documents on the Laws of War, 2000, p. 157.   37 The Declaration only applies to military operations. Handguns using bullets that expand or flatten on impact are used legally by police forces around the world on the basis that (1) used at short range, most likely in self defence, the chance of incapacitating an attacker is increased; and (2) the risk of the bullet passing through an attacker and injuring others is minimized. Coupland & Loye, ‘The 1899 Hague Declaration Concerning Expanding Bullets’, p. 141.   38 A former serving officer in the South African Defence Force (SADF ), speaking anonymously, confirmed that during counter-­insurgency activities in Namibia this practice was common in preparing for military operations. Instructions on the

Notes   143   39   40   41   42   43   44   45   46   47

  48

  49   50

  51   52   53   54   55   56   57   58   59   60   61   62   63   64   65   66

p­ rocedure to improvise dum dum bullets can also be found easily on the internet. See: www.totse.com/en/bad_ideas/guns_and_weapons/162568.html. R.J. Mathews and T.L.H. McCormack, ‘The Influence of Humanitarian Principles in the Negotiation of Arms Control Treaties’, International Review of the Red Cross 834, June 1999, p. 336. A. Roberts and R. Guelff (eds) Documents on the Laws of War, 2000, p. 156. United Nations, Chemical and Bacteriological Weapons, Report of the Secretary-­ General, UN Doc. A/39/488, Annex II (1984). A. Roberts and R. Guelff (eds) Documents on the Laws of War, 2000, p. 157. As of 15 November 2011 there are 137 states parties. A. Roberts and R. Guelff (eds) Documents on the Laws of War, 2000, p. 157. 149 states were represented at the conference. A. Roberts and R. Guelff (eds) Documents on the Laws of War, 2000, p. 157. A complete prohibition on biological weapons was provided for in the Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction, signed on 10 April 1972 and entering into force in 1975. See: ‘Report of the ICRC for the Review Conference of the 1980 UN Conventions on Prohibitions or Restrictions on the Use of Certain Conventional Weapons which may be Deemed to be Excessively Injurious or to have Indiscriminate Effects’, International Review of the Red Cross 299, April 1994, pp. 123–182. See: E. Prokosch, ‘The Swiss Draft Protocol on Small-­Calibre Weapon Systems’, International Review of the Red Cross 307, August 1995, 411–425. For an extensive discussion of this issue see L. Moosberg, ‘Does the Swedish Use of the 12.7 mm Multipurpose Projectile Undermine the St. Petersburg Declaration?’, Masters Thesis (Spring 2003). Available online at: www.teol.uu.se/noha/masters/ moosberg.pdf (accessed 10 September 2004). See: R. Coupland and P. Herby, ‘The SIrUS Project: Progress Report on Superfluous Injury or Unnecessary Suffering’, International Review of the Red Cross 835, 1999, 583–592. D. Fleck (ed.) The Handbook of Humanitarian Law in Armed Conflicts, Oxford: Oxford University Press, 1995, p. 149. G. Best, Humanity in Warfare, p. 140. G. Best, Humanity in Warfare, p. 164. W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 52. B.W. Tuchman, The Proud Tower. A Portrait of the World Before the War, p. 241. W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 58. B.W. Tuchman, The Proud Tower. A Portrait of the World Before the War, p. 239. PRO, FO 30/40. 3/36 (2 July 1899), as quoted in G. Best, Humanity in Warfare, p.139 W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 183. W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 184. W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 185. W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 68. W.I. Hull, The Two Hague Conferences and their Contribution to International Law, p. 68. B.W. Tuchman, The Proud Tower. A Portrait of the World Before the War, p. 234. J.B. Scott, Proceedings of the Hague Peace Conference of 1899, New York: Garland, 1920, p. 9.

144   Notes   67 G. Best, Humanity in Warfare, p. 133.   68 A.P. Zimmern, The League of Nations and the Rule of Law 1918–1935, New York: Macmillan, 1936, p. 102.   69 B.W. Tuchman, The Proud Tower. A Portrait of the World Before the War, p. 248.   70 B.W. Tuchman, The Proud Tower. A Portrait of the World Before the War, p. 258.   71 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 100.   72 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 101.   73 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 106.   74 U.S. Congressional Record, 69th Congress, 2nd Session., LXVIII Part I, p. 149).   75 Ibid., p. 157.   76 F.J. Brown, Chemical Warfare; a Study in Restraints, pp. 108–109.   77 J.W. Hammond, Poison Gas: Myths Versus Reality, Westport, CT: Greenwood Press, 1999, p. 21.   78 J.P. Zanders, ‘The Chemical Weapons Convention in the Context of the 1925 Geneva Debates’, p. 41.   79 J.P. Zanders, ‘The Chemical Weapons Convention in the Context of the 1925 Geneva Debates’, p. 41.   80 J.P. Zanders, ‘The Chemical Weapons Convention in the Context of the 1925 Geneva Debates’, p. 41.   81 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 109.   82 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 178.   83 G. Woker, ‘Chemical and Biological Weapons’, in Inter-­Parliamentary Union, What Would be the Character of a New War?, London: Victor Gollancz, 1933, p. 359.   84 Proceedings, 12th International Red Cross Conference, Geneva, 1925.   85 W. Tcheou, Proceedings of the Conference for the Supervision of the International Trade in Arms and Ammunition and in Implements of War, Doc A 13, Geneva, 1925, p. 313.   86 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 294.   87 R.J. Mathews and T.L.H. McCormack, ‘The Influence of Humanitarian Principles in the Negotiation of Arms Control Treaties’, p. 5.   88 F.J. Brown, Chemical Warfare; a Study in Restraints, pp. 133–134.   89 Named after Corson and Stoughton, the scientists who first developed the gas.   90 This resolution (Resolution 2603A (xxiv) 16 December 1969) was adopted by 80 votes to 3 (Australia, Portugal, USA) with 26 abstentions, including Great Britain.   91 S.D. Bailey, Prohibitions and Restrictions in War, p. 134.   92 New Statesman ‘The CS Gas Muddle’ (31 July 1970): p. 109.   93 I. Detter, The Law of War, Cambridge: Cambridge University Press, 2000, p. 256.   94 J.W. Hammond, Poison Gas: Myths Versus Reality, p. 22.   95 Only China, Japan, Persia (Iran) and Siam (Thailand) from Asia were present.   96 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 292.   97 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 290.   98 F.J. Brown, Chemical Warfare; a Study in Restraints, pp. 292–293.   99 F.J. Brown, Chemical Warfare; a Study in Restraints, p. 293. 100 Italian forces invaded Abyssinia (now Ethiopia) on 2 October 1935, completing the conquest and annexing the country by March 1936. See: ICRC ‘Abyssinian War (1935–1936)’. Available online at: www.icrc.org/Web/Eng/siteeng0.nsf/iwpList74/ A0B2CAE68823612241256C8700332B7B. 101 Reprisals are a term of legal art referring to acts which although illegal by themselves may be permissable as a deterrent in response to illegal acts by an enemy. G. Best, War and Law Since 1945, Oxford: Clarendon Press, 1997, p. 192. 102 K. Rutherford, ‘The Hague and Ottawa Conventions: A Model for Future Weapon Ban Regimes?’, The Nonproliferation Review, Spring/Summer 1999, p. 42. 103 G. Best, Humanity in Warfare, p. 159. 104 T. Pakenham, The Boer War, London: Weidenfeld and Nicolson, 1979, p. 251.

Notes   145 105 T.J. Lawrence, The Principles of International Law, Lexington, MA: D.C. Heath & Co, 1895, pp. 439–440. 106 K. Rutherford, ‘The Hague and Ottawa Conventions: A Model for Future Weapon Ban Regimes?’, p. 41. 3  The emergence of the landmine regimes   1 Capt. J.A. Roach, ‘Certain Conventional Weapons Convention: Arms Control or Humanitarian Law?’, Military Law Review 105, Summer 1984, p. 4.   2 Lt-­Col. B.M Carnahan, ‘The Law of Landmine Warfare: Protocol II to the United Nations Convention on Certain Conventional Weapons’, Military Law Review 105, Summer 1984, p. 95.   3 E. Prokosch, The Technology of Killing: A Military and Political History of Anti-­ Personnel Weapons, London: Zed Books, 1995, p. 160.   4 ICRC, Draft Rules for the Limitation of the Dangers Incurred by the Civilian population in Time of War, Geneva: ICRC, 2nd Edition 1958.   5 R.J. Mathews, ‘The 1980 Convention on Certain Conventional Weapons: A Useful Framework Despite Earlier Disappointments’, International Review of the Red Cross 844, December 2001, p. 993.   6 E. Prokosch, The Technology of Killing, p. 160.   7 ‘Militarily significant states’ is a term found in paragraph 6 to the preamble of the CCW.   8 Capt. J.A. Roach, ‘Certain Conventional Weapons Convention: Arms Control or Humanitarian Law?’ p. 14.   9 United Nations Press Release DC/2556 ‘Further Restrictions on Anti-­Personnel Landmines Adopted by Review Conference on Injurious Weapons.’ 10 ICRC, ‘Statement of the ICRC at the United Nations General Assembly’, 51st Session, First Committee, International Review of the Red Cross 315, 1 November 1996, pp. 631–636. 11 Capt. J.A. Roach, ‘Certain Conventional Weapons Convention: Arms Control or Humanitarian Law?’ p. 53. 12 The origins of the NGO-­led campaign to ban APMs, which brought this issue to international public attention, can be traced to three organizations working in Cambodia: the Coalition for Peace and Reconciliation, Handicap International (HI) and the Mines Advisory Group (MAG). At the beginning of the 1990s, through newsletters, publications and reports, the profile of the landmine issue was raised and the foundations set for the emergence of the ICBL in 1992. 13 This expression was first used at the last major preparatory meeting held in Brussels during 24–27 June 1997, prior to the Oslo negotiating conference. 14 Full title: The Convention on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and on their Destruction. 15 J. Williams, and S. Goose, ‘The International Campaign to Ban Landmines’, in M.A. Cameron, R.J. Lawson and B.W. Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, Oxford: Oxford University Press, 1998, p. 36. 16 Burkina Faso deposited the 40th ratification on 17 September 1998, bringing the treaty into force. 17 E. Prokosch, The Technology of Killing, p. 162. 18 The Arms Project, Landmines: A Deadly Legacy, Human Rights Watch and Physicians for Human Rights, 1993, p. 272. 19 Lt-­Col. B.M. Carnahan, ‘The Law of Landmine Warfare’, p. 88. 20 Lt-­Col. B.M. Carnahan, ‘The Law of Landmine Warfare’, p. 92. 21 Lt-­Col. B.M. Carnahan, ‘The Law of Landmine Warfare’, pp.  82–83, quoting G. Schwarzenberger, International Law as Applied by International Courts and Tribunals 1945–1988, London: Sweet and Maxwell, 1968.

146   Notes 22 R.J. Mathews, ‘The 1980 Convention on Certain Conventional Weapons’, p. 997. 23 For Finland, the geopolitical reality of a long land border with another non-­signatory – Russia – represents a Realpolitik-­driven exception to Finland’s otherwise notable humanitarian foreign policy tradition. Finland finally joined the mine ban treaty in January 2012. 24 E. Prokosch, The Technology of Killing, p. 149. 25 E. Prokosch, The Technology of Killing, p. 151. 26 Capt. J.A. Roach, ‘Certain Conventional Weapons Convention: Arms Control or Humanitarian Law?’ p. 4. 27 Capt. J.A. Roach, ‘Certain Conventional Weapons Convention: Arms Control or Humanitarian Law?’ p. 54. 28 As an example of the relationship between regime rules and effectiveness, it has taken 12 years (1997–2009) until the CD has been able to agree on a programme of work. 29 M.J. Matheson, ‘Filling the Gaps in the Conventional Weapons Convention’, Arms Control Today, November 2001. 30 It should, however, be noted that much of this extant work, in particular the seminal reference work on the Ottawa Process – To Walk Without Fear, the Global Campaign to Ban Landmines – was edited and written almost exclusively by Canadian officials, experts and scholars. Of the three editors, Maxwell Cameron and Brian Tomlin are academics at Canadian universities and Robert Lawson works with the Canadian government. 31 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, Humanitarianism and War Project, Occasional Paper 42, 2000, p. 20. 32 S. Neil MacFarlane, ‘Foreword’, in D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, xiii. 33 Capt. J.A. Roach, ‘Certain Conventional Weapons Convention: Arms Control or Humanitarian Law?’ p. 11. 34 Swedish Ministry of Foreign Affairs, ‘Conventional weapons. Their Deployment and Effects from a Humanitarian Aspect: Recommendations for the Modernisation of International Law’, A Swedish Working Group Study, 1973. 35 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, p. 17. 36 E. Prokosch, The Technology of Killing, pp. 154–155. 37 The Arms Project, Landmines: A Deadly Legacy, p. 303. 38 ICRC, ‘Statement of the ICRC at the United Nations General Assembly’, 51st Session, p. 632. 39 J. Williams and S. Goose, ‘The International Campaign to Ban Landmines’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, p. 30. 40 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, International Organization 52 (3), Summer 1998, p. 622. 41 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, p. 32. 42 The core of the ICBL was built around no more than 24 full time activists with expenditures of US$1–2 million per annum in the latter stages of the campaign. See: S.D. Goose, ‘Strategising About International Citizens Campaigns’, The Progressive Response 2 (8), March 20 1998. 43 J. Williams and S. Goose, ‘The International Campaign to Ban Landmines’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, p. 22. 44 R.J. Mathews, ‘The 1980 Convention on Certain Conventional Weapons’, p. 994. 45 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, p. 624. 46 For a full analysis of the role of the ICRC in the mine ban process see: L. Maresca and S. Maslen (eds) The Banning of Anti-­personnel Landmines. The Work of the

Notes   147 International Committee of the Red Cross 1955–1999, Cambridge: Cambridge University Press, 2000. 47 P. Blagden, Antipersonnel Mines: Friend or Foe? Geneva: ICRC, 1996. 48 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, p. 40. 49 J. Williams and S. Goose, ‘The International Campaign to Ban Landmines’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, p. 43. 50 Selected survey findings are described in M.A. Cameron, R.J. Lawson and B.W. Tomlin, ‘To Walk Without Fear’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, pp. 7–12. 51 In this book, which remains the only authoritative analysis of the Ottawa Process, those interested in a summary of the statistical data are asked to contact the editors (no contact details are given). M.A. Cameron, R.J. Lawson and B.W. Tomlin, ‘To Walk Without Fear’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, p. 17. 52 B. Boutros-­Ghali, ‘Foreword’, in K. Cahill (ed.), Clearing the Fields. Solutions to the Global Landmines Crisis, New York: Basic Books, 1995, xiv. 53 UN General Assembly, ‘An International Agreement to Ban Anti-­Personnel Landmines’, Resolution 51/45S, adopted 10 December 1996. The Resolution was adopted by 156 states in favour, ten abstentions and none against. 54 Final Report of the Governmental Experts, Article 49, CCW/CONF.I/GE/23. 55 E. Prokosch, The Technology of Killing, p. 185. 56 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, p. 12. 57 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, p. 13. 58 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, p. 624. 59 K. Anderson, ‘The Ottawa Convention Banning Landmines, the Role of International Non Governmental Organisations and the Idea of International Civil Society’, European Journal of International Law II (1), 2002, p. 112. 60 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, p. 24. 61 O.R. Young, ‘Political Leadership and Regime Formation: On the Development of Institutions in International Society’, International Organization 45 (3), 281–308. 62 D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, p. 26. 63 K. Anderson, ‘The Ottawa Convention Banning Landmines, the Role of International Non Governmental Organisations and the Idea of International Civil Society’, p. 105. 64 F. Kalshoven, ‘The Conventional Weapons Convention: Underlying Legal Principles’, International Review of the Red Cross 279, 1990, p. 517. 65 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, p. 630. 66 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, p. 631. 67 M. Finnemore and K. Sikkink, ‘International Norm Dynamics and Political Change’, Paper presented at the Ideas, Culture and Political Analysis Workshop, Princeton University, May 15–16 1998. Available online at: www.ciaonet.org/conf/ssr01/ ssr01ak.html (accessed 30 April 2002), p. 7. 68 K. Anderson, ‘The Ottawa Convention Banning Landmines, the Role of International Non Governmental Organisations and the Idea of International Civil Society’, European Journal of International Law II (1), 2002, p. 106. 69 J. White and K. Rutherford, ‘The Role of the Landmine Survivors Network’, in Cameron, Lawson and Tomlin (eds) To Walk Without Fear: The Global Movement to Ban Landmines, p. 111. 70 J. Williams and S. Goose, ‘The International Campaign to Ban Landmines’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, p. 43.

148   Notes 71 J. Williams and S. Goose, ‘The International Campaign to Ban Landmines’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, p. 46. 72 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, p. 640. 73 N. Stott, ‘The South African Campaign’, in Cameron, Lawson and Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, p. 73. 4  Implementing the landmine regimes   1 E.M. Filippino and T. Paterson, ‘Mine Action Lessons and Challenges: Is Mine Action Making a Difference. . . . Or Avoiding the Question’, Journal of Mine Action 9.1, July 2005.   2 See: International Mine Action Standards (IMAS), Edition 2, January 2003. Standard 04.10.   3 S. Maslen, Mine Action after Diana: Progress in the Struggle against Landmines, London: Pluto Press, 2004, pp. 62–63.   4 Final Report of the First Review Conference of the States Parties to the Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of APMs and on their Destruction; Nairobi Action Plan 2005–2009 APLC/CONF/2004/2005, Nairobi, 3 December 2004.   5 From an initial compliance rate of 63 per cent in 2001, subsequent annual rates were 75 per cent, 88 per cent, 91 per cent, 96 per cent, 96 per cent and 97 per cent in 2007. International Campaign to Ban Landmines, Landmine Monitor 2008, ICBL, p.  1. Available online at: www.the-­monitor.org.   6 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  2. Available online at: www.the-­monitor.org   7 T. Findlay, ‘Verification of the Ottawa Convention: Workable Hybrid or Fatal Compromise?’, Disarmament Forum 4, 1999.   8 ICBL ‘Letter to States Parties to the MBT regarding the Intersessional Standing Committee Meetings in May 2001’. Available online at: www.icbl.org/SC/y2/letter_to_ statepart_May_2001.php (accessed 7 July 2003).   9 ‘Austrian Intervention During the Standing Committee on the General Status and Operation of the Convention’, Statement by a representative of the Government of Austria, Geneva, 31 May 2002. 10 R.J. Mathews, ‘The 1980 Convention on Certain Conventional Weapons: A Useful Framework despite Earlier Disappointments’, International Review of the Red Cross 844, December 2001, p. 1007. 11 International Campaign to Ban Landmines, Landmine Monitor 2006, ‘About Landmine Monitor’, ICBL. Available online at: www.the-­monitor.org 12 Seven states contributed financial support to the first (1999) Landmine Monitor: Austria, Belgium, Canada, Ireland, Norway, the Netherlands and the United Kingdom. Of these, only the United Kingdom is not listed as a contributor according to the December 2010 schedule of government grants provided by the ICBL. See: International Campaign to Ban Landmines, Landmine Monitor 1999, ICBL, www.icbl.org/ index.php/LM/Our-­Research-Products/Landmine-­Monitor, and ICBL 2010 Annual Report. See: www.icbl.org/index.php/icbl/Library/About-­Us/Annual-­Reports/ ICBL_2010_report (accessed 15 February 2012). 13 S. Maslen, Mine Action after Diana: Progress in the Struggle against Landmines, p. 71. 14 At the 11 November 2011 meeting of APII states parties, ICBL Chair Steve Goose introduced his remarks with the statement that ‘the protocol is largely irrelevant – largely irrelevant due to the overwhelming success of the mine ban treaty’. See: www. icbl.org/idex.php/icbl/Library/News-­Articles/APII-­intervention.

Notes   149 15 K.E. Kjellman, ‘Norms, Persuasion and Practice: Landmine Monitor and Civil Society’, in K.B. Harpviken, (ed.) The Future of Humanitarian Mine Action, Basingstoke: Palgrave, 2004, p. 182. 16 K. Brinkert, ‘The convention banning anti-­personnel mines: applying the lessons of Ottawa’s past to meet the challenges of Ottawa’s future’, in Harpviken (ed.) The Future of Humanitarian Mine Action, p. 14. 17 ICBL, Mine Ban Treaty 11th Meeting of States Parties, Phnom Penh, Cambodia, ‘Summary of Thursday 1 December 2011’. Available online at: www.icbl.org/index. php/icbl/Treaty/MBT/Annual-­Meetings/11msp/Summary-­1Dec2011 (accessed 8 March 2012). 18 S. Nellen, ‘Lessons from the Implementation of the Anti-­Personnel Mine Ban Convention’, Disarmament Forum 1, 2005, pp. 66–67. 19 J.M. Beier, ‘ “Emailed Applications are preferred”: ethical practices in mine action and the idea of global civil society’, in Harpviken (ed.) The Future of Humanitarian Mine Action, p. 26. 20 2005 meeting of States Parties to the Conference on Certain Conventional Weapons, CCW/MSP/2005/2 para 38. 21 Vice-­presidents were nominated from Bulgaria, China, Cuba, the Czech Republic, Germany, Japan, Morocco, the Philippines, Poland and Switzerland. 22 R. Lloyd, ‘Strengthening the Inhumane Weapons Convention: Second Review Conference of the Certain Conventional Weapons Convention’, ISIS Policy Paper on Qualitative Arms Control 2, December 2001. 23 F. Faulkner, Moral Entrepreneurs and the Campaign to Ban Landmines, Amsterdam: Rodopi, 2007, pp. 38–39. 24 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, International Organization 52 (3), summer 1998, p. 620. 25 The composition of the board, minutes of its meetings, work plan and outputs are available on a website that also houses the updated IMAS. See: www.mineactionstandards.org. 26 E-­MINE can be found at: www.mineaction.org. 27 Examples include the MgM network established by German mine clearance organization Menschen Gegen Minen (People Against Landmines) at www.mgm.org/e/index. htm and the Inter-­Galactic EOD Forum moderated by mine action expert Andy Smith at: http://groups.google.com/group/igeod?hl=eng. 28 P.M. Haas, ‘Introduction: Epistemic Communities and International Policy Coordination’, International Organization 46 (1), 1992, p. 3. 29 P.M. Haas, ‘Introduction: Epistemic Communities and International Policy Coordination’, p. 4. 30 H. Muller, ‘The Internalization of Principles, Norms, and Rules by Governments: The Case of Security Regimes’, in V. Rittberger (ed.) Regime Theory and International Relations, Oxford: Clarendon Press, 1993, p. 386. 31 GICHD, A Study of the Development of Indigenous Mine Action Legislation, Geneva: GICHD, 2004. 32 N. Grobbelaar, (ed.) Mine Action in Southern Africa: Instrument of Development, Johannesburg: South African Institute of International Affairs, 2003, p. 134. 33 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  2. Available online at: www.the-­monitor.org. 34 GICHD, A Study of the Development of Indigenous Mine Action Legislation, p. 1. 35 GICHD, A Study of the Development of Indigenous Mine Action Legislation, p. 2. 36 This is typically under the auspices of an inter-­ministerial national mine action authority (NMAA). A Mine Action Centre (MAC) is responsible for day-­to-day coordination and implementation of mine action policy and activities. 37 United Nations General Assembly, Mine Action and Effective Coordination: The United Nations Policy, UN Doc. A/53/496, UNMAS: New York, 1998.

150   Notes 38 S. Maslen, Mine Action after Diana: Progress in the Struggle against Landmines, p. 103. 39 GICHD, A Study of Local Organisations in Mine Action, Geneva: GICHD, 2004. 40 P. Nergaard, ‘An Operator’s Perspective on Ottawa’s Article 5’, Journal of Mine Action 9.2, February 2006, p. 35. 41 UNMAS, 2004 Annual Report, New York: UNMAS, pp. 8–9. 42 Sixth meeting of States Parties, ‘Achieving the Aims of the Nairobi Action Plan: The Zagreb Progress Report’, APLC/MSP.6/2005/5 p. 21. 43 ICBL, Mine Ban Treaty 11th Meeting of States Parties, Phnomh Penh, Cambodia, ‘Summary of Friday 2 December 2011’. Available online at: www.icbl.org/index.php/ icbl/Treaty/MBT/Annual-­Meetings/11msp/Summary-­2Dec2011 (accessed 8 March 2012). 44 S. Maslen, Mine Action after Diana: Progress in the Struggle against Landmines, p. 71. 45 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  3. Available online at: www.the-­monitor.org. 46 ICBL Statement to the First Conference of States Parties to Amended Protocol II of the Convention on Certain Conventional Weapons. 15 December 1999. 47 See: www.the-­monitor.org. 48 International Campaign to Ban Landmines, Landmine Monitor 2000, ICBL, p.  748. Available online at: www.the-­monitor.org. 49 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  43. Available online at :www.the-­monitor.org. 50 DfID, ‘Landmines and Poverty: Breaking the Link’, DFID Issues, August 1998, p. 3. 51 DfID, ‘Landmines and Poverty: Breaking the Link’, p. 3. 52 For further information on IMSMA, see: www.gichd.org. 53 The World Bank, ‘Landmine Contamination: A Development Imperative’, Social Development Notes, Conflict Prevention and Reconstruction 20, October 2004, p. 2. 54 Reuters, ‘EU to launch into landmine removal’, 18 November 1997. 55 See: http://:ec.europa.eu/news/external_relations/110404_en.htm. 56 ICRC, ‘Mine Action 1999’, ICRC Special Report, Geneva: ICRC, August 2000. 57 A list of ITF donors can be found at: www.itf-­fund.si/Donors/ (accessed 7 February 2012). 58 K. Brinkert, ‘Paving the Bridge between Disarmament and Development: Resources Generated by the Anti-­Personnel Mine Ban Convention’, Disarmament Forum 3, 2003, p. 44. 59 The primary source of US government support – through the creative use of a public-­ private partnership – is the US Agency for International Development/Patrick J. Leahy War Victims Fund (USAID/LWVF ). 60 Statement by James F. Lawrence, Deputy Director, Office of Weapons Removal and Abatement, Bureau of Political-­Military Affairs, US Department of State, Standing Committee on Mine Clearance, Mine Risk Education and Mine Action Technologies, Geneva, 14 June 2005. 61 Brinkert, K. ‘Paving the Bridge between Development and Disarmament: Resources Generated by the Anti-­Personnel Mine Ban Convention’; Disarmament Forum 3, 2003, p. 50. 62 A. Hurrell, ‘International Society and the Study of Regimes: A Reflective Approach’, in Rittberger (ed.) Regime Theory and International Relations, p. 71. 63 International Campaign to Ban Landmines, Landmine Monitor 2004, ICBL, p.  254. Available online at: www.the-­monitor.org. 64 For a politically nuanced look at mine action in Afghanistan and Bosnia, see: M. Bolton and H. Griffiths, ‘Bosnia’s Political Landmines’, London: Landmine Action, 2006 and M. Bolton, ‘Goldmine? A Critical Look at the Commercialization of Afghan Demining’, Centre for the Study of Global Governance, Research Paper 01/2008, London School of Economics.

Notes   151 65 International Campaign to Ban Landmines, Landmine Monitor 2008, ICBL, p.  5. Available online at: www.the-­monitor.org/index.php/LM/Our-­Research-Products/ Landmine-­Monitor. 66 According to the Geneva Call website, as of 8 February 2012, forty-­one non-­state armed groups have agreed to ban the use of APMs through this mechanism. A template Deed and complementary information can be found at: www.genevacall.org. 67 Geneva Call, Armed Non-­State Actors and Landmines, A Global Report Profiling Non-­State Actors and their Use, Acquisition, Production, Transfer and Stockpiling of Landmines, Volume I, Geneva: Geneva Call, 2006, p. 14. 68 Geneva Call, Armed Non-­State Actors and Landmines, p. 31. 69 A US diplomatic cable published by Wikileaks describes how the Deputy Director General for Arms Control in the Turkish Ministry of Foreign Affairs lobbied the United States and others to prevent Geneva Call from obtaining consultative status within the UN Economic and Social Council (ECOSOC). The reason given was that the work of Geneva Call bolstered the presence and visibility of the PKK. See: http://:metaleaks.net/document.php?id=253655 (accessed 19 February 2012). 70 K.E. Kjellman, ‘Norms, persuasion and practice: Landmine Monitor and civil society’, in Harpviken (ed.) The Future of Humanitarian Mine Action, p. 185. 71 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  1. Available online at: www.the-­monitor.org. 72 G. Harris, ‘The Economics of Landmine Clearance: Case Study Cambodia’, Journal of International Development 12 (2), 2000, pp. 219–225. 73 C. Horwood, ‘Humanitarian Mine Action: The First Decade of a New Sector in Humanitarian Aid’, Relief and Rehabilitation Network 32, March 2000, p. 30. 74 The Paris Declaration on Aid Effectiveness was endorsed on 2 March 2005 by over one hundred Ministers, Heads of Agency and senior officials. The complementary Accra Agenda for Action was drawn up in 2008. 75 C. Horwood, ‘Ideological and analytical foundations of mine action: human rights and community impact’, in Harpviken (ed.) The Future of Humanitarian Mine Action, p. 167. 5  Humanitarian demining   1 As defined by the IMAS, humanitarian demining involves the range of ‘activities which lead to the removal of mines and UXO hazards, including technical survey, mapping, clearance, marking, post-­clearance documentation, community mine action liaison and the handover of cleared land’. International Mine Action Standards IMAS, Edition 2, January 2003. Standard 3.42.   2 M. Croll, The History of Landmines, Barnsley: Leo Cooper, 1988, p. 136.   3 The 2008 Edition of Landmine Monitor identifies more than 70 mine affected countries around the world. International Campaign to Ban Landmines, Landmine Monitor 2008, ICBL, p. 19. Available online at :www.the-­monitor.org.   4 GICHD, A Guide to Mine Action, Geneva: GICHD, 2003, p. 65.   5 C. King, ‘The Demining Toolkit’, in GICHD, Mine Action: Lessons and Challenges, Geneva: GICHD, 2005, p. 34.   6 Lt-­Col. B.M. Carnahan, ‘The Law of Landmine Warfare: Protocol II to the United Nations Convention on Certain Conventional Weapons’, Military Law Review 105, Summer 1984, p. 84.   7 Permanent Mission of India to the United Nations, ‘Statement by Ambassador Savitri Kadi, Permanent Representative of India to the First Annual Conference of the States Parties to the Amended Protocol II to the CCW’, Geneva, 15 December 1999.   8 D. Banerjee and M. Joseph, Anti-­Personnel Landmines: a South Asian Regional Survey, New Delhi, 1999, pp. 20–21.   9 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  29. Available online at: www.the-­monitor.org.

152   Notes 10 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  29. Available online at: www.the-­monitor.org. 11 S. Maslen and P. Herby, ‘An International Ban on Anti Personnel Mines. History and Negotiation of the Ottawa Treaty’, International Review of the Red Cross 325, December 1998, pp. 693–713. 12 ICRC, ‘The Weapons Issue: Statement of the ICRC at the United Nations General Assembly’, 18 October 1996 before the First Committee, International Review of the Red Cross, 315, 1 November 1996, pp. 631–636. 13 E. Prokosch, The Technology of Killing: a Military and Political History of Anti-­ Personnel Weapons, London: Zed Books, 1995, p. 182. 14 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p. 8 and Landmine Monitor 2010, ICBL, pp. 2–4. Available online at: www.the-­monitor.org. 15 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  23. Available online at: www.the-­monitor.org. 16 Nairobi Action Plan, Action no. 27. Final Report of the First Review Conference. APLC/CONF/2004/5, 9 February 2005, p. 99. 17 N. Grobbelaar, (ed.) Mine Action in Southern Africa: Instrument of Development, Johannesburg: South African Institute of International Affairs, 2003, p. 41. 18 D. Barlow, ‘The Parable of the Two Sons’, Journal of Mine Action 11 (2), Spring 2008, p. 11. 19 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  17. Available online at: www.the-­monitor.org. 20 ICBL, Mine Ban Treaty 11th Meeting of States Parties, Phnomh Penh, Cambodia, ‘Statement on Mine Clearance’. Available online at: www.icbl.org/index.php/icbl/ Treaty/MBT/Annual-­Meetings/11msp/11MSP_Statement_Mine_Clearance (accessed 8 March 2012). 21 ICBL, ‘Fulfilling the obligations of Article 5 of the mine ban treaty’, Food for Thought Paper, November 2011. Available online at: www.icbl.org/index.php/icbl/ Treaty/MBT/Annual-­Meetings/11msp/Art5-Paper (accessed 8 March 2012). 22 ICBL, Mine Ban Treaty 11th Meeting of States Parties, Phnomh Penh, Cambodia, ‘Statement on Mine Clearance’. 23 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  30. Available online at: www.the-­monitor.org. 24 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  20. Available online at: www.the-­monitor.org. 25 Interview with an (anonymous) military official, 20 April 2005. International Campaign to Ban Landmines, Landmine Monitor 2005, ICBL, p. 612. Available online at: www.the-­monitor.org. 26 See: ‘Country and Area Reports’, International Campaign to Ban Landmines, Landmine Monitor 2007. Available online at: www.the-­monitor.org. 27 See: ‘Venezuela’, in International Campaign to Ban Landmines, Landmine Monitor 2011. Available online at: www.the-­monitor.org. 28 ICBL Statement to the First Conference of States Parties to Amended Protocol II of the Convention on Certain Conventional Weapons, 15 December 1999. 29 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  29. Available online at: www.the-­monitor.org. 30 D. Attwood ‘Implementing Ottawa: Continuity and Change in the Roles of NGOs’, Disarmament Forum 4, 1999, p. 29. 31 According to the 2006 edition of Landmine Monitor, armed non-­state groups used APMs in ten countries: Burma, Burundi, Colombia, Guinea-­Bissau, India, Iraq, Nepal, Pakistan, Russia/Chechnya and Somalia. The 2010 edition identifies continued use in Afghanistan, Colombia, India, Myanmar, Pakistan and Yemen. 32 Handicap International, ‘What was said at the Summit’. Available online at: www. handicap-­international.org.uk/page_394.php; (accessed Wednesday 1 December 2004).

Notes   153 33 GICHD, Mine Action: Lessons and Challenges, pp. 32–33. 34 C. Taylor, ‘Humanitarian Demining Programs: The First Decade of a New Sector in Humanitarian Aid’, Relief and Rehabilitation Network Paper 32, March 2006, p. 8. 35 Norway, ‘Resources to Achieve the Convention’s Humanitarian Aims’, Non-­paper presented at the Fourth Meeting of the States Parties, Geneva, 18 September 2003. 36 The Praxis Group Ltd., ‘Willing to Listen: An Evaluation of the United Nations Mine Action Programme in Kosovo 1999–2001’, February 2002, p. 20. 37 The Praxis Group Ltd., ‘Willing to Listen’, p. 8. 38 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  32. Available online at: www.the-­monitor.org. 39 International Campaign to Ban Landmines, Landmine Monitor 2011. Available online at: www.the-­monitor.org. 40 See: ‘Myanmar’ and ‘Niger’, in International Campaign to Ban Landmines, Landmine Monitor 2011. Available online at: www.the-­monitor.org. 41 K. Brinkert, ‘Paving the Bridge between Disarmament and Development: Resources Generated by the APMBC’, Disarmament Forum 3, 2003, p. 44. 42 K. Brinkert, ‘Paving the Bridge between Disarmament and Development’ in Disarmament Forum 3, pp. 44–45. 43 Government of Peru, Response to the Questionnaire distributed by the Norwegian Coordinator of the Resource Mobilisation Contact Group, 27 January 2003. 44 Croatian Ministry of Foreign Affairs and Croatian Mine Action Centre, ‘Funding Demining in the Republic of Croatia’, Presentation to the Standing Committee on the General Status and Operation of the Convention, Geneva, 3 February 2003. 45 See: BBC News, ‘Cambodian Mine Action Agency Accused of Corruption’, World News: Asia Pacific (Friday 9 July, 1999). 46 Information obtained from the website of the NGO Forum on Cambodia. Available online at: www.ngoforum.org.kh (accessed on 21 December 2003). 47 See: www.apminebanconvention.org (accessed 10 January 2012). 48 Leo Mong Hay, ‘Where is Cambodia’s anti-­corruption law?, UPI Asia.com, Available online at: www.upiasia.com/Human_Rights/2008/05//27where_is_cambodias_anti-­ corruption_law/2456/ (accessed 18 November 2011). 49 R. Eaton, C. Horwood and N. Niland, ‘The Development of Indigenous Mine Action Capacities’, Study Report Prepared for the United Nations Department of Humanitarian Affairs, 1997, p. 67. 50 The Praxis Group Ltd., ‘Willing to Listen’, p. 80. 51 Contamination was subsequently found in six districts declared free of mines and UXO, leading to new surveys and additional clearance. S. Maslen, Mine Action after Diana: Progress in the Struggle against Landmines, London: Pluto Press, 2004, p. 93. 52 Interview with anonymous mine action expert, Siem Reap, 1 July 2001. 53 N. Grobbelaar, (ed.) Mine Action in Southern Africa: Instrument of Development, 2003, p. 60. 54 The Pentagon has requested $1.66 billion for research and production of two new landmine systems between 2006 and 2013 which, according to Landmine Monitor, are incompatible with the mine ban treaty. International Campaign to Ban Landmines, Landmine Monitor 2007. Available online at: www.the-­monitor.org. 55 K. Anderson, ‘The Ottawa Convention Banning Landmines, the Role of International Non Governmental Organisations and the Idea of International Civil Society’, European Journal of International Law II (1), 2002. 56 European Parliament, ‘A World Without Landmines’, Resolution # P6_TA-­PROV, 0298, 2005. 57 A. Bryden and A. McAslan, Mine Action Equipment: Study of Global Operational Needs, Geneva: GICHD, 2002, p. 6. 58 German Initiative to Ban Landmines, Mine Action Programmes from a Development-­ Oriented Point of View: The Bad Honnef Framework, Berlin: GIBL, 1999, p. 42.

154   Notes 59 N. Grobbelaar, (ed.) Mine Action in Southern Africa: Instrument of Development, 2003, p. 16. 60 J.M. Beier, ‘ “Emailed Applications are Preferred”: ethical practices in mine action and the idea of global civil society’, in K.B. Harpviken (ed.) The Future of Humanitarian Mine Action, Basingstoke: Palgrave, 2004, p. 28. 61 Lardner is quoted by M.J. Flynn, ‘Political Minefield’, in R. Mathew, B. McDonald and K.A. Rutherford, Landmines and Human Security, New York: Suny Press, 2004, p. 121. 62 A. Bryden and A. McAslan, Mine Action Equipment: Study of Global Operational Needs, pp. 7–8. 63 M.J. Matheson, ‘Filling the Gaps in the Conventional Weapons Convention’, Arms Control Today, November 2001. 64 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, pp. 771–772. Available online at: www.the-­monitor.org. 65 Further details on the Falklands Initiative are available at: www.landmineaction.org. 66 See: www.landmineaction.org. 67 International Campaign to Ban Landmines, Landmine Monitor 2011, available online at: www.the-­monitor.org. 68 ICBL, Mine Ban Treaty 11th Meeting of States Parties, Phnomh Penh, Cambodia, ‘Summaries of Wednesday 30 November and Thursday 1 December 2011’. 69 Final Report of the 7th Meeting of the States Parties, Geneva, 18–22 September 2006, APLC/MSP.7 2006/5 p. 14. 70 United Nations Mission in Kosovo, Mine Action Coordination Cell, Mine Action Programme Annual Report’, 2001. 71 Quote taken from The HALO Trust website: www.halotrust.org/Kosovo.html (accessed 10 January 2007). 72 International Campaign to Ban Landmines, Landmine Monitor 2011, available online at: www.the-­monitor.org. 73 For humanitarian demining in the context of post-­conflict peacebuilding see: A. Bryden and H. Hänggi (eds) Security Sector Reform and Post-­Conflict Peacebuilding, Munich: Lit Verlag, 2005. 74 S. Maslen, Mine Action after Diana: Progress in the Struggle against Landmines, p. 34. 75 K.B. Harpviken and R. Roberts (eds) Preparing the Ground for Peace – Mine Action in Support of Peacebuilding, Oslo: PRIO, 2004, p. 58. 76 See: United Nations, ‘Mine Action Guidelines for Ceasefire and Peace Agreements’, United Nations Inter-­Agency Coordination Group on Mine Action, New York, 2003. 77 K.B. Harpviken and B.A. Skara, ‘Humanitarian Mine Action and Peacebuilding: Exploring the Relationship’, in Harpviken (ed.) The Future of Humanitarian Mine Action, p. 37. 78 Final Report of the 7th Meeting of the States Parties, Geneva, 18–22 September 2006, APLC/MSP.7/2006/5 p. 25. 79 M. Dahinden, ‘Humanitarian Demining at a Crossroads – A Farwell Lecture’, unpublished, 1 July 2004. 6  Stockpile destruction   1 International Mine Action Standards (IMAS), Edition 2, January 2003. Standard 3.202.   2 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  15. Available online at: www.the-­monitor.org.   3 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  4. Available online at: www.the-­monitor.org   4 IRIN, ‘Laying Landmines to Rest: Humanitarian Mine Action’, www.newsite.

Notes   155 irrinnews.org/InDepthMain.aspx?InDepthID=19&ReportID=62807 (accessed 5 January 2003).   5 Mine Ban Treaty, Article 4.   6 S. Maslen, ‘Destruction of Anti-­personnel Mine Stockpiles’, in GICHD, Mine Action: Lessons and Challenges, Geneva: GICHD, 2005, p. 194.   7 Figures drawn from International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p. 16. Available online at: www.the-­monitor.org.   8 Final Report of the First Review Conference of the States Parties to the Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of APMs and on their Destruction; Nairobi Action Plan 2005–2009 APLC/CONF/2004/2005; Nairobi, 3 December 2004, part III, Actions #9–16.   9 Nairobi Action Plan 2005-2009 APLC/CONF/2004/2006, Nairobi, 3 December 2004, part III, Action 15 10 International Campaign to Ban Landmines, Landmine Monitor 2008, ICBL, p.  9. Available online at:www.the-­monitor.org 11 Presentation by the Democratic Republic of the Congo, Standing Committee on Stockpile Destruction, Geneva, 11 May 2006. 12 United Nations, Statement on the Issue of Stockpile Destruction to the Intersessional Programme, 2005. 13 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  326. Available online at: www.the-­monitor.org. 14 In June 2005, the United Nations Mission in DRC (MONUC) reported that the Union of Congolese Patriots – Lubamba (UPC/L) had been responsible for mine laying. UN OCHA, ‘Monitoring de la Situation Humanitaire en RDC’, 11–17 June 2005. 15 ‘Rapport General de la Table Ronde sur la Mise en Oeuvre de la Convention d’Ottawa Sur l’Interdiction des Mines Antipersonnel’, Republique Democratique du Congo, 6–7 December 2005. 16 Standing Committee on Stockpile Destruction, Meeting Report, 11–12 May 2006, p. 3. 17 International Campaign to Ban Landmines, Landmine Monitor 2008, ICBL, p.  79. Available online at: www.the-­monitor.org. 18 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  14. Available online at: www.the-­monitor.org. 19 Turkmenistan, Article 7 Report submitted 14 November 2001, Forms B-­D. 20 Standing Committee on Stockpile Destruction, ‘Update on Implementation of Article 4’, 30 May 2002. 21 S. Maslen, Mine Action after Diana: Progress in the Struggle against Landmines, London: Pluto Press, 2004, pp. 68–69. 22 International Campaign to Ban Landmines, Landmine Monitor 2005, p.  594. Avail­ able online at: www.the-­monitor.org. 23 T.M. Franck, The Power of Legitimacy Among Nations, Oxford: Oxford University Press, 1990, p. 184. 24 Article 7 paragraphs 1b, d and f. 25 UN Final Report of the First Meeting of the States Parties to the Convention on the Prohibition of the Use, Stockpiling, Production and Trade of Anti-­personnel mines and on their Destruction, May 1999, www.disarmament.un.org/mineban.nsf (accessed 5 June 2003). 26 Amended Protocol II, Article 13 c. 27 See: www.mineaction.org. 28 Statement by Sergei Ivanov, Minister of Defence, Parliamentary Hearings, 23 November 2004. 29 The PFM-­1 remotely-­deliverable APM provides a very specific destruction challenge. It contains hydrogen chloride so open detonation as a destruction method would lead to environmental pollution.

156   Notes 30 International Campaign to Ban Landmines, Landmine Monitor 2005, ICBL, p.  854. Available online at: www.the-­monitor.org. 31 R. Dogov, ‘Landmines in Russia and the Former Soviet Union: A Lethal Epidemic’, Medicine and Global Survival 7 (1), pp. 40–41. 32 Afghanistan, Colombia, India, Iraq, Myanmar, Pakistan and Turkey. International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p. 16. Available online at: www.the-­monitor.org. 33 Geneva Call, Armed Non-­State Actors and Landmines, A Global Report Profiling Non-­State Actors and their Use, Acquisition, Production, Transfer and Stockpiling of Landmines, Volume I, Geneva: Geneva Call, 2006, p. 28. 34 Entry into force of APII for the Russian Federation took place on 2 September 2005. 35 O.R. Young, (ed.) ‘Global Governance. Drawing Insights from the Environmental Experience’, An Occasional Paper from the Dickey Center, Dartmouth College, Hanover, New Hampshire, 1995, p. 19. 36 Standing Committee on Stockpile Destruction (11–12 May 2006): p. 4. 37 International Campaign to Ban Landmines, Landmine Monitor 2002, ICBL. www. the-­monitor.org. 38 K. Dansereau, ‘NATO-­sponsored Project Reduces Massive Mine Stockpile across Eastern Ukraine’. Journal of Mine Action 7.2, 2003. 39 For Belarus see: www.the-­monitor.org/index.php/cp/display/region-­profiles/find_ profile/BY/2011 (accessed 10 January 2012). For Ukraine see: www.the-­monitor.org. 40 Mine ban treaty, Article 3 paragraph 1. 41 Algeria, Bangladesh, Brazil and Turkey. International Campaign to Ban Landmines, Landmine Monitor 2008, ICBL, p. 11. Available online at: www.the-­monitor.org. 42 International Campaign to Ban Landmines, Landmine Monitor 2010, ICBL, p.  8. Available online at: www.the-­monitor.org. 43 United States Department of State Factsheet, 27 February 2004. 44 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, pp. 11–13. Available online at: www.the-­monitor.org. 45 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  132. Available online at: www.the-­monitor.org. 46 International Campaign to Ban Landmines, Landmine Monitor 2006, ICBL, p.  83. Available online at: www.the-­monitor.org 47 S. Maslen, ‘Destruction of Anti-­personnel Mine Stockpiles’, in GICHD, Mine Action: Lessons and Challenges, p. 196. 48 S. Maslen, ‘Destruction of Anti-­personnel Mine Stockpiles’, in GICHD, Mine Action: Lessons and Challenges, p. 197. 49 UK Permanent Representation to the Conference on Disarmament, ‘Anti-­personnel Landmine Stockpiles and their Destruction: A Progress Report’, 11 May 2001. 50 Parliamentary answer by Foreign and Commonwealth Office (FCO), Hansard, 11 September 2003, col. 400W. 51 OECD DAC, ‘Conflict Prevention and Peacebuilding: What Counts as ODA?’, Paris: OECD, 3 March 2005. 52 In order to comply with the UN Millennium Development Goals (MDGs), donors have committed to raising their levels of ODA to 0.7 per cent of gross national income by 2015. Each year, many donors fail to achieve this level of ODA. 53 Statement by United Nations Development Programme on the status of UN support to Stockpile Destruction to the Intersessional Standing Committee on Stockpile Destruction, Geneva, 6 February 2003. 54 Standing Committee Meeting on Stockpile Destruction, Geneva 12 February 2004, ‘European Commission Project for the destruction of PFM-­1 stockpiles in Ukraine’, Statement by P.Krejsa, p. 2.

Notes   157 7  The effectiveness of global landmine regimes   1 www.halotrust.org.   2 Prominent examples include: K. Anderson, ‘The Ottawa Convention Banning Landmines, the Role of International Non Governmental Organisations and the Idea of International Civil Society’, European Journal of International Law II (1), 2002; D. Hubert, ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, Humanitarianism and War Project, Occasional Paper 42, 2000; and K. Rutherford, ‘The Hague and Ottawa Conventions: A Model for Future Weapon Ban Regimes?’, The Nonproliferation Review, Spring/Summer 1999.   3 The Conference on the Arms Trade Treaty is scheduled to convene between 2–27 July 2012.   4 The Rome Statute of the ICC entered into force on 1 July 2002. As of 21 November 2011, the ICC has 119 states parties.   5 The Optional Protocol entered into force on 12 February 2002. As of 1 November 2011, it has 143 states parties.   6 R. Price, ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, International Organization 52 (3), Summer 1998, p. 617.   7 The Arms Project, Landmines: A Deadly Legacy, Human Rights Watch and Physicians for Human Rights, 1993, p. 266.   8 Full title: Protocol on Explosive Remnants of War, (Protocol V to the 1980 CCW Convention).   9 P. Ellis, ‘ERW States Conclude a Protocol’, Journal of Mine Action 8.1, 2004, p. 52. 10 Protocol V entered into force on 12 November 2006 and currently has 76 states parties. The Convention on Cluster Munitions entered into force on 1 August 2010 and has 66 states parties (figures accurate as of 21 November 2011). 11 D. Kaye and S.A. Solomon, ‘The Second Review Conference of the 1980 Convention on Certain Conventional Weapons’, American Journal of International Law 96 (4), October 2002, p. 927.

Bibliography

Aldrich, G.H. and C.M. Chinkin, ‘Symposium: The Hague Peace Conferences, A Century of Achievement and Unfinished Work’, The American Journal of International Law, January 2000. Alvarez, J., ‘Multilateralism and its Discontents’, European Journal of International Law 11 (2), 2000, 393–411. Anderson, K., ‘The Ottawa Convention Banning Landmines, the Role of International Non Governmental Organisations and the Idea of International Civil Society’, European Journal of International Law II (1), 2002. Araujo, R.J., ‘Anti-­personnel Mines and Peremptory Norms of International Law: Argument and Catalyst’, Vanderbildt Journal of Transnational Law 30, January 1997. Arend, A. and R. Beck, International Law and the Use of Force, London: Routledge, 1993. Arts, B., ‘Regimes, Non-­State Actors and the State System: A ‘Structurational’ Regime Model’, European Journal of International Relations 6 (4), 513–542. Attwood, D., ‘Implementing Ottawa: Continuity and Change in the Roles of NGOs’, Disarmament Forum 4, 1999. Attwood, D., ‘Promoting Compliance: Observations on the Anti-­Personnel Mine Ban Convention’, GCSP Arms Control and Disarmament Cluster. Available online at: www.gcsp.ch/e/research/ACCluster/Papers99–2000/Atwood.htm (accessed 18 November 2001). Bailey, S.D., Prohibitions and Restrictions in War, Oxford: Oxford University Press, 1972. Banerjee, D. and M. Joseph, Anti-­Personnel Landmines: a South Asian Regional Survey, New Delhi, 1999. Barlow, D., ‘The Parable of the Two Sons’, Journal of Mine Action 11 (2), Spring 2008. Barnett, M.N. and M. Finnemore, ‘The Politics, Power and Pathologies of International Organizations’, International Organization 53 (4), 699–732. Baxter, R.R. ‘Multilateral Treaties as Evidence of Customary International Law’, British Yearbook of International Law 41 (275), 1965–1966. Baxter, R.R. and T. Burgenthal, ‘Legal Aspects of the Geneva Protocol of 1925’, Amer­ ican Journal of International Law 64, 1970. Baylis, J. and S. Smith, (eds), The Globalization of World Politics, Oxford: Oxford University Press, 2001. BBC News, ‘Cambodian Mine Action Agency Accused of Corruption’, World News: Asia Pacific, Friday 9 July, 1999. Beach, Gen. Sir H., ‘Qualitative Arms Control and International Humanitarian Law applicable to Armed Conflict’, ISIS Policy Paper on Qualitative Arms Control 1, July 2001.

Bibliography   159 Beier, J.M., ‘Siting Indiscriminacy: India and the Global Movement to Ban Landmines’, Global Governance 8 (3), 2002. Best, G., Humanity in Warfare, London: Weidenfield and Nicolson, 1980. Best, G., War and Law Since 1945, Oxford: Clarendon Press, 1997. Blagden, P., Antipersonnel Mines: Friend or Foe?, Geneva: ICRC, 1996. Bleicher, M., ‘The Ottawa Process: Nine Day Wonder or a New Model for Disarmament Negotiations?’, Disarmament Forum 4, 1999. Blix, H., ‘Current Efforts to Prohibit the Use of Certain Conventional Weapons’, Instant Research on Peace and Violence 4 (1), 1974, 21–30. Bolton, M. ‘Goldmine? A Critical Look at the Commercialization of Afghan Demining’, Centre for the Study of Global Governance, Research Paper 01/2008, London School of Economics, 2008. Bolton, M. and H. Griffiths, ‘Bosnia’s Political Landmines’, London: Landmine Action, 2006. Bottigliero, I., 120 Million Landmines Deployed Worldwide: Fact or Fiction? Barnsley: Leo Cooper, 2000. Boulden, L., The Politics of Demining: Mine Clearance in Southern Africa, Johannesburg: South African Institute of International Affairs, 1999. Boutros-­Ghali, B., ‘Agenda for Peace: Preventive Diplomacy, Peacemaking and Peacekeeping’, Report of the Secretary-­General pursuant to the Statement adopted by the Summit Meeting of the Security Council on 31 January 1992, A747/277-S/24111. Brem, S. and K. Rutherford, ‘Walking Together or Divided Agenda? Comparing Landmines and Small-­Arms Campaigns’, Security Dialogue 32 (2), 169–186. Brierly, J.L., The Law of Nations, 6th edn, Oxford: Oxford University Press, 1991. Brinkert, K., ‘Paving the Bridge Between Disarmament and Development: Resources Generated by the APMBC’, Disarmament Forum 3, 2003. Brown, F.J., Chemical Warfare; a Study in Restraints, Princeton: Princeton University Press, 1968. Bryden, A. and A. McAslan, Mine Action Equipment: Study of Global Operational Needs, Geneva: GICHD, 2002. Bryden, A. and H. Hänggi (eds), Security Sector Reform and Post-­Conflict Peacebuilding, Munich: Lit Verlag, 2005 Burnett, M.N. and M. Finnemore, ‘The Politics, Power and Pathologies of International Organizations’, International Organization 53 (4), 699–732. Byers, M., Custom, Power and the Power of Rules, Cambridge: Cambridge University Press, 1999. Cahill, K., (ed.), Clearing the Fields. Solutions to the Global Landmines Crisis, New York: Basic Books, 1995. Cameron, M.A., R.J. Lawson and B.W. Tomlin (eds), To Walk Without Fear: The Global Movement to Ban Landmines, Oxford: Oxford University Press, 1998. Camilleri, J.H. and J. Falk, The End of Sovereignty. The Politics of a Shrinking and Fragmenting World, Cheltenham: Edward Elgar, 1992. Carnahan, Lt-­Col. B.M., ‘The Law of Landmine Warfare: Protocol II to the United Nations Convention on Certain Conventional Weapons’, Military Law Review 105, Summer, 1984. Cassese, A., ‘Means of Warfare, the Present and Emerging Law’, Revue Belge de Droit International, XII, 1976. Cassese, A., International Law in a Divided World, Oxford: Oxford University Press, 1988.

160   Bibliography Chayes, A. and A.H. Chayes, ‘On Compliance’, International Organization 47 (2), 175–205. Chomsky, N., The New Military Humanism: Lessons from Kosovo, Monroe, ME: Common Courage Press, 1999. Cirincione, J. (ed.), Repairing the Regime, London: Routledge, 2000. Claude, I.L., Swords Into Ploughshares. The Problems and Progress of International Organization, 3rd edn, New York: Random House, 1964. Coghlan, D., ‘Progress on Mine Action Principles’, Disarmament Forum 4, 1999. Cooley, A., and J. Ron, ‘The NGO Scramble. Organizational Insecurity and the Political Economy of Transnational Action’, International Security 27 (1), 5–39. Cooper, S., Arbitration or War: Contemporary Reactions to the Hague Peace Conference of 1899, New York: Garland, 1972. Cooper, S., Internationalism in Nineteenth Century Europe: The Crisis of Ideas and Purpose, New York: Garland, 1976. Corkson, J. and J. Nottingham, The Control of Chemical and Biological Weapons, New York: Carnegie Endowment for International Peace, 1971. Croatian Ministry of Foreign Affairs and Croatian Mine Action Centre, ‘Funding ­Demining in the Republic of Croatia’, Presentation to the Standing Committee on the General Status and Operation of the Convention, Geneva, 3 February 2003. Croll, M., The History of Landmines, Barnsley: Leo Cooper, 1988. Coupland, R. and P. Herby, ‘The SIrUS Project: Progress Report on Superfluous Injury or Unnecessary Suffering’, International Review of the Red Cross 835, 1999, 583–592. Coupland, R. and D. Loye, ‘The 1899 Hague Declaration Concerning Expanding Bullets. A Treaty Effective for more than 100 Years Faces Complex Contemporary Issues’, International Review of the Red Cross 85 (849), 2003, 135–142. Dahinden, M., ‘Humanitarian Demining at a Crossroads – A Farewell Lecture’, unpublished, 1 July 2004. Dansereau, Kevin., ‘NATO-­sponsored Project Reduces Massive Mine Stockpile across Eastern Ukraine’, Journal of Mine Action 7.2, August 2003. Davenport, D., ‘The New Diplomacy’, Policy Review 11 (6), 2002. Detter, I., The Law of War, Cambridge: Cambridge University Press, 2000. DfID, ‘Landmines and Poverty: Breaking the Link’, DFID Issues, August 1998. Dogov, R., ‘Landmines in Russia and the Former Soviet Union: A Lethal Epidemic’, Medicine and Global Survival 7 (1). Draper, G.I.A.D. ‘Implementation of International Law in Armed Conflicts’, International Affairs 48, 1972, 46–59. Draper, G.I.A.D., ‘The Emerging Law of Weapons Restraint’, Survival, January/February 1997. Durham, H. and T.L.H. McCormack (eds), The Changing Face of Conflict and the Efficacy of International Humanitarian Law, The Hague: Kluwer Law International, 1999. Eaton, R., C. Horwood and N. Niland, ‘The Development of Indigenous Mine Action Capacities’, Study Report Prepared for the United Nations Department of Humanitarian Affairs, 1997. Efaw, Capt. A.C.S., ‘The United States Refusal to Ban Landmines: The Intersection Between Tactics, Strategy, Policy and International Law’, Military Law Review 159, 1999, 87–151. Efinger, M., M. Mendler and V. Rittberger, ‘Towards an East–West Security Regime: The Case of Confidence and Security Building Measures’, Journal of Peace Research 27 (1), 1990, 55–74. Ellis, P., ‘ERW States Conclude a Protocol’, Journal of Mine Action 8. 1, 2004.

Bibliography   161 European Commission, ‘The Response of the European Union to the Anti-­personnel Landmines Challenge’, Luxembourg: Office for Official Publications of the European Communities, 2000. Faulkner, F., ‘Anti-­personnel Landmines: A Necessary Evil?’, International Relations 8 (4), April 1987. Faulkner, F., Moral Entrepreneurs and the Campaign to Ban Landmines, Amsterdam: Rodopi, 2007. Feld, W. and R. Jordan, International Organizations: A Comparative Approach, 3rd edn, Westport, CT: Praeger, 1999. Filippino, E.M. and T. Paterson, ‘Mine Action Lessons and Challenges: Is Mine Action Making a Difference. . . . Or Avoiding the Question’, Journal of Mine Action 9.1, July 2005. Final Report of the First Review Conference of the States Parties to the Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of APMs and on Their Destruction; Nairobi Action Plan 2005–2009 APLC/CONF/2004/2005; Nairobi, 3 December 2004. Final Report of the Governmental Experts, Article 49, CCW/CONF.I/GE/23. Findlay, T., ‘Verification of the Ottawa Convention: Workable Hybrid or Fatal Compromise?’, Disarmament Forum 4, 1999. Finnemore, M. and K. Sikkink, ‘International Norm Dynamics and Political Change’, Paper presented at the Ideas, Culture and Political Analysis Workshop, Princeton University, May 15–16 1998. Available online at: www.ciaonet.org/conf/ssr01/ssr01ak. html (accessed 30 April 2002). Fleck, D. (ed.), The Handbook of Humanitarian Law in Armed Conflicts, Oxford: Oxford University Press, 1995. Franck, T.M., The Power of Legitimacy Among Nations, Oxford: Oxford University Press, 1990. Geneva Call, Armed Non-­State Actors and Landmines, A Global Report Profiling Non-­ State Actors and their Use, Acquisition, Production, Transfer and Stockpiling of Landmines, Volume I, Geneva: Geneva Call, 2006. German Initiative to Ban Landmines, Mine Action Programmes from a Development-­ Oriented Point of View: The Bad Honnef Framework, Berlin: GIBL, 1999. GICHD, A Study of Socio-­economic Approaches to Mine Action, Geneva: GICHD, 2001. GICHD, A Guide to Mine Action, Geneva: GICHD, 2003. GICHD, A Study of the Development of Indigenous Mine Action Legislation, Geneva: GICHD, 2004. GICHD, A Study of Local Organisations in Mine Action, Geneva: GICHD, 2004. GICHD, Mine Action: Lessons and Challenges, Geneva: GICHD, 2005. Goldblat, J., Agreements for Arms Control – A Critical Survey, Stockholm: SIPRI, 1982. Goldblat, J., Arms Control. A Guide to Negotiations and Agreements, Thousand Oaks, CA: Sage Publications Ltd, 1994. Goldblat, J., ‘Ways to Improve the Implementation and Enforcement of Arms Control Agreements – Role of Verification’, GCSP Arms Control & Disarmament Cluster. Available online at: www.gcsp.ch/e/research/ACCluster/Papers99–2000/Goldblat.htm (accessed 22 January 2002). Goose, S.D., ‘Strategising About International Citizens Campaigns’, The Progressive Response 2 (8), March 20 1998. Goose, S.D., ‘The Ottawa Process and the 1997 Mine Ban Treaty’, Yearbook of International Humanitarian Law 1, 1998.

162   Bibliography Gordenker, L. and T. Weiss (eds), NGOs, the United Nations and Global Governance, Boulder, CO: Lynne Rienner Publishers, 1996. Green, L.C. ‘What One May Do in Combat – Then and Now’, in A.J.M Delissen and G.J. Tanja (eds), Humanitarian Law of Armed Conflict: Challenges Ahead, Leiden: Martinus Nijhoff Publishers, 1991. Greenwood, C., Humanitarian Law and Laws of War. Centennial of the First International Peace Conference, Preliminary Report, June 1998. Grobbelaar, N. (ed.), Mine Action in Southern Africa: Instrument of Development; Johannesburg: South African Institute of International Affairs, 2003. Haas, E.B., ‘Why Collaborate? Issue Linkage and International Regimes’, World Politics 32, 1980, 357–405. Haas, P.M., ‘Introduction: Epistemic Communities and International Policy Coordination’, International Organization 46 (1), 1992. Haas, P.M., ‘Knowledge, Power and International Policy Coordination’, International Organization 46 (1), 1992. Haas, P.M., R.O. Keohane and M.A. Levy (eds), Institutions for the Earth: Sources of Effective International Environmental Protection, Cambridge, MA: MIT Press, 1993. Haggard, S. and B.A. Simmons, ‘Theories of International Regimes’, International Organization 41 (3), 1987. Hall, B. and T.J. Biersteker (eds), The Emergence of Private Authority in Global Governance, Cambridge: Cambridge University Press, 2002. Hammond, J.W., Poison Gas: Myths Versus Reality, Westport, CT: Greenwood Press, 1999. Handicap International, ‘What was said at the Summit’. Available online at: www. handicap-­international.org.uk/page_394.php; (accessed Wednesday 1 December 2004). Harpviken, K.B., ‘The Landmines Campaign: Representations and Reflections’, Cooperation and Conflict: Journal of the Nordic International Studies Association 37 (1), 101–108, 2002. Harpviken, K.B. (ed.), The Future of Humanitarian Mine Action, Basingstoke: Palgrave, 2004. Harpviken K.B. and R. Roberts (eds), Preparing the Ground for Peace – Mine Action in Support of Peacebuilding, Oslo: PRIO, 2004. Harris, G., ‘The Economics of Landmine Clearance: Case Study Cambodia’, Journal of International Development 12 (2), 2000, 219–225. Hasenclever, A., P. Mayer and V. Rittberger, Theories of International Regimes, Cambridge: Cambridge University Press, 1997. Herrmann, R.K. and V.P. Shannon, ‘Defending International Norms: The Role of Obligation, Material Interest, and Perception in Decision-­making’, International Organisation 55 (3), 621–654. Hettne, A., A. Inotai and O. Sunkel (eds), Globalisation and the New Regionalism, New York: Palgrave Macmillan, 1999. Higgins, R., Problems and Process. International Law and How We Use It, Oxford: Clarendon Press, 1994. Hill, M. and P. Hupe, Implementing Public Policy, Thousand Oaks, CA: Sage Publications, 2002. Hobsbawm, E.J., The Age of Revolution 1789–1848, London: Weidenfeld and Nicolson, 1962. Horwood, C., ‘Humanitarian Mine Action: The First Decade of a New Sector in Humanitarian Aid’, Relief and Rehabilitation Network 32, March 2000.

Bibliography   163 Hubert, D., ‘The Landmine Ban: A Case Study in Humanitarian Advocacy’, Humanitarianism and War Project, Occasional Paper 42, 2000. Hull, W.I., The Two Hague Conferences and their Contribution to International Law, New York: Garland, 1972. Hulme, D. and M. Edwards, Non-­Governmental Organizations – Performance and Accountability, London: Earthscan publications, 1995. Hulme, D. and M. Edwards, NGOs, States & Donors: Too Close for Comfort, New York: Palgrave Macmillan, 1996. Human Rights Watch, Landmines – A Deadly Legacy, New York: Human Rights Watch, 1993. ICBL, Landmine Monitor Reports. (1999–2011) (International Campaign to Ban Landmines). Available online at: www.the-­monitor.org/index.php/LM/Our-­ResearchProducts/Landmine-­Monitor. ICBL, Statement to the First Conference of State Parties to Amended Protocol II of the Convention on Certain Conventional Weapons, 15 December 1999. ICBL, ‘Nations Refusing to Ban Landmines Meet Again’, ICBL Press Release. Available online at: www.icbl.org/prelease/1999/dec15.html (accessed 8 May 2000). ICBL, ‘Letter to States Parties to the MBT regarding the Intersessional Standing Committee Meetings in May 2001’. Available online at: www.icbl.org/SC/y2/letter_to_ statepart_May_2001.php (accessed 7 July 2003). ICBL, Mine Ban Treaty 11th Meeting of States Parties, Phnom Penh, Cambodia, ‘Summary of Thursday 1 December 2011’. Available online at: www.icbl.org/index. php/icbl/Treaty/MBT/Annual-­Meetings/11msp/Summary-­1Dec 2011 (accessed 8 March 2012). ICBL, Mine Ban Treaty 11th Meeting of States Parties, Phnom Penh, Cambodia, ‘Summary of Friday 2 December 2011’. Available online at: www.icbl.org/index.php/ icbl/Treaty/MBT/Annual-­Meetings/11msp/Summary-­2Dec 2011 (accessed 8 March 2012). ICISS, ‘The Responsibility to Protect’, Report of the International Commission on Intervention and State Sovereignty, Ottawa: International Development Research Centre, 2001. ICRC, ‘Abyssinian War (1935–1936)’, Available online at: www.icrc.org/Web/Eng/ siteeng0.nsf/iwpList74/A0B2CAE68823612241256C8700332B7B (accessed 5 July 2003). ICRC, Draft Rules for the Limitation of the Dangers Incurred by the Civilian population in Time of War, Geneva: ICRC, 2nd edn, 1958. ICRC, ‘The Weapons Issue: Statement of the ICRC at the United Nations General Assembly’, 51st Session, First Committee, International Review of the Red Cross, 315, 1 November 1996. ICRC, ‘Statement of the International Committee of the Red Cross, Preparatory Committee for the 2001 Review Conference of the United Nations Convention on Certain Conventional Weapons’, Geneva, 14 December 2000. ICRC, Submunitions and other Unexploded Ordnance. A Study, Geneva: ICRC, 2000. ICRC, Clusterbombs and Landmines in Kosovo, Geneva: ICRC, 2000. ICRC, Implementing the Ottawa Convention: National Legislation, Geneva: ICRC, 1999. ICRC, ‘Mine Action 1999’, ICRC Special Report, Geneva: ICRC, August 2000. ICRC, Banning Anti-­Personnel Mines. The Ottawa Treaty Explained, Geneva: ICRC. 1998. ICRC, The Geneva Conventions of August 12 1949, Geneva: ICRC, 1997 edn.

164   Bibliography ICRC, Protocols Additional to the Geneva Conventions of August 12 1949, Geneva: ICRC, 1997 edn. ICRC, Landmines: Time for Action, Geneva: ICRC, 1994. ICRC, ‘Report of the ICRC for the Review Conference of the 1980 UN Conventions on Prohibitions or Restrictions on the Use of Certain Conventional Weapons Which may be Deemed to be Excessively Injurious or to have Indiscriminate Effects’, International Review of the Red Cross 299, April 1994, 123–182. ICRC, ‘The Vienna Review Conference: Success on Blinding Laser Weapons but Deadlock on Landmines’, International Review of the Red Cross 309, 31 December 1995. ICRC, ‘Report of the International Committee of the Red Cross to the Preparatory Committee for the 2001 Review Conference of the United Nations Convention on Certain Conventional Weapons: Scope of Application and Explosive Remnants of War’, Geneva, 14 December 2000. International Peace Research Association, ‘Arms Control Agreements and Humanitarian Laws of War’, Bulletin of Peace Proposals 17 (3–4), 1986. International Mine Action Standards, edn 2, 2003. Available online at: www.mineactionstandards.org, (accessed 4 June 2004). IRIN, ‘Laying Landmines to Rest: Humanitarian Mine Action’. Available online at: www.newsite.irrinnews.org/InDepthMain.aspx?InDepthID=19&ReportID=62807 (accessed 5 January 2003). Jones, B.D. and C.K. Cater, ‘Civilians in War: 100 Years After the Hague Peace Conference’, New York: International Peace Academy, 1999. Kalshoven, F., ‘Arms, Armaments and International Law’, Recueil des Cours 191, 1985 II. Kalshoven, F., ‘The Conventional Weapons Convention: Underlying Legal Principles’, International Review of the Red Cross 279, 1990. Kaye, D. and S.A. Solomon, ‘The Second Review Conference of the 1980 Convention on Certain Conventional Weapons’, American Journal of International Law 96 (4), October 2002. Keck, M.E. and K. Sikkink, Activists Beyond Borders. Advocacy Networks in International Politics, Ithaca: Cornell University Press, 1998. Keegan, J., The First World War, London: Pimlico, 1999. Keohane, R.O., After Hegemony, Princeton: Princeton University Press, 1984. Keohane, R.O., International Organizations and State Power: Essays in International Relations Theory, Boulder, Colorado: Westview Press, 1989. Keohane, R.O., Power and Governance in a Partially Globalized World, London: Routledge, 2002. King, C., ‘Legislation and the Landmine’, Janes Intelligence Review Special Report 16, 1997. Kongstad, S., ‘The Continuation of the Ottawa Process: Intersessional Work and the Role of Geneva’, Disarmament Forum 4, 1999. Krasner, S.D. (ed.), International Regimes, Ithaca: Cornell University Press, 1983. Krasner, S.D., Structural Conflict: The Third World Against Global Liberalism, Berkeley: University of California Press, 1985. Kratochwil, F. and J.G. Ruggie, ‘International Organization: A State of the Art on the Art of the State’, International Organization 40 (4), Autumn 1986, 753–775. Lachowski, Z., ‘The Ban on Anti-­Personnel Mines’, in The SIPRI Yearbook 1998: Armaments, Disarmament and International Security, Stockholm: SIPRI, 545–558, 1998. Laffey, M. and J. Weldes, ‘Beyond Belief: Ideas and Symbolic Technologies in the Study

Bibliography   165 of International Relations’, European Journal of International Relations 3 (2), 1997, 193–237. Latham, A., ‘Towards an Effective Verification Regime for the Convention on Certain Conventional Weapons: The Outline of an Incremental Approach’, Centre for Defence and Security Studies, Winnipeg, December 1994. Latham, A., ‘The Politics of Stigmatisation: Global Cultural Change and the Transnational Campaign to Ban Antipersonnel Landmines’, International Studies Association 40th Annual Convention (Feb 16–20 1999). Available online at: www.ciaonet. org/isa/laa01/index.html. (accessed 15 February 2002). Lauterpacht, H., ‘The Problem of the Revision of the Law of War’, British Yearbook of International Law, 29, 1952. Lawrence, T.J., The Principles of International Law, Lexington, MA: D.C. Heath & Co, 1895. League of Nations, ‘Proceedings of the Conference for the Supervision of the International Trade in Arms and Ammunition and in Implements of War’, League of Nations Document A 13, 1925. Lenarcic, D.A., ‘Knight Errant? Canada and the Crusade to Ban Anti-­personnel Landmines’, Contemporary Affairs 2, 1998. Levie, H.S., ‘Prohibitions and Restrictions on the Use of Conventional Weapons’, St John’s Law Review 68, Summer 1994. Levie, H.S., ‘History of the Law of War on Land’, International Review of the Red Cross 838, June 2000, 339–350. Levy, M.A., O.R. Young and M. Zurn, ‘The Study of International Regimes’, European Journal of International Relations 1 (3), 1995, 267–330. Li Ping, X., ‘The Latest Hot Topic in the Field of International Arms Control – Landmines’, International Prospect, Shanghai, August 1997. Lloyd, R., ‘Strengthening the Inhumane Weapons Convention: Second Review Conference of the Certain Conventional Weapons Convention’, ISIS Policy Paper on Qualitative Arms Control 2, December 2001. MacFarlane, S.N., ‘Politics and Humanitarian Action’, Occasional Paper 41, Thomas J. Watson Institute for International Studies and the United Nations University, 2000. MacFarlane, S.N., ‘Humanitarian Action: The Conflict Connection’, Occasional Paper 43, Thomas J. Watson Institute for International Studies, 2001. Maresca, L., ‘Inhumane Weapons: The Second Review Conference on the CCW and Developments in International Humanitarian Law’, ISIS Policy Paper on Qualitative Arms Control 3, April 2002. Maresca, L. and S. Maslen (eds), The Banning of Anti-­personnel Landmines. The Work of the International Committee of the Red Cross 1955–1999, Cambridge: Cambridge University Press, 2000. Martin, L., ‘Interests, Power and Multilateralism’, International Organization 46 (4), September 1992, 765–792. Maslen, S. and P. Herby, ‘An International Ban on Anti Personnel Mines. History and Negotiation of the “Ottawa Treaty” ’, International Review of the Red Cross 325, December 1998. Maslen, S., Mine Action after Diana: Progress in the Struggle against Landmines, London: Pluto Press, 2004. Matheson, M.J., ‘Filling the Gaps in the Conventional Weapons Convention’, Arms Control Today, November 2001.

166   Bibliography Mathew, R., and K. Rutherford, ‘Banning Landmines in the American Century’, International Journal on World Peace, 16 (2), 1999. Mathew, R., B. McDonald and K.A. Rutherford, Landmines and Human Security, New York: Suny Press, 2004. Mathews, R.J., ‘The 1980 Convention on Certain Conventional Weapons: A Useful Framework Despite Earlier Disappointments’, International Review of the Red Cross 844, December 2001. Mathews, R.J. and T.L.H. McCormack, ‘The Influence of Humanitarian Principles in the Negotiation of Arms Control Treaties’, International Review of the Red Cross 834, June 1999, 331–352. McCormack, T.L.H., ‘International Law and the Use of Chemical Weapons in the Gulf War’, California Western International Law Journal 21, 1990–91, 5–10. McGinnis, M. and Elinor Ostrom, ‘Design Principles for Local and Global Commons’, Paper prepared for presentation at a conference on linking local and global commons, Harvard University, 23–25 April 1992. McGrath, R., ‘Cluster Bombs. The Military Effectiveness and Impact on Civilians of Cluster Munitions’, The UK Working Group on Landmines, 2000. Meron, T., ‘International Criminalisation of Internal Atrocities’, American Journal of International Law, 89, July 1995. Molander, J., ‘Strengthening International Humanitarian Law: The Review Conference of the 1980 Conventional Weapons Convention’, UNIDIR Newsletter 28/29, December 1994–May 1995. Moosberg, L., ‘Does the Swedish Use of the 12.7 mm Multipurpose Projectile Undermine the St. Petersburg Declaration?’, Masters Thesis (Spring 2003). Available online at: www.teol.uu.se/noha/masters/moosberg.pdf (accessed 10 September 2004). Muller, H., ‘Compliance Politics: A Critical Analysis of Multilateral Arms Control Treaty Enforcement’, GCSP Arms Control & Disarmament Cluster. Available online at: www. gcsp.ch/e/research/ACCluster/Papers99–2000/Muller.htm. (accessed 22 January 2001). Nairobi Action Plan, Final Report of the First Review Conference. APLC/CONF/2004/5, 9 February 2005. Nellen, S., ‘Lessons from the Implementation of the Anti-­Personnel Mine Ban Convention’, Disarmament Forum 1, 2005, 66–67. Neumann, I.B., ‘Harnessing Social Power: State Diplomacy and the Landmines Issue’, in A. Cooper (ed.), Enhancing Global Governance: Towards a New Diplomacy, Tokyo: UN University Press, 2002. Nergaard, P., ‘An Operator’s Perspective on Ottawa’s Article 5’, Journal of Mine Action 9.2, February 2006. New Statesman, ‘The CS Gas Muddle’, 31 July 1970. New York Times, ‘Letter to President Clinton’, 3 April 1996, A9. Nicholson, M., Rationality and the Analysis of International Conflict, Cambridge: Cambridge University Press, 1995. OECD DAC, ‘Conflict Prevention and Peacebuilding: What Counts as ODA?’, Paris: OECD, 3 March 2005. Pakenham, T., The Boer War, London: Weidenfeld and Nicolson, 1979. Paris, R., ‘Human Security: Paradigm Shift of Hot Air?’, International Security 26 (2), Fall 2001, 87–102. Peas, Kelly-­Kate, International Organizations: Perspectives on Global Governance, Upper Saddle River NJ: Prentice Hall, 4th edn, 2009. Prakash, A. and J.A. Hart (eds), Globalization and Governance, London: Routledge, 1999.

Bibliography   167 Pressman, J. and A. Wildavsky, Implementation, Berkeley: University of California Press, 1973. Price, R., ‘Reversing the Gunsights: Transnational Civil Society Targets Landmines’, International Organization 52 (3), Summer 1998, 613–644. Prokosch, E., The Technology of Killing: A Military and Political History of Anti-­ Personnel Weapons, London: Zed Books, 1995. Prokosch, E., ‘The Swiss Draft Protocol on Small-­Calibre Weapon Systems’, International Review of the Red Cross 307, August 1995, 411–425. Putnam, R.D., ‘Diplomacy and Domestic Politics: The Logic of Two-­level Games’, International Organization 42, 1988, 427–460. ‘Rapport General de la Table Ronde sur la Mise en Oeuvre de la Convention d’Ottawa Sur l’Interdiction des Mines Antipersonnel’, Republique Democratique du Congo, 6–7 December 2005. Reinicke, W.H., Global Public Policy. Governing Without Government?, Washington D.C.: Brookings Institution Press, 1998. Reuters, ‘EU to launch into landmine removal’, 18 November 1997. Rieff, D., ‘The Precarious Triumph of Human Rights’, New York Times Magazine, Section 6, 76, 8 August 1999. Risse-­Kappen, T., ‘Ideas Do Not Float Freely: Transnational Coalitions, Domestic ­structures, and the End of the Cold War’, International Organization 48, 1994, 185–214. Rittberger, V. (ed.), Regime Theory and International Relations, Oxford: Clarendon Press, 1993. Roach, Capt. J.A., ‘Certain Conventional Weapons Convention: Arms Control or Humanitarian Law?’, Military Law Review 105, Summer 1984. Roberts, A. and R. Guelff (eds), Documents on the Laws of War, Oxford: Clarendon Press, 2nd edn, 1989. Roberts, A and R. Guelff (eds), Documents on the Laws of War, Oxford: Oxford University Press, 3rd edn, 2000. Roht-­Arriaza, N., ‘Institutions of International Justice’, Journal of International Affairs 52, Spring 1999, 474–491. Rosenau, J.N. and E.O. Czempiel (eds), Governance Without Government. Order and Change in World Politics, Cambridge: Cambridge University Press, 1992. Roth, K., ‘New Minefields for NGOs: After the War on Landmines, These Organisations Started New Campaigns’, Nation, 13 April 1998, 22. Ruggie, J., ‘International Responses to Technology: Concepts and Trends’, International Organization 29 (3), 1975. Ruggie, J.G., ‘Multilateralism: The Anatomy of an International Institution’, International Organization 46 (3), Summer 1992. Rutherford, K., ‘The Hague and Ottawa Conventions: A Model for Future Weapon Ban Regimes?’, The Nonproliferation Review, Spring-­Summer 1999. Rutherford, K., ‘The Evolving Arms Control Agenda: Implications of the Role of NGOs in Banning Anti-­Personnel Landmines’, World Politics 53, 2000, 74–114. Rutherford, K., Disarming States: The International Movement to Ban Landmines, Santa Barbara CA: Praeger, 2010. Sandoz, Y., ‘ICRC Involvement in Banning or Restricting the use of Certain Weapons’, Presentation at the Seventeenth Annual Seminar for Diplomats on International Humanitarian Law, 2000. Sandoz, Y. ‘The Importance of Humanitarian Law and Action for Security’, Abstract of

168   Bibliography presentation at 4th International Security Forum, Geneva 15–17 November 2000. Available online at: www.isn.ethz.ch/isn/Digital-­Library/Publications/Detail/?id=81. Schonberg, K., ‘The Generals’ Diplomacy: US Military Influence in the Treaty Process, 1992 to 2000’, Presented at the Annual Meeting of the American Political Science Association, August 30 to September 3, 2000. Washington D.C. Schwarzenberger, G., International Law as Applied by International Courts and Tribunals 1945–1988, London: Sweet and Maxwell, 1968. Scott, J.B., Proceedings of the Hague Peace Conference of 1899, New York: Garland, 1920. Scott, J.B., The Hague Peace Conferences of 1899 and 1907, New York: Garland, 1972 edn. Self, W.A. and J.A. Roach, Index of International Humanitarian Law, Geneva: ICRC, 1992. Serge, S., ‘Vers un Cour Pénale Internationale: La Convention de Rome entre les ONG et le Conseil de Securité’, Revue Générale de Droit International Publique 103 (1), 1999, 29–45. Short, N., ‘The Role of NGOs in the Ottawa Process to Ban Landmines’, International Negotiation 4, 1999, 481–500. Simmons, P.J. and Chantal De Jonge Oudraat (eds), Managing Global Issues: Lessons Learned, Washington D.C.: Carnegie Endowment for International Peace, 2001. Smilie, I. and H. Helmich (eds), Government-­NGO Partnerships for International Development, Earthscan, 1999. Snidal, D., ‘The Relative-­Gains Problem for International Cooperation: Response’, American Political Science Review 87, 1993. Sommaruga, C., ‘Eliminating Anti-­personnel Landmines: Continuing the Process’, Address by Dr Cornelio Sommaruga, President of the ICRC, Zagreb, 28 June 1999. Sommaruga, C., ‘Without Reservation – Addressing the Challenges of Achieving a Landmine-­Free World’, Speech given at an International Symposium to mark the 5th Anniversary of the Ottawa Convention (November 30–1 December 2002). Available online at: www.gichd.ch (accessed 8 January 2003). Swedish Ministry of Foreign Affairs, ‘Conventional weapons. Their Deployment and Effects from a Humanitarian Aspect: Recommendations for the Modernisation of International Law’, A Swedish Working Group Study, 1973. Taylor, C., ‘Humanitarian Demining Programs: The First Decade of a New Sector in Humanitarian Aid’, Relief and Rehabilitation Network Paper 32, March 2006. Tcheou, W., Proceedings of the Conference for the Supervision of the International Trade in Arms and Ammunition and in Implements of War, Doc A 13, Geneva, 1925. The Arms Project, Landmines: A Deadly Legacy, Human Rights Watch and Physicians for Human Rights, 1993. The Economist, ‘Sins of the Secular Missionaries’, The Economist, 29 January 2000. The Economist, ‘Survey Human Rights Law: The Conscience of Mankind’, The Economist, 5 December 1998. The European Commission, ‘Mine Actions in the World’, Luxemburg: Office for Official Publications of the European Communities, 2004. The Praxis Group Ltd., ‘Willing to Listen: An Evaluation of the United Nations Mine Action Programme in Kosovo 1999–2001’, February 2002. The Spectator, ‘The Tyranny of Human Rights’, The Spectator, 28 August 1999, 11. The World Bank, ‘Landmine Contamination: A Development Imperative’, Social Development Notes, Conflict Prevention and Reconstruction 20, October 2004.

Bibliography   169 Tomlin, B., M. Cameron and B. Lawson (eds), To Walk Without Fear: The Global Movement to Ban Landmines, Oxford: Oxford University Press, 1998. Tuchman, B.W., The Proud Tower. A Portrait of the World Before the War: 1890–1914, New York: Macmillan, 1966. Underdal, A., ‘The Concept of Regime Effectiveness’, Cooperation and Conflict 27, 1992, 227–240. UNIDIR, ‘Framework for a Mine-­free World’, Disarmament Forum 4, 1999. UN Final Report of the First Meeting of the States Parties to the Convention on the Prohibition of the Use, Stockpiling, Production and Trade of Anti-­personnel mines and on their Destruction, May 1999. Available online at: www.disarmament.un.org/mineban. nsf (accessed 5 June 2003). UN General Assembly, ‘An International Agreement to Ban Anti-­Personnel Landmines’, Resolution 51/45S, adopted 10 December 1996. United Nations, ‘Mine Action Guidelines for Ceasefire and Peace Agreements’, United Nations Inter-­Agency Coordination Group on Mine Action, New York, 2003. United Nations General Assembly, Mine Action and Effective Coordination: The United Nations Policy, UN Doc. A/53/496, UNMAS: New York, 1998. United Nations General Assembly, Report of the Panel on United Nations Peace Operations, A/55/305 S/2000/809, August 2000. UN OCHA, ‘Monitoring de la Situation Humanitaire en RDC’, 11–17 Juin 2005. UNMAS, 2004 Annual Report, New York: UNMAS, 2004. United Nations Press Release DC/2556, ‘Further Restrictions on Anti-­Personnel Landmines Adopted by Review Conference on Injurious Weapons’, Available online at: www.un.org/News/Press/docs/1996/ (accessed 9 May 2000). U.S. Congressional Record, 69th Congress, 2nd Session, LXVIII Part I. Vagts, D.F., ‘Taking Treaties Less Seriously’, American Journal of International Law 92 (3), July 1998. Vakil, A.C., ‘Confronting the Classification Problem: Towards a Taxonomy of NGOs’, World Development 25 (12), 1997. Van Tuijl, P., ‘NGOs and Human Rights: Sources of Justice and Democracy’, Journal of International Affairs 52 (2), Spring 1999, 474–491. Vazquez, C.M., ‘The Four Doctrines of Self-­executing Treaties’, American Journal of International Law 89 (4), October 1995. Viotti, P. and M. Kauppi, International Relations Theory. Realism, Pluralism, Globalism, New York: MacMillan, 2nd edn 1993. Woker, G., ‘Chemical and Biological Weapons’, in Inter-­Parliamentary Union, What Would be the Character of a New War?, London: Victor Gollancz, 1933. Young, O.R., ‘Political Leadership and Regime Formation: On the Development of Institutions in International Society’, International Organization 45 (3), 281–308. Young, O.R. (ed.), ‘Global Governance. Drawing Insights from the Environmental Experience’, An Occasional Paper from the Dickey Center, Dartmouth College, Hanover, New Hampshire, 1995. Young, O.R. and Gail Osherenko (eds), Polar Politics: Creating International Environmental Regimes, Cornell University Press, 1993. Zanders, J.P., ‘The Chemical Weapons Convention in the Context of the 1925 Geneva Debates’, The Nonproliferation Review, Spring–Summer 1996. Zimmern, A.P., The League of Nations and the Rule of Law 1918–1935, New York: Macmillan, 1936.

Index

1899 Hague Peace Conference 17–19, 21, 27, 30, 36 1925 Gas Protocol 2–4, 14, 17–40, 51, 60, 98, 123, 128, 130, 136 advocacy 3, 5–6, 10, 15, 58, 61, 65, 67, 80, 85, 87, 106, 119, 121, 126–7, 129, 136 Afghanistan 24, 32, 53, 77, 80–1, 87, 90, 95, 109–10, 118–19 Anderson, Ken 58, 61, 99 Angola 53, 77, 91, 96, 109, 115 anti-personnel mine (APM): ban vii–viii, 3, 5–6, 13, 41, 44–6, 53, 58–61, 63, 87, 127–9, 132; export 50–1; production 50, 53–5, 62, 82, 129; trade 54, 77, 82, 129; transfer 51, 77, 108, 118; use 48, 50–1, 53–4, 77, 81–4, 91, 94, 107–8, 133 anti-vehicle mine (AVM) 4, 46, 48, 65, 107, 116 Ardagh, General Sir John 20, 23, 28–9 armed non-state actor (ANSA) 50, 81–3, 85, 92, 94, 99, 106, 109–10, 112, 115, 117, 121, 125 arms control 3, 5, 21, 35, 44, 47, 49, 59, 71 Austria 45, 52–3 Axworthy, Lloyd 5, 44, 59 Beier, J. Marshall 73, 100 Belarus 82, 107, 114–15, 120–1, 131 Belgium 52–3 Best, Geoffrey 18, 30 booby trap 4, 48, 52–3, 65 Bosnia-Herzegovina 81 Boutros-Ghali, Boutros 5, 44, 56 Brinkert, Kerry 72, 80, 96 Brown, Frederic J. 32, 34, 37 Brussels Declaration 45, 58 Cambodia 24, 54, 81, 97–8, 131

Cambodian Mine Action Centre (CMAC) 97 Cameron, Maxwell 5 Canada 44–5, 52–3, 55–7, 59, 103, 111, 114 Caribbean Community (CARICOM) 54 Chechnya 77, 93–4 chemical weapons 11, 21, 24–6, 33–7, 40, 128 Chemical Weapons Convention (CWC) 25–6, 45, 68, 128 child soldier 135–7 China 5, 27, 34, 50–1, 57, 67, 82, 98–9, 101, 107, 129, 133 civil society viii, 3–5, 30–1, 33–4, 36, 39, 46–7, 50–1, 54, 56–63, 67, 69–71, 73, 81, 85, 93–4, 111–13, 117, 126, 129–30, 136 cluster munitions viii, 102, 137 Cold War 114 Colombia 75, 82, 92 Conference on Disarmament (CD) 51–3, 58, 146n28 Convention on Certain Conventional Weapons (CCW) 2, 4, 7, 15, 43–4, 46–7, 49–57, 60, 63, 67, 71, 73, 80, 85, 94–5, 111, 127–8, 136–7 Coupland, Robin 23 Croll, Mike 87 Deed of Commitment 82, 99, 115, 117 defection: involuntary defection 7–8, 65, 74–7, 85, 93, 105, 109, 111, 121, 133; voluntary defection 39, 65, 74–5, 78, 82, 93, 105, 109 Democratic Republic of the Congo (DRC) 89, 96, 109–10, 121 disarmament 26, 35, 49, 59, 71, 79 disarmament, demobilization and reintegration (DDR) 97, 103, 109, 119

Index   171 dum dum bullets 3, 11, 17, 19–20, 23–4, 26, 28–9, 31, 36, 38–40, 129–30, 142n27, 142n38 epistemic community 15, 73–4, 126 expanding bullets see dum-dum bullets explosive remnants of war (erw) 137 European Commission 79, 92, 110, 114–15, 119–20 European Parliament 99 European Union 45, 52, 79, 128, 135 fact finding mission (FFM) 49, 68–9, 75 Falkland Islands 89, 101–2, 134 Finnemore, Martha 9, 11 First World War 4, 20–1, 33, 37, 39, 56 France 23–4, 32, 43, 50, 61, 88, 92, 118

International Campaign to Ban Landmines (ICBL) vii, 5, 44–7, 52–8, 60–2, 64, 67, 68–73, 80, 92, 109, 120 International Court of Justice (ICJ) 27 International Criminal Court (ICC) vii, 25, 135–7 international humanitarian law (IHL) vii, viii, 1–7, 9, 11, 13–14, 17, 21–2, 25–7, 33–4, 38–44, 48, 52, 56, 59–60, 62, 66, 119, 130, 133–5, 137–8 International Mine Action Standards (IMAS) 74–5, 107 International Red Cross (IRC) 33–4, 36, 39, 56 Iraq 24–5, 50, 77, 80, 82, 95, 107, 119 Israel 50, 77, 82, 117 Italy 37–8, 50

Geneva viii, 21, 31, 33, 36, 42 Geneva Call 81–3, 85, 96, 99, 112, 115, 117, 121, 151n66 Geneva Conventions 2, 42 Geneva International Centre for Humanitarian Demining (GICHD) vii, x, 64, 67, 71, 75, 78, 86 Germany 27–8, 31–2, 37, 52, 92, 118 Goldblat, Jozef 17, 23–4 Goose, Steve 56, 95 Great Britain 20, 27–9, 32, 38; see also United Kingdom Grobelaar, Neuma 91, 98 Guelff, Richard 25

Kosovo 94–5, 97–8, 102, 137, 144n100, 153n51 Kosovo Liberation Army (KLA) 97 Kosovo Protection Corps (KPC) 97–8

The Hague 19, 22, 30–1, 36, 137 Hague Declaration III 3–4, 13–14, 17–40, 123, 130, 142n18, 142n37 HALO Trust x, 102, 127 Handicap International 57, 61 Horwood, Chris 84 Hubert, Don 55 human security 120, 130, 135 humanitarian demining 15, 65–6, 74, 81, 87–105, 131–2; see also mine clearance Hungary 92

McGrath, Rae 100 Mali 92 Maslen, Stuart 76, 108, 118 Matheson, Michael 51, 100–1 middle powers 49, 52, 56 military necessity 1, 25, 29, 32, 38, 55, 92 military utility 5, 17, 47, 50, 101, 136 mine action: capacity building 76–8, 85, 97, 111, 124, 131–3; corruption 76, 81, 83, 85, 97, 125, 131; donors 63–4, 70, 74–7, 79, 81, 84–5, 97, 100, 113, 119, 124, 129; funding 78–9, 80–1, 94–6, 103, 121, 131, 150n59, see also resources; impact 10, 47, 83–4, 87, 95, 99–103, 106–7, 118–20, 131, 133–4; technology 75, 79, 100, 132 mine awareness see mine risk education mine clearance 48, 68 mine risk education (MRE) 65, 68 mined area 88–9, 91–3, 102, 109 minefield 38–9, 48–9, 89–90, 101–2

Implementation Support Unit (ISU) 67, 71, 127 India 5, 50–1, 77, 89, 107, 129 Information Management System for Mine Action (IMSMA) 78 inhumane weapons 14, 39, 60 International Committee of the Red Cross (ICRC) 5, 26, 37, 42–6, 53–7, 59, 61–2, 68, 72, 79, 90, 102, 137

Landmine Action 102 Landmine Monitor 65, 67–8, 70, 72, 83–4, 88, 91–6, 107, 109–10, 112–13, 118, 125, 127, 131, 135 Landmine Survivors Network (LSN) 61 Lardner, Tim 100 Lawson, Robert 5 Leahy, Patrick 51 Loye, Dominique 23

172   Index Mines Action Canada 57 Mozambique 24, 53–4, 91, 103 Nergaard, Per 76 Niger 92, 96 Nobel Peace Prize 30, 62 non-governmental organisation (NGO) vii, 4–5, 10, 44–6, 49, 52, 54–7, 59, 71, 73–4, 93, 98, 102, 117, 127, 136–7, 141n34 norms: anti-APM norm 2, 11, 45, 50, 60–3, 79, 83, 91, 98, 102–3, 118, 121, 128, 132, 134–5; norm bandwagon 11, 20, 36, 45, 114, 128–9, 132; stigmatisation 4, 11, 18, 27, 37–8, 40, 55–6, 60–1, 99, 128, 130, 135–6 North Atlantic Treaty Organization (NATO) 50–1, 94, 110, 114–15, 118, 120, 137 Norway 19, 52–3, 95–7, 115, 118, 131 Norwegian Peoples Aid (NPA) 56, 76 Organization of African Unity (OAU) 54 Organization of American States (OAS) 54, 114 Organization for the Prohibition of Chemical Weapons (OPCW) 25, 39 Oslo 5, 11, 45–6, 50, 53–4, 58, 62, 137 Ottawa 5, 44–6, 53 Ottawa Convention see mine ban treaty Ottawa Process viii, 3, 5, 7, 41, 49–63, 67, 72–3, 78–81, 84–5, 87, 93, 99–100, 103, 118, 121, 126, 128, 130, 134–6 Pakistan 5, 50–1, 77, 82, 107, 130 Pakistan Ordnance Factories (POF) 77, 82–3, 130 Peace Movement 27, 29–31, 33, 38–9 peacebuilding 76, 103, 135 peacekeeping 103, 118 Price, Richard 54, 60, 62, 73, 137 Princess Diana 11, 61–2 Prokosch, Eric 26, 47, 53, 90 Putnam, Richard 7, 10, 36, 61, 129 regime: clustering 4, 6, 10, 13, 45, 58, 63, 72, 76, 93, 106, 115, 121, 124, 126–7, 134–6; consequences 9, 12; emulation effects 50, 73, 80, 98, 104, 117–18, 127, 129, 131, 134–5, 137; interplay 4, 7, 11, 38, 52, 83, 85, 87, 94, 106, 117–8, 124, 128–30, 133, 135; learning 7, 9, 12, 57, 63, 88, 121, 124–7, 134; nesting 3–4, 7, 9, 11, 14, 17, 25, 38, 41, 57, 60, 62, 119,

123, 129–30, 135; network effects 46, 72, 74, 85, 104, 122, 149n27; spillover effects 4, 9, 11, 27, 50, 82, 103, 130, 135–7 regime theory viii, 1–2, 4, 6–7, 13, 107, 130, 138, 140n14, 140n17 resources 64, 76, 78–80, 85, 95–7, 100–2, 104–5, 109, 112–15, 121, 131–2 review conference 4, 26, 43–4, 51, 55, 68–9, 71, 73, 82, 91, 101, 137 Roberts, Adam 25 Russia 5, 18–19, 27–8, 35, 50–1, 67, 77, 93–4, 98–9, 101, 107, 112–13, 116–18, 129, 133 Rutherford, Ken 38 Second World War 4, 11, 36–7, 48, 66 security sector reform (SSR) 103 Selebi, Jacob 53 Sikkink, Kathryn 9, 11 small arms and light weapons (SALW) 135–6 South Africa 38, 52–4, 62, 75, 128 South Korea 50, 107, 116–17 Southern African Development Community (SADC) 91 Soviet Union 24, 52, 53, 90, 110, 114 Sri Lanka 82, 103 stockpile destruction 15, 65–6, 68, 90–1, 106–22, 124–5, 129, 131–3 Stott, Noel 62 Sweden 19, 42, 52 Switzerland 19, 52–3, 78–9, 96 Tomlin, Brian 5 Turkey 27, 50, 77, 82, 91, 107 Turkmenistan 110–11, 121–2 Ukraine 82, 107, 114–15, 120–1, 131 Underdal, Arild 8, 85, 131 unexploded ordnance (UXO) 64, 101 United Kingdom vii, 11, 23, 35–6, 40, 50, 61, 77–8, 89, 101–2, 118–20, 128, 130, 134 United Nations 25, 43, 65, 68–9, 74, 76–7, 88–9, 93, 95, 100, 103, 111, 117, 119–20, 135 United Nations Development Programme (UNDP) 115, 119 United Nations General Assembly (UNGA) 24, 35, 42, 45, 51, 56, 61, 136 United Nations Mine Action Service (UNMAS) 67, 74, 77, 102, 111, 117 United States (US) 5, 20–22, 24, 27, 29,

Index   173 31–2, 39, 43, 58–9, 61, 63, 67, 77, 79–80, 98–100, 107, 110, 116–19, 128–30, 132, 136 US Department of Defense (DoD) 50, 153n54 US Department of State 34, 37, 50 US policy 31–36, 40, 44, 46, 50–1, 98, 116–17, 128–9 US Senate 31–2, 35, 51 US War Department 31, 34

Venezuela 92 verification 7, 15, 24–5, 33, 39, 45, 49, 54, 67–70, 75, 93–5, 99, 111–13, 115, 125 victim assistance 61, 65–6, 68, 132 weapons of mass destruction (WMD) 25 Williams, Jody 44, 56, 62 World Bank 57, 78–9 Young, Oran 8–9, 11–12, 58

Taylor & Francis

eBooks FOR

liBRARIES

Over 23,000 eBook titles in the Humanities, Social Sciences, STM and Law from some of the world's leading imprints.

Choose from a range of subject packages or create your own! ~ Free MARC records

~ COUNTER-compliant usage statistics ~ Flexible purchase and pricing options ~ Off-site, anytime access via Athens or referring URL

~ Print or copy pages or chapters

~ Full content search ~ Bookmark, highlight and annotate text ~ Access to thousands of pages of quality research

at the click of a button For more information, pricing enquiries or to order a free trial, contact your local online sales team. UK and Rest of World: online.sales®tandf.co.uk US, Canada and Latin America: e-reference®taylorandfrancis.com

www.ebooksubscriptions.com