War and State-Building in Afghanistan: Historical and Modern Perspectives 9781472572172, 9781350012561, 9781472572189

Table of contents :
Cover
Half-title
Title
Copyright
Contents
List of Maps
Notes on the Contributors
Preface
Acknowledgements
List of Abbreviations
Introduction: Armies, Warfare and the State in Afghanistan from Pre-modernTimes to the Present Era
1 Continuity and Change in Asymmetric Warfare in Afghanistan: From the Mughals to the Americans
Introduction
The ability to hurt vs. the ability to hide
Hiding and hurting: geography
Willingness to fight
Time preferences
Asymmetric war in Afghanistan: from the Mughals to the Americans
Conclusion: Afghanistan, the end of the beginning or the beginning of the end?
Acknowledgements
Notes
2 Great Mughals, Warfare and COIN in Afghanistan, 1520–1707
Introduction
Beginning of Mughal intervention in Afghanistan: from Babur to Humayun, 1504–56
Pacification and consolidation of Mughal rule in Afghanistan from Akbar to Aurangzeb: 1556–1707
Conclusion
Notes
3 Counterinsurgency and Empire: The British Experience with Afghanistan and the North-west Frontier, 1838–1947
Afghanistan and counterinsurgency
Power and perception
Invading Afghanistan
Influencing Afghanistan
On the north-westfrontier: ideology, institution and insurgency
Air control and control of the air
Frontier arithmetic
Notes
4 The Conflict of War and Politics in the Soviet Intervention into Afghanistan, 1979–89
Introduction
The war that the Soviet generals had never expected to fight
A war to start all wars
Conclusions
Notes
5 Al-Qaeda versus Najibullah: Revisiting the Role of Foreign Fighters in the Battles of Jalalabad and Khost, 1989–92
Introduction
Background
The Soviet withdrawal and its aftermath
The Battle of Jalalabad
The role of Arabs in the Battle of Jalalabad
Consequences of the defeat
The Battle of Khost
The role of Arabs in the Battle of Khost
Conclusion
Acknowledgement
Notes
6 The Afghan National Army and COIN: Past, Present and Future Reconsidered
Introduction
The Afghan National Army at present and COIN
The early history of the Afghan Army
The Afghan Army of the 1860s and 1870s
Servants of the dynasty: the Afghan Army, 1881–1901
The decay of the army
The Democratic Republic of Afghanistan Army (DRAA)
The Afghan National Army after 2001
Conclusion
Notes
7 Revising COIN: The Stakeholder-centric Approach
Introduction
COIN in theory
What is an insurgency?
What is counterinsurgency?
COIN and its critics
Towards a revised COIN theory: a stakeholder-centric approach
Conclusion
Notes
8 The Country as a Whole: Imagined States and the Failure of COIN in Afghanistan
Introduction
Arguments
The priority of the ‘state’ in Afghanistan
Public servants in Afghanistan
The USSR in Afghanistan, 1978–89
Operation Enduring Freedom
Conclusions
Notes
9 Heart or Periphery? Afghanistan’s Complex Neighbourhood Relations
Introduction
Regional Security Complexes
Positioning Afghanistan
South Asia and Afghanistan
The Persian Gulf and Afghanistan
Central Asia and Afghanistan
Competing policy visions
Conclusion
Notes
Select Bibliography
Glossary
Index

Citation preview

War and State-Building in Afghanistan

Bloomsbury Studies in Military History Series Editor: Jeremy Black Bloomsbury Studies in Military History offers up-­to-date, scholarly accounts of war and military history. Unrestricted by period or geography, the series aims to provide freestanding works that are attuned to conceptual and historiographical developments in the field while being based on original scholarship. Published: The 56th Infantry Brigade and D-Day, Andrew Holborn (2010) The RAF’s French Foreign Legion, G.H. Bennett (2011) Empire and Military Revolution in Eastern Europe, Brian Davies (2011) Reinventing Warfare 1914–1918, Anthony Saunders (2011) Fratricide in Battle, Charles Kirke (2012) The Army in British India, Kaushik Roy (2012) The 1711 Expedition to Quebec, Adam Lyons (2013) Britain, Germany and the Battle of the Atlantic, Dennis Haslop (2013) Military Transition in Early Modern Asia, 1400–1750, Kaushik Roy (2014) The Role of the Royal Navy in South America, Jon Wise (2014) Scotland and the British Army 1700–1750, Victoria Henshaw (2014) Forthcoming: Conflict and Soldiers’ Literature in Early Modern Europe, Paul Scannell (2015) Postwar Japan as a Sea Power, Alessio Patalano (2015) Youth, Heroism and Naval Propaganda, Douglas Ronald (2015) Reassessing the British Way in Warfare, Keith McLay (2015) The D-Day Landing on Gold Beach, Andrew Holborn (2015) Australasian Propaganda and the Vietnam War, Caroline Page (2015) William Howe and the American War of Independence, David Smith (2015) Australian Soldiers in the Boer and Vietnam Wars, Effie Karageorgos (2016)

War and State-Building in Afghanistan Historical and Modern Perspectives Edited by Scott Gates and Kaushik Roy

Bloomsbury Academic An imprint of Bloomsbury Publishing Plc

Bloomsbury Academic An imprint of Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

1385 Broadway New York NY 10018 USA

www.bloomsbury.com BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2015 Paperback edition first published 2016 © Scott Gates, Kaushik Roy and Contributors, 2015 Scott Gates and Kaushik Roy have asserted their right under the Copyright, Designs and Patents Act, 1988, to be identified as the Editors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. No responsibility for loss caused to any individual or organization acting on or refraining from action as a result of the material in this publication can be accepted by Bloomsbury or the authors. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: HB: 978-1-4725-7217-2 PB: 978-1-4742-8635-0 ePDF: 978-1-4725-7218-9 ePub: 978-1-4725-7219-6 Library of Congress Cataloging-in-Publication Data War and state-building in Afghanistan : historical and modern perspectives / edited by Scott Gates and Kaushik Roy. pages cm. – (Bloomsbury studies in military history) Includes bibliographical references and index. ISBN 978-1-4725-7217-2 (hardback : alk. paper) – ISBN 978-1-4725-7218-9 (epdf) – ISBN 978-1-4725-7219-6 (epub) 1. Afghanistan–History, Military. 2. Nation-building–Afghanistan. I. Gates, Scott, editor, author. II. Roy, Kaushik, 1971- editor, author. DS357.5.W37 2014 355.009581–dc23 2014019626 Series: Bloomsbury Studies in Military History Typeset by RefineCatch Limited, Bungay, Suffolk Printed and bound in Great Britain

Contents List of Maps Notes on the Contributors Preface Acknowledgements List of Abbreviations

Introduction: Armies, Warfare and the State in Afghanistan from Pre-­modern Times to the Present Era  Scott Gates and Kaushik Roy 1

2 3 4 5 6 7 8 9

Continuity and Change in Asymmetric Warfare in Afghanistan: From the Mughals to the Americans  Scott Gates, Kaushik Roy, Marianne Dahl and Håvard Mokleiv Nygård Great Mughals, Warfare and COIN in Afghanistan, 1520–1707  Kaushik Roy Counterinsurgency and Empire: The British Experience with Afghanistan and the North-west Frontier, 1838–1947  John Ferris The Conflict of War and Politics in the Soviet Intervention into Afghanistan, 1979–89  Pavel K. Baev Al-Qaeda versus Najibullah: Revisiting the Role of Foreign Fighters in the Battles of Jalalabad and Khost, 1989–92  Anne Stenersen The Afghan National Army and COIN: Past, Present and Future Reconsidered  Rob Johnson Revising COIN: The Stakeholder-centric Approach  Erik Reichborn-Kjennerud, Karsten Friis and Harald Håvoll The Country as a Whole: Imagined States and the Failure of COIN in Afghanistan  Ivan Arreguín-Toft Heart or Periphery? Afghanistan’s Complex Neighbourhood Relations  Kristian Berg Harpviken

Select Bibliography Glossary Index

vi vii x xi xii

1

21 43 79 113 131 147 191 221 245 281 291 293

List of Maps Map 1 Map 2 Map 3 Map 4

The Safavid and Mughal Empires Soviet USSR Afghanistan 2007 Mughal–Safavid Uzbek Great Game in Afghanistan, c. 1550–1700 Borders Map 5 Theatre of Yusufzai Uprising, 1585–6 Map 6 Afghanistan and its Neighbourhood

30 33 34 44 45 246

Notes on the Contributors Ivan Arreguín-Toft is currently a Fellow at the Oxford Martin School at the University of Oxford, where he also serves as co-Chair of Dimension One (cyber security policy and strategy) of Oxford’s Global Cyber Security Capacity Building Centre. He is also Departmental Lecturer in Public Policy at the Blavatnik School of Government, University of Oxford, while on leave from his primary post as Assistant Professor of International Relations at Boston University. Arreguín-Toft is an authority on asymmetric politics and conflict (insurgency–counterinsurgency and terrorism; cyber politics; gender politics; and Russian foreign policy), as well as strategy, strategic interaction and human security. Pavel K. Baev is a Research Director and Professor at the Peace Research Institute Oslo. He is also associated with the Brookings Institution (Washington, DC) and French Institute for International Affairs (IFRI, Paris). Support for his research on security matters from the Norwegian Defence Ministry is greatly appreciated. Marianne Dahl is a Doctoral Researcher at the Peace Research Institute Oslo. She works on topics related to non-­violent and violent conflict, authoritarian stability and democratic change. Her work has appeared in the Journal of Peace Research. John Ferris is a Professor of History at the University of Calgary. He has published several books and more than 60 academic articles and chapters in books on diplomatic, intelligence and military history, as well as contemporary strategy and intelligence. His work has appeared in Diplomacy and Statecraft Historical Journal, Intelligence and National Security, International History Review, Journal of Military History and Journal of Strategic Studies. Karsten Friis is Head of the Security and Defence Research Group at the Norwegian Institute of International Affairs (NUPI). He holds the equivalent of an MPhil in Political Science from the University of Oslo and an MSc in International Relations from the London School of Economics.

viii

Notes on the Contributors

Scott Gates is a Research Professor at the Peace Research Institute Oslo and Professor of Political Science at the Norwegian University of Science and Technology. He has published nine books and more than 20 peer-­reviewed articles. He is an Associate Editor of the Journal of Peace Research and is the Editor in Chief of the International Area Studies Review. Gates is the 2014 co-­ recipient of the Herbert Simon Award for the scientific study of bureaucracy. He is a member of the Norwegian Academy of Science and Letters (DNVA) and the Royal Norwegian Society of Science and Letters (DKNVS). Gates’ current research interests include conflict dynamics and governance. Kristian Berg Harpviken is Director of the Peace Research Institute Oslo. He is a sociologist with particular competence on wartime migration, transnational movements and mobilization, the dynamics of civil war, and on peace processes at large. Harpviken is a frequent media commentator, and has published in a wide range of academic and policy outlets. His first monograph was Social Networks and Migration in Wartime Afghanistan (2009). Harald Håvoll is a retired Lieutenant-Colonel from the Norwegian Armed Forces and a former Head of the Armed Forces’ Joint Doctrine Centre. He is now a consultant and military analyst associated with the Norwegian Institute of International Affairs (NUPI) and Centre for International and Strategic Analysis (SISA) in Oslo. Håvoll is the author of the NUPI Security in Practice report no. 13 in 2008: ‘COIN Revisited – Lessons of the Classical Literature on Counterinsurgency and its Applicability to the Afghan Hybrid Insurgency’. Rob Johnson is the Director of the Oxford Changing Character of War Programme and Senior Research Fellow at Pembroke College, Oxford University. Specializing in the History of War, his principal fields of research are insurgency and counterinsurgency, strategy and conflicts in Asia and the Middle East. He is author of The Afghan Way of War (2011). Håvard Mokleiv Nygård is a Senior Researcher at the Peace Research Institute Oslo. He works on issues related to conflict, repression and democratization. Nygård’s research has appeared in International Studies Quarterly, International Area Studies Review, Journal of Conflict Resolution, World Development and Terrorism and Political Violence.

Notes on the Contributors

ix

Erik Reichborn-Kjennerud is a Researcher at the Norwegian Institute of International Affairs specializing in military theory and strategic thinking. He holds an MA in Security Policy Studies from the George Washington University. Kaushik Roy is Guru Nanak Professor at the Department of History, Jadavpur University, Kolkata, India and Global Fellow, Peace Research Institute Oslo (PRIO), Norway. He has written and edited 24 books and more than 50 articles in peer-­reviewed journals and edited volumes. Anne Stenersen (MPhil, PhD) is a Research Fellow at the Norwegian Defence Research Establishment (FFI). She has studied militant Islamism since 2006, with particular emphasis on the Al-Qaeda network and militant groups in Afghanistan and Pakistan. Her most recent publications include a book chapter in Talibanistan: Negotiating the Borders between Terror, Politics, and Religion, edited by Peter Bergen (2013).

Preface Why another book on Afghanistan? The uniqueness and strength of this volume is its interdisciplinary appeal: it brings under one umbrella experts from the fields of history, sociology, political science, economics and international relations. The diversity of opinion among the contributors reflects the richness of this volume. The Introduction has two interrelated themes. On the one hand, it provides a summary of the essays; on the other, it shows how the contributors engage with the principal debates. Some of the chapters are historical and thus based on solid archival sources. The other chapters, by sociologists, political scientists and international relations experts tend to be more theoretical in focus and follow a social science approach. We do not know of any volume on Afghanistan with such broad methodological fusion. In fact, the first essay of this edited collection is a joint product of an economist/political scientist, two political scientists and a historian. Overall, the edited collection merges regional/area studies with global studies approaches by fusing local studies against a backdrop of an international studies paradigm. Such an approach opens new vistas in the study of Afghanistan in particular and other regions in general. We hope that the select bibliography will aid further research. Scott Gates, Oslo, and Kaushik Roy, Kolkata, 2014

Acknowledgements The editors want to thank PRIO, which served as the host institution for this project. The Norwegian Ministry of Defence by funding the projects entitled ‘COIN in Afghanistan’ in 2012 and on ‘the Future of Warfare’ in 2013 significantly aided our research on Afghanistan. The Norwegian Ministry of Defence also provided support for the workshop, ‘COIN in Afghanistan: From the Mughals to the Americans’, held at Oslo, Norway during 12–13 February 2012, Oslo, Norway. Inputs from the participants at the conference significantly improved this collection. Some of the papers of the conference have been published in a special issue of International Area Studies Review (‘Statebuilding and War in Afghanistan’, Volume 15, Issue 3), published by Sage. Special thanks to the Director of PRIO, Kristian Berg Harpviken, who supported the project and also presented a paper in the above-­mentioned seminar. The editors take this opportunity to thank all the participants who travelled all the way from Canada, India, USA and UK to attend the seminar and in the last two years painstakingly agreed to update their essays in the light of ongoing research. One scholar at PRIO, Marianne Dahl, deserves special mention. Besides her intellectual contribution, she provided essential editorial and logistical inputs that to a great extent sustained this project. We also thank Andreas Forø Tollefsen for help in making the maps used in this volume. Thanks to Claire and Emma of Bloomsbury for showing interest in publishing this volume. Lastly, special thanks to Professor Jeremy Black for his enthusiasm and support for this project. Inputs from the three referees of Bloomsbury are also duly acknowledged by the two editors.

List of Abbreviations AFV ANA ANP ANSF APC CIA CIS COIN DRA DRAA EU FATA GOI HN IED ISAF ISI MOD MTT NATO NCO NCS OEF OMLT PDPA PRIO PRT RAF RSC SCO SOF UN

Armoured Fighting Vehicle Afghan National Army Afghan National Police Afghan National Security Forces Armoured Personnel Carrier Central Intelligence Agency of USA Commonwealth of Independent States Counterinsurgency Democratic Republic of Afghanistan Democratic Republic of Afghanistan Army European Union Federally Administered Tribal Areas in Pakistan British Government of India Host Nation Improvised Explosive Device International Security and Assistance Force Inter-Services Intelligence of Pakistan Ministry of Defence Military Training Team North Atlantic Treaty Organization Non Commissioned Officer National Commanders’ Shura Operation Enduring Freedom Operational Mentoring and Liaison Team People’s Democratic Party of Afghanistan Peace Research Institute Oslo Provincial Reconstruction Team Royal Air Force Regional Security Complex Shanghai Cooperation Organization Special Operation Force United Nations

Introduction: Armies, Warfare and the State in Afghanistan from Pre-­modern Times to the Present Era Scott Gates and Kaushik Roy

The Mughals, British and Soviets all failed to subjugate Afghanistan. These failures offer valuable lessons for today. While military technology has changed, the physical geography, social and cultural characteristics have not. These unchanging factors limit the extent to which war and state-­building policies can be modified. This constancy over time makes the experiences of the past especially relevant for today. Taking a long historical perspective from 1520 to 2012, this edited volume examines the Mughal, British, Soviet and US-NATO efforts at counterinsurgency (COIN) and state-­building in Afghanistan. Our analysis begins in 1520 when Afghanistan serves as a pawn in the Great Game between Mughal India, Safavid Persia and the Uzbek Khanate of Central Asia. The volume ends in 2012, paying witness to the ongoing operations by the US and NATO forces in Afghanistan. Special emphasis is given to ecology, terrain and logistics to explain sub-­ conventional operations and state-­building in Afghanistan. Drawing on new archives and a synthesis of previous COIN experiences and state-­building literature, this edited collection addresses several debates on Afghanistan. The volume engages with three important debates in the field. First, whether COIN can be traced back to nineteenth-­century imperial campaigns in the colonies; second, how far the principle of ‘minimum force’ is applicable to the case of present-­day COIN in Afghanistan; and third, whether the stability–instability paradox in Afghanistan can be linked to the evolution of the ‘rentier’ state or not. Barnett R. Rubin writes that ‘locational’ rents during the height of the Cold War accelerated the expansion of state apparatus in Afghanistan after 1955. Foreign funding enabled the polity to be autonomous from society. Without capital accumulation, rent extraction and bargaining between the local power holders and the state’s elite, the rentier polity builds up an infrastructure due to

2

War and State-Building in Afghanistan

inflow of foreign aid. According to Rubin, a rentier polity exercises despotic power through arbitrary coercion, but is unable to develop infrastructural power. Infrastructural power is the capacity of the state to penetrate civil society and to implement decisions taken by the central government throughout the realm. Infrastructural power requires development of apparatuses and institutions for monitoring and surveillance rather than merely coercion. To sum up, a rentier state is not accountable to society and is incapable of penetrating and monitoring a non-­capitalist, non-­monetized society as in Afghanistan. In such cases, state officials might seek autonomy in their own corporate interests and their behaviour becomes predatory. Rubin notes that the rise of the rentier state in Afghanistan can be traced back to the end of the nineteenth century under Amir Abdul Rahman Khan (r. 1881–1901), who received largesse from the Government of India (GOI).1 The second chapter in this edited volume by Kaushik Roy argues that a sort of nascent rentier regime emerged in Afghanistan due to financial subsidy annually provided by the Mughals to the Pathan (Pashtun) tribes of Kabul and Kandahar regions. While most Western commentators agree that the origin of present-­day COIN can be traced back to nineteenth-­century imperial warfare by the Western powers in the colonies, the first two chapters in this edited collection strike a different note. In the first essay titled ‘Continuity and Change in Asymmetric Warfare in Afghanistan: From the Mughals to the Americans’, Scott Gates, Kaushik Roy, Marianne Dahl and Håvard Mokleiv Nygård argue that adaptations and innovations are the hallmarks of warfare. Weapons, logistics, transport and other forms of military technology change considerably from decade to decade. For every innovation comes the diffusion of the new technology, nullifying the advantage held by the first adaptor. A group lacking the means or the ability to exploit a new technology will adapt strategically to stay in the fight. This form of adaptation, the negation of technological advantage, is the essence of asymmetric warfare. This chapter theoretically examines the strategic dynamics of asymmetric warfare while drawing specifically on the case of Afghanistan. In Afghanistan, physical geography, cultural characteristics and historical factors have limited the extent to which insurgency and COIN policies can be modified. This had led to substantial continuity over time, which makes the experiences of the past especially relevant for today. Taking a long historical perspective from 1520 to 2011, this chapter examines the Mughal, British, Soviet and American/ NATO experiences with asymmetric warfare in Afghanistan. In Chapter 2, Kaushik Roy examines several aspects of British COIN that can be traced back to Mughal pacification policy in Afghanistan during the sixteenth

Introduction

3

and seventeenth centuries. In turn, present-­day COIN’s lineages can be traced back to nineteenth-­century British COIN. Hence, the seeds of modern-­day COIN may be traced back to one of the pre-­modern empires of Asia. For almost two centuries, the Mughals were engaged in both conventional war and unconventional operations in Afghanistan. Certain regions in which the Mughal forces operated became important for the British Indian units of the Raj, the Soviet armoured columns, and also for the NATO-US forces in the new millennium. Hence, a brief description of the physical geography of the theatre is necessary. For all the powers interested in intervening in Afghanistan from the medieval era to the present day, the theatre extends from the River Oxus in the north to the Arabian Sea in the south. The older names of Oxus were Jaihun and Amu Darya. The Oxus makes a curve for its great bend northwards east of Badakshan. Tirmiz was an important fortress north of Oxus under the ruler of Balkh. During 1646–7, when the Mughals attacked Nazar Muhammad, the Uzbek ruler of Balkh, the Mughals garrisoned Tirmiz.2 Badakshan included the districts of Andarab, Kulyab/Khutlan, Hissar/Gissar, and so on. Qunduz was the capital of Badakshan under the Uzbek Khanate. In the Shingan Mountains in Khutlan, rubies have been mined from medieval times. These ruby mines sustained many Afghan warlords during the 1980s. The important towns of Balkh are Andkhud, Shibarghan, Sar-­iPul, Fatehabad, and Astana-­i Imam (its modern name is Mazar-­i Sharif). The important districts of Balkh are Sriram and Gurzigan. Balhab is the principal river in Balkh. In order to cross Balhab, an invasion force (whether Mughal or Soviet) would have to move through the narrow valley of Darra Gaz. There are three stretches of desert within Balkh. From Bamian, the major route connects Kabul with Balkh. The route runs through Gumbazak/ Katar Sum Pass. The shortest route between Kabul and Balkh was through Khinjan and Ghori but it was extremely difficult owing to steep ascents and descents and narrow paths.3 Despite being Sunni Muslim themselves, the Mughals faced tough opposition from the Afghans, especially the Sunni Pathans (Pashtuns). Local clan based resistance to a centralizing power comprised the backbone of resistance in Afghanistan. This is not to argue that Islam has played no role in Afghan insurgency. Rather, the point we are making is that Islam in general and jihad in particular are not crucial components of Afghan opposition to external invaders in their homelands. From the medieval era, the control of Afghanistan represented a sort of ‘Great Game’ for controlling the heart of Eurasia. The Mughals, argues Roy in Chapter  2, were more successful than the British in

4

War and State-Building in Afghanistan

maintaining a permanent presence in Afghanistan. Nevertheless, the Mughals were never able to control western and north-­west Afghanistan. British control over Afghanistan during the nineteenth century was more transitory. Moreover, unlike the British the Mughals used Afghanistan as a springboard for projecting power in Central Asia. Road-­building, subsidies to the tribes, the construction of military outposts and forts, the establishment of thanas and military recruitment of several Afghan clans were the characteristic features of Mughal COIN. Certain Mughal COIN techniques remained operational even under the British Indian Empire. For instance, both the Mughals and the British through limited military recruitment tried to co-­opt the potentially rebellious manpower of Afghanistan. The Mughals realized that border management of the frontier tribes was inseparable from the related and greater Afghan ‘problem’. The Mughal technique of subsidies to placate the Afghan tribes continued in the British period. When the amount of subsidy was cut, rebellions broke out, which in turn cost the imperial powers much more. So, the seeds of formation of a rentier state in Afghanistan could be traced back to the seventeenth century. The Abdalis/Durranis and the Ghilzai/Gilzai confederacies comprise the two major components of the Pathans. After the Mughal collapse due to Nadir Shah’s invasion, Ahmad Khan, a member of the Saduzoi subdivision of the Abdalis, established a unified Afghan state in 1747. However, in the early nineteenth century, the Saduzoi ruling family was replaced by another subdivision of the Abdalis: Muhammadzai Barakzais. Under Ahmad Shah (r.  1747–72) and his successor Timur Shah (r.  1772–93), the Alikozai, Popalzai and Barakzai subdivisions of the Abdali tribe monopolized most of the power and privileges. The Alikozais became so powerful in Herat that after the reign of Timur Shah they established an independent emirate at Herat.4 In the early nineteenth century, the Alikozais numbered some 10,000 families. Around the same time, the Popalzais numbered to 12,000 families. Within the Popalzais, the Sadozai and Bamizai sub-­groups were most powerful, being closely related with Ahmad Shah due to genealogical links. The Muhammadzai sub-­group within the Barakzai subdivision numbered 4,000 to 5,000 families in the eighteenth century. But thanks to royal patronage, their numbers rose to 30,000 families during Amir Dost Muhammad Khan’s (b. 1792, d. 1863) second reign (1842–63). In order to balance the overtly powerful Durrani and Gilzai confederacies, both Ahmad Shah and Timur Shah maintained a non-­tribal bodyguard of outsiders: the Qizilbash of Persia/Iran. The term Qizilbash technically meant red head due to the red headgear worn by members of the Turcoman/Turkoman tribes who supported the Shia leader Shaikh Haider (d. 1488). The Qizilbash dominated

Introduction

5

Herat and Kandahar during the Safavid occupation of West and South Afghanistan. In return for tuyul grants, the Qizilbash tribes provided cavalry contingents to the Safavids. Nadir Shah and then the Saduzoi rulers pampered the Qizilbash. As a result, some 12,000 Qizilbash families were settled in Kabul. The policy is somewhat similar to Hamid Karzai depending on the western private security companies for furnishing his personal bodyguards. And when Dost Muhammad Khan took over as Amir of Afghanistan he drastically reduced the power of the Qizilbash and disbanded the special royal bodyguard corps. In the 1830s, there were only 1,000 Qizilbash sowars5 in Dost’s cavalry, which numbered 12,000 sowars. And during the First Anglo-Afghan War (1839–42), the Qizilbash supported the British puppet Shah Shuja.6 Thus, we see that the ‘state’ in Afghanistan represented a fragile fabric based on a complex network of tribes, sub-­tribes and clans whose positions and loyalties were continuously fluctuating with time and context – thereby hardly constituting a state at all. After the Mughals, the British colonial power from India intervened in Afghanistan. The first British intervention in Afghanistan came in the shape of First Anglo-Afghan War, when the British tried to put the Saduzoi expelled ruler Shah Shuja on the throne of Kabul.7 Thomas Rid emphasizes that present-­day US COIN has its lineages in nineteenth-­century French colonial warfare. Lieutenant-Colonel David Galula’s (b. 1919, d. 1967) writings function as a bridge between the French theorist and practitioners of COIN in nineteenth-­ century Afro-Asia and modern American practitioners of pacification. Galula had served in Algeria. Rid notes the influence of Galula’s writings in General David Petraeus’s COIN field manual. The three most important nineteenth-­ century French COIN practitioners were Marshals Thomas Robert Bugeaud (b. 1784, d. 1849), Gallieni (b. 1849, d. 1916) and Hubert Lyautey (b. 1854, d. 1925). These figures’ writings and actions influenced Galula’s conceptual framework. Gallieni emphasized that successful COIN requires a combination of political and military actions. Lyautey emphasized that a successful COIN doctrine must display suppleness, elasticity and adaptability in accordance with the requirements of time and context.8 In Chapter 3, John Ferris assesses British military operations and pacification during the nineteenth and twentieth centuries, in Afghanistan and the NorthWest Frontier of India. Like Christian Tripodi, Ferris does not overemphasize the cultural factors in his analysis of frontier management policies. He considers how far these actions can be understood through the modern terminology of ‘counter-­insurgency’, and challenges the universality of these concepts. Britain was driven to intervene in these territories because they were located on its

6

War and State-Building in Afghanistan

frontier, while British decision makers believed that threats or defeats could shake its hold over India. On two occasions Britain attempted to conquer Afghanistan, and to rule the country through puppets. It failed to achieve these ends, and instead turned to a more successful policy, which was to maintain influence in Afghanistan through politics, bribery and latent military power. Ferris argues that in Afghanistan, Britain rarely practiced anything like COIN. The limits to British success in that country, however, forced Britain to rule the north-­west frontier region, i.e. the territories between British India and Afghanistan (present-­day North-West Frontier Province and FATA in Pakistan). The British GOI did so through several means. In Baluchistan, it followed a policy of pacification, relying essentially on politics and financial aid to local leaders, with a minimal reliance on force. The GOI attempted to pursue similar practices among Pashtuns on the north-­west frontier, with mixed success, thus forcing the British in India to rely far more heavily on coercion. From 1920, the GOI also applied airpower to support pacification, though generally with restraint, for fear that inflicting civilian casualties would provoke attacks on British-­occupied regions in India. Through these means, Britain managed to establish a stalemate in these territories, which met its minimal needs there. The British experience demonstrates the history of Western pacification and COIN in these territories. It also shows how one can leave Afghanistan and still achieve one’s aims there, even after failing to master the country through force. Elsewhere, Tripodi writes that there were two partly divergent views within the British Empire about the Russian threat to Afghanistan. While the London Government saw Czarist Russia through the prism of a European balance of power paradigm, British India perceived Russia as encroaching on Afghanistan from its springboard in Central Asia. And any extension of Russian influence in Afghanistan, assumed the strategic managers of British India, would have a negative effect on the fragile imperial rule within the Indian subcontinent. To a large extent, Ferris agrees with Tripodi’s assertion. However, with the aid of hindsight, Tripodi continues that the British failures during the First and Second Anglo-Afghan Wars were actually a blessing in disguise. This was because conquest of Afghanistan would have required troops that Britain could not afford. Furthermore, such a gigantic task would have weakened British hold on India by denuding military reserves from India to barren and unproductive Afghanistan. In 1919, after gaining victory over the Afghans during the Third Anglo-Afghan War, British India displayed strategic maturity by following a ‘hands-­off approach’ towards Kabul. This was because the Russian threat had vanished and Afghanistan was only of marginal importance to British India’s

Introduction

7

security.9 Such strategic acumen was probably not present at least in the official mind of British imperialism in the second and third decades of the twentieth century. One could argue that financial stringency of the Raj plus financial problems of the London Government, demobilization of the bloated British Indian and British armies, requirements for frequent aid to civil duties within India, and rising nationalist pressure within the Indian subcontinent forced London and Delhi to go ‘soft’ vis-à-vis Afghanistan in late 1919.10 The traditional view is that the British COIN until this date had been most successful mainly because of the ‘minimum force’ philosophy. The other characteristics of British COIN were civil–military cooperation, the integration of civil and military intelligence, and a small unit approach to combat the irregulars. For instance, Raffi Gregorian writes that decentralized small unit jungle warfare patrols (a continuation of Tiger patrols of the British Army in Burma during the Second World War) and a focus on intelligence gathering enabled the British to defeat the Chinese guerrillas in Malaya during 1948 and 1951.11 Keith Jeffery agrees that in the history of modern British COIN, the Malayan model is the best. And the chief characteristic of the Malayan model is integration of police and military intelligence agencies under one roof, which prevented excessive militarization of the pacification policy.12 Tim Jones notes that the British Army is a learning organization. For instance, the counter-­ guerrilla patrolling based on small units emerged from the manuals issued in the Burma theatre in 1944 and reached its logical culmination in the mid-1950s in Malaya.13 In 1954, writes Randall W. Heather, when the British Army gave up its large unit sweeps in favour of decentralized small unit patrols by junior commanders and revamped the intelligence gathering organization, the Mau Mau insurgency in Kenya was crushed.14 In contrast, Soviet/Russian COIN from the nineteenth century to the present day is regarded as most brutal. In between these two extreme poles lie the American and French COIN. French pacification policy in Algeria and IndoChina, runs the legend, failed because of French military officers’ failure to accept political subordination. Being firepower heavy (due to techno-­savvy American culture), US COIN creates considerable collateral casualties, which ultimately defeats the policy of winning over the people inhabiting the disturbed regions. The US Army focused on ‘find, fix and destroy’ the insurgents and neglected the broader political dimension. Again, the US failed to raise, train and utilize properly indigenous police in a systematically efficient manner. Both the US and the Soviet/Russian armies used large unit sweeps against the nimble irregulars, which is like using a sledgehammer to kill a fly. Dissenting voices within the US

8

War and State-Building in Afghanistan

military organization like the Marines in Vietnam and David Petraeus’s lobbying during the Iraq and Afghan ‘surge’ advocated an alternate pacification policy based on small unit patrols and winning the ‘hearts and minds’ of the populace (note the similarities with the traditional interpretation of British COIN). And when such alternate COIN policies have been followed, the results have been good.15 In general accounts, the Russian Army is treated as something worse than the Bourbons. While the Bourbons learnt nothing, forgot nothing, the Russian military commanders seem to have been operating on the principle of learnt nothing but forgot everything. From Kabul to Grozny, the Russian generals were bent to refight the Second World War. With mechanized forces, heavy artillery and air strikes, the Russians conducted a sort of ‘Battle of Berlin’ both in Kabul and Grozny.16 A re-­evaluation of the traditional picture is already underway, however. In recent times, both the minimum force philosophy and effectiveness of British COIN have come under attack. In his study of British COIN in Palestine during 1945–47, David A. Charters claims that the British Army’s failure to procure real time intelligence about the Jewish guerrillas and terrorists were the principal limitations. The British Army’s organizational culture was against giving top priority to the collection and dissemination of intelligence. Both in Palestine and in Kenya (like the US Army in Vietnam), the British security personnel suffered from linguistic inability to deal with the locals.17 By making a case study of British COIN in Kenya between 1952 and 1956, British scholar Huw Bennett asserts that instead of minimum force, the British Army used ‘exemplary force’ against the Kikuyus. A deliberate policy of beatings, tortures, murders and forced population movement was introduced to force the Kikuyu civilian population to support the government. Inducing fear among the civilian population was deployed as a strategic instrument. Repression based on indiscriminate use of force was considered the best way of dealing with the Mau Mau insurgency. Mistreatment of the civilian population by the British soldiers was not considered a crime by the British Army. And the policy of terror actually succeeded in alienating the Mau Mau insurgents from the Kikuyu civilian population. The former were driven into the jungle where the hunter-­killer squads of the security forces hunted them down.18 Similarly, David A. Percox concludes that brute repression and violent policing actions rather than any ‘hearts and minds’ strategy resulted in the collapse of the insurgency in Kenya in 1956.19 Nevertheless, Ferris’s chapter in this collection accepts that the British COIN along the north-­ west frontier during the late nineteenth and early twentieth century was characterized by a minimum force philosophy. In recent times, British COIN in

Introduction

9

Afghanistan was not that stellar. In 2006, the British took charge of Helmand. Several villages were razed due to massive use of firepower by the British soldiers. The British troops relied on firepower to kill the enemy from a distance. In a COIN task, the British used Multiple Launch Rocket System and Javelin Surface to Air Missiles. Like the US and other NATO troops, the British in Afghanistan also called for close air support, which caused further unnecessary civilian deaths.20 In Chapter  4, Pavel Baev challenges the established perspective of Soviet COIN being ineffective. He argues that despite the traditional accounts, Soviet COIN against the mujahideen had been quite successful at the operational– tactical levels. This was despite fewer Soviet troops being deployed compared with the US and NATO troops in Afghanistan in the first decade of the new millennium, and the mujahideen were better trained and better equipped than the present neo-Taliban. The Soviets were let down, however, by the grand strategy of Moscow, which opposed deploying the required numbers of military assets in Afghanistan. Baev notes that the self-­propelling dynamics of violence in Afghanistan, which appeared set to outlast the as yet ongoing peace-­making, is rooted in the impact of the Soviet intervention, in which fighting was only one element of the complex political drama of destruction of the Afghan state. The intervention was launched in response to the escalation of domestic crisis in Afghanistan, about which the Soviet leadership knew much but understood little. The Soviet Army showed the capacity for learning but the improved tactical skills and upgraded operations brought only greater destruction, which was counter-­productive in the absence of a coherent strategy and turned out to be politically unsustainable. No retrospective analysis can establish with any certainty whether the war had a ‘military solution’ or not, but it is quite clear that the USSR in its autumnal decade had neither Stalinist determination nor Leninist ingenuity to find one. The Soviet military machine was not over-­burdened by the peripheral war and could have absorbed the defeat, but the consequences of the mujahideen victory for Afghanistan were truly devastating. The US helped to mobilize the most aggressively radical forces and supported their anti-­modernization agenda, assuming that the ‘black hole’ that would emerge after the Soviet withdrawal would be an isolated problem of no global significance. Soviet COIN in Afghanistan by all means had been brutal. And both Baev and Ivan Arreguin-Toft substantiate this. By 1990, there were 1.3 million dead Afghans (pre-­war population was 15.5 million) and 3.5 million had become refugees (mostly in Pakistan and to a lesser extent in Iran). The Red Army suffered less

10

War and State-Building in Afghanistan

than 15,000 dead and some 51,000 wounded.21 While Baev asserts that Soviet strategy prevented further inflow of military assets to Afghanistan, which in turn prevented complete pacification, Ivan Arreguin-Toft writes that lack of adequate infantry on the part of the Soviets prevented them from sealing the porous AufPak border, which helped keep the ‘Afghan pot’ boiling at the right temperature.22 Once the Soviets withdrew from Afghanistan in 1989, the United States lost all interest in that poor landlocked country. This was mostly because Washington viewed the ‘Afghan problem’ through the bipolar lens of systemic struggle against the good and bad (a product of Ronald Reagan’s Manichean world-­view). The mujahideen left to themselves with tacit support of Pakistan’s ISI fought against Dr. Najibullah’s regime and also fought among themselves. In such a context, Pakistan backed the Taliban (a group of radical Islamists) and Osama bin Laden established his base with its network of training camps in Afghanistan. In Chapter 5, Anne Stenersen covers the war between Afghan mujahideen and Najibullah’s Communist government between 1989 and 1992. She analyses the mujahideen’s war fighting strategies and the Afghan Government’s responses to those strategies. Four cases in particular are examined, namely, the battles of Jalalabad, Khost, Herat and Kabul. Each of the cases represents different fighting strategies on part of the mujahideen. In the Battle of Jalalabad in 1989, the mujahideen sought to conquer a major population centre by way of conventional warfare, but failed drastically. In contrast, the city of Khost was captured through the successful use of military force two years later. Herat was captured through negotiations, while Kabul fell due to a complex set of reasons including attritional warfare and internal weaknesses in the Afghan Government. The chapter concludes that in spite of their much-­publicized victory at Khost in 1991, the main reason for the mujahideen’s successes was not their military strength, but their patience and ability to exploit internal weaknesses in the Afghan Government. The case study provides examples of how local guerrillas confronted the armed forces of their own country to oust what they viewed as a hostile ‘puppet regime’. Like the mujahideen in 1989, the Taliban guerrillas today are likely to include elements that will continue to fight against the Afghan regime after international forces leave in 2014. The Afghan War of 1989–92 may therefore contain important lessons for the future. From the first decade of the first millennium, the United States attempted an ‘Afghanization’ of the war in Afghanistan, a process somewhat similar to ‘Vietnamization ‘of the Vietnam War from the late 1960s. Afghanization of the Afghan War involved building up the Afghan National Army (ANA) and Afghan National Police (ANP). At the beginning of 2007, the ANP trained and equipped

Introduction

11

by US and ISAF forces numbered about 36,000 men. By 2009, they had expanded to 80,000 personnel and the plan was to expand them further to 135,000 personnel. The ANP numbered 70,000 in early 2009, of whom some 50,000 were trained and equipped.23 In Chapter 6, Rob Johnson traces the historical evolution of the Afghan Army from late the nineteenth century, its approach to internal security and war fighting, and suggests how these legacies would influence the ANA’s approach to COIN. Johnson writes that Western strategy towards the end of their campaign in Afghanistan envisaged a steady drawing in of ANA numbers after 2014 as the combined effect of three years of operations by American ‘surge’ troops and an enlarged ANSF suppressed and exhausted the insurgents. American and other ISAF forces are to be the screen behind which the ANA can be built up rapidly and developed for its role of providing internal security. However, despite a rapid expansion in numbers of Afghan troops, there are signs that the general scheme is in jeopardy. Indeed, in 2012, many insurgent groups appeared to be as defiant as ever, arguing that they need only wait until the withdrawal of ISAF before launching a more significant offensive against the inexperienced and less well-­ equipped ANA and the insecure Afghan Government. Nevertheless, the priorities for Western commanders are the capability of the ANA, its effectiveness in combat, leadership development, and retention through improvements in the terms and conditions of service. Embedded ISAF personnel, known as OMLTs or MTTs, initially assisted the ANA by operating alongside and within small Afghan units. These were gradually morphed into larger units, provided ‘overwatch’, and attempted to move into a training role. Yet, Afghan military personnel expressed concern that they lacked the equipment to confront their ‘real’ enemy, namely Pakistan, and hinted that their own approach to countering insurgency might take a radically different form to the Western armies. Being capital-­intensive, the Western armies lack adequate infantry to conduct ground patrolling of the disturbed zones. And an adequate number of ‘boots on the ground’ is a necessary prerequisite of COIN. Due to lack of foot soldiers, the Western armies depend on long-­distance artillery, close air support, APCs, AFVs and high-­tech weapons for short-­range fire like mortars. Instead of manpower, they resort to firepower. The net result is massive collateral casualties, which instead of winning the ‘hearts and minds’ of the populace further alienates them. This is evident in Afghanistan as well as in Iraq. Despite Petraeus’s surge, neither the NATO nor the US Army could swamp these two countries with infantry for a long period. Inadequate manpower also dogged the Soviet 40th Army in Afghanistan in the early 1980s.

12

War and State-Building in Afghanistan

It could be argued that the so-­called British success in post-Second World War COIN operations was because the process involved major political concessions: decolonization and dissolution of the overseas empire.24 Such concessions took the wind out of the sail of the insurgents and would-­be terrorists. To apply such an approach in Afghanistan would mean leaving the country, including the Karzai government, at the mercy of the neo-Taliban and a resurgent Al-Qaeda. What, then, are the alternatives? Western COIN theories are most likely at fault.25 US-NATO COIN has not been successful in Afghanistan. Does that mean that insurgencies are unwinnable and the new millennium belongs to the non-­state actors? Has COIN both in theory and practice reached a dead end? In Chapter  7, Erik ReichbornKjennerud, Karsten Friis and Harald Håvoll offer an alternative to this pessimistic picture. They start by making a critical estimation of the US FM 3–24. They assess the existing COIN theories and point out that both the population-­centric and enemy-­centric approaches to COIN have their own strengths and limitations. Such a division in the COIN theories, they rightly point out, is actually a red herring. Rather, they campaign for a holistic approach. The stakeholder approach to COIN as propounded by them is characterized by its focus on peace-­building and state-­building. The authors warn that the Western concept of a monolithic unified state with a strong central authority might be unworkable in Afghanistan with its historical tradition of weak centre, strong periphery and dynamic clan connections. Their warning is similar to the point raised by Ivan Arreguin-Toft in Chapter  8 that Western ideas about post-Westphalian state structure is unworkable in the highly fragmented Afghan society with its historical tradition of weak state structure. More importantly, such an approach highlights that greater focus is required on how to maintain the peace and what to do after fighting ends. Limited by Eurocentricism, most COIN studies focus on the Western armies alone. They fail to see alternative models in non-Western settings, which offer valuable lessons. COIN theorists, for example, forget the Indian Army, which is manpower-­intensive. Thanks to India’s demographic resources, its army can afford a significant number of ‘body bags’ returning without political upheavals. Moreover, the Indian Army since 1947 has been conducting COIN in a region much bigger than Afghanistan and Iraq combined. Furthermore, the Indian Army conducts COIN in its domestic space unlike the US and NATO, which carries out peacekeeping operations on foreign soil (except the British Army in Ireland). Hence, due to media pressure and political sensibilities, the Indian Army has to be extra careful about collateral casualties. The Indian troops are

Introduction

13

only given handheld firearms to combat the insurgents. They are not given artillery and air support while countering the insurgents.26 To a great extent, the Indian Army’s COIN can be traced back to the British Indian Army’s (not the British Army’s) tradition of Small War. During the nineteenth and early twentieth century, the British Indian Army was one step ahead of the British units while conducting Small War/COIN along the North-­West Frontier.27 As a reminder to the Western policy-­makers, India’s COIN has relevance for other regions and other powers. Jugdeep S. Chima compares and contrasts India’s successful COIN in Punjab during the 1980s with the US and NATO’s unsuccessful COIN in Iraq. In the early 1990s, about 120,000 Indian Army personnel and 160,000 security personnel were deployed in Punjab (19,445 square miles). As a point of comparison, in 2007, there were 160,000 US troops and 130,000 Iraqi soldiers and between 79,000 to 140,000 Iraqi police personnel in Iraq (168,743 square miles). Size and numbers do matter. The Indian state could actually swamp the disturbed zone with army and security personnel on a long-­term basis. As a result, by the mid 1990s, after the death of some 25,000 people, the insurgency in Punjab died down.28 Generally, Western COIN theorists accept that a successful COIN requires ten to one numerical superiority over the insurgents.29 The territory of Afghanistan includes 245,000 square miles.30 Let us have a look at the Soviet and US troop deployment in Afghanistan. In 1980, the Soviet 40th Army had 81,800 personnel (of whom 61,800 were in the combat branches). In 2006, the ISAF pledged the deployment of 31,000 soldiers provided by 35 countries. At the end of 2007, there were 25,000 US troops and by January 2009, the numbers rose to 36,000. The US administration’s attempt to cut back the size of its contingent to 17,000 troops failed due to the rise of neo-Taliban in South Afghanistan and along the Auf-Pak border.31 How far is cultural understanding of the insurgents necessary to implement a successful COIN operation that ultimately would aid in stable state-­building? In a 2008 article, Tripodi presents a case study of British India’s North-­West Frontier. Through the agency of political officers who interacted with the tribal sirdars, the British attempted to acquire cultural awareness of the mosaic of tribes inhabiting the western bank of the River Indus. Whether the British political officers followed good or bad anthropology, writes Tripodi, is irrelevant as regards the prosecution of a successful state-­building enterprise along the North-­West Frontier of India. Financial problems, distrust and tension between the civilian and military officers of the Raj and a quest for immediate results ultimately shaped British India’s policy towards the North-­West Frontier. And this created a climate in which the non-­military options that required time and

14

War and State-Building in Afghanistan

money to generate successful results in the long run were marginalized by military options, which at least offered the chimera of subduing unrest in the immediate context. The net result was militarization of British India’s COIN.32 Militarization of COIN is a characteristic that has continued in state-­building enterprises in turbulent frontier regions from the nineteenth century to the present day. Hence, the contributors to this edited volume focus mostly on the military aspects of COIN and state-­building. Of the great powers that have conducted COIN in Afghanistan, the exceptional case would be the Mughals. The Mughals, Safavid and the Uzbeks who attempted to control Afghanistan as part of the pre-­modern ‘Great Game’ realized the long-­ term benefits of economic prosperity in creating stability in a turbulent zone. The Mughals especially were pioneers in promoting overland trade as a motor of economic growth in Afghanistan. Because of the proximity of the passes through the Hindu Kush and Suleiman mountains, Kabul enjoyed a premier position in the trading relations with Central Asia and India. Samarkhand, the principal city of Transoxiana (Central Asia), was the junction of main trade routes from Kashmir and Persia (via Kabul and Merv). The two cities of Samarkhand and Bokhara were principal marts for the Indian traders. Goods (coarse and fine calico, silk brocade, etc.) manufactured in Bengal and Gujarat reached Central Asia and Persia through Afghanistan. Cotton goods from Punjab and the Coromandel Coast were also exported to Central Asia, as were Kashmiri shawls. India also exported sugar and medicinal herbs. India imported dry fruits, fish fruits (grapes and apples), musks, furs, falcons and corals from Central Asia. Indian traders from Astrakhan also imported sables, birds’ feathers, white fur coats, mirrors, copper and iron from Russia. The principal import of India from Afghanistan was war horses. More than 100,000 horses were imported annually and the profit in this trade was estimated at 2,500 per cent. The pastoral nomads between the Indus and Oxus functioned as traders. The trading nomads were known as Powindahs. Mostly, they were of the Gilzai and Lodhi tribes of which the Lohanis, Nasiris and Niyazis were important subgroups. The Afghans of Kandahar came to Mughal Sind as carriers of trade. The Khatris (Hindu Multanis) functioned as traders and moneylenders, which kept the wheels of commerce turning smoothly. The Mughals especially took particular care to build and maintain highways and rest houses for the traders. Brisk trade accelerated the growth and expansion of cities along the main trading routes. Some of the towns also became centres for manufacturing export goods. Lahore, Sialkot and Shikarpur in Sind manufactured textiles and indigo, which were exported via Kandahar to Isfahan and Aleppo. The spin-­off effects of such

Introduction

15

merchandise on the Afghans was substantial. The pre-­modern Asian rulers showed more political wisdom than present-­day presidents, prime ministers and generals turned dictators. Even during inter-­state war, the Mughal, Safavid and Uzbek rulers took special care to promote and maintain long-­distance trading relations. In fact, foreign merchants were taken under special royal protective custody to prevent any harm falling on them. Furthermore, in case of any difficulties faced by foreign traders, the rulers deputed commissions and wrote to the opposite rulers for factoring out the difficulties. It is time for White House officials and the EU to think of such long-­term economic benefits for barren and unproductive Afghanistan. Just attempting to destroy the poppy cultivation will not work. The Afghans need alternative economic employment. And the great powers of the world should attempt to initiate a long-­term strategy for Afghanistan by launching the pipeline project for extracting natural gas from Central Asia. Such a project would have beneficial effects for Pakistan and India’s Kashmir. The collapse of the Safavid, Uzbek and Mughal power in the early eighteenth century, the rise of Nadir Shah and the silting of the Indus disrupted this long-­distance overland trade. This in turn partly accelerated the outbreak of tribal violence in Afghanistan.33 Thus, Rubin is erroneous in his assertion that after the discovery of a sea-­oceanic connection between Europe and Asia in the late fifteenth century (Vasco Da Gama influence), the Central Asian heartland including Afghanistan experienced a spate of economic decline.34 The reality was much more complex. The Eurocentric tendency to see the state as the source of stability is also problematic. In Chapter 8, Ivan Arreguin-Toft focuses on this issue. He advances three arguments. First, Afghanistan is, to paraphrase Benedict Anderson’s famous formulation, an imagined state. Just as individuals often derive considerable satisfaction in imagining themselves to be a part of a larger, yet effectively unknowable community, so too does the community of states – in particular Western states – deriving satisfaction from imagining that all the world’s peoples reside in similar political units. Second, because Afghanistan is instead a contingent collection of semi- and micro-­states rather than a single state as such, ‘order’ (understood as a government in possession of a monopoly on the legitimate use of violence) cannot be restored to it, regardless of the quantity of resources applied toward that goal, and regardless of the strategy employed to make use of those resources. Third, and finally, the theoretical and policy implications have significance for other imagined states, and lead to a crucial yet poorly understood dilemma, whose best resolution may lie in the community of established states looking to its own defences.

16

War and State-Building in Afghanistan

In his monograph on Afghanistan, Rubin blames the nature of the international system for fragmentation of Afghanistan during the dying days of the Cold War.35 Kristian Berg Harpviken turns the limelight on the troubled neighbourhood of Afghanistan in Chapter 9. Harpviken examines Afghanistan’s historical status as a buffer state, at the interface of empires. His point of departure is to see Afghanistan less as part of a larger South-Central Asia (the main premise for current policy thinking) and more as a waste ground for the internal conflicts in the three surrounding regional security complexes. This implies a critique of the dominant stance of the ‘international community’ over the past decade. It provides leads as to what to expect in the context of NATO’s exit and gives the basis for alternative policy formulations. We now turn to take a look at the chief characteristics regarding the evolution of polity in Afghanistan from historical times to the modern era. Rubin asserts that the rentier state being a weak state in order to compensate its fragility encouraged further fragmentation of Afghanistan’s society. He continues that traditionalism and localism of Afghanistan are not survivors of ancient traditions, but the result of Afghanistan’s forced integration into the contemporary state system.36 One is reminded of Terence Ranger, E.J. Hobsbawm and Hugh Trevor Ropers’ argument about construction and manipulation of identity formation. In 1978, on the eve of the Saur (April) Revolution, 85 per cent of the populace of Afghanistan were peasants and nomads and agriculture accounted for 60 per cent of production. Most of the agricultural lands were divided into small landholdings. The peasants paid almost no taxes. The state paid its bureaucrats, police and army personnel from mainly foreign aid.37 Almost all the inhabitants of Afghanistan are Muslims, of whom some 90 per cent are Sunni and the rest mostly Shias. Among the Afghans, the biggest ethnic group is the Pathans, who constitute some 40 per cent of the populace. Next come the Tajiks (20 per cent). The other important groups are the Uzbeks, Aimaqs and Hazaras.38 Afghanistan remains a patrilineal society in which property is inherited by agnatic kin. For the people of Afghanistan, kinship remains an essential vehicle to mobilize political and economic resources.39 Rubin and Harpviken highlight the baneful influence of the big powers’ divide et impera policy to explain Afghanistan’s fragmentation. Indeed, the great powers’ power-­politics has resulted in the division of countries like Korea. Even in the pre-­modern era as we have seen, Afghanistan remained divided among three great powers: Mughals, Uzbeks and the Safavids. Nonetheless, conflict, tribalism and the lack of state development in Afghanistan cannot be explained soley in terms of outside intervention and geopolitical power-­politics. A society based on

Introduction

17

subsistence farming which generates no significant taxable surplus cannot develop a Westphalian nation-­state. An insignificant taxation base prevents the central government from providing meaningful social services, which undermines the very legitimacy of the state. Armed tribals also prevent the central government from establishing a monopoly over armed power. This further weakens the functioning of a state in Afghanistan. The polity built by Ahmad Shah was a loose structure dependent on loot and plunder gathered from India. In order to widen the fiscal base, Dost Muhammad Khan, the ruler of Kabul in 1834–5, gave a call for jihad against the Khalsa Kingdom of Punjab. However, Ranjit Singh’s Punjab was able to defeat the Afghan invasion.40 Dost Muhammad Khan’s regime was already tottering when the British invaded Afghanistan in 1839. Western imperial powers like Britain and France played a divide et impera policy in South Asia and Southeast Asia. Yet, difficult geography and lack of sustainable economic growth probably made the case of Afghanistan much more problematic. A combination of unique domestic and foreign factors explain the turbulence in Afghanistan from historical to modern times. To conclude, from a long-­term perspective, Afghanistan is characterized by a weak state system and great power intervention. Attempts to introduce Western-­style state institutions in multi-­ethnic and multi-­religious societies of Asia will result in chaos if not disaster.

Notes 1 Barnett R. Rubin, The Fragmentation of Afghanistan: State Formation and Collapse in the International System (1995, reprint, New Haven, CT: Yale University Press, 2002), pp. 9, 12–14, 19. 2 Tirmiz retained its strategic value during the Soviet invasion of Afghanistan. 3 Irfan Habib, An Atlas of the Mughal Empire: Political and Economic Maps with Detailed Notes, Bibliography and Index (1982, reprint, Delhi: Oxford University Press, 1986), Notes, Sheet 1A-B, Northern Afghanistan, Political and Economic. 4 Christine Noelle, State and Tribe in Nineteenth-Century Afghanistan: The Reign of Amir Dost Muhammad Khan (1826–1863) (1997, reprint, London: Routledge, 2008), pp. xiii, 1–4. 5 A sowar is a cavalryman equivalent to a sepoy in the infantry. 6 Noelle, State and Tribe in Nineteenth-Century Afghanistan, pp. 1–2, 5, 7, 9, 25–6, 29. 7 Noelle, State and Tribe in Nineteenth-Century Afghanistan, p. xv. 8 Thomas Rid, ‘The Nineteenth Century Origins of Counterinsurgency Doctrine’, Journal of Strategic Studies, vol. 33, no. 5 (2010), pp. 727–58.

18

War and State-Building in Afghanistan

9 Christian Tripodi, ‘Grand Strategy and the Graveyard of Assumptions: Britain and Afghanistan, 1839–1919’, Journal of Strategic Studies, vol. 33, no. 5 (2010), pp. 701–25. 10 Kaushik Roy, The Army in British India: From Colonial Warfare to Total War, 1857–1947 (London: Bloomsbury Academic, 2013), pp. 102–22. 11 Raffi Gregorian, ‘Jungle Bashing in Malaya: Towards a Formal Tactical Doctrine’, in Ian Beckett (ed.), Modern Counter-Insurgency (Burlington, VT: Ashgate, 2007), pp. 29–50. 12 Keith Jeffery, ‘Intelligence and Counter-Insurgency Operations: Some Reflections on the British Experience’, in Beckett (ed.), Modern Counter-Insurgency, pp. 107–38. 13 Tim Jones, ‘The British Army, and Counter-Guerrilla Warfare in Transition, 1944–52’, in Beckett (ed.), Modern Counter-Insurgency, pp. 139–81. 14 Randall W. Heather, ‘Intelligence and Counter-Insurgency in Kenya, 1952–56’, in Beckett (ed.), Modern Counter-Insurgency, pp. 79–105. 15 Christopher C. Harmon, ‘Illustrations of “Learning” in Counterinsurgency’, in Beckett (ed.), Modern Counter-Insurgency, pp. 351–70. 16 Carl Van Dyke, ‘Kabul to Grozny: A Critique of Soviet (Russian) Counter-Insurgency Doctrine’, in Beckett (ed.), Modern Counter-Insurgency, pp. 467–83. 17 David A. Charters, ‘British Intelligence in the Palestine Campaign, 1945–47’, in Beckett (ed.), Modern Counter-Insurgency, pp. 3–28. 18 Huw Bennett, ‘Minimum Force in British Counterinsurgency’, Small Wars and Insurgencies, vol. 21, no. 3 (2010), pp. 459–75. 19 David A. Percox, ‘British Counter-Insurgency in Kenya, 1952–56: Extension of Internal Security Policy or Prelude to Decolonisation?’, in Beckett (ed.), Modern Counter-Insurgency, pp. 263–318. 20 Huw Bennett, ‘The Other Side of the COIN: Minimum and Exemplary Force in British Army Counterinsurgency in Kenya’, Small Wars and Insurgencies, vol. 18, no. 4 (2007), pp. 638–64. 21 Stephen L. Melton, The Clausewitz Delusion: How the American Army Screwed Up the Wars in Iraq and Afghanistan (A Way Forward) (Minneapolis, MN: Zenith Press, 2009), p. 138. 22 Ivan Arreguin-Toft, How the Weak Win Wars: A Theory of Asymmetric Conflict (Cambridge: Cambridge University Press, 2005), pp. 169–99. 23 Melton, The Clausewitz Delusion, p. 143. 24 Richard Popplewell, ‘ “Lacking Intelligence”: Some Reflections on Recent Approaches to British Counter-­insurgency, 1900–1960’, in Beckett (ed.), Modern CounterInsurgency, pp. 319–35. 25 The latest, most informative and searing critique of Western COIN is provided by Douglas Porch in his Counterinsurgency: Exposing the Myths of the New Way of War (Cambridge: Cambridge University Press, 2013). Porch asserts that present-­day Western COIN could be traced back to late nineteenth-­century imperial campaigns in Afro-Asia and Eurasia. American, British and French COIN remains based on

Introduction

19

notions of racial supremacy and brute force. And present-­day Western COIN, in the eyes of Porch, is actually neo-­imperialism in a new garb. 26 For theory and practice of COIN by the Indian Army after 1947, see Scott Gates and Kaushik Roy, Unconventional Warfare in South Asia: Shadow Warriors and Counterinsurgency (Burlington, VT: Ashgate, 2014). 27 T.R. Moreman, The Army in India and the Development of Frontier Warfare, 1849–1947 (London: Macmillan, 1998). 28 Jugdep S. Chima, ‘Controlling the Sunni Insurgency in Iraq: “Political” and “Military” Strategies from Successful Counterinsurgency in Punjab-India’, Small Wars and Insurgencies, vol. 18, no. 4 (2007), pp. 615–37. 29 Rod Paschall, ‘Marxist Counterinsurgencies’, in Beckett (ed.), Modern CounterInsurgency, p. 455. 30 Arreguin-Toft, How the Weak Win Wars, p. 170. 31 Melton, The Clausewitz Delusion, pp. 140, 143; The Soviet–Afghan War, How a Superpower Fought and Lost: The Russian General Staff, trans. and ed. by Lester W. Grau and Michael A. Gress (Lawrence, KS: University Press of Kansas, 2002), p. 12. 32 Christian Tripodi, ‘Peacemaking through Bribes or Cultural Empathy? The Political Officer and Britain’s Strategy towards the North-West Frontier, 1901–1945’, Journal of Strategic Studies, vol. 31, no. 1 (2008), pp. 123–51. 33 MuzaffarAlam, ‘Trade, State Policy and Regional Change: Aspects of Mughal–Uzbek Commercial Relations, c. 1550–1750’, Journal of the Economic and Social History of the Orient, vol. 37, no. 3 (1994), pp. 202–27; Noelle, State and Tribe in NineteenthCentury Afghanistan, p. 23. 34 Rubin, The Fragmentation of Afghanistan, p. 19. 35 Rubin, The Fragmentation of Afghanistan, p. x. 36 Rubin, The Fragmentation of Afghanistan, p. 15. 37 Rubin, The Fragmentation of Afghanistan, p. 19. 38 Arreguin-Toft, How the Weak Win Wars, p. 170. 39 Rubin, The Fragmentation of Afghanistan, pp. 22–3. 40 Noelle, State and Tribe in Nineteenth-Century Afghanistan, pp. 15, 17.

1

Continuity and Change in Asymmetric Warfare in Afghanistan: From the Mughals to the Americans Scott Gates, Kaushik Roy, Marianne Dahl and Håvard Mokleiv Nygård

Introduction From 1520 to 2014 the Mughal, British, Soviet and ISAF (International Security Assistance Force, i.e. NATO) armies have engaged in asymmetric conflict in Afghanistan. Despite superior strength, none of them succeeded.1 Like former rebel groups opposing previous foreign invasions in Afghanistan, the Taliban is both willing and able to continue fighting. In this chapter, we argue that this is due to physical geography and strategic and tactical adaptation, which have favoured the continued survival and fighting of the weaker side. These factors have remained constant, or at least remained relatively stable, over the centuries in Afghanistan. The reasons why ISAF has not succeeded in defeating the Taliban are similar to those why the Soviets, British and Mughals never succeeded.2 One important factor that has varied over time, however, is the intensity of differential time preferences. Afghanistan offers a unique ‘laboratory’ in which to analyse asymmetric conflict. In this chapter, we draw on and extend arguments developed by Andrew Mack3 and Ivan Arreguín-Toft4 on how the weak win wars, and combine them with Christopher Butler and Scott Gates’ theory of strategic adaptation, featuring the concepts of the ability to hurt and to hide.5 Despite considerable technological change, the nature of the conflict in Afghanistan has exhibited a remarkable level of continuity from the Mughal (sixteenth century) to the recent era. The implication is that some aspects of insurgency–counterinsurgency dynamics tend to be relatively static compared with other theatres of war. Nevertheless, the strategic nature of armed conflict is defined by change, adaptation and learning. Indeed, the characteristics of asymmetric war in Afghanistan are characterized

22

War and State-Building in Afghanistan

by both continuity and change. Pacification policies adopted by intervening forces tend to be repeated over the course of Afghanistan’s history. The main question we seek to answer is, given that one is a much weaker power, why continue fighting? We offer two explanations. First, the weaker party derives some form of benefit from fighting that exceeds the costs. This relates to the ability to hide and hurt, which in turn relates to the unchanging nature of geography. The second reason is that in the long run, the weaker party believes that the stronger will leave the fight. You need not defeat the stronger; one just needs to last longer. This is the essence of differential time preferences. The first reason has remained constant over the centuries; the second has intensified over time.

The ability to hurt vs. the ability to hide There are different perspectives from which to analyse asymmetric warfare, depending on the nature of the question asked: those interested in how the weak win wars focus on material and technical gaps; those interested in insurgency– counterinsurgency dynamics focus on the tactics used in asymmetric conflict (i.e. guerrilla tactics, insurgency, terrorism, etc.). The fact is that the two aspects of asymmetric war fit hand in glove: the tactics used in asymmetric war stem from the material imbalance between forces. Adaptation and innovation are the hallmarks of warfare. Weapons, logistics, transport and other forms of military technology have changed considerably over the course of history. For every innovation comes the diffusion of new technology, nullifying the advantage held by the first adaptor. Military history abounds with examples of technological innovation providing an advantage to one belligerent over another – for example, the Ottoman defeat of the Safavids with the use of gunpowder in 1514 at the Battle of Chaldiran.6 Another example is the defeat of the Polish armies by the German panzers, which played such a prominent role in the Blitzkrieg described by Liddell Hart and Taylor.7 Yet, technological innovation is only part of the story. Tactical innovation is also a central aspect of military history. Strategic adaptation of tactics can and often has mitigated the technological edge of more powerful armies. A group lacking the means or the ability to exploit a new technology will adapt strategically to stay in the fight. If they don’t, they will be defeated. This form of adaptation, the negation of technological advantage, is essential to sustaining an asymmetric war. Indeed, ‘the choice of tactics involves a trade-­off between the

Continuity and Change in Asymmetric Warfare

23

rebels’ ability to hurt the government and the government’s ability to hurt the rebels’.8 If the parties in an asymmetric conflict were to meet in a conventional battle, the weaker party would be defeated – and, in most cases, cease to exist. In order to survive, the weaker party must adapt materially or strategically. In the short run, material adaptation is rarely an opportunity; instead, the weaker party will tend to adopt non-­conventional tactics to put the conflict on a more even footing. This tends to involve prioritizing hiding – and it’s own survival – over the ability to hurt. As we demonstrate below, successful strategic adaptation hinges on the ability to hide while continuing to hurt. This depends to a large extent on the natural terrain. A rebel group is defined by its members, their distribution throughout the country, and their density among the civilian population. All these aspects affect the rebel group’s ability to fight and survive. It is much harder to identify and capture a small, dispersed group, hiding either among civilians or in the natural terrain. However, a large and concentrated group located close to the government will be much better placed to hurt the government, and in the end, win the conflict.9 To win the conflict, the rebel group will in the long run have to adapt materially, and choose a tactic that benefits the ability to hurt over the ability to hide. The weaker party, typically an insurgent group fighting an asymmetric war, will tend to adopt non-­conventional tactics so as to put the conflict on a more even footing. Systematic analysis of all wars over time demonstrates that military tactics play a significant role in determining victory in battle.10 To overcome superior troop strength and technology, weaker parties turn to guerrilla warfare and terrorism. Both guerrilla warfare and terrorism involve tactics that significantly reduce the government’s ability to hurt the rebels. Indeed, guerrilla tactics constitute the quintessential form of asymmetric warfare. Guerrilla warfare is fought by small groups of combatants employing tactics involving mobility and surprise. Ambushes, raids and sabotage are used in an effort to cripple the state, particularly the state’s military capacity. Guerrilla tactics focus more directly on the infrastructure and agents of the state. Terrorism, in contrast, targets not the state itself, but non-­combatants. The tactic is designed to instil fear and intimidation among the general public to further a political agenda. It affects change indirectly through the creation of a widespread sense of fear among the general populace. When the weaker side chooses tactics that benefit the ability to hide, the superiority of the stronger side becomes less prominent. The success of such a tactical adaptation hinges on whether external factors allow for it.

24

War and State-Building in Afghanistan

To understand the inherent logic tying tactics to asymmetric unequal access to resources with which to devote to the conflict and unequal military technology, we turn to the contest success functions (CSF) technology widely employed by economists to analyse conflict.11 The CSF model assumes that each belligerent has some initial endowment of resources, which are allocated to guns or butter (or into fighting or productive effort). The relative capabilities of different armed groups play a critical role in the contest success function. But more important, capability is not the only factor being modelled. Contest success functions explicitly account for the allocation of resources to guns and to non-­fighting effort. Few wars, if any, are true total wars (in which nations devote their total economic capacity to fighting). Most wars involve only a portion of a nation’s total capacity dedicated to war capacity.12 This decision of how much to allocate to the fighting effort (or military expenditure) can be modelled explicitly. This is especially relevant for understanding asymmetric warfare. The proportion of collective value that a group gets is a function of its fighting effort divided by the total fighting effort of both groups. Given this division mechanism, each group maximizes its share of the collective value, which implies that all productive value is commonly pooled and can be captured by either side through fighting. In this regard, fighting is costly. We can think of the sum of the fighting efforts of the players as the internalized costs of war. In addition, of course, war imposes all kinds of negative externalities on non-­combatants. This essentially means that groups are subject to a budget constraint, which we model as a resource endowment. For our model, we assume that both groups simultaneously make their decisions about fighting effort. We find that each group’s equilibrium level of fighting effort is a function of its initial resources. The relative distribution of resources available to each group, in turn, shapes each group’s allocation decision. Given a relative disparity in resources to devote to ‘guns and butter’ and the associated asymmetric distribution of power, a weaker group will obtain a higher marginal utility to fight rather than engage in other economic activities. As the weaker power has a higher marginal benefit from fighting, it expects to get more from fighting than from not fighting. The incentive to fight increases as the weaker group becomes comparatively weaker. Indeed, a weaker group will devote a greater proportion of its resources to fighting to continue the fight. Jack Hirshleifer, who originally developed this model, labelled it the ‘paradox of power’,13 while Butler and Gates14 refer to it as the ‘nothing left to lose’ result.

Continuity and Change in Asymmetric Warfare

25

Given limited resources, a weaker party will employ the tactics that most efficiently consume their limited resources and sap the strength of the more powerful enemy. This is the essential strategic element defining asymmetric war. Consequently, it should not be surprising that throughout history rebel groups in Afghanistan have made use of strategies that allow them to hide instead of engaging in conventional warfare.

Hiding and hurting: geography The geography of Afghanistan offers great opportunities for hiding, making it possible to survive and continue fighting. The importance of geography has persisted from the age of the Mughals to the present day. The Taliban can hide in the mountains and launch surprise attacks, making it difficult for the ISAF forces to identify and target the rebels. In addition, Afghanistan offers many opportunities for financing rebellion, especially poppy cultivation. This creates a situation where rebel groups benefit from the fighting itself. The implication is that the Taliban are not fighting simply to win. The Taliban know it can never decisively defeat the coalition forces, but, as we argue below, fighting itself has benefits for the Taliban. Unlike the ISAF, the Taliban continue to fight, even if the chances of military success are remote as long as ISAF forces oppose them. Parties that fight simply to win and parties that benefit from the conflict itself value time differently. The conflict itself is a burden to the ISAF forces, and in the long run they will have to exit the conflict. In contrast, the Taliban are in no rush to get out. Consequentially, as long as the Taliban forces survive, it does not matter that they continue to be weaker than the ISAF forces. At some point in the future the ISAF forces will exit Afghanistan, and the Taliban will have a much easier battle to fight. A failure to take this fact properly into consideration leads to flawed COIN tactics evidenced in Afghanistan with the over-­reliance on ‘surge’-type tactics. Geography and relative fighting capability are inherently interrelated. Empirical cross-­national analyses of armed conflict by Halvard Buhaug and colleagues (Buhaug and Gates,15 Buhaug and Rød,16 Buhaug, Gates and Lujala,17and Buhaug18) show that difficult terrain, porous boundaries and access to lootable resources also affect the relative strength of a government and rebel group engaged in conflict. Moreover, power decays as it is projected across distance. In Boulding’s terminology, rough terrain increases the loss-­ofstrength gradient (LSG) of military power projection, particularly for heavily

26

War and State-Building in Afghanistan

armed and mechanized regular forces.19 Rough terrain where the government has little or no control makes it possible for insurgents to hide, while continuing to hurt. Additionally, rebel access to neighbouring countries tends to benefit the rebels, and makes the conflict even more difficult to end. This stands in stark contrast to countries where the state is strong and poses more or less complete control. Under such circumstances it is extremely difficult to continue hurting while hiding. Government forces have a clear disadvantage if they are required to operate over long distances. Remote and sparsely populated regions are more difficult for a government to control, and governments often face significant logistical obstacles when involved in a conflict in distant areas.20 These include physical barriers for transportation of troops and equipment (such as mountains and the lack of a proper transport network), higher costs associated with longer distance, limited knowledge of the local environment and, as is often the case, lack of support from the local population. Indeed, the ability to project power is likely to affect both how and for how long a war is fought.21 A rebel group that has knowledge of the terrain and knows how to benefit from it is better equipped to succeed on the battlefield. Nowhere has this been demonstrated with more force than in Afghanistan, where mujahedeen groups successfully fought off the Soviet invasion, and more recntly the Taliban insurgents have managed to evade massively superior NATO forces. Access to natural resource wealth, location and terrain also affect the local balance of power between the rebels and the government.22 Valuable and easily extractable resources, such as alluvial gemstones and narcotics, may increase the rebels’ funds available for the purchase of arms, and the prospect of personal enrichment may also make rebel recruitment easier. Access to such wealth also affects how troops are deployed and engaged in battle, the characteristics of warfare and the prospects for an early settlement to the conflict. Such wealth should make the rebel group stronger. If a rebel group can seek protection in a neighbouring country that can serve as a sanctuary and base for revenue raising, and porous borders that offer access to safe havens in foreign territory and to international markets, the national army may find it impossible to operate beyond the state boundaries, obviously reducing the chances of defeating the rebel group. Safe havens help facilitate sustained opposition,23 though it is also consistent with arguments regarding the role of foreign markets24 and supportive ethnic kin.25

Continuity and Change in Asymmetric Warfare

27

Where a rebel group chooses to recruit and to fight depends on the strategic ambitions of the group. But where a group is operating can also be a powerful determinant of its strategic ambitions. If the group aims to capture the state, it eventually must aim at the capital, and in general this will give the government the advantage with regard to its ability to project force. If the group’s aim is to secede, the group is more likely to be in a more favourable position vis-à-vis the government (presuming that the seceding province is relatively remote from the capital). Therefore, only the strongest opposition movements, enjoying support from the masses or key elements of the military, attempt to challenge the centre on its home ground – unless the state is unusually weak.

Willingness to fight A rough terrain or a supportive civilian population can provide relatively weak rebels with the opportunity to adapt strategically and continue fighting. However, given that the rebels know that they will never be able to adapt materially, and defeat the enemy, it might not be in their interest to do so. To understand why the weaker side continues to fight, we must look not only at the opportunity to do so, but also at the circumstances that make them willing to do so. We argue that a weak group will be willing to continue fighting, despite knowing that they will never defeat the enemy, if either: (1) they benefit from fighting itself, or (2) they know that the stronger side will exit the conflict in the long run. If the latter is the case, all they have to do is to outlast the stronger side, not defeat it. Relative military capabilities may affect the chances of victory or defeat, but they definitely do not determine winning or losing. Similarly, possessing a strong (but not overwhelming) military advantage may not necessarily lead to a short war. First, a low-­intensity conflict itself may be the objective of a rebel group. Small-­scale violence may provide the ideal setting for illegal gemstone mining and drug cultivation – opportunities that may be considerably more limited in more peaceful times. Therefore, excessive wartime income may reduce incentives for a peaceful settlement. Second, groups that resort to extensive looting of resources may concentrate on revenue extraction and use less time and effort to fight the state army. This would make the rebel movement a smaller and less immediate threat for the government, which may then tolerate the group as long as it is not a direct threat to its existence. Finally, it is possible that natural resources provide opportunities for weak, cash-­strapped movements to emerge

28

War and State-Building in Afghanistan

that do not have the potential to grow strong. There is no shortage of long­lived resource-­funded insurgencies: examples include Afghanistan (opium), Colombia (cocaine), Myanmar (timber, opium, gems) and Cambodia (timber and gemstones). The few empirical studies that account for rebel access to contraband are generally consistent with the proposed conflict-­prolonging effect of contraband resources.26

Time preferences A weak party unable to adapt materially can still win a conflict if it survives to the point where the stronger party gives up. The critical distinction has to do with time preferences, which dictate how long a group will engage in conflict. Groups are expected to continue fighting when the net costs from fighting are lower than the expected gain from winning, controlling for the likelihood of actually winning. A coarse distinction can be made between two types of groups: (1) those that only experience gains if they win the conflict, and (2) those that also benefit from the conflict while it is being fought. Some rebel groups benefit from the conflict itself through looting. This is more likely to be the case when natural resources such as drugs and diamonds involve significant income opportunities.27 If unemployment is high, the opportunity cost of fighting might be negative.28 When that is the case, whether the group continues fighting or not does not depend on the size of the prize of winning or on the likelihood of winning. Instead, it depends on whether the group can gain more by putting its weapons down. This is rarely the case for a government, especially when fighting abroad. Time preferences shape the outcome when working in combination with opportunities to hide and hurt, as well as the ability to sustain costs. Fighting will stop if hiding opportunities end or if the relative benefits from conflict are reduced. Moreover, a group that cannot sustain casualties from fighting will in the long run exit the conflict. In contrast, a group that can sustain casualties can take a long-­term perspective, and remain in the conflict. In general, foreign powers engaged in COIN on distant shores will have shorter time horizons than native insurgent groups. Democracies, in particular, facing inpatient voters will not be able to sustain high costs in the long run. Voters are likely only commit to the conflict in the short run, while rebels and their supporters are likely to commit in the long run. In other words, time preferences reflect notions of resolve.29

Continuity and Change in Asymmetric Warfare

29

Asymmetric war in Afghanistan: from the Mughals to the Americans We now enter the laboratory of Afghanistan to study asymmetric conflict. Lacking economic opportunities, the mainstay for the male populace of Afghanistan was (and to some extent still is) military service along with poppy cultivation and looting and plundering the surrounding sedentary civilizations. If there ever was an environment that reflected Hirshleifer’s ‘dark side of economics’ – one of producers and predators, or a mix of both, it is Afghanistan.30 Our analysis begins in 1550 when Afghanistan serves as a pawn in the Great Game between Mughal India, Safavid Persia and the Uzbek Khanate of Central Asia (three of the great Islamic empires). We end our analysis in 2012, paying witness to the ongoing operations by the US and NATO forces in Afghanistan. Myriad volumes have been written on the British, the Russians and NATO’s contemporary COIN operations in Afghanistan, but few have systematically compared these experiences. Historians and political scientists have tended to focus on one period of history. Some even argue that earlier periods are irrelevant for understanding the current war. Many of these contemporary works, moreover, have not addressed the problem of integrating military and political approaches. We address this problem by taking a long historical perspective, examining strategic changes in the fighting of asymmetric war in Afghanistan, from the Mughals to the Americans. Five invaders mark the history of Afghanistan: Alexander the Great, the Mughals, the British, the Soviets and today the ISAF forces – all tried either to take control of Afghanistan or to defeat specific groups, but none succeeded. From 1550 to 2014, the Mughals, British, Soviets and Americans/NATO engaged in asymmetric warfare in Afghanistan. In each era, physical geography has remained constant or at least remained relatively stable over the centuries. In turn, geography has limited the degree to which insurgency and COIN policies can be modified. The relative constancy of these central factors has led to substantial continuity over time, which allows us to compare the nature of asymmetric war over the centuries. We find the experiences of the past especially relevant for today. Afghanistan offers a perfect example of how being the stronger side is no guarantee for success. Despite fierce advantages when it comes to weapons, logistics and transport, ISAF has not succeeded in Afghanistan. Owing to its geography and ethnic composition, Afghanistan has offered an extremely favourable environment for making it possible for the weaker side to survive in conflict through tactical adaptation. The consequence in the short run

30

War and State-Building in Afghanistan

is continued conflict. As the strong side in these conflicts has always come from abroad, and has had a vested interest in leaving the conflict, the weaker side has always been likely to win in the end. To end this circle of history, the ISAF forces have to embed the dynamics of conflict into their analysis of how to win.

Afghanistan and the ‘Great Game’ in Asia The Mughals, like all subsequent foreign powers in Afghanistan, faced the problem of asymmetric war. Also as in subsequent periods of history, Afghanistan geographically was positioned between two great powers, the Safavid Persian Empire and the Mughal Empire (see Map 1). The Safavids’ power was in decline, while the Mughals were expanding their influence. This resulted in contestation over Kandahar. A unified Afghan monarchy emerged following this inter-­state rivalry between Persia and Mughal India over Kandahar. The Mughal Empire invaded Afghanistan, but failed to control the amirs and the maliks at the local levels. These local power holders sponsored insurgencies in the north-­western edge of the Mughal Empire (i.e. the present-­day ‘badlands’ of Pakistan). In many ways, this pattern is similar to the warlords that exercise local authority in Afghanistan today. In the same way as the Karzai regime is confronted today by these warlords, these local authorities fundamentally challenged the Mughal Empire. These disruptions resulted in the Mughals launching regular annual expeditions against the tribal areas. The rough terrain of Afghanistan played a

Map 1  The Safavid and Mughal Empires

Continuity and Change in Asymmetric Warfare

31

fundamental role. Nimble Afghans fighting from the mountaintops ambushed the heavy Mughal cavalry columns along the narrow mountainous tracks. Despite superior forces and military technology, the Mughals were never able to control Afghanistan. Strategic adaptation and the problems of logistics in this terrain proved to be favourable to the weaker power.

Britain’s small wars in Afghanistan The British doctrine of ‘small war’ developed in relation to operations in Afghanistan during the late nineteenth century. This doctrine gave birth to Britain’s COIN doctrine of the twentieth century (which was used in Malaysia, for example). Just like the Mughals, the British realized that continuous instability in Afghanistan resulted in small wars becoming hot wars, more specifically resulting in the three Afghan Wars of 1842, 1882 and 1919. Like the Mughal mansabdars, the British generals discovered the importance of logistics in rough terrain. During their operations in Afghanistan, operating from bases in north-­west India (now north-­west of Pakistan), more than half of the British units were tied up guarding the long line of communications. The British were unable to hold their long line of communications, which was continuously harassed by the Afghan lashkars. Indeed, the British invasions of Afghanistan were costly. Moreover, unlike conventional warfare there was no value to capturing territory. Land was infertile and unable to support a large invading army. In addition, not even the capture of Kabul offered military advantage, since Afghanistan was in no way a Westphalian state. Even the defeat of the Afghan regular army did not lead to victory. Like a hydra, innumerable war bands under the clan leaders emerged throughout the countryside. This lack of central political control still describes the political environment in Afghanistan. Both the British and the Mughal invading armies had to retreat because there was no objective to capture and the land was infertile to support a large invading army. During the 1919 Waziristan Campaign and the subsequent Third Afghan War, the British found out that light tanks and armoured cars were useless. The Afghans had adapted tactically and blocked the narrow roads in the valleys with boulders. Adapting to this tactic, the British relied on light infantry skilled in road-­opening duties and convoy escort. The British mounted a continuous pacification campaign in Waziristan to seal off the border with Afghanistan, in order to prevent the local Pushtuns in Waziristan from providing logistical support to the insurgents roaming southern Afghanistan. A similar kinship

32

War and State-Building in Afghanistan

dynamic is evident in recent years as the Taliban/Al-Qaeda Pushtun personnel acquire logistical support from the Pushtuns of north Pakistan. This is evident today in Waziristan in a massive pacification campaign by the Pakistan Army (due to US pressure) to prevent infiltration of the insurgents from southern Afghanistan into Pakistan. The British failed despite their relative strength and tactical adaptation to the insurgents. In the long run, they were unable to sustain the costly occupation that offered few if any material benefits. On their side, the Afghan insurgents took advantage of the possibility to hurt and hide offered by the rough Afghan terrain. The British experiences with the decentralized political nature of Afghanistan and the role of outside forces, particularly with regard to the shared Pushtun kinship in Pakistan and Afghanistan, can help us better understand the situation in Afghanistan today. These two issues are particularly relevant when it comes to a discussion of a political solution for the present.

The bear in Afghanistan The USSR intervened in Afghanistan in 1979. A modern armoured, conventional army failed against its weak irregular opponents (at its peak, there were about 70,000 lightly armed mujahideen, of whom 20,000 were fighting at any one time), thus vindicating the British and Mughal experiences. Take, for instance, the Battles of Zhawar (September 1985 to April 1986) fought near the Pakistan– Afghanistan border. The mujahideen complex centred round the caves was supplied from Pakistan through Miran Shah. Even the Soviet air attacks on the caves with smart munitions did not prove effective, just as British artillery attacks against the sangars proved to be ineffective during the second decade of the twentieth century. The Soviet ground force attacked in the traditional style of the Second World War. First, heavy artillery opened up and then the infantry advanced. The mujahideen were alerted by the Soviets’ opening artillery barrage that a ground attack was imminent. With the beginning of the artillery barrage, the mujahideen, taking advantage of the ability to hide, took cover in the caves and after the end of the barrage they opened up on the advancing Soviet infantry in the open with machine guns and rocket launchers. The Soviets also experienced difficulty in logistic support. Even the technological advancement of using helicopters for transport proved inefficient. For pricking the Bear, the CIA funded and equipped the jihadis with STINGERS and other ‘high-­tech’ short arms. These weapons were used to great effect, and Several MI–8 helicopters used for ferrying Spetsnaz troops were lost during the course of the operation.

Continuity and Change in Asymmetric Warfare

33

Mughal heavy cavalry, British heavy field artillery and Soviet AFVs all proved useless in the roadless and bridgeless Afghan terrain. Geography again trumped technology in the battlefields of Afghanistan. Between 1980 and 1985, the Soviets bombed villages, destroyed granaries and the irrigation systems in order to eliminate villagers’ support for the mujahideen. This resulted in the displacement of 7 million Afghani refugees to Iran and especially Pakistan. Many of them later swelled the mujahideen ranks. Soviet firepower, heavy COIN operations (using fixed-­wing assault aircraft, helicopter gunships and commandos) and forcible transfer of the population (similar to the British technique used in Malaysia against the Chinese immigrants) backfired, as even moderate Afghans were alienated. Lack of respect for Afghan culture and Islam and a rigid imposition of inflexible Marxist doctrine made the situation even worse for the Soviets. The role of the USA in strengthening the mujahideen played a significant role in altering the nature of the asymmetric war being fought by the Soviets. Geopolitics again played a role, as was the case with the Mughals and the Persians, the British and the Russians, and now the Soviets and the Americans. Ultimately, the cost of the Soviet intervention was too high. The spread of madrassas in Zia ul Haq’s Pakistan constituted the seedbed for the rise of Al-Qaeda and the Taliban. Initially, the jihadis/mujahideen fought the ‘godless’ Soviets. After the retreat of the Soviet Union from Afghanistan and the failure to properly build a functioning and effective Afghan national army capable of controlling Afghan territory, the Taliban took over the Afghan state. Under the Taliban regime, Afghanistan became a sanctuary for extremist

Map 2  Soviet USSR

34

War and State-Building in Afghanistan

groups. Subsequently, these groups turned their attention from the godless Soviets to the decadent capitalist West. The lesson to be learned here regards the role of third parties to the conflict. The massive number of troops deployed in Afghanistan by the Soviets also provides some lessons for the American ‘surge’.

The USA’s misadventure in Afghanistan Operation Enduring Freedom (OEF) proved to be a hollow victory for the USA. In Iraq, the full blast of insurgency hit US forces after the regular forces of Saddam Hussein folded up during Operation Iraqi Freedom (or the Second Gulf War). In both Afghanistan and in Iraq, there was to be no quick victory, instead the slow slog of COIN. Indeed, the Americans are still debating with NATO and themselves about how many troops are required to police Afghanistan and its 31 million inhabitants. Technology is no panacea for tackling the problems posed by the light-­footed armed insurgents. Cruise missiles, smart weapons and Black Hawk helicopters have proven useless against the AK 47 of the Afghan jihadis just as the Sniders and Krupp guns of the British proved useless against the jezails of Afghan jihadis in the late nineteenth century. Once again taking advantage of mountainous terrain, Afghan insurgents are able to hide and wait out the enemy, periodically extracting losses from the ISAF forces and making sure the war grows increasingly unpopular in the West. Indeed, much of Afghanistan is considered high risk by ISAF forces (see Map 3).

Map 3  Afghanistan 2007

Continuity and Change in Asymmetric Warfare

35

There are still options for ISAF forces to adapt to insurgent tactics. Instead of armoured/motorized formations, air mobile brigades should be used for rapid transport of troops to quell disturbances in the inaccessible regions. The USNATO forces, like the Soviets, shouldn’t use ground-­attack helicopters and aircraft. These generate collateral damage and loss of legitimacy for the Coalition forces. ISAF should take a cue from Commander-­in-Chief, British India Field Marshal Rawlinson, who did not allow the RAF to bomb the Afghan and Indus tribesmen during the 1920s and the 1930s. A successful COIN operation must be a ‘just war’. Stability in Afghanistan requires holding the ground with a large infantry army. What is required is a ‘bite and hold’ strategy with large ground forces. The US should use local Afghan armies for conducting close-­quarter, small-­scale combat with Al-Qaeda and the Taliban, exemplified by these ‘hugging tactics’. Fighting from a distance with missiles will not be successful. Like the British Indian infantry, the US-NATO troops require light ambush drill, picketing the mountaintops and sides of the hills. The US-NATO military planners would do well to read Charles Callwell’s chapter on mountain war in his magnum opus titled Small Wars published in 1896. Within the US military and NATO at least some officers are tinkering with new ideas about how to win COIN operations and the ways to engage in nation-­ building. One American officer has touted the formula of these insurgencies being a ‘three block war’: a combination of high-­intensity warfare, humanitarian aid and low-­intensity operation all within the three blocks of a city, and all these happening more or less simultaneously. Lieutenant-General Peter W. Chiarelli and Major Stephen M. Smith,31 influenced by Douglas M. Macgregor,32 argue that the hierarchical structure of the US military forces need to be flattened to enable a quick flow of information from top to bottom and vice versa and this would allow the men on the ground to take quick decisions that would allow them to get inside the enemy’s OODA (Observe, Orient, Decide and Act) loop. Here we are back to John Boyd’s33 OODA loop that was initially designed to implement rapid manoeuvre campaigns (involving air–land battles) based on Auftragstaktik (mission-­oriented command system) against the Soviet forces in central Europe. Several American officers are arguing that they need a new COIN doctrine that focuses not on the enemy (i.e. not concentrating on the number of enemy soldiers killed) but on the common people of the regions dominated by insurgency. So, the focus is on a people-­centric COIN, a concept somewhat similar to Rupert Smith’s concept of fighting the ‘war amongst the people’.34 Retired colonel Mike Capstick argues that the three pillars of COIN should be security, governance and economic plus social developments.35 And

36

War and State-Building in Afghanistan

he rightly says that of these three pillars, security is most important, as without it, economic and social developments and proper governance in the ‘disturbed’ region would not be possible. Local Afghan forces played a vital role in ousting Al-Qaeda and the Taliban especially from north Afghanistan during OEF. These forces should be equipped and trained by the NATO-US forces and officers from the NATO and US armies should learn the local vernaculars and function as trainers and officers. The British practised this technique and built up the combat-­efficient and loyal Indian Army during the late nineteenth century. At present, there is a debate whether to organize the Afghan National Army as mixed units (various communities recruited from diverse regions organized in companies of a particular regiment) or class units (comprising a particular community from a particular region constituting a regiment). The Soviet experiment with mixed units failed miserably during the 1980s. Class units have better motivational strength. The British Indian Army of the nineteenth century and the present Indian Army with their mixed and class regiments could provide a model for the security managers associated with organizing the Afghan National Army. The Afghan security forces (war bands of the warlords) could function as paramilitaries. The British had a number of them during the late nineteenth century, such as the Khyber Rifles and Tochi Scouts. At present, instead of equipping and training the Afghan National Police (ANP) and Afghan National Army (ANA), the US is pleading for more military manpower from the NATO states. Due to lax discipline, inadequate training and a paucity of funds, the desertion rates from the ANP and the ANA are quite high. The establishment of stable local networks in Afghanistan depends on solid indigenous leaders. A mechanical divide et impera strategy in the long run will further fragment the Afghan polity and will backfire on the protagonists. A nationwide, loose federal structure needs to be set up in Afghanistan. One technique for ensuring local security is the establishment of Village Defence Militia. These local leaders should be given a stake in the economic developments – for instance, revenues from the projected pipeline of natural gas from Central Asia. The current literature on counterinsurgency is for the most part focused on how to win asymmetric conflicts, or more broadly on how asymmetric conflicts end. This has been addressed by looking at the balance of power between the belligerents, and the amount of resources committed to the fight. The idea of overcoming an insurgency through a ‘surge’ of troops flows naturally from such an analysis. By committing more troops to the fight, eventually a tipping point will be reached when the enemy will be completely overpowered, resulting in a

Continuity and Change in Asymmetric Warfare

37

counterinsurgency victory. Such a strategy proved successful in Iraq, but has so far not led to anything near a decisive victory for coalition forces in Afghanistan. It could be argued that the reason the surge worked in Iraq but not in Afghanistan is simply that it involved too few additional troops in Afghanistan. We argue, in contrast, that such an argument misses a fundamental aspect of the dynamics of asymmetric conflicts. The surge mentality that pervades the debate on asymmetric conflict has not taken into consideration the simple fact that many of the groups engaged in asymmetric conflicts do not expect to win. Bargaining theory, the pre-­eminent model of studying conflict in both political science and economics, focuses heavily on the outcome of conflict, and to a large extent ignores the dynamics of conflict. A focus on the outcome in a conflict naturally implies an expectation that the stronger side will win. A uni-­ dimensional focus on the balance of power neglects that even when faced with an overwhelming enemy, an insurgency could still retain the ‘power to hurt’.36 Hence, spending more in Afghanistan should generate a victory. This is hardly the case for the Taliban for whom the conflict itself involves opportunities for income, such as from opium production.37 The Taliban therefore are in no rush to end the conflict. In stark contrast, the ISAF forces are under constant political pressure to withdraw. The conflict has become immensely unpopular in the home countries of the ISAF forces, and in the long run, as it has already been announced, that ISAF will pull out of Afghanistan. Thus, the Taliban know that as long as they can keep fighting, they will eventually succeed. When facing an enemy who does not derive benefits from fighting, as does the Taliban when fighting ISAF, this ability to hurt is fundamental and creates a situation where some gains for the weaker side, and a lack of gains for the stronger side, both substantially benefit the weak side. Moreover, a balance-­ofpower focus naturally leads to an analysis of the number of boots on the ground, leading to what people have called a ‘surge mentality’. However, as long as the Taliban benefit from the fighting itself, ISAF would have to deploy enough forces to completely subdue them to achieve a decisive victory. In reality, this would entail the deployment of many hundreds of thousands of troops. There is absolutely no political will for this among the coalition partners. The Taliban know this, and the consequence is that a surging of troops did not have the intended effect. Here we see the interaction of geography with tactics. In Iraq, where the population is much more heavily centralized, a situation was created whereby fewer troops were needed to control more or less the entire country and population. In Afghanistan, an order of magnitude more forces would be needed to achieve the same results.

38

War and State-Building in Afghanistan

Conclusion: Afghanistan, the end of the beginning or the beginning of the end? From the time of Alexander’s army to ‘Obama’s War’, Afghanistan has tested the limits of great powers. For our period (1550–2014), Islam has been an ever present factor, while the tribal structure has defined patterns of mobilization and recruitment. From these cultural and geographical constants, we can compare the experiences of foreign forces in Afghanistan across time. Such a comparison offers some lessons, including: a conventional military operation in Afghanistan is a sure recipe for disaster. If one intervenes, the invading power burns its fingers. If Afghanistan is neglected, then troubles from the Afghan border can spill over. If the US and NATO treat Afghanistan as a pawn in the New Great Game with a declining Russia and a resurgent China, then the present chaos will be followed by catastrophe. In response to a realization that the military strategy under the Obama Administration is afflicted by conflicting goals of force protection, civilian protection (and hearts and minds) and defeating the enemy, a political solution has received stronger support. Peace-­ building requires the painstaking design of credible institutions for civilian, and preferably democratic, rule. These issues of governance do not replace, but are superimposed upon, conflict resolution and prevention. A durable peace requires a sustained provision of public goods as well as measures to contain and prevent violence.

Acknowledgements This research received support from the Norwegian Ministry of Defence. The Norwegian Ministry of Defence also provided support for the workshop, ‘COIN in Afghanistan: From the Mughals to the Americans’, 12–13 February 2012, Oslo, Norway. Comments received from the participants at the conference significantly improved this chapter.

Notes 1 Asymmetric warfare is defined in myriad ways. Here we feature material disparity. Different researchers feature different strength ratios. For example, Ivan ArreguínToft in ‘How the Weak Win Wars: A Theory of Asymmetric Conflict’, International

Continuity and Change in Asymmetric Warfare

39

Security, vol. 26, no. 1 (2001), pp. 93–128, and T.V. Paul in Asymmetric Conflicts: War Initiation by Weaker Powers (Cambridge: Cambridge University Press, 1994) use a power ratio of one to two. The precise power ratio in Afghanistan for the Mughal, British, Soviet and American interventions all exceed ten to one. 2 For a fuller treatment, see Kaushik Roy, War and Society in Afghanistan: From the Mughals to the Americans, 1520–2012 (New Delhi: Oxford University Press, forthcoming). 3 Andrew Mack, ‘Why Big Nations Lose Small Wars: The Politics of Asymmetric Conflict’, World Politics, vol. 27, no. 2 (1975), pp. 175–200. 4 Ivan Arreguín-Toft, ‘How the Weak Win Wars: A Theory of Asymmetric Conflict’, pp. 93–128. 5 Christopher Butler and Scott Gates, ‘Asymmetry, Parity, and (Civil) War: Can International Theories of Power Help Us Understand Civil War?’, International Interactions, vol. 35 (2009), pp. 330–40. 6 Kaushik Roy, ‘Horses, Guns, and Governments: A Comparative Study of the Military Transition in the Manchu, Mughal, and Ottoman and Safavid empires, circa 1400 to circa 1750’, International Area Studies Review, vol. 15, no. 2 (2012), pp. 99–121. 7 John Mearsheimer, Liddell Hart and the Weight of History (Ithaca, NY: Cornell University Press, 1988). 8 Butler and Gates, ‘Asymmetry, Parity, and (Civil) War’, p. 333. 9 Ibid. 10 Allan C. Stam, Win, Lose, or Draw: Domestic Politics and the Crucible of War (Ann Arbor, MI: University of Michigan Press, 1996); Arreguín-Toft, ‘How the Weak Win Wars: A Theory of Asymmetric Conflict’, pp. 93–128. 11 Mathematically, the basic CSF can be expressed as:

. S. Skaperdas, ‘Contest

Success Functions’, Economic Theory, vol. 7, no. 2 (1996), pp. 283–90; S. Skaperdas, ‘On the Formation of Alliances in Conflict and Contests’, Public Choice, vol. 96, nos 1/2 (1998), pp. 25–42. 12 The Second World War could be taken as the almost logical culmination of Total War, a process which, despite teleological critique, could be traced back to the onset of the French Revolution of 1789 and the Industrial Revolution of the nineteenth century. 13 Jack Hirshleifer, ‘The Paradox of Power’, Economics and Politics, vol. 3 (1991), pp. 177–200; Jack Hirshleifer, The Dark Side of the Force: Economic Foundations of Conflict Theory (Cambridge: Cambridge University Press, 2001). 14 Christopher Butler and Scott Gates, ‘African Range Wars: Climate, Conflict, and Property Rights’, Journal of Peace Research, vol. 49, no. 1 (2012), pp. 23–34. 15 Halvard Buhaug and Scott Gates, ‘The Geography of Civil War’, Journal of Peace Research, vol. 39, no. 4 (2002), pp. 417–33.

40

War and State-Building in Afghanistan

16 Halvard Buhaug and Jan Ketil Rød, ‘Local Determinants of African Civil Wars, 1970–2001’, Political Geography, vol. 25, no. 3 (2006), pp. 315–35. 17 Halvard Buhaug, Scott Gates and Päivi Lujala, ‘Geography, Rebel Capability, and the Duration of Civil Conflict’, Journal of Conflict Resolution, vol. 53, no. 4 (2009), pp. 544–69. 18 Halvard Buhaug, ‘Dude, Where’s My Conflict? LSG, Relative Strength, and the Location of Civil War’, Conflict Management and Peace Science, vol. 27, no. 2 (2010), pp. 107–28. 19 K.E. Boulding, Conflict and Defense: A General Theory (New York: Harper & Row, 1962). 20 J. Herbst, States and Power in Africa: Comparative Lessons in Authority and Control (Princeton, NJ: Princeton University Press, 2000). 21 Ivan Arreguín-Toft, How the Weak Win Wars: A Theory of Asymmetric Conflict (New York: Cambridge University Press, 2005); Idean Salehyan, ‘Transnational Rebels – Neighboring States as Sanctuary for Rebel Groups’, World Politics, vol. 59 (2007), pp. 217–42. 22 Philippe Le Billon, ‘The Political Ecology of War: Natural Resources and Armed Conflicts’, Political Geography, vol. 20, no. 5 (2001), pp. 561–84. 23 Arreguín-Toft, How the Weak Win Wars. 24 Le Billon, ‘The Political Ecology of War’, pp. 561–84. 25 Kristian S. Gleditsch, ‘Transnational Dimensions of Civil War’, Journal of Peace Research, vol. 44, no. 3 (2007), pp. 293–309; Stephen M. Saideman, ‘Discrimination in International Relations: Analyzing External Support of Ethnic Groups’, Journal of Peace Research, vol. 39, no. 1 (2002), pp. 27–50; Idean Salehyan, ‘Transnational Rebels’, World Politics, vol. 59, no. 2 (2007), pp. 217–42. 26 Buhaug, Gates and Lujala, ‘Geography, Rebel Capability, and the Duration of Civil Conflict’, pp. 544–69; James D. Fearon, ‘Why do Some Civil Wars Last so Much Longer than Others?’, Journal of Peace Research, vol. 41, no. 3 (2004), pp. 275–301; Päivi Lujala, ‘The Spoils of Nature: Armed Conflict and Rebel Access to Natural Resources’, Journal of Peace Research, vol. 47, no. 1 (2010), pp. 15–29. 27 Fearon, ‘Why do Some Civil Wars Last so Much Longer than Others?’, pp. 275–301. 28 Paul Collier and Anke Hoeffler, ‘Greed and Grievance in Civil War’, Oxford Economic Papers, vol. 56, no. 4 (2004), pp. 563–95. 29 Mack, ‘Why Big Nations Lose Small Wars’, pp. 175–200. 30 Hirshleifer, The Dark Side of the Force. 31 Peter W. Chiarelli and Stephen M. Smith, ‘Learning from Our Modern Wars: The Imperatives of Preparing for a Dangerous Future’, Military Review (September/October 2007) [http://www.army.mil/professionalWriting/volumes/ volume6/january_2008/1_08_1.html; accessed 22 February 2013]. 32 Douglas A. Macgregor, Breaking the Phalanx (New York: Praeger, 1997). 33 John R. Boyd, The Essence of Winning and Losing (1986/2010) [http://tobeortodo. com/wp-­content/uploads/2011/11/essence_of_winning_losing.pdf].

Continuity and Change in Asymmetric Warfare

41

34 Rupert Smith, The Utility of Force (New York: Penguin, 2007). 35 Michael Capstick, ‘Reviewing Canada’s Afghan Mission’, Policy Options (April 2008) [http://www.irpp.org/po/; accessed 15 February 2013]. 36 Branislav L. Slantchev, ‘The Power to Hurt: Costly Conflict with Completely Informed States’, American Political Science Review, vol. 97 (2003), pp. 123–33. 37 Jo Thori Lind, Karl Ove Moene and Fredrik Willumsen, ‘Opium for the Masses? Conflict-­induced Narcotics Production in Afghanistan’, Review of Economics and Statistics (2013; DOI: 10.1162/REST_a_00418).

2

Great Mughals, Warfare and COIN in Afghanistan, 1520–1707 Kaushik Roy

Introduction Afghanistan has an area of about 250,000 square miles. The north and the east consist of a mountain system that emerges from the tablelands of the Pamirs.1 The eastern fringe of the country is mountainous, especially the north-­east where the protruding Wakhan Corridor joins the Pamirs. The elevation here frequently exceeds 10,000 feet and forests appear at intermediate elevations. The central mountainous region, the Hindu Kush, obstructs travel across the centre of the country. The Turkoman Plain, characterized by sandy desert and scattered scrub grasses, dominates the northern edge of Afghanistan. The Herat-Farah Lowlands in the west are part of the Iranian Plateau and some regions of it are suitable for cultivation. South-­western Afghanistan is mostly sandy desert.2 Iran’s frontier lies west of Herat, extending from Zulfiqar in the north to Koh-­iMalik Siah in the south, a distance of about 400 miles.3 The Kabul River forms part of the Indus River system, whereas the dry western and south-­western parts of Afghanistan geographically are a continuation of the Iranian Plateau. The area north of the Hindu Kush, known as Afghan Turkestan, is part of the Central Asian plains.4 In the absence of navigable rivers, the principal means of transport in Afghanistan until around 1900 were mules, camels and pack-­horses.5 The origin of the Afghans is still debated. The Iranians, Greeks and Scythians (Aryan stock) settled in Afghanistan between the third and first century bce. The nineteenth-­century British scholar-­officials of the Raj (British Government in India) believed in the Armenian and Jewish origin of the Afghans.6 H.W. Bellew, the Sanitary Commissioner of Punjab in an article dated 1881 asserted that the Afghans were partly Israelites – the Afghans belonged to the ‘lost tribes of Israel’. The Assyrian ruler Tiglath Pilesar captured many Israelites and they

44

War and State-Building in Afghanistan

Map 4  Mughal–Safavid Uzbek Great Game in Afghanistan, c. 1550–1700 Borders

Great Mughals, Warfare and COIN, 1520–1707

45

Map 5  Theatre of Yusufzai Uprising, 1585–6

were settled in Ghor in around 700 bce. The Israelites intermixed with the ‘Arians’ (Bellew meant Aryans) who were the original inhabitants of Afghanistan. Bellew divided the Arians into eastern (Indian) Arian and western (Persian) Arian. The Persian Arians were Tajiks and inhabited western Afghanistan. The Arians who inhabited eastern Afghanistan were later known as Pathans. They were very similar to the Kshatriyas (Rajputs) of India. And in the mountainous region lived the Turanian (Mongol) people. Furthermore, intermixture occurred with the arrival of the Scythians (Sakas, Parthians, Huns, etc.) from north of the River Oxus into Afghanistan in the Common Era (ce). The Scythians settled in particular in Seistan and in the Kandahar region. Thus overall, in Bellew’s view, the Mongols were in northern Afghanistan, the Arians in southern Afghanistan and the Semitic people were sandwiched between them in central Afghanistan.7 Writing around the tenth century ce., one Muslim chronicler noted that the inhabitants of Kabul were Buddhists, Jews and Muslims. The Arabs entered

46

War and State-Building in Afghanistan

Afghanistan after the seventh century ce. Nirodbhusan Roy asserts that the Afghan ‘race’ sprang from the intermixture of Scythian and Jewish colonists. In the nineteenth century, the peoples who inhabited eastern Afghanistan, especially the region west of the Indus River, were called Pathans by British officials. Today, they are called Pashtuns. According to one tradition, the word Pathan is a corrupted version of Pahtan that was derived from the word Batan, which means delivered, set free.8 The Pathans would show this by repeatedly rising up against various foreign regimes throughout history. From the medieval era, most inhabitants of Afghanistan were Muslims, the majority belonging to the Sunni faith. Among the various tribes, the Durranis from Zamindwar and Ghor were dominant. Many of them settled in Derajat and Multan in India.9 Until the nineteenth century, the Afghans were addicted to a local wine, referred to as ‘grape juice’. It was purified in clay vats and then boiled and poured into goatskins. It was said to improve after two years and one British officer declared during the early nineteenth century that it tasted like Madeira. Most Afghans were intensely Islamic and generally wore long beards.10 Jos Gommans writes that the pre-­modern Afghans had the triple identities of being shepherds, traders and cavalrymen. The Afghans of the Gilzai and Lodhi tribes participated in long-­distance trade with Central Asia and India.11 This chapter describes attempts at empire-­building in Afghanistan under the great Mughal rulers. For almost two centuries, the Mughals engaged in both conventional and unconventional (counterinsurgency) operations against external enemies as well as the Afghans in order to check unrest and establish public order.

Beginning of Mughal intervention in Afghanistan: from Babur to Humayun, 1504–56 The Mughal intervention in Afghanistan began with the Chaghtai Turk named Zahir-­ud-din Muhammad Babur, the founder of the Mughal Empire. In June 1494, at the tender age of 11, Babur (b. 14 February 1483, d. 1530) became the ruler of Ferghana. Ferghana is an oasis between the deserts of Khiva and the Takla Makan and now in Uzbekistan. To the east of Ferghana is Kasghar, to the west Samarkhand and to the south the mountains of Badakshan.12 When Babur was 19, he captured Samarkhand.13 In 1501, at the Battle of Sar-­i-Pul, the Uzbek Sirdar Shaibani Khan defeated Babur.14 In 1502, Shaibani Khan was able to mobilize some 60,000 soldiers.15 He was the most powerful man in Central

Great Mughals, Warfare and COIN, 1520–1707

47

Asia. After failing to found a kingdom in Central Asia, in 1504 Babur moved into Afghanistan. Due to Shaibani Khan’s depredations, many Badakshanis also joined Babur’s party.16 Babur provides a description of Afghan defensive combat techniques: ‘I first heard the word sangar after coming to Kabul where people describe fortifying themselves on a hill as making a sangar.’17 At that time, Sayyid Ali Khanwas was the ruler of Kabul. He owed allegiance to Sultan Sikander Lodhi of Delhi (r.  1489–1517).18 By September 1504, Babur had captured Kabul (the biggest city in Afghanistan) and made it the capital of his Afghan principality. In his memoir, Babur describes Kabul as a multi-­ethnic city where various languages were spoken, including Arabi, Farsi, Turki, Hindi, Afghani, Lamghani, Pashai and Paraji. Afghans, Turks, Mongols, Arabs and Iranians, some of them peasants, others merchants who moved in armed caravans, inhabited Kabul.19 Ghazni (at that time also known as Zabulistan) was captured in October 1504.20 At that time, Herat was under the Timurid Sultan Husain Bauqara, whose sons controlled parts of Khorasan.21 In 1505, Babur captured the Fort of Khilat-­i-Ghilzai, situated midway between Ghazni and Kandahar. Babur’s victory was due to his generalship as well as the technological advantage that his army enjoyed over the tribes of Afghanistan. While the Afghan tribes were mostly armed with bows, Babur’s soldiers possessed firearms.22 Babur’s Kabul vilayat (realm) extended from Lamghan in the east including Peshawar to Kohistan in the west comprising the mountainous region of Ghor. In the north, Babur’s control extended to the Hindu Kush Mountains including Andarab and Kunduz provinces, and in the south to Farmul, Bannu, Nanghar and the Suleiman Mountains south-­east of Kabul.23 Babur rewarded his loyal begs (chieftains) by distributing the villages to them. In 1506, Babur conducted a COIN campaign against the rebellious Hazaras, who practised pastoral nomadism. The Hazaras are actually a Central Asian Turkish group who inhabited the Hazara District east of the River Indus. As punishment, he collected sheep and horses from them and about 80 Hazaras were executed. While coming back from the Hazara expedition, Babur collected revenue from Baran. A detachment under the joint command of Jahangir Mirza and Kasim Beg stormed the sangars on the hilltops that were defended by the rebellious Afghans.24 Babur levied 400,000 shahrukis (coins) on Bhira on the left bank of the River Jhelum as the price for protection and presented the region to Hindu Beg, assigning its revenues to him for maintaining the troops that policed the region. The headmen of Bhira were assembled and Babur fixed the ransom as

48

War and State-Building in Afghanistan

1,000 shahrukis per head. Khushab on the right bank of the Jhelum was assigned to Shah Hasan. Shah Hasan was ordered to assist Hindu Beg in maintaining law and order in the region.25 Nevertheless, Babur did not tolerate his soldiery unnecessarily disturbing the Afghans. For instance, in February 1519, when it was reported to him that some of his soldiers without reason were robbing the people of Bhira, he ordered the ringleaders to be executed. And the other guilty soldiers had their noses slit before being marched through the camp and discharged from the army.26 Babur also adopted a marital alliance in order to co-­opt potentially rebellious subjects within his realm. For example, Babur married the daughter of the Yusufzai sirdar in order to conciliate the Yusufzais. Actually, in 1519, Babur was contemplating an expedition against the Swatis and the Yusufzais. But, Taus Khan, the younger brother of Shah Mansur (the chieftain of the Yusufzais) bought the latter’s daughter and submitted to the Mughals.27 During June 1507, Shaibani Khan took Herat, which posed a threat to Persia as well as to Babur’s fledgling kingdom of Kabul.28 Initially, Babur attempted to build a power base in Kabul in order to defend Timurid power south of Amu Darya and then build a state in Central Asia. By 1512, he failed to achieve these two goals and turned his attention to the east (i.e. South Asia).29 Since Afghanistan did not yield much revenue, the only way for Babur to ward off the threats to his Kingdom of Kabul from Central Asia and Persia was either to launch plundering raids into the Punjab and north India or better extend his rule to west and north India. In 1519, Babur raided Bajaur,30 which later was to operate as the staging post for military operations in north-­west India. Afghanistan was important for the rulers of north India because it was the transit point for India’s horse trade. The climate of India was not suited for breeding high-­quality horses. And warfare in Central and South Asia placed a high premium on good-­quality horses. So, horse merchants moved about 10,000 horses annually through Kabul into north India. The horses were brought from Balkh and Turkestan and fattened in Kabul. The dry and sandy soil of the Central Asian arid zone generated high-­quality Turkoman horses. This trade was highly lucrative for the merchants as they were able to make a profit of 300–500%.31 The rulers of India required a constant supply of horses not only to replace the casualties of war but also because the horses did not flourish in the climate of India. The heat, dust and heavy rainfall of Hindustan did not suit the Central Asian horses. Babur jotted in his memoirs: ‘Sometimes it rains 10, 15 or 20 times a day; torrents pour down all at once . . . The fault is that the air becomes very soft and damp.’32 By imposing taxes, the ruler of Afghanistan could raise the price of military horses in India. Even worse, by stopping the horse trade a hostile ruler

Great Mughals, Warfare and COIN, 1520–1707

49

of Afghanistan could choke the supply of military horses to India, making her militarily weak. In addition, until the British conquest of India, all invasions of the subcontinent had taken place through Afghanistan. During the pre-British era, there were four roads between Kabul and India. One was through the Khaibar Mountains. After crossing the Khyber Pass from Afghanistan, the first big city on the Indian side was Peshawar in west Punjab. The Nur-Dara Mountains near the Safed Koh is in the neighbourhood of Khaibar. Another road was by way of Bangash. The third was by way of Naghar. Naghr is Baghzan, about 70 miles from Kabul. And the fourth road was through Farmuli along the Urghun.33 Hence, the security of India demanded control over Afghanistan. In 1525, Babur invaded India with 12,000 soldiers (Turks, Tajiks, etc.) from Afghanistan.34 Babur’s army comprised many Afghan soldiers, especially Hazaras and Baluchis.35 Babur politically co-­opted the potentially rebellious Hazaras by recruiting them in his army of invasion. The prospect of pillage and plunder of Hindustan had always attracted the Afghans and had prevented any uprising by the ‘unruly’ tribes within Afghanistan against the central government. Military recruitment for the Afghans was important because only 12 per cent of Afghanistan’s land is arable even today.36 Political co-­option of potentially rebellious Afghans through military recruitment was also a technique followed by the British. Babur’s eldest son Humayun recruited troops from Badakshan for the army that was preparing to invade Hindustan. The Badakshanis were experts in mounted archery.37 While proceeding towards India, in January 1526, Babur took the Fort of Bajaur with the aid of firearms.38 At that time, north India was under the Lodhi Dynasty (1450–1526). The Lodhis were also Afghans who had migrated to India during the thirteenth century. From the thirteenth century onwards, Afghans from the Roh, Suleiman and Ushtughar mountain ranges moved into India in search of military employment. Many of them settled in the Ganga-Jamuna doab and formed military outposts of the Muslim sultanates in north India. Roh is roughly the region from Bajaur to Siwi in Bhakkar and from Hasan Abdal to Kabul. Those Pathans/Pushtuns who hailed from Roh and settled in the western part of north India (Rohilkhand) were known as Rohillas.39 As Babur moved into west Punjab, two Lodhi nobles named Daulat Khan and Alam Khan raised an army of about 30,000 from mostly the mercenary Afghans. Thus, we see that the Afghans, not for the last time, were fighting on both sides for pay.40 Daulat Khan, the Lodhi Governor of Punjab, was at odds with Sultan Ibrahim Lodhi due to the latter’s centralizing tendencies. In fact, Daulat Khan’s son Dilawar Khan invited Babur

50

War and State-Building in Afghanistan

to attack the Lodhi Sultanate.41 Babur raised the money for the Indian campaign by plundering Peshawar and from the 10,000 gold ashrafis given to him by Daulat Khan. On 21 April 1526, in the First Battle of Panipat, Babur defeated and killed Sultan Ibrahim Lodhi of Delhi (r. 29 December 1517 to 21 April 1526). According to one estimate, while the Mughal Army was 24,000 strong, Sultan Ibrahim commanded 50,000 men and 2,000 elephants. However, the Lodhi Army was marked by internal dissensions.42 To a great extent, Ibrahim’s downfall was due to clan rivalries among the Afghans. The various Afghan clans such as the Lodhis, Sahukhails, Nuhanis and Sarwanis fought among each other.43 Clan rivalries prevented the emergence of a strong state in Afghanistan in the historical context. After the battle, Babur captured Delhi and Agra and founded the Mughal Empire in India.44 Better equipment (crossbows in the hands of mounted archers and handheld firearms for the infantry)45 and Babur’s generalship resulted in Mughal victories over the Lodhis. Not for the last time, a warlord from Afghanistan had established an empire over Hindustan. In 1526, Babur’s empire included Kabul, Kandahar, Kunduz, Badakshan and the Delhi– Agra region.46 Babur’s position was anything but secure. He had written in his autobiography that while at his rear the Uzbek chieftains posed a threat, in the east, Bihar and Bengal remained under the Afghan sirdars. The Uzbeks controlled Turan, which means Uzbekistan, Tajikistan and Kazakhstan.47 Babur had good reason for feeling insecure. He commented that if the Uzbek sirdars (chieftains) united, they would be able to mobilize about 100,000 cavalry. Luckily for Babur, the various Uzbek sirdars were busy fighting against each other. For instance, in October 1526 the Uzbek sirdar Ubaidullah Khan of Bokhara attacked Merv.48 According to a Persian estimate, in 1528 the Uzbek Khanate had some 80,000 standing troops and some 40,000 auxiliaries. The Uzbek army also included the Mongols who were settled between the Volga River and Kasghar.49 In 1539, Ubaidullah, the Uzbek Khan died. In 1545, Abdul Aziz declared himself an independent ruler despite the reigning khan being alive.50 In 1530, Babur was succeeded by his eldest son Nasir-­ud-din Humayun (1508–56). Humayun’s brothers got parts of Babur’s empire. The brothers were practically independent in their principalities but accepted the nominal sovereignty of Humayun. Mirza Kamran (died 1557) obtained Kabul and Kandahar as his fief and Badakshan was given to Mirza Sulaiman. In 1535, when Humayun was busy annexing Gujarat, the Safavids threatened Kabul. Mirza Kamran marched from Lahore to Kabul and on 25 January 1536 defeated Sam Mirza, the brother of Shah Tahmasp, the Safavid monarch. Meanwhile, Humayun

Great Mughals, Warfare and COIN, 1520–1707

51

had to deal with the rising power of Sher Khan, the Afghan warlord of Bihar. Sher Khan belonged to the Afghan tribe named Sur. His grandfather was an Afghan horse dealer who settled in Bihar. At that critical juncture, Kamran, who commanded 20,000 cavalry and controlled the region between Kabul and Zamindawar (north-­west of Kandahar) in the north to Sirhind in the south, refused to come to Humayun’s aid. On 17 May 1540, Sher Khan defeated Humayun at the Battle of Chausa. Humayun pursued by the Suri forces retreated to Agra and then through Sirhind to Sind in 1541. In September 1541, Humayun marched into Thatta (Sind). However, he failed to take the Shewan Fort because the ruler of Thatta prevented any grain from reaching the besieging Mughal force by laying to waste the surrounding countryside.51 Meanwhile, Sher Khan proclaimed himself ruler of Hindustan and took the title of Sher Shah.52 In 1543, the fugitive Humayun reached Seistan, an eastern province of Persia, an area of 300 by 160 miles. The western part of Seistan consists of a vast arid plain. In 1544, Shah Tahmasp, the ruler of Persia, granted an audience to Humayun and agreed to provide the latter with 14,000 Turcoman cavalry for recapturing Kabul.53 Humayun’s aim was to capture Kabul and use it as a launching pad to reconquer the lost Mughal dominion in Hindustan. With the Persian force, Humayun marched from Seistan towards Kandahar, which was held by his rebellious brother Mirza Askari (1518–51). Kandahar was captured in 1545 and Bairam Khan, a noble loyal to Humayun, was put in charge of the fort. Then, Humayun raised a force from the Afghans. In accordance with the deal struck between Shah Tahmasp and Humayun, Kandahar Fort was turned over to the Persians. However, Humayun planned to acquire Kandahar for himself.54 Then, Humayun turned against his rebellious brothers to establish his rule in Afghanistan. The Afghans joined both sides as mercenaries during the Mughal civil war in Afghanistan. Kamran laid siege to the Fort of Zamindawar with the help of Khizr Khan, a Hazara chieftain. However, Kamran’s attempt was foiled due to the vigorous action of Bairam Khan and the reinforcements sent by Humayun. The Hazaras then deserted Kamran. Throughout history, Afghan mercenaries have displayed the tendency to desert the losing side and join the other even in the midst of a conflict. In 1545, Humayun marched towards Badakshan, which was held by Mirza Sulaiman. However, Humayun fell ill at Shakhdan and the Mughal officials panicked and abandoned their posts in Badakshan.55 In 1546, during winter when snow was falling, Humayun marched towards Kabul with a cavalry force and baggage carried by camels. Jouher, the Private

52

War and State-Building in Afghanistan

Secretary of Humayun had written in his history: ‘such a quantity of snow had fallen that the roads were quite blocked up; we were therefore obliged to ram the snow well down; after which the horses and camels were enabled to move on’.56 In fact, Babur points out in his memoirs that snow made all but one of the passes of Hindu Kush impassable for four to five months during the winter. Before the construction of the Salang Tunnel in 1964 with Soviet aid, only the Shibar/ Shibartu Pass could be crossed in winter for reaching Kabul from north Afghanistan. This pass was reached through Charikar and the Ghorband Valley and led into Bamian. After the snow started to melt in April, the rivers in the narrow mountain valleys were flooded, which in turn prevented deployment of large forces for another two months. Thus, the months of May to October seemed to be best for conducting military operations in the Kabul region. But, Babur warned in his memoir that Kabul could be reached from the west through the Herat–Kandahar road, which was comparatively flat and easily traversed even during the winter.57 At the vicinity of Kabul, Humayun’s force was attacked by a Afghan mercenary named Sher Afghan (not to be confused with Sher Khan/ Sher Shah Suri) and Khizr Khan Hazara, who were in the pay of Kamran. However, Humayun defeated the Afghan mercenaries.58 In April 1547, after a siege of three months, Humayun was able to capture Kabul from his brother Mirza Kamran.59 Kamran escaped to Ghakkar territory but the Ghakkar chieftain handed him over to Humayun based on the promise of a large reward. Humayun later blinded Kamran.60 Thus, we see that financial incentives sometimes overcame the frontier tribes’ sense of hospitality to a fugitive in their territory. Not all the tribes accepted Humayun’s rule without a struggle. In 1548, one Mughal official named Ulugh Mirza was killed during an encounter with the Hazaras.61 In 1553, Humayun captured Peshawar and ordered the construction of a fort there for pacification of the surrounding region.62 Sher Shah, the founder of the short-­lived Suri Empire, was aware that the Mughals might stage a comeback. He had formulated a strategy to prevent this, but died suddenly in 1545 due to the accidental explosion of a mine during the Siege of Chanderi. Sher Shah’s unworthy successors failed to implement Sher Shah’s defensive strategy. In Sher Shah’s own words: . . . I wished to have depopulated the country of Roh, and to have transferred its inhabitants to the tract between the Nilab and Lahore, including the hills below Ninduna, as far as the Siwalik; that they might have been constantly on the alert for the arrival of the Mughals, and not allow anyone to pass from Kabul to Hind, and that they might also keep the zamindars of

Great Mughals, Warfare and COIN, 1520–1707

53

the hills under control and subjection. Another is to have entirely destroyed Lahore, that so large a city might not exist on the very road of an invader, where immediately after capturing it on his arrival, he could collect his supplies.63

Sher Shah’s scheme of ethnic transplantation and a scorched earth policy for checking both the conventional threat of a foreign power and to suppress the unruly frontier chieftains was novel in the medieval context. In the 1980s, the Soviets pursued an ethnic transplantation policy in Afghanistan to suppress the mujahids (warriors of Islam). In 1555, Humayun crossed into India and reached Sirhind. Meanwhile, Sultan Sikander Suri (Sher Shah’s successor) advanced with an army of about 80,000 cavalry. Humayun’s army meanwhile had also recruited some Uzbek freebooters. In the ensuing battle, a charge at the rear of the Suri army resulted in its dissolution.64 The Mughal victory at the Battle of Machiwarra on the bank of the Sutlej was made possible due to the deployment of mounted archers. Since Humayun controlled Afghanistan, unlike the Lodhis he could recruit mounted archers.65 While Humayun was engaged in re-­establishing the Mughal Empire in north India, one of his officers, Selim Khan, built a fortress with four strong forts at Mankot. Each fort was constructed on the top of a hill. The forts were built of stone and mortar. The forts had a good water supply and were very difficult to approach. A garrison was installed in the fortress to pacify the warlike Ghakkar tribe. The fortress was also supposed to act as a refuge for the Mughal Army in case it met with defeat in north India and had to retreat to Kabul. After Humayun’s death, from his base in this fortress Selim Khan started plundering the environs of Lahore. However, he soon died a natural death. The fortress came under control of the rebel Sikander. Humayun’s son Akbar sent the imperial army to capture the fortress. The Mughals constructed redoubts and drew lines of circumvallation around the fort. However, the rebel garrison defended itself with cannons and muskets.66

Pacification and consolidation of Mughal rule in Afghanistan from Akbar to Aurangzeb: 1556–1707 Under Sher Shah, the Salt Range in north-­west Punjab was the border of his empire and the north-­west frontier defence was centred round the Rohtas Fort.67

54

War and State-Building in Afghanistan

Under Mughal Emperor Akbar (ruled 1556–1605), the north-­west frontier defence centred round the Attock Fort which guarded the main ferry across the Indus. Akbar realized, writes M. Athar Ali, that Mughal India’s ‘scientific frontier’ should run along the Kabul–Kandahar axis. So, Akbar shifted the principal line of defence from the Indus further to the north-­west across the Hindu Kush Mountains.68 Later, some hardliners of British India also accepted the Kabul– Kandahar axis as the scientific frontier of India. Akbar pre-­empted the British in realizing that Sind constituted the crucial southern component of the ‘scientific frontier’. The route from India to southern Afghanistan (i.e. Kandahar) passed through Siwi and the Bolan Pass, which made Sind a strategic prize. Furthermore, sulphur, which was required for the manufacture of gunpowder, was available in Sind.69 The Mughals had to deal with three clan groupings in Sind, the largest of which was the Kalmati Baluch tribe. One section among them inhabited Siwistan and Lakhi and they maintained 20,000 cavalry. Another section of this tribe inhabited the Karah Hills near Kachh Gandawa and was able to mobilize about 4,000 cavalry. The Nahmardi who inhabited the territory from Sehwan to the Kirthar Range had 3,000 cavalry and 7,000 infantry, while the Baluch Nazhari had about 1,000 cavalry.70 In 1586, the Mughals invaded Sind. The Mughal commander Sadiq Khan defeated Jani Beg’s force and then marched towards Sehwan and besieged the fort there. The garrison was well equipped with cannons and muskets. And the guns at the fort proved too much for the besieging Mughal Army. Mughal mining techniques proved to be of no avail. The earthen parapet of the fort was quite high and the defenders constructed a new wall. Thus, Sadiq Khan was forced to retreat. Furthermore, Mirza Jani Beg’s war boats equipped with cannons and musketeers harassed Sadiq Khan’s force, which had retreated near the Lakhi Hills. In 1590, another Mughal force under Khan-­i-Khanan was sent to Sind. Meanwhile, Jani Beg had bolstered his army with the aid of Ottoman and Portuguese mercenaries. In response, the Mughals made greater use of cannons to destroy Jani Beg’s war boats.71 After Humayun’s death, Kabul fell into chaos. Mirza Sulaiman, the ruler of Badakshan, his wife Haram Bagum and their son Mirza Ibrahim planned to occupy Kabul. Munim Khan, the Mughal noble was in charge of Kabul. When he learnt of the advance of the Badakshanis, he strengthened the bastions and the bulwarks of the fort. With cannons and muskets, the garrison fought the 10,000 Badakshanis. Meanwhile, the main Mughal Army was engaged in the Delhi– Agra region against the Hindu General Hemu. Many Afghans fought on the side of Hemu. On 5 November 1556, at the Second Battle of Panipat, Hemu was

Great Mughals, Warfare and COIN, 1520–1707

55

defeated and Mughal rule was secured in north India. Before the main Mughal Army in north India could redeploy in Afghanistan, without any substantial aid from the central government, the Mughal regional government in Kabul was able to defeat the Badakshanis. Nevertheless, troubles continued in Afghanistan. Khan Zaman rebelled in Zamindawar. Khan Zaman’s brother Bahadur Khan gathered an Afghan militia and tried to capture Kandahar, which was Emperor Akbar’s regent Bairam Khan’s jagir (land given in lieu of salary). When Shah Muhammad (Bairam Khan’s deputy) who was in charge of the Kandahar Fort heard of the impending invasion by Bahadur Khan, the former realized that Akbar was not in a position to send help from India. In desperation, Shah Muhammad requested assistance from the ruler of Persia. This was an instance of weak Mughal decision-­making, encouraging Persian intervention in Afghanistan. The Shah of Persia sent 3,000 Turcoman cavalry gathered from Seistan, Farah and Gurmsir under the command of Ali Yar Beg Afshar. Afshar’s force defeated Bahadur Khan’s Afghan auxiliaries. Shah Muhammad however did not hand over Kandahar to the Persians. In the future, the Persians would attempt to capture Kandahar whenever the Mughals were engaged in troubles in India. Bahadur Khan submitted to Akbar and the latter transferred him to a jagir in Multan in order to prevent him doing any further mischief in Afghanistan. To exert greater control over Kabul, Akbar decided to appoint his court official Munim Khan as the guardian of Mirza Hakim, who was Akbar’s brother and who ruled in a semi-­autonomous manner.72 Due to Akbar’s centralizing administrative measures and religious reforms, a large chunk of the Mughal nobility rebelled against Akbar. Many rebellious nobles were Uzbeks. The Uzbeks were quite powerful in Balkh, Bokhara, Khorasan and Ferghana. Many immigrants from these regions had entered Akbar’s service.73 In 1563, Iskander Khan became the ruler of the Uzbek Khanate and in 1583, his son Abdullah Khan took over.74 By early 1581, Akbar was able to crush the rebellion by the Mughal nobles against his authority. Then, Akbar decided to cross swords with his brother Mirza Hakim, the ruler of Kabul. When Akbar was engaged against the rebels in north India, Hakim had invaded Punjab. Akbar marched through Ambala and Sirhind. When Akbar reached Payal beyond Sirhind, he heard the news that Hakim had evacuated Punjab. Akbar then crossed the Sutlej and Beas rivers via a bridge of boats. He avoided the direct road through Lahore in order to keep near the base of the hills. The River Ravi was crossed using a bridge of boats but when the army reached the River Chenab, boats were at a premium. A collection of ferry boats for transporting the whole army took about three days. The Mughal Army reached the Rohtas Fort,

56

War and State-Building in Afghanistan

which was held by Yusuf against Hakim. On arrival at Indus, Akbar stopped for 50 days. This was necessary because the construction of a bridge during the rainy season was impossible and a small force taking a defensive position on the opposite bank could have prevented the passage of the army across the flooded Indus. Most of the imperial officials were unwilling to face the hardship and dangers of an invasion of Afghanistan. However, Akbar overruled them. Akbar ordered a fort to be built at Attock. Kasim Khan (an engineer who constructed the Agra Fort) was left in charge of this fort with the order to subdue the refractory tribes in the surrounding region and to construct a bridge. On 12 July, Akbar finally crossed the Indus. On 1 August, when Prince Murad’s advance guard was attacked by Hakim’s force, the Rajput Mansabdar Raja Man Singh arrived with the main force and saved the day. After a short halt at the junction of the Kabul River and the Indus, the Mughals marched to Peshawar, which had been evacuated and burnt by Hakim’s officials. Prince Salim (later Emperor Jahangir) entered the Khyber Pass in advance of his father Akbar and marched to Ali Masjid and Jalalabad. On 3 August, the Mughals entered Kabul and Hakim fled to the hills. Hakim never submitted to Akbar. And Akbar knew that he could not remain in Kabul with a big army indefinitely – sooner or later, he had to go back to north India. Akbar was sure that as soon as he would cross the Indus with the main body of the Mughal Army, Hakim would create problems in Kabul. Akbar tried to play off the Badakshanis against Hakim. Bakht-­un-Nissa, the half sister of Akbar was the wife of Khwaja Hasan, the ruler of Badakshan. Akbar offered the province of Kabul to this lady. Then, the Mughal Army left Kabul and Akbar reached Agra on 1 December 1581.75 Though Hasan did not occupy Kabul, the very threat of the Badakshanis prevented Hakim from creating a large-­scale threat to the Mughals. The Hindu Kush separated the Mughal Suba (province) of Kabul from Balkh and Badakshan. Badakshan is on the bank of the River Oxus. To the east of Badakshan lies the Pamirs. In earlier times, the capital of the region was a city also named Badakshan. But, in the late medieval era, Badakshan’s capital was Faizabad. In general, the Mughals did not give overt importance to Badakshan because they did not consider the Oxus a good defensive barrier. Also, hilly Badakshan’s revenue was considered inadequate even for maintaining a high-­ranking mansabdar.76 In 1585, Abdullah Khan Uzbek conquered Badakshan and in the next year occupied Balkh. In response, Akbar strengthened Mughal rule in Kabul. By this time, Hakim had died and Akbar established direct administration over Kabul, which was organized into a suba.

Great Mughals, Warfare and COIN, 1520–1707

57

In response to the rising Uzbek threat, Akbar decided to strengthen Mughal control over the Indus tribes. Zain Khan was ordered to annex Swat and Bajaur. About 30,000 Yusufzai families inhabited Bajaur, and another 40,000 Yusufzai families inhabited Swat. On the banks of the Indus, Zain Khan attacked the Yusufzais and the Mughal altamash (advance guard) dispersed the tribesmen. To ensure permanent control over this region, Zain Khan constructed a fort named Jagdara in midst of the Yusufzai territory. When the Yusufzais realized that the Mughals were going to stay permanently in their region, they attacked the Mughals repeatedly. Yusufzai resistance was intense in the Hills of Karagar and in the Buner region. Owing to continuous fighting and the need to patrol the mountainous country, Zain Khan requested reinforcements. Akbar sent Raja Birbal, a Hindu Mansabdar with a detachment and another force with Hakim Abul Fath. However, Birbal, Abul Fath and Zain Khan failed to cooperate with each other. Personal disagreements among these three Mughal commanders further hampered the Mughal COIN campaign. In 1586, the three Mughal commanders met at the Hills of Malak and to hammer out a coherent strategy. Despite Zain Khan’s attempt, cooperation was not forthcoming from the other two Mughal commanders. Then, the three commanders marched towards Jagdara Fort. Zain Khan argued that he would have the final say in the campaigns against the Yusufzais but Birbal and Fath disagreed. Zain Khan emphasized that while one strong Mughal detachment should occupy Jagdara, the other two detachments should launch mobile patrols for flushing out the insurgents. Zain Khan added that such a strategy would be time-­consuming. However, Birbal and Fath Khan disagreed. They asserted that the proper strategy should not be permanent control of the Yusufzai territory but destruction of the hostile force. Hence, the three detachments should advance and meet the Yusufzais in a decisive battle. And after destruction of the Yusufzais, Birbal and Fath Khan argued that with their contingents they should retire to Delhi. Zain Khan pointed out that to control the Yususfzais, permanent occupation of their homeland was necessary. Furthermore, moving into the defiles of the mountains without proper reconnaissance might be dangerous. In the end, Birbal-Fath Khan’s view prevailed. The whole army moved out from Jagdara with the right and left wings under the command of Birbal and Fath Khan respectively. On the first day of advance, the army covered five kos and halted at the village of Kandak. Next day, the army moved into the narrow defiles of the mountains and the right wing was left as rear guard. On the third day, the Yusufzais attacked. As the Mughal advance guard attacked the Yusufzias and drove them away, the tribesmen plundered the pack animals of the Mughal force. Birbal and

58

War and State-Building in Afghanistan

Fath Khan mistakenly believed that they had the Yusufzais on the run. So, the Mughal force moved another six kos to Khanpur. Zain Khan advised that the Mughal should not advance further into the mountainous terrain because of the lack of water supply. Furthermore, Zain Khan commented that due to the absence of level plains, the Mughals would be unable to deploy their superior cavalry force against the Yusufzais on the narrow mountainous paths. However, Birbal and Fath Khan disagreed and next day the Mughals penetrated further into the mountains. As the Mughals reached the mountains of Bulandari, the Yusufzais attacked from the hills above. Zain Khan commanded the rear guard. Fighting continued until nightfall. Next morning, the Yusufzais attacked again and as the Mughal advance guard tried to clear the defile, it was repulsed. The Afghans then attacked from all sides with arrows and stones. In the narrow defiles, the Mughal force lost all order. Taking advantage of the terrain, the Yusufzais launched nocturnal attacks and the Mughals panicked. In their attempt to retreat during the night, many Mughal soldiers and their animals fell off the precipice. The 8,000-strong Mughal detachment under Birbal was wiped out in the Swat defile. Zain Khan and Abul Fath were able to retreat with their contingents.77 Absence of unity of command and a failure to conduct a thorough reconnaissance of the terrain before advancing into the narrow mountainous country held by the hostile tribesmen spelled doom for the Mughals. In the narrow winding mountainous tracks, heavy Mughal cavalry was wiped out by the Afghans attacking from the hills above just as 400 years later, heavily laden Soviet armoured columns would be ambushed on the narrow mountainous roads by the mujahideen perched on the hills. Akbar was angry with Zain Khan and Abul Fath for running away from the battlefield. The emperor appointed another Rajput, Mansabdar Raja Todar Mal. He implemented Zain Khan’s strategy. Todar Mal constructed several forts and launched several detachments simultaneously to flush out the Yusufzais from their mountain strongholds. And Todal Mal was successful in crushing the Yusufzai insurgency.78 For Akbar, the principal threat remained the Uzbeks. And to ensure secure lines of communications between Lahore and Kabul, Akbar overlooked tribal depredations and patched a fragile peace with them.79 In 1598, Abdullah Khan died and civil war broke out in Transoxiana (Central Asia). This provided the opportunity to Shah Abbas, the ruler of Safavid Persia, to occupy Khorasan.80 Its principal cities were Herat and Meshed/Mashad. Gurmsir, an Afghan district, formed the boundary between Kandahar and Khorasan. Gurmsir is located on

Great Mughals, Warfare and COIN, 1520–1707

59

both banks of the River Helmand and inhabited by pastoral tribes.81 In March 1603, in an attempt to maintain good relations with Persia, Husain Beg was sent as the Mughal ambassador to Persia by the Mughal emperor.82 Under Emperor Akbar, the Mughals constructed a highway along the Khyber/ Khaibar Pass to maintain secure military connections with Afghanistan. Every spring, the stones were removed and the scrub and other vegetation cleared. The road was made suitable for the bullock carts, which carried the baggage of the Mughal Army.83 Akbar’s successors repeatedly tried to develop secure communications between India and Afghanistan. Joannes De Laet, the Flemish geographer, philologist and naturalist was born in Antwerp in 1593 and died at Leiden in 1649. In 1625, he was the Director of the Company of the West Indies. Later, he became one of the directors of the Dutch East India Company. In 1631, he published his observations on the Mughal Empire in Latin.84 De Laet describes the internal security situation of the Mughal North-­West Frontier under Emperor Jahangir: ‘The road from Lahore to Kabul is infested by Pathan brigands; and although the king has established 23 guard stations of troops at regular intervals, none the less travellers are frequently robbed by these brigands, who in the year 1611 actually attacked and looted the city of Kabul.’85 Usually, the journey from Lahore (the important Mughal military outpost in western Punjab) to Kabul took about three months, because of the vast detours travellers had to make to avoid the robber-­infested regions and a lack of good tracks across the mountains. However, the Mughal royal road from Lahore to Kabul with guard posts constructed at regular intervals reduced the journey time to 20–25 days. De Laet provides details about the Mughal road from Lahore to Kabul. The different stages along the road with corresponding distance are given in Table 2.1. The road from Lahore to Kabul was designed to strengthen Mughal control over central and northern Afghanistan. Another road was constructed from Lahore to Kandahar to secure Mughal domination over southern Afghanistan and especially to guard against any possible Persian incursion. The Mughals maintained about 15,000 cavalry in Kandahar. Kandahar Fort was protected on the west by a steep rugged mountain, and in the south and east by a strong wall. The details of the royal highway from Lahore to Kandahar as described by De Laet are given in Table 2.2. The construction of the royal roads assisted the transfer of military assets from the core of the empire to deal with frontier uprisings and to check the external threat posed by the Uzbeks and the Persians. The road building and maintenance of internal peace also bolstered long-­distance commerce. Large caravans from Kabul journeyed for

60

War and State-Building in Afghanistan

Table 2.1  Outposts along the Mughal Road from Lahore to Kabul Outposts

Distance Remarks

Lahore to Kacha Sarai Kacha Sarai to Aminabad town where the River Ravi was crossed Aminabad to Chima Gakhar city Chima Gakhar to Gujerat, a trading post

10 kos 8 kos 12 kos 14 kos

Not to be confused with the Mughal Suba Gujarat bordering on the Arabian Sea in west India

7 kos

Gujerat to the crossing of River Chenab Crossing point at Chenab to Khawaspur Khawaspur to Rohtas-­i-Khurd

15 kos

Rohtas-­i-Khurd to Hatea Hatea to Pakka Pakka to Rawalpindi Rawalpindi to Hasan Abdal Hasan Abdal to Attock

15 kos 4 kos 14 kos 4 kos 15 kos

Attock to Peshawar Peshawar to Ali Masjid

36 kos 10 kos

Ali Masjid to Dakka Dakka to Basawal Basawal to Bariku Bariku to Ali Boghan (on River Kabul) Ali Boghan to Jalalabad Jalalabad to Charbourg Khurd Charbourg Khurd to Nomle Nomle through Gandamak to Surkhab Sarai Surkhab Sarai to Jagdalak

12 kos 6 kos 6 kos 11 kos

12 kos

14 kos 4 kos 14 kos 4 kos 8 kos

A fort constructed on top of a mountain. Sher Shah Suri’s frontier fort was still in operation under the Mughals

A city on the Indus with a strong fort The path between Peshawar and Ali Masjid was extremely dangerous because here the tribal sirdars could effectively mobilize somewhere between 10,000 to 12,000 cavalry

Great Mughals, Warfare and COIN, 1520–1707 Jagdalak to Abi Barik Abi Barik to Dowabad Dowabad to Butkhak Butkhak to Bikrami Bikrami to Kabul

61

8 kos 8 kos 8 kos 3 kos 3 kos

Source: The Empire of the Great Mogul, De Laet’s Description of India and Fragment of Indian History, tr. by J.S. Hoyland and annotated by S.N. Banerjee (1928, reprint, New Delhi: Munshiram Manoharlal, 1974), pp. 55–6.

Table 2.2  Mughal posts in the road from Lahore to Kandahar Mughal Posts

Distance

Lahore to Chak Sunder, a small town Chak Sunder to Naushera/Nowshera Naushera to Mopalki Kamal Khan Mopalki Kamal Khan to Qamal Chan Qamal Chan to Harappa Harappa to Chak Ali Shah Chak Ali Shah to Tulamba Tulamba to Siddhu Sarai Siddhu Sarai to Khatti Churikabadai Khatti Churikabadai to Multan (a big city situated 3 kos from the banks of the Indus) Multan to the small village of Petto Ali Petto Ali to Katzai Duki Katzai Duki to Secota Secota to the defiles of Khoja Amran Mountains Khoja Amran Mountains to Peshingaon Fort Peshingaon Fort to Kandahar

11 kos 15 kos 8 kos 19 kos 16 kos 12 kos 12 kos 14 kos 15 kos 12 kos 20 kos 72 kos 14 kos 24 kos 23 kos 60 kos

Source: The Empire of the Great Mogul, De Laet’s Description of India and Fragment of Indian History, tr. by J.S. Hoyland and annotated by S.N. Banerjee (1928, reprint, New Delhi: Munshiram Manoharlal, 1974), pp. 69–70.

two to three months to reach Kasghar. The chief trading centre was Yarkhand, where large quantities of silk, musk and rhubarb from China were exported to Afghanistan.86 After Akbar’s death, his eldest son Prince Salim ascended the Mughal throne as Jahangir on 23 October 1605.87 By May 1606, the Persians posed a threat to Kandahar. Jahangir jots in his memoirs:

62

War and State-Building in Afghanistan

I had gathered from the reports of Shah Beg Khan, the Governor of Kandahar, that the amirs of the Qizilbash frontier were going to make a move owing to the corrupting influence of several men remaining from the army of the mirzas of that area who were always shaking the chain of contention and strife and writing letters encouraging [the Qizilbash] to take Kandahar.88

The death of Humayun and the rebellion of Jahangir’s son Khusrau encouraged the Persians to make a move against Kandahar.89 Jahangir tells us: The Governor of Farah, the malik of Seistan, and the jagirdars of that region attacked Kandahar with the assistance of Husayn Khan, the Governor of Herat. Thanks to his courage and bravery, Shah Beg Khan made a manly defence by fortifying and securing the fortress.90

Jahangir congratulated Shah Beg for not engaging in a pitched battle with the Persians in front of Kandahar. In his memoirs Jahangir wrote that Shah Beg’s decision to shut himself up in the Kandahar Fort and prepare the garrison for a siege was the right decision.91 At the time of the Persian Siege of Kandahar, the Mughal garrison was lucky because both the Mughal Emperor Jahangir and the Mughal court were located at Lahore. And from Lahore, a relief force was easily dispatched to relieve the siege of Kandahar. Jahangir himself was aware of the Mughal’s good luck. He notes in his memoirs: ‘By chance, the imperial forces, that had been sent after Khusrau from Agra were camped in Lahore, just then.’92 The emperor describes the preparation of a relief force: I immediately appointed a large contingent under the leadership of Mirza Ghazi, who was accompanied by a number of officers and servants of the court such as Qara Beg . . . Tokhta Beg . . . Khwaja Aqil was appointed as bakshi to this campaign. To cover expenses, forty-­three thousand rupees were given to Qara Khan, and fifteen thousand rupees were given to Naqdi Beg and Qilich Beg, who accompanied Mirza Ghazi. I decided to stay in Lahore in order to settle this matter, and to visit Kabul.93

When the Persian force learnt that a Mughal relief force was marching towards Kandahar, they lifted the siege and retreated across the River Helmand, some 56 kos from Kandahar. The Mughal relief army entered Kandahar on 31 January 1607. In this siege, the regional governor of Farah was involved; the Persian Royal Army did not take part. Jahangir came to believe that the governors of Farah and Herat undertook the Kandahar expedition without any direct

Great Mughals, Warfare and COIN, 1520–1707

63

order from Shah Abbas I (r. 1587–1629). In December 1607, the defence of the Mughal North-­West Frontier was put in the hands of Mirza Ghazi, the commandant of the relief force. He was given charge of the subas of Thatta, Multan and Kandahar.94 About 20 kos beyond Kabul was the town of Charikar and 20 kos further to the north, Ghorband. Under Jahangir, it was at the boundary of the Uzbek Empire whose capital was Samarkhand.95 In 1645, Shah Jahan (b. 1592, d. 1666, r. 1628–58) planned to conquer Badakshan and Balkh by launching operations from Kabul. Not only did the Mughals start to assemble military contingents from various parts of India to Kabul but the imperial court itself moved to Kabul. On 6 April 1645, Shah Jahan ordered Asalat Khan, the Mir Bakshi (Commander-­ in-Chief), to collect mansabdars, ahadis and musketeers in Kabul for the forthcoming campaign.96 On the utilization of the Afghan mercenaries by the Mughals, Inayat Khan writes: He [Shah Jahan] moreover instructed the said Khan [Asalat Khan] to recruit a band of gallant and sturdy youths from amongst the Oymaqs, Chaghtais, and other tribes dwelling in the neighbourhoods of Kabul on the Badakshan frontier. With amir-­ul umara’s concurrence, he was to recommend the worthiest for suitable mansabs and enlist the rest into the ranks of the ahadis.97

The amir-­ul umara and Asalat Khan were ordered to send a party to widen and level the road leading to Badakshan and also to construct bridges along the route of invasion. The emperor further informed them that reinforcements from Punjab would soon reach them. During August 1645, Nazar Muhammad and his son Abd al-Aziz Khan, who were in possession of Samarkhand, Bokhara and parts of Transoxiana, were fighting each other. The civil war in Central Asia provided an opportunity to the Mughals to intervene in the region north of Afghanistan.98 In fact, Nazar Muhammad sent an envoy to the Mughal durbar to plead Mughal support for his cause.99 The amir-­ul-umara dispatched Khalil Beg with 2,000 cavalry and 1,000 infantry to seize the Kahmard Fort. After capturing this fort, Khalil Beg returned with most of his force to Zuhak to collect supplies and munitions of war. A detachment from Zuhak, which was escorting the heavy baggage and military stores being transported to Fort Kahmard, was attacked by the Uzbeks at Bamiyan. Before Khalil Beg could move with reinforcements to aid the Zuhak detachment, the Uzbeks withdrew. Meanwhile, Abd al-Rahman and Tardi Ali with their Uzbek soldiers captured Kahmard Fort from the Mughal garrison.

64

War and State-Building in Afghanistan

Khalil Beg wrote to the amir-­ul-umara that no further reinforcements could be sent for the assault on the Kahmard Fort because the route between Zuhak and Kahmard was devoid of grass for the horses and provisions for the men due to frequent to and fro movement of the armies along it. Moreover, the tracks between the mountains were not only narrow but also very difficult to traverse. The local Afghan chieftains who had transferred their loyalty to the Mughals in response to the changing military balance advised the amir-­ul-umara that a full-­scale invasion of Badakshan during the winter was not possible owing to logistical problems. In response, the amir-­ul-umara sent a force of 10,000 cavalry to Khinjan in order to reconnoitre the terrain for conducting military operations after the winter. Asalat Khan led a force over the Hindu Kush in an expedition that lasted for 16 days. He bought back many prisoners and also horses, camels, cattle and sheep from the inhabitants in retaliation. Shah Jahan ordered the amir-­ul-umara to send masons, carpenters and sappers in order to improve the road by Thul. The emperor took up winter quarter at Peshawar and the Rajput contingent was stationed at Attock. While Rustam Khan guarded Rohtas, Qilij Khan was stationed with a detachment at Bhera.100 Raja Jagat Singh, a Mughal mansabdar, with 1,500 cavalry and 2,000 infantry recruited from Rajasthan was deployed in Kabul. After they were financially recompensed from the Mughal treasury at Kabul, Jagat Singh’s force crossed the Thul Pass and sent an advance guard under Bhao Singh to ravage Khost.101 Khost is a district on the northern slope of the Hindu Kush between south and south-­ east of Kunduz and lies near the hill tracts that the British described as Kafiristan. It is in the west of Badakshan.102 During November 1645, Jagat Singh advanced towards Sirab and Indarab. When the Uzbeks attacked, he built a series of stockades made of wood and successfully engaged in a defensive close quarter battle. The Hazara infantry allied with the Uzbek cavalry was driven back by the matchlock men in Jagat Singh’s force. Then, 4,000 cavalry (2,000 Rajputs under Rajrup and 2,000 raised from the Afghans in Kabul) with firearm-wielding infantry attacked the Uzbek cavalry, which though numbering 20,000, was driven back.103 The Uzbeks were masters of mobile battles, as their mounted archers adopting Parthian tactics were simply unbeatable from the time of Alexander onwards. But, in a static defensive battle, lacking handheld firearms, the Uzbek cavalry was unable to defeat a well-­armed adversary (especially infantry) taking advantage of the terrain and field fortifications. In 1646, a 60,000 strong force armed with artillery was assembled under Prince Murad. The imperial objective was either to restore Nazar Muhammad as a tributary ruler in Balkh or to annex his kingdom to the Mughal Empire. In July

Great Mughals, Warfare and COIN, 1520–1707

65

1646, Murad and his commander Mardan Khan occupied Balkh. The Mughals got hold of Nazar Muhammad’s treasury with 12 million rupees but Nazar fled the city. Within a month, Murad retreated from Balkh due to logistical difficulties and Shah Jahan ordered his Wazir Sadullah Khan to reorganize the army for another invasion. Prince Aurangzeb was appointed as commander of the second imperial invading force and Shah Jahan himself moved towards Kabul to provide logistical and moral support to the invading force. The Uzbeks failed to stand against the musketeers and field artillery of the Mughals but the light Uzbek horse proved adept at ambushing and harassing the lumbering Mughal columns. During the summer of 1647, the Mughals in Balkh opened negotiations with Abdul Aziz at Bokhara. The Mughal Army were unable to live off the land. The sparsely populated desolate countryside did not generate an adequate surplus for sustaining the Mughal troops. There were no banjaras (mobile traders who carried grain and livestock for feeding medieval Indian armies) in Central Asia to supply the Mughal soldiers and their horses with grain and fodder. In October 1647, with the winter approaching, Prince Aurangzeb offered Balkh and the surrounding districts to Nazar Muhammad in return for nominal submission of the latter to the Mughal Emperor. As the Mughals retreated to Kabul over the snow-­filled passes, the Uzbek cavalry took a heavy toll on them. Overall, the Mughal–Uzbek treaty resulted in the shifting of the Mughal frontier some 50 miles north of Kabul. However, the Mughals had failed to capture Bokhara and Samarkhand. The two-­year campaign cost the Mughal court some 40 million rupees (mid-seventeenth century calculation).104 To put this figure in context, the annual Mughal revenue in 1590 was about 110 million rupees.105 As soon as the Mughal power projection in Central Asia failed, the Safavid threat in Afghanistan reappeared. In 1648, Shah Abbas II planned to capture Kandahar. From the border of Khorasan, Shah Abbas recruited matchlock men and pioneers. Agents were sent to Farah, Seistan and Herat to collect grain to feed the army on its march. The Persians planned to attack in the winter when the Mughals would be unable to send reinforcements through Multan and Kabul to Kandahar. Nevertheless, the Mughal court appointed Prince Aurangzeb as commander of the relief force, which numbered 50,000 cavalry and 10,000 infantry (including matchlock men and soldiers equipped with bans). Aurangzeb was ordered to proceed along the shortest route, which was from Bangash-­i-Bala to Bangash-­i-Payin, Kabul, Ghazni and then to Kandahar. Meanwhile, Shah Abbas marched from Herat to Tus (Mashhad-­i-Mukaddas) and then to Farah. After halting at Farah for some days, he then advanced towards Kandahar. The main body of the Persian Army was under Shah Abbas. He detached

66

War and State-Building in Afghanistan

two contingents comprising the Persian advanced guard. One detachment consisting of 8,000 cavalry and some matchlock men under Mehrab Khan was ordered to besiege the fortress of Bust. Another detachment comprising some 6,000 Qizilbash cavalry under Saz Khan Balik was ordered to occupy Zamindawar.106 Daulat Khan, the Mughal commandant of Kandahar, took up position in the interior of the fort. Besides depending on the Mughal garrison, he tried to strengthen his position by levying the Afghans from Kandahar. He stationed a party of imperial and locally raised matchlock men at the top of the Kambul Hill. Kakar Khan defended the towers with some matchlock men. Mughal troops as well as local levies raised in Kandahar garrisoned the fortifications at Daulatabad and Mandavi. However, the towers on the top of the hill of Chihal Zinah (40 steps) constructed by Kalich Khan were not guarded properly. From these towers, guns and matchlocks could be fired into the interiors of the forts of Daulatabad and Mandavi. The Qizilbash captured these towers and posted some matchlock men who opened fire on the garrisons of the two above mentioned forts. Meanwhile, a defeatist attitude spread among the Mughal mansabdars, ahadis and the matchlock men. They argued that since no aid from India would reach them, if they continued to fight the Persians they would either be killed by the Qizilbash or become prisoners and be captives for life. Daulat Khan failed to provide resolute leadership. As a result, most of the Mughal soldiers deserted the entrenchments. Hence, the Persians quite easily entered the Sher Haji. The Siege of Kandahar, which lasted for 54 days, cost the Persians some 2,000 dead (including 600 Qizilbash) and about 400 men of the Mughal garrison died. The Persian ruler with most of his cavalry left Kandahar due to lack of forage and grain. However, before leaving for Khorasan, the Shah appointed Mehrab Khan with 10,000 Qizilbash cavalry and matchlock men to garrison Kandahar. Dost Ali Uzbek and another detachment were left to garrison Bust.107 Aurangzeb reached Multan with the relief force and then proceeded to Kohat, where he halted and sent agents to determine the depth of snow fall before advancing further. Khalil Beg was sent in advance to level the road and construct bridges. He send an intelligence report that the route through the hilly terrain of Kohistan was so deep with snow that it could not be used for a further month. Aurangzeb then decided to advance through Peshawar by way of the rugged Sendh-Basta Pass and finally reached Kabul. From Kabul, he marched to Kandahar and made his headquarter half a kos from the fort. In 1649, when the Mughal Siege of Kandahar Fort had been in place for three and half months, the Mughal besieging army began to suffer from lack of grain and fodder. Owing to

Great Mughals, Warfare and COIN, 1520–1707

67

a lack of a siege train of battering guns and skilled artillerymen, the Mughals failed to capture the fort. Aurangzeb could not bring the heavy guns with him owing to the difficult route through which he entered Kabul from Peshawar. When the winter approached, Aurangzeb decided to withdraw from Kandahar due to mounting logistical difficulties.108 In 1651, Shah Jahan sent Sadullah Khan and Aurangzeb together with 50,000 cavalry and 10,000 infantry (including gunners, matchlock men and rocket men) to capture the forts of Kandahar, Bust and Zamindawar. This time, the emperor decided to fit out a grand artillery train with the relief force. Sadullah Khan was given 8 heavy and 20 light guns. The latter fired projectiles that weighed between four and five pounds. In addition, 20 hathnals (field guns drawn by elephants) and 100 shutarnals (small field guns mounted on the back of camels – the equivalent of the Western armies’ regimental guns) were present. About 3,000 camels were employed for carrying lead, powder and iron shot. When the siege had dragged on for 68 days, one Mughal gun through a lucky shot was able to kill the Persian Commandant of Artillery, Muhammad Beg. The Qizilbash suddenly burst out of the Kandahar Fort and attacked the Mughal trenches. The Mughals, who prevailed, had with them seven breaching guns. They were able to destroy the parapets and bastions of the fort. Of these seven guns, the barrels of two cracked due to continuous firing. The gunners of the other five breaching guns were not skilled enough to fire the guns with effect. In his account of the Siege of Kandahar, the imperial chronicler Inayat Khan noted the technical inferiority of the Mughals regarding siege guns and unskilled gunners vis-à-vis the Safavids. Meanwhile, news reached Aurangzeb that the Uzbeks had reached the vicinity of Ghazni. Aurangzeb was afraid that his line of communications with Kabul might be severed. So, he decided to retreat and raised the siege.109 Shah Jahan could not accept that the Mughals had lost Kandahar to the Safavids. In 1652, Aurangzeb’s brother Prince Dara Shikoh was given charge of the subas of Kabul and Multan with the express order of recapturing Kandahar. A heavier siege train accompanied Dara. Two breaching guns were given to him, which fired iron shot. The gunners of these two guns were under Kasim Khan. Another big gun that accompanied Dara fired shot of one hundredweight. About 30,000 cannon balls, 5,000 mans of gunpowder, 2,500 mans of lead and 14,000 bans also accompanied Dara’s army. In 1653, Rustam Khan, the Mughal commander captured the Bust Fort. He dismantled the fort and joined the main army near Kandahar. However, Kandahar Fort still eluded the Mughals. The siege dragged on for five months. By this time, the Mughal Army had exhausted

68

War and State-Building in Afghanistan

all lead, powder and iron shot. Due to the presence of a huge force in the environs of Kandahar for such a long time, the supply of forage also grew scarce. As winter approached, Dara, like his brother Aurangzeb in 1651, decided to retreat.110 The Mughal adventure in Kandahar was over. Kandahar being a strategic province, the Mughals expended much blood and resources in trying to recover the region. The Khyber Pass was the gateway of India for any power marching into the subcontinent from either Persia or Central Asia. Security of the Khyber Pass required control over Jagdalak and Kabul. Furthermore, the Mughals imported Turki horses through Kabul. The possession of Turki horses, which were superior to all types of horse bred in India, gave the Mughal cavalry superiority in the subcontinent. And the cream of the Mughal Army comprised cavalry. During the late seventeenth century, the Mughals maintained some 100,000 Turki horses.111 Lastly, the Mughals also had a sentimental attachment to Kabul where Babur first established his kingdom after being expelled from his ancestral Ferghana. To secure the southern approaches to Kabul from the west, control over Kandahar was necessary. From an economic point of view, the Mughals had to spend more than the income they derived from Afghanistan in maintaining rule over the country. In fact, Kabul was one of the poorest subas of the Mughal Empire. And the military expenditure was highest for the Afghanistan Suba. A revenue statement of Shah Jahan’s reign shows that the provinces of Agra and Lahore yielded the greatest revenue. Each of these provinces yielded 822,500,000 dams annually.112 Kabul Suba was divided into 40 mahals. The revenue from Kabul amounted to 157,625,380 dams.113 Most of the revenue from Afghanistan was derived from poppy cultivation. Like the present scenario, 450 years ago the principal profitable crop for Afghan cultivators was poppies. However, unlike the present US-NATO dominated Hamid Karzai Government, the Mughals encouraged poppy cultivation for several reasons. Military campaigns were dependent on majun (opium). Both the Muslim (Turkish and Afghan) and Rajput troopers and the Hindu and Muslim mansabdars took opium during campaigning and especially before battle. In fact, sometimes campaigns ceased due to an inadequate supply of opium. The ladies of the royal harem (household) were also heavily addicted to opium. Furthermore, the Mughal aristocrats were addicted to a drink that was a mixture of water, opium, apricots and plums.114 Thus, poppies remained Afghanistan’s principal exportable cash crop from medieval times to the rise of the Taliban. In general, the Afghans hated Mughal overlordship in Afghanistan. Indigenous Pashtun patriotism was expressed by the Pathan warrior poet Khusal Khan (b. 1613), a member of the Khatak tribe.115 In the 1660s and the 1670s, the Afghan

Great Mughals, Warfare and COIN, 1520–1707

69

tribes proved to be a problem for Emperor Aurangzeb (r.  1659–1707). The situation would have been far more dangerous for the Mughals had the Afghan uprising occurred simultaneously with the Persian invasion of Kandahar. Tarbiyat Khan, the Mughal envoy in Persia reported negatively about Shah Abbas. The latter was reported to be preparing for war against the Mughals from Khorasan. Aurangzeb sent Prince Muhammad Muazzam and the Rajput Mansabdar Maharaja (big Hindu ruler) Jaswant Singh to Kabul, and he himself decided to march to Punjab from Deccan. On 22 August 1666, the Persian monarch died while traveling from Farrukhabad to Isfahan. The danger of war with Persia being over, Aurangzeb ordered Muazzam to halt at Lahore.116 The Afridis concentrated around the Khaibar Pass and threatened Mughal control in this region.117 In 1667, the Yusufzais who inhabited the region of Swat and the Bajaur Valley, and the plains north of Peshawar rebelled under Mulla Chalak. The latter appointed Bhagu as Wazir of his new fangled dominion. Bhagu with 5,000 clansmen crossed the Indus above Attock and invaded Pakhli. Pakhli is the plain that lies east of the Pakhli River in the Hazara District and through this region ran the principal road from Punjab to Kashmir. Bhagu captured the Shadman Fort. Encouraged by the Mughal reverses, Yusufzai bands began to plunder the western Peshawar and Attock districts.118 Kamil Khan the Faujdar of Attock was ordered to concentrate the contingents of all the Mughal jagirdars near Nilab and crush the rebellious tribesmen. Nilab was a town close to the River Nilab (Blue water) in the upper Indus. Amir Khan, the Subadar of Kabul sent Shamsher Khan with 5,000 men to cooperate with Kamil Khan. However, Kamil Khan did not wait for the reinforcements under Shamsher Khan but attacked the Yusufzais.119 On 2 May 1667, Shamsher Khan crossed the Nilab River and on towards Attock. The Yusufzais retreated to the hills. On the same day, Aurangzeb sent Muhammad Amir Khan the Mir Bakshi with 9,000 cavalry to crush the Yusufzais. However, before the imperial reinforcements arrived, Shamsher Khan was able to defeat the rebellious tribe and took 300 of their maliks as prisoners.120 Muhammad Amin Khan, the Subadar of Lahore was appointed as Subadar of Kabul. When the winter set in and there was heavy snowfall, all the roads in the Kabul region were blocked. To avoid the privation of facing the Kabul winter and also to maintain a line of communications with India, especially Punjab, Amin Khan left his naib (deputy) in charge of Kabul and left for the milder climate of Peshawar. While stationed in Peshawar, Amin Khan had to deal with the Afridi Uprising in 1672. He sent a strong force against the Shipa Sept near Shahbazgarhi and the Bajaur tribes. Shahbazgarhi was plundered and 6,000 cattle

70

War and State-Building in Afghanistan

were taken from the tribe as punishment. Looting the livestock of the rebellious tribesmen was a technique that the British would also follow in the nineteenth century. During the Rabi (spring) season, Amin Khan began preparing for a return to Kabul. Mirza Ali Beg, the Faujdar of Jalalabad was harsh towards the Pathans of his district. In retaliation, the Pathans blocked the road to Kabul to obstruct the journey of Amin Khan to Kabul. When the advance guard of Amin Khan reached Jamrud, about seven kos from Peshawar, news reached the Mughals that the Afghans had blocked the road further up. One lobby within the Mughal camp was for conciliation and negotiation with the Afghans. However, Amin Khan refused to meet with the rebellious Afghans and marched towards Jamrud. When the Mughals reached Ali Masjid, news filtered back to the Mughal camp that the Afghan attack was imminent. Amin Khan hastily prepared his artillery morchas (batteries) to meet the imminent attack. At the Tarta Pass (3,400 feet), the Afghans descended from the hillside in several waves and attacked Amin’s camp. The imperial camp was cut off totally by the Afghans who surrounded it. Amin Khan made a mistake in not placing guards crowning the heights on both sides of the road. The Afghans cut off the water supply and the Mughals suffered from a shortage of drinking water. Amin Khan sent feelers to the Afghan camp. The Afghan sirdars demanded not only restoration of their annual subsidies but also inams (special rewards). Humiliated by the demand for extra subsidy, Amin Khan decided to attack the Afghans again.121 On 21 April 1672, from the top of the hills the Afghans, equipped with bows, arrows and muskets attacked the Mughals from all sides. Some veteran Mughal soldiers tried to drive the Afghans away. However, caught in a disadvantaged lower position the Mughals were defeated. The Mughal artillery was unable to fire uphill. Thus, terrain negated the advantage of firepower. Utter confusion prevailed in the Mughal camp. Amin Khan’s son Abdullah Khan died. Though Amin Khan was able to escape to Peshawar, there were about 10,000 Mughal casualties. Furthermore, 20 million rupees in cash and kind was looted from the Mughal camp. To cap it all, the Afghans captured 20,000 men and women who were sold as slaves in Turan (Central Asia).122 As a mark of imperial resentment, Amin Khan was transferred from the Kabul Suba. Then, Aurangzeb gave the task of preparing an army of retaliation to Muazzam Khan (the Subadar of Lahore) and the Mughal Mansabdar Mahabat Khan who was transferred from Deccan to the North-­West Frontier. The Mughal Army under their joint command marched to Peshawar where they made camp because the winter snowfall had blocked all routes. In the following spring, the Mughal Army with an advance guard of 5,000 cavalry under Raja Mandhata moved into Jamrud. Aurangzeb decided to

Great Mughals, Warfare and COIN, 1520–1707

71

strengthen the army moving towards Kabul. Jaswant Singh, the Subadar of Gujarat was assigned the province of Kabul and Shujat Khan with a strong park of artillery was appointed the leader of his vanguard.123 Shujat Khan made the mistake of advancing too rapidly and became separated from the main body under Jaswant Singh. Shujat Khan crossed Gandak and while passing through the Kharapa Pass was ambushed by the Afghans on 14 February 1674. In a daring night attack, the Afghans charged from the hills around. The next morning, Shujat Khan’s force was attacked again.124 When Jaswant heard of the night attack he immediately dispatched the artillery under two Rajput sirdars named Bachraj and Raghunath. A contingent of 200 cavalry was ordered to guard the artillery park and another contingent of 300 cavalry was ordered to attack the Afghans. The Afghans, who had surrounded Shujat Khan’s camp, were attacked suddenly and simultaneously from all sides and were defeated.125 However, Shujat Khan died during this encounter.126 Hearing of the continuous Afghan attacks on the imperial contingents, Aurangzeb decided to intervene personally. On 7 April 1674, with another army and a strong park of artillery he began to march to Kabul. On reaching Lahore, Aurangzeb stopped for 20 days to recruit more soldiers. With 2,500 troopers and artillery Sarbuland Khan was ordered to move along the foothills. Meanwhile, Jaswant Singh came from Jamrud and paid his respects to Aurangzeb at Rawalpindi. In September 1674, Prince Akbar and Asad Khan were ordered to move quickly to Kabul through Kohat. With the main army Aurangzeb crossed the Indus at Attock. News of the entry of the Mughal Emperor into Kabul created a flurry among the rulers of Persia and Turan and they sent their ambassadors. Prince Muhammad Akbar was appointed as the Subadar of Kabul.127 On 14 June 1675, Mukarram Khan and Muhammad Yaqub were ordered to attack the Afghans at Khapash. The tribesmen’s houses were plundered and many males were taken prisoner. However, the imperial force was ambushed in rugged terrain. Mukarram Khan escaped to Izzat Khan, the Thanadar of Bajaur. On 14 June 1675, with 9,000 troops Sarbuland Khan was sent against the rebellious Afghans. The imperial troops plundered the Afghans and destroyed their homes. In an attempt to consolidate Mughal control over the recently subdued regions, Aghar Khan and Hazbar Khan were appointed thanadar of Jalalabad and Jagdalak respectively.128 The Mughals realized that the Afghan chieftains rule was based on the sufferance of their followers. Their power over the tribes was nominal. So, instead of propping up the tribal sirdars with imperial support, by the late seventeenth century Mughal policy was to deter the tribesmen from rebellion by ‘showing the

72

War and State-Building in Afghanistan

flag’. In the nineteenth century, the British went to great lengths to support the authority of the tribal sirdars. However, the Mughals did not close the door to political concessions. The Mughal durbar allowed the Afridis, Shinwaris, Khattaks and Yusufzais to levy tolls on the trade caravans that passed between Lahore and Kabul. Furthermore, 600,000 rupees a year were granted to the various tribes as political pension. However, several Afghan sirdars considered the financial package inadequate. Occasionally, leaders emerged among the tribes who claimed princely or religious authority and led rebellions against imperial authority. Then, the Mughal Army had to be used to crush the rebels. Mughal counterinsurgency techniques involved sending a military detachment to crush the uprising, and then to establish thanas with garrisons in each of them in order to consolidate Mughal control over the region and establish security in the erstwhile troubled zone. However, most of the Mughal outposts in Afghanistan were withdrawn during winter and the pacification process had to start anew in spring.129 The Mughals sent mobile columns to suppress the rebellious Ghurids, Gilzai and Yusufzai tribes. The rebellious tribesmen were expelled from their villages. The seats of the tribal sirdars were raided and thanas with strong garrisons were established there. Prince Shah Alam was appointed the Subadar of Kabul and then Aurangzeb left for Agra.130

Conclusion Despite being Muslim themselves, the Mughals, Safavids and Uzbeks faced tough opposition from the Afghans. This is not to argue that Islam has played no role in Afghan insurgency. Rather, the point we are trying to make is that Islam in general and jihad in particular are not crucial components of Afghan opposition to external invaders in their homelands. From the medieval era, the control for Afghanistan represented a sort of ‘Great Game’ for controlling the heart of Eurasia. The Mughals were more successful than the British in maintaining a permanent presence in Afghanistan. The two principal Mughal bases for controlling Afghanistan were Kabul and Kandahar. The distance between Kabul and Kandahar was about 325 miles.131 The Mughals were never able to capture Herat and control west and north-­west Afghanistan. This region was alternately under the influence of the Persians and the Uzbeks of Central Asia. British control over Afghanistan during the nineteenth century was more transitory. Moreover, unlike the British, the Mughals used Afghanistan as a springboard for power projection in Central Asia.

Great Mughals, Warfare and COIN, 1520–1707

73

As this chapter has shown, road building, subsidies to the tribes, construction of military outposts, forts, the establishments of thanas and military recruitment are the characteristic features of Mughal counterinsurgency, which evolved through one and half centuries. Despite the passage of time and changes in military hardware, certain Mughal COIN techniques remained operational even under the British Indian Empire, the imperial successors of the Mughals. The Mughals realized that border management of the frontier tribes cannot be separated from the related and greater Afghan ‘problem’. The Mughal technique of granting subsidies to placate the Afghan tribes continued in the British period. When the subsidy was cut, rebellions broke out, which in turn cost the imperial power a great deal more. So, reduction of tribal subsidies proved to be a false economy. Also, both the imperial powers through limited military recruitment tried to co-­opt the potentially rebellious manpower of Afghanistan. Nevertheless, during the nineteenth century, under the British, certain innovative COIN techniques also evolved. The following chapter will focus on that issue.

Notes 1 George MacMunn, Afghanistan from Darius to Amanullah (1929, reprint, Lahore: Sang-­e-Meel Publications, 2002), pp. 2–3. 2 Robert F. Baumann, Russian-Soviet Unconventional Wars in the Caucasus, Central Asia, and Afghanistan, Leavenworth Papers, No. 20 (Fort Leavenworth, KS: Combat Studies Institute Press, 1993), p. 134. 3 MacMunn, Afghanistan, p. 6. 4 Jos J.L. Gommans, The Rise of the Indo-Afghan Empire, c. 1710–1780 (Leiden: E.J. Brill, 1995), p. 11. 5 MacMunn, Afghanistan, p. 13. 6 Nirodbhusan Roy, Niamatullah’s History of the Afghans (n.d., reprint, Lahore: Sang-­e-Meel Publishers, 2002), p. 17. Mohan Lal considers the theories of an Israelite or Armenian origin of the Afghans as spurious. Lal, Life of the Amir Dost Mohammaed Khan of Kabul, 2 vols (1846, reprint, New Delhi: Asian Educational Services, 2004), vol. 1, pp. 1–7. 7 H.W. Bellew, ‘A New Afghan Question’, Journal of the United Service Institution of India, vol. 47 (1881), pp. 49–97. 8 Roy, Niamatullah’s History of the Afghans, pp. 8, 20–1. 9 Major G.F. MacMunn, The Armies of India (1911, reprint, New Delhi: Heritage Publishers, 1991), pp. 153–4. 10 Patrick Macrory, Kabul Catastrophe: The Story of the Diasastrous Retreat from Kabul, 1842 (Oxford: Oxford University Press, 1986), p. 19.

74

War and State-Building in Afghanistan

11 Gommans, The Rise of the Indo-Afghan Empire, pp. 16, 21. 12 Babur-Nama (Memoirs of Babur), tr. from the original Turki text of Zahiruddin Muhammad Babur by A.S. Beveridge, 2 vols (n.d., reprint, New Delhi: Saeed International, 1989), vol. 1, p. 1; Stanley Lane-Poole, The Emperor Babur (n.d., reprint, New Delhi: Sunita Publications, 1988), p. 28. 13 Babur-Nama, vol. 1, pp. 105, 134. 14 Lane-Poole, Babur, pp. 56–7. 15 Mansura Haidar, ‘Military Organization Under the Uzbeks’, in Haidar, Medieval Central Asia: Polity, Economy and Military Organization (Fourteenth to Sixteenth Centuries) (New Delhi: Manohar, 2004), p. 304. 16 Babur-Nama, vol. 1, pp. 185, 196. 17 Babur-Nama, vol. 1, p. 232. 18 Roy, Niamatullah’s History of the Afghans, p. xxxiv. 19 Stephen F. Dale, The Garden of the Eight Paradises: Babur and the Culture of Empire in Central Asia, Afghanistan and India (1483–1530) (Leiden: E.J. Brill, 2004), pp. 187–8. Pashai and Paraji were dialects of Persian or other Iranian languages. 20 Babur-Nama, vol. 1, pp. 199, 217. 21 Dale, The Garden of the Eight Paradises, pp. 187–8. 22 Lane-Poole, Babur, pp. 98–9, 101. 23 Dale, The Garden of the Eight Paradises, p. 189. 24 Babur-Nama, vol. 1, pp. 227, 253. 25 Abul Fazl, The Akbar Nama, tr. from the Persian by H. Beveridge, 3 vols (1902–39, reprint, New Delhi: Saeed International, 1989), vol. 1, p. 238. 26 Babur-Nama, vol. 1, p. 383. 27 Jayashree Vivekanandan, Interrogating International Relations: India’s Strategic Practice and the Return of History (London: Routledge, 2011), p. 161; Abul Fazl, The Akbar Nama, vol. 1, pp. 236–7. 28 Babur-Nama, vol. 1, pp. 327–8. 29 Dale, The Garden of the Eight Paradises, pp. 188–9. 30 Lane-Poole, Babur, p. 137. 31 Babur-Nama, vol. 1, p. 202; Macrory, Kabul Catastrophe, p. 20; Gommans, The Rise of the Indo-Afghan Empire, pp. 16–17. 32 Babur-Nama, vol. 2, p. 519. 33 Zain Khan, Tabaqat-­i-Baburi, tr. S. Hasan Askari and annotated by B.P. Ambastha (Delhi: Idarah-­i-Adabiyat-­i-Delli, 1982), p. 187; Henry Steinbach, The Country of the Sikhs (n.d., reprint, New Delhi: KLM Book House, 1977), p. 6. 34 Babur-Nama, vol. 2, p. 452; Abul Fazl, The Akbar Nama, vol. 1, p. 240. 35 Babur-Nama, vol. 2, p. 457; Roy, Niamatullah’s History of the Afghans, p. liv. 36 William Maley, The Afghanistan Wars (2002, reprint, London: Palgrave Macmillan, 2009), p. 10.

Great Mughals, Warfare and COIN, 1520–1707

75

37 Khan, Tabaqat-­i-Baburi, pp. 7, 21. 38 Roy, Niamatullah’s History of the Afghans, p. 208. 39 Roy, Niamatullah’s History of the Afghans, pp. 12–13, 23, 32. 40 Khan, Tabaqat-­i-Baburi, p. 25. 41 Roy, Niamatullah’s History of the Afghans, p. xlii. 42 The History of India as Told by its Own Historians: The Muslim Period, The Posthumous Papers of the Late H.M. Elliot, ed. and continued by John Dawson, 8 vols (1876–7, reprint, Delhi: Low Price Publications, 2001), vol. 5, Tarikh-­i-Salatin-­iAfghana of Ahmad Yadgar, pp. 25, 28. 43 Roy, Niamatullah’s History of the Afghans, pp. xxviii, xxx. 44 Babur-Nama, vol. 2, pp. 472–5. 45 Khan, Tabaqat-­i-Baburi, p. 22. 46 Babur-Nama, vol. 2, p. 480. 47 The Jahangirnama: Memoirs of Jahangir, Emperor of India, tr., ed. and annotated by Wheeler M. Thackston (New York: Oxford University Press in association with the Smithsonian Institution, Washington DC, 1999), p. 23. 48 Babur-Nama, vol. 2, pp. 480, 534. 49 Haidar, ‘Military Organization Under the Uzbeks’, in Haidar, Medieval Central Asia, pp. 304–5. 50 Haidar, ‘The Administrative Structure Under the Uzbeks’, in Haidar, Medieval Central Asia p. 122. 51 Abul Fazl, The Akbar Nama, vol. 1, pp. 287, 307, 326–7, 346, 367. 52 The Tezkereh Al Vakiyat or Private Memoirs of the Mughal Emperor Humayun, written in Persian Language by Jouher, tr. Major Charles Stewart (London: Oriental Translation Fund, 1832), pp. 20–38. 53 Tezkereh Al Vakiyat, pp. 54, 68, 71, 73, 77. 54 Tezkereh Al Vakiyat, pp. 78–9; Iqtidar Alam Khan, The Political Biography of a Mughal Noble: Munim Khan-­i-Khanan, 1497–1575 (New Delhi: Munshiram Manoharlal, 1991), p. 7. 55 Khan, The Political Biography of a Mughal Noble, pp. 7–8, 11. 56 Tezkereh Al Vakiyat, p. 85. 57 Dale, The Garden of the Eight Paradises, pp. 187–8. 58 Khan, The Political Biography of a Mughal Noble, p. 12. 59 Tezkereh Al Vakiyat, p. 87. 60 The History of India as Told by its Own Historians, vol. 5, Tazkirat-­ul-Wakiat of Jauhar, pp. 147–9. 61 Khan, The Political Biography of a Mughal Noble, p. 10. 62 Tezkereh Al Vakiyat, p. 108. 63 The History of India as Told by its Own Historians, vol. 5, Tarikh-­i-Jahan Lodhi, p. 108. 64 Tezkereh Al Vakiyat, pp. 108–16.

76

War and State-Building in Afghanistan

65 Andre Wink, Akbar (Oxford: Oneworld, 2009), p. 24. 66 Abul Fazl, The Akbar Nama, vol. 2, pp. 80–1. 67 Abul Fazl, The Akbar Nama, vol. 1, p. 357. 68 M. Athar Ali, ‘Jahangir and the Uzbeks’, in Ali, Mughal India: Studies in Polity, Ideas, Society, and Culture (New Delhi: Oxford University Press, 2006), p. 316. 69 The History of India as Told by its Own Historians, vol. 6, Appendix, Note A, p. 457. 70 Sunita Zaidi, ‘Akbar’s Annexation of Sind – An Interpretation’, in Irfan Habib (ed.), Akbar and his India (1997, reprint, New Delhi: Oxford University Press, 2002), pp. 25, 27. 71 Fatima Zehra Bilgrami, ‘The Mughal Annexation of Sind – A Diplomatic and Military History’, in Habib (ed.), Akbar and his India, pp. 33–6, 39–43. 72 Abul Fazl, The Akbar Nama, vol. 2, pp. 39–41, 43, 45, 56, 59–60, 81–4, 289. 73 Wink, Akbar, p. 22. 74 Haidar, ‘The Administrative Structure under the Uzbeks’, in Haidar, Medieval Central Asia, p. 123. 75 Vincent A. Smith, Akbar the Great Mogul: 1542–1605 (n.d., reprint, New Delhi: S. Chand, 1962), pp. 141–4. 76 M. Athar Ali, ‘The Objectives behind the Mughal Expedition into Balkh and Badakshan, 1646–7’, in Ali, Mughal India, pp. 327–9. 77 The History of India as Told by its Own Historians, vol. 6, Akbar Nama of Abul Fazl, pp. 80–4. 78 The History of India as Told by its Own Historians, vol. 6, Zubdatut Tawarikh of Shaikh Nurul Hak, pp. 191–2. 79 Jadunath Sarkar, A Short History of Aurangzeb (1930, reprint, New Delhi: Orient Longman, 1979), p. 116. 80 Ali, ‘Jahangir and the Uzbeks’, in Ali, Mughal India, p. 317. 81 Tezkereh Al Vakiat, pp. 53, 59. 82 History of India as Told by its Own Historians, vol. 6, Wakat-­i-Jahangiri, p. 302. 83 MacMunn, Afghanistan, p. 17. 84 The Empire of the Great Mogul, De Laet’s Description of India and Fragment of Indian History, tr. J.S. Hoyland and annotated by S.N. Banerjee (1928, reprint, New Delhi: Munshiram Manoharlal, 1974), p. iii. 85 The Empire of the Great Mogul, De Laet’s Description of India and Fragment of Indian History, p. 55. 86 The Empire of the Great Mogul, De Laet’s Description of India and Fragment of Indian History, pp. 56–7, 70. 87 The Jahangirnama, p. 21. 88 The Jahangirnama, p. 58. 89 The Jahangirnama, p. 58. 90 The Jahangirnama, p. 58.

Great Mughals, Warfare and COIN, 1520–1707

77

The Jahangirnama, p. 66. The Jahangirnama, p. 66. The Jahangirnama, p. 59. The Jahangirnama, pp. 66, 89. The Empire of the Great Mogul, De Laet’s Description of India and Fragment of Indian History, p. 56. 96 Inayat Khan, The Shah Jahan Nama, An Abridged History of the Mughal Emperor Shah Jahan, compiled by his Royal Librarian, ed. and completed by W.E. Begley and Z.A. Desai (New Delhi: Oxford University Press, 1990), p. 323. 97 Khan, The Shah Jahan Nama, p. 323. Comments in brackets are mine. 98 Khan, The Shah Jahan Nama, pp. 323, 327–8. 99 John F. Richards, The Mughal Empire, The New Cambridge History of India, I: 5 (1993, reprint, New Delhi: Foundation Books, 2002), p. 132. 100 Khan, The Shah Jahan Nama, pp. 328–30. 101 Khan, The Shah Jahan Nama, p. 331. 102 Abul Fazl, The Akbar Nama, vol. 1, p. 250. 103 Khan, The Shah Jahan Nama, p. 332. 104 Richards, The Mughal Empire, pp. 132–3. 105 Wink, Akbar, p. 78. 106 History of India as Told by its Own Historians, vol. 7, Shah Jahan Nama of Inayat Khan, pp. 87–9. 107 History of India as Told by its Own Historians, vol. 7, Shah Jahan Nama of Inayat Khan, pp. 89–91, 93–4. 108 History of India as Told by its Own Historians, vol. 7, Shah Jahan Nama of Inayat Khan, pp. 95–6. 109 History of India as Told by its Own Historians, vol. 7, Shah Jahan Nama of Inayat Khan, pp. 99–101. 110 History of India as Told by its Own Historians, vol. 7, Shah Jahan Nama of Inayat Khan, pp. 101–2. 111 Wink, Akbar, pp. 23–4. 112 History of India as Told by its Own Historians, vol. 7, Majalis-­us-Salatin of Muhammad Sharif Hanafi, p. 138. 113 History of India as Told by its Own Historians, vol. 7, Miral-­i-Alam, Miral-­i-Jahan Nama of Bakhtawar Khan, p. 164. 114 Wink, Akbar, pp. 57–9. 115 Stephen Tanner, Afghanistan: A Military History from Alexander the Great to the Fall of the Taliban (New York: Da Capo Press, 2002), p. 113. 116 Saqi Mustad Khan, Maasir-­i-Alamgiri, tr. and annotated by Jadunath Sarkar (1947, reprint, Calcutta: The Asiatic Society, 1990), pp. 37–8. 91 92 93 94 95

78

War and State-Building in Afghanistan

117 George MacMunn, Vignettes from Indian War (1901, reprint, New Delhi: Low Price Publications, 1993), p. 184. 118 Sarkar, A Short History of Aurangzeb, p. 116. 119 Khan, Maasir-­i-Alamgiri, pp. 40–1; The Empire of the Great Mogul: De Laet’s Description of India and Fragment of Indian History, p. 5. 120 Khan, Maasir-­i-Alamgiri, p. 41. 121 Ishwardas Nagar, Futuhat-­i Alamgiri, tr. and ed. Tasneem Ahmad (New Delhi: Idarah-­i- Adabiyat-­i-Delli, 1978), pp. 103–5, 124. 122 Nagar, Futuhat-­i Alamgiri, pp. 105–6; Khan, Maasir-­i-Alamgiri, p. 72. 123 Nagar, Futuhat-­i Alamgiri, pp. 106–8. 124 Nagar, Futuhat-­i Alamgiri, pp. 108–10; Khan, Maasir-­i-Alamgiri, p. 81. 125 Nagar, Futuhat-­i Alamgiri, p. 110. 126 Khan, Maasir-­i-Alamgiri, p. 81. 127 Nagar, Futuhat-­i Alamgiri, pp. 111–12; Khan, Maasir-­i-Alamgiri, pp. 82–4. 128 Khan, Maasir-­i-Alamgiri, pp. 89–90. 129 Sarkar, A Short History of Aurangzeb, pp. 115–16. 130 Nagar, Futuhat-­i Alamgiri, pp. 113–14. 131 MacMunn, Afghanistan, p. 115.

3

Counterinsurgency and Empire: The British Experience with Afghanistan and the North-west Frontier, 1838–1947 John Ferris

The formula which always guided me when giving advice about our dealings with the tribesmen was: Fear, fence, face, fanaticism. Fear, in that the tribesman is in no sense of the term a superman and is, like anyone else, anxious to preserve his life and subject to fear of losing it. Fence. With the exception of the immediate offenders, the neighbouring sections will as a rule sit on the fence in the critical stages waiting for the result of the opening engagement. Face. They are always on the lookout for some excuse to save their faces. Fanaticism. Once Jehad is declared they become mad dogs and reason goes to the wind. The problem is to provide them, and as rapidly as possible, with the means of saving their faces, and so induce them to come down on our side of the fence, and the means, surely, an early display of adequate force. General Sir Kenneth Wigram (1937)1 But Afghan wars are not like wars in Europe. The trouble only became serious when the war was over. Sir Evelyn Howell, Resident in Waziristan (1930)2

Afghanistan and counterinsurgency This chapter assesses the British experience with counterinsurgency (COIN) in Afghanistan, from the perspective of strategy over the long term. The question, ‘how did Britain handle counterinsurgency in Afghanistan?’, sounds simple.

80

War and State-Building in Afghanistan

Actually, it is problematic. Britain rarely conducted anything like COIN in the modern territories of Afghanistan. It did so constantly in nearby locales, including some once regarded as part of Afghanistan: the modern Pakistani territories of Baluchistan, Federally Administered Tribal Areas (FATA) and Khyber-Pakhtunkwa Province. British officials described these locales through loose terms like the ‘North-­West Frontier’, or by the names of specific areas and peoples, such as Waziristan or Afridis. This chapter will follow British usage, with a critical edge. To authorities in Britain and Kabul, and to local peoples, these regions were linked. To divide them, even for purely conceptual purposes, presents difficulties. The ‘borders’ – a problematic concept in itself – between Afghanistan, the North-­West Frontier and the Punjab differed significantly in 1790, 1842, 1892 and today. So did politics within them. Neither British governors nor Afghan amirs (nobles) really controlled the populations of the North-­West Frontier, which often disregarded those rulers and their borders. Western concepts of legitimacy and sovereignty clashed with those held by Pashtuns and Muslims. Many of these characteristics apply today to Afghanistan and the FATA. Counterinsurgency, moreover, whether gauged by modern definitions or those used by Britain, is a complex matter. Between 1838 and 1947, British officers did not use that term. They did sometimes refer to ‘guerrillas’, whom they purported to despise (Garnett Wolseley, the best of Victorian generals, described them as ‘brigands and assassins’) or to ‘guerrilla’ wars.3 Britons did not describe Pashtuns as guerrillas, however, but as irregulars, rightly enough. British officers expressed the sense of COIN through terms like ‘small wars’ and ‘imperial policing’, one of the few areas where they had anything like a modern concept of doctrine, complete with manuals which distilled experience and guided action.4 They also linked these matters to issues involving police, politics and imperial policy. For Britain, COIN was a tool of empire, one of many means to create a rule that it regarded as good, but which, if pursued today, we would not. British ideas on these issues varied widely by time, and institution. Their understanding and practice of COIN did not always fit our views. Even when it does, they described matters in different language than we use. These problems are redoubled because of the peculiarities in modern ideas of COIN. The latter emerged as European imperialism eroded, during an age of liberal decolonization, in reaction to anti-­imperialist agitation and Mao Tse-­tung’s theory of guerrilla warfare. These ideas cannot merely be projected backward in time, before these empires existed, when insurgents and counterinsurgents took different forms than they do today. Not all insurgents are

Counterinsurgency and Empire, 1838–1947

81

guerrillas, forces that refuse to engage conventional forces in battle, but instead endure the occupation of their villages while harassing the occupiers. They may be irregular forces that deliberately avoid a guerrilla strategy, because it exposes their people to attack, but instead stall invaders on their frontiers, through battles where they use conventional weapons unconventionally, creating killing zones through ground and fire.5 To use modern jargon, during the nineteenth century, insurgents rarely could move from Mao’s first to third stage of guerrilla warfare, but those starting from that upper level could mount ferocious resistance, centring on battle with Western forces rather than evading them. Guerrillas became common problems for Britain only in the twentieth century, when it occupied territory where political movements could emerge and invoke a military threat. Compared with previous generations, they had better strategy and means to raise military and political support, through conspiratorial or mass movements, so to enter Mao’s first stage, and then to advance. Again, wars of conquest, and COIN, are not identical. Counterinsurgents can only defeat peoples who already have been conquered. The term ‘COIN’ conflates responses (perhaps involving different forces and strategies) to two distinct threats, guerrillas and irregulars. Where the modern concept unifies actions involving police, pacification, struggle against guerrillas and small unit operations, Victorians would have divided these matters between two poles, pacification by police and political officers, and irregular operations by regular soldiers. Modern schools of thought on guerrilla warfare and COIN, whether Maoist or liberal, are also ideologically loaded, treating peoples not as people but rather as conceptual categories. While people are real, ‘the people’ is an idea. Abstractions do not fight wars. There is no such thing as ‘the people’, with just one set of aims or grievances. The theories of Maoists and liberal counterinsurgents work only when people behave as these ideas insist; often, they do not. Both theories assume that all guerrilla wars are won or lost through the application of the same universal principles and techniques. Both are wrong: guerrilla wars always must be seen in the particular. These schools imagine that guerrilla wars are easier and simpler, and sometimes nicer, than they really are. Whether a COIN campaign against a serious foe can ever be both ‘liberal’ and ‘effective’, however, is dubious. The ‘hearts and minds’ school of COIN purports to draw inspiration from British practices after 1945, which actually were more unpleasant than it understands.6 Nor do these categories exhaust all of the aims and means of COIN. The Soviet Union, for example, had a unique model of COIN. It aimed to turn struggles for national liberation into class wars, by combining ruthless attacks on

82

War and State-Building in Afghanistan

insurgents and their friends; with efforts to split societies and bribe support into existence, by redistributing land from richer to poorer peasants; and by using amnesties to control some guerrillas while isolating the intransigent. This approach beat serious guerrillas with strong support in Europe, but failed in Afghanistan between 1979 and 1986.7 COIN may aim not only to stabilize matters but to change them, perhaps fundamentally, even when authorities, as with British ones in Baluchistan, claim merely to be restoring an old order. Intention and effect have paradoxical relations in COIN – a failed campaign may have great, perhaps counterproductive, consequences. Effects flow from whatever any ruler tries to do, actually achieves or does without trying, and from the resistance it inspires, silent or violent. These paradoxical and reciprocal relationships were doubly complicated in Afghanistan and on the North-West Frontier, where politics involved combat, and at the highest level, often meant civil war. No outside power seeking to shape events in this fragmented polity could succeed through normal diplomacy, but only through actions rather like the techniques of COIN. To help allies or oppose enemies might involve bribery and foment civil war. Equally, in 1893, so as to ensure a defensible northern frontier given its inability to control Afghanistan, Britain annexed territory populated by Pashtuns who felt some loyalty to Kabul, and even more to kin across the Khyber. Britain had to conduct COIN on the North-­West Frontier because it could not do so in Afghanistan.

Power and perception Any assessment of British COIN in Afghanistan and on the North-­West Frontier must address all of its aims and means, in both areas. Those matters were linked to the greatest strategic competitions which Britain faced, power politics across Eurasia and rule in India. Only then, a gentlemanly third come with the details of who controlled what, where and how, in Afghanistan and on the North-­West Frontier. COIN was just one means to achieve those last aims. Britons believed that their Raj, an empire of opinion, rested on reputation more than strength. As the Foreign Secretary to the Indian Government wrote in 1887, ‘our position in India is essentially artificial without a solid material base’.8 A sense that they did not know what dangers lay beneath their rule caused fear and uncertainty. In 1852, one Viceroy, the chief British official in India, the Marquis of Dalhousie wrote, ‘in India one is always sitting on a barrel of

Counterinsurgency and Empire, 1838–1947

83

gunpowder, and no one knows how close or where a spark may be concealed’.9 The Indian ‘Mutiny’ sharpened this sense. Britons believed that their rule might be threatened from within, through a popular rising motivated by religion and guided by the indigenous elites. They thought they had no friends in India, but many sorts of potential enemies. Hindus, the majority of the population, were divided by caste and elites. Muslims, however, members of an Islamic Ocean stretching across continents, shared an underlying unity. Calls for jihad could shake them as one, whether like a jelly or a tsunami. The impact of Muslim fraternity was uncertain. Many decision-­makers doubted its power, but few before 1930 disregarded the danger. Even John Lawrence, Viceroy, iron man of the Indian ‘Mutiny’ and founder of the ‘Close Frontier’ School, accepted that ‘fanaticism, the pride of race, the feeling of strength, and the inclination to combine against us in the Mahommedan Races living in the mountainous countries between India and the Oxus . . . render them formidable in themselves, and the feeling thus engendered re-­acts on the Mahommedan population in India’.10 Fear of jihad shaped British thinking. Thus, in 1875 one senior official, Bartle Frere, drawing an analogy with the medieval head of the assassins, the Ismaili ‘Old Man of the Mountain’, and the broader body of Shi’ites, wrote that Indian Wahabis in their madrasa at Patna ‘are held by the learned and orthodox to be dangerous and fanatical heretics, but are dreaded and courted by all classes. They are the natural vent for the undying fanaticism of Islam, requiring at all times to be watched, and in troubled times becoming a political force of much importance.’11 Such concepts were linked to a doctrine taught to all Indian officials and soldiers in the nineteenth century. The idea of an internal–external threat assumed that Britain’s hold in India was weak, while a small detonator fired beyond the Hindu Kush Mountain Range could spark the explosive in the magazine, the population of India. Lawrence defined the optimistic interpretation of this issue. The population of India was divided, as was the Indian Army. Under ‘certain circumstances’ they might ‘unite’ against Britain, but ‘so long as we are strong and able to maintain our authority, we shall have plenty of supporters. It is only when we are weak that our friends will fall away.’12 Other views were darker. In 1891, one commander of the Indian Army, Lord Roberts, wrote that any Russian victory over Britain in Afghanistan would trigger an avalanche of enemies, Cossacks joined by ‘almost every Afghan capable of bearing arms’ and many tribesmen on the North-­West Frontier, while Indian soldiers and people of all ethnicities might rise against Britain.13 By this logic, external threat was a euphemism for an internal one. The Russian menace was an Indian one. What

84

War and State-Building in Afghanistan

usually is taken as fear of Russia was one of Islam. The unity of Islam, the combination of an amir able to declare jihad, Pashtun tribes under his influence, and Muslim brethren to the south, was the powder trail connecting detonator and magazine. Such ideas are easy to lampoon. In fact, they were not irrational, as shown by analysis of their times, and our present dilemma. The rising of 1857 convinced Britons that peril could rise from the blue. The Raj faced many internal enemies. A sense of fraternity bound the Muslims. During 1914–21, Pashtuns on the north-­west frontier and Indian Muslims responded to calls for jihad and hejira, voluntary emigration from lands ruled by infidels, to those of the faithful. Calls for jihad were loud in every challenge to British rule on the frontier until 1947. Russia and other countries (Germany and Italy), especially between 1914 and 1939, sought to turn Afghanistan on India, and to spark revolts against the Raj. Britons, however, usually controlled these fears, viewing internal threat as latent rather than large. These concerns damaged but did not wreck policy, although after 1919 the focus on jihad, Muslim solidarity, an internal–external threat and the north-­west frontier, led decision-­makers to misconstrue the threat to the Raj of colonial nationalism.14 As the Permanent Under Secretary to the India Office, Arthur Hirtzel, wrote in 1919: Pan Islamism is undoubtedly a danger – or rather a potential danger . . . The antidote is nationalism . . . we ought to . . . make the Moslem of India find his spiritual home in India, and regard himself not as an Indian Moslem but as a Moslem Indian. Similarly in Arabia – the nationalist movement if properly guided and controlled will tend against Pan Islamism by keeping the Arab’s thoughts at home.15

From this perspective, the greatest threat to India was internal, then internal– external, and only then, purely external. Afghanistan and the North-­West Frontier were just parts of those problems, generally small ones, though potentially crucial. Britain could suffer losses in these areas without caring, so long as they did not shake its position in India, which essentially meant prestige, whether by reinforcing the lever of jihad or by destabilizing confidence in the subcontinent. Equally, local victory might not solve bigger problems. A chronic but cheap stalemate could be a success, better than a complete but costly effort. Victory in Baluchistan was useful and preferable to stalemate in Waziristan; but British failure, or success, to the same degree in both areas, mattered little to major games. Had such victories been acquired at great financial and military costs, Britain could have lost big competitions precisely by winning small ones.

Counterinsurgency and Empire, 1838–1947

85

Too much expense abroad might destabilize India. Control over Afghanistan, with its call on garrisons, might so have overstretched British power as to break it during 1857. Instead, Britain bought Afghanistan for pennies on the pound. To Britain, failures happened on the north-­west frontier and Afghanistan, as in 1841, 1881, 1898 and 1919, but no disasters. This orientation also precluded costly options. Britain had strategic patience for time, but not money. So long as the expense, and the relationship between cost and benefits, remained favourable, it could maintain stalemates for generations, but expensive and unnecessary wars always eroded Britain’s will. Britain conquered so much because it cost so little. Success in COIN requires strength, conquest even more so. Britain lacked such strength on the North-­West Frontier, which shows its estimate of the problem. Local strength stemmed from global strategy. British and British-­led armies had to handle several competitions, roughly equal in moment, where irregular operations occurred regularly, and enemies combined conventional and unconventional forces, or used weapons in unorthodox ways. Britain needed forces able to manage all these competitions, be recalibrated from one competition to another, and then to combine their strengths so to defeat any competitor.16 All land forces of the Crown evolved to fill a strategic niche across continents, combining politics, power and technology, in which they had superiority over most competitors. These elements included general-­purpose forces (the British and Indian armies) and scores of specialized (regional, paramilitary or gendarmerie) units for particular problems, like the Khyber Rifles. Redeployable forces were tiny in number, rarely reaching 10,000 solders for any campaign, though able to rise higher in case of crisis, their power multiplied by quality and technology. For good and ill, this system limited Britain’s ability to solve problems by power, and drove it to search for political solutions, which matched force as a weapon. Britain rarely was too strong for its own good, that classic problem of the United States. Britain’s sensitivity to local politics was high, even when its understanding was not. British decision-­makers understood and co-­opted individuals but were merely good in handling movements that linked elites and masses, whether open or conspiratorial. This approach led Britain down a path that generally was fruitful. By working with local elites and interests, Britain created tolerance of and support for its presence, allies and instruments. This approach, however, automatically created problems whenever Britain got the politics wrong. No place better illustrates these conditions than Afghanistan and the North-­West Frontier.

86

War and State-Building in Afghanistan

Invading Afghanistan Britain first invaded Afghanistan through habit and hubris. Since 1757, expansionist forces within the administration in India had driven its empire. Danger always stood beyond its borders, or could be made to look that way. War suited frontier officials rating promotion above peace.17 They had a record of success. Indian authorities invaded Afghanistan in 1838 because they thought it easily could be made a protectorate, and used to dominate Central Asia. They were misinformed – this invasion stemmed from trivial politics, confused policy and failure in net assessment, comparing the will and power of competitors against one’s own. Indian authorities knew little about Afghanistan, which they expected to control as easily as other states they had conquered in Asia – not surprisingly, given the chaos that engulfed that country. They were overly impressed by the promises of the ally they hoped to place on the throne of Kabul, the exiled ex-Shah Shuja. During the occupation, his servants fed false intelligence to British authorities, leading them to destroy his enemies, and their credibility.18 Even more, Indian authorities monitored internal plots against their rule, who they thought might be inspired by an external enemy.19 The Viceroy, Lord Auckland, believed: The agents of Russia were openly striking the credit and power of this country, in political schemes, fraught with danger, not to our interests only, but to the safety, of the British Indian Empire . . . The most vague and wild alarms prevailed throughout India. Every element of malignity and disaffection within the vast limits of our supremacy, was called into eager action.

Invading Afghanistan, he claimed, had killed these dangers.20 Auckland talked defence, but took the offence. He and his servants acted more from desire than fear, believing Afghanistan the key to Asia, and easy to turn. Britain’s first occupation of Afghanistan failed, but not in a simple way. Twenty-­two thousand British and Indian soldiers entered a country divided between hundreds of independent units. Many supported the British advance, few opposed it. The strongest leader, the Amir of Kabul, Dost Mohammed, fled. British forces garrisoned the cities, while political officers maintained influence in rural areas through irregular campaigns, bribes and paid levies. For two years, occupation faced little opposition. The Afghans were willing to let Britain intervene in dynastic politics, but not to conquer them. As it began to do so, opposition rose. Yet, British officials wanted Shuja to establish a real regime,

Counterinsurgency and Empire, 1838–1947

87

but a protected one. Only when they saw that he could not, or would not, do so, did they make their own, motivated by a search for what they regarded as good government. Thus, Shuja became a puppet, was seen as one, and for his own reasons, so described himself to leading Afghans. As in all areas of Asia they ruled directly, Britons raised taxes and attacked existing customs and interests, especially of chiefs and warriors. In 1841–2, these matters sparked a combination of jihad, coup, Pashtun war of national resistance, and banditry, not a national revolution, but dozens of loosely related risings. The British had no friends; once they ceased to be feared, and free-­handed, many enemies struck opportunistically. Alexander Burnes, the Chief Political and Intelligence Officer in Kabul, wrote, ‘More fighting still! When will this country be pacified?’21 Panicked Afghani leaders caused the revolt in Kabul by murdering Burnes. A private vendetta became political when their acts were not punished. Pashtuns saw threats to their society and religion, a chance to loot and means to mobilize. In the nineteenth century, prolonged guerrilla or irregular campaigns were hard to mount from peoples with loose organization beyond lineage or village. This required a fear by elites and people of threat to their way of life, and some widespread bodies, usually religious ones, to overcome social atomization and unify resistance. In Afghanistan during 1841–2, chiefs, Dost Mohammad’s son and Sufi orders provided inspiration, legitimacy and leadership.22 This combination of popular and elite resistance was unique in Asia at that time, but similar characteristics motivated contemporary opposition to European imperialism in Chechnya-Daghestan, Algeria, Punjab and Awadh. These cases, however, mattered so much that despite heavy losses, above anything Britain suffered in Afghanistan, European powers fought until they won. Right after abandoning Afghanistan, Britain crushed a greater power, the Sikh kingdom. What saved Afghanistan then, as later, was the fact that Afghans could make occupation expensive, while their country was worthless to foreigners. Each stage of the rising surprised the British, who were stunned into unusual steps: one political officer offered 10,000 rupees each, ‘or even 15,000’, for the heads of several Pashtun leaders.23 The order for assassinations was rare, and even more the high price on offer. British policy stranded forces, tripped them, and encouraged warriors to leap for their fallen throat. The British force at Kabul was smashed when its leaders were murdered. Their successors, naive and paralysed, accepted an offer of free passage, which became a pitiless massacre of 4,500 soldiers and more followers through snowdrifts down the Khyber Pass. Most other garrisons stood until relieved by another British force, which attempted to

88

War and State-Building in Afghanistan

recoup the disaster by demonstrating its ability to devastate Pashtun territory, and left. After 1842, Britain abandoned attempts to establish a forward defensive system for India in Afghanistan. A new view arose on Indian security, a defensive one. The Close Frontier School held that India was best defended on the North-­West Frontier itself, without commitments beyond. It did not view Russian expansion in Asia as a threat. Exponents of these views pursued little contact with states or peoples beyond India’s borders. Lawrence, for example, held that Britain could exert little influence on Afghanistan because of the ‘fanaticism, the pride of race, the feeling of strength and the inclination to combine against us’ of its people, the divisions among its princes, and ‘the conflicting passions and interests which convulse the body politic’.24 Meanwhile, Dost Mohammed returned to Kabul where, for 20 years, he created a new state, borrowing heavily from British ideas and aid.25 Exploiting British subsidies of seven lakhs of rupees (one lakh equals Rs 100,000 or £5,000) and 8,000 flintlock muskets, he balanced the old military system, of feudal cavalry and tribal levies, or lashkars, by a new army, armed and trained on Western lines. He held no grudges against Britain that money could not buy. That subsidy, and recognition of British power recently deployed in Punjab and Persia, kept him quiet during 1857. Britons encouraged these events through means they disliked, subsidies. The latter left an amir dependent on a continuation of largesse, but also stronger, with them looking suspiciously like paying blackmail. In 1867, during the civil war following Mohammed’s death, the Indian Foreign Secretary was not in any hurry to subsidize the Cabul ruler with either money or arms. A subsidy of money vamps up their pride and pretension, is never applied to the purpose for which it is given, nor, indeed, to any useful or healthy purpose, but stimulates their insatiable avarice, and is regarded as a tribute extracted from our weakness and apprehensions. Aid in arms is less objectionable, but even in this form of affording help to the Cabul ruler, it is hardly expedient to be forward in pressing such assistance on his acceptance . . . A supply of arms is really the most valuable assistance that can be given to an Afghan ruler, and would be generally understood as a very decisive mode of ranging British aid on his side.26

After that civil war, Dost Mohammad’s son and successor, Sher Ali, continued his work. He bolstered his army, aided by annual British subsidies of £60,000 (Rs 1.2

Counterinsurgency and Empire, 1838–1947

89

million) and 17,000 rifles. Under these amirs, Afghanistan became more prosperous than it had been for decades. Around 1874, British views on these matters changed, as a new government took power in London, while Russia leapt cross central Asia, bringing its borders alongside Afghanistan. It also threatened the Ottoman Empire and Persia, strategic buffers for the British Empire. In response, a ‘Forward School’ emerged among Indian officials. They agreed that Russian expansion in central Asia threatened the Raj, which must establish a defence zone beyond the North-­West Frontier.27 Lord Salisbury, Secretary of State for India, held that Britain must make Afghanistan a protectorate, a shield against St. Petersburg rather than an agent of subversion. This aim would require overcoming the Afghan refusal to accept British diplomats. A few British officials stationed in Afghanistan would become the power behind the throne through ‘the pacific invasion of England . . . The principle of it is, whenever you bring the English in contact with inferior races, they will rule.’28 This policy rested on excellent intelligence about Russia, drawn from state papers stolen in St. Petersburg, which Salisbury used well, combined with crass stereotypes taken from imperial anthropology, and poor understanding of how Afghans would react to his pressure. Though Sher Ali communicated with Russian authorities, nothing suggests that he wanted war with Britain. He did not wish to be a puppet. His efforts to avoid that fate, and Russia’s pursuit of influence in Kabul, so to shape British policy in Turkey, including a military march on and the dispatch of a mission to Afghanistan, led Britain to invade the country in 1878. This decision was made by authorities in India, triggering a machine they had built to British orders, but against the will of many ministers in London, including Salisbury. Britain wanted to prevent Russia from influencing Afghanistan, but invasion was partly an accident. It might not have happened without ostentatious Tsarist efforts. Conquest proved easy. About 37,000 British and Indian soldiers marched, Sher Ali’s army broke and he fled, to die in exile. Control proved hard. British forces withdrew, leaving behind an allied Amir, Muhammad Ya’qub, Sher Ali’s son, with a British advisor, Louis Cavagnari. When a rising in Kabul killed Cavagnari, British forces returned and Ya’qub abdicated, saying he would rather be a peasant than a puppet. Chaos emerged. British security rested on subsidies to tribes. One staff officer, an advocate of annexing Afghanistan, wrote, ‘all this black-­mail paying is very wrong in principle, but our force is so small that it requires all such help’.29 The enemy, although politically divided, pursued an effective strategy, combining lashkars and a decent regular army, with firearms to match Britain. Some 30,000 gazis and soldiers, united under the banner of jihad

90

War and State-Building in Afghanistan

and Ya’qub’s son, besieged Kabul, catching British forces by surprise again. In December 1879, a third British offensive up the Khyber Pass in 13 months dispersed the besiegers, but several pretender cousins built local power bases and grabbed for the throne. Efforts at regime change in Afghanistan caused one in Britain. A new British government altered its strategy. Britain could neither control Afghanistan nor let it splinter. It needed to find an Afghan prince able to keep his country united and quiet, strong enough to rule Afghanistan, smart enough not to bother Britain. Fortuitously, such a marvel was at hand, Abd-­al Rahman, an able man with some power, in need of more. He offered Britain political support in return for military aid. For Britain, this step was a gamble: it distrusted any Afghan prince, and years of exile in Tsarist territory raised concerns that Abd al-Rahman was a Russian ally. Precisely for that reason, Russia aided his return with a gift of 200 muskets and Rs 300,000. Had Abd al-Rahman wished to do so, only another British invasion could have stopped him. The British could not be sure of the truth, that experience in exile had taught him cunning, and suspicion of St. Petersburg. The issue ceased to be national resistance to Britain, becoming instead a struggle between princes. By stating its intention to leave and to support a widely acceptable candidate, Britain negated opposition to its presence, while retaining some ability to shape Afghan politics. So long as Britain would go, even central figures in the jihad were willing to accept its candidate as amir, so long as they did too.30 Abd al-Rahman’s cousin, Muhammad Ayyub, with 20,000 men was not: he started a rebellion and beat a British brigade at Maiwand on 27 July 1880. This revolt forced an alliance between political and military power. Abd alRahman used his influence to help Britain destroy Ayyub, neutralizing his rival’s support and helping his friends acquire supplies on the march. British forces, moving fast and hard, smashed Ayyub’s forces, and helped their ally overawe other rivals. Then they left the country, handing power to him, and proof of their utility: five lakhs and 1,000 rifles, far more than the aid Russian authorities had given him, and nearly half its annual subsidy to Sher Ali.31

Influencing Afghanistan Over the next generation, Britain created a new policy towards Afghanistan, based on a version of the ‘Close Frontier’. It lasted as long as the Raj. Britain abandoned offensive attitudes towards Afghanistan. The latter became a strategic buffer, which Britain could defend against Russia, or not, as it chose. Through

Counterinsurgency and Empire, 1838–1947

91

the declaration of the Durand Line and the annexation of Pashtun territories, Britain created a ‘scientific frontier’, a tactical buffer with a strong defensive position and great garrisons south of Afghanistan. That policy kept Britain and Russia from war, despite constant rivalry and occasional hostility, and enabled the management of Afghanistan. Meanwhile, Abd al-Rahman, his power primed by British aid, created an effective army and a state, as against a confederation of tribes, loose but more powerful than any seen in Afghanistan. It rested – relied – on unprecedented levels of British aid, Rs 12 lakhs per year, rising to Rs 18 lakhs per year in 1893, in return for not fighting the declaration of the Durand Line, and over 40,000 rifles. He accepted Britain’s demand that Afghanistan have foreign relations only with India, and not Russia. Abd al-Rahman was willing to cooperate with Britain, which he found a useful counter to Russian expansion.32 This system was imperfect, because intelligence was so. British authorities knew they could not be sure to spot a Russian attack before it began. They disliked the limits to their knowledge in Kabul, provided only by an Indian Muslim agent, ‘surrounded by spies’: ‘anyone suspected of coming near us is at once arrested’. Nor was his information always significant. In 1893, Abd al-Rahman’s chief wife became ‘very angry with the British Agent, who has been bribing her sweeper women to get news. She wants to know whether the great Government of Britannia thinks it a noble thing to bribe her women to tell “when she is naked”.’33 This ignorance left uncertainty, fear and loathing to haunt the North-­West Frontier, though the concern was secondary. This policy lasted under the reign of Abd al-Rahman’s son, Habibullah, even during great tests of the internal–external threat to India. During the First World War, Germany and Turkey sought to spread jihad and revolution throughout the British Empire. They sent 1,000 soldiers, and some diplomats, to Kabul, to induce Afghanistan to attack India. The British officials feared that the ‘popular attitude in Afghanistan . . . is pro-Turkish, fanatical and inflammable to a degree’, pressed by ‘an anti-British faction’ led by Habibullah’s younger brother, ‘largely supported by the priests and the fanatical element in the population’. Habibullah, however, had no desire to become a pawn, and then be sacrificed. Britain judged his aims and influence correctly.34 It carefully monitored all activities in Afghanistan and intercepted German communications from Kabul. So confident was their judgement, that when Habibullah requested a larger subsidy, British officials raised it only by two lakhs, from 18 to 20 – just £10,000, at a time when grants to tribes in the Khyber Pass rose six-­fold.35 In May 1919, four months after Habibullah was assassinated, his son and successor, Amir Amanallah, declared war and jihad against the Raj. That

92

War and State-Building in Afghanistan

declaration might have been more inconvenient in 1916. British security blocked Amanallah’s efforts to explode the magazine of Muslims within India, but he sparked a revolt that broke Britain’s position in Waziristan. Britain, concerned with Bolshevik efforts to subvert India and manipulate Muslims, faced the greatest embodiment of the internal–external threat ever seen.36 The Raj responded carefully, seeking a victory of politics rather than force. It merely defended its frontiers and restored its position in Waziristan, while forcing Amanallah to terms by stalling his armies and levies, bombing Kabul, bankrupting his regime, and agreeing that Afghanistan could have diplomatic relations with anyone it chose. When the Third Anglo-Afghan War was over, the Government of India wished, as its Foreign Secretary noted, ‘to let Afghanistan down light; to forgive and if possible to educate its truculent young Ruler, who has already been tainted by the catchwords of Bolshevism’.37 A desire to defeat Bolshevik subversion among Muslims drove British policy towards Afghanistan during 1920–1, which widened the breach Kabul and Moscow were building between themselves. British officials gave solutions of Soviet telegrams to Amanallah, which showed Russian cynicism towards him, and Britain’s knowledge of his intrigues. They gained from his shock – one witness thought Amanallah ‘evidently upset’, especially about a Soviet message stating ‘that treaty would make Afghanistan financially and militarily dependent on Bolsheviks’ – but this material merely aided a rearguard action, the abandonment of British hegemony over and subsidies to Kabul, and its replacement with education.38 An independent Afghanistan posed new problems. Britain’s response was effective, because its intelligence was excellent. With Afghan independence finally came the chance to establish a British legation in Kabul, able to acquire information through normal diplomacy. British cryptanalysis also mastered the codes of many countries, including Afghanistan and the USSR. The signs were complex. Amanallah did not wish bad relations with Britain, and often had poor ones with the USSR. Yet, the USSR increasingly was involved in Afghanistan, so to limit its aid to the basmachi, Muslim guerrillas in Central Asia, and to support Amanallah, whom Soviet diplomats and intelligence officers saw as an ally. Britain thought Amanallah erratic and unfriendly, and the USSR hostile. In two, revealingly contradictory, ways, Afghanistan became central to British strategy. British planners prepared to dispatch mechanized and air forces to aid Afghanistan against Soviet attack. Meanwhile, Amanallah’s effort to create an air force, aided by the USSR, raised fears of an ‘Afghan air menace’, an aerial assault on India as the razor edge for an internal–external attack. Multiplying concern with cynicism, Royal Air Force (RAF) Headquarters in

Counterinsurgency and Empire, 1838–1947

93

India argued that ‘the Russian-­on-­the-­frontier-­of-India bogey is due for revival, and that the threat is largely in the air, in view of the vast moral effect a small effort can create’, especially on ‘the mass of an illiterate, excitable and unstable Eastern population’.39 Despite these concerns, British policy towards Afghanistan remained calm, even during 1929, when Amanallah’s reforms sparked a civil war, which, intelligence showed, might cause Soviet intervention. As his regime collapsed, Soviet leaders advocated military intervention to save Amanallah. Otherwise, their agent in Kabul warned, the USSR must abandon all the progressive elements in AFGHANISTAN, which, increasing from day to day, are rallying round him, and [involves] in fact support to Afghan reaction. Leaving these decisions in force means canceling the whole of our ten years work in AFGHANISTAN . . . we will for long be compromised not only in AFGHANISTAN but also in the whole of the Middle East and in INDIA. In any case no one will believe in our neutrality and what no one will be able to swallow is that after ten years we have thrown overboard the progressives of Afghanistan and gone back on all our declarations.40

Ultimately, the USSR did not invade Afghanistan, though it did attack basmachi there, and supported a faction that lost a civil war to one backed very cautiously by Britain. Throughout these events, Britain pursued its aims through diplomacy rather than force. Intelligence illuminated Amanullah’s erratic path, Soviet policy and perceptions, and subversion of India. During the crisis of 1929, code breaking revealed that the Soviet Foreign Minister, Maxim Litvinov, told the Turkish ambassador that ‘unless AMANULLA’s situation improved (to such an extent that there was some hope of his final success) no help would be given, since any such help would only compromise RUSSIA without result’. One Afghan ally of the USSR told another, ‘the more these Russians promise, the less they fulfill. They do not realize what kind of Nemesis will overtake them at last.’ The Afghan Foreign Minister warned that unless the Soviets ceased intervention, the government ‘would be compelled to unite whole population of AFGHANISTAN with intention of declaring Holy war against U.S.S.R. and . . . would have to ask ENGLAND for assistance’.41 Had the USSR invaded, it might have created the counter-­weight of a coalition between a British Army and an anti-Soviet jihad. This intelligence helped British authorities to avoid alarmism, and guided a cautious policy of avoiding any actions that might ‘scare her [Russia] into dangerous counter-­action. We are convinced by far our safest course is “business

94

War and State-Building in Afghanistan

as usual” as far as possible within our own borders.’ Had such actions occurred in 1880, they would have maddened British authorities; those of 1929 were sufficiently informed, able and confident to watch the Soviet position collapse, and to help an acceptable alternative, Nadir Khan, become Amir, with little British intervention. As the Viceroy, Lord Halifax, noted, ‘I hope we may succeed in getting Nadir Khan in without giving rise to too intensive an anti-British agitation, and indeed it seems as if in this respect too things were working not unfavourably for us in that he appears to be generally regarded by all parties in Afghanistan as the deus ex machina.’42 A decade later, during the last internal– external challenge to the Raj, British code breaking and human intelligence, aided by the USSR, blocked a risible Italian effort to restore Amanullah (who, ageing and exiled, volunteered to return to his homeland by parachute) to the throne through a coup, and then to declare jihad and, in conjunction with an insurgent leader in Waziristan, the Faqir of Ipi, to foment revolution in India.43 British experience illustrates how one may win politically by leaving Afghanistan, after failing to do so through military means during an occupation. Withdrawal from Afghanistan actually boosted Britain’s leverage. When occupying Afghanistan, one plays one’s weaknesses against their strengths; when manipulating Afghanis from outside, their weaknesses become paramount. Little power is needed to achieve goals against them, though not against outside rivals attracted by its weakness and significance, who must be checked through other means. Britain attempted to conquer Afghanistan, or to practise something like COIN there, only during 1838–41 and 1878–81. Otherwise, it treated Afghanistan as an independent polity on imperial frontiers, using means like indifference, aid, annexation, bullying and bribery, while foreswearing conquest and rule behind the throne. After its first shock, and only true failure, in 1841, Britain learned to live with civil war and regime change in Afghanistan. During the tests of 1880, 1915–17 and 1930, it identified a strong ally, even men like Abd al-Rahman and Nadir Khan whom it once had distrusted, and helped him to defeat his rivals. Britain learned to manage the amirs, to ensure they walked the line between too much strength and too little, to judge their character and intentions, and to develop means of influence. Since British policy in Afghanistan was linked to rivalry with other powers, it needed means to monitor and manage them, to solve problems in Kabul abroad, by diplomatic pressure in St Petersburg, or by countering subversion launched by Germany, Turkey, Japan and the USSR. Such actions were hard with Russia between 1830 and 1914, largely because of problems with intelligence. They became easier after 1914, when British intelligence became excellent.

Counterinsurgency and Empire, 1838–1947

95

On the north-­west frontier: ideology, institution and insurgency Britain could leave Afghanistan and achieve its aims there, but not the North-­ West Frontier. Failure to conquer Afghanistan created a need to control that territory, which included positions required to block external threat. Britain had to do so from a weak position, unable to use the upper levels of its escalatory ladder, terror or empire. The attempt at control routinely is characterized by terms like ‘butcher and bolt’. While having some truth, such language is caricature. Along the North-­West Frontier, ideology, doctrine, experience and institutional interest, produced several approaches towards irregular warfare and pacification. These practices were distinct, though all shared aims – to keep a frontier quiet enough, to cause no problems elsewhere. Often, those aims had little importance: to be responsible for achieving them meant more than actually doing so. That process took different forms by time and space. Ultimately, Britain achieved a cheap and moderate success in Baluchistan, and a more costly stalemate among Pashtuns. Nineteenth-­century commentators distinguished two views of frontier management, the Bombay and Punjab schools. The ‘Bombay School’ stemmed from General John Jacob’s confrontation with hillmen raiding lowland people, and each other, in Sind. He found solutions from British experience in pacifying India, the Scottish highlands, and London. Jacob sought to create order, develop support and intelligence from the population, and immediately to pursue raiders, through a soldierly version of the policeman’s hue and cry. His policies required cooperation from the population, and found enough of it to succeed. Jacob committed his few military resources to the problem, transferring forces from garrisons to the field. He linked force and politics in the classic British tradition of pacification (and, later, of counterinsurgency), combined with an ardent sense of imperialism, western superiority and paternalistic idealism.44 It is moral more than physical force which is required to control predatory tribes: both are doubtless necessary, but the latter is so, chiefly, to enable us to apply the other. Justice, honesty, high principles, unswerving firmness, and force without violence, succeed best with these men, as with others. If we imitate their crimes, on pretence of retaliation, we only perpetuate the evil . . . A few words will sum up the whole system. At first, put down all violence with the strong hand; then, your force being known, felt and respected, endeavour to excite men’s better natures, till all men, seeing that your object

96

War and State-Building in Afghanistan

is good, and of the greatest general benefit to the community, join heart and hand to aid in putting down or preventing violence.45

In 1856, for purposes of external security, Jacob recommended that Britain use similar techniques, combined with a small garrison at Quetta, to dominate Baluchistan, territory between Sind, Persia and Afghanistan. This proposal languished during the days of Close Frontier. In 1875, however, the Government of India adopted Jacob’s plan, as part of its efforts to manage Afghanistan, and Asia. The Viceroy, Lord Lytton, held that agents at Quetta ‘would be the sentry of British interests, and might eventually become the head of an important Intelligence Department’.46 Indian authorities sent one of Jacob’s students, a man with prior experience in pacification, to execute the aim.47 Colonel Sandeman, a Scot, was the only man to achieve Salisbury’s ‘pacific invasion of England’. He aimed to keep Baluchistan quiet on the cheap, while indirect rule in Kabul managed Afghanistan and the North-­West Frontier. That ambition failed, while Sandeman achieved success in pacification. His system minimized the use of force, instead relying on politics and subsidies. Force, while necessary, was used primarily to make Baluchs take bribes. He was authorized to offer Baluch chiefs Rs 67,000 annually, and to raise the subsidy for their leading figure, the Khan of Khelat, which was doubled to one lakh. These figures were large, given the size of populations, compared with those offered to Sher Ali, or spent in the region between Peshawar and Kabul. During the operations of 1878–80, the latter sums reached Rs 170,000, and then for the decade after, approximated Rs 80,000, about 20 per cent of which went to leaders.48 In the Khyber Pass, this money barely bought a bargaining position. In Baluchistan, Sandeman used this power to build influence and maintain balance, playing chiefs against each other while strengthening their hold over their peoples, by enabling patronage networks that depended on a continuation of subsidies, and using that power to tame recalcitrant leaders, all the while minimizing British interference in internal affairs. His system kept Baluchistan cheap and cheerful until 1947. The Punjab School was less charismatic than corporative. It confronted Pashtun societies, divided, independent and often hostile. They were harder to shape than those in Baluchistan, while British means were weaker. It disciplined them through the same modes as Sandeman, politics, force and subsidies, but in different proportions, changing over time. Between 1848 and 1878, British authorities largely ignored Pashtuns. When it could not do so, Britain handled them through small subsidies for good behaviour and protection, embodied in levies that hired local men to police the

Counterinsurgency and Empire, 1838–1947

97

roads, rather than to rob them, backed by punitive expeditions. The latter often were driven by fear of jihad, which could be self-­reinforcing.49 Thus, the Ambala campaign of 1863 aimed to disperse the ‘Hindustani fanatics’, Indian Wahabis living upcountry from Peshawar, who preached jihad and found local support. After it was over, the Secretary of State for India, Charles Wood, observed, It really does seem to me that we have magnified a small matter, and by doing so have created a danger ten times as great as the one we have had to deal with at first. I am willing to believe that they have experienced our power, – that they have learnt the infinite superiority of our arms, – and that their loss, which is said to be great, will indispose them to try conclusions with us again. The danger which might have been great of a Mahomedan crusade is over, and I trust that we may have peace on our borders for some time to come. In order to achieve victory, the tribes, having been beaten, must be conciliated.50

As with Afghanistan, the trouble happened after the war was done. During 1877–8, British authorities sought to solve their problems in Waziristan through a Sandeman solution, guided by an able officer and backed by subsidies of Rs 54,000 to local leaders and tribes. This effort failed, partly through fluke, but also because Sandeman’s methods had less power over Pashtuns than Baluchs. If subsidies were used or refused to bolster one leader and weaken another, the latter retaliated by attacking the former and Britain, so to break his rival and boost himself. Interference had unintended and unfortunate consequences.51 In response, across the frontier over the next generation, during the classic days of ‘butcher and bolt’, the Punjab School emphasized negative against positive reinforcement. It spent less money and more time fighting, pursuing not pacification, but coercive diplomacy through militarized means. The latter involved taking hostages, imposing fines, economic blockade and punitive expeditions, throwing regular forces against irregular ones, with the aim of combining surprise in operations and firepower in tactics. Its means were more effective against the weak than the strong, the near than the far. Targets were hard to reach, and harder to teach. Attack provoked reciprocity, as each side adjusted to the other. Initially, foes charged formations with swords, but then, despite British efforts at blockade, adopted modern rifles and tactics.52 Operations became increasingly hard, because the population, including veterans of the Indian Army, played its hand well. During the Tirah Campaign in 1897–98, snipers fought 59,000 British and Indian soldiers to a standstill, ambushing units

98

War and State-Building in Afghanistan

and picking off officers. Mahsud and Waziri lashkars hammered British divisions during the rebellion of 1919–20. The Indian Army classed these warriors, ‘operating in their own country . . . among the finest fighters in the world’, many possessing ‘an intimate knowledge of our habits and tactics. The tribesmen have been described as the best umpires in the world, for they seldom allow a tactical error to go unpunished.’53 During the 1890s, Pashtuns found answers to the strategy of ‘butcher and bolt’. Britain responded by returning to reliance on politics, rather than punitive expeditions. It finally spent as much money among Pashtuns, as Baluchs, adding Rs 184,000 to annual subsidies, with another Rs 54,000 following between 1900 and 1910. Though it never established a Sandemen system among Pashtuns, it applied his methods among them as far as possible. They had fair success until the end of the Raj. Over the frontier, British officers created and commanded paramilitary forces, such as the Khyber Rifles, which paid Pashtuns to be police, rather than bandits. Political officers raised local forces, levies and khassadars, the latter selected by local maliks, clan leaders. So, to build influence and to achieve negative and positive aims, Britain subsidized large fractions of the local male population and their maliks, both directly and by enabling them to offer patronage to their following: ‘the model rested on the conciliation of the Pakthun elites rather than a culture of egalitarianism and consensus-­driven action. Maliks and khans who accepted allowances were further rewarded with privileges that accrued to them personally.’54 The scale of subsidies rose dramatically during the First World War. By 1918, in Waziristan alone, Britain paid tribal subsidies of Rs 125,865, and employed thousands of men (perhaps 20 per cent of the male population) part time as khassadars and levies, for another Rs 1.3 million.55 The 1919–20 rising signalled the limits to this policy, and the dangers of the frontier. After it was suppressed, the Indian Army pursued firmer control across the frontier, building elaborate road systems and deploying large garrisons behind barbed wire in advanced cantonments. A division, 15,000 regular soldiers, was deployed permanently in Waziristan. These activities supported a policy of ‘peaceful penetration and civilization of the tribes’. Financial officials opposed this occupation because of its cost, which many political officers thought futile.56 Later events might seem to support these views. This policy did not pacify the frontier. During 1937–8, three divisions, about 50 per cent of the Indian Army’s strength, and its entire field force, were committed against the Faqir of Ipi, in perhaps the most costly stalemate of the Raj’s history.57 Yet, in fiscal, political and military terms, British India weathered these commitments, its forces handled these conditions, and the tribes tolerated this expanded presence. That policy

Counterinsurgency and Empire, 1838–1947

99

was backed by greater tribal subsidies, rising to Rs 276,836 in 1923, and Rs 2.8 million for 4,600 khassadars in Waziristan alone, alongside enhanced efforts to work with and through local elites. Maliks received 7 per cent, or Rs 65,000, of the Rs 886,000 allocated in tribal allowances across the North-West Frontier Province during 1931. The true figure for allowances to chiefs probably is larger than we can know. Annually between 1926 and 1931, and probably long before and after, about 66 per cent of India’s secret service budget, some Rs 450,000, carried off the books, was allocated to the North-West Frontier Province for ‘protection and raids’, and especially to ‘Entertainment, Envoys and Chiefs’.58 Much of this money no doubt went to chiefs, though some may have gone to other purposes, including the acquisition of intelligence, influence in Afghanistan, and propaganda among mullahs. The costs of subsidies skyrocketed, exceeding the levels ever allocated to Afghanistan. During 1901–2, all forms of political expenses, excluding military costs, in North-West Frontier Province and Baluchistan, totaled 10.83 and 16.59 lakhs, respectively. By 1929–30, they reached 172.66 (including 50.13 connected to building roads, fundamental to British policy) and 82.88 lakhs.59 These expenses, about £1.3 million per year, were substantial, roughly equalling the maintenance costs of all RAF squadrons in India though only 5 per cent of the budget of the Indian Army, rising but still cheap for the price. The Indian authorities thought that allowances varied from the innocuous to blackmail, while ‘their utility as a steadying factor is well attested in all but times of the greatest excitement, when all material considerations are apt to be suddenly burnt up in the fire of tribal feeling or religious enthusiasm’. They hoped to escape the system of ‘paying the tribesmen not to make trouble’, but never succeeded.60 From 1920 to 1947, Britain managed Pashtuns as David Lloyd George did the press, through the maxim, ‘What you can’t square, you squash; what you can’t squash, you square’; with the same limited success. During the inter-­war years, British expenses on the North-­West Frontier reached their peak, absorbing perhaps 20 per cent of Indian revenues. Much of this expenditure, however, was discretionary, a means to justify the maintenance of forces at the level authorities thought necessary to deter Russians, Afghans or Indians; nor was it unbearable. Britain’s power on the North-­West Frontier broke because it lost India, not the other way around. Equally, the frontier never was pacified, merely managed, with new policies driven by lessons drawn from a previous one, which seemed just to have failed. This process was complex because Britain confronted a competition, a rival that adapted to defeat its actions, and a condition to be endured, rather than a problem to be solved. The game was

100

War and State-Building in Afghanistan

shaped no more by Britain than Pashtuns, who saw it as enemy, scenery, tool and bankroll. They gained more from this relationship than they lost, or Britain won. Only an academic could call Pashtuns essentially victims. Britain could live with this complexity and stalemate, because political officers always were able to negotiate a settlement. For all sides, strategy involved creating and maintaining an ability to talk. War was a form of politics by other means, shooting a mode of discourse. This outcome was scarcely satisfactory. Officers with universalist ideas of COIN claimed that Sandeman’s solutions would work better.61 Most officers of the Indian Army disagreed. Some, combining imperial anthropology and the concept of ‘martial races’, held that Baluchs simply were less formidable than Pashtuns. In 1879, Roberts denied that most regiments of the Bombay Army, the force handling Baluchistan, could ‘cope with Afghans’.62 Others used sociological arguments: Baluch society was more hierarchical, thus manageable, than fragmented, egalitarian and democratic Pashtuns. As one general wrote in 1932: It is comparatively easy to keep a chief well disposed by giving him moral support, as well as an allowance, and occasionally perhaps some arms. He can then take the edge off his tribesman’s warlike proclivities, while the mutual jealousies of the chiefs are likely to keep the whole in a state of some equilibrium – not too stable perhaps but a great deal better than among the irresponsible democracies . . . Sandeman’s policy was to support the Khans and make them control their clans. If the Khans disobeyed or abused their position, he deposed them. 63

All found Pashtuns tough customers. In 1924, the Chief Commissioner of the North-West Frontier Province noted, ‘Their inter-­tribal relations, their dealings with each other within the tribe and even within the family are regulated almost entirely by force. The average Pathan tribesman is cruel, treacherous, almost inhuman in his tormentation of a weaker rival. Such a race cannot be ruled by kindness and tact alone, though much good can be done by these qualities if the backing of force is there.’ In 1931, the Tribal Control and Defence Committee, consisting of leading political and military officials, noted the ‘virile and martial qualities and the predatory instincts of the tribes’. Pashtuns were made of unusually ‘intractable . . . human material’, and ‘prone to outbursts of fanaticism when stirred by the preaching of their mullas or by hopes of loot in times of excitement’.64 Such ideas, central to British views of controlling the North-­West Frontier, and to any judgement of them, sound easy to discount. They reek of Orientalism

Counterinsurgency and Empire, 1838–1947

101

and empire. Yet, British views of all peoples, Muslim or otherwise, combined competent observation with interpretation by stereotype and overgeneralization. The latter were influenced by many systems of ideas, complementary and contradictory: imperial anthropology, racism, Orientalism, Whig paternalism, concepts of national characteristics, cultural ethnocentrism, and a model of the evolutionary modernization of all peoples on British lines. These concepts are hard to handle as a whole, or individually. Racism had simple effects at lower levels of decision-­making, but complex results at higher ones. The evidence is vast, and filled with language or logic that was, or can be seen as being, racist or Orientalist, but they must be treated as problems, not as self-­evident. British assessments of any people, white or otherwise, was driven most by ethnocentrism, not racism, placing environmental above genetic factors, all embodied in ideas of national character. Britons had many views about any group of foreigners. Every group had its own gaze. Muslim peoples were not seen as one unchanging ‘other’. Imperial knowledge altered over time. Often it simply was knowledge. Officers of the Indian Army were perhaps the group of British decision-­makers most marked by racism, while political officers drank deep from imperial anthropology. Ideology shaped British views of peoples on the North-­West Frontier more than usual with them, but still their stereotypes had force. They were marked as much by romanticism as racism, and even more, by observation.65 Subsequent Pakistani, Russian and NATO experience suggests that no form of externally imposed police, or COIN, easily can master Pashtuns.

Air control and control of the air Nor did ideology alone shape approaches to COIN. Institutional interests had a greater impact. Thus, Sandeman’s system was dominated by political officers. Its soldiers, few in number with a secondary role, stemmed from the Bombay Army, marginalized in the military politics of the Raj. The victor in that struggle, the Bengal Army, dominated the united service after 1895. It had institutional reasons to keep its fighting edge keen for real soldiering, which frontier war did better than pacification, and to justify its existence. Punjab and the North-­West Frontier became the centre of its dispositions, and two of its main assignments, to control the border and defeat external assault.66 These forces did not practise COIN because they did not want to. Nor did they have to, so long as they kept responsibility for the task.

102

War and State-Building in Afghanistan

In the last years of the Raj, they faced competition. After 1919, British authorities expected mechanical devices to replace manpower in imperial policing with economy and effect. From this concept, the RAF generated a strategy for expanding at the expense of the older services, substitution, and a theory and practice of COIN, air policing. The RAF provided a key link in the process from 1911 in which western airpower struck Islamic insurgents. Its earliest uses by Britain, between 1916 and 1921, involved full-­scale attacks on villages, killing hundreds of civilians. Thereafter, Britain applied air policing less ruthlessly, because authorities believed their public would not tolerate indiscriminate attacks on civilians.67 Still, Britain used the tool effectively across the Middle East. Thus, during 1920, in its attack against the so-­called ‘Mad Mullah of Somaliland’, an airstrike on Mohammed bin Abdullah Hassan’s encampment wounded him, killed some of his lieutenants and scattered his flocks, a base for his power, which hostile tribesmen seized. He fled and died. The precision of this strike, while illuminating, proved impossible to replicate during the inter-­war years. India’s experience with air policing was different from that of the Middle East. In 1920, Indian authorities opposed the initial attempt to apply air control, in Iraq. As the Military Advisor in the India Office, H.V. Cox, noted, aircraft ‘can do nothing but act in one of two ways (a) pay a visit (b) drop a bomb. To rule and administer a great deal is necessary between these two acts!’68 Indian officials believed that air policing violated central principles for tribal control. Cox denied ‘that all that the usual invasion of tribal territory means is the destruction of villages and crops. In the past we have very seldom been able to convince the fighting tribes that they have got to come in and submit, until we have had a fair and square fight in the open and our men, British and Indian, have proved to them that they are no match for the disciplined soldier.’69 The RAF’s proposals raised the darkest fear in the Anglo-Indian soul, of watching brown men rape white women. In 1921, when pressed to police the North-­West Frontier by air, the Viceroy, Lord Reading, replied that the ‘principle of retaliation’ was so ‘deep-­ rooted’ in the tribes . . . that attacks on our women might easily follow sooner or later if we resorted to indiscriminate bombing of tribal villages. C-in-C holds with me that both psychologically and militarily it would be unsound to wage war on women and children of tribesmen who are already in a sense our subjects and who may some day become such in reality and who are fanatically susceptible as regards their women folk. We are therefore against bombing

Counterinsurgency and Empire, 1838–1947

103

of frontier villages. If we are forced to it in exceptional circumstances we should always endeavour to give women and children time to depart.70

Commenting on air control two years later, Hirtzel feared ‘the general political effect which a reputation for “methods of barbarism” may have on the frontier and in Afghanistan – to say nothing of the rest of the world’, and ‘the expense and inconvenience that would be caused if the life of an Englishwoman led to a punitive expedition or to a public demand that officers’ wives should not be allowed to live at frontier stations’. Moreover, ‘easy recourse to force tends to undermine the individual character by which the best political officers have controlled the frontier’. Ultimately, the Government of India concluded that these fears were overstated, and pledged to keep the RAF from damaging ‘sound old frontier methods’.71 It minimized attacks on women and children, through such means as insisting that no village be attacked by air without receiving warning of at least 24 hours, giving ample time for evacuation. These rules rarely were violated.72 Indian authorities rejected RAF proposals that gas be used in air control.73 Imperial Britain took more care to avoid civilian casualties on the North-­West Frontier after 1919, than NATO did in Afghanistan and FATA during 2012. For the RAF, the North-­West Frontier was the imperial locale where it could most gain from air policing, especially by using the same squadrons to replace soldiers there and other forces elsewhere. Thus, in 1921 the Chief of the Air Staff, Hugh Trenchard, claimed that seven RAF squadrons simultaneously could replace the Indian Army in tribal areas and as a field force against Afghanistan, and all divisions held in Britain to reinforce India during war.74 In 1925, he argued that ten RAF squadrons simultaneously could do the work of warships in the Indian Ocean, and police the North-­West Frontier, replacing the seven cavalry regiments, 34 batteries and 46 infantry battalions presently there – 50 per cent of the Indian Army, which would become vulnerable to reduction. In 1931, the RAF argued that if the present seven air squadrons were raised to 11, they could replace 40 per cent, or 17, of the infantry battalions on that frontier.75 Air control might have worked in Waziristan: after all, it did so, combined with versions of the Sandeman system, in Kurdistan and southern Iraq. Against this, air control always failed whenever it confronted any troubles in towns, or major insurgencies in the countryside, as in Palestine during 1929 and 1936–39, and Iraq during 1941. Then, only soldiers could save the situation, their numbers few, because of substitution engendered by the RAF. Given Britain’s need for the Indian Army in 1939–45, adopting air control on the North-­West Frontier would have ranked high among the false economies of empire.

104

War and State-Building in Afghanistan

The RAF pursued this aim, against great resistance. In 1922, it advocated a trial involving ‘intensive bombing by 30–40 machines for months if necessary just to make life intolerable for the offending sections by preventing them from moving freely about, grazing cattle, tilling fields, etc., – upsetting, in fact, as completely as possible their normal mode of life’.76 The RAF was not allowed such a trial until 1925, when for 54 days, 26 aircraft attacked villages and ‘any targets which might present themselves’, meaning livestock and armed men. Yet, it avoided attacks on civilians, failed in attempts to set fire to crops, and carefully controlled damage, such as destroying terraced fields, which might ‘prejudice the resumption of normal economic life when peace should be made’. Privately, Edward Ellington, the RAF commander in India, admitted that the campaign had not achieved the basic aim, to make tribesmen submit.77 Meanwhile, soldiers conceded that the RAF had made an immense difference on the North-­West Frontier, yet had to keep it under their control. Despite constant struggles, they succeeded.78 Until 1947, air support became a routine but secondary element of military control on the frontier, restrained by political officers and controlled by the army. It bolstered the Indian Army’s power in blockade, and the ability to strike isolated foes and their flocks. Airpower dominated the campaign against the Faqir of Ipi, and sometimes killed scores of civilians, yet always was restrained. Britain treated the Muslims more gently than the Germans. The RAF understood the roots for its failures. In 1922, a senior airman, John Salmond, noted that soldiers were ‘weary of their lives in awful birdcages of wire’ in Waziristan, but would offer ‘the most uncompromising hostility’ to air policing, so long as it threatened massive cuts in the army.79 On this point, the British Indian Army had much support. John Maffey, then the Chief Commissioner of North-West Frontier Province, was the official most firmly in favour of air control, partly because he loathed the army’s forward policy. A ‘well-­funded and efficient’ RAF, he held, was ‘destined to furnish our true offensive in the tribal areas’, without need for ground support. Yet, even Maffey agreed that the RAF could not replace a field force against Afghanistan, which remained a significant role.80 A decade later, the Government of India rejected proposals for whole-­ scale substitution, which would have gone far to eliminate its massive budgetary deficit. The Indian Army said the proposal was infeasible, the RAF could not prove it would work. Hence, it ‘would endanger the security of the whole frontier and involve a risk, which would have serious reactions in India, and might implicate His Majesty’s Government by compelling the Government of India to ask for military assistance from the United Kingdom. The essence of the risk lies in the substitution of an unknown for a known factor.’81

Counterinsurgency and Empire, 1838–1947

105

Air policing in India was governed by political contradictions above, and below. Almost from the start, Britain built political systems in newly conquered Iraq, Palestine and Trans-Jordan around air control. To adopt it on the North-­ West Frontier would require a revolution in existing systems and relations with Pashtuns. Thus, in 1931, the Tribal Control and Defence Committee noted, ‘If the slate were clean, now that the Government has the power of punishment by air, it might be possible to dispense with payments for good will. But, the slate is not clean.’ To abandon allowances, central to frontier management, would be ‘an unforgiveable insult’; and ‘Pathan pride is proverbial’.82 Again, the RAF could wreck the economies of tribes, but Indian authorities, thinking poverty the root of instability on the frontier, sought to reduce it. One minister to Kabul warned that where punitive expeditions inflicted some economic damage and seized guns, air attack ‘is purely destructive, and since the problem of the frontier is largely economic, a punishment which destroys resources without reducing fighting power only aggravates the difficulties of that problem’.83 The RAF emphasized ‘that air action achieves its effect without causing much loss of life to the enemy’, but as Ellington noted, ‘P.O.s on the frontier are not altogether agreed that this is a good point. Many of them hold that it is a good thing to kill a considerable number of tribesmen for the moral effect it has on the tribes as a whole.’84 Conversely, killing civilians raised howls from British and Indian newspapers, and generals. Ironically, Orientalist views sheltered civilians from air attack, in part, because of the imprecision of contemporary airpower. In 1922, army officers thought that with good intelligence, ‘ “mullah-­chasing” should be a profitable employment for one or two aeroplanes – i.e. the instigating man who is behind every “jihad” should have his life made a burden to him’.85 Had Britain possessed aircraft as precise as Predators, it would have used them, as the West now does.

Frontier arithmetic This chapter has explained a negative: Why did Britain use COIN so rarely as a solution to its problems in Afghanistan, and the North-­West Frontier? These experiences regarding Afghanistan of a Western empire spread across the world, with one pillar in the Indian sub-­continent, are unique, but they have similarities to those of the Mughals and Pakistan. The root of British policy, its views of an internal, or an internal–external threat, are hard to generalize. No modern leader of India would say, as Britons did, that ‘India is as quiet as a barrel of gunpowder’.

106

War and State-Building in Afghanistan

Many Indian and Western leaders, however, might say that Pakistan is as quiet as a nuclear stockpile. For Pakistan, Pashtuns pose an atomic version of an internal– external threat. For all parties, COIN in Afghanistan and the FATA carries the risk of destabilizing more important places. British concerns with the links between Afghanistan, the North-­West Frontier and global jihad parallel modern views. These links are as strong as Britain ever faced, or feared, partly because its actions created these connections. Its operations in Afghanistan and the North-­ West Frontier encouraged the rise of a new ethno-­religious identity among Pashtuns, which linked manliness, resistance to occupation and the idea of holy warrior, or gazi. British diplomatic experience in managing Afghanistan, in Kabul and other capitals, easily can be extrapolated elsewhere. Britain showed an example of what to do with Afghanistan when one cannot conquer it or conduct COIN there. Britain succeeded when three conditions coincided: it had something to offer, was feared, and found Afghans willing to cooperate. Britain found solutions to some problems, along with a costly condition, which it could not overcome. Britain contained that condition, which was all it needed to do. Their condition is our inheritance.

Notes 1 General Sir Kenneth Wigram, ‘Defence in the North-West Frontier Province’, Journal of the Royal Central Asian Society, vol. 24, no. 1 (1937), pp. 74–5. 2 Sir Evelyn Howell, Mizh: A Monograph of Government’s Relations with the Mahsud Tribe (1931, reprint, Karachi: Oxford in Asia Historical Reprint Series, 1979), p. 80. 3 Garnet Wolseley, ‘South African Journal, 1879/80’, entry 24 October 1879, WO 147/7/ TNA, Kew, Surrey, UK. 4 For classic views, cf. Lord Wolseley, The Soldier’s Pocket Book for Field Service (1869); Charles Calwell, Small Wars, Their Principles and Practice (1896) and C.W. Gwynn, Imperial Policing (London: Macmillan, 1934). The best recent studies of these matters are T.R. Moreman, ‘ “Small Wars” and “Imperial Policing”: The British Army and the Theory and Practice of Colonial Warfare in the British Empire, 1919–1939’, Journal of Strategic Studies, vol. 19, no. 4 (1996), pp. 105–31, and David French, ‘The British Army and the Empire, 1856–1956’, in Greg Kennedy (ed.), Imperial Defence, The Old World Order: 1856–1956 (London: Routledge, 2008), pp. 91–110. 5 James Belich, The New Zealand Wars and the Victorian Interpretation of Racial Conflict (Auckland: Auckland University Press, 1980).

Counterinsurgency and Empire, 1838–1947

107

6 David French, The British Way in Counter-­insurgency, 1945–1967 (Oxford: Oxford University Press, 2011). The classic critique of the liberal theory of COIN, especially in its American form, is D. Michael Shafer, Deadly Paradigms (Princeton, NJ: Princeton University Press, 1988). 7 Alex Statiev, Social Conflict and Soviet Counterinsurgency in the Western Borderlands, 1944–50 (Cambridge: Cambridge University Press, 2009), especially the comparative analysis of modern counterinsurgencies, pp. 310–38. 8 Memorandum by H.M. Durand, 21 January 1887, in Viceroy’s Private Secretary to Bradford, 18 April 1887, L/PS/7/49, India Office Records and Library (IORL), British Library (BL), London, UK. 9 John Baird (ed.), Private Letters of The Marquis of Dalhousie (reprint, Shannon: Irish University Press, 1972), p. 189. 10 Lawrence to Wood, 27 May 1865, John Lawrence Papers, F 90/30, IORL, BL. 11 John Martineau, The Life and Correspondence of the Right Hon. Sir Bartle Frere (London: John Murray, 1895), p. 131; John Ferris, ‘The Internationalism of Islam: The British Perception of a Muslim Menace, 1840–1951’, Intelligence and National Security, vol. 24, no. 1 (2009), pp. 57–77. 12 Lawrence to Canning, 16 November 1858, Lawrence Papers, IORL, F 90/13, 13 Memorandum by Roberts, ‘The dangers to which a reverse would expose us’, 27 January 1891, L/MIL/17/14/80, IORL, BL. 14 Ferris, ‘Internationalism’, and Brandon Douglas Marsh, ‘Ramparts of Empire: India’s North-West Frontier and British Imperialism, 1919–1947’ (PhD dissertation, University of Texas, Austin, TX, 2009). 15 Minute by Hirtzel, 26 August 1919, L/PS/10/576, IORL, BL. 16 John Ferris, ‘Small Wars and Great Games – The British Army and Hybrid Warfare, 1700–1970’, in Williamson Murray and Peter Mansoor (eds), Hybrid Warfare: Fighting Complex Opponents from the Ancient World to the Present (Cambridge: Cambridge University Press, 2012), pp. 199–224. 17 For the background to these events, cf. Garry Alder, ‘Big Game Hunting in Central Asia’, Journal of Imperial and Commonwealth History, vol. 9, no. 3 (1981), pp. 318–30; David Gillard, The Struggle for Asia, 1828–1914 (London: Methuen, 1977); Edward Ingram, The Beginning of the Great Game in Asia: 1828–1834 (Oxford: Clarendon, 1979); M.E. Yapp, Strategies of British India, Britain, Iran and Afghanistan, 1798–1850 (Oxford: Clarendon, 1980). D.M. Peers, Between Mars and Mammon: Colonial Armies and the Garrison State in India, 1819–1835 (London: I.B. Tauris, 1995). 18 Mohan Lal, Life of the Amir Dost Mohammed Khan of Kabul, vol. 2 (1846, reprint, Karachi: Oxford University Press, 1978), pp. 42–5. 19 Auckland to Hobhouse, 10 May 1839, 1 April 1839, Elpinstone to Hobhouse, 29 September 1839, Broughton de Gyfford MSS, IORL, BL, F 213/8, IORL, BL. 20 Auckland to Elphinstone, 13 November 1839, IORL, BL, F 213/10. 21 Mohan Lal, Dost Mohammed Khan of Kabul, vol. 2, p. 388.

108

War and State-Building in Afghanistan

22 The best accounts are William Dalrymple, Return of a King: The Battle for Afghanistan, (London, 2013); Yapp, Strategies, pp. 307–460; Rob Johnson, The Afghan Way of War, How and Why they Fight (New York: Oxford University Press, 2011); Senzil Nawid, ‘The State, the Clergy, and British Imperial Rule in Afghanistan during the 19th and Early 20th Centuries’, International Journal of Middle Eastern Studies, vol. 29, no. 4 (1997), pp. 581–605. 23 Hari Ram Gupta, Life and Work of Mohan Lal Kashmiri, 1812–1877 (Lahore: Minerva Book Shop, 1943), pp. 337–59. 24 Lawrence to Wood, 27 May 1865, vol. 30, and Lawrence to Eastwich, 15 July 1866, vol. 39, John Lawrence MSS, F 90, IORL. 25 The best accounts are Christine Noelle, State and Tribe in Nineteenth Century Afghanistan (Padstow: RoutledgeCurzon, 1997), especially pp. 250–97, and Vartan Gregorian, The Emergence of Modern Afghanistan: Politics of Reform and Modernization, 1880–1946 (Stanford, CA: Stanford University Press, 1969), pp. 52–90. 26 H.M. Durand, The Life of Major-General Sir Henry Marion Durand, vol. 1 (London: W.H. Allen, 1883), p. 358. 27 John Robert Ferris, ‘Lord Salisbury, Secret Intelligence, and British Policy toward Russia and Central Asia, 1874–1878’, in Ferris, Intelligence and Strategy (London: Routledge, 2005), pp. 8–44. 28 Ferris, ‘Lord Salisbury, Secret Intelligence, and British Policy toward Russia and Central Asia, 1874–1878’, in Ferris, Intelligence and Strategy, p. 18. 29 Lady MacGregor (ed.), The Life and Opinions of Major-General Sir Charles Metcalfe MacGregor, vol. 2 (Edinburgh: William Blackwood, 1888), p. 107. 30 Nawid, ‘State’, pp. 589–90, is useful, despite errors of fact about British policy in 1878–80. 31 D.P. Singhal, India and Afghanistan, 1876–1907: A Study in Diplomatic Relations (Melbourne: University of Queensland Press, 1963), pp. 85, 89. 32 Gregorian, Emergence of Modern Afghanistan, pp. 129–62. 33 Durand to Lansdowne, 11 October 1893, 17 November 1893, Lord Lansdowne Papers, D 558, vol. 25, IORL. 34 No. 644, 8 July 1915, L/PS/10/473; Memorandum by India Office, 24 May 1916, ‘Situation at Kabul’, L/PS/10/593, IORL. 35 Minute by TWH (Holderness), 25 October 1915, L/PS/10/459, IORL. 36 John Ferris, ‘The British Empire vs. The Hidden Hand: British Intelligence and Strategy and “The CUP–Jew–German–Bolshevik Combination”, 1918–1924’, in Keith Neilson and Greg Kennedy (eds.), The British Way in Warfare, Power and the International System, 1856–1956, Essays in Honour of David French (Farnham: Ashgate, 2010), pp. 325–46. 37 Hamilton Grant to Arthur Balfour, 7 July 1919, Arthur Balfour Papers, The British Library, Add Ms 49749, BL.

Counterinsurgency and Empire, 1838–1947

109

38 Ferris, ‘The British Empire vs. The Hidden Hand: British Intelligence and Strategy and “The CUP–Jew–German–Bolshevik Combination”, 1918–1924’, pp. 325–46. 39 Chamier to Trenchard, 6 August 1925, RAF India to Air Ministry, 14 October 1925, AIR 5/608, TNA, Kew. 40 George Agabekov, O.G.P.U., The Russian Secret Terror (New York: Brentano’s, 1931). A tolerably accurate memoir by the Soviet intelligence chief in Kabul; solutions of Soviet diplomatic and intelligence telegrams from Afghanistan, by British code breakers, HW 12/117, No. 34443; HW 12/119, Nos 35132, 34998, passim, TNA. 41 HW 12/117, No. 34443; HW 12/119, Nos 35132, 34998; HW 12/85, No. 023721, 31 August 1926; HW 12/117, 34315, 4 April 1929, TNA. 42 Telegram from Viceroy to India Office, 14 February 1929, passim, FO 371/13992, N 1037; cf. FO 371/13995, FO 371/14003, TNA; Halifax to Peel, 13 February 1919, 4 April 1919, passim, Lord Halifax papers, C 152/5, IORL. 43 P(S) 2293/1941, Indian Political Intelligence (IPI) to Mr. Silver, 9 October 1941, IPI to Silver, 10 October 1941 P(S) 2293, L/PS/12/3249; Memorandum by IPI to Mr. Silver, 8 March 1946, L/PJ/12/615, TNA; Alan Warren. Waziristan, the Faqir of Ipi, and the Indian Army: The North West Frontier Revolt of 1936–37 (New York: Oxford University Press, 2000); Milan Hauner, India in Axis Strategy, Germany, Japan, and Indian Nationalists during the Second World War (Stuttgart: Klett-Cotta, 1980). 44 H.T. Lambrick, John Jacob of Jacobabad (Karachi: Oxford University Press, 1960). 45 Captain Lewis Pelly (ed.), The Views and Opinions of Brigadier-General John Jacob, C .B. (London: Smith, Elder, 1858), pp. 4, 349–51. 46 Lord Lytton, ‘Memorandum on Khelat’, 1 July 1876, Lord Lytton Papers, E 218/141, IORL, BL. 47 John Lowe Duthie, ‘Failure and Success: John Jacob’s Quetta Project, 1856–76’, Asian Affairs, vol. 10, no. 3 (1979), pp. 272–91; Thomas Henry Thornton, Colonel Sir Robert Sandeman, His Life and Work on our Indian Frontier (London: John Murray, 1895); Simanti Dutta, Imperial Mappings in Savage Spaces: Baluchistan and British India (Delhi: B.R. Publishing, 2002). 48 Singhal, A Study in Diplomatic Relations, p. 20; Thornton, Colonel Sir Robert Sandeman, His Life and Work on our Indian Frontier, p. 65; Lady MacGregor (ed.), Life and Opinions, p. 107; Sana Haroon, Frontier of Faith: Islam in the Indo-Afghan Borderland (New York: Columbia University Press, 2007), p. 22; ‘Statement Showing Tribal Allowances in the North-West Frontier Province’, 2 June 1931, passim, L/ PS/12/3150, IORL. 49 Keith Surridge, ‘The Ambiguous Amir: Britain, Afghanistan and the 1897 NorthWest Frontier Uprising’, Journal of Imperial and Commonwealth History, vol. 36, no. 3 (2008), pp. 417–34; Michael D. Kasprowicz, ‘1857 and the Fear of Muslim Rebellion

110

War and State-Building in Afghanistan

on India’s North-West Frontier’, Small Wars & Insurgencies, vol. 8, no. 2 (1997), pp. 1–15. 50 Wood to Lawrence, 15 January 1864, Charles Wood Papers F 78 L.B. 15, IORL, BL. 51 Hugh Beattie, Imperial Frontier: Tribe and State in Waziristan (Richmond: Curzon Press, 2002), pp. 120–41. 52 Captain H.L. Nevill, Campaigns on the North-West Frontier (London: John Murray, 1912); T.R. Moreman, ‘The Arms Trade and the North-West Frontier Pathan Tribes, 1890–1914’, Journal of Imperial and Commonwealth History, vol. 22, no. 2 (1994), pp. 187–216. 53 General Staff, Army Headquarters India, Operations in Waziristan: 1919–1920 (Calcutta: Superintendent Government Printing, 1921), p. 5; T.R. Moreman, The Army in India and the Development of Frontier Warfare: 1849–1947 (London: Macmillan, 1998); Christian Tripodi, Edge of Empire: The British Political Officer and Tribal Administration on the North West Frontier, 1877–1947 (Farnham: Ashgate, 2011); Michael Barthorp, The Frontier Ablaze: The North-West Frontier Rising, 1897–98 (London: Windrow & Greene, 1996). 54 Haroon, Frontier of Faith, p. 22. 55 Warren, Waziristan, p. 56. 56 Marsh, ‘Ramparts of Empire: India’s North-West Frontier and British Imperialism, 1919–1947’; ‘Report of the Tribal Control and Defence Committee’, 1931 L/ PS/12/3171, IORL. 57 Marsh, ‘Ramparts of Empire: India’s North-West Frontier and British Imperialism, 1919–1947’; Warren, Waziristan. 58 Haroon, Frontier, p. 22; Appendix to Finance Department, 14 January 1931, passim, R/1/4/1028, IORL. 59 ‘Report of the Tribal Control and Defence Committee’, 1931, L/PS/12/3171. 60 Government of India to India Office, No. 2 of 1931, Foreign and Political Department, 15 September 1931, L/PS/12/3171, IORL. 61 Chrstian Tripodi, ‘ “Good for One but Not the Other”: The “Sandeman System” of Pacification as Applied to Baluchistan and the North-West Frontier, 1877–1947’, Journal of Military History, vol. 73, no. 3 (2009), pp. 767–802. 62 Lord Roberts, Forty-One Years in India, From Subaltern to Commander-­in-Chief (London: Richard Bentley, 1897), pp. 336–7. 63 Major-General S.F. Muspratt, ‘The North-West Frontier of India’, Royal United Services Institution Journal, vol. 77, no. 507 (1932), pp. 472–4. 64 No. 688, 53-P.S., 6 March 1924, by H.N. Bolton, Chief Commissioner North West Frontier Province, L/PS/12/3260; Report of the Tribal Control and Defence Committee, 1931, L/PS/12/3171, IORL.

Counterinsurgency and Empire, 1838–1947

111

65 Paul Titus, ‘Honour the Baluch, Buy the Pushtun: Stereotypes, Social Organization and History in Western Pakistan’, Modern Asian Studies, vol. 32, no. 3 (1998), pp. 657–87; Hugh Beattie, ‘Negotiations with the Tribes of Waziristan 1849–1914: The British Experience’, Journal of Imperial and Commonwealth History, vol. 39, no. 4 (2011), pp. 571–87. 66 Tan Tai Yong, The Garrison State: The Military, Society and Government in Colonial Punjab, 1849–1947 (London: Sage, 2005). 67 John Ferris, The Evolution of British Strategic Policy: 1919–1926 (London: Macmillan, 1989); David Omissi, Air Power and Colonial Control: 1919–1939 (Manchester: Manchester University Press, 1990). 68 Minute by H.V. Cox, 4 August 1920, L/PS/10/766, IORL. 69 Memorandum by H.V. Cox, ‘Notes on Colonel Johnson’s Report on the Use of Tanks in India’, 6 August 1920, L/MIL/7/17133, IORL. 70 Viceroy, Telegram to India Office, 28 November 1921, L/PO/415, IORL. 71 Minutes by Hirtzel, 29 March 1923, 30 April 1923, passim, L/PS/12/3260, IORL. 72 ‘Instructions Governing the Employment of Armed Forces in the Maintenance of Tribal Control on the North-West Frontier of India and in Baluchistan’, 1940, passim, L/PS/12/3260; cf. L/PS/12/3187, passim, IORL. 73 Memo from the Resident in Waziristan to the Secretary of the Chief Commissioner NWFP, No. 1066, 13 May 1925, L/PS/12/3260, IORL. 74 2nd Meeting of the Indian Military Requirements Committee, 25 November 1921, CAB 16/38, TNA. 75 Memorandum by Deputy Director of Operations and Intelligence, 29 June 1925, ‘Outline Scheme for the Control of the North West Frontier of India by RAF’, AIR 5/413, TNA; ‘Air Staff Proposals for the Fuller Employment of Air Power on the North-West Frontier of India’, by W.G.H. Salmond, 16 January 1931, L/PS/12/3171, IORL. 76 Salmond to Trenchard, 21 July 1922, AIR 2/232 S. 19755, TNA. 77 Commander in Chief India to India Office, 29 June 1925, L/MIL 7/16950, IORL; Undated memorandum, c. July 1925 by internal evidence, no author cited, ‘Verbal Notes by Sir E. Ellington’, AIR 5/413, TNA. 78 Rawlinson to Hailey, 16 December 1921, Malcolm Hailey Papers, E 220/3B; Chetwode to Findlater Stewart, 23 November 1931, General Philip Chetwode Papers, IORL, D 714/17; Rawlinson to Montgomery-Massingberd, 5 September 1922, Montgomery-Massingberd Papers, vol. 135; Kirke to Skeen, 13 May 1926, Walter Kirke Papers, E 396/3, IORL. 79 Salmond to Trenchard, 21 July 1922, AIR 2/232 S. 19755, TNA. 80 Salmond to Trenchard, 21 July 1922, AIR 2/232 S. 19755, TNA; John Maffey, 8 September 1922, ‘Memoranda on India Report’, John Salmond Papers, AC 73/14, RAF Museum, Hendon.

112

War and State-Building in Afghanistan

81 Government of India to India Office, No. 2 of 1931, Foreign and Political Department, 15 September 1931, L/PS/12/3171, IORL. 82 ‘Report of the Tribal Control and Defence Committee’, 1931, L/PS/12/3171, IORL. 83 British Legation, Kabul, 18 April 1923, ‘Note’ by Humphreys, 16 April 1923, L/PS/12/3260, IORL. 84 AIR 5/413, TNA. 85 No. 6, Diary of Inspection Tour by A.V.M. Sir J. Salmond, n.d., c. July 1922, AIR 2/232 S. 19755, TNA.

4

The Conflict of War and Politics in the Soviet Intervention into Afghanistan, 1979–89 Pavel K. Baev

Introduction The failure of state-­building in Afghanistan, from which the United States and its coalition partners are presently withdrawing, may be sui generis in its unique complexity of causes, from ethnic tensions to opium economy, but it is too convenient to explain it away as the eternal feature of this ‘black hole’ between several historic civilizations. To foreign invaders, there is nothing particularly unruly or alien in the character of the peoples who inhabit this landlocked dead-­ end region, but many drivers of the still deepening disaster can be traced back to the national catastrophe caused by the ill-­conceived Soviet intervention.1 That unilateral and broadly condemned blunder is certainly very different from the ongoing UN-mandated operation that was started for all the right reasons and sustained with the best multilateral intentions, but the deep traumas of the former have badly complicated the latter and quite possibly condemned it to a sad end. This makes a re-­examination of the 10 years long and 30 years old Soviet attempt to provide ‘brotherly aid to the friendly Afghanistan’ not only academically interesting but also politically relevant, because it could provide some clues about the further trajectory of the humanitarian disaster that might disappear from the front pages after the Western disengagement but is certainly not going to sort itself out. There is a significant and expanding body of research focused on the impact of the Afghan war on the disintegration of the Soviet state and society, from the fateful decision-­making in the Politburo in late 1979 to the forlorn but still proud homecoming in early 1989.2 This chapter seeks to approach that war from a different angle, focusing on the shortcomings in projecting power that caused the escalation of rebellion and the subsequent defeat of the most powerful

114

War and State-Building in Afghanistan

military organization in the world. The aim is not to compile a list of limitations and mistakes but rather to examine how their interplays condemned the intervention to disaster and the Afghan state to failure. The first part deals with military matters on the tactical, operational and strategic levels. The second part re-­examines the ‘big picture’ of that war, including its political content, cultural meaning and historic setting. The conclusion offers a reflection on the fact that the lessons of that tragic and protracted cataclysm were never learned.

The war that the Soviet generals had never expected to fight The Soviet leadership was adamant to deny the plain fact that it was waging a war in Afghanistan, hiding behind figures of speech like ‘performing international duty’, and that ideological denial had significant consequences for strategic planning and for resource allocation to that peripheral but crucial theatre. The deployment of troops was expected to be a demonstration of commitment to ensuring Afghanistan’s stable development after the perfectly executed ‘special operation’ of replacing a dangerously radical usurper with a more reasonable leadership. The Soviet Army had had some tradition (going back to the civil war in Turkestan in the 1920s) but no experience in ‘unconventional’ warfare at the start of the 1980s, or for that matter fighting experience of any kind, while its training remained predominantly focused on confronting NATO. It showed the capacity for learning but the improved tactical skills and upgraded operations brought only greater destruction, which was counter-­productive in the absence of a coherent strategy and turned out to be politically unsustainable.

The tactics of holding the forts and hunting the hunters Unlike the muddled strategy, the Russian military establishment self-­critically evaluated the tactics of combat operations in Afghanistan both during the war and after its inglorious end.3 The ‘enemy’ in Afghanistan has not changed that much in the operational sense, so the Soviet experience in hitting and hammering on its weaknesses is worth a second look. What immediately strikes an observer who attempts a look back is the huge increase in tactical capabilities of the expeditionary forces produced by the two leaps in the ongoing Revolution in Military Affairs (RMA): one related to the introduction of high-­precision weapons, and the other driven by the arrival and continuous upgrade of information technologies.4 A Soviet razvedchik stumbling along some mountain

The Conflict of War and Politics, 1979–89

115

trek in Hindu Kush doesn’t look that that different from a British scout a hundred years before (Kalashnikovs make a difference, but Martini-Henri rifles were used by mujahideen to good effect), but has very little in common with a US ranger with his (or her) body armour, night-­vision goggles and Global Positioning System (GPS) navigator. Besides technical equipment, a major difference is that Soviet soldiers, even those of the better-­trained elite units, were conscripts, while the US and NATO forces comprise volunteers (the British Army in the nineteenth century did not use conscripts either), and the private security contractors also employ seasoned professionals. This very superficial comparison suggests as a rough assessment that on the same strength of deployed force (about 100,000–120,000 in total), the Western coalition in the early 2010s has at least three times greater tactical capabilities than the Soviet ‘contingent’ of the early 1980s. We might also assume that the current strength of the Taliban groups is several times (if not an order of magnitude) less than the grand total of mujahideen forces 25 years ago, while available modern communications have provided only a modest increase in their hit-­and-­run capabilities. Thus even more impressive are the tactical achievements of the Soviet 40th Army, which was able to prevail in the vast majority of tactical encounters and didn’t lose a single zastava (guard post) to enemy assault.5 In the initial phase of the war, that success was mostly the result of a complete lack of experience on the part of the newly formed rebel gangs, but as they gained skill, the Soviet command learned that regular army regiments were not a particularly useful asset, except for garrisoning key strategic points and for convoy duties. After the hollow victory of capturing the Panjsher Valley in May 1984 by a large armoured corps, from which the fighters of Ahmad Shah Massoud easily escaped, most combat operations were performed by various mobile units, such as airborne brigades and reconnaissance battalions, but above all by the famous Spetsnaz groups.6 Many operations involved demonstrative movement of heavy armour towards a suspect kishlak (village) with the Spetsnaz groups intercepting the retreating dushmans along their route of escape. On many other occasions, these units laid ambush to caravan trails in an attempt to cut supply to mujahideen strongholds in the mountains, thus engaging in guerrilla-­type operations against a guerrilla adversary.7 The effectiveness of these operations depended to a crucial degree upon the close air support provided by the tactical aviation and the helicopters of the army aviation; the Soviet command viewed better integration of land and air strikes as the right tactic against an elusive enemy, but this ‘lessons learned’

116

War and State-Building in Afghanistan

process was spoiled by the appearance of the STINGER portable surface-­to-­air missiles in autumn 1996. The MI-24 helicopter gunships were never able again to attack with impunity, so the STINGERs made a difference, even if they never were the ‘miracle weapon’ that decided the outcome of the war.8 To convert the tactical advantages into a tangible strategic success, the USSR needed a series of ‘surges’ in the mid-1980s, and had plenty of troops to reinforce the ‘limited contingent’ but was unable to follow this course owing to political considerations. The rebel forces grew in both numbers and experience, and even if they still lost most of the tactical encounters in the last phase of the war, a shift in balance was unmistakable.

The operations never went by the book – and never failed Soviet military operations in Afghanistan are so dissimilar to the ongoing ISAF activities that it would not be an over-­statement to suggest that the USSR invaded an entirely different country and conducted a counterinsurgency (COIN) campaign of a different nature.9 For one thing, it was physically far easier to move the troops in because there were two convenient entry points along the 2,350 km long border, at Kushka and Termez/Tirmiz. The supply routes were also a great deal shorter but the main focus of combat operations was nevertheless very firmly on the control of key roads. There is a certain similarity with current efforts at making the roads secure, particularly since the NATO troops require a lot more supplies, including water, but the difference is that the 40th Army faced a far higher risk of ambush, while the ISAF convoys are hit mostly by mines and improvised explosive devices (IEDs). The task that the Soviet command had to prioritize above everything else was keeping the Salang highway to Kabul open; in a typical episode of that ‘road war’, several hundred of Ahmad Shah Massoud’s fighters ambushed a column of tankers on that highway on 16 October 1986, but were counter-­attacked and pursued by mobile reinforcements.10 What is astounding, in hindsight, about the pattern of Soviet operations is that there were no terrorist attacks of the now so familiar kind, so the COIN of the 1980s had no counter-­terrorist dimension. This counter-­intuitive feature constitutes a part of a larger phenomenon. The guerilla warfare that the Soviet forces were unable to suppress was almost entirely rural in nature; there were no uprisings (like in Herat in March 1979) or street-­fighting in the cities, so the task of garrisoning them was entirely delegated to the ever unreliable Afghan Army and only marginally more capable police (Tsarandoi).11 It is significant that in the 1980s Kabul had only a third of its present-­day three million plus population

The Conflict of War and Politics, 1979–89

117

and that refugees were fleeing out of it rather than in. Yet, the proposition that the political agenda of the mujahideen alliance was not particularly attractive for the urban population, which was hardly that responsive to the Communist ideology but still remained committed to the ideas of modernization, appears entirely plausible. It is also the case that the operational task of interdicting the enemy supply routes from Pakistan, which was a high priority in the Soviet war, is less a concern for the coalition forces.12 The point is that the Afghan villagers had limited access to guns at the time of the Soviet invasion, so the supply of small arms and ammunition was crucial for keeping the guerilla campaign going, particularly in the isolated Panjsher Valley. Since that time, Afghanistan has become awash with weapons, making local supplies plentiful. What is in even sharper contrast, however, is the simple fact that the 40th Army did not have to perform any operations aimed at destroying the poppy harvest or eradicating opium production. The culture of poppy cultivation had existed for centuries but the colossal explosion in heroin export occurred only in the 1990s. Although the incentives for the victorious and feuding mujahideen to engage in this profitable business are clear, traditional agriculture was devastated during the course of the Soviet COIN.13 Exiting from the southern mouth of the Salang tunnel, the Soviet convoys came to the green Shomali Plain covered in orchards, olive groves and vineyards all the way to Kabul; nothing but barren hills meets a traveller’s eye nowadays.14 Overall, the Soviet operations were of higher intensity than those executed by the ISAF and the US forces, and consistently caused greater damage to the rebels. The paradox that the heavy losses sustained by the mujahideen translated into greater support for their cause was the result of a combination of a string of strategic blunders and disastrous political leadership.

The strategy of winning the time that was always on the enemy’s side Strategy was so clearly the weakest element in the Soviet war effort that it is fair to say that the USSR had several significant strategic advantages over US and NATO peace-­making, state-­building or terrorist-­hunting.15 First, the USSR opted for a unilateral intervention and so did not have to worry about keeping a coalition together and accommodating the preferences and limitations of maverick partners. The commander of the 40th Army had authority over all Soviet military units deployed in Afghanistan, and the length of his low-­speed

118

War and State-Building in Afghanistan

(by modern standards) line of communication to Moscow meant that he had more autonomy in making decisions and flexibility in implementing them than would normally have been the case in the rigidly hierarchic Soviet military organization. He and his subordinates, unlike those of General Stanley McChrystal, did not have to worry about pesky journalists, since all reporting from the ‘invisible’ war (the ‘CNN factor’ was yet to be introduced) was strictly censored. Other advantages included the rather high tolerance of its own casualties (though there was pressure to keep them below 150 a month) and the generally low burden of the combat operations for the Soviet military machine. The level of militarization of the Soviet Union was so high in the 1970s, that the Afghan War required no additional mobilization; the losses of equipment and munitions used were negligible compared with the total reserves accumulated for a hypothetical protracted conventional war. The army, which had started to suffer from the ‘force-­without-­war’ syndrome, gained valuable combat experience, which then was largely lost in the implosion of the USSR and massive downsizing of the military structures.16 Absorbing its own casualties, the Soviet command demonstrated complete indifference to the so-­called ‘collateral damage’ and had no problem ordering indiscriminate air strikes on villages and turning the Panjsher Valley into ‘scorched earth’. One significant limitation in this regard was the ban on using chemical weapons, despite many claims to the contrary in Western analyses.17 Another crucially important limitation was the firm political prohibition of any ‘hot pursuit’ into, or special operations in, Pakistan or Iran; thus the huge refugee camps were off-­limits to the Spetsnaz and provided perfect ‘safe havens’ for the mujahideen. The key problem about the Soviet strategy, however, was not the strict order forbidding cross-­border raids or the inability to close those borders, or even the inability to beef up the ‘limited contingent’ as the guerilla war was escalating, but the complete failure to define and assess the enemy. Lester Grau, one of the sharpest analysts of the Soviet intervention, wrote of an ‘ideological blind spot’ in the strategic thinking produced by the sheer impossibility in the MarxistLeninist theory of war of such a phenomenon as a popular uprising against ‘brotherly help’ by the leader of the socialist camp.18 Since it was theoretically inconceivable that the common people could take up arms against the Soviet intervention, it was impossible to propose that indiscriminate use of deadly force against civilians would multiply the support for the dushman. The empirical fact of escalation of insurgency contradicted the unshakable doctrinal proposition that the ‘working masses’ would inevitably come to recognize the

The Conflict of War and Politics, 1979–89

119

good and truth of socialism, in much the same way as in the 2000s they were expected to see the light of democracy. Political orthodoxy prescribed that time was on the side of ‘forces of progress’, and that wishful thinking dictated a rather passive approach to operations (prevalent in 1981 and in 1983–4), since it was more important to win time than battles, while minimizing the burden was seen as an essential condition for sustaining the deployment. The pragmatic assessment of the situation by the General Staff, however, dictated a pro-­active strategy (adopted in 1982 and again in 1985), since only by defeating the elusive enemy in as many encounters as possible could the resistance be suppressed before it engulfed one province after another. Each turn to offensive operations inflicted heavy losses on the mujahideen and massive casualties among the civilian population, and the anger over the latter helped the rebels to recover their strength in the period when the operations were curtailed, so the swings – determined to a large degree by the series of funerals in the geriatric Politburo – were highly unhelpful. It is interesting to note that the 40th Army had no responsibility for, or tasks related to, the proverbial policy of ‘winning hearts and minds’, which was implemented by other institutions, from the KGB to civil engineers constructing and restoring the electricity grids. That division of labour translated into bitter inter-­service rivalry, which resonated in the Kremlin, culminating in the idea of whether to withdraw troops, which was fully supported by the post-Ustinov Ministry of Defence but firmly opposed by the post-Andropov KGB.19 The USSR wasn’t able to apply the most efficient tool in the ‘hearts and minds’ policy – money, although having the financial resources hasn’t secured success for the international coalition, which has had to deal with many embittered hearts and traumatized minds. As for the Soviet military, its major contribution to building a ‘socialist-­oriented Afghanistan’ was supposed to be the equipping and training of the Afghan Army, and its achievements in fulfilling this task are rarely appreciated.20 The problem of recruiting (and desertion) was never satisfactorily resolved but the education of officers was rigorous (as was the KGB-directed training of the secret police, or State Information Agency, KhAD), and in the second half of the 1980s many units, such as the 53rd Infantry Division commanded by General Abdul Rashid Dostum, developed strong cohesion and loyalty to their commanders. The crucial flaw in the poorly developed Soviet strategy, which generally followed the pattern of ‘deterrence by punishment’, was the inability to follow its own logic that prescribed that the escalation of insurgency should be countered by an escalation of counter-­measures. The military would have had no problem

120

War and State-Building in Afghanistan

doubling the size of its contingent in Afghanistan and increasing the scale of atrocities, but the political leadership – first disorganized by natural causes, and then stunned by the fast-­unfolding implosion of the ‘one-­dimensional superpower’ – had no stomach for that.21 A connoisseur of historic irony would appreciate the paradox that the USSR collapsed under the burden of its own military machine – yet it did not dare to make full use of it in the only place where it was usable.

A war to start all wars The Soviet war in Afghanistan is often compared with the US war in Vietnam and sometimes also with the French War in Algeria, yet it is unique in one respect, which is central to the learned analyses presented in this volume; it was indeed ‘a war like no other’.22 The US-supported and internationally recognized victors did not even try to make Afghanistan into a functioning state but had continued the brutal destruction of their own country until the Taliban produced a vision of unified militarized theocracy and came close to implementing it across most of the ‘scorched earth’. This self-­propelling dynamics of violence, which appears set to outlast the ongoing peacemaking, is rooted in the Soviet debacle, which therefore deserves some benefit of hindsight.

Clausewitz stands to be slightly corrected From the very start, the Soviet deployment of a ‘limited military contingent’ was such a senseless mess that it appears to defy the Clausewitzean axiom about ‘war as a continuation of politics by other means’. It is even more ambiguous whether that piece of strategic wisdom is relevant for the mujahideen, if only because the Prussian theorist was not inclined to give serious consideration to guerrilla-­type wars, where the ‘other side’ rarely is a unified actor with a coherent agenda. There is little doubt, however, that fighting was only one – though maybe the most dramatic – driver in the complex political drama of destruction of the Afghan state. The well-­researched inevitability of the Politburo decision to intervene in the escalating crisis in Afghanistan supports the ‘policy-­first’ proposition, and invites the thought that the intervention was in fact a continuation of several different policies.23 On the highest level, we find the bi-­polar confrontation policy of the Cold War, which I examine briefly in the next sub-­section. There

The Conflict of War and Politics, 1979–89

121

was also a combination of a ‘historic-­progress’ ideological policy that promoted a particular path of modernization and the party-­political intrigues that sought to establish the CPSU leadership among the Communist parties of the world. On that basis, the former was perfectly content with the reformist effort of President Mohamed Daud, but the latter compelled Moscow to welcome warmly the Saur Revolution in April 1978.24 There was also the policy of cultivating close personal ties with the leaders of satellite states shaped by the authoritarian nature of the Soviet regime. It is this big-­brotherly ‘wet-­kiss’ policy that tipped the Kremlin scales in favour of a military intervention after the murder of Nur Mohamed Taraki in October 1979, while his urgent request for intervention prompted by an uprising in Herat in March of the same year was duly considered but turned down.25 This convoluted combination of disagreeable policies can supply an answer of sorts to the bewildering question: How could the Soviet leadership in the fateful year of 1979 know so much about Afghanistan and understand so little? Indeed, there was a mass of academic expertise at the Institute of Oriental Studies, a plethora of networks connecting the many Afghan doctors and engineers trained in Soviet universities, and particularly useful ties with hundreds of officers who graduated from the Frunze Academy and other military colleges, so information was plentiful and up to date. Processing was obviously the problem, and the ideological filter that eliminated all doubt that the working masses would internalize the communist doctrine was only a part of it. Another part was the performance of the overgrown bureaucratic nomenklatura where every subordinate learned to report upwards only what the boss wanted to hear, so the unpleasant truth never made it to the top.26 It could be logically established that the defeat in the war was a continuation of a larger and a more profound failure of the Soviet policy of building a military-­ bureaucratic ‘socialist’ superpower, but the second half of Clausewitz’s dictum is more problematic. The parallels with Vietnam and Algeria are quite misleading because the so-­called ‘national-­liberation movements’ led by Ho Chi Minh and Ahmed Ben Bella had a clear and popular anti-­colonial agenda, which was only marginally relevant for the constantly feuding mujahideen. At the same time, the quasi-­cultural explanations of the fierce tenacity of the resistance, following the ‘primordial-­predisposition-­to-­defiance’ line of reasoning, are definitely unsatisfactory. What offers a better clue is the investigation of how the rational but over-­zealous attempts to enforce modernization, which actually started with Daud’s tentative reforms, generated a strong anti-­modernization momentum in the conservative and predominantly rural society, particularly

122

War and State-Building in Afghanistan

when these attempts were accompanied by a massive foreign intervention. The counter-­modernization policy became a common platform for the motley crew of warlords, gained increasing support in the abused kishlaks, and was the only survival strategy for the refugees – leading straight to the Taliban.

War as a breaker of cultural codes Proceeding as a continuation of two incoherent and incongruent policies, the Soviet–Afghan war was also a continuation of two vastly different cultures, even if Clausewitz had little to say about this dimension of war. While the Soviet political culture was experiencing a fast and irreversible decomposition (in which the Afghan debacle was only a minor catalyst) in the 1980s, its military culture showed remarkable integrity. Many ugly features of that culture, including the inherently low value of human life, manifested themselves in the character of operations (as described above), where the so-­called ‘collateral damage’ was never an issue. Soviet casualties were a major concern primarily because there was a strict political directive on their minimization – but it concerned only combat losses, so the marked impact of infectious diseases was bracketed out. At any given moment, up to a third of the rank-­and-­file of the 40th Army was incapacitated by illness (hepatitis being the main culprit), so that 88 per cent of the total figure of 620,000 men and women who served in Afghanistan were recorded as seriously sick or wounded, while there were only 15,051 fatalities.27 Despite the widespread perceptions of the senselessness of brutality and sacrifices and despite the humiliation of defeat, this culture survived and was even strengthened by combat experience (that was mostly lacking in the Soviet Army in the 1970s), so the troops retreated home in early 1989 with flying colours. Paradoxical as it may seem, it was the culture of the winning side that was destroyed beyond repair. Claiming no expertise on the matter, I rely on the opinions of historians who describe the peculiar cultural patchwork of the old Afghan state as a mixture of self-­governed village-­communities (generally opposed or indifferent to the seat of power in Kabul), mild and moderate Islamic tradition, and a very particular code of conduct known as Pushtunwali, to which other tribes generally concurred.28 Many families owned guns (mostly from the early 1900s), but the blood feuds were moderated by customs of paying compensation, which could be characterized as a form of peace bargaining. These unique features remained entirely incomprehensible to the Soviet advisors and state-­builders who saw the weakness of the central authority as an easily

The Conflict of War and Politics, 1979–89

123

amendable deficiency rooted in the heritage of ‘medieval fragmentation’ and the influence of Islam as an archaic habit that would fade away under the influence of communist ideas. The clash of cultures was only partly mitigated by the strong discouragement of officers and men to communicate with the locals, which did not stop the illegal trade in all sorts of military supplies and the spread of drug abuse in the ranks.29 The delicate Afghan culture was, however, badly distorted by the protracted and all-­penetrating war. The kishlak lost its traditional integrity as a self-­ reproducing political and social unit; the fear of indiscriminate bombing drove millions of villagers to abandon their homes, and they moved not to the cities (as is often the case in civil wars) but to Northern Pakistan and Eastern Iran. The huge refugee camps became the hothouses in which a new culture was breeding. The totality of war determined the austere warrior character of this culture, which revived and relied upon the myths of Afghan ferocious defiance against invaders. Some warrior features had certainly been present in the Pushtunwali as well but more as a tribute to tradition, because Afghanistan actually had been a mostly peaceful country for three-­quarters of the twentieth century. The new code of conduct in bello combined a complete de-­humanization of the enemy with a stoic acceptance of the miniscule price of human life – and incorporated a recognition of the importance of ‘good publicity’ in the West acquired through emphasizing the suffering of innocent civilians. It is also possible to suggest that unlike the traditional hierarchic family–clan–tribe identity structure, the new culture was more fluid and stimulated the growth of networks based on shared wartime experiences. The destruction and breakdown of traditional cultural values and ties made Islam into the central element of the new culture, and the tendency towards radicalization was a natural response to the escalation of violence. The Soviet ideology was aggressively atheist, and the Afghan communists saw the influence of Islam as a sign of backwardness, so their policies of ‘winning hearts and minds’ centred on land reform and education had nothing to offer in terms of engaging the crucially important beliefs. The thin layer of civic urban sub-­ culture that many Western travellers to Kabul found pretty attractive in the mid-1970s was wiped away;30 the Muslim clergy became a committed ally for the rebels, which paved the way for the emergence of such a unique phenomenon as the Taliban in the mid-1990s, when the children born in the refugee camps became teenagers. A sharp reflection on this culture can be found in Joseph Brodsky’s short poem ‘On negotiations in Kabul’, where a meeting between Western ambassadors

124

War and State-Building in Afghanistan

and ‘cruel-­hearted mountain tribes’ is vividly described, resulting in the ‘serpent signature on the agreement’ – ‘and then there is nothing, none to see there, none at all . . .’31

The distorted perspective of the Cold War The particular place of the Soviet–Afghan War in the larger context of the Cold War is carefully researched, yet a bit of re-­examination is warranted.32 The point is not to savour the historic irony of the motivation in the geriatric Politburo to intervene in the self-­propelling crisis in Afghanistan in order to prevent the neighbour from ‘falling into American hands’ (like Somalia or Egypt did earlier in the 1970s), while nowadays Moscow would rather prefer these hands to remain tied with this problem for as long as possible.33 What really needs some re-­evaluation is the eagerness of the US to exploit the Soviet blunder and even to facilitate it by providing aid to the rebels since the summer of 1979, according to the self-­congratulatory Zbigniew Brzezinski.34 His idea of turning Afghanistan into a ‘Soviet Vietnam’ appeared dubious to President Jimmy Carter but gained perfect traction in Washington with the arrival of President Ronald Reagan.35 The proposition of adding to your mortal enemy’s troubles appeared – and still appears to many Cold War warriors – perfectly reasonable, and the risks involved were assessed as low-­to-­non-­existent compared with the expected returns on the entirely affordable investment. Certainly, the career of the late Osama bin Laden casts a shadow over the logic of this proposition, but the extermination of the ‘evil one’ makes it possible for the patriotically minded US politicians to dismiss any doubt that the adventures of Rambo and James Bond among the Afghan mujahideen were entirely appropriate. The real problem, however, was greater than merely misjudging one manic-­terrorist character among dozens of ‘freedom fighters’ and ‘noble savages’. The US leadership consciously and strategically built an alliance with the forces that pursued a counter-­modernization agenda and preached an increasingly radical version of Islam. The collapse of the state institutions in Afghanistan in the mid1990s was therefore not a consequence of Western neglect of the humanitarian needs of the war-­traumatized country but a pre-­determined result of the mujahideen victory. This assertion might appear far-­fetched but it emerges with undeniable clarity from the records of Soviet–American diplomatic and political exchanges in the second half of the 1990s.36 Mikhail Gorbachev was desperate to explain that the

The Conflict of War and Politics, 1979–89

125

new Soviet leadership actually did care about Afghanistan and had neither imperial ambitions nor concerns about face-­saving (which was perhaps a bit of a stretch), but failed to make an impression. It was the December 1987 Washington Summit that proved to be the crucial moment as Gorbachev developed a distinct impression that Reagan agreed to stop supplying arms to the rebels – only to be bitterly disappointed the following month.37 Gorbachev’s main argument was the inevitable seizure of power by Islamic radicals. And he hoped the unhappy experience in Iran would make Washington receptive, so prior to the summit the Chairman of the KGB Vladimir Kryuchkov tried to impress upon CIA Director Robert Gates that, ‘You seem fully occupied in trying to deal with just one fundamentalist state.’38 In reality, it was the non-­ fundamentalist Iraq that shaped up as the main troublemaker at that time, so Gorbachev’s argument fell perfectly flat. One of the reasons for Reagan’s team’s reluctance to look one step further was the non-­negotiable demand from Pakistan (firmly backed by Saudi Arabia) to ensure the destruction of the Najibullah regime, and that influence revealed a very dubious value of the strategic alliance between the USA and Pakistan. Washington had to turn a blind eye to the advancement of the nuclear programme by this ally and to pretend that the fast growth of Islamic radicalism was neutralized by the expanded ties with the Directorate for Inter-Service Intelligence (ISI), and those compromises set the bilateral relations on the trajectory that has brought them to the present-­day deep crisis of trust.39 Much the same way as the Politburo was unable to see the real picture of popular uprising against the intervention, the US administration was in denial of the true agenda of the forces it helped to mobilize and provided political protection for.

Conclusions The desire to have complete closure on the Afghan debacle prevailed in Moscow over the rational choices for learning some political and strategic lessons from that sustained effort or using the accumulated experience to reform the army. For the Russian leadership, the limited impact of that misadventure on the spontaneous implosion of the USSR was quite obvious, while the consequences were seen as the problem for concerned neighbours – and the newly reconstituted Russian state was no longer one of those.40 The complexity of the dual task of advancing economic reforms and stabilizing the political

126

War and State-Building in Afghanistan

processes inside the turbulent space delimited with uncomfortable borders determined the pronounced intention in Moscow to disengage from Central Asia. That political retreat, however, turned out to be short-­lived, and already in late 1992, Russia had found itself executing a military intervention in the ferocious civil war in Tajikistan.41 In the late 1990s, Moscow tried in vain to raise the alarm about the threat to international peace represented by the Taliban and, in a remarkable turn of strategic priorities, helped Ahmad Shah Massoud to defend his stronghold in the Panjsher Valley. The Russian Army never had an opportunity to upgrade its rapid-­deployment capabilities as it was pushed into massive downsizing and simultaneously had to perform a series of experiments with post-Soviet peace-­enforcement in the first half of the 1990s.42 It was only the humiliating defeat in the First Chechen War that sought reflection on the politically misguided ‘conquest’ of Afghanistan, in which multiple operational victories shaped a strategic defeat.43 No retrospective analysis can establish with any certainty whether the war had a ‘military solution’ or not, but it is quite clear that the USSR in its autumnal decade had neither the Stalinist determination nor the Leninist ingenuity to find one.44 It was able to absorb the defeat, but Gorbachev’s team, which replaced the cohort of elders responsible for launching the experiment, found little glee in observing how their predictions of the bloody mess in Afghanistan following a rebel victory came true. The rather typical description of the catastrophe that Afghanistan went through in the 1980s as a ‘national liberation’ war, in which the fearless mujahideen warriors stood against and defeated the imperialistic aggressor is patently – and conveniently – simplistic. The more sophisticated interpretation that adds the Cold War context so that the US comes out as the patron of ‘freedom fighters’ and the localized jihad becomes a part of the global Crusade against the ‘evil empire’, is also far from satisfactory, even if it serves the never slackening American demand for self-­glorification. What is missing from these depictions of a ‘good war’ is the acknowledgement of the failure of the Afghanistan modernization project that had been set in motion in the 1970s, and that the USSR sought to advance according to its distorted understanding of ‘historic progress’. The 40th Army was a formidable fighting force, and it took a mobilization of the most conservative groups and ideas in the shattered Afghan society to countervail it. The US fostered that mobilization empowering the most rabid of Islamic radicals and exploiting the traumas of destruction of the peace-­bargaining culture; the consequences of a merciless victory appeared insignificant – until proven otherwise.

The Conflict of War and Politics, 1979–89

127

Notes 1 This chapter is based on my presentation at the PRIO conference ‘COIN in Afghanistan: From the Mughals to the Americans’ (Oslo, 12–13 February 2012). It draws on my earlier research into this theme presented in Pavel K. Baev, ‘Sad Wisdom of Hindsight: Soviet Intervention in Afghanistan, 1979–1989’, in Cheryl Bernard, Ole Kværnø, Peter Dahl Thruelsen and Kristen Cordell (eds), Afghanistan: State and Society, Great Power Politics and the Way Ahead (Santa Monica, CA: Rand, 2008), pp. 37–46. 2 The most valuable recent addition to this literature is Rodric Braithwaite, Afgantsy: The Russians in Afghanistan 1979–89 (London: Profile Books, 2011). A unique source on the decision-­making in 1979 is David A. Welch and Odd Arne Westad (eds), The Intervention in Afghanistan and the Fall of Détente, Nobel Symposium 95 (Oslo: The Nobel Institute, 1996). On the war ending, see Artemy M. Kalinovsky, A Long Goodbye: The Soviet Withdrawal from Afghanistan (Cambridge, MA: Harvard University Press, 2011). 3 Detailed examination of many tactical episodes by the faculty of the Frunze Military Academy can be found in Lester W. Grau, The Bear Went Over the Mountain: Soviet Combat Tactics in Afghanistan (London: Frank Cass, 1998; the Russian original text is from 1991). 4 A thoughtful examination of the RMA debates can be found in Colin S. Gray, Strategy for Chaos: Revolution in Military Affairs and the Evidence of History (London: Routledge, 2004). A useful recent analysis is Barry D. Watts, The Maturing Revolution in Military Affairs (Washington, DC: Center for Strategic and Budgetary Assessments, 2011). 5 Popular action film The 9th Company is loosely based on a real combat episode but only six soldiers were actually killed [see http://en.wikipedia.org/wiki/9th_ Company]. 6 A remarkable collection of first-­hand accounts, particularly on the Afghan war, is S. Kozlov (ed.), Spetsnaz GRU-2: Voina ne okonchena, istoriya prodolzhaetsya (Spetsnaz GRU-2: The War is Not Over, History Continues) (Moscow: Russkaya Panorama, 2002). 7 The popular Russian TV series Hunters for Caravans (2010) waxes heroic about those operations. 8 Edward B. Westermann, ‘The Limits of Soviet Airpower: The Failure of Military Coercion in Afghanistan, 1979–1989’, Journal of Conflict Studies, vol. 19, no. 2 (1999), pp. 39–71. 9 One careful examination of the Soviet operations is Gregory Feifer, The Great Gamble: The Soviet War in Afghanistan (New York: Harper Collins, 2009).

128

War and State-Building in Afghanistan

10 Ruslan Aushev, unaware at that time that he would become the President of Ingushetia, was wounded in that action and made the Hero of the Soviet Union; see Mark Franchetti, ‘Can the West Avoid Russia’s Fate in Afghanistan?’, The Sunday Times, 3 January 2010. 11 Instances of attacks in cities are meticulously recorded in Chapter 15 of Ali Ahmad Jalali and Lester W. Grau, Afghan Guerilla Warfare: In the Words of the Mujahideen Fighters (London: Compendium, 2001). 12 Kristian Berg Harpviken gives an expert examination of the cross-­border networks of the Taliban in his Social Networks and Migration in Wartime Afghanistan (London: Palgrave Macmillan, 2009). 13 The authoritative source on the drug economy is Afghanistan Opium Survey 2011 (Vienna: UNODC, December 2011). A useful journalistic account is Alissa J. Rubin, ‘In Afghanistan, Poppy Growing Proves Resilient’, New York Times, 1 January 2012. 14 Braithwaite, Afgantsy, pp. 206–7. 15 Scott R. McMichael, ‘The Soviet Army, Counter-Insurgency, and the Afghan War’, Parameters, vol. 19 (December 1989), pp. 21–35. 16 One of the key reasons for the under-­utilization of that experience was the absence of a corps of professional sergeants in the Russian army. A very useful memoir is Boris Gromov, Ogranichennyi Kontingent (The Limited Contingent) (Moscow: Kultura, 1994). 17 One example is David C. Isby, ‘Soviet Tactics in the War in Afghanistan’, Jane’s Defence Review, vol. 4, no. 7 (July 1983), pp. 681–92. 18 Lester W. Grau and Michael A. Gress, The Soviet–Afghan War: How the Super-Power Fought and Lost (Lawrence, KS: University Press of Kansas, 2002), ‘Editor’s Preface’, p. xix. 19 Dmitry Ustinov, the Soviet Defence Minister since 1976, died on 20 December 1984, in the same year as Yuri Andropov, the head of the KGB since 1967 and the General Secretary of the CPSU from 12 November 1982 (after Leonid Brezhnev’s death) to 9 February 1984. Both had a crucial role in the decision to invade Afghanistan in December 1979. On the disagreements between the military and the KGB, see Chapter 7 (‘The Army Withdraws and the Politburo Debates’) in Kalinovsky, A Long Goodbye. Also useful is Diego Cordovez and Selig S. Harrison, Out of Afghanistan: The Inside Story of the Soviet Withdrawal (Oxford: Oxford University Press, 1995). 20 One self-­serving but still useful account is Mahmud Gareev, Afganskaya Strada (Hot Times in Afghanistan) (Moscow: Insan, 1999/2002). 21 A noteworthy assessment is Paul Dibb, The Soviet Union: The Incomplete Superpower (Chicago, IL: University of Illinois Press, 1986). 22 Victor Davis Hanson used this definition for the title of his excellent book on the Peloponnesian War (A War Like No Other: How the Athenians and the Spartans

The Conflict of War and Politics, 1979–89

129

Fought the Peloponnesian War (New York: Random House, 2005); no analogy is implied. 23 As it happened, all key decision-­makers had died (leaving no memoirs) before the war came to the end; one good source is Alexander Lyakhovsky, ‘Inside the Soviet Invasion of Afghanistan and the Seizure of Kabul’, Cold War International History Working Paper 51 (Washington, DC: Woodrow Wilson Center, 2007). 24 Henry Bradsher, Afghan Communism and the Soviet Intervention (Oxford: Oxford University Press, 1999). 25 Those deliberations are evaluated in Alexander Lyakhovsky, Tragediya i Doblest Afgana (Tragedy and Valor of Afghanistan) (Moscow: Nord, 2004). 26 On the stifling of dissent, see Odd Arne Westad, ‘Concerning the Situation in “A”: New Russian Evidence on the Soviet Intervention in Afghanistan’, Cold War International History Project Bulletin, 8/9 (Washington, DC: Woodrow Wilson Center, 1996). 27 This official figure was questioned by many Western scholars who were inclined to grant a modicum of credibility to the data supplied by mujahideen; see, for instance, Antony Arnold, The Fateful Pebble: Afghanistan’s Role in the Fall of the Soviet Empire (Novato, CA: Presidio, 1993). By now, however, the questions about this exercise in Mikhail Gorbachev’s glasnost have been eliminated. He also describes the collapse of the medical services; see Braithwaite, Afgantsy, pp. 173–5, 323–30. 28 On the micro-­communities, see Amin Saikal, Modern Afghanistan: A History of Struggle and Survival (London: Taurus, 2004). He also elaborates on such a peculiar factor as royal polygamy. On the Pushtunwali, see ‘Honor Among Them’, The Economist, 19 December 2006; and also David Loyn, Butcher and Bolt: Two Hundred Years of Foreign Engagement in Afghanistan (London: Hutchinson, 2008). 29 Artyom Borovik, The Hidden War: A Russian Journalist’s Account of the Soviet War in Afghanistan (New York: Grove Press, 1990). 30 Braithwaite, Afgantsy, pp. 34–5 reminds that Kabul was a colourful stop on the Hippie Trail. 31 My translation from Joseph Brodsky, ‘K peregovoram v Kabule’ (1992); the topic is highly unusual for the poet, who didn’t live to see that far-­away war coming to his New York. 32 One impeccably precise analysis is Odd Arne Westad, The Global Cold War: Third World Interventions and the Making of Our Times (Cambridge: Cambridge University Press, 2005). 33 One useful reflection on old mistakes influencing current perceptions is Dmitri Trenin and Aleksei Malashenko, Afghanistan: A View from Moscow (Washington: CEIP, 2010). As this chapter was finalized, Russian Foreign Minister Sergei Lavrov scolded NATO for setting the ‘artificial’ deadline for withdrawal in 2014 and urged the Alliance to remain engaged in Afghanistan; see Fred Weir,

130

War and State-Building in Afghanistan

‘Russia Urges NATO to Stay in Afghanistan Beyond 2014’, Christian Science Monitor, 19 April 2012. 34 In the famous interview with Le Nouvel Observateur (15–21 January 1998), he didn’t claim full credit for prompting the Soviet intervention but asserted that ‘we knowingly increased the probability that they would’; one sharp analysis is Steve Coll, Ghost Wars: The Secret History of the CIA, Afghanistan, and bin Laden, from the Soviet Invasion to September 10, 2001 (New York: Penguin Press, 2004). 35 Gabriella Grasselli, British and American Responses to the Soviet Invasion of Afghanistan (Aldershot: Dartmouth Publishing, 1996). 36 Chapter 5 (‘Engaging with the Americans’) in Kalinovsky, A Long Goodbye, examines those miscommunications in great detail. 37 David Ottaway, ‘US Links Aid Cutoff to Pullout by Soviets’, Washington Post, 8 January 1988. 38 Robert Gates, From the Shadows: The Ultimate Insider’s Story of Five Presidents and How They Won the Cold War (New York: Touchstone, 1997), p. 425. 39 For an elegant examination of this trajectory, see Anatol Lieven, Pakistan: A Hard Country (London: Allen Lane, 2011). 40 For a thoughtful examination of that impact and the domestic consequences, see Mark Galeotti, Afghanistan: The Soviet Union’s Last War (London: Frank Cass, 1995). 41 That intervention turned out to be the only successful Russian effort at terminating the post-Soviet conflicts, and the 1997 peace agreement consolidated the clan-­based authoritarian regime of Emomali Rahmon. See Shirin Akiner, Tajikistan: Disintegration of Reconciliation? (London: Royal Institute of International Affairs, 2001). 42 My analysis of that de-­militarization and interventionism is in Pavel K. Baev, The Russian Army in a Time of Troubles (London: Sage, 1996). 43 Robert M. Cassidy, Russia in Afghanistan and Chechnya: Military Strategic Culture and the Paradoxes of Asymmetric Conflict (Carlisle Barracks, PA: US Army War College, 2003). 44 An interesting counterfactual case for achieving a ‘pacification’ of Afghanistan by provoking a new war between India and Pakistan is developed by David C. Isby, ‘Afghanistan: The Soviet Victory’, in Peter G. Tsouras, Cold War Hot: Alternative Decisions of the Cold War (London: Greenhill Books, 2003), pp. 208–32.

5

Al-Qaeda versus Najibullah: Revisiting the Role of Foreign Fighters in the Battles of Jalalabad and Khost, 1989–92 Anne Stenersen

Introduction After the Soviet withdrawal from Afghanistan in 1989, groups of local insurgents continued the fight against the Soviet-­sponsored Afghan regime led by President Najibullah. At the centre of this alliance were the four so-­called ‘fundamentalist’ mujahideen parties that had been based in Peshawar during the war: Jamiat-­eIslami led by Burhanuddin Rabbani, Hizb-­e-Islami Hekmatyar led by Gulbuddin Hekmatyar, Hizb-­e-Islami Khalis led by Yunus Khalis, and Ittehad-­e-Islami led by Abd Rabb al-Rasul Sayyaf. These were also the parties that had cooperated most closely with the ‘Afghan-Arabs’, the Arab volunteers who had come to the region during the 1980s to support the mujahideen’s war against the USSR. The ‘Afghan-Arabs’ movement has been subject to much analysis since 2001, not least because it gave rise to Al-Qaeda, one of the world’s most dangerous terrorist organizations. Al-Qaeda was formed by a group of Afghan-Arabs in Peshawar in 1988. After the Soviet Union had pulled out of Afghanistan, and Al-Qaeda had suffered bitter losses in the battle against Najibullah’s army at Jalalabad in 1989, Osama bin Laden decided to move his headquarters to Sudan. He did not return to Afghanistan until 1996, when Sudanese authorities requested him and his organization to leave the country. Less well-known is the fact that a number of Afghan-Arabs stayed in Afghanistan and Pakistan during this period, where they ran training camps, took part in local battles, or simply hid from being prosecuted in their home countries. The activities of these Afghan-Arabs have generally been under-­studied in existing literature. The activities of Afghan-Arabs in Afghanistan after the Soviet withdrawal are interesting to study for several reasons. First and foremost, it provides a unique

132

War and State-Building in Afghanistan

example of how the withdrawal of a common, external enemy (the Soviet Union) affects the activity of foreign jihadi fighters in a conflict. This may make us better equipped at understanding today’s challenges in Afghanistan, especially as the Taliban and Al-Qaeda’s common, external enemy, the US and NATO, is scheduled to leave Afghanistan by 2014. One should be careful at drawing direct parallells between the two conflicts. As argued by Bruce Riedel and others, there are probably more differences than likenesses between the two wars.1 Yet, there are some striking similarities with regard to the foreign fighters’ preferred areas of operation (southeastern and eastern Afghanistan), ideological driving forces (pan-Islamism and jihadism), and choice of allies (local commanders with a direct connection to the 1980s ‘fundamentalist’ mujahideen parties). To bridge an existing gap in Al-Qaeda and Afghanistan-­focused literature, this paper analyses the role of the Arab fighters in Afghanistan in the period from 1989 to 1992, when Afghanistan was ruled by the Soviet-­sponsored Najibullah regime. The main research question is: How did the withdrawal of the Soviet Union in 1989 affect the military role of the Arab fighters in Afghanistan? Two battles in particular will be examined: the Battle of Jalalabad in 1989 and the Battle of Khost in 1990–1. The reason for choosing these two battles in particular is that the participation of independent, Arab-­led fighting groups is well documented in both cases. Moreover, the battles represent two different stages of the Afghan conflict, and this enables us to compare and contrast the military participation of Arabs on the Afghan battlefield during changing external circumstances.

Background A paradox yet to be explained fully in existing literature is that the largest military participation of Afghan-Arabs in the Afghan–Soviet war took place after the Soviet forces had withdawn from Afghanistan. It is a paradox because the presence of a ‘foreign occupier’ on Muslim soil is generally the main motivator for jihadists who go abroad to take part in other Muslim struggles. If this were the case, one would expect the foreign fighters to lay down their arms as soon as the foreign occupier leaves the country. However, the reality is more complex. In Afghanistan, the issue was probably related to timing and local conditions. In the first years of the war in Afghanistan, it was hard for Afghan-Arabs to gain access to the battlefield. The mujahideen commanders were interested in money and weapons, not volunteer fighters. Foreign volunteers were often inexperienced

Al-Qaeda versus Najibullah, 1989–92

133

and unfamiliar with local terrain, language and culture. Moreover, there was no organization to receive, organize and train Arab volunteers until the Services Office was established by Abdullah Azzam and Osama bin Laden in 1984. The Services Office started to organize military training camps for Arabs, but it was still hard to get accepted as fighters at the front. Jalaluddin Haqqani’s base in Khost was an exception – he was one of the few Afghan commanders who were willing to use Arab fighters as a military asset during the war. Then, in 1987 bin Laden took the initiative to establish an all-Arab base inside Afghanistan called Al-Ma’sada – ‘The Lion’s Den’. In May 1987, the Soviets carried out a three-­week offensive in the Jaji area known as the ‘Battle of Jaji’. During this offensive, Al-Ma’sada was besieged by Soviet troops, but the Arabs managed to withstand the attack. The event earned bin Laden and his fighters great respect. This was a great boost of self-­confidence for the Afghan-Arabs, who now believed themselves capable of taking on the Soviets directly. This was late in the war – only one year before the Red Army started pulling out from Afghanistan – which was probably the main reason for why the largest battles of the Arabs in Afghanistan took place after 1988. It is worth noting that the Arabs’ military participation in the Jaji battle was neither very large nor significant. The Battle of Jaji was not simply an attack against Al-Ma’sada – it was a broad Soviet-­led offensive into Jaji to secure one of the Soviet border forts there. The offensive was met with fierce resistance from thousands of Afghan mujahideen that included groups from many – if not all – of the seven resistance parties. Moreover, it included commanders who joined the battle from outside the Jaji area, most notably Jalaluddin Haqqani and commanders based in Kabul province. Due to the defeat of the Soviets, the battle stands out as one of the most important victories of the mujahideen during the entire Soviet–Afghan War.2 The Arabs contributed by holding a small piece of territory, but this hardly had any strategic significance for the mujahideen as a whole. On the other hand, the ‘victory’ at Al-Ma’sada had several ramifications for the Afghan-Arabs movement. Arab journalists, some of whom had been present during the battle, wrote vivid accounts of the ‘divine miracles’ that had ensured the Arab victory at Al-Ma’sada. Such myth-­making was important for the recruitment of new foreign volunteers to Afghanistan. The battle also benefitted bin Laden personally. He became known as a military hero in Afghanistan, not solely a financier.3 This contributed to empowering him and his new organization, Al-Qaeda, which was formed in Peshawar in 1988.

134

War and State-Building in Afghanistan

The Soviet withdrawal and its aftermath The withdrawal of Soviet troops from Afghanistan started in 1988 and was completed on 15 February 1989. However, the Soviet Politbureau’s decision to withdraw had been taken already back in 1986, before bin Laden’s guerrilla groups started playing an independent role in the war. Thus, it is hard to argue that bin Laden’s fighters played any role in the military defeat of the Soviets, although bin Laden and his network of Gulf financiers made substantial material contributions to the war. After Al-Qaeda was established in 1988, it was decided that the organization should focus its efforts on international jihad, while Afghanistan should mainly be used as a venue for training and recruitment. Still, bin Laden put considerable resources into supporting the Battle of Jalalabad, which took place between March and June 1989. This was the first major attempt by mujahideen to seize power in Kabul after the departure of the Soviet Army. The plan, which was prepared by the Seven-Party Union in Peshawar and the Pakistani Inter-Services Intelligence (ISI), was to seize the eastern Afghan city of Jalalabad and use the site as a launch pad for a military offensive on Kabul itself. The ultimate goal was to oust the Najibullah regime and install a mujahideen government in its place.4 The event illustrates that the departure of a ‘foreign occupier’ does not necessarily mean that the so-­called ‘ideologically motivated’ jihadi fighters will leave the conflict. This gives reason to question the role of ideological motivation for foreign fighters in a conflict area over time. To be sure, the mujahideen who continued to fight Najibullah after 1989 claimed that there was still an ideological justification for doing so. They viewed Najibullah as a ‘Soviet puppet’ with whom reconciliation was unacceptable. Fighting Najibullah was a religious duty, it was argued, because he was not a real Muslim but an ‘agent’ of the Soviet Union. Moreover, Najibullah had been responsible for the death of hundreds of Afghan Muslims during his time as chief of the Afghan intelligence bureau (KhAD). He was therefore seen as a criminal who should be punished by death according to Islamic law. In reality, Najibullah sought to distance himself from the Soviet Union and Communist ideology after 1989 – for example, by declaring Afghanistan an ‘Islamic State’ – for the purpose of accommodating Islamist parties into the government. Many former mujahideen parties, especially the so-­called ‘moderates’, did in fact put down their weapons after the Soviet withdrawal, and supported a negotiated solution. But this was not enough for the fundamentalist mujahideen factions who believed that ‘the blood of Afghan martyrs would be

Al-Qaeda versus Najibullah, 1989–92

135

shed in vain’ unless a fundamentalist Islamic government was installed in Afghanistan. The departure of the Soviet Union from Afghanistan created splits in the Afghan-Arab community in Peshawar regarding the future role of Arabs in the region. The question of military participation with Afghan groups, and in what form, was a particular area of dispute. But the dispute went deeper than merely being a question of whether ‘jihad’ against Najibullah was warranted or not. This becomes clear when we examine the role of the Arabs in the battles around Jalalabad and Khost between 1989 and 1992. The accounts start with a general overview of the battles, followed by a description of the Arab participation in each battle and the debates this created in the Afghan-Arab community in Peshawar.

The Battle of Jalalabad As stated above, the Battle of Jalalabad was the first major attempt by the mujahideen to seize power in Kabul after the departure of the Soviet Army from Afghanistan. The offensive started on 6 March 1989 – a mere three weeks after the last Soviet troops had left Afghanistan – with mujahideen groups attacking from the east with tanks and covered by mortar fire. Sources have estimated that the mujahideen started out with some 5,000–7,000 men amassed outside Jalalabad. Other sources estimated the number of mujahideen to total 15,000.5 The Kabul regime, on their side, sent at least 15,000 men to Jalalabad, although some units reportedly deserted. Thus, it appears that numerically, the forces were not that different. However, Najibullah’s army had numerous advantages, the most obvious being their firepower and level of organization. The mujahideen managed to seize Samarkhel village east of Jalalabad, and pushed on towards Jalalabad airport, only two miles from the town. They temporarily seized the airport, but were forced to abandon it after a heavy counter-­attack from DRA (Democratic Republic of Afghanistan) forces. In the following days, the mujahideen tried attacking again from the north and south, but without success, mainly due to heavy enemy artillery and air support. Another obstacle faced by the mujahideen was the well-­fortified government positions and the minefields around the city. After this, the battle turned into a stalemate. In April 1989, DRA forces went on the offensive, and by the end of that month the mujahideen had been driven back to the outskirts of Jalalabad.6 By June, most of the mujahideen had pulled out of the battle. Sources estimated that mujahideen incurred some 3,000 dead

136

War and State-Building in Afghanistan

and injured – one of the largest single losses during the entire war.7 A number of reasons why the mujahideen were defeated at Jalalabad have been discussed in previous literature. They are summarized below. First, there was an obvious imbalance of forces, and the mujahideen were unable to take advantage of the element of surprise. The plan to attack Jalalabad had been known for some time, and the DRA had had plenty of time to put up fortifications and minefields around the city.8 While both sides reportedly had tanks (the mujahideen had a few T-55s previously captured from the Soviets), the DRA had heavy artillery, including SCUD missile systems operated by Soviets from Kabul, and air support. As the mujahideen had no effective air cover, the DRA’s air attacks caused the majority of the mujahideen’s casualties.9 Second, the mujahideen’s fighting units were badly coordinated, making it impossible to carry out coordinated attacks of any strength. While at least eight mujahideen groups were involved in the battle, they carried out their own ambushes and skirmishes, without coordinating with the other groups.10 One commander explained that due to the lack of coordination, ‘one group could be attacking, while the group on the other side was sleeping’.11 There were ISI officers among the mujahideen, but this was not sufficient to make the groups coordinate their actions with each other.12 Third, the political leadership of the mujahideen was deeply split and weakened. During the Battle of Jalalabad, Hekmatyar and Massoud’s forces clashed in Tokhar province in the north – indicating how little unity there was between the mujahideen parties at the time.13 One source reported that Massoud allowed government supplies pass through his area during the battle, because he disagreed with the ISI’s extensive support to Hekmatyar.14 Many other mujahideen commanders also did not support the battle, predicting that it would result in disastrous losses for their private armies.15 Fourth, the mujahideen lacked an effective logistics system, especially when it came to re-­supply of ammunition. According to the Pakistani Brigadier M. Yousaf, the mujahideen were only supplied with enough ammunition to fight intensively for a week – and their supply chain was not effective enough to last a long stalemate.16 Finally, the mujahideen gave the DRA forces no option to surrender. During the battle, the mujahideen brutally executed a number of Afghan Army soldiers. This was a bad strategic decision because it gave extra motivation to the DRA forces to continue fighting. As we shall see, the prisoner of war (POW) policies of the mujahideen were very different at Khost, which encouraged a number of DRA forces to surrender or voluntarily join the mujahideen.

Al-Qaeda versus Najibullah, 1989–92

137

The role of Arabs in the Battle of Jalalabad The Afghan-Arabs were divided on whether – and with what means – to participate in the Battle of Jalalabad in 1989. Even bin Laden’s newly established organization, Al-Qaeda, was split on the issue. The ‘general view’ of Al-Qaeda’s shura was to stay out of the battle, especially with ground forces. The battle was seen as ‘politically dubious’ (due to the involvement of the ISI) and the Arabs did not have the ‘adequate skills’ to contribute on the ground. They could perhaps support the Afghan mujahideen with artillery, but nothing more. In the end, bin Laden insisted that the Arabs should participate in the battle with guerrilla fighters.17 A group of several hundred Arabs, commanded by Osama bin Laden, entered the Battle of Jalalabad during the spring of 1989. Details of their tactical movements and contributions are scarce. We know that the Arabs operated as an independent fighting group under Arab command, rather than being integrated in Afghan units. The Arabs had skilful and experienced field commanders, including two Egyptians who had previously fought with Haqqani’s units in Khost. According to one eyewitness, the Arabs continued fighting ‘as the fronts became half-­empty’ when Afghan groups deserted due to heavy human losses.18 In the end, bin Laden ordered the Arabs to withdraw. The Arabs suffered several hundred casualties – the largest number of losses during one single battle in Afghanistan – mostly as a result of enemy air bombardments. Due to the many weaknesses of the mujahideen coalition who ran the battle, the Arab participation was doomed to fail. Jihadi news magazines in Peshawar were filled with glossy propaganda praising the brave Arab ‘martyrs’ who had died at Jalalabad.19 But behind the scenes, a critical leaflet was being circulated. It argued that the Arab fighters had died in vain due to the lack of proper leadership and coordination. It called for an immediate pullout of Arabs from the Jalalabad area. The Arabs should halt further operations until a proper organization had been established that could make sound decisions regarding when and how they should contribute to future military campaigns. Apparently, neither the Services Office nor Al-Qaeda was qualified to fulfil that role at the time. The leaflet did not mention bin Ladin directly, but was an indirect criticism of him as he had been the one who had pushed for the Arabs to enter the Jalalabad battlefield.20 The leafleat was written by Mustafa Hamid, an Egyptian journalist and jihadi fighter who was a good friend of Jalaluddin Haqqani. He would later become a leading figure among the Arabs who participated in the Khost battles in 1990–1.

138

War and State-Building in Afghanistan

In that battle, Al-Qaeda would play a much less prominent role than at Jalalabad. Before going on to describe the battles of Khost, let us look briefly at some of the political consequences of the mujahideen’s defeat at Jalalabad.

Consequences of the defeat ‘The Jihad has never recovered from Jalalabad’ wrote Mohammed Yousaf, head of the ISI from 1983 to 1987, in his memoirs published in 1992.21 Indeed, the Battle of Jalalabad was a watershed in the history of the mujahideen movement. Failure in this battle appeared to have brought a change of view among the mujahideen: that ‘a mujahideen victory began to seem improbable in the short run’. As a consequence, Najibullah’s ‘National Reconciliation’ process was strengthened, as many local mujahideen agreed to a cease-­fire with the government.22 This loss of motivation to fight is also signified by the fact that mujahideen military activity dropped considerably after the battle.23 The Afghan warlord Ismail Khan said, ‘The Battle of Jalalabad lost us the credit won in ten years of fighting.’ Consequently, the strategies of the mujahideen parties changed towards a coup d’état.24 There was indeed a coup attempt, carried out by then defence minister in the Najibullah Government, General Shahnawaz Tanai, in cooperation with Gulbuddin Hekmatyar, on 6 March 1990. However, the coup failed.25 In spite of the setback after Jalalabad, some commanders, like Jalalauddin Haqqani, continued to believe in military victory because they refused the National Reconciliation plan presented by Najibullah. In 1990, Haqqani initiated the forming of a National Commanders’ Shura (NCS), a council of senior mujahideen commanders to coordinate military efforts inside Afghanistan. It was indicative of the deep split that the Jalalabad battle had caused between the Seven-Party Union and the commanders inside Afghanistan. The commanders wanted to act independently of the political leaders in Peshawar, whom they saw as increasingly ineffective and corrupt. The NCS sought to turn the tide of the mujahideen movement and motivate commanders to continue the fight. As part of the effort, they issued propaganda claiming that Russia and the West had exaggerated the significance of the Battle of Jalalabad, in order to create an impression that the Afghan War could not be won militarily.26 But, there is little doubt – considering other sources – that the mujahideen wished to seize Jalalabad, and believed it would pave the way to the quick capture of the Afghan capital Kabul.

Al-Qaeda versus Najibullah, 1989–92

139

The first meetings to set up the NCS took place in Paktiya, 7–9 May and 22– 24 June 1990. A declaration issued after the last meeting stated, ‘the purpose for the formation of this shura [sic] is in no way aimed at creating a gap between the leadership cadre and the jihad commanders and both [these groups] are the real representatives of the people of Afghanistan and inseparable part of each other’.27 Naturally, the party leaders in Peshawar were suspicious of Haqqani’s intentions. Haqqani was accused of wanting to establish his own political party, which he strongly denied. A third shura meeting took place in Shah Salim at the border with Badakhshan between 9 and 13 October 1990. Some 1,000 commanders attended the third shura, including Ahmed Shah Massoud.28 The October meeting ‘passed resolutions against the ISI-sponsored strategy of a direct attack on Kabul by militia units based in Pakistan. Instead, it outlined a plan to capture provincial outposts of the regime and set up regional administrations (base areas) in nine zones.’ They declared that the commanders would take care of military affairs, and jihad leaders of political affairs. However, if the jihad leaders were unable to do so, the commanders would take care of political affairs too.29 Initially, the NCS was boycotted by Hekmatyar and Sayyaf, and the ISI attempted to dissuade people from attending it. However, both Hekmatyar and the ISI eventually came to support Haqqani’s military campaign against DRA forces at Khost in 1991. The campaign led to a much-­needed victory for the mujahideen.

The Battle of Khost The Battle of Khost took place in the second half of March 1991. Unlike the Battle of Jalalabad, which dragged on for several months, the mujahideen seized the city of Khost in just two weeks. However, one should keep in mind that the geography of the two battles was fundamentally different. Jalalabad is situated in the plains, on the main highway from Kabul to the Khyber Pass border crossing with Pakistan. This made Jalalabad an important strategic location for the Kabul Government. Khost, on the other hand, is situated close to the Pakistani border in the south-­east and isolated from the main highways in the country. ‘This was a town which had been completely surrounded since the beginning of the war and was not of great strategic significance.’30 This made it harder for reinforcements and supplies to reach the city. As long as the Soviet Army was stationed in the country, it was ‘possible to organize rescue operations in order to resupply it and break the siege’, as happened at the end of 1987. But ‘once the Red Army was

140

War and State-Building in Afghanistan

gone, Khost became off limits for Government supply columns’, and could only rely on air supplies.31 The rugged terrain of Khost province provided the mujahideen a great advantage, allowing them to rely on their well-­known guerrilla tactics. The mujahideen in Khost had been fighting Soviet and Afghan forces for more than ten years, and the former had already built a network of underground tunnels, caves and trenches that provided crucial shelter for artillery and air strikes from Kabul. During 1990, the mujahideen carried out several advances that were important prerequisites for the final success at Khost. In early 1990 they captured Torkhar, a mountain post south of Khost city. The mountain offered a free view to Khost airport. After capturing the mountain, the mujahideen were able to intensify their attacks against cargo planes attempting to land at the airport to supply the DRA forces in Khost. As we shall see later, Arab fighters played an important role in this effort. On 14 March 1991, mujahideen began their final assault on Khost. The perimeter around the city was attacked from at least three different sides with tanks, rockets and heavy artillery, including the Soviet rocket system Sakr-20 (BM-21).32 One of the mujahideen tactics was to sneak up close to enemy positions during the night and then shell them with rockets until they abandoned their position. They also attacked directly with tank fire.33 After capturing DRA Army outposts on the perimeter, and gaining control over the airport, mujahideen advanced towards the city itself. The goal was eventually to capture the DRA’s main army garrisons at Takhta Beg and Matoon, which were manned by some 3,000–4,000 government troops. It is unknown how many mujahideen participated in the assault. One source said that 1,500 mujahideen participated on the ‘south-­western front’. So, taking into account that there was at least two other ‘fronts’ (south-­eastern and north-­western), it is reasonable to assume that the total number of mujahideen was on a par with the number of DRA troops. Khost fell on 31 March 1991 after the DRA troops in the city surrendered to the mujahideen.34 The main reason for the surrender, according to one captured Afghan general, was lack of supplies and morale among the troops.35 By that time, the mujahideen had also become known for treating captives well. For example, in February 1990, a group of DRA soldiers were captured during the conquest of the Torghar Mountain. After being interrogated by Haqqani’s forces, they were released and each one ‘was given 200 Pakistani rupees to help him go back to his village’.36 The lack of supplies was probably one of the key reasons why Khost fell so easily.37 The fall of the city was indeed due to a mass surrender of DRA troops;

Al-Qaeda versus Najibullah, 1989–92

141

the mujahideen claimed to have taken more than 2,000 prisoners of war. As the mujahideen controlled the roads, supplies could only be flown in by air. Due to the mujahideen’s anti-­aircraft capabilities,38 the supplies had to be parachute-­ dropped from high altitudes; and at least once, the supplies fell into the hands of the mujahideen, not the DRA.39 A captured Afghan general said that his garrison had run out of supplies two days prior to the surrender. Another reason for the mujahideen’s success was their access to cover and concealment, which exposed them less to air and missile attacks. It is unclear whether the artillery and air attack capability of the DRA at Khost matched that at Jalalabad. But, according to mujahideen sources, the mujahideen were bombarded by both SCUD missiles and targeted by airplanes during their advance towards Khost. The bombs and missiles were described as inaccurate, rarely hitting their target. One commander exclaimed, ‘they [the Scud missiles] cannot harm us in any way. They smash to pieces on colliding with the mountains.’40 Mujahideen propaganda tends to explain these misses as a result of incompetence or – perhaps equally common – divine intervention.41 A more accurate explanation is probably that the geography of Khost, and pre-­made trenches and caves, gave the mujahideen good opportunities at cover and concealment. Moreover, mujahideen tactics were based on besieging the city for a long time and taking few risks with regards to being exposed; for example, they preferred to carry out attacks in cloudy weather, when air support was impossible.42 According to one estimate, the Battle of Khost only led to 120 killed in total, 10 of whom were Arabs.43 It has been claimed that bad weather made it impossible for the DRA to launch airstrikes on the mujahideen during their final campaign. However, this is contradicted by the eyewitness account of Mufti Usmani, which specifies that the bad weather came only after Khost had been conquered. While it prevented Kabul from bombing the city and destroying the heavy military equipment that had been captured by the mujahideen, it does not seem to have prevented the DRA from bombing during the mujahideen’s advance.44 A third reason for the mujahideen’s success at Khost – often emphasized by the mujahideen themselves – was their high degree of unity and cooperation on the ground. This was a factor that had been greatly lacking in the Battle of Jalalabad.45 The Battle of Khost was initiated and organized by the NCS, which had been created in 1990. The NCS was led by Jalalauddin Haqqani, a highly respected commander and a native to the area of Khost. The offensive had been thorougly planned. The areas around Khost were divided into different ‘fronts’ and each commander was assigned a specific sector. At least seven different

142

War and State-Building in Afghanistan

commanders led the battle on the ground.46 The ISI was also said to have officers on the ground, although the mujahideen officially denied this.47 Arab accounts say that ISI officers were present from time to time to observe the battle, but that they did not interfere in tactical decisions.48 The battle was coordinated by a ‘High Council’, led by Jalalauddin Haqqani, and with representatives from all parties in the Seven-Party Union. Hekmatyar and Sayyaf, who had initially boycotted the NCS, had eventually decided to join the battle. The mujahideen groups were able to communicate and coordinate on wireless radios, allegedly supplied by the US. Mufti Usmani confirmed in his account that the mujahideen ‘were in contact with each other over the wireless’ and this enabled them to attack on several fronts at the same time.49

The role of Arabs in the Battle of Khost By the end of 1990, the Afghan-Arab movement in Peshawar had seen some drastic changes. The controversies surrounding the Arabs’ participation in the Battle of Jalalabad had created deep splits in the movement. The coherence of the movement was further damaged when Abdullah Azzam, the chief ideologue of the Afghan-Arabs, was assassinated in December 1989. Bin Laden left the scene and went to Saudi Arabia, leaving a fragmented movement in Peshawar with no clear leading figure. After the fiasco at Jalalabad, bin Laden decided to disengage from the Afghan conflict, and used Afghanistan mainly for training cadre for his international organization. He initially expressed a desire to fight against the Communist regime in Yemen, but this plan was derailed in 1990 when Saddam Hussein invaded Kuwait and threatened Saudi Arabia. At this point, bin Laden offered to raise an army of mujahideen fighters to defend the Kingdom. However, the Saudi king turned down the offer, and decided to invite American troops to protect the country instead. It is a well-­known fact that this infuriated bin Laden. The event was a watershed in the history of Al-Qaeda, as it marked the beginning of bin Laden’s anti-Saudi and anti-American campaign. The nature of the Arab participation in Khost in 1990–1 must be understood in light of this context. Overall, the nature of Arab participation was very different in Khost than in Jalalabad. While the Arabs at Jalalabad had been under the command of bin Laden, the Khost operations were led by independent Arabs such as the Jordanian Abu al-Harith and the Egyptian Mustafa Hamid. Both of them had fought in the region for years, and had an extensive network of local

Al-Qaeda versus Najibullah, 1989–92

143

contacts in the area. Although Al-Qaeda had a presence in Khost mainly because it was running training camps for its own recruits, it did support the other Arab groups at Khost with manpower and logistics. The nature of the cooperation between Al-Qaeda and the other Arabs at Khost is well illustrated through the so-­called ‘Airport Project’, an operation that was carried out in the autumn of 1990. It was led by Mustafa Hamid, and aimed to deter enemy planes from landing at Khost Airport by bombarding them from positions on the Torkhar Mountain. Al-Qaeda had promised to deliver ammunition to the project, but failed to do so. Mustafa Hamid explains: The issue of ammunition was very important to me during the operation. The way the administrators in Al-Qaeda acted did not give me reason to have any confidence in them. During the long period we spent preparing for the operation they only supplied us with a third of the weapons that were needed . . . I had plenty of others from whom I could obtain the required weapons and men, especially Haqqani and al-Harith . . . Had it not been for their help I would not have been able to complete the project.50

By the end of 1990, Al-Qaeda had distanced itself from the Arabs involved in Khost, partly because they saw the Arab participation on the Khost battlefield as futile,51 but also because bin Laden had other priorities. In August 1990, bin Laden ordered Al-Qaeda’s personnel in Khost to return to Saudi Arabia to prepare for a possible war against Saddam Hussein. The decision damaged the ‘Airport Project’ because there was already a shortage of personnel in Mustafa Hamid’s ranks. Hamid was astonished that bin Laden had decided to pull out Al-Qaeda’s field commanders, but to leave the trainers, so that Al-Qaeda’s training camps could continue running. In Mustafa Hamid’s view, the trainees in these camps were nothing but ‘tourists’ from Saudi Arabia and Yemen. It clearly illustrates the different priorities of Al-Qaeda and the other Arabs in Afghanistan at the time.

Conclusion After the Soviet withdrawal from Afghanistan in 1989, the Arabs were split on whether they should continue to play a military role in Afghanistan. The original justification for the Arabs to participate in the conflict had been the presence of a ‘foreign occupation force’ on Muslim soil. When the Soviets departed, it was natural for the Arabs to question their future participation in the conflict.

144

War and State-Building in Afghanistan

However, this chapter has argued that the participation of Arab fighters in Afghanistan after 1989 was determined by local group dynamics, rather than ideology. It is well illustrated by the role of Al-Qaeda in Afghanistan. Bin Laden decided to fight the Najibullah regime in Jalalabad in 1989, but refused to send his fighters to fight the same regime in Khost in 1991. It was clearly a strategic descision on bin Laden’s part, not an ideological one. In a similar fashion, Mustafa Hamid and other ‘independent’ Afghan-Arabs refused to participate in the battle of Jalalabad, yet they gave their full support to Haqqani’s forces in Khost a year later. The continued participation of foreign fighters in a conflict cannot simply be explained by referring to ideological reasons, or the presence of a common, external enemy. This must be kept in mind when analysing the future development of Al-Qaeda in Afghanistan and Pakistan.

Acknowledgement A draft of this chapter was presented at the conference, ‘COIN in Afghanistan: From Mughals to the Americans’, Peace Research Institute Oslo (PRIO), 12–13 February 2012. The author would like to thank the conference participants for useful feedback on the draft.

Notes 1 Bruce Riedel, ‘Comparing the US and Soviet Experiences in Afghanistan’, CTC Sentinel, vol. 2, no. 5 (May 2009), pp. 1–4. 2 David C. Isby, ‘Four Battles in Afghanistan’, Soldier of Fortune (April 1988), pp. 32–4. For another account of the Battle of Jaji, and especially events around Al-Ma’sada, see Lawrence Wright, The Looming Tower: Al-Qaeda and the Road to 9/11 (New York: Vintage Books, 2007), pp. 116–20. Wright adds interesting details to the account. In the first phase of the Soviet attack, Al-Ma’sada was evacuated and partly destroyed, and the Arabs retreated to Sayyaf ’s Jaji base. However, as Al-Ma’sada was situated in a strategic position, Sayyaf told the Arabs to return to the base, supported by Sayyaf ’s own men. The rest of the battle ‘was actually waged more by Sayyaf . . . than bin Laden’. The Arabs nevertheless gained a reputation after the battle. 3 Mustafa Hamid, salib fi sama’ qandahar, Book 6 of tharthara fi saqaf al-­alam, pp. 21–2. Downloaded from MAFA (Mustafa Hamid’s Blog), 12 October 2009 [http://mafa.maktoobblog.com].

Al-Qaeda versus Najibullah, 1989–92

145

4 Barnett R. Rubin, The Fragmentation of Afghanistan (New Haven, CT: Yale University Press, 2002), p. 250; Olivier Roy, Islam and Resistance in Afghanistan (Cambridge: Cambridge University Press, 1990), p. 211. 5 Giles Dorronsoro, Revolution Unending: Afghanistan, 1979 to the Present (London: Hurst, 2005), p. 228; Brigadier Mohammad Yousaf and Mark Adkin, The Bear Trap: Afghanistan’s Untold Story (London: Leo Cooper, 1992), pp. 227, 230–1. 6 Dorronsoro, Revolution Unending, p. 228; Stephen Tanner, Afghanistan: A Military History from Alexander the Great to the Fall of the Taliban (Cambridge, MA: Da Capo Press, 2002), pp. 271–2. 7 Dorronsoro, Revolution Unending, p. 228; Yousaf and Adkin, The Bear Trap, p. 231. 8 Yousaf and Adkin, The Bear Trap, pp. 229–30; Tanner, Afghanistan: A Military History, p. 271. 9 ‘Fighting Continues All Over Afghanistan: Two Coups Against Najib’, Afghan Jehad, Quarterly Magazine of the Cultural Council of the Afghanistan Resistance, vol. 3, no. 2 (January/March 1990), p. 16; Yousaf and Adkin, The Bear Trap, p. 228. 10 Tanner, Afghanistan: A Military History, pp. 271–2. 11 Yousaf and Adkin, The Bear Trap, p. 228. 12 Dorronsoro, Revolution Unending, p. 228. 13 Yousaf and Adkin, The Bear Trap, p. 231. 14 Dorronsoro, Revolution Unending, p. 228. 15 Rubin, Fragmentation, p. 250; Robert D. Kaplan, Soldiers of God: With Islamic Warriors in Afghanistan and Pakistan (New York: Vintage Books, 2001), p. 166. 16 Yousaf and Adkin, The Bear Trap, pp. 229–31. 17 Yousaf and Adkin, The Bear Trap, p. 62. 18 Mustafa Hamid, al-­hamaqa al-­kubra, aw al-­harb al-­ma’iz, Book 4 of tharthara fi saqaf al- alam. Downloaded from MAFA (Mustafa Hamid’s Blog), 12 October 2009 [http://mafa.maktoobblog.com], p. 32. 19 See, for example, Al-Jihad, Year 7, no. 77 (April/May 1991), pp. 46–7. It is a print journal issued by Maktab al-Khidmat in Peshawar. 20 Quoted in Hamid, al-­hamaqa al-­kubra, p. 77. 21 Yousaf and Adkin, The Bear Trap, p. 220. 22 Dorronsoro, Revolution Unending, pp. 201–2. 23 Dorronsoro, Revolution Unending, pp. 185, 187, 229. 24 Dorronsoro, Revolution Unending, pp. 228–9. 25 Tanai started the rebellion because after the failure of the Battle of Jalalabad, the US and Pakistan agreed to accept Najibullah to be part of the interim/transition government. Mian Faqir Ahmad, ‘What Difference did the Kabul Uprising Make?’, Mujahideen, Year 4, no. 2 (April/May 1990), p. 15. It is a print journal issued by Hezb-­i-Islami Afghanistan in Peshawar.

146

War and State-Building in Afghanistan

26 ‘Facts about Jalalabad’, Khuddam ul-Furqan (December 1989), pp. 12–13. It is a print journal issued by Harkate-Inqilabe-Islami Afghanistan in Peshawar. 27 ‘Commanders Call for Another Nation Wide Meeting’, Afghan Jehad, vol. 3, no. 3 (April/June 1990), pp. 30–1. 28 For more details, see Afghan Jehad, vol. 3, no. 3 (April/June 1990), pp. 19, 34–5. 29 Ibid. 30 Dorronsoro, Revolution Unending, p. 229. 31 Dorronsoro, Revolution Unending, p. 103. 32 Mujahideen (October/November 1991), p. 10. 33 Mufti Mohammad Rafi Usmani, Jihad in Afghanistan Against Communism (Karachi: Darul-Ishaat, 1992), pp. 294–305. 34 ‘How Khost was Liberated? Special Afghan Jehad Report,’ Afghan Jehad, vol. 4, no. 2 (January/March 1991), pp. 18–20. On numbers, see also Dorronsoro, Revolution Unending, p. 80. 35 Usmani, Jihad in Afghanistan, p. 334. 36 Mustafa Hamid, al-­matar 90, Book 5 of tharthara fi saqaf al-­alam, p. 56. Downloaded from MAFA (Mustafa Hamid’s Blog), 12 October 2009 [http://mafa.maktoobblog. com], p. 80. 37 As described, for example, in Dorronsoro, Revolution Unending, p. 219. 38 Usmani, Jihad in Afghanistan, pp. 316–17. 39 Usmani, Jihad in Afghanistan, p. 311. 40 Usmani, Jihad in Afghanistan, p. 301. 41 Usmani, Jihad in Afghanistan, pp. 305–10. 42 Usmani, Jihad in Afghanistan, pp. 300–1. 43 Hamid, al-­hamaqa al-­kubra, p. 91. 44 Usmani, Jihad in Afghanistan, p. 332. 45 See, for example, ‘How Khost was Liberated? Special Afghan Jehad Report’, Afghan Jehad, vol. 4, no. 2 (January/March 1991), pp. 18–20. It says that ‘the main factor which led to the liberation of Khost was the alliance and solidarity among the mujahideen groups and their joint command’. 46 Usmani, Jihad in Afghanistan, pp. 294–5. 47 Ibid. 48 Usmani, Jihad in Afghanistan, p. 335; Hamid, al-­matar 90, p. 26. 49 Usmani, Jihad in Afghanistan, pp. 303, 315–16. 50 Mustafa Hamid, ahzan al-­umm al-­kabira (A Mother’s Deep Sorrow), The Harmony Programme Database at CTC West Point, Doc. No. AFGP-2002-600092, p. 22 [http://www.ctc.usma.edu/wp-­content/uploads/2010/08/AFGP-2002-600092Orig-Meta.pdf]. 51 Hamid, ahzan al-­umm al-­kabira, p. 24.

6

The Afghan National Army and COIN: Past, Present and Future Reconsidered Rob Johnson

Introduction There are several distinct Western assumptions that underpin American and European approaches to countering irregular resistance. From constraints on the use of force, manifest in particular ‘rules of engagement’, to limits on the handling of civilians and population control measures, Western practitioners of COIN are quick to identify the need to conduct a ‘population-­centric’ campaign.1 Western forces use a particular vocabulary to describe their operations, but differentiate their values and methods from local actors, sometimes acknowledging that their approach is the result of particular historical experiences, but also normative Western political standards. Western military personnel thus emphasize the importance of understanding an indigenous ‘culture’ that is quite different from their own.2 American and British doctrines, for example, stress the importance of ‘influence’ operations, the winning of civilian sympathy (hearts) and rational choice (minds), or isolating and neutralizing the insurgent. These relative benign terms mask the business of forcing portions of a civilian population to yield to force, imposing limits on their freedom of action, and obfuscates the rather irksome fact of the existence of foreign armies in the sovereign state of Afghanistan, which is one the primary drivers of local resistance.3

The Afghan National Army at present and COIN There is greater consciousness, indeed, some unease about less proscribed Afghan approaches to countering insurgency, including the willingness to use

148

War and State-Building in Afghanistan

excessive force, the more robust handling of prisoners and support for vested interests such as the relative power positions of regional warlords.4 There has been great concern that Afghan troops, particularly those from regions of Afghanistan that have little sympathy for the people of the south and east, may be more coercive towards the population and undo the attempts of the International Security Assistance Force (ISAF) to win hearts and minds through development work and the ‘courageous restraint’ of its own troops. Others have expressed doubts that the Afghan Army would remain cohesive in the face of insurgency after the change of posture of Western forces, from combat to training, after 2014. The various ethnic components of the army may, it is thought, simply break up or perhaps attempt to seize power in the regions of the country, returning Afghanistan to a condition of fiefdoms and weak central authority that characterized the early nineteenth century. There is little doubt that Western forces have possessed the military capability and skills required to overmatch their insurgent adversaries at a tactical level. An array of technological advantages, from surveillance equipment to attack helicopters, has enabled ISAF to defeat small insurgent teams whenever they become visible. Less overt forms of resistance, such as improvised explosive devices (IEDs), proved more difficult to combat but detection equipment and better drills lessened their impact considerably. Only the suicide IED delivered in a clandestine manner by an individual (SIED) and single round sharpshooter attack or sniping (SRSA) defied ISAF counter-­measures. Nevertheless, for strategic success, the West needed to build institutions that were capable of preventing the return of international jihadist organizations. The only durable institution constructed by 2012 was the Afghan National Army (ANA). Much depended on its ability to provide security with only a minimal degree of external support. However, the Afghan Government’s national revenue has been dwarfed by the cost of sustaining a large army.5 The Western strategy envisaged a steady drawing down in ANA numbers after 2014 as the combined effect of three years of American ‘surge’ troops and a further three years of an enlarged ANSF suppressed and exhausted the insurgents. However, by 2012, there were only tentative signs that this linear evolution was being realized. Indeed, many insurgent groups appeared as defiant as ever, arguing that they needed only wait until the withdrawal of ISAF before launching a more significant offensive against the ANA and the Afghan Government. The surge of American troops announced in 2009 was designed to provide a screen behind which the ANA could be built up rapidly. Conscious of the problem of developing Afghan leadership and professionalism according to

The Afghan National Army and COIN

149

Western standards, the risk of fuelling rivalries between ethnic groups, and of low levels of literacy, extensive national drug use and engrained irregular modes of waging war, the Americans knew that there were significant obstacles responsible for holding back development.6 Ali A. Jalali, the former Afghan Minister of the Interior, believed that the planned increase of the ANA required ‘the mobilization of resources that are unlikely to occur within the announced timeframe’ and he listed severe constraints in terms of pay and benefit costs, ethnic divisions, Taliban infiltration and corruption issues.7 The evolution and expansion of the ANA was nevertheless designed to allow the surge forces, and the rest of the ISAF contingent, to begin a staged withdrawal that would be complete by 2014.8 In August 2009, General Stanley McChrystal, the Commander of the UN mandated ISAF, advocated a system of ‘Embedded Partnering’ for the ANA.9 He envisioned a ‘trust-­based, habitual and enduring relationship’ that was undoubtedly modelled on the recent American experience in Iraq. The aspiration was to ensure joint command at every level, and he sought to eradicate the differences of expectation and culture that had characterized previous operations. Nevertheless, the emphasis was on training the ANA such that it could lead operations on its own and sustain itself indefinitely without ISAF input.10 This was to be achieved in stages. Among the priorities were the capability of the ANA, its effectiveness in combat, its scale, leadership development, and retention through improvements in the terms and conditions of service (TACOS). Embedded ISAF personnel, known as Operational Mentoring and Liaison Teams (OMLTs) or Military Training Teams (MTTs), initially assisted the ANA by operating alongside and within small Afghan units. These were gradually organized into larger Afghan units and formations, while the embedded ISAF teams drew back into Brigade Advisory Teams (BATs) or provided ‘overwatch’ for Afghan kandaks that had already proven their ability to operate independently.11 The experience of partnering and supporting Afghan units proved to be complex. Up to four British battlegroups, a significant portion of its entire deployed force in Afghanistan, were dedicated to this role, with a further battalion sized team allocated to the Afghan National (uniformed) Police (ANP). The OMLT teams, some of them only four men strong, found themselves exposed and forced to conduct operations practically singled-­handed when inexperienced and reluctant Afghan troops failed to engage in the routines of patrolling or small skirmishes with insurgents. By 2011, the situation had improved significantly, with larger numbers of Afghan troops available and their competence much developed. In February 2012, General Seren Shah and the

150

War and State-Building in Afghanistan

ANA conducted their own COIN operations in central Helmand with the British in a supporting role, although the winter time, traditionally a far quieter period, was selected deliberately to test the Afghan troops in a relatively benign and low-­ tempo setting. Nevertheless, the aspiration to have ISAF and ANA fighting shoma ba shoma (shoulder to shoulder) as equal partners seemed to have been achieved, at least in principle. The ANA and ISAF units in Helmand were co-­ located in a ‘rule of three’ (one ANA unit, with one ANP and one ISAF equivalents). The aim was to conduct joint training, share intelligence and deploy on operations together. In these operations, ISAF had to ensure that it passed command responsibility to Afghan officers and above that, to the Afghan Government itself. In order to create the idea of restoring Afghan sovereignty, it was imperative to ensure the ‘host nation’ could determine the timing, location, and planning of operations. This was not without difficulty. President Hamid Karzai had to be pressured and requested three times to give his approval for Operation Moshtarak in Helmand in 2009.12 His anxiety about civilian losses and therefore the erosion of his base of support in Pashtun areas was considerable. Part of the transfer of command involved clarification in the distinctions between mentoring and advising. Advice while embedded as an OMLT was often robust, but mentoring required a more subtle approach of developing potential, encouragement, guidance and management. British officers found that developing a relationship or rapport took time, but the six-­monthly rotation of brigades precluded a deeper mutual understanding. One interviewee, conscious that the local units were just getting to know him before he departed again, remarked that he felt embarrassed to be an Afghan officer’s ‘next best friend’.13 Like many of their ISAF comrades, Pashto and Dari language skills were considered to be deficient among British personnel, but even senior officers admitted that sending their soldiers and marines on extended linguistic training would be detrimental to an officer’s long-­term career. Interpersonal skills were thought to be more important and were regarded as integral to every British officer who had attended the Royal Military Academy Sandhurst, but cultural training as part of the preparatory component for deployment to Helmand was limited and tended to be reduced to simplistic stereotypes. Afghan officers expressed frustration that British personnel dwelt on pashtunwali, the code of the Pashtun, at the expense of more practical issues, such as rivalry over shares of the narcotics business. The British Army regarded these concerns about drugs and land as peripheral to their core work of providing security. Several of the shortcomings of the Afghan Army were the product of its history.

The Afghan National Army and COIN

151

The early history of the Afghan Army Like the Military Revolution in the West, state formation in Afghanistan was dependent on the army, but also driven by the need to furnish its costs. The Durrani Empire was built on the successes of its army not its administration, and it took some considerable time, and much foreign influence, before a functioning state bureaucracy emerged at Kabul. Ahmad Shah Durrani, the founder of the Pashtun dynasty in the late eighteenth century, admitted to his treasurer in 1751 that his regime was, in effect, a military state that depended on its warriors: ‘it is by their unanimity, and with the help of their swords, that I have been raised to this position’ and their importance meant: ‘I should certainly look upon my soldiers as partners in this wealth.’14 This was a nation in arms. Practising a form of military-­fiscalism, the imperial coffers had to be replenished with raids on the Mughal territories to the east. The Afghan rank-­and-file were enthusiastic about these attacks since they offered large personal profits through looting, the chance to exhibit their courage and gain promotion, and the legitimacy of waging ‘Holy War’ against Hindus and Sikhs. Indeed, the anatomy of the Afghan Army in this period underscores the character of the regime and its methods of war. This was not a Sipahi i-­munazzam (regular army), but a Sipahi ghair-­i-munazzam (irregular force). The core of the troops under Ahmad Shah Durrani’s control numbered 16,000 feudal cavalry, known as the sawara-­i kushada or sawara-­i khudaspa, but this was expanded when he took power to 40,000. The army was composed of Durrani and Ghilzai Pashtuns with a bodyguard of Kizilbash (‘redhead’ Turkic mercenary cavalry), augmented on campaign with more irregulars to reach a maximum strength of 120,000. The Durrani cavalrymen were supplied by the clan elders in return for grants of rent-­free land. If their land produced insufficient revenue, they could be granted barats (stipends) from other sources, such as transit duties. Additional cash allowances were received if they increased the number of horsemen in Ahmad Shah’s service. Because of their association with the ‘royal’ clan, the Durrani elders assumed the mantle of a noblesse oblige, and, like the baronial feudal order of Europe in the thirteenth century, they took the view that they had the right to govern in the name of their selected monarch. If that monarch failed to serve their interests, they took it as a sine qua non that they could intervene to remove that head of state. Thus, from the outset the army was an important element in Afghan politics. Ahmad Shah’s capital at Kandahar was guarded by 12,559 cavalry, of whom 5,710 were Durrani loyalists.15 The rest were non-Durrani qawm from

152

War and State-Building in Afghanistan

neighbouring districts, such as the Ghilzai Hotaks, and the separate Kakars, Darwais and Barechis. When his son Timur Shah moved the capital to Kabul, additional cavalry were raised under the title Kara Nokars. Again, non-Durrani Pashtuns and ethnic groups like the Tajiks were strongly represented to achieve some balance and to prevent any one faction from controlling the army. In an effort to extend the monarch’s reach, in theory every landowner was supposed to provide soldiers, and pay them, in return for a remission of land taxes. In reality, only the region around Kabul ever submitted their full quota, and then only when an emergency arose. Remote areas either sent a token few men or none at all. Nevertheless, the Kara Nokars were formed into units under the command of their own maliks, and these were grouped under five Durrani sirdars.16 These men, like all those paid in some way by the Shah, were expected to act as civil administrators in times of peace, fusing the military and civil structures still further. While ensuring a single line of authority, the risk was that, in periods of tribal unrest or rebellion, the Shah was often unable to enforce their authority without bringing in troops from elsewhere and so relatives were selected for their loyalty, but also to keep them at arm’s reach from the capital. Logistics had always been the weakest component of Afghan forces. Supplies were stripped from the lands through which the force passed, but this imposed limits of campaigning seasons and the length of time the army could remain in the field. Again, the structure of the army was oriented towards the acquisition of plunder. A third of the ‘regular’ army was mounted, and a variety of light guns were drawn. But the forces were raised only for a particular campaign and the sipah-­i ghair-­i munazzam were a mixture of cavalry and dismounted warriors. Some accounts suggest that a majority of these irregulars were also mounted to suit their role as raiders, emphasizing mobility and mass over strength: corps consisting of thousands of horsemen could generally swarm around enemy garrisons and overwhelm them before they could be reinforced, or cascade through the countryside to acquire booty and then ride away before their enemies could concentrate against them. Each cavalryman was armed with a variety of weapons for the diverse military tasks he might have to perform on extended expeditions, including a shield, long spear, matchlock, sword, pistol and long knife. Officially, the army could be paid in cash or granted jagirs or lucrative appointments in Afghanistan, or in conquered parts of India. Irregular forces were raised not just by asking for levies of men from chieftains in return for an annual tax remission, but through the appeal of loot (that would be shared by the victorious troops). Irregulars might also be granted the same jagirs and cash

The Afghan National Army and COIN

153

rewards enjoyed by the standing army. On the borders, irregular light cavalry were maintained more or less permanently to patrol the frontier, impose taxation (from which they took a share for themselves) and to pursue enemies of the Shah, but officially they were paid in the same way as other tribal levies. In an attempt to garrison the troublesome north, a permanent military colony was established at Balkh and here too the Kohna Nokars, as the northern cavalry were known, were paid by assigning them land which they could pass on to their sons and heirs.17 Elsewhere, fortresses were supposed to control entry points into the Afghan realm. At Herat, the earthen walls were many feet thick and provided with a variety of cannons; and the citadel, the Maruchaq, occupied an area that was equivalent to two-­thirds of the size of the city itself. The fort of Dehdadi in Mazar-­i Sharif was also notable for its size and strength, and the Bala Hissar of Kabul was famous throughout Central Asia.18 The relatively lawless nature of the rural areas meant that almost every village community and family house resembled a small fortress, known as the qala (tower or compound). Compound walls not only concealed women and valuables, but also kept at bay rival families or clans. Towers were built at the corners of compounds for observation and for shooting platforms and every family was armed. The thick vegetation of cultivated terraces or ditch-­lined fields could also provide cover. However, it was the mountainous nature of much of Afghanistan that rendered the country a natural fortress. High summer temperatures, freezing winter conditions, long routes without abundant pasture or water, high altitudes and narrow passes and defiles closed by winter snows all meant that even relatively poorly armed Afghans could harass a regular army even if they could not halt its progress. Like the desert raider, safety lay in the environment and one’s ability to raid and slip away, rather than in the strength of arms. Nevertheless, the precarious nature of Afghanistan’s food supply and infrastructure meant that invaders practising a policy of scorched earth could defeat Afghan resistance, even if they could not sustain a large army of occupation in the country for a long time. There were three significant problems, though, that affected Afghan military forces specifically. First, as noted above, they lacked any secure financial– logistical system, so that if the amirs or chieftains failed to pay their men, or there were sustained periods without the rewards of loot, there were mutinies. Second, the frequency of raids fell hardest on the Sikhs of the Punjab, and they reorganized their forces in such a manner that resistance against the Afghans stiffened. Third, their loyalty was often in doubt. Afghan forces were raised by local chieftains and communities, and their first loyalty was to them and not the

154

War and State-Building in Afghanistan

Durrani rulers. The governors of regional cities were also aware that loyalty could be purchased. Durrani Shahs faced the dilemma of staffing governorships with weak men who might be unable to rule effectively, or stronger ambitious relatives who might be tempted to rebel. The loyalty of the various armed groups was thus the deciding factor in the Afghan political economy. To offset this problem of disloyalty, Ahmad Shah had established another independent military force, the ghulam-­i shah. This bodyguard, based on the Persian shah-­sevan, excluded Durranis from its ranks and recruited Kizilbash. Troopers were enlisted for life and paid in cash, their units commanded by respected officers, or qular aghasis. Yet, for an army made of feudal levies without professionalized systems such as regular terms of service, the Afghan Army’s strength often ebbed and flowed. Although it was traditional to try to garrison regions with troops drawn from other parts of the country, to avoid sympathies and fraternization developing, in practice this was far harder to achieve. Soldiers would simply vote with their feet if conditions did not suit them. Officers might use greater coercion to prevent desertion, but irregular pay or simply a desire to return to one’s home could not prevent the army from fading in strength. Far more serious was the problem of local insurrection, and Afghan troops drawn from the same communities as those in revolt were not reliable. Ghilzais, who had traditionally made up the majority of the Afghan forces, were often in the forefront of resistance to the Shah. Although Ahmad Shah, like those who followed, tried to adopt Persian practices, such as the reading of statements of fealty to the monarch at the Khutba (Friday prayers), the Ghilzai elders were aware that the Shah did not control their districts. Financial rewards might induce cooperation, but they did not ensure loyalty. Ghilzai clans were also conscious that they could exact tolls on caravans that made their way from Kabul and Kandahar or across the Suleiman Range into the subcontinent, and this financial independence reinforced the sense that they were politically autonomous too. A Pashtun proverb concludes: ‘You can hire an Afghan, but you can never buy him.’ Given the proximity of the rich plains of the Punjab, many of the Afghan eljaris were drawn from Ghilzai clans, and their forces were a mix of clan groups organized in tens, fifties, hundreds and thousand-­strong formations under leaders correspondingly styled as dahbashi, pinjabashi, sadbashi (yuzbashi) and hazarbashi (mingbashi). Governors of the provinces were given overall jurisdiction, but real authority rested with the elders who selected them and paid them.19 Real money, as noted before, came from looting, and eljaris were permitted to retain four-­fifths of what they stole, the rest being handed over to

The Afghan National Army and COIN

155

the Shah, unless they could get away with keeping more. It was an expectation, long engrained, that enduring the risks of military service entitled one to be rewarded with the spoils of war. Exhibitions of personal courage were deeply engrained in the Pashtun code, and were common to other Afghan minorities. It was considered a means to acquire greater honour and prestige by performing acts of daring, either against rival clans or in wartime against an enemy. There was much to be gained therefore by being at the front, and no honour at all in serving as some logistician or victualler. Equally, accepting orders from others was seen as a limitation on their opportunity to demonstrate their courage. Every man thought himself a khan, ‘har saray khan deh’, as every man thought himself the equal of anyone else. This had a number of unfortunate effects. In battle, many would try to get straight into the fight, regardless of the odds against them, and there would be little restraint or patience. Equally, once a line started to collapse, the whole army tended to fragment and melt away. In periods of stalemate, the absence of any logistical support meant that the army had to break up for foraging.20 For greater internal security, Ahmad Shah’s son, Timur, looked to develop the Kizilbash to provide the core of an army that would be loyal to him, through cash rewards, and not to Afghan clan affiliations. This attempt to break with tribal linkage was significant, and Timur, like his father, was clearly trying to establish a hereditary monarchy that could bind all Afghans under a single ruler in emulation of the Persian or the Mughal empires. Yet, the Pashtun clans of the south distrusted, collectively and universally, any attempt to assert one individual or family over all others. Pashtunwali emphasized egalitarianism, and, despite the creation of a majlis or advisory council of elders, Durrani and Ghilzai elders were not convinced that Timur was content to be a mere primus inter pares. Nevertheless, although he was forced to deploy the Kizilbash cavalry to break a series of Pashtun revolts, Timur knew he had to continue the policy of avoiding taxation and to foster support through consensus and a network of patronage. He also ensured that garrisons were maintained to prevent any large faction, like the Ghilzai, taking possession of Kandahar, which they might use as a base for operations and a recruiting ground to overthrow him. Most importantly, he tried to keep the Durranis, the closest qawm to the seat of power, divided. A British observer wrote: ‘the King’s policy is to keep the Douranees in subjection to himself, while he exalts them over other Afghauns’, continuing: ‘For this purpose he protects the Taujiks, and all others whose power he can use to depress the nobles, without endangering the ascendancy of his tribe.’21 In practice, this meant intervening in the internal politics of the Durranis, and keeping up the demands

156

War and State-Building in Afghanistan

for soldiers in return for grants of land, whereas, with non-Pashtun groups, Timur could afford to spend less time concerning himself with their affairs.22 In the nineteenth century, the Amir of Kabul, Dost Mohammad, used the coerciveness of the army to extract funds in the environs of Kabul. The Tajik Kohistanis, the Ghilzais of Laghman and Ghazni, and the Hazaras of Bamian were all forced to pay into the Amir’s coffers by soldiers-­turned-tax collectors. Formations of 1,000 cavalry were divided into detachments in each district, and these billeted themselves in houses and villages, eating everything, until the villagers paid the revenue required. The largesse was then ferried back to Kabul, turned into cash for the troopers or exchanged for goods, while a portion of collected grain was either stored or given to the soldiers’ families. Slaves too could be taken and sold for the Amir’s profit. Villagers and clans therefore took the view that their army was a predatory organization to be resisted if possible. The greatest threat to Afghan security, however, was foreign intervention. The Afghans managed to resist a Persian invasion at Herat in 1837 by sheltering behind the massive walls of the city and making use of the surrounding countryside for supplies, but they proved no match in the field against the British Army of the Indus just two years later. In open battle, when massing in greater numbers and when besieged at Ghazni, they were simply unable to overcome the disciplined British and Indian regulars. However, when the British occupation appeared to be permanent and vested financial interests were threatened, the defeated were stirred into rebellion. The trigger was less the presence of foreigners, but more the British attempts to change the system of feudal cavalry and thus the financial-­patronage network, which affected the Afghan elites collectively.23 In addition, the sudden British cut to allowances of the Ghilzai Pashtuns in the Khyber region, a revenue stream on which the impoverished local khans and khels depended, ignited an additional revolt along the British line of communication into India. Opportunism and fears of treachery fuelled the Kabul–Khyber uprising, and ensured its spread across the northern-­central provinces. The reduced British garrisons at Kohistan, Ghazni and Kabul were thus overwhelmed. The Afghan irregulars were still unable to defeat the British when they occupied strong fortified positions. The British force at Sherpur had held their walls successfully in 1841–2, while Afghan investments and storming parties had failed to carry the defences at Kandahar or Jelalabad. To make matters worse, the tenuous alliance between ethnic groups in Kabul in January 1842 broke up as soon as the British had been ejected, the primary source of dispute being the raising and control of the Afghan military forces. Most of the khans had

The Afghan National Army and COIN

157

exhausted the cash they had acquired in the early stages of the revolt, and many turned to looting Kabul and the surrounding area in order to raise money to pay for troops. A similar situation plagued Afghanistan in the early 1990s when the irregular mujahideen forces that emerged after the Soviet withdrawal effectively sacked the country. The restored Amir, Dost Mohammed, was aware that Afghanistan lacked an army that could ensure both internal security and hold off the British, a point driven home by the second invasion from India in 1842, which had stormed through the Khyber Pass despite attempts to hold it. The difficulty was that, without changes in the means to acquire more revenue, including the taxation structure, it was difficult to raise a larger or more professional force. Nevertheless, the legacy of the British occupation and its injection of cash into the local economy was that the amount of tax being collected from the realm in the vicinity of Kabul had risen from 2.5 million to 7 million rupees. This was sufficient to raise a large enough force to wage war against his rivals in the north. Acquisition of the revenues of the north then offered the chance to sustain operations against the southern provinces, and subsequently against Herat. Nevertheless, the Amir realized that he needed a professional army led by an officer corps loyal to him and his dynasty. Sher Ali, his successor, continued to consolidate the Mohammedzai grip on power, emphasizing the modernization of the Afghan Army as the means to defeat internal unrest. A cabal of Kandaharis attempted to overthrow Sher Ali under the leadership of Mohammed Amin Khan in 1865, but in a vindication of Dost Mohammed’s vision, Sher Ali’s trained and better equipped army ensured victory and thus his political survival.24

The Afghan Army of the 1860s and 1870s In Dost Mohammed’s second reign, regiments had been formed neither by conscription nor voluntarism, but by seizing quotas of men with threats to the family that they would either be imprisoned or ruined. Pay for foot soldiers, when it was paid at all, amounted to 5 rupees a month. Discipline was harsh with capital punishment or torture for minor infractions. Command was devolved to provinces and districts with no single chief of staff who might threaten the political order. For all its faults, this had produced a regular army that was, to quote Sir Bartle Frere, ‘quite equal in armament, skill, and drill to any corps in our service’.25 Each of the large cities was garrisoned and troops were held at strategically important locations ready to respond to any internal unrest or

158

War and State-Building in Afghanistan

foreign incursion.26 By 1879, Sher Ali had built up the army to a force of 56,173 troops, divided into 73 regiments of infantry, 42 regiments of cavalry and 48 batteries of artillery.27 All of these units were based on the British Indian Army model, but the bulk of the soldiers and the officer corps were drawn from a narrow ethnic band of Ghilzais and from Pashtuns of Wardak. Overall command of the army was given to Sipah Salar Husayn Ali Khan Kizilbash, and training advice was provided by two former Muslim Indian Army instructors, General Karim Bakhsh and Sirdar Ghulam Bahadur Naqshband.28 Although soldiers were expected to serve for life, their conditions of service were marginally better than in the past, since infantry soldiers’ pay was increased to 7 rupees a month, while troopers of the regular cavalry were given 14 rupees a month. But this was not a ‘quota army’: there were very few Durrani and Zirak Pashtuns or minorities, reflecting the perceived disloyalty of many factions in Afghanistan. The majority of fighting men were formed into local forces. Afghan qawm militia, especially in Ghilzai districts, believed they were ‘born soldiers, excellent skirmishers and experienced foragers’.29 However, they were ‘impatient to restraint’ and possessed no answer to artillery, believing that ‘if other nations were, like themselves, armed only with swords instead of guns and other sophisticated weapons, Afghanistan could conquer the world’. There was still a strong faith in the idea that gazis, the religious warriors of a determined frame of mind, could overwhelm greater numbers of infidels.30 The Amir knew he could not control these forces, but he also regarded them as a useful screen that could provide a buffer against the British. He warned the Government of India, when it complained about potential foreign influences in Afghanistan: ‘Before looking elsewhere, the English had better attend to their own house. Should they dare to move in this direction, they will first have the frontier tribes like hornets about their ears before they can ever attempt to cross the border.’31 The regular army consumed between 25 and 40 per cent of Afghanistan’s revenue in the 1870s, and the extra expenditure had required a more efficient system of taxation.32 Even with the growth and development of the army in the 1870s, Sher Ali feared it was no match for the modern forces of the British or the Russians for this was a decade of rapid technological change. The British had armed themselves with single-­shot breech-­loading rifles, rifled artillery and there were experiments with machine guns. The British possessed telegraph communications, as well as tinned rations to sustain themselves on campaign all year long, and the railway network in India would soon terminate on the Afghan frontier.33 In short, the Europeans were opening up a technology gap in warfare.34 The jezail was rendered obsolete by rifles with much greater range and accuracy.

The Afghan National Army and COIN

159

Furthermore, a company of disciplined European troops was able to lay down a volume of fire that could cut down charges by ghazis before they could close with the British in hand-­to-hand fighting. Howard Hensman, a journalist in Kabul in 1879, noted that, despite a mass attack involving thousands, only a single gazi reached the British defensive line at Sherpur, but he was shot down too.35 Success for the Afghans was dependent on the use of surprise or mass. In open battle, they were simply outmatched. There were also structural fissures that threatened to break the army. There was considerable rivalry between regiments recruited from different localities, and the infrequent and irregular nature of soldiers’ pay fostered desertion or indiscipline. Lists subsequently seized by the British in Kabul suggested that Sher Ali had wanted to introduce universal military service, but there was a lack of officers and trainers who could convert large numbers of conscripts into the personnel of a modern army. There were, in common with other Asian countries in this period, attempts to emulate the Western model, even to the extent that British uniforms were copied, but this outward change concealed the lack of real transformation.36 One British journalist wrote: ‘Shere Ali might be able to distribute Enfield and Snider rifles amongst his sepoys, fit out batteries with every kind of shot and shell, and teach his men such rudimentary discipline as would enable them to march in fairly good order; but he could not get beyond this.’37 There was no military education, no attachés, no visits to foreign armies and no military academies or staff courses. The British believed that the Afghan fighter was just unsuited to service in a regular army. One observer concluded: ‘The Afghan does not lack native courage, and in hill warfare he is unrivalled so long as it takes the shape of guerrilla fighting; but once he is asked to sink his identity and to become merely a unit in a battalion he loses all self-­confidence, and is apt to think more of getting away than of stubbornly holding his ground as he would have done with his friends, led by his own malik or chief.’38 There was a practical reason for this attitude. If the battle swung against him, it was far easier to retreat to one’s local community and blend in to ‘play the part of a peaceful peasant’, whereas regular troops could be pursued to destruction. Nevertheless, a number of Pashtuns along border districts were enlisted in British frontier formations and line regiments because of their perceived martial qualities as warriors. Pashtuns were especially admired for their ‘strong, self-­sufficient’ attributes, although the internecine feuding between khels and sections was a threat to unit cohesion. The British Indian Army tried to circumvent local rivalries by grouping personnel into specific frontier corps, separate ‘class’ companies and fostered regimental loyalty, while

160

War and State-Building in Afghanistan

Political Officers were despatched to work with the irregular khassadars along the frontier. But despite the imposition of European disciplinary measures, they never overcame the first loyalties of the Pashtuns they employed.39 Other than the development of shoulder arms technology, it was in the artillery that the most significant evolution was taking place in the Afghan army in this period. Unsurprisingly, given the technical expertise required, the topkhana was regarded as the elite arm of the Afghan forces. Traditional brass smooth-­bore guns of small calibre had been held in the various fortresses of Afghanistan, but the Amir was eager to replicate as many of the British rifled guns as he could and British-­made guns were introduced from 1869. Ironically, Sher Ali had benefited from sending observers into British arsenals in India, and the British permitted former Indian Army NCOs to assist in the training of the Afghan Army. However, despite the enthusiasm for armaments, the quality of the imitation was low. Gunpowder manufacturing was flawed, and, in the case of rifle cartridges, storage was badly managed resulting in rapid deterioration of stocks. Charges were invariably lower than British ones, which tended to diminish the range of Afghan weapons. Fuses for exploding shells were also of poor design, causing accidents and duds. However, these defects did not detract from the Afghans’ determination to develop their own armaments industry. The British underestimated the number of modern rifles the Afghans possessed, and were surprised in 1879 by the weight of fire that the Afghans could lay down. The biggest problem was the lack of training. Gunners were issued a single artillery round with which to test the barrel, but there was no ‘live-­firing’ practice. Consequently, artillerymen were trying to learn how to operate their guns for the first time when engaged in combat. On 8 October 1879, Afghan gunners fought an artillery duel with the British in Sherpur from their position on the Asmai Heights. They did not inflict any casualties whatsoever. By contrast, at Sang-­i-Nawishta, British gunners were able to score direct hits on the Afghan guns, which severely demoralized the Amir’s men.40 In the same vein, Afghan infantrymen were issued with only three live rounds per year for training, possibly because it was feared they might constitute a threat to the Amir if they were supplied with more, so fire control and marksmanship suffered accordingly. The British noted that fire tended to be too high ‘and our men passed safely upwards with the storm of bullets rushing far above their heads’.41 Lack of familiarity with the capabilities and ranges of the modern rifles meant that Afghan troops tended to fight at close quarters as they had always done. What the Afghan did possess was greater mass than their neighbouring adversaries. A British Intelligence report observed: ‘In the hour of trouble every

The Afghan National Army and COIN

161

headman and chief of every clan or tribe is called upon and made to furnish levies . . . and 100,000 armed men could be assembled at Kabul without much difficulty.’42 It was generally felt that, having dispersed and abandoned the formal discipline of the regular army when defeated in conventional operations, Afghan soldiers reorganized themselves as irregular fighters. Hensman observed: ‘Occasionally we saw some sort of marshalling going on in the leading lines, in which the best-­armed men were placed, but this was due more to the desire on the part of the leaders to make the most of their strength than to any idea of forming the mob into battalions.’43 Despite the belief that the Afghans ‘trusted to numbers and fanaticism’, Hensman concluded that ‘Afghanistan is a nation of soldiers, every adult being (apart from any military training he may receive) a ready swordsman and a fair shot.’44 Yet, the Afghans were not just inherently hardy warriors schooled in warfare by blood feuds or the harsh environment. They showed they were capable of rapid tactical innovations against British firepower. In late 1879, they adopted dispersed formations to avoid risking heavy casualties from artillery fire. They formed up in ‘small scattered groups or on a line extending many miles across the country’. By doing so, they maximized the number of firearms that could be brought to bear and took advantage of the cover afforded by the terrain. However, organized resistance could also collapse suddenly, reflecting a lack of cohesion. The Amir’s flight from Kabul in 1879 was a prelude to the collapse of resistance in the capital: ‘Before his departure the Amir appeared to have lost all authority at Kabul, while his army had been weakened by numerous desertions.’45 It seemed that ‘The country between Jalalabad and Kabul was . . . in a state of anarchy.’ Yakub Khan took command at Kabul and issued orders to the Ghilzai khans to assemble lashkars at Jagdalak, but they refused and sent ‘submissive letters’ to the British in January 1879, as did the inhabitants of Kabul. One or two serious defeats in open warfare had been sufficient to deter further resistance and prompted the break-­up of the army.

Servants of the dynasty: the Afghan Army, 1881–1901 To the Afghan people, Amir Abdur Rahman Khan always argued that his army was organized purely for the defence of Afghanistan against foreign invasion, but in reality the troops were designed for internal security. The army crushed more than 40 separate disturbances between 1881 and 1901. The direction of these operations Abdur Rahman retained himself. It was noticeable that senior

162

War and State-Building in Afghanistan

officers were selected because they were mediocre and they were given no training so as to prevent any chance of a coup d’état. The British Consul at Meshed in 1886 noted: ‘The company officers . . . seem far superior to the General Officers, for the former do appear to have the instincts of soldiers, but General Officers [are appointed] merely as an act of favour bestowed on any but soldiers; a slave, a surveyor, or a contractor suddenly finds himself raised to the rank of General, and given to the command of an army.’46 The sons of clan elders, khanzadas, were recruited and their promotion accelerated over that of career officers, peshkhidmat, in the hope that ‘in times of emergency the leading men, having regard for the lives of their relatives, would not venture to rise against him’.47 To protect the sensibilities of the clans, Abdur Rahman avoided foreign experts or military advisors. His aim was to ensure that the clan hierarchies became synonymous with the Afghan military. The core of the army consisted of eight regiments, largely volunteers, that had accompanied Abdur Rahman in 1880 towards the end of the Second AngloAfghan War, and he made no attempt to retain the troops that had served his dynastic rival Sher Ali.48 Hazaras and Kizilbash were also dismissed. Durrani Pashtuns were similarly generally excluded because they had fought for Ayub Khan, the main pretender to the throne, in 1880–81. Returning to tradition, it was the Ghilzais who made up the bulk of the junior officers and the rank-­andfile. Nevertheless, the outbreak of the Ghilzai revolt in 1886 had necessitated the mobilization of a larger force, particularly when many Ghilzai regular soldiers deserted, and it was easier to use Durrani Pashtuns, the old rivals of the Ghilzais, than to organize regiments from around the country. Nevertheless, despite the initial surge of recruits that year, service in the army was not popular among the Durrani Pashtuns.49 The exception was a small corps of royal cavalry, the Risala-­i Shahi Kandahari, which was drawn from the families of the Mohammedzais, and retained in order to ensure loyalty to Kabul. By the 1890s, the composition of the Afghan Army was far more balanced. Areas conquered after unrest were called upon to supply recruits, again to ensure future loyalty and to absorb some of the young men of particularly troublesome communities. The exception was the Hazaras, where only a handful of Besud and Jaghuris were enlisted for reasons of sectarian prejudice, which might affect cohesion within the army. Some volunteers from beyond the Durand Line were incorporated, including Khattaks, Waziris and Afridis, and there was an increase in their numbers after the 1897 Pashtun Revolt. The army continued to be deployed with regiments stationed outside their place of origin in order to prevent fraternization with the locals, although intermarriages did occur. All

The Afghan National Army and COIN

163

regiments were organized on the qawm but it is striking that there was only one purely Pashtun regiment.50 In order to increase the size of the army, and make it the ‘strongest in Central Asia’, Abdur Rahman could neither rely on a purely voluntary system, nor on levying traditional quotas of men from pacified districts. It was during the Penjdeh Crisis of 1885, when a Russian invasion seemed imminent, that he attempted to impose a system of conscription on Herat, the hasht nafari, that demanded 15 per cent of all eligible males between 14 and 50 years old. Unfortunately, the local economy could not function under such manpower demands and the regime could not afford the cost.51 In 1887, the system was changed to one man in twenty, selected by the local mullahs and elders, while in Kandahar a system of shasht nefari, one man in six, was imposed.52 By the mid-1890s, this system was being used among irregular levies as well as for the regular army. In 1894, in Logar, Nangahar and Ghazni, families had each been ordered to supply one man, but the unpopularity of the measure led to a steady emigration from border districts into British India. However, the Amir retained the notion of irregular levies that could be deployed in an emergency and he encouraged Afghan elders to ‘keep their young men acquainted with the tactics of war’. The following year, Abdur Rahman introduced a more universal system of char nafari, the service of one in four, modified in 1896 to a selective system of hasht nafari. The British Government of India supported and funded the military expansion of Afghanistan as the best means by which to deny the country to Russian encroachments and therefore protect British India’s landward frontier. In 1883, the regular army had reached 43,000, with a further 3,000 irregulars on a standing basis.53 By 1890, the regular army stood at 60,000 and may have reached 90,000 by the time of Abdur Rahman’s death in 1901.54 There were around 100 infantry regiments and 24 cavalry regiments stationed around the country monitoring the threat of foreign invasion, but focused just as strongly on internal security in order to consolidate Abdur Rahman’s authority and that of his dynasty. The army was, according to one biographer, ‘the high point of his accomplishments’.55 The Afghan Army also changed its role and structure as it expanded. When Abdur Rahman had made his bid to seize the throne with his army from the north in 1880, he had under his command a force of infantry and cavalry that fought as guerrillas even though they had a nominal regular army structure. Once in power, however, the army adopted the standard pattern of the period, with piyada (infantry) and sawara (cavalry) organized in companies, battalions or regiments, brigades and divisions, and the topkhana in batteries. The infantry

164

War and State-Building in Afghanistan

was the largest and cheapest branch of the army. The smallest subdivision was the paira, a section of eight men under a hawaldar. Although theoretically infantry companies consisted of 100 sipahis (soldiers), few reached that strength and consequently the battalion, which should have been made up of six companies, was often less than the established strength of 600 men. That said, in some areas battalions were made up of 1,000 men, depending on local recruiting patterns and the state of the economy. The Afghan regiments were identified by clan, city or region, but sometimes adopted a religious title, such as the Mohammadiyya, or simply the colour of their uniform, such as the Zardposh. The fact that the troops were responsible for providing their own uniforms and clothes led to great variety in appearance. The only item that differentiated a regular soldier from an irregular was a leather belt issued by the Amir. All regulars were supposed to be armed with the modern Martini-Henry rifle, and irregulars with the Snider, but in reality there was again a great diversity in weaponry. The Afghan cavalry regiments had an established strength of 400 sowars, each divided into four troops and commanded by the Zir Mishar (leader of the thousand) or Karnail (colonel).56 Most units, though, appeared to have 316 in a regiment and 80 in each troop until the expansion of the 1890s when cavalry regiments were established at 600 men each. The horsemen were each armed with a carbine and sword, and, until the 1890s, some of these were still muzzle-­ loading Enfields. A few carried lances or pistols but, like the infantry, there were a variety of weapons in each troop. Sowars appear to have provided their own horses like the sillidar system in British India, but there also appear to have been some 30,000 horses of good quality provided by the government while ‘Kataghani horses were most prized’.57 Certainly if a horse was killed in action, the Afghan Government would pay for its replacement; but in peacetime the sowar was expected to pay half the cost if a horse died, and each horse cost about 100 kabulis, although they were far cheaper in the provinces. British observers criticized the efficiency of the cavalry. It was noted that Afghan cavalry never did drill and were scattered into small troops across the country to collect taxes rather than prepare for war. Nevertheless, Colonel Yate, the British Consul in Meshed, described Afghan cavalrymen as uncomplaining types and he praised the Kabuli sowars as ‘less bigoted and more energetic’, noting that they ‘learn quickly’ and are ‘strong, sturdy men as a rule’.58 The Afghan sowars could receive additional pay for bravery, or bahadari, equivalent to one kabuli rupee per month, but the chance of loot far outweighed the meagre pay he could receive every two months from the government. Yate was impressed that

The Afghan National Army and COIN

165

most Afghan sowars were different from the ‘Pathans’ enlisted in the British Indian Army: ‘An Afghan will eat with anybody and do anything, whereas I have seen frontier Pathans call a native officer to account for sitting down to drink tea with his European officers [since] they are as bigoted in their habits as any Hindu.’59 As in the reign of Sher Ali, under Abdur Rahman the artillery was the most prized branch of the Afghan Army. It was estimated that the Amir possessed 860 guns of various types by 1891.60 It was also estimated that there were 5,500 trained gunners, but the quality of their training varied considerably. Some guns were confined to the defence of fortified positions, but of the mobile artillery there were some machine guns, field guns and light mountain guns. Each of these was drawn by mules, horses or elephants, and there was some attempt at organization into field artillery batteries of six guns, as in the Indian Army. In addition to the main branches of the army, the royal bodyguard consisted of three infantry battalions and four cavalry regiments, all of whom were better armed, well fed and paid higher salaries. The Anduran-­i Khas (the household guard) and the Hazirbash (bodyguard of the Amir) were handpicked khanzada men from across the country, including Tajiks and Uzbeks, but with a disproportionate number of Kandahari Mohammedzais, Gardezis and Safis who were considered the most loyal. By contrast the Hazaras, who were regarded as an inferior minority of the country, were formed into pioneer battalions to carry out road building, mining and construction tasks. Despite their status they were constantly at work and therefore considered an efficient element of the army.61 Abdur Rahman gave special privileges to loyal khans and elders, including refined clothing, allowances or remissions on taxes, and tried to reinforce the sense of obligation to him with a notion of hierarchy within the clans.62 Khans were given military ranks including adjutant, colonel and brigadier, and the numbers holding rank increased throughout Abdur Rahman’s reign.63 In return they were expected to continue the tradition of raising a force of mounted levies, to parade these men annually for inspection and to hold arms and ammunition sold to them by the government. Some received payment for retaining a standing force of 3–15 light horsemen, the sawar-­i kushada, which were loyal and could provide the sort of mobility needed to crush an insurrection in its early stages. Abdur Rahman was eager to embrace the idea of a ‘nation in arms’ to strengthen the defences of his country.64 Following the destruction and forced conversion of Kafiristan in the 1890s, the expansion of the regular army, and the

166

War and State-Building in Afghanistan

establishment of a system of espionage and surveillance, the Amir felt that there was no threat from internal opponents that could not be contained. However, he was still concerned about threats to the northern territories and the passes of the east.65 In the north, every third man was supposed to serve as an irregular fighter, and elsewhere there was supposed to be a man in service for every 30 jeribs of land.66 The estimated number of eljaris across the country in 1894 was 90,000, but the comprehensive system was dropped in favour of a more selective hasht nafari. The sheer expense may have played a part in the decision.67 To meet the costs of the more modest hasht nafari, it was suggested that if one man in eight was picked by the khans or elders, then the other seven men should support the selected man while he served in the ranks.68 The overall effect of the system, which appeared to vary across the country, was to create a cadre of trained men who could act as reservists in the event of a war. In 1897, when there seemed to be a chance of war with Britain, military cantonments were established and Afghan men were expected to attend training sessions.69 Regular officers were also sent to villages to train villagers in the arts of musketry and some drill. Uniforms and rifles were issued, although as soon as the crisis had passed the practice was stopped. It was claimed that 100,000 reserves could be called upon in an emergency that year, but as before, the quality of the forces varied and it was unlikely they could hold ground for long. Moreover, suspicions about the government’s intent were aroused and there was opposition to the haft nafari proposals in Khost and Kunar, which resulted in some elders being imprisoned.70 The critical weakness of the Afghan Army, both regular and irregular, was the social division between qawm, clan and ethnic groups. Skirmishes between regiments made up of rival groups were not uncommon.71 In episodes of rebellion, soldiers were also sometimes over-­enthusiastic in their repression. However, the Amir tried to create national unity through religion and a fear of foreign invasion. He repeatedly emphasized the need for bonds between his rule, the khans and the people to withstand the pressures of confrontation and war. A national army held out the hope that this might be possible, but ultimately his regime was held together by coercion and terror. He attempted to establish himself as the sole authority on Islam in the state, punished rebels as enemies of his divinely ordained rule, and crushed armed resistance with a severity that was shocking even by the standards of customary and Islamic law in rural Afghanistan. He revived historic punishments, including the construction of kalla minar (towers of skulls) and claimed to have killed no fewer than 120,000 rebels in his reign.72

The Afghan National Army and COIN

167

The decay of the army In 1919, Amir Amanullah came to power amidst intrigue and with the precarious support of the army that Abdur Rahman had created. Amanullah had refused to accept the accession of his uncle, Nasrullah, and his command of the military garrison at Kabul proved to be decisive.73 To win over the army, he promised to pay each soldier 20 rupees a month and not 11 rupees as offered by his rival. The troops at Jalalabad, where Nasrullah had established himself, immediately declared themselves for Amanullah. Nevertheless, the mullahs of Afghanistan, although largely unaffected by the court coup, favoured Nasrullah because he was known to support the interests of the ulema. Expressions of concern spread rapidly to the rest of the population and even the army began to look unreliable. Amanullah needed to establish his legitimacy with the people. His long and outspoken criticism of his father’s close relations with the British offered the chance to harness public support. Within days Amanullah declared a jihad, and Britain and Afghanistan were at war. The conflict of 1919 was short, and despite local numerical superiority, the Afghan Army was defeated on every front. Even Amanullah’s claim that Afghanistan had actually ‘won’ the war could not conceal the fact that there were no territorial gains to show for it. Only the British concession to relinquish control of Afghanistan’s foreign policy, in order to develop better relations in the long term against Soviet influence, gave the Amir an opportunity to retain sufficient legitimacy to rule. Amanullah exploited the opportunity to the fullest extent, and agreed to terminate his support for the Basmachi rebellion in Central Asia in return for better relations with Moscow. The Afghan Army had performed as badly as predicted. The Afghan Army of 1919 consisted of 78 infantry battalions, 21 cavalry regiments and 280 breech-­ loading artillery, with another 300 obsolete muzzle-­loading smooth-­bore cannons, giving a total strength of 50,000 men. In peacetime, 35 battalions and 3 cavalry regiments were distributed around 10 military districts for internal security while 23 battalions and 8 regiments of cavalry garrisoned positions along the Persian and Russian borders. A reserve was held at Kabul. There was no attempt to create formations larger than an ad hoc brigade of infantry, cavalry and artillery, and there was no staff corps to manage logistics. Afghan regulars had little training, even in ‘skill at arms’, while ‘tactical exercises were unknown’.74 The Afghan regulars were unable to manoeuvre in large formations and consequently they tended to wait for an attack and they focused on occupying defensive positions. Most of the Afghan artillerymen lacked any training, and gunnery practice was again unknown. Nevertheless, the topkhana was issued

168

War and State-Building in Afghanistan

with some German 10 cm field howitzers, and a few Krupp pack guns of 75 mm calibre. There were a few vintage machine guns that required a crank-­handle to operate, but only a handful of units carried ammunition that had smokeless powder, which was all imported or smuggled. Although a smokeless propellant factory had been established in 1912, it had been supplied with the wrong chemicals and was never used. Clouds of smoke simply gave away the positions of every Afghan unit to British gunners. Afghan regular units thus lacked the competence to do any more than occupy hastily constructed defences where they were shelled and outflanked in every battle. What had begun as a limited offensive into India ended with defeat on Afghan soil and an ignominious withdrawal. Desperate to modernize his country, Amanullah’s decision to pension off many senior Afghan officers in favour of Turkish military advisors soon caused deep resentment in the army. Moreover, junior officers were alarmed by the whirlwind of proposed secular reforms after the war that was generating resistance among the ulema and conservative rural khans and elders. The new Administrative Code, drafted by Turkish advisors and which ended the practice by which women and daughters were treated as goods, was condemned by the ulema as a foreign intervention, and there was a great deal of anxiety about the arrival of foreign doctors. In March 1924, a serious revolt broke out in Khost, led by mullahs and maliks, and the Afghan Army units were unable to defeat the rebels.75 In October 1924, one military detachment was destroyed, and there seemed a chance that the rebels would gather more support and march on Kabul. However, Shinwaris, transborder Pashtuns including Waziris, and Hazaras were enticed to enlist as irregular fighters in defence of Afghanistan with generous subsidies, and members of the ulema were persuaded by similar means to condemn the rebels, and even declared a jihad against them. The government’s irregular forces swarmed over Khost and sacked the province. A second revolt in 1928 was not so easily contained. The Afghan Army deteriorated in quality after the Khost rebellion, despite the purchase of some new technologies.76 Pay for the soldiers became erratic and wages were so low that many could hardly feed themselves. There were still no regular training packages, and no medical support services. Equipment was in poor repair. That said, young officers were sent on exchanges with foreign forces and were enthusiastic about new techniques. Senior commanders resented these ‘upstart’ young officers, although often older clan rivalries underpinned the recriminations. The result was that internal security was being weakened precisely when Amanullah needed it most. He could not afford both a strong army and a package

The Afghan National Army and COIN

169

of modernizing reforms, and hoped that the reforms would create bureaucratic rationalization and greater efficiency, which would generate more revenue in the long term. But Britain had ceased paying its generous subsidies when Amanullah had insisted on independence and the country faced a short-­term fiscal crisis.77 In November 1928, the unrest developed into serious outbreaks of fighting.78 By December, the rebel forces, including elements of the army, reached Kabul and Amanullah was forced to abdicate. Nevertheless, for the army, the process of decay and neglect continued until after the Second World War.

The Democratic Republic of Afghanistan Army (DRAA) The chief characteristic of the Civil War period in Afghanistan from 1978, including the years of Soviet military intervention, was the gradual disintegration of the Afghan regular army and the emergence of militias led by paramilitary freebooters popularly styled as ‘warlords’. The Soviets had established links with Afghanistan as early as 1921 but Cold War rivalry meant that by the 1960s and 1970s, Moscow saw the Afghan state as the outer rampart of its vulnerable southern flank. As the Afghan Government’s relations with Pakistan soured, the government in Kabul saw advantage in seeking more support from the northern superpower, but when the number of Soviet advisors increased, particularly for the armed forces, there was some resentment at the potential loss of sovereignty this could entail. Outwardly, the country appeared strong, not least because the Afghan Army and Air Force were equipped with Soviet weapons, vehicles and aircraft, but declining state revenue and nepotism within the bureaucracy created resentment and fear. The fractious and ethnic character of Afghan politics was deepened by a new ideological consciousness throughout the country in the 1970s, which even affected the Communist People’s Democratic Party of Afghanistan (PDPA). The result was that the PDPA Parcham and Khalqi factions, led respectively by Babrak Karmal and Mohammed Nur Taraki, launched a coup d’état against President Daoud on 27 April 1978, an event subsequently edified as the Saur Revolution. In the mêlée, Daoud was murdered together with his family. In the aftermath, thousands of alleged ‘bourgeoisie’, a euphemism for all enemies of the Communist state, were arrested. Led by young Marxist officers, the troops of the 4th Brigade, a commando unit, and elements of the air force, had taken part in the coup, while only 1,800 soldiers and guards had stayed loyal to Daoud. Nevertheless, the coup was hardly swift: the operation had begun in the morning

170

War and State-Building in Afghanistan

at Kabul Airport, but it was not until 16.00 that resistance at Daoud’s palace was crushed, and the strike aircraft appear to have been called in because of the strength of the defences.79 Any observer could see that force, particularly control of the armed forces, would be vital to the success of a future bid for government, and Khalq had already set up a secret ‘United Front of Afghan Communists’ among the officer corps to facilitate their control of the armed forces before they took power. When the government failed to secure any popular backing for its programmes of reform, and opposition increased, Hafizullah Amin took over as Prime Minister. His method of restoring order was to inflict greater terror. He executed his rivals, including Taraki. He sent his secret police to suppress, arrest or murder all opposition. Villages that supported the embryonic resistance were razed and the inhabitants butchered, with the connivance of internal security troops and the secret police. By 1978, with episodes of armed resistance increasing, the country was in crisis. Moscow felt it had to act to restore order and influence, and it launched a coup de main operation in December 1979, using its widespread influence and knowledge developed over the previous two decades. The new President, Karmal, was informed that his role was to reunify the political parties, modernize the army, popularize the regime, extend the government’s revenue and carry out economic development. The opposition to the Afghan Government and its Soviet military support nevertheless assumed the character of a struggle for liberation from foreign ideologues in defence of Islam. The resistance believed they were defending the people from corrupt and coercive government, which the Soviets were supporting, and the very fabric of their life was at stake in what rapidly became a people’s war. There were between 1,600 and 1,800 Soviet military advisers in Afghanistan by the end of 1980, some of whom had served in the country before the intervention. The Soviets retained most advisory influence at senior levels with at least 60 generals, while the rest of the advisors were distributed at a ratio of 11 officers to each Afghan Army division, four to each regiment and four to every battalion.80 Of the Soviet ‘aid’ budget, a significant amount went to the Afghan Army. The total value of aid to the Afghan military and the Soviet Limited Contingent was 1,578.5 million roubles in 1984, increasing to 2,623.8 million in 1985, 3,197.4 million in 1986, and 4,116 million in 1987, which equates to $7.5 billion over four years.81 The Soviet forces struggled to locate and defeat insurgent groups and they resorted to raiding villages where resistance fighters were thought to be

The Afghan National Army and COIN

171

concealed, or advanced into the mountains to destroy mujahideen bases, mounted ambushes and perfected counter-­ambush drills, and mined routes which they suspected the resistance used. In contrast to the fluid, lightly equipped resistance, many Afghan and Soviet troops were in fact pinned to the defence of roads, settlements and communications. More than a third were tied up in this way and it is estimated that combat troops, that is those who took part in regular, offensive operations, amounted to only 20 per cent of the total strength of the 40th Army and the DRAA. Only 51 battalions were engaged in frequent operations out of the 133 battalions in the 40th Army, and, of these, some battalion commanders complained that it was almost impossible to get all the men to take part in combat.82 Furthermore, the sheer effort of getting ammunition and combat supplies into the country by road and air absorbed much of the Soviet 40th Army and a significant part of the Afghan forces. The garrisoned security outposts, the zestavo, particularly those near the border, were manned by Afghan troops less than enthusiastic about their duties. While some remote posts were attacked, other garrisons exercised a ‘live and let live’ system. The journalist Edward Giradet visited one in 1982 and found ‘a group of bored conscripts . . . kicking an empty can about. We met several soldiers hauling water from a nearby desert well. We shook hands and they watched as dozens of guerrillas marched by, leading strings of pack mules loaded with arms, ammunition and other supplies.’83 Information, guides, routes through minefields and safe conduct passes were exchanged. The Afghan Army, which was supposed to support the Soviets, was plagued by desertions, defections, informers, ghost pay rolling and even murdered some of its Russian officers. Many battalions were unwilling to engage in combat. In theory, the army of 1979 consisted of 10 divisions, armed with modern Soviet weapons, with its own integral aircraft, armour and artillery.84 By 1989, the army had expanded to 12 divisions, with several specialized brigades and support services. However, Afghan units were often under strength. A division might only be able to muster 1,000 men. In the early years of the conflict, the desertions reached epidemic proportions: by 1980 it was about 25 per cent of their strength. Two brigades of the 9th Division in Kunar, a brigade in Badakhshan and another brigade of the 11th Division at Jalalabad abandoned the army and either went home or joined the resistance. Loyalty depended on the role of each unit, as specialist, technocratic roles were less prone to desert than the infantry. In the Afghan Air Force many of the officers, pilots and ground crew spoke Russian and had received their specialist training inside the Soviet Union.85 However, most officers of the army and the air force were not trusted by their own

172

War and State-Building in Afghanistan

government or by the Soviets. Amin, and then Karmal, had sacked, arrested or executed officers for disloyalty. Some therefore defected and their expertise proved invaluable to the mujahideen. To increase counterinsurgent manpower, the government lowered the age of conscription from 18 to 15, forced villages to provide manpower at gunpoint, increased service to a three-­year term and mobilized reservists up to the age of 39, although it should be noted that sometimes men of 55 were still serving. Those who refused to enlist were given a four-­year gaol sentence. Absenteeism without authorization could carry a sentence of five years. Desertion could carry a penalty of 15 years or execution.86 Conscription was extended to four years during the conflict, a measure that provoked mutinies in some battalions. By early 1984, attempts to conscript at gunpoint were failing even in permissive areas, as elders tried to spirit the young men away to the hills in order to preserve precious agricultural manpower.87 After their first three-­year stint, soldiers were granted two years’ leave, and then they were required to serve a further three or four years. The only exemption was if soldiers were married and had children. The measures succeeded in stemming the flood of 1980, although ‘attrition’ was never resolved. The available pool of manpower was also drained by clearance operations. The army increased to 40,000 by 1982 and, theoretically, to 150,000 by 1989. Nevertheless, each of the divisions rarely exceeded 5,000 men.88 Possibly about 50 per cent of the conscripts deserted and most of them absconded with their weapons. Not all joined the resistance, and there were reports of some soldiers changing sides up to seven times. Money was often the determining factor. The tradition of low pay in the army was never improved upon, and soldiers earned 200 Afghanis a month, the equivalent of $2. The result was that, in combat, Soviet commanders could not be sure whether the Afghan troops would either sit out the whole action or else simply run off. Garrisons were at least hemmed in by their own barbed wire and mines. By 1987, Afghan officers began to complain that they were being used for all the most dangerous missions, while the Soviets preserved their manpower. And there were clear tensions when Afghan personnel felt as if they were treated as second-­class citizens in their own country. General Kutsenko, a senior military advisor to the Afghan Army, concluded: ‘The Soviet military [officers] served only two years and were then replaced. Few of them learned the customs of the local tribes. But the Afghan commanders had been fighting for 5 to 8 years and they well understood the psychology of their people.’89 Believing that the Soviets should have allowed Afghan units to conduct their own operations, he observed

The Afghan National Army and COIN

173

a lack of cooperation between Afghan and Soviet commanders and a general deterioration in morale: ‘Soviet officers began to say that if the Afghan forces did not want to fight the mujahideen, why should they be doing so?’ Of course, this was not true in every case. Given that many believed the Americans and Pakistanis were behind the resistance, the defence of the country against foreign influences was the motivating factor for many in the government forces. The Afghan and Soviet agencies were supposed to facilitate defections by mujahideen commanders and reconcile them to the government. They practised ‘Psyops’ and attempted to fuel the rivalry between clans to the point where mujahideen groups would fight each other. Intelligence teams were also expected to conduct covert operations: to kill or capture foreign advisors attached to the resistance. Nevertheless, there were serious problems with counter-­intelligence. The resistance maintained a number of sympathetic personnel within the ranks of the Afghan Army or the Tsrandoi (gendarmerie). In May 1985, the head of the Military Intelligence Department of the Afghan General Staff, General Khalil, was arrested with 10 of his officers. He was charged with supplying intelligence to Ahmad Shah Masoud, the Afghan–Tajik mujahideen commander in the Panjshir Valley.90 Nevertheless, much faith was placed in the continued strengthening of the Afghan Army: discipline was tightened and desertion curtailed. Mullahs were incorporated into the Afghan Army’s administration. However, the mujahideen did their best to spread their influence within the army. There were serious breaches of operational security, such that Soviet officers began to inform their Afghan partners about operational objectives only once the action was already underway, or they would make last minute changes, both of which had a deleterious effect on Afghan Army preparation, morale and efficiency. In keeping with the plan for transitioning, and ‘Afghanizing’ the conflict, there were occasional operations led by the Afghan Army that were supported by Soviet firepower, but Moscow had ordered a withdrawal that was well underway in 1987. That year, aside from Operation Majistral in Khost and Operation Typhoon in the Panjshir, Soviet forces remained largely on the defensive. The Soviets were, like the British, eager to employ locals as part of the security apparatus and they set up militias.91 One example illustrates the contours of the Afghan decision to collaborate. The Ismailis in the Kayan Valley of Baghlan Province responded positively to the call of Dr. Najibullah, the Soviets’ final surrogate leader, for a militia, because they were a marginalized group. Surrounded by sectarian or ethnic rivals, they were eager to defend their lands and property. When the Tajiks and Pashtuns near the Salang Tunnel aligned themselves either

174

War and State-Building in Afghanistan

with the mujahedeen Jamiat or with the jihadist Hizbi-­i Islami, Sayyed Mansur Naderi of Kaihan, brother of the pir of the Ismailis, organized his community to arm and defend themselves and the road between Kabul and the Soviet border. By 1989, Naderi had 13,000 troops organized in the ‘80th Division’ under the command of his son, Jaffar, and a place in the local government council. To improve security, the Afghan Army also engaged in negotiations with the aim of establishing territorial units among local leaders who had been former resistance. These groups were essentially co-­opted to regular units or given regular army designations. The local communities decided who would serve, and only one-­third would have to be on duty at any one time, the rest being able to remain at home in civilian employment. The system worked as long as the funds continued to flow in, but allegiances were never guaranteed and many changed sides the moment the Soviets withdrew, although the continued flow of money was the key to ensuring they remained on the Afghan Government’s side. Eventually, 100,000 former insurgents were in government service. The 17th Division in Herat, for example, was made up of 3,400 regular troops and 14,000 militiamen. At Jelalabad in 1989, the mujahideen suffered a major defeat when the DRAA held the city and its approaches with determination. The further deterioration of the stability of the Afghan state was nevertheless exemplified by the attempted coup by General Tenai in 1990. The failure of his plot led to an erosion of confidence in the regular army. The government, now unable to rely on the DRAA, and looking to achieve consensus with rural leaders, formed larger and more numerous militias to replace the regular troops. These forces were soon so large that they were unaccountable to their local areas, in contrast to traditional lashkars. When the funding stopped abruptly with the collapse of the USSR, the militias abandoned the government. The most powerful commanders became ‘warlords’ with personal fiefdoms across the country, but there was a multitude of small-­scale groups that attempted to control their own areas. The Civil War was still unresolved but a new layer of violence had been added as militias sought to expropriate money and property wherever it could be found. At Kandahar, for example, mujahideen factions had been prevented from taking control of the city by militias, such as the Jawzjani of General Dostum. Their habit of looting earned them the sobriquet ‘the carpet stealers’ from the Kandaharis. When the government funds ran out in 1992, the Jawzjani withdrew, and the mujahideen factions were involved in a chaotic takeover. Mullah Naqib and his followers seized the old Afghan Army base; Amir Lalai took control of the commercial district; Haji Ahmad held the airport; Uztaz Abdul Aleem took possession of the police headquarters and prison; and Haji Sarkateb held Bagh-­e

The Afghan National Army and COIN

175

Pul. When it was clear that the city could no longer support these fighting units, the mujahideen commanders turned to robbery. Within a month, the city had been systematically looted and there was barely anything left to sell. Tanks were sold to the Tajik–Afghans in the north and even the hospital was stripped of its beds. When the city had been sacked, the mujahideen turned to banditry on the highways. The lawlessness generated a great deal of anger among southern Pashtun clans.

The Afghan National Army after 2001 In the wake of Operation Enduring Freedom, the Taliban’s military formations were destroyed, driven into Pakistan or simply dispersed among the population. It took three years for the Talib commanders to reorganize, and a further three years before they could form units that could take the field to orchestrate operations against Western forces. A major offensive through the Helmand and Arghandab valleys towards Kandahar was blunted by the simultaneous arrival of British and Canadian ISAF troops in the south, but the resulting fighting and disruption, particularly to the narcotics trade, stirred local resistance. To augment the number of counterinsurgents in Afghanistan, and to pave the way for the Afghan Government to take control of its own security, in 2005 the Americans launched an ambitious plan to increase the size of the ANA to 100,000 by 2011 and eventually to 250,000, to build a fully functioning ANP force of 82,000, and to expand the National Afghan Air Corps. The much publicized ‘surge’ was not only about Western forces, but involved a growth in the ANA from under 79,000 in 2008 to 95,000 in February 2009, and the final scale of the ANSF was planned to be 400,000. The rapid growth of the ANSF was to be matched by a corresponding improvement in quality and professionalism, measured by ‘Capability Milestones’. In July 2009, it was claimed that 47 per cent of ANA units had reached CM1, the ‘ability to operate independently’. However, only 24 police units had reached this total in the same period, while 447 remained at CM4, ‘formed but not yet capable of conducting operations’, reflecting the slower rate of development of the police in Afghanistan more generally. These improvements in quality were considered essential when police abuses and army heavy-­handedness were blamed for antagonizing the population, thereby fuelling the insurgency. However, building an army from scratch in the midst of a war was a challenging undertaking. The ANA recruits were young, 3 in 10 were drug addicts, 9 out of 10 were illiterate and 25 per cent per year were prone to

176

War and State-Building in Afghanistan

desertion.92 Among the embedded training teams (the OMLTs or MTTs), few Western military advisors believed that Afghan units were capable of fighting the insurgents without Western forces being present. One veteran Afghan Major told this author that if the Western forces left Afghanistan, his feeling was the ANA would last two days.93 Others were not so pessimistic and pointed to the domination of non-Pashtun ethnic groups in the ranks of the ANA, which would provide a trained bulwark to any Taliban offensive. However, Pashtuns in the south believed the ANA was a Tajik–Afghan militia designed to oppress them.94 In fact, 40 per cent of the rank-­and-file and 70 per cent of the officer corps was Tajik, which reflected no more than the legacy of the Northern Alliance victory and the relationship of the north with the Karzai Government, rather than conspiracy. The Afghan recruits were not untypical in their expectations upon enlistment. They looked forward to paid employment, regular meals, welfare services for themselves and their families, clothing, status and opportunities. Once in the army, ‘warriors’ acquired the benefits of a new identity, enhanced social status and other rewards. As in many other countries, military service was encouraged through family tradition, or by influential local or feudal leaders, particularly where bounties were on offer if a critical number could be encouraged to serve. Marginalized communities have sometimes sought to enlist in order to achieve a greater balance vis-à-­vis other communities, but it was a common theme to find communities seeking to get their sons some form of education or skills training, and to acquire some of their pay in order to benefit their community (by buying land, for example). Military service involved a social contract between rural communities and the state, and different military forces were raised to fulfil this contract on terms acceptable to both parties, from home defence by local auxiliaries to more elite units. We should also acknowledge that Afghan recruits, with very low levels of education, were no different from other nationalities confronted by the shock of basic training in a regular army. Many found it hard to adjust to life in barracks and a timetable that planned their day from dawn until dusk, and ‘Some were mystified by the socks that came with their uniforms.’95 Food was, inevitably, a common cause of complaint. The training programme was nevertheless relentless: trainers had to turn a raw recruit into a ‘warrior’, capable of reacting to ambushes, shooting, marching, maintaining battlefield discipline, and calling in fire missions in just 18 weeks. One American Staff Sergeant noted that: ‘The hardest lesson is getting through the idea of “one target, one shot”. They tend to go black on ammo.’ Other military trainers called it the ‘spray and pray’ school of target practice.

The Afghan National Army and COIN

177

Because of the relative inexperience, the insurgents preferred to target ANSF rather than ISAF, but it was also a question of firepower. The ANA had fewer armoured vehicles and found it more difficult to call for air support. It is challenging to call in an air strike if one cannot read, or navigate with a map. Crucially, the insurgents’ targeting resulted in a rising casualty toll that affected recruiting. Colonel Karimullah, the officer in command of army recruiting in Kabul remarked: ‘The boys themselves are not afraid, but it is their parents who make the decisions to let them join, and when they see all this on TV, they don’t think it’s worth it.’96 While recruitment rates rose from 600 to 2,000 a month in 2007 because of attractive rates of pay, re-­enlistment remained a problem. Only half the soldiers renewed their contracts once their three-­year terms of service expired. Some soldiers reported that their $100 monthly salaries were less than what they could make growing poppies or smuggling.97 In 2009, there was great concern that expanding the ANA by 50,000 required the recruitment of 100,000 because of the numbers abandoning the army.98 In 2011, the figure was lower, but 24,000 were absent without leave and 5,000 deserted the ANA in June that year alone. The Minister of Defence Abdul Rahim Wardak was nevertheless upbeat about the period after transition, suggesting that the numbers leaving would be reduced as the Afghans assumed more responsibility for their own security.99 The key issue for the ANA was cohesion. Historical examples suggest that cohesion is generated by external pressures, shared experiences (introduced through progressive stages of inoculation) and internal values, ethics and culture. The internal competition within the ANA reflects rivalries within Afghan society, but it is worth highlighting that Afghan national consciousness is also strong. The regimental system preferred by the Europeans, with clear and differentiated identities, has also proved important in the past, particularly when combined with the unifying and mobilizing tenets of Islam. Perhaps the greatest challenge for the ANA, however, has been that the most significant threats in the last thirty years have been from internal sources. Corruption and a lack of a service ethos in government and society are serious concerns. However, it seems that centrifugal forces that threaten the unity and cohesion of Afghanistan may be exaggerated. It remains to be seen whether the ANA can sustain itself into the future, but collapse is surely not inevitable. What is of greater concern is that Western assumptions about the ANA are rarely challenged. The Western policy has been to create a large voluntary service army in its own image and not to develop a ‘nation in arms’. The American military has adopted a corps system of organization with mixed personnel in contrast to Afghan historical models of regional units, or even the colonial Indian Army approach of forming ‘class’

178

War and State-Building in Afghanistan

companies within regional regiments to foster competition but preserve local identities. The Western armies have assumed, without much contestation, that their technological solutions, including the metrics of ‘capability’, are superior to identity and cultural criteria. Developing professionalism, particularly among Afghan officers and NCOs, was a long-­standing issue for the Western training teams of the National Training Mission Afghanistan, Combined Security Transition Command Afghanistan (NTMA-CSTC-A). There was a great deal of confidence that the ANA could be expanded, its logistical, medical and air support elements could be developed, and that its training centres could deliver skilled personnel, but there was some doubt whether there could be a lasting impact on developing the ministerial systems that would oversee the ANA in the time allocated and against the entrenched patronage practices of Afghanistan. There were indeed significant achievements in recruitment and retention. Pay was increased to compete with the market rates of employment in civilian life, but also to match the material benefits of working with the insurgent groups. Literacy training was introduced, with 130,000 Afghan ‘Warriors’ receiving an average of 65 hours literacy instruction by 2011.100 Death or wounds among ANA personnel carried a gratuity or pay continuation for family members, a meritorious promotion system was introduced and improved barrack facilities were constructed. To address the specific problem of professionalism in the officer corps, the National Military Academy of Afghanistan (NMAA) was established on the basis of a four-­year graduate programme like that in the United States. In 2009, 84 graduated out of 360 original applicants, rising to an anticipated 600 graduates per annum from over 3,000 applicants in 2014.101 Integral to the curriculum was a focus on character development and the ethos of honour. Specific classes and invited speakers, addressing the issues of ethics and honour, were particular highlights of the programme. The focus on qualities in leaders was a reaction to the perceived negative impact of warlordism in the civil war period. While commanding personal loyalty and exercising power through patronage, a system built on warlord commanders was vulnerable, particularly if the commander himself was killed (as in the case of Ahmad Shah Masoud). Many warlord commanders were concerned with their own regional fiefdom and not the state per se, placing their narrow economic self-­interests before those of the nation. The need to pay for their personal military forces had led to brigandage and punitive taxation. However, while an emphasis on qualitative leadership was admirable, it was systems of command and the bureaucratization of the army that was of equal

The Afghan National Army and COIN

179

importance in countering insurgency.102 Leadership qualities are a very appealing and reassuring aspect of military history, offering a human dimension to the impersonal nature of war, but it is striking how many apparently ‘great’ leaders were personally flawed. Insurgencies were not always neutralized or crushed by commanders with enhanced features of charisma, courage, determination, knowledge or integrity. Leadership is a function and part of a system of military efficiency; leaders are required to understand their men in order to motivate them, overcome their instinctive fears, and ensure they endure the hardships and psychological pressures imposed by the tedious work of combating insurgents. Leaders are required to reach decisions, plan and possess the knowledge of the tools and processes of war. They are required to give instructions, to synchronize and to control. In countering insurgency, command must be devolved to the lowest level, implying small units with a limited ‘span of command’ able to operate independently and with initiative across a widely dispersed area of responsibility. Moreover, leaders in counterinsurgency have to maintain a regular and analytical approach to communication, up and down the chain of command, but they also have to manage public perceptions through direct contact with the population, keeping one’s word, and acting with courage and restraint. An Afghan Minister expressed concern in 2012 that the rapid expansion of the ANA was not allowing time for the required evolution of a professional culture.103 For the same reason he opposed giving the ANA the heavy weapons and airpower it desired, for, without stating it directly, he implied that there was a risk of a coup d’état. Critics had long claimed that the haste to create a large army meant that it was not possible to develop competent leaders at the brigade level along with their staffs.104 Rather than creating entirely new units with inexperienced officers, it was suggested it would have been far better simply to increase the scale of existing structures first. Nevertheless, there were exceptions to this record of criticism and from 2009 there were efforts to mentor Afghan military staffs through joint planning and operations. The 3rd Brigade of the 201st Corps was able to operate entirely independently from a very early stage and its ability to work alongside local populations and political leaders was far superior to that of ISAF.105 Brigadier General Zameri, the Brigade Commander, won influence with locals in Sarobi Province near Kabul and through the Tagab Valley of Kapisa. Adopting multiple lines of operation, Zameri ensured security for the population but insisted on the eradication of poppy in favour of saffron. He encouraged the digging of irrigation systems, and pomegranate production. His men, without assistance, established their own observation posts above villages, set up their own forward

180

War and State-Building in Afghanistan

operating bases, improved roads with their own engineering equipment, and called public meetings convened by the use of radio broadcasts, the corps’ own imam and through district governors. Improved security enabled the delivery of humanitarian assistance, and local people would phone Zameri directly to inform him of the location of IEDs left by insurgents. Afghan engineers defused these IEDs, adding to the brigade’s reputation. Zameri was also willing to operate with a force of French paratroopers in the north of Tagab, and there was joint planning of operations based on their relative strengths, specifically where the ANA led the operations there was likely to be contact with the civilian population. Instead of heavily armoured and distant ISAF personnel moving through villages, civilians would meet and greet dismounted Afghan troops. Patronage, ethnic partiality and personal connections were still the concern of many trainers and mentors working with the ANA.106 Favouritism, fear of reprisals, appointments and the need to repay ‘benefactors’ affected much of the ANA’s administration from leave allocation and deployments to fuel distribution and issuing of operational orders. One example illustrates the general issues. In Korengal Valley in Nuristan in 2008, the ANA brigade commander had secured his appointment through his political connections, and his corps commander was apparently powerless to replace this less than competent officer. It took the threat of withdrawal by the American training team to persuade the Kabul military authorities to dismiss the brigadier. The answer to this problem lay in selection of officer candidates on the basis of merit or professionalism rather than political clout, the promotion of more experienced and proven NCOs, the expansion of the NCO training programmes and rotation of units and sub-­units from the fourteen most violent districts to quieter areas. Setting an example through demonstration of best practice, in part through visits of Afghan staff to military academies in the Western nations, proved very effective in generating new aspirations in the ANA’s officer corps. Junior officers graduating from the NMAA could be relied upon to adopt the right ethos but the approach of senior commanders was the key to improvement. Continuation of training and education, following the model of officer career development in the West, was also a missing component, even though the value of well-­trained troops in countering insurgency has been proven historically. It is a maxim that inexperienced troops and their less than competent commanders will invariably create friction and abuse of the civilian population, undermining the operational requirement to win hearts and minds. Mentoring and training advisors to the ANA made a number of important observations about the competence of the Afghan forces. The Australian mentors

The Afghan National Army and COIN

181

with the 205th Corps noted that Soviet practices, particularly a reluctance to devolve command responsibility, was characteristic of the older DRAA veterans, while former mujahideen, while lacking knowledge of staff work, were well connected with local politics.107 Ethnic affiliation made a dramatic impact on the organization of the ANA, and despite the Afghan Ministry of Defence policy on balancing quotas of ethnic composition in the ANA, it had no influence in daily administration. The dictates of the Tashkiel, the force established size, was either regarded as sacrosanct or used as an excuse not to make troops available for certain operations. Language was also a barrier to cooperation as some staff spoke only Dari or Pashto, but not both. Regardless of the language, Afghan Ministry of Defence orders were often vague and gave little idea of the actual mission ANA units were supposed to fulfil. This led to further deliberate sabotage of orders, as commanders chose to ignore or reinterpret their tasks accordingly. Some of this can be explained by a desire not to make an error and lose prestige. However, it was also the case that staff work was carried out badly with complacency and incompetence at all levels. Maintenance of vehicles and equipment, for example, was of a low standard and as a result there was a strong reliance on civilian contractors to provide transport and logistical support. Contract management processes were nevertheless weak, resulting in duplication and waste. The Afghan commanders also exhibited a tendency to be reactive and to make sudden decisions to deal with short-­term crises. Less urgent but critical issues received less attention, reflecting a generally bad record of longer-­term planning. Shortages in a number of areas, including manpower, meant that relief, rotation and other management practices were absent or inadequate. Competing agendas, resentment at a flood of enthusiastic ISAF ideas, tokenism in terms of commitment, and a tendency to blame ISAF as an excuse for poor planning were common frustrations. However, the mentors learned to accept ANA preferences and those that followed Lawrence of Arabia’s enduring advice that, ‘it is their war, and you are there to help them, not win it for them’, were the most successful. It was evident as final transition approached that capability milestones used to establish ANA units could not be used as a measure of competence on operations against insurgents.108 Rather than the ability to operate independently, the true criteria were the ability to reduce levels of violence, to secure development work, and to support the work of Afghan political leaders. There was some concern in 2009–12 that for too long, ISAF units co-­opted ANA formations in order to carry out their own operational plans, but this was gradually replaced by a genuine desire to let the ANA lead and conduct its own missions. In 2012, there were still

182

War and State-Building in Afghanistan

discussions about the degree to which ISAF should provide ‘overwatch’ or quick reaction forces, who owned responsibility for the Provincial Reconstruction Teams and how to share intelligence. The greatest misunderstanding was over the meaning of ‘inteqal’/’transition’ itself. To ISAF, it was a gradual process of handing over responsibility while continuing mentoring, training and assistance. To Afghans, it simply meant a handover and the end of an ISAF presence. Reflecting the euphoria of ‘Tranche 1’, which included the handing over of Lashkar Gah, the capital of Helmand Province, Governor Gulab Mangal stated: ‘Today sees the first area in Lashkar Gah transfer to the sons of Afghanistan, who have trained hard for this job and already fight bravely to provide security over many months in Lahskar Gah, defeating the enemy everywhere.’109 Critics continued to point to the lack of preparedness of the ANA, drawing attention to inadequate logistics, poor training programme capacity (particularly continuation training), the lack of basic education and hygiene of soldiers, the risk-­aversion of some European training teams which prevented true partnering, delays in resourcing the training mission at least until the announcement of the ‘surge’ in 2009, the small numbers of trainers available until that point in the campaign, and the continuing lag in development of the ANP, arguably a more important component in countering insurgency, compared with the ANA.110 Moreover, the cost of Afghan security forces was well known to be beyond the capacity of the Afghan Government, suggesting that the United States would have to continue to pay for the ANA and ANP for years to come. The estimate for 2014, given in 2011, was US$6 billion per annum, but given the sensitivity of this amount in a period of austerity, ISAF produced revised figures of US$1.4 billion. This represented a considerable saving compared with the cost of the American military presence, which, in 2011, totalled US$120 billion. The strategic vision was nevertheless that an ANA of 200,000 would be unnecessary once the insurgency was defeated. One Afghan Minister also expressed the view in 2012 that the ANA should be reduced to no more than 100,000, although he was vociferous in his condemnation of Pakistan military interference in Afghanistan, which, he believed, was the real reason for the continuation of the insurgency.111

Conclusion At the time of writing, the ANA has not yet been fully tested as an independent force. Its reliance on Western military assistance has been evident hitherto, but

The Afghan National Army and COIN

183

there have been some encouraging signs of how the ANA might perform beyond the ISAF deadline for combat operations in 2014. This chapter has dealt specifically with the history and evolution of the Afghan regular forces in the context of internal security and the problems that have beset the organization. This is, of course, only part of the story of countering insurgency, a process involving multiple lines of operation and agencies, the most important of which is, invariably, the civil police. Strategic success for the ANA depends on the security envelope it can offer within which a political solution can be arranged and a return to civil policing can be set in place. Armies are, by definition, rather blunt instruments for the restoration of order and security. There are many Western concerns that the ANA may not remain cohesive, that it will be an unsustainable financial burden, and that it may try to usurp power for itself, particularly when threatened with a reduction in scale after c. 2016. However, it is impossible to predict what trajectory the ANA will follow, not least because war is a dynamic contest of wills where friction will always create for us an impenetrable fog. One can state with confidence that various parties of Afghans will look for security, but whether that is provided by a central state or by independent non-­state actors remains to be seen.

Notes 1 COMISAF, HQ ISAF, COIN Guidance 2009; COMISAF, HQ ISAF, COMISAF’s Counterinsurgency Guidance, 1 August 2010; [British] Army Field Manual, vol. 1, Part 10 – Counter Insurgency (Warminster: Land Warfare Centre, January 2010), pp. 3–9 ff.; The US Army and Marine Corps, Counterinsurgency Field Manual FM3–24 (Chicago, IL: University of Chicago Press, 2007); Robert R. Tomes ‘Relearning Counterinsurgency Warfare’, Parameters (Spring 2004), pp. 16–28. 2 See David Kilcullen, ‘Anatomy of a Tribal Revolt’, Small Wars Journal, 29 August 2007 [smallwarsjournal.com/blog/2007/08/anatomy-­of-­a-­tribal-revolt; accessed 2009]. 3 Antonio Giustozzi, War, Politics and Society in Afghanistan, 1978–1992 (Washington, DC: Georgetown University Press, 2000). 4 Anthony Cordesman, Shaping Afghan National Security Forces: What it will Take to Implement President Obama’s New Strategy (Washington, DC: Centre for Strategic and International Studies, 10 December 2009), p. 58. 5 Obaid Younossi, Peter Dahl Thruelsen, Jonathan Vaccaro, Jerry Sollinger and Brian Grady, The Long March: Building an Afghan National Army (Santa Monica, CA: Rand, 2009), p. 60.

184

War and State-Building in Afghanistan

6 Anne Flaherty, ‘Afghanistan Security Force More than a Year Away’, Associated Press, 23 August 2010; Younossi, Thruelsen, Vaccaro, Sollinger and Grady, The Long March, p. 2. 7 Ali A. Jalali (Afghan Minister of the Interior, January 2003 to September 2005), ‘Afghanistan in Transition’, Parameters (Autumn 2010), pp. 17–31. 8 ‘Obama’s Address on the War in Afghanistan’, New York Times, 2 December 2009. 9 COMISAF HQ ISAF, Embedded Partnering Directive 15 August 2009; Younossi, Thruelsen, Vaccaro, Sollinger and Grady, The Long March, pp. 12–16. 10 Younossi, Thruelsen, Vaccaro, Sollinger and Grady, The Long March, pp. 43–8. 11 Younossi, Thruelsen, Vaccaro, Sollinger and Grady, The Long March, p. xiv. 12 Source withheld. 13 Interview of an OMLT officer with the author [name withheld], August 2009. 14 Ganda Singh, Ahmad Shah Durrani (Bombay: Asia Publishing House, 1959). 15 Hasan Kawun Kakar, Government and Society in Afghanistan: The Reign of Amir Abd’al Rahman Khan (Austin, TX: University of Texas Press, 1979), p. 93. 16 The Sadr-­i-A’zam was Mir Abdul-Qasim; the two na’ib al-­sultanat were Naib Sarwar, the Governor of Herat, and Muhammad Nabi; the Na’ib al-Dawla was Sardar Ghulam Rasul, the Governor of Jellalabad; and the Wazir was Sipah Salar Charkhi. 17 Kabul Diary, 37, 25 September 1895, Section F, November 1898, 5, 147–82, NAI (National Archives of India), New Delhi. 18 Ghulam Mohammed Ghobar, Afghanistan der Masir-­i Tarikh (Afghanistan on the Highway of History) (Kabul: Book Publishing Institute, 1967), p. 653. 19 Kakar, Government and Society, p. 111. 20 David B. Edwards, Heroes of the Age: Moral Fault Lines on the Afghan Frontier (Berkeley, CA: University of California Press, 1996), p. 197. 21 Mountstuart Elphinstone, An Account of the Kingdom of Caubul (London: Longmans, Green & Co., 1815), p. 514. 22 Thomas Barfield, Afghanistan: A Cultural and Political History (Princeton, NJ: Princeton University Press, 2010), p. 105. 23 The khans received 6 tomans (120 Kham rupees) per horseman, or sometimes a grain equivalent and the khans would keep a proportion of the allowance for themselves. If the khans did not receive their stipend, they in turn could not pay off their own retainers or other sub-­contracted followers. In Kabul, the muster rolls indicate that in August 1839 there were 5,662 cavalrymen whereas in 1840 there were 5,797. Kandahar raised 1,218. The numbers were limited because they represented what Dost Mohammad, or the Kandahari leaders, could afford. In times of war, these troops were augmented by larger numbers of ‘irregulars’. See also Malcolm Yapp, ‘The Revolutions of 1841–1842 in Afghanistan’, Bulletin of the School of Oriental and African Studies, vol. 27, no. 2 (1964), p. 338.

The Afghan National Army and COIN

185

24 Foreign Department, A of 1 July 1865, no. 3, 26 June 1865, nos 319, 594, 13–15 June, Cabul Diaries, Appendix I, Kabul Munshee received 12 June, L/PS/5/257, vol. 176, enclosures to secret letters, 1863–66, IORL, BL, London. 25 Frere, cited in Vartan Gregorian, The Emergence of Modern Afghanistan (Stanford, CA: Stanford University Press), p. 84. 26 Lambert, ‘Statement of Revenue and Expenditure of Afghanistan, 1877–78’, 6 March 1884, Foreign Department, Secret-F, 1886, NAI. 27 Strength and Distribution and Armament of the Afghan Army, Foreign Department, Secret-F, February 1893, nos 224–9, NAI. 28 Ghobar, Afghanistan dar Masir-­i Tarikh (Afghanistan on the Highway of History), p. 596. 29 Kakar, Government and Society in Afghanistan, p. 96. 30 J.P. Ferrier, History of the Afghans (London: John Murray, 1858), pp. 312–13. 31 Report on the Second Afghan War, strictly confidential, Captain Oliver for Major General MacGregor, Simla, 1885–86, p. 3, L/MIL/17/14/29/1, IORL. 32 Various estimates have been offered for state expenditure in this period. Lieutenant N.F. FitzG. Chamberlain estimated that the Afghan military budget totalled 1,921,195 Kabul rupees (Kr) and was broken down as follows: Kr 1,781,233 on pay; Kr 120,235 Kabul Arsenal; Kr 19,727 Military Clothing Department. The total Afghan budget was thought to be Kr 7,982,390. Howard Hensman, The Afghan War of 1879–80 (London, 1881, reprinted New Delhi: Lancer, 2008), p. 320; C.M. MacGregor, The Second Afghan War: Official Account (London: Intelligence Branch, 1908), pp. 634–5. 33 See P.S.A. Berridge, Couplings to the Khyber (Newton Abbot: David & Charles, 1969). 34 The ‘gap’ was not purely technological but also involved new forms of organization and impulses to imperial possession, but, in many instances, these changes revisited older concepts and should be seen in context. See Jeremy Black, Western Warfare: 1775–1882 (Chesham: Acumen, 2001), p. 187. 35 Hensman, The Afghan War of 1879–80, p. 330. 36 Lord Roberts, Forty-One Years in India: From Subaltern to Commander-­in-Chief (1897, reprint, New Delhi: Asian Educational Services, 2005), pp. 559–63. 37 Hensman, The Afghan War of 1879–80, p. 321. 38 Hensman, The Afghan War of 1879–80, p. 321. 39 Christian Tripodi, ‘Peacemaking through Bribes or Cultural Empathy? The Political Officer and Britain’s Strategy towards the North-West Frontier, 1901–45’, Journal of Strategic Studies, vol. 31, no. 1 (2008), pp. 137–8. 40 Hensman, The Afghan War of 1879–80, p. 322. 41 Hensman, The Afghan War of 1879–80, p. 322. 42 Police Report, no. 27, 14 April 1878, in MacGregor, The Second Afghan War: Official Account, p. 635.

186

War and State-Building in Afghanistan

43 Hensman, The Afghan War of 1879–80, p. 323. 44 Hensman, The Afghan War of 1879–80, p. 213. 45 Hensman, The Afghan War of 1879–80, p. 40. 46 Colonel J.W. Ridgeway to Foreign Office, 20 December 1886, cited in Kakar, Government and Society, pp. 96–7. 47 Kabul Diary, 20–23 February 1892, p. 959, L/PS/7/65, IOR; Kayz Mohammed, Saraj al-Tawarikh (The Lamp of History) (Kabul: Afghan Government Publication, 1915), p. 642; Kakar, Government and Society, p. 97. 48 Newswriter in Kandahar to Government of India, Simla, Telegram, 5 November 1880, no. 8, p. 1945, L/PS/7/26; Peshawar Agency Diary, 20 December 1899, no. 179, L/PS/7/119 IORL. 49 Kayz Mohammed, Saraj al-Tawarikh (The Lamp of History), p. 554. 50 Kakar, Government and Society, p. 98. 51 Peshawar Diary, 31 May 1885, p. 225, L/PS/7/44, IORL. 52 Kabul Diary, 5 August 1887, p. 1843, L/PS/7/50, IORL. 53 Kandahar Diary, 1 February 1883, p. 762, L/PS/7/35, IORL. 54 Kabul Diary, 26 March 1890, p. 1241, L/PS/7/36; Kabul Diary, 11 July 1896, Letter no. 2090-F/96, L/PS/7/87, IORL. 55 Kakar, Government and Society, p. 98. 56 The colonel was paid a handsome 200 kabulis, with a sliding scale of pay for each of the subordinate ranks. The second in command was the Nap or Adjtan; there were four Sil Mishar, or Ressaldars; four Parak Mishar (commander of fifty), or jemadars; one Mirza (head of admininistration); one Kotnap (Major) who ran the pay and accounts; one Landakwar (Sergeant), or sajan, who was in charge of rations and forage. Some sub-­ranks, like Mir Akhors, were excused guard duty or given other privileges; Sowars were paid 20 kabulis. Lt. Col. Charles Edward Yate, Afghan Cavalry (Simla: Government of India, 1893), p. 1, L/MIL/17/14/19, IOR. 57 J.W. Ridgeway, Strength of Afghan Troops in Turkestan and Badakshan, September 1886, L/PS/18/A70, IOR; Yate, Afghan Cavalry, p. 4, L/MIL/17/14/19, IOR. 58 Yate, Afghan Cavalry, p. 2, L/MIL/17/14/19, IORL. 59 Yate, Afghan Cavalry, p. 3, L/MIL/17/14/19, IORL. 60 Military Report on Afghanistan (Calcutta: Government of India, 1906), pp. 201–2, L/MIL/17/14/4, IOR. 61 Military Report on Afghanistan, p. 197, L/MIL/17/14/4, IOR. 62 David B. Edwards, Heroes of the Age: Moral Fault Lines on the Afghan Frontier (Berkeley, CA: University of Califormia Press, 1996), pp. 118–19. 63 Mir Agha Sahibzada to the Peshawar Commissioner, Peshawar Diary, 8 October 1891, p. 764, L/PS/7/64, IORL. 64 Military Report on Afghanistan, p. 191.

The Afghan National Army and COIN

187

65 Peshawar Diary, 29 October 1895, no. 20, L/PS/7/83; Kabul Diary, 28 April to 1 May 1894, p. 757, L/PS/7/74, IORL. 66 Kabul Diary, 21–24 April 1894, p. 647; Kabul Diary, 28 April to 1 May 1894, p. 757, L/PS/7/74, IORL. 67 The Military Resources of Afghanistan, 23 May 1887, L/PS/18/A78, IORL. Kitab-­i Qanun-­i Afghanistan (Books on the Payment of the Army of Afghanistan) (Kabul, n.d.) cited in Kakar, Government and Society, p. 308. 68 Kayz Mohammed, Saraj al-Tawarikh (The Lamp of History), p. 1172. 69 Kabul Diary, 11 July 1896, Letter no. 2090 F (96), L/PS/7/87, IORL. 70 Kakar, Government and Soviety, p. 112. 71 Riyazi, ‘Ayn Waqai’ (Meshed, 1907), p. 245. 72 Riyazi, ‘Ayn Waqai’, p. 111. 73 General Staff, Army Headquarters, India, The Third Afghan War, 1919: Official Account (Calcutta: Government of India, 1926), p. 13. 74 General Staff, Army Headquarters, India, The Third Afghan War, 1919: Official Account, p. 23. 75 The Khost Rebellion, L/PS/10/1112, IORL. 76 Afghanistan: Purchase of Tanks, Armoured cars, etc., File 4658, pt. 2, 1924–1925, L/PS/10/1130, IORL. 77 Afghanistan: The Amir’s Subsidy, 11 February 1919 to 22 July 1919, L/PS/11/154, file P3845/1919, IORL. 78 Afghan Rebellion, L/PS/10/1285–95, IORL. 79 Mohammad Yousaf and Mark Adkin, Afghanistan: The Bear Trap (London: Leo Cooper, 1992), p. 142. 80 A. Maiorov, Pravda ob afganskoi voine (Moscow, 1996), p. 110. 81 Artemy Kalinovski, The Blind Leading the Blind: Soviet Advisors, Counter-­ insurgency and Nation-Building in Afghanistan, CWIHP Working Paper 60 (January 2010); Memorandum from Chairman of USSR GOSPLAN, N.K. Baybakov, ‘Regarding Additional Aid to the DRA’, 8 October 1980, International Department (RGANI), fund 5, Op. 77, D. 802, p. 44, Harvard University, Cambridge, MA. 82 Lester Grau, The Soviet–Afghan War: Superpower Mired in the Mountains [http://www.smallwars.quantico.usmc.mil/search/LessonsLearned/afghanistan/ miredinmount.asp; accessed December 2010]; V. Korolev, Uroki voiny v Afganistane 1979–1989 godov [http://www.sdrvdv.org/node/159; accessed December 2010]. 83 Edward Giradet, Afghanistan: Soviet War (London: Croom Helm, 1985), p. 61. 84 According to Gorelov, the senior Soviet military advisor in Afghanistan, the Afghans had 145,000 men, 650 tanks, 87 infantry fighting vehicles, 780 armoured personnel carriers, 1,919 guns, 150 aircraft and 25 helicopters.

188

War and State-Building in Afghanistan

85 Mark Galeotti, Afghanistan: The Soviet Union’s Last War (London: Frank Cass, 1995), p. 7. 86 Yousaf, Afghanistan: Bear Trap, p. 57. 87 G. Bobrov, Soldatskaya Saga (Moscow: AST Publishing Group, 2007), pp. 237–40. 88 Mark Urban, War in Afghanistan (London: Macmillan, 1990), p. 106; Gilles Dorronosoro, Revolution Unending: Afghanistan, 1979 to the Present (London: Hurst, 2005), p. 188. Yousaf, Afghanistan: Bear Trap, p. 58. 89 V. Ogryzko, Pesni afganskogo pokhoda (Moscow, 2000), p. 49, cited in R. Braithwaite, Afgantsy (London: Profile Books, 2011), p. 143. 90 D. Gai and V. Snegirev, ‘Vtorzhenie’, Part I, Znamia, no. 3 (March 1991) and Part II, Znamia, no. 4 (April 1991), p. 137. 91 Angelo Rasanayagam, Afghanistan, p. 130; Barnett R. Rubin, The Fragmentation of Afghanistan: State Formation and Collapse in the International System (1995, reprint, New Haven, CT: Yale University Press, 2002), pp. 159–60. 92 United Nations Assistance Mission in Afghanistan (UNAMA), Annual Report on Protection of Civilians in Armed Conflict, 2009 (New York: UNAMA, 2010). 93 Interview with the author, Kabul Military Training Centre, August 2010. 94 James Fergusson, Taliban (London: Bantam Press, 2010), p. 158. 95 Aryn Baker, ‘Taking Aim at the Taliban’, Time Magazine, 16 August 2007. 96 Baker, ‘Taking Aim at the Taliban’. 97 For the ‘pay crisis’, see Anthony Cordesman, Shaping Afghan National Security Forces: What it will Take to Implement President Obama’s New Strategy, 10 December 2009, p. 54. 98 Thom Shanker and John H. Cushman, ‘Reviews Raise Doubts on Training of Afghan Forces’, New York Times, 6 November 2009; Joshua Partlow, ‘More Afghan Soldiers Deserting the Army, NATO Statistics Show’, Washington Post, 2 September 2011; Younossi, Thruelsen, Vaccaro, Sollinger and Grady, The Long March, p. 19. 99 Partlow, ‘More Afghan Soldiers Deserting the Army’. 100 Anthony Cordesman, Shaping Afghan National Security Forces: What it will Take to Implement President Obama’s New Strategy, 10 December 2009, p. 56. 101 Author interviews and NMAA statistics, NMAA, Kabul, March 2010. 102 For an advocate of the qualities approach, see Mark Moyar, A Question of Command (New Haven, CT: Yale University Press, 2009). 103 Author interview with Afghan Minister [name and location withheld], March 2012. 104 Jeff Haynes, ‘Reforming the Afghan National Army: Getting the Most out of the ANA So We Can Do Less’, Foreign Policy Research Institute, 13 November 2009. 105 Haynes, ‘Reforming the Afghan National Army’. 106 Anthony Cordesman, Shaping Afghan National Security Forces: What it will Take to Implement President Obama’s New Strategy, 10 December 2009, pp. 26, 52.

The Afghan National Army and COIN

189

107 LTC Gavin Keating, ‘ “Living in the Twilight Zone”: Advising the Afghan National Army at the Corps Level’, Australian Army Journal, vol. 8, no. 3 (Duntroon: Land Warfare Studies Centre, Summer 2011). 108 For CM improvements in 2009, see Younossi, Thruelsen, Vaccaro, Sollinger and Grady, The Long March, p. 47. 109 ‘Afghans Take the Lead in Lashkar Gah Security’, 20 July 2011, MOD DNI, London. 110 Anthony Cordesman, Shaping Afghan National Security Forces: What it will Take to Implement President Obama’s New Strategy, 10 December 2009, pp. ii, 38–9. 111 Interview with the author (name withheld), March 2012.

7

Revising COIN: The Stakeholder-centric Approach Erik Reichborn-Kjennerud, Karsten Friis and Harald Håvoll

Introduction With the apparent lack of progress and success in Afghanistan, COIN, both as a theory and a practice, is falling out of favour with the political and military establishment in the United States. This comes at a time when the US is redirecting its geopolitical focus away from global instability towards the Asia-Pacific and the ‘New Great Power Game’. The 2012 US Defense Strategic Guidance clearly states that US forces ‘no longer will be sized to conduct large-­scale, prolonged stability operations’ like those in Iraq and Afghanistan. Instead, they will ‘emphasize non-­military means and military-­to-military cooperation to address instability and reduce the demand for significant US force commitments to stability operations’. It goes on to explain that they will be ready to conduct limited COIN operations if required, but emphasizes that this will mostly be done by operating alongside coalition forces, meaning that ‘helping others defend themselves’ will be the new mantra for reducing instability around the world.1 In essence, the 2012 Strategic Guidance calls for an end to COIN operations. In addition, the operations in Afghanistan have taken on a new phase that focus primarily on capture/kill operations and Foreign Internal Defence (FID), so-­called ‘COIN-lite’, rather than population security, good governance and nation-­ building. Although this points to the demise of COIN as policy and military practice, the US military is currently re-­writing its COIN doctrine and the Defense Strategic Guidance points to the need to ‘retain and continue to refine the lessons learned, expertise and specialized capabilities that have been developed’2 over the past decade in Iraq and Afghanistan.

192

War and State-Building in Afghanistan

While COIN seems to have fallen out of grace due to the apparent lack of success in Afghanistan and Iraq, one should maybe not be as quick to throw the baby out with the bath water. COIN theory in itself may not be at fault for the lack of results. Regardless of the desire not to engage in lengthy, large-­scale stability operations, history tells us that expertise and specialized capabilities to conduct such operations will be needed in the future. By analysing and comparing COIN as theory and COIN in practice, this chapter seeks to understand what can be drawn from existing theory and from its critics in order to inform how COIN can be revised to guide future stability and counterinsurgency operations. While this article will not offer any panaceas for COIN operations, it will be argued that a focus on stakeholders in the conflict rather than on the population or the enemy is a better approach for countering insurgencies and ensuring long-­term stability in war-­torn states.

COIN in theory3 When what started as a conventional war turned irregular in late 2003 after the invasion of Iraq, there was no updated doctrine available to turn to when faced with a growing insurgency. The US Army and Marine Corps were organized, trained and equipped for fighting conventional wars against regular enemies. From fighting its preferred wars against formed units in the open the US now were faced with individual enemies fighting in and from the shadows. This deficit had to be remedied quickly and the work gained momentum when Lieutenant-General David Petreaus returned from his second tour of duty in Iraq in October 2005 to take command of the Combined Arms Center (CAC) in Fort Leavenworth. He soon collected a group of competent personnel to start working on a revised COIN doctrine4 and at the same time built a strong rapport with his US Marines counterpart Lieutenant-General James Mattis.5 In December 2006, the new doctrine was published as a combined US Army and US Marines product, and it immediately had an impact on the conduct of operations in Iraq as well as on education and training in the US Army. This chapter takes as a point of departure COIN theory as it is presented in the US doctrine FM 3–24. The role of doctrines varies from country to country. In an ideal world, doctrines would drive decisions on how the armed forces of a country should be organized, what missions it should train to accomplish and what equipment it needs.6 This in turn points towards a prescriptive role of doctrines. In the US

Revising COIN: The Stakeholder-centric Approach

193

military in general and in the US Army in particular, doctrines are very important and come close to this ideal, especially since the establishment of the Army’s Training and Doctrine Command (TRADOC) in 1973 and the pursuant issue of the famous Air Land Battle doctrine of 1982. In other countries, like Norway, doctrines do not have the same tradition and roles. The Norwegian Joint Doctrine leans more to the descriptive side of a descriptive–prescriptive scale. Within the Norwegian Armed Forces, doctrine is not something you bring with you to the battlefield or conflict area but something you use in your education and preparation for deployment. Doctrine is not about how to conduct war and military operations but how to think about war and operations and only to a very limited degree guides equipment procurement. The US doctrines have, however, tended to be more like guidelines to help the commanders at all levels in their actual conduct of operations as well as guiding the structuring and training of the force – thus more on the prescriptive side. The FM 3–24 appears to have taken a step towards the descriptive side compared with previous US doctrines. However, the impact the doctrines have had on the organization, training and hardware of the US Army still points towards a strong prescriptive role. A central imperative of COIN as presented in FM 3–24 is to learn and adapt.7 ‘In COIN, the side that learns faster and adapts more rapidly – the better learning organization – usually wins.’8 It would seem that a doctrine that is too prescriptive in its guidance to the actual conduct of operations would inhibit learning and adaptation while a more descriptive one better allows for this. In this chapter, we will analyse COIN theory and FM 3–24 against recent practice as prescriptive rather than descriptive theory.

What is an insurgency? An insurgency is first and foremost a struggle for the political power over the allegiance of the population in a given territory. It is a method employed by a non-­state actor to challenge the existing political authority. According to FM 3–24, an insurgency is about the overthrow of a government through the use of subversion and armed conflict.9 Actually, FM 3–24 is somewhat ambivalent in its description of what an insurgency is, as it also states: ‘an insurgency is an organized, protracted politico-­military struggle designed to weaken the control and legitimacy of an established government, occupying power, or other political authority while increasing insurgent control’.10 If to weaken the control and legitimacy of the government is included as an aim of insurgents, it would

194

War and State-Building in Afghanistan

significantly expand the scope of cases that can be included in the insurgency category. What separates insurgents from criminals is primarily the political aim of the insurgents and it is their means and ways that separates them from terrorists. Organized crime is purely parasitic – the only aim of criminals is self-­ aggrandisement, and they do not serve a constituency other than themselves.11 Although insurgents and terrorists sometimes employ similar methods (for example, suicide bombings), the main difference lies in their size and organization. Both groups are fighting a political struggle but the insurgent’s main method of armed struggle is through guerilla warfare primarily against enemy military forces. A terrorist group, on the other hand, is normally numerically too small to wage a guerilla war. They seldom operate as more than a handful in each action and their targets are primarily civilian. Unlike an insurgent group, they are neither able nor willing to seize and hold territory and to exercise some form of control over a defined territory.12

What is counterinsurgency? There is no generally agreed upon definition of counterinsurgency. FM 3–24 states that ‘Counterinsurgency is military, paramilitary, political, economic, psychological, and civic actions taken by a government to defeat insurgency.’13As such, it is simply ‘an umbrella term that describes the complete range of measures that governments take to defeat insurgencies’.14 All insurgencies are sui generis – of its own kind – and there are numerous ways to defeat them. In order to find the most appropriate way of conducting a specific counterinsurgency operation, one must understand each particular conflict with reference to three defining factors: the nature of the insurgency being countered, the nature of the government being supported, and the environment – especially the human environment – in which the conflict takes place.15 This chapter focuses on counterinsurgency waged by external forces in support of a host nation government. There is a clear distinction between a counterinsurgency waged by a local government against domestic insurgents and counterinsurgency fought mainly by external, foreign forces supporting a host nation government. The struggle for legitimacy, a centerpiece in FM 3–24, is considerably more difficult for foreigners, particularly if they are of a different ethnicity, religion or culture. The challenge for external forces is that COIN is a

Revising COIN: The Stakeholder-centric Approach

195

protracted struggle and what might have been a positive attitude among the local populace at the outset may wear thin with time, especially if no apparent progress is produced. Eventually their legitimacy may crumble and what were once seen as liberators may be regarded as occupiers. ‘Victory’ in COIN is an elusive concept. It is very difficult to define what constitutes success and how to know when an end state has been reached. Some would claim that annihilation of the insurgents is the goal, while others assert that success is when the insurgency has become ‘manageable’ by the government. Yet, others would hold that the best we can hope for is to change an old process into a new one.16 According to FM 3–24, victory is achieved ‘when the populace consents to the government’s legitimacy and stops actively and passively supporting the insurgency’.17 Success depends on the goals set by the politicians, but the field manual suggests that to ‘defeat’ an insurgency, the purpose is to address the underlying conditions for the insurgency. This is to be done through reforming and strengthening the existing political order so it will be seen as legitimate in the eyes of the population. The ability of a political authority to deliver public goods thus becomes an integral part of the strategic objective to equip the host nation to govern and secure itself. This refers to the need for reform of the host nation, something that requires the coordinated efforts of the whole range of political tools available to the counterinsurgents. For an intervening force, the purpose necessarily includes national interests. They share the goal of sustainable stability by a host nation government able to govern and secure itself, but the purpose of this stability for an external actor is to prevent local and regional instability and to ensure future threats to its interests do not emanate from that state, something that will allow the intervening force to exit. The field manual defines legitimacy as the primary objective of any COIN operation. ‘A COIN effort cannot achieve lasting success without the HN [host nation] government achieving legitimacy.’18 Long-­terms success ‘depends on the people taking charge of their own affairs and consenting to the government’s rule’.19As such, COIN puts the population at the centre of its strategy in order to achieve its end state of a legitimate host nation government able to govern and secure itself. The manual thus frames counterinsurgencies as contests for legitimacy between the insurgents and the counterinsurgents. ‘At its core, COIN is a struggle for the population’s support.’20 ‘Political power is the central issue in insurgencies and counter-­insurgencies; each side aims to get the people to accept its governance or authority as legitimate.’21

196

War and State-Building in Afghanistan

Though legitimacy is not directly defined in the manual, it describes legitimate governments as those that rule with consent of the governed while illegitimate ones rely mainly on coercion to keep control of its population.22 The emphasis on legitimacy as the main objective of any COIN operation is based on the idea that legitimate governance is inherently stable because ‘the societal support it engenders allows them to adequately manage internal problems, change, and conflict that affect individual and collective well-­being’. This is contrasted to illegitimate states that are seen as inherently unstable and unable to regulate society or can only do so by applying overwhelming coercion.23 In essence, COIN theory focuses on the underlying factor of bad governance as the source of instability and root cause of the insurgency. Building or restoring legitimacy thus becomes the way to achieve one’s desired ends. Legitimacy, according to the field manual, is to be achieved through a balanced application of both military and non-­military means. This is because military means alone can only address the symptoms of a loss of legitimacy and not restore or enhance the legitimacy necessary to achieve durable peace. The field manual lists six possible indicators of legitimacy that can be used to analyse threats to stability: (i) ability to provide security for the populace; (ii) selection of leaders that are considered just and fair by a substantial majority of the populace; (iii) high level of popular participation in or support for political processes; (iv) culturally acceptable level of corruption; (v) culturally acceptable level and rate of political, economic and social development; (vi) high level of regime acceptance by major social institutions.24 These indicators are deemed important to achieve the support of a sufficient majority of the population. Although different societies and cultures may place different emphasis on the various indicators, these indicators point to the need for security, elections and welfare for the population and consequently also reform of governance. As such, COIN is as much about state-­building and social re-­engineering as it is about fighting the enemy. ‘Counter-­insurgents aim to enable a country or regime to provide the security and rule of law that allow establishment of social services and growth of economic activity.’25 Since ‘the primary objective of any COIN operation is to foster development of effective governance by a legitimate government’,26 the focus in COIN is not control of territory or purely the destruction on the enemy’s capacity to fight, but for the ‘minds’ of the population. Thus the activities of the foreign intervening forces must serve to alter the population’s perception of the government through reforming its governance capacity. This means that the relationship between the intervening forces, the government and the population in large part will

Revising COIN: The Stakeholder-centric Approach

197

determine how legitimacy is perceived. In this sense, it is vital that the intervening forces communicate and interact with the host nation population in order to determine what they define as effective and legitimate governance and that all commanders must consider how operations contribute to strengthening the host nation’s legitimacy. If the demands of the population, be it security, welfare or elections, are met by the government, legitimate control can be achieved; if not, legitimacy is at risk. ‘In the end, [the population] determine the ultimate victor.’27 The field manual’s approach to countering an insurgency – often termed population-­centric COIN for its focus on the population – uses military force to foster the conditions for long-­term economic development and good governance to make the central government of a host nation legitimate in the eyes of the general population. Population-­centric COIN is often conducted through what the field manual calls a clear-­hold-build operation that has three objectives: create a secure physical and psychological environment by clearing out the insurgents, establish firm government control over the populace and area by holding territory (preferably by host nation government security forces), and gain the populace’s support by building up support for the host nation government through delivering essential services. This approach is aimed at developing a long-­term effective host nation government framework that secures the people and their basic needs, thus reinforcing the government’s legitimacy.28 By controlling key areas, security and influence will then spread out into other areas.29 ‘Clear-­hold-build objectives require lots of resources and time. The US and HN commanders should prepare for a long-­term effort.’30 COIN is a political-­military struggle, and although military efforts are necessary and important to COIN, it is only effective when integrated into a shared strategy with the other elements of national power, making unity of effort an essential element.31 While COIN is not primarily a military fight, ‘controlling the level of violence is a key aspect of the struggle’, as a ‘more benign security environment allows civilian agencies greater opportunity to provide their resources and expertise’.32 This makes the military an enabling factor in COIN – not a solution in its own right – as without security few other lines of operation can be initiated or sustained. David Galula’s ‘formula’ of 20 per cent military and 80 per cent civilian effort in COIN is not to be taken literally but rather as an indication of the resources and efforts needed over time to produce a sustainable stability.33 At certain times and in certain areas, the military effort will be the main one with civilian efforts in support. In other areas and at different times, the opposite will be the case. FM 3–24 states that the military effort is a

198

War and State-Building in Afghanistan

combination of offensive, defensive and stability operations and that the weight of each type of operation is at the commander’s discretion dependent on the situation and the mission.34 The purpose of the use of military force in COIN according to FM 3–24 is to create legitimacy through protection of the population and to enable development and rule of law. As such, the ‘counter-­insurgents take upon themselves responsibility for the people’s well-­being’.35 The underlying logic can be represented by a metaphor: ‘If you have a mosquito problem the solution lies in the swamp – not in swatting as many mosquitoes as possible.’ As Bernard Fall argued almost 50 years ago, ‘when a country is being subverted it is not being outfought; it is being out-­administered’.36 This means that the fight is really a competition over government and not about who can outfight the other. According to FM 3–24, military forces contribute to the legitimacy of the host nation government by providing security to the population.37 The most cost-­ effective way of achieving this is by securing the main population centres. To build legitimacy, the use of force must be constrained, proportionate and discriminate. The idea is that collateral damage has a more negative impact on legitimacy than the positive effect the impression of strength has. In order to achieve a precise effect by military force, timely and correct intelligence is paramount. And to get to such information the analysts must understand the local context – in particular the so-­called ‘human terrain’. This collection of actionable intelligence in turn requires the forces to interact with the local population thus increasing the risk to the troops. The close interaction with the people also enhances the legitimacy of the COIN forces as the COIN forces’ safety becomes the people’s safety – and vice versa. In addition, the field manual ascribes stability operations – civil security, civil control, essential services, governance, and economic and infrastructure development – as a vital part of the COIN effort in order to gain legitimacy for the host nation government. Although these are mainly civilian tasks, it is expected that military forces contribute either in support of civilians, or directly when civilian experts are not available to undertake such tasks. As such, the field manual states that ‘Soldiers and Marines are expected to be nation builders as well as warriors.’38

COIN and its critics Although COIN as the guiding principle for the operation was not implemented until the arrival of General Stanley McChrystal in June 2009,39 COIN has been

Revising COIN: The Stakeholder-centric Approach

199

widely criticized after its limited success in Afghanistan. The apparent failure to provide victory or basic security in the country after a decade of fighting indicates a significant fault, but whether it is the implementation of the COIN doctrine or the doctrine itself that is at fault in Afghanistan is a moot question. While it is commonly understood that many aspects of the implementation has been flawed, owing to limited resources, coalition-­related caveats, resistance to the doctrine, and so on, several critics have also concluded that the doctrine and the entire COIN theory has been proved wrong. Some of the critique of the doctrine has been aimed at the limited scope of historical cases it is based on, and that both older and newer conflicts not presently included should inform a revision of the doctrine. The COIN theory adopted by FM 3–24 is primarily based on case studies from the Cold War – in particular, Malaya, Algeria and Vietnam as presented through the seminal works of David Galula, Roger Trinquier, Frank Kitson, John Nagl and others. This critique of the background and production of the doctrine have some merit. The cases referred to are too homogenous as the sole basis of a general theory of COIN. They all took place within the same limited time frame (1950s to 1975), within the same geopolitical setting (the Cold War) and within similar local political settings (an insurgency against a colonial or puppet rule countered by external forces). The dilemma for the writers of doctrine, however, is that the more specific the doctrine, the more it is relevant only to a limited spectrum of scenarios. The more wide-­ranging and general the doctrine, the more it risks being relevant to none. Sebastian L.v. Gorka and David Kilcullen simply state that this dilemma cannot be solved within one unified doctrine: ‘it becomes evident that a single unified counter-­insurgency doctrine is not possible, that there can be no universal set of best practices evolved over time that can cover such diverse starting points, end-­states, and local context’.40 The critics of the doctrine’s recommendations can be roughly divided into two groups. On the one hand, there are those that agree with the overall tenets of population-­centric COIN as described in the field manual, but see flaws in the theory and argue for revision of the manual, especially regarding recent empirical evidence from Iraq and Afghanistan and other historical case studies. On the other hand are those who argue for a so-­called enemy-­centric approach, that the population-­centric COIN approach is flawed and that the focus should be on the insurgents, not the insurgency. In the following, we discuss some of the critiques of COIN theory based primarily on the Afghanistan experience, and the alternative approaches that are being launched. It is not a discussion of all the things that have gone wrong in

200

War and State-Building in Afghanistan

Afghanistan, but of those aspects emphasized by its critics to conclude that COIN has been proved wrong.

Revising population-­centric COIN The first type of critique can be roughly divided into two parts. First, it is claimed that the concept of legitimacy is too Western-­centric, founded on what the West view as universal norms, based on the rights of the individual over the community as a form of social contract between the ruler and the ruled. Rather than being based on universal values and norms, the critics claim, the foundation of legitimacy is found in the specific culture of the society in question.41 This critique can be illustrated by the six indicators of legitimacy in the field manual mentioned above. Concepts such as popular selection of leaders and ‘high level of participation in political processes’ are typically based on Western liberal democratic ideas. They are relatively irrelevant in most of the places in which Western militaries have fought insurgencies over the past decades. Although the manual mentions different forms of legitimacy, this is not sufficiently covered because it only spends a paragraph on explaining different forms of legitimacy. If different cultures and societies have different forms of legitimacy and subsequently different ways of achieving legitimacy, this is of paramount importance to the manual as it makes legitimacy the primary objective in COIN. In addition, imposing Western norms and values on societies may not only be difficult in practice, but may in fact be counter-­productive, as it can offend the locals and thus serve to delegitimize the counterinsurgents or upset traditional power balances that can lead to more instability. Moreover, the field manual assumes that the population will accept the form of central authority as legitimate as long as it provides the population with what they deem are their needs. In this way, FM 3–24 only offers benefits to the society as a method of gaining legitimacy. This rational social contract model, where the state buys legitimacy by providing services, ignores the host of other mechanism through which legitimacy is built and maintained in most societies. Furthermore, such a form of legitimacy may only work as long as the external forces are present and able to meet local expectations. It can thus be seen as an artificial form of legitimacy that may prove unstable when the external involvement eventually comes to an end. If legitimacy is purely built on providing benefits in a conflict situation, a new legitimacy system based on traditional norms and values may resurface when violence fades. Another problem is the way the military forces have gone about building this legitimacy. There are many examples when

Revising COIN: The Stakeholder-centric Approach

201

winning ‘hearts and minds’ has meant handing out toys to kids or conducting ‘quick impact’ programmes, aimed at short-­term popularity boosts and force protection, but which simultaneously have undermined long-­term development programmes.42 It is also unlikely that it has contributed to the standing of local authorities among the population. Another problem with the focus on legitimacy is the problem it creates for the intervening forces to actually enhance or create legitimacy for the host nation government or the local people. The perceived legitimacy of the intervening force – a function of the intervening force’s conduct, identity and ability to meet local expectations – is thus vital to COIN.43 If the external forces that are supposed to gain legitimacy for the host nation government are seen as illegitimate in the eyes of the population, the whole task may prove to be futile. Furthermore, time is rarely on the intervening actors’ side. Eventual popular support from the local populace tends to be reduced over time even in the most permissive environments. If this is the case, building legitimacy for a third party may be extremely challenging. If the host nation government is seen as working with the illegitimate intervening forces, they will also be deemed illegitimate and thus the only way to achieve some sort of order will be through the use of overwhelming coercion, which the field manual explicitly considers to be unstable. Second, it is claimed that the field manual is too reliant on a central state as a basis for governing the society.44 In relation to Afghanistan, a country in which legitimacy rarely has been centralized, the critics claim that the government-­ centric idea of legitimacy is entirely flawed. Instead, legitimacy flows from religion, ethnicity, clan and tribe and other forms of local allegiance, something that has largely been ignored in the field manual. Hence, it is claimed that a bottom-­up approach focused on local governance rather than the central government will have a better chance of succeeding. This approach agrees with the field manual that legitimacy should be the main objective, but rather than focusing on building legitimacy for the central government in the eyes of the population, the focus should be on getting the population on your side by providing them with local level governance that is deemed acceptable to them. To the proponents of this approach, COIN can work in the absence of a legitimate host nation central government as long as the local authorities are deemed legitimate. In Afghanistan, for instance, the central government is seen as corrupt and incapable of providing the population with security and essential services on the local level, making it impossible to gain the legitimacy needed for success according to the field manual.45

202

War and State-Building in Afghanistan

In addition, the general population may be largely irrelevant in order to achieve stability as a result of legitimacy. ‘If counter-­insurgency is, in the final analysis, about which side has the greatest legitimacy, then we cannot simply measure that legitimacy as a function of political recognition by the majority of the population.’46 In societies where the central government has based its rule on coercion or there has been no central government, allegiances to leaders of a tribe, clan, ethnicity or religion are likely to be a much stronger foundation of legitimacy. If legitimacy flows from other allegiances, this has to be taken into account and different approaches to achieving this must be examined.

Enemy-­centric approach to COIN In general, the enemy-­centric group is critical of COIN due to what they see as an over-­ambitious strategy to build states and re-­engineer entire societies to achieve the political objectives. US Army Colonel Gian P. Gentile, for instance, argues that COIN has become such a dominant way of thinking in the American military that they do not see any other, more limited ways of dealing with instability and insurgencies, leading the US Army into never-­ending campaigns of nation-­building and attempts to change entire societies to achieve the loyalty of populations.47 This, the critics claim, is too costly in both blood and treasure and has achieved very little success over the past decade. Moreover, they attack what they see as a very narrow and flawed understanding of war and warfare, claiming that the field manual’s view of insurgencies as caused by bad governance is not necessarily accurate and may indeed not be the cause of many insurgencies. Limiting the understanding of causes of insurgencies to bad governance runs the risk of neglecting that the conflict may be a result of other factors such as ethnic antagonisms, ideological disputes, old-­fashioned power struggles or simple greed, and that it may be that the real challenge comes from the adversary and not from the inability to provide the population with certain services. Indeed, if history is an indicator, successful COIN campaigns have rarely been won by ‘out-­administering’ the insurgents, but by outfighting them.48 The proponents of enemy-­centric COIN argue for a more narrow approach that focuses on the insurgents rather than the insurgency, the enemy rather than the population. This they claim is not only the most cost-­effective way of conducting a counterinsurgency operation, but also the one that is most likely to bring success. For them, legitimacy is thus not the means to stability as the population-­centrists would argue, but rather a by-­product or consequence of killing the insurgents in the first place. Also, unless the government can

Revising COIN: The Stakeholder-centric Approach

203

demonstrate the ability to secure and control its population, well-­meaning efforts to appear legitimate are likely to fail because security is of primary concern to the population. As such, stability does not flow from legitimacy, but the other way round. A focus on the causes of violence rather than the causes of the insurgency allows the enemy-­centric approach to argue for a more limited end-­state. Thus stability becomes an end in itself, rather than a way to an end-­ state and subsequently that reform of governance and societal re-­engineering should be processes separate from COIN. To be fair, the FM 3–24 gives consideration to the idea that the ability of the state to provide security to its population can give it enough legitimacy to govern in the people’s eyes. However, the field manual explicitly states that coercive states are inherently unstable in the long run and thus stability based on coercion may only be short-­lived.

Critique of the use of military force in COIN In his seminal book The Logic of Violence in Civil War, Stathis Kalyvas argues that people, irrespective of their pre-­war sympathies, ‘prefer to collaborate with the political actor that best guarantees their survival rather than defect by helping the rival actor’. In war-­torn societies where the population’s primary concern is security, control is likely to shape collaboration because political actors who enjoy substantial territorial control can protect civilians in that territory, giving survival-­oriented civilians a strong incentive of collaboration irrespective of their initial preferences. As such, military resources generally trump pre-­war political and social support in spawning control. This means that collaboration is largely endogenous of control and that the two are self-­ reinforcing, as more collaboration leads to greater control and so on. Through control, political actors try to shape popular support and deter collaboration with their rivals.49 However, if collaboration is part of control, the question of how to gain control in the first place arises. Kalyvas’s insight that support follows strength is important for both proponents and critics of COIN.50 Kilcullen, for instance, uses this idea to argue for a theory of competitive control: ‘whoever does better at establishing a resilient system of control, that gives people order and a sense of security where they sleep, is likely to gain their support and ultimately win the competition for government’.51 By protecting the population under rule of law, one will increase collaboration and deter defection and ultimately win the competition for governing the people. However, Kalyvas argues, ‘the military resources that are

204

War and State-Building in Afghanistan

necessary for the imposition of control are staggering and, hence usually lacking, (. . .) and rival actors are therefore left with little choice but to use violence as a means to shape collaboration’.52 This argument is often used by proponents of the enemy-­centric approach to contend that even if we had the resources required to protect the population in a COIN operation, or more importantly the political will to use such resources, it would be too costly and thus to be avoided. Thus, we should not conduct such operations by focusing on passively protecting the population to increase collaboration, but rather actively focus on the insurgents,‘killing your way to control’. As William F. Owen argues, protecting the population ‘should not be the activity, but should be the benefit from destroying the enemy’.53 Therefore, the argument goes, going after the enemy will showcase the strength of our forces to the population and thereby secure collaboration as the population understands that we can best guarantee their survival. Earlier studies have shown that targeting the insurgency or a terrorist organization’s leadership through kill-­or-capture missions has little or even negative effect. However, more recent studies seem to contradict this, showing that removing insurgent leaders increases governments’ chances of defeating insurgencies because they increase the mortality rates of the insurgent groups when experienced commanders are lost, leading to fewer insurgent attacks, and reducing overall levels of violence.54 These arguments have some merit particularly when targeting the military leaders of the insurgency. It takes years of combat experience to produce an effective military commander but it takes only limited training for a foot soldier to do simple insurgency work. Led by an experienced commander these foot soldiers can operate as a fairly competent combat unit, while they might have close to no combat effectiveness with an inexperienced leader. In addition to the effect of diminished combat effectiveness there is also the possible effect of ‘support following strength’. When the population see that the COIN forces are able to eliminate core insurgent commanders and thereby reduce the overall levels of violence, some individuals will be inclined to collaborate with the counterinsurgents, rather than the insurgents. The problem with this enemy-­centric approach, however, as Kalyvas points out, is that the effective use of violence to establish control is highly dependent on applying force selectively. ‘Indiscriminate violence is of limited value since it decreases the opportunity costs of collaboration with the rival actor’,55 thereby providing a reason for the population to passively or actively support the other side. The so-­called ‘night-­raids’ in Afghanistan are illustrative of the unintended effects of kill-­and-capture operations. Despite their obvious success in hunting

Revising COIN: The Stakeholder-centric Approach

205

down and killing or capturing insurgents (about 1,500 insurgents killed or captured by early 2011 and 80 per cent of the night-­raids were conducted without firing a shot), the raids enraged neighbours and locals.56 According to the Pasthun social code, the Pasthunwali, one who intrudes on a man’s property uninvited is doing so either to rob or to dishonour the person, and Pashtuns are obliged to come to that person’s assistance. On some occasions, neighbours rushing to the scene to help have been shot by Special Operations Forces (SOFs) because they were suspected of being fellow Taliban. The embarrassment and perceived humiliation has probably driven several locals into the Taliban camp. This negative effect of the night-­raids may have been the trigger behind the spring 2012 agreement between the Karzai Government and the US that all night-­raids in the future should either be led or approved by Afghan authorities. The dilemma is that selective violence is dependent on information from the populace in order to capture or kill the insurgents, something that is difficult to obtain if one does not have control. Individuals only want to provide information when it is safe for them to do so, as they wish to maximize their chances of survival. The paradox is that political actors do not need to use violence in areas where they have control, and cannot use selective violence in areas of no control, having no or limited access to information.57 This dilemma makes the enemy-­ centric approach difficult to operationalize. Although both the proponents and critics of COIN agree that control is vital, they do not agree on how to achieve this. A compromise of the two approaches, and one that is gaining increasing support within the US establishment, is what has been dubbed ‘COIN lite’.58 This approach to counterinsurgency is more limited, as its focus is on stability and it does not contain a state-­building component. Within such a hybrid approach, the focus is on offensive operations by SOFs against insurgents while regular forces primarily undertake Foreign Internal Defence (FID), training and mentoring of the host nation’s own security forces. The idea is to let the host nation government be responsible for the protection of its population while more competent SOFs do the hunting and capturing/killing of insurgents. This, they argue, will provide the host nation with more visibility and ownership of the struggle and reduce the negative effect of foreign troops seen as ‘occupiers’. It also reduces significantly the level of force required by external forces, making the engagement more palatable for the external force’s domestic audience (politicians and population in general), a point that has largely been neglected in the COIN field manual. In addition, it reduces the risk to the regular external forces, thus making it easier for the political leadership in the troop-­contributing countries to stay the course.

206

War and State-Building in Afghanistan

The problem with this approach is that it takes a very long time to undertake FID and can thus only work if one is extremely proactive towards the insurgency and act before the insurgents are able to gain the momentum. This could work if one is able to muster enough political support for early involvement in conflicts, but this is challenging. If a full-­fledged insurgency is underway, a ‘COIN-lite’ approach is less likely to succeed because one would not necessarily be able to fend off the insurgents, risking the need to escalate and thereby be dragged into the conflict with a much larger presence. This is much like what happened in Vietnam. The polarizing debate between the population-­centrists and the enemy-­ centrists may be a result of arguing from two different analytical perspectives. As Kalyvas notes, ‘asking what causes a civil war is not the same as asking what causes violence within a civil war’.59 While the population-­centric group focuses on the causes of the insurgency, and thereby views the solution to the insurgency as reform and strengthening of governance, the enemy-­centric advocate focuses on the causes of the violence, the insurgents, and thus argues for a narrower end-­state, stability. The different starting points for arguing their case leads to different end-­states and the ways and means of achieving these. Clearing up this confusion may be a way to bring the two groups closer and reinvigorate the debate about COIN.

Towards a revised COIN theory: a stakeholder-­centric approach The population-­centric versus enemy-­centric debate is deeply polarizing and has led to a stalemate that is hampering any intellectual progress on how to counter insurgencies. While the enemy-­centric promoters have raised a lot of good arguments about the problems with the field manual and COIN theory in general, relying primarily on going after the enemy is a very narrow approach that favours short-­term gains over long-­term efforts to secure a durable peace that is necessary for long-­term stability. If the aim of all wars is a ‘better peace’, then, as Beatrice Heuser argues, a Clausewitzian brutal imposition of one’s will upon the enemy is ‘unlikely to lead to a lasting peace, unless the enemy is annihilated . . . A peace with which the defeated side cannot live in the long term will necessarily engender a new war to reverse the situation.’60 Since annihilation of the enemy is not a realistic option for Western governments, due to moral considerations, a negotiated peace that

Revising COIN: The Stakeholder-centric Approach

207

all the belligerents can live with is, in most cases, the best solution one can get out of a COIN operation. Gorka and Kilcullen support this view in their study of numerous different insurgencies, noting that a government usually wins if it is eventually prepared to negotiate with its non-­state enemy.61 An intervening force needs to consider the social cohesion of a state and what can plausibly be constructed from the old order. The more sweeping the destruction of the existing order (short of total annihilation of one side) and the more fragmented a society gets, the more difficult the establishment of domestic order is likely to be. Although a negotiated solution with insurgents and others may not be in line with the norms and values of Western liberal democracies, such an outcome is most likely to benefit the population because it may end the violence more quickly than a legitimacy or war-­fighting contest. If the population-­centric approach is too ambitious and the enemy-­centric approach unsustainable, what would then be the solution for future COIN operations? How can we find a middle ground that does not require unrealistically high political investments in resources for military forces, development aid and long-­term state-­building efforts, while simultaneously recognizing that security and peace require a certain degree of political legitimacy to be sustainable? While both the proponents and critics of COIN have their merits, they do not disentangle the difficult question on how to build peace after war, which should be of paramount importance in any counterinsurgency. In other words, what is it that keeps the weapons silent after the secession of hostilities? To answer this, we turn to the peace-­building/state-­building literature. Insights from this literature have tended to be neglected by COIN research, but it draws on experiences from many conflicts and may offer some relevant clues, even if not explicitly addressing counterinsurgencies. The question of how to build peace after war is after all a key component in any kind of warfare. Just as in COIN, state-­building is regarded as a key element for success. This has emerged over the last decades so that state-­building has become an integral part of peace-­building. When the UN revised its approach to peace-­keeping and peace-­building in 2001, Kofi Annan, then Secretary General of the UN, stated that peace becomes sustainable ‘not when all conflicts are removed from society, but when the natural conflicts of society can be resolved peacefully through the exercise of state sovereignty and, generally, participatory governance’.62 In the academic literature, state-­building is considered to be a particular approach to peace-­building, ‘premised on the recognition that achieving security and development in societies emerging from civil war partly depends on the existence of capable, autonomous and legitimate government institutions’.63 This is often

208

War and State-Building in Afghanistan

labelled a ‘liberal-­peace’, in other words that a liberal economy and political system are preconditions for lasting peace. Simply put, through representative institutions and a free economy, conflicts are expected to be resolved peacefully. Building these institutions thus becomes a central tenet of peace-­building. However, the literature is critical of the merits of much of the liberal state-­ building efforts, due to mixed results, the tendency to create ‘neo-­imperial’ relationships and ‘cultures of dependencies’, as well as an inclination to ‘one-­sizefits-­all’ and Western-­based ‘templates’ to good governance and institution-­ building.64 Furthermore, while democracy may be regarded as the most stable way of governance, the process of democratization has often turned out to destabilize fragile peace agreements instead of cementing them.65 This is a vast literature, but in the following we will primarily draw on two texts relevant for the current discussion, Mats Berdal’s Building Peace after War and Alex de Waal’s Mission Without End? Peacekeeping in the African Political Marketplace.66 To begin, there are often clear similarities between the critique leveraged against COIN and critique against peace-­building, such as the Western state-­template that is being promoted. Berdal argues that approaches to peace-­ building have displayed a marked tendency to abstract the task of peace-­building from their political, cultural and historical context . . . The result has been an ahistorical and static view of the challenges posed to outside intervention in war-­torn societies and a consequent failure to take account of the variety of ways in which the past constrains, shapes and imposes limits on what outsiders can realistically achieve. This tendency has encouraged a social-­ engineering approach to the concept of peace-­building. External actors have failed to gauge the extent to which their own actions, policies and historical baggage necessarily contribute to shaping the ‘post-­conflict environment’, whether through the stirring of nationalisms or through the legitimization or delegitimization of indigenous power structures, or by empowering or disempowering what are, for better or worse, key local actors.67

By switching the word ‘peace-­building’ with ‘COIN’ and placing this text in the context of Afghanistan, one gets a pretty good picture of what has gone wrong in our efforts to stabilize and bring peace to the country. Furthermore, Berdal presents three priority tasks for an outside intervening force – a secure environment, stabilization of governance structures and the provision of basic services – all of which are very similar to those advocated in the COIN field manual. Additionally, Berdal argues, the driving force behind

Revising COIN: The Stakeholder-centric Approach

209

these activities should always be the building of legitimacy, both for the intervening forces and for the administrative and governance structures on which a durable peace depends. However, it is the lack of an understanding of the local context mentioned above that has ‘too often doomed peace-­building endeavours to ineffectiveness’.68 According to Berdal, if there is one overarching lesson from the post-­conflict interventions in the 1990s it is that stability cannot be imposed on war-­torn societies from the outside. This is recognized in the COIN field manual as well, which states that, ‘in the end, the host nation has to win on its own’.69 This is mainly due to two factors: the limited political will of intervening forces for an open-­ended commitment that any attempt to impose durable peace would require and, more importantly, limits to what can be imposed from the outside. Stability, Berdal argues, has to be elicited, and the key to this lies in the notion of legitimacy.70 For Berdal, it is vital that the governance structures put in place and promoted by the external forces command legitimacy in the eyes of the local parties, neighbouring states and the wider international community.71 Of these, local and regional legitimacy is of primary importance. The international community’s tendency to focus on central governments, creating power-­sharing mechanisms in the capitals combined with a ‘social engineering’ approach, has at times ignored local power structures and tensions and overlooked potential alternative paths to peace. Institutional and governmental models that may appear legitimate and just from the outside may not be regarded so by those impacted by it. As pointed out by many analysts of peace-­building, domestic legitimacy is crucial.72 Another aspect Berdal points to is that interveners often conclude that an absence of (central) government implies absence of governance. However, local forms of governance may very well be in place, notwithstanding the presence of formal government institutions.73 The very notion of a ‘failed state’, as it is often referred to in Western media and academia, presupposes a ‘state-­template’ or a ‘functioning state’ to contrast it with. This is usually defined as a state with a potent central government, basic services and institutions, and monopoly of the use of force – in other words, a Western-­style Weberian state.74 The question is whether intervening state-­builders or COIN fighters accord sufficient importance to these local governing structures when looking for political end-­states. Political structures that emerge from existing forms of governance, rather than being imposed from outside, are more likely to be regarded as legitimate and thereby last longer. This is Alex de Waal’s starting point when he criticizes the very idea of functioning state institutions as the core of peace-­building. He questions whether

210

War and State-Building in Afghanistan

state-­building in the Weberian sense is the right remedy for war-­torn societies with limited historical experience with centralized states. He points out that ‘many of the world’s most difficult conflicts occur in countries where any such state institutions are subordinate to social affinities and patronage networks, and are likely to remain so for the foreseeable future’.75 Wars in these places are not between hierarchically organized armies or groups, but loosely connected groups, held together through systems of loyalties and trade-­offs. Both rebels and government are unlikely to be very disciplined or coordinated, but rather to ‘operate in the same way: using kinship and patronage, and licensing proxies’.76 The key for any political solution to such conflicts lies in these various relationships and their fluctuating evolution. De Waal describes this as a ‘patrimonial marketplace’, governed by socio-­cultural rules: In the patrimonial political marketplace, the only semi-­stable outcome is an inclusive buy-­in of all elites by the best-­resourced actor in the marketplace. Military victories are rarely decisive. More often, members of the losing side quickly negotiate a lower price for their loyalty. The best outcome falls short of stability because all loyalties are provisional pending shifts in the value of allegiances in the political marketplace. It follows that a successful international peace engagement will be one that supports the most inclusive and robust buy-­in – one that is sufficiently well grounded in the relative value of the parties to survive the withdrawal of its international sponsors.

The term ‘marketplace’ thus turns politics into a trade, where loyalty and legitimacy are fragile and rest on various forms of balance of power and rational interests. This model applies both between regional leaders and ‘their’ population and between regional leaders and the central authorities. ‘Political life can be described as an auction of loyalties in which provincial elites seek to extract from one or other metropolitan centre the best price for their allegiance.’77 In this model, legitimacy is vested in the relevant stakeholders or power brokers, not every individual citizen of the state. There are no notions of popular support of a central government, or loyalty emerging out of the provisions of government services, as in the COIN field manual. On the contrary, de Waal argues that ‘in a weakly institutionalized country in which patrimony rules, any attempts to address supposed root causes such as injustice, lack of liberal democracy and unequal development may not help – or may even hinder –the achievement of more modest but realizable goals based on elite bargains’.78 According to de Waal, the basic rules of political bargaining are simple: ‘Provincial elite members seek to maximize the price they can obtain for their

Revising COIN: The Stakeholder-centric Approach

211

loyalty from metropolitan elites (mostly governments) . . . using the tools at their disposal, which include votes, extending or withdrawing economic cooperation, and the use of violence.’79 Using examples from, among others, Tanzania, Nigeria, Sudan and the Democratic Republic of Congo, de Waal argues that such bargaining is the most common way of resolving conflict in these societies. Since the political environment is so fluid in a marketplace, stability cannot rest on formal institutions or justice systems. Loyalties shift, so ‘peace must be made and kept on a retail basis’.80 A crucial insight from these cases is that foreign-­brokered peace accords may be less durable than a purely domestic one: ‘In a purely domestic bargaining exercise, the parties will approximate their true respective values and agree a price which reflects that.’81 This is not a ‘give war a chance’ kind of argument, that wars should be allowed to conclude by themselves and the subsequent fatigue and exhaustion will be the best platform for sustainable peace.82 De Waal argues in favour of political solutions based on a trade-­offs and legitimacy, not merely war-­exhaustion. However, the presence and engagement by outsiders will always alter the local power-­balance and distort the local political marketplace. As a result, the exit of the foreigners – or the expectation thereof – will create waves of re-­positioning and bargaining. Less is therefore more when it comes to foreign engagement, and for the armed forces it is a good argument for keeping a distance from local politics – but never ignoring it or being unaware of it. The mere presence of foreign troops will impact on the marketplace, and the troops need to be aware of how, but that does not entail explicit engagement in local brokering. As Kalyvas argues, reducing violence requires as much local action as action at the centre. At least in the short and medium term, tinkering with local control could be a more efficient way to achieve peace and stability than investing in mass attitudinal shift . . . The allocation of troops and, especially, administrative resources should be based on a clear understanding of the local balance of control.83

Recognizing that military victories are unlikely to be decisive in such societies, de Waal provides us with an approach that retains the crucial element of legitimacy recognized by COIN. However, instead of seeking to build legitimacy of the political system from every single individual (‘the population’), he focuses on the relevant stakeholders in the marketplace. And instead of building legitimacy through government structures and provision of services, he emphasizes the power-­relationships between the stakeholders in the political marketplace.

212

War and State-Building in Afghanistan

Conclusion By analysing COIN and its critics, this chapter has sought to develop a revised theory of COIN to better inform future stability and counterinsurgency operations. In order to overcome the polarizing debate between the population-­ centric approach and the enemy-­centric approach, our aim was to find a middle-­ ground between them. Through the insights offered from the peace-­building literature, we argue that a focus on stakeholders rather than the population or the enemy in a COIN operation is more likely to succeed in bringing long-­term peace and stability to war-­torn countries. It shares with the population-­centric approach the recognition of political legitimacy for sustainable peace, but not that it should stem from the wider population. Furthermore, it shares with the enemy-­centric approach that intervening forces’ primary focus should be on the insurgents and not the insurgency, but that this is not a narrow focus on defeating the enemy, but rather on the use of military force in order to create the conditions that allow for a stable political order. Although the intervening forces should primarily play a military role and not be state-­builders, we argue that conclusive results cannot be achieved through military means alone, and that a negotiated solution to bring stability is the best way to ensure a durable peace. This means that the focus of all COIN operations should be on what comes after the end of violence. Although this approach offers no panacea on how to go about countering insurgencies, some key insights have been developed. Of primary importance is the need to understand that all insurgencies are sui generis – of its own kind – and each one is filled with different incentives and disincentives for the continuation of violence. This means that there can be no ‘one-­size-fits-­all’ approach to counterinsurgencies and that a focus on both the causes of the insurgency and the drivers of the insurgents are vital for understanding the nature of the insurgency. Such an understanding is paramount to inform the counterinsurgents on how to create the conditions that will allow for a peace process. Understanding the nature of the insurgency and providing a solution to countering an insurgency is dependent upon what questions one asks. Depending on whether one focuses on the causes of an insurgency or what causes the violence, one will get very different answers that lead to different end-­states and the ways and means of achieving these. The lack of a common starting point for debating COIN has, in many respects, led to stalemate in furthering our understanding on how to counter an insurgency. Clarifying what can reasonably be achieved with the available resources is in many respects a good start.

Revising COIN: The Stakeholder-centric Approach

213

While agreeing with the enemy-­centric approach that the proposed end-­state of the population-­centric approach is too ambitious, its narrow focus of defeating the enemy to achieve the limited end-­state of stability will most likely not produce a lasting peace. However, we agree that the focus of a counterinsurgency operation should be to facilitate a lasting stability and not a legitimate host nation government able to secure and govern itself. As such, we propose an end-­ state that is: a political agreement between the main stakeholders in the conflict that is regarded as legitimate and ensures stability that is acceptable to all.84 The goal is to enable a political process that leads to an agreement between the main stakeholders that will allow the external forces to withdraw. Although this may appear a very limited objective, simply focusing on ending hostilities and not aiming to reform the governance of society, it is paramount for an intervening force to consider the social cohesion of a state and what can plausibly be constructed from the old order. The more sweeping the destruction of the existing order, the more likely it is that that society will become fragmented, and subsequently, the more difficult it will be to establish domestic order. Thus, this approach to COIN is aimed at consolidating the different factions in the conflict rather than an extensive re-­engineering of society. A COIN operation should therefore be stakeholder-­centric, meaning that the focus of the effort should be on all the relevant military, political, social, religious (etc.) stakeholders in the society that may impact on a future political agreement. This will help in shoring up the legitimacy of the political agreement while it does not require a full-­scale COIN operation aimed at protecting the population and reforming governance. Building legitimacy through a political process is thus the way to achieve the ends in a stakeholder-­centric approach. In this sense, legitimacy has to be thought of as what the stakeholders would most likely support or accept based on their standing in that particular society. Basing the political process on de Waal’s local political marketplace rather than a top-­down imposed negotiation from the centre makes this process more legitimate providing a better chance of stability over the long term. In stakeholder-­centric COIN, the military objective is not limited to protecting the population or defeating the enemy but facilitating a political process, adapted to the local political marketplace, which is deemed legitimate by all parties to the conflict. This means that the intervening force’s military objective is to stop violent conflict and create the conditions for a political process. This is based on the argument that an intervening force can neither protect the population nor achieve unconditional surrender from the warring parties. Thus, a negotiated solution is the best one can hope for.

214

War and State-Building in Afghanistan

Following on from Kalyvas, we argue that military force is instrumental in influencing the decision-­making calculus of the different stakeholders in order to compel them to enter into negotiation and eventually compromise. Compellence is military speak for the use of threats, or some degree of direct action to induce the opponent into giving up what is desired,85 and create a desirable action. While deterrence, the other side of coercion, is concerned with maintaining the status quo and preventing change, compellence is more or less the opposite, as it desires a change in the status quo or a return to the previously disrupted status quo. Because one is confronted with forces that already have changed the status quo, this is why compellence and not deterrence should be the main idea behind the use of force in a stakeholder-­centric approach to countering insurgencies. In this sense, military force becomes instrumental in changing the balance of power on the battlefield and inducing action that is desirable to the peace process. This change in turn has to be followed up by a concrete plan of negotiation that has to acknowledge the need to offer the stakeholders more than just an opportunity to disarm. Insights from the peace-­ building literature and especially de Waals’ concept of the local political marketplace, makes it possible to argue that such an approach should be the basis of any COIN operation. The problem here is that the balance of power is fragile. This means that an initial response to the insurgency will have to consider its actions carefully. As all belligerents in a conflict should be regarded as stakeholders, only supporting the host nation government against armed opposition would seriously hamper any later efforts at consolidating peace through negotiations. The more you support one side, the less the chance of success. If one goes too far, the host nation government is likely to push for more compromises than the insurgent leaders would have been willing to accept had the situation on the ground been different. If a negotiated solution is made on these terms, it may not be a lasting solution, as the stakeholders will not consider it legitimate. The negotiated solution would have to be something all parties can live with in order for it to be sustainable and survive the withdrawal of external forces. As such, one needs to take into account that the host nation government is part of the host nation stakeholders and that the more you support one side, the less chance of success. Also, as an external intervention into any conflict is based on national interests, one has to understand where one’s own objectives overlap with those of the host nation stakeholders – and where not. A carefully thought out strategy on how to achieve one’s own objectives according to one’s interests is crucial at the start of any involvement.

Revising COIN: The Stakeholder-centric Approach

215

A stakeholder-­centric approach can only work if the intervening forces are able to overcome what Berdal identifies as a lack of understanding of the local context. If one is to influence and compel local stakeholders in a conflict to engage in a peace process, a careful analysis, especially of the incentives and disincentives for violence of all the stakeholders, must be in place. Without a proper appreciation of the political and socio-­cultural context in which military forces are being applied, this approach will be doomed to ineffectiveness. The lack of a clear strategy from the beginning of the American interventions in both Afghanistan and Iraq has led to a number of scathing critiques. The American Way of War has been critiqued for its ‘tactical and apolitical orientation’ as well as neglecting ‘the political and socio-­cultural context’ in which military force was used.86 In relation to what Antulio J. Echevarria II views as a narrow focus on defeating the enemy rather tthan achieving political goals, he argues that: the new American way of war considers . . . post-­conflict operations not as a part of war itself, but something belonging to its aftermath. This unhelpful distinction obscures the fact that the principal condition for strategic success in the wars in Afghanistan and Iraq was the establishment of a political (and to a certain extent an economic) order favourable to the United States. Failure to see the purpose for which a war is fought as part of war itself amounts to treating battle as an end rather than a means.87

The failure to have a clear strategy for consolidating the initial military success into political success in the two interventions contributed to the growing problem of insurgencies in both countries. The COIN field manual – which grew out of the chaos in Afghanistan and especially Iraq – can be seen as a reaction to the failure to see post-­conflict operations as part of war itself, and subsequently as an effort to amend some of this narrow focus by including stability operations and state-­building as part of COIN. However, as the wars have dragged on, the field manual and its implementation have increasingly come under attack for being too ambitious and too costly. Instead, critics have increasingly turned back to the narrow approach of defeating the enemy, once again neglecting the post-­ war phase and overlooking war’s purpose – the creation of a ‘better peace’. Although strategy is difficult to do well,88 it is crucial because it is the cornerstone for connecting political goals with military means. Thus the essence of operational art – translating the political and strategic aims into operational and tactical objectives – in any counterinsurgency operation must be to create the conditions favourable for political order. This means that one must appreciate that

216

War and State-Building in Afghanistan

the establishment of order is central for a strategic victory and must be viewed as part of war itself, and that any use of force must be applied to attain political goals rather than tactical military aims. In a stakeholder-­centric approach, this means that a good strategy must account for a dynamic political context that can easily change when force is applied to an opponent with its own options and goals, whose own behaviour is shaped by local political, social and cultural norms.

Notes 1 Sustaining U.S. Global Leadership: Priorities for 21st Century Defence (Washington, DC: US Department of Defense, 2012), p. 6. 2 Ibid. 3 In the following, the acronym COIN will refer to the doctrine (FM 3–24) and its theory, recommendations and principles. The term counterinsurgency will refer to the phenomenon of countering an insurgency in general. 4 An interim COIN doctrine was issued in October 2004 as Field Manual (Interim) 3–07.22. 5 John Nagl, ‘The Evolution and Importance of the Army/Marine Corps Field Manual 3–24, Counterinsurgency’, U.S. Army and Marine Corps, Foreword, in Counterinsurgency, FM 3–24/MCWP (Chicago, IL: University of Chicago Press, 2007), pp. xv–xvi. 6 Foreword, Counterinsurgency, FM 3–24/MCWP, p. xiv. 7 U.S. Army and Marine Corps, Counterinsurgency, FM 3–24/MCWP (Chicago: University of Chicago Press: 2007), p. 46. 8 FM 3–24, p. ix. 9 FM 3–24, p. 2. 10 FM 3–24, p. 2. 11 Harald Håvoll, ‘COIN Revisited: Lessons of the Classical Literature on CounterInsurgency and its Applicability to the Afghan Hybrid Insurgency’, Oslo: NUPI, Security in Practice, no. 13 (2008), p. 6. 12 Bruce Hoffman, ‘Terrorism Defined’, in R. Howard and R. Sawyer (eds), Terrorism and Counterterrorism (Guilford, CT: McGraw-Hill/Dushkin, 2002), p. 22. 13 FM 3–24, p. 2. 14 David Kilcullen, Counterinsurgency (London: Hurst, 2010), p. 1. 15 David Kilcullen, Counterinsurgency, p. 2. 16 Robert B. Polk, ‘Interagency Reform: An Idea Whose Time has Come’, in Joseph R. Cerami and Jay W. Boggs (eds), The Interagency and Counterinsurgency Warfare: Stability, Security, Transition and Reconstruction Roles (Carlisle, PA: Strategic Studies Institute, US Army War College, 2007), p. 319.

Revising COIN: The Stakeholder-centric Approach

217

17 FM 3–24, p. 6. 18 FM 3–24, p. 39. 19 FM 3–24, p. 2. 20 FM 3–24, p. 51. 21 FM 3–24, p. 2. 22 FM 3–24, p. 37. 23 FM 3–24, pp. 37–8. 24 FM 3–24, p. 38. 25 FM 3–24, p. 2. 26 FM 3–24, p. 37. 27 FM 3–24, p. 38. 28 FM 3–24, pp. 174–84. 29 This is sometimes referred to as the ‘ink-­spot strategy’. 30 FM 3–24, p. 175. 31 FM 3–24, p. 39. 32 FM 3–24, p. 54. 33 David Galula, Counterinsurgency Warfare: Theory and Practice (1964, reprint, Westport, CT: Praeger Security International, 2006), p. 63. 34 FM 3–24, p. 34. 35 FM 3–24, p. 55. 36 Bernard B. Fall, ‘The Theory and Practice of Insurgency and Counterinsurgency’, Naval War College Review, vol. 1, no. 1 (Winter 1998, reprinted from April 1965 issue), pp. 53–4 (emphasis in original). 37 FM 3–24, p. 38. 38 FM 3–24, Foreword. 39 For General Stanley McChrystal’s COIN approach to the war in Afghanistan see, for instance, Stanley McChrystal, COMISAF USFOR-A Counterinsurgency Training Guidance, Headquarters USFOR-A/ISAF, Kabul, Afghanistan, Memorandum 10 November 2009. 40 Sebastian L.v. Gorka and David Kilcullen, ‘An Actor-­centric Theory of War: Understanding the Difference between COIN and Counterinsurgency’, Joint Forces Quarterly, issue 60, 1st quarter (2011), p. 16. 41 See, for instance, Robert Egnell, ‘Winning “Hearts and Minds”? A Critical Analysis of Counter-Insurgency Operations in Afghanistan’, Civil Wars, vol. 12, no. 3 (2010), pp. 282–303. 42 Andrew Wilder and Gordon Stuart, ‘Money Can’t Buy America Love’, Foreign Policy (1 December 2009) [http://www.foreignpolicy.com/articles/2009/12/01/money_ cant_buy_america_love?hidecomments=yes; accessed 25 July 2012]. 43 Mats Berdal, Building Peace after War (London: IISS, Routledge, 2009), p. 98.

218

War and State-Building in Afghanistan

44 See, for instance, William Rosenau, ‘Counterinsurgency: Lessons from Iraq and Afghanistan’, Harvard International Review, vol. 31, no.1 (Spring 2009), pp. 52–6. 45 David C. Ellis and James Sisco, ‘Implementing COIN Doctrine in the Absence of a Legitimate State’, Small Wars Journal (13 October 2010) [http://smallwarsjournal. com/jrnl/art/implementing-­coin-doctrine-­in-the-­absence-of-­a-legitimate-­state; accessed 25 July 2012]. 46 Gorka and Kilcullen, ‘An Actor-­centric Theory of War: Understanding the Difference between COIN and Counterinsurgency’, p. 17. 47 Gian P. Gentile, ‘A Strategy of Tactics: Population-­centric COIN and the Army’, Parameters (Autumn 2009), pp. 5–17. 48 Bernard Finel, ‘A Substitute for Victory’, Foreign Affairs (8 April 2010) [http://www. foreignaffairs.com/articles/66189/bernard-­finel/a-­substitute-for-­victory; accessed 25 July 2012]. 49 Stathis N. Kalyvas, The Logic of Violence in Civil War (Cambridge: Cambridge University Press, 2005), pp. 12–13. 50 For an enemy-­centric view, see William F. Owen, ‘Killing Your Way to Control’, The British Army Review, no. 151 (Spring 2011), pp. 34–7. For a more population-­centric view, see Kilcullen, Counterinsurgency, pp. 152–4. 51 Kilcullen, Counterinsurgency, p. 152. 52 Kalyvas, The Logic of Violence in Civil War, p. 12. 53 Owen, ‘Killing Your Way to Control’, p. 151. 54 B. Price, ‘Targeting Top Terrorists’, International Security, vol. 36, no. 4 (Spring 2012), pp. 9–46. 55 Kalyvas, The Logic of Violence in Civil War, p. 144. 56 Jonathan Smith, ‘We Own the Night’, Small Wars Journal (22 February 2012) [http:// smallwarsjournal.com/jrnl/art/we-­own-the-­night; accessed 25 July 2012]. 57 Kalyvas, The Logic of Violence in Civil War, p. 13. 58 Crispin Burke, ‘Like it or Not, Small Wars will Always be Around’, World Politics Review (24 January 2012) [http://www.worldpoliticsreview.com/articles/11238/ like-­it-or-­not-small-­wars-will-­always-be-­around; accessed 25 July 2012], and O. Manea and J. Nagl, ‘COIN is not Dead’, Small Wars Journal (6 February 2012) [http:// smallwarsjournal.com/jrnl/art/coin-­is-not-­dead; accessed 25 July 2012]. 59 Kalyvas, The Logic of Violence in Civil War, p. 392. 60 Beatrice Heuser, ‘Ends, Ways, Means: Clausewitz and Other Prophets’, Infinity Journal Special Edition, Clausewitz and Contemporary Conflict (February 2012), p. 18 [http://www.infinityjournal.com/article/48/Ends_Ways_Means_Clausewitz_ and_Other_Prophets; accessed 25 January 2012]. 61 Gorka and Kilcullen,  ‘An Actor-­centric Theory of War: Understanding the Difference between COIN and Counterinsurgency’, p. 17.

Revising COIN: The Stakeholder-centric Approach

219

62 ‘No Exit without Strategy: Security Council Decision-­making and the Closure or Transition of United Nations Peacekeeping Operations’, Report of the Secretary General, UN S/2001/391, New York (20 April 2001), p. 2. 63 Roland Paris and Timothy D. Sisk, ‘Introduction: Understanding the Contradictions of Postwar Statebuilding’, in Roland Paris and Timothy D. Sisk, The Dilemmas of Statebuilding: Confronting the Contradictions of Postwar Peace Operations (London: Routledge, 2009), pp. 1–2. 64 Paris and Sisk, ‘Introduction: Understanding the Contradictions of Postwar Statebuilding’, in Paris and Sisk, The Dilemmas of Statebuilding: Confronting the Contradictions of Postwar Peace Operations, pp. 11–14. See also Oliver Richmond and Jason Franks, Liberal Peace Transitions: Between Statebuilding and Peacebuilding (Edinburgh: Edinburgh University Press, 2009); Michael Pugh, Neil Cooper and Mandy Turner, Whose Peace? Critical Perspectives on the Political Economy of Peacebuilding (New York: Palgrave Macmillan, 2011). 65 Roland Paris, At War’s End: Building Peace after Civil Conflict (Cambridge: Cambridge University Press, 2004). 66 Alex de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, International Affairs, vol. 85, no. 1 (2009), pp. 99–113. 67 Berdal, Building Peace after War, p. 19. 68 Berdal, Building Peace after War, p. 25. 69 FM 3–24, p. 47. 70 Berdal, Building Peace after War, p. 97. 71 Berdal, Building Peace after War, pp. 98–100. 72 Richmond and Franks, Liberal Peace Transitions: Between Statebuilding and Peacebuilding. 73 Berdal, Building Peace after War, pp. 123–4. 74 See, for example, the ‘Failed State Index’, of the US Fund for Peace [http://www. fundforpeace.org]. 75 de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, p. 99. 76 de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, p. 101. 77 de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, p. 103. 78 de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, p. 110. 79 de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, pp. 103–4. 80 de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, p. 110.

220

War and State-Building in Afghanistan

81 de Waal, ‘Mission without End? Peacekeeping in the African Political Marketplace’, p. 109. 82 Edward N. Luttwak, ‘Give War a Chance’, Foreign Affairs, vol. 78, no. 4 (1999), pp. 36–44. 83 Kalyvas, The Logic of Violence in Civil War, p. 391. 84 While the term all depends on the context in which the counterinsurgency takes place, it necessarily includes the local, regional and international actors that have a stake in the conflict. 85 Thomas C. Schelling, Arms and Influence (New Haven, CT: Yale University Press, 2008), p. 79. 86 Frank G. Hoffman, ‘Re-Assessing the American Way of War’, Orbis (Summer 2011), p. 525. 87 Antulio J. Echevarria II, Toward an American Way of War (Carlisle, PA: Strategic Studies Institute, US Army War College, 2004). 88 Colin S. Gray, ‘Why Strategy is Difficult’, Joint Forces Quarterly, issue 22 (Summer 1999), pp. 80–6.

8

The Country as a Whole: Imagined States and the Failure of COIN in Afghanistan Ivan Arreguín-Toft

Introduction The place we think of as Afghanistan has been riven by insurgency for much of its existence. In this chapter, I argue that contemporary COIN in Afghanistan has been a problem because Afghanistan is an ‘imagined state’ rather than a real one. Afghanistan, Somalia and the Democratic Republic of Congo are but three contemporary examples of places which once may have possessed a critical mass of indigenous citizens desirous of community within a state, but where there is no longer sufficient indigenous demand for a state as such. This is partly due to the same developments that have worked to escalate terrorism – at least in the lay mind – from an annoying law enforcement problem to a vital national security issue. More on this below. The ISAF’s COIN efforts in Afghanistan, which began in earnest in 2006, have enjoyed increasing micro success – success in specific valleys or provinces in Afghanistan – but at a cost in macro success. This matters for two reasons. First, macro success – the rehabilitation of an Afghan state (much less one with a democratic government and a viable non-­criminal economy) – was the original political objective of Operation Enduring Freedom (OEF). If a viable and sustainable Afghan state does not emerge following the departure of ISAF in 2013–14, OEF more broadly, and COIN more narrowly, may be said to have failed. Second, macro success – the re-­creation of a state governed from Kabul – remains existentially impossible because the diverse peoples who reside within the territory marked on maps as Afghanistan no longer evidence sufficient demand for a government capable of governing more than a fraction of the territory. This leads to interlinked foreign policy, development and COIN dilemmas not unique to Afghanistan but particularly intense there: because the

222

War and State-Building in Afghanistan

foreign demand for government exceeds the domestic supply (itself an empirical marker of lack of indigenous demand), well-­intentioned foreigners are forced to import their own civil servants1 in order to secure the population and build a non-­criminal economic infrastructure (both necessary to the reconstruction of a state in Afghanistan). However, imported government depends on a continuous flow of resources and legitimacy; and the latter is impossible to secure until the imported government’s personnel exit. This chapter begins by briefly reprising the argument that public servants are the sine qua non of states, showing how as a result of the timing of historical traumas such as major interstate war, civil war and famine, in imagined states ranging from Somalia to Afghanistan, a minimum critical mass of public servants has proven prohibitively difficult to recruit, train and maintain. I then highlight the period following the end of the Second World War to show how in Afghanistan’s case, a long-­suffering and fragile national unity was shattered by superpower rivalry, culminating in the Soviet Union’s assassination of Mohammed Da’ud Khan in 1978, and accelerated by its decade-­long occupation and COIN efforts there. I then advance to a brief history of OEF, and ISAF’s contribution to Balkanization of Afghan identity before concluding with theoretical and policy implications which extend from the Afghan case to a larger class of similar cases.

Arguments The principal argument in this chapter is that Afghanistan is one instance of a broader class of instances which include places marked on maps as states, and which possess many of the markers of a state – passports, government ministries, currencies, and so on – but which lack the sine qua non of the state as such: a monopoly on the legitimate use of violence within an internationally recognized bounded territory. In advancing this argument, I make two key assumptions, which it makes sense to highlight before moving on. First, I assume that Max Weber’s definition of the state (paraphrased above), while admittedly minimalist and possibly masculinist,2 is a useful one. That is, I wish to build a general explanation of a contemporary foreign policy dilemma, and thus it makes sense to invoke a generally shared assumption about what a state is. Most arguments about the state as a form of political association share this metric. For those who share the long tradition of the necessity or at least desirability of the state as such, the general form of the argument is that without

Imagined States and the Failure of COIN

223

a reasonable guarantee of physical security (at a minimum), no other socially valued goods or ends may be pursued. This is where Thomas Hobbes begins his argument in Leviathan: he creates a harrowing vision of life without the state, and then employs the order-­producing capacity of the state to persuade the reader that ‘the state’ and ‘absolute monarchy’ are the same thing.3 The point is that the starting point of Hobbes’s argument rests on the state as order-­producing (where ‘order’ is in turn framed in terms of physical security) because there was (and remains) widespread consensus on the necessity and desirability of the state as a form of political association. Consensus on the quantity and quality of socially valued goods the state should guarantee beyond physical security does not exist. Second, I assume that although social and political institutions possess inertia, their existence depends ultimately on demand. In other words, the supply of any institutional form, including the state, is dependent on a demand for it. I argue that in an increasing number of the world’s states, indigenous demand for a state has dropped below the point at which even habit and tradition can sustain the state as such. This is the chief underlying cause of what political scientists over the past two decades have mis-­named ‘failed states’.4 Taking these assumptions into account, the logic of my argument follows rather simply. First, I argue that public servants are vital for states, and that therefore they may serve as a useful empirical indicator of indigenous demand for states. Where demand for a state is sufficient, the supply of public servants will be sufficient, and the state will be real rather than imagined. This should not be taken to imply anything about the quality of the state: its leaders may be few or many, corrupt or honest, tyrannical or consensual. Where demand for the state drops, the supply of public servants drops, and government of the state carries on in name only, or fractures into micro states. Second, I argue that the now widespread prohibition against war as an advance of policy by other means has had the unintended consequence of eroding indigenous demand for the state as a form of political association. Third, extra-­territorial demand for the state and the state system remains high, but cannot suffice to resolve the problem of imagined states due to nationalism: foreign public servants are foreign first and public servants second; and this renders them odious locally, irrespective of the very real services and benefits they provide. In this sense, it is easier to understand what I mean by ‘the problem of imagined states’. I invoke the term in the sense that Benedict Anderson does in reference to nationalism:

224

War and State-Building in Afghanistan

In an anthropological spirit, then, I propose the following definition of the nation: it is an imagined political community – and imagined as both inherently limited and sovereign. It is imagined because the members of even the smallest nation will never know most of their fellow-­members, meet them, or even hear of them, yet in the minds of each lives the image of their communion . . . The nation is imagined as limited because even the largest of them, encompassing perhaps a billion living human beings, has finite, if elastic, boundaries, beyond which lie other nations. No nation imagines itself as coterminous with mankind . . . It is imagined as sovereign because the concept was born in an age in which Enlightenment and Revolution were destroying the legitimacy of the divinely-­ordained, hierarchical dynastic realm . . . The gauge and emblem of this freedom is the sovereign state. Finally, it is imagined as a community, because, regardless of the actual inequality and exploitation that may prevail in each, the nation is always conceived as a deep, horizontal comradeship. Ultimately it is this fraternity that makes it possible, over the past two centuries, for so many millions of people, not so much to kill, as willingly to die for such limited imaginings.5

Of course, Anderson was writing at a time in which the problematique – to borrow a French construction – surrounds the fact that many people who were living within nations qua ‘imagined communities’, seemed willing to escalate their desire to live in states – other than the ones in which they currently resided – to war. My argument is that following the end of the Cold War, an increasing number of people living within nations or states wished to live in no state, rather than ‘their own’ state. Thus whereas in Anderson’s formulation the imagining gets done by nationals (inside-­out), in my argument the imagining is done by the community (outside-­in). The ‘international community’ – itself an imagining – wants the globe to be entirely quilted with states. But it has effectively outlawed interstate war, and the more interesting question then becomes, ‘do peoples in Afghanistan want to be a state?’ I argue that in most cases they do not. This is why COIN has succeeded in a micro sense but failed in a macro sense. Such is the general form of the argument I wish to advance here. My aim for the present is to take this general argument and apply it to the particular case of COIN in Afghanistan (with a focus on OEF). I should add that I restrict myself to deriving good questions and teasing out their theoretical and policy implications. The equally interesting and crucial collection of empirical data and their analysis must await subsequent treatment.

Imagined States and the Failure of COIN

225

The priority of the ‘state’ in Afghanistan It is useful to remind readers that ‘once upon a time’ there were no states in Europe (to cite but one area of the world where this was true). While the development and contemporary evolution of states in Europe (and elsewhere) remains a subject of vibrant scholarly interest, for our purposes it is sufficient to point out that since the late seventeenth century, the ‘state’ as we think of it today slowly expanded to quilt the globe. The United Nations (which really is a misnomer, since the organization is actually an aggregation of states) currently recognizes 193 sovereign states, which occupy the entire globe’s surface save for Antarctica and parts of the Arctic. The state, in other words, has become the norm; so much so that in the past two decades since the end of the Cold War (and the disintegration of the USSR), we can and do speak of ‘failed’ and ‘failing’ states.6 Again, my argument for why this entire rubric is misleading is expanded upon elsewhere, but the logic may be summarized simply. Essentially, a sine qua non of the state as such is the public servant (where the public is contiguous with the territorial boundary of the polity). Even before the industrial revolution, the increasing complexity and cost of inter-­polity warfare in Europe created irresistible demand for ministers and administrators capable of managing state finances. Ministers made ministries, and states with the most effective administrators were more easily able to turn taxes and other inputs into fighting power; thus securing the state or facilitating its expansion by conquest. The industrial revolution – in particular the advent of the railroad – dramatically accelerated this trend, which culminated in the capture of the state’s leadership by representatives of the middle classes and the destruction (as in France and Russia) or warehousing (as in Britain and the rest of Europe) of aristocrats. This ideal-­type state emerged from a kind of Darwinian ‘survival of the fittest’ competition to become the most lethal form of political association known to human history.7 As Charles Tilly famously put it, ‘war made the state, and the state made war’.8 It spread across the globe along three parallel axes: missionary work, commerce and conquest. Huge and sophisticated African tribes, clans and confederations did not arrive on Europe’s shores and conquer European states, nor did Asian Empires or Latin American civilizations such as Inca, Maya and Aztec. Rather everywhere it was the reverse: even small European states such as Portugal and the Netherlands were able to conquer and (at least initially) control large swathes of territory in Africa,

226

War and State-Building in Afghanistan

Asia and Latin America; even those populated by sophisticated peoples with organized militaries. Europeans brought the institutions of the state (and their own public servants) with them, and demarcated their territories with lines on maps. In this way, the state covered the globe and became a kind of super-­norm: so axiomatic in most minds that departures from either a desire to live in a state or an ability to maintain one are instantly problematized as ‘failure’. It follows that in places in the world in which the state as such is less natural (and inevitable), and where political independence from European colonial domination gave legitimacy to the notion of a return to older forms of association, it is demand for the state as an institution that should be problematized rather than its so-­called failure. Add to this the contemporary observation that the likelihood of interstate war (or in Stephen Pinker’s version, violence itself) is withering away, and we arrive at serious questions about the presumed universal distribution of a demand for states.9 A second problem has less to do with whether people in a given territory want to live in a state as opposed to some other form of political association, such as a clan or tribe10 – a question of willingness – and more to do with the communications costs of state services; the most rudimentary of which is the provision of physical security – a question of capacity. It may well be, for example, that if we ranked all territories by the cost per unit of distance of building a road, we would find a strong correlation between lower costs and the likelihood of a centralized, European-­style state. By contrast, where topography and climate make road building (or other communications, such as port facilities) difficult, we would expect the form of political association to have adapted, and a corresponding dearth of centralized, Euro-­style states. It is worth recalling, by way of example, that Paris has the network of roads and traffic circles it does not out of some aesthetic consideration and not as a natural evolution in demography or commerce, but specifically to aid in the imposition of government control on otherwise costly-­to-reach areas of the city.11 Afghanistan is a place where topography and climate make centralized government very difficult, and where social and political institutions, practices and traditions have evolved to take account of this reality. It remains possible to construct a real state on the territory (it has been done at least once already), but in Afghanistan a series of exogenous shocks over three decades have made the prospect prohibitively costly, especially for foreigners.

Imagined States and the Failure of COIN

227

Public servants in Afghanistan Elsewhere I have advanced the argument that public servants make a reliable empirical indicator of indigenous demand for a state.12 I have also agreed with many others that the intensity of the negative externalities generated by contemporary failed states – narco trafficking and terrorism to cite but two – has risen in the past two decades, offering an interest-­based account of the ‘international community’s’ attempts to reconstruct a centralized state in Afghanistan. It is this external demand for a state (not matched by an internal demand) that underpins my claim that as a state, contemporary Afghanistan is imagined rather than real. What follows is an historical account of the flight of public servants from Afghanistan; offering an explanation for why ISAF’s COIN efforts must fail. This chapter starts with a discussion of the impact of Soviet intervention in Afghanistan after the Second World War and then follows with like summaries of two successive interventions: the Taliban in the 1990s, and NATO and ISAF in the 2000s. It is worth remembering that Afghanistan was in fact a state from about 1928 until the assassination of Mohammed Da’ud Khan and his family by Soviet-­ supported Afghan Marxists in 1978. But the end result of three decades of war and successive foreign occupations has been the devolution of loyalty from the state we call Afghanistan, to intense loyalties at the sub-­state level. The bad news is that this has made NATO’s broad strategy of rebuilding a state existentially impossible because for the time being, there is no state to rebuild (or insufficient indigenous demand). The good news is that historically, other states with complex topography and powerful sub-­national identities have proven able to work together constructively within a larger state, so there is theoretically nothing to prevent Afghanistan from becoming more akin to Switzerland at some point in the future.

The USSR in Afghanistan, 1978–89 By now the history of the Soviet Union’s ill-­fated intervention in the Afghan civil war that followed the murder of Mohammed Da’ud Khan and his family in 1978 is better known.13 However, here I want to focus on three key aspects of that intervention that are most relevant to OEF, and what I have asserted will be the necessary failure of COIN in Afghanistan: (1) the asymmetry of effect which is

228

War and State-Building in Afghanistan

graven into the very topography of Afghanistan makes it very costly to secure communications and concentrations of resources and people, and very cheap to interdict communications and thereby threaten concentrations of resources and people; (2) the strong tendency to techno-­fetishism on the part of intervening polities; and (3) the adaptive but bewildering complexity of patterns of authority and legitimacy in Afghanistan. However, I begin with some background history, starting with a brief summary of the salient events that followed the independence of India from Great Britain in 1947. I will follow this with another brief introduction to the Soviet Union’s military intervention in Afghanistan from 1979 to 1989.

From the Second World War to 1978 Afghanistan’s prior status as a genuine sovereign state had taken generations to engineer: Afghanistan became a state by virtue of a succession of skilful leaders dedicated to maintaining a careful balance between tradition and modernization. Their success forced the centuries-­long patchwork-­quilt identities of tribe and valley to slowly yield to a collective Afghan identity. The violent division of the Indian subcontinent into Pakistan and India following Britain’s departure in 1947, however, provoked the president of Afghanistan’s constitutional monarchy, Sirdar Muhammad Da’ud, into seeking development aid to strengthen his country’s economy and military. When the Eisenhower Administration demurred, the Soviet Union stepped into the vacuum, offering billions of rubles to Afghanistan with ‘no strings attached’. Da’ud was not fooled. He understood that Soviet aid would be a Trojan horse, and he continued to outlaw communist participation in government. In 1963, Afghanistan’s king fired him, and Da’ud’s successors allowed multiple political parties to have a say in Afghanistan’s government. In 1965, the PDPA was formed, and after steadily gaining power in Parliament, in 1976 it split into two factions. This factionalization accelerated, as did opposition to Afghanistan’s constitutional monarchy. In 1973, Da’ud, with Soviet support, staged a bloodless coup and attempted to restore the equilibrium between modernization and tradition. But in doing so, Da’ud sought to diversify aid sourcing (Da’ud alarmed the Soviets by seeking aid from Iran and Saudi Arabia) and slowing the pace of political liberalization (including adding restrictions on communist and Islamist political parties). In April 1977, Brezhnev summoned Da’ud to Moscow for a dressing down. He was bluntly ordered to cease permitting foreign advisors on the grounds that they were NATO spies. Da’ud hotly retorted that the Soviets had no business

Imagined States and the Failure of COIN

229

meddling in Afghanistan’s politics. In July of the same year, the Soviets forced the rival communist factions to unite, and a year later Da’ud and his family were murdered. But the forced unification of the two communist factions – the Parchamis, led by Babrak Kamal, and Khalq, led by Muhammad Taraki – proved unstable following the murder of Da’ud. As one historian who lived through these events reported, infighting between these rival groups (each seeking both to enrich themselves personally and to murder each other) was only one of two problems. The second and bigger problem is that having imprisoned, murdered or lost through flight all of Afghanistan’s public servants, the ruling Marxists now faced the daunting task of attempting to govern with zero government experience and in possession of a template for government that was entirely alien to the social, historical and political context of Afghanistan: each [of the men who replaced Da’ud] was convinced that the PDPA blueprint was the guideline for reorganizing both society and the state. Thus, they relied on Soviet, not Afghan, experience, and thus, too, they broke with the Afghan past. This may explain why, after they rose to power, they became ever more alienated from their own people and ever more disunited among themselves.14

Thus, with the Marxists taking power, all hope of restoring the equilibrium that had made Afghanistan a genuine state disappeared. The process of destroying the old order was only the first stage in what would soon dramatically underline the destruction of Afghan national identity and replace it with the complex and interlocking patchwork of provincial, ethnic, religious and geographical sub-­ loyalties that matter most today. The evolution of Soviet COIN strategy over the next decade aided further destruction of the state.

From 1979 to 1989 Most observers of contemporary Afghanistan are not well versed in the Soviet experience of COIN. After the Second World War, the United States and its NATO allies found the disparagement of Soviet military accomplishments in that war simpler and more politically palatable than the difficult alternative of giving the Soviets their full military due (patient scholarship and the collapse of the USSR in 1991 have led to a reassessment, but the politics of memory still tend to unwisely discount Soviet military effectiveness and innovation; and now do the same with Russian military effectiveness and innovation).

230

War and State-Building in Afghanistan

As a result, the core memory of the Soviet experience surrounds the issue of US covert support of allied Afghan fighters (collectively remembered as mujahideen) through the impossible-­to-seal border with Pakistan; and in particular, the role of Stinger MANPADs in causing the Soviet COIN effort’s ignominious collapse. Both factors played a key role, but that shallow understanding led to two unfortunate consequences that bedevilled early COIN operations against the Taliban in OEF. First, Soviet military failure was attributed mainly to the notion that communists were rigid and inflexible and driven by ideology. In other words, the Soviets lost Afghanistan. This obscured the operational and tactical sophistication of the mujahideen fighters who opposed the Soviets. Second, the victory of the mujahideen was attributed to Stingers (whose effectiveness was not attributed to their skilful use by the mujahideen, but to the superiority of US over Soviet technology). I do not intend here to do justice to the full historical record of Soviet COIN efforts; my aim instead is to call attention to the Soviet COIN effort’s contribution to the destruction of the Afghan state as an empirical reality, and its replacement with a kind of virtual place-­holder; an imagined state. Upon reflection, it should not surprise us that the Soviet Union’s original mission, and the nature and quantity of the forces authorized to support that mission, were limited, in the important sense of limited war. The Soviets did not wish to risk a confrontation with the US or Pakistan, which might inadvertently escalate to a nuclear crisis. More importantly, the forces assigned to Afghanistan were limited because the task at hand was thought to be relatively simple and negative: secure the nascent Marxist government in Kabul from collapse and defeat, and thereby maintain stability on the Soviet Union’s southern flank while protecting Marxist ‘inevitability’ from an embarrassing and potentially delegitimizing defeat. Operationally, the goal was to use mainly third-­echelon forces of the Soviet Union’s 40th Army to secure Kabul, and Afghanistan’s few major communications assets, until the new Democratic Republic of Afghanistan (DRA) could right itself and use its own forces to establish order throughout the country. However, Soviet military doctrine and forces had been designed to preserve Soviet security from a conventional assault by another state. They were heavily mechanized, and included heavy weapons (mortars, artillery and crew-­served machine guns). This created two sets of difficulties that it would take the Soviets years to overcome. First, vehicles require vehicle-­passable terrain (roads), a lot of fuel and spare parts. Heavy weapons (in particular artillery) require heavy ammunition, best transported by rail or truck. All of these would be forced to traverse ill-­

Imagined States and the Failure of COIN

231

developed roads and bridges within easy range of mujahideen soldiers on foot. Tactically, this placed the relatively inexperienced Soviet forces at a lethal disadvantage, but that disadvantage was difficult for them to anticipate in part because the Soviets had something the mujahideen would never have: air support. Second, Soviet doctrine, training, leadership and technology, in tandem with its NATO counterpart, assumed that killing necessarily resulted in victory. Its armed forces were designed, in other words, to kill their way to victory just as they had against the Third Reich in 1945. This made them ill-­suited for COIN in Afghanistan, because killing could not conduce to victory.15 As individuals, Afghanistan’s diverse peoples fear death just as anyone else might, but not above all else. Killing Afghan mujahideen would never convince them to give up, and beyond the threat of death, the Soviets and their DRA clients could offer the mujahideen nothing. None of this was obvious in the first few years, however. What quickly developed was what we now call ‘mission creep’. Bases – whether Kabul itself, supply depots, or any town or site of value – could only be secured by fortification and garrison, and each required large quantities of supplies in order to maintain them. Those supplies, in turn, would have to traverse highly vulnerable lines of communication. Securing those lines of communication would require more combat forces, which were not available because the overall mission had been strictly limited (recall that the intervention was begun on Leonid Brezhnev’s watch, but he died soon after, and was succeeded by two General Secretaries before the accession of Mikhail Gorbachev in 1985). This meant that local Soviet commanders were going to have to improvise, and their first innovation was an obvious one: concentrate your forces on the greatest threat to communications of your most valuable base (Kabul). Once secured, shift available forces to the next greatest threat, and so on, until the threat to your communications is lifted. The logic was sound, and from a military standpoint there was only one other option (which I will come to presently). But, it didn’t work, at least not until very late in the COIN effort. From 1980 to 1985, concentrating forces on a particular valley (say, the Panjsher Valley, which was subject to not one but seven separate assaults over the decade of Soviet support for the DRA) eliminated the threat to communications only so long as operations continued. Once they ceased, the mujahideen would move back in and begin interdicting communications and ambushing bases all over again. The topography of Afghanistan also turned Soviet mobility assumptions on their head: the mujahideen, mostly on foot, and armed with small arms (and portable crew-­served weapons) proved more

232

War and State-Building in Afghanistan

nimble than the Soviets, whose vehicles were forced to stay on roads, and whose only hope of salvation in case of ambush was air support.16 I have recounted the various strategic innovations of both sides, and the vulnerabilities inherent in the Soviet’s growing dependency on air mobile operations elsewhere.17 For our purposes, it is only necessary to point out that due to a structural condition – limited aims and a limited contingent – Soviet forces were not sufficient to provide communications security between key values (Kabul, Bagram air base, and so on). As a result, they tended to concentrate their available forces in spring offensives, each aimed at a different valley or province within Afghanistan (Kunar Valley, Panshjer Valley, and so on). This made military sense, but politically it solidified the tendency of Afghan resistance to identify with the territory on which it fought, rather than Afghanistan as a whole: to paraphrase Charles Tilly’s formulation, Soviet war was making – not a state – but micro states. Each would emerge from Soviet failure and withdrawal with a separate sub-­national identity, complete with a skilled local political leadership (including public servants) capable of maintaining order and security within each respective province or valley. These micro states each came to possess modestly different justice systems too. Following the war, this Balkanized Afghan identity aided the Taliban in conquering the entire country, because unlike the Soviets (who had unified resistance against themselves and, by association, the DRA), the Taliban were not ‘godless communists’ (in addition to many of the Taliban having been mujahideen and being ethnically Pashtun, many also made strategic use of marriage as a way to establish themselves as non-­foreigners). More importantly, the mujahideen fighters that had finally ejected the Soviets and then destroyed the DRA and its leadership, could never agree on a plan of reconstruction and reform, and instead set about attacking each other and shamelessly looting what was left of Afghanistan’s assets. Their venality and infighting, combined with the lack of a ‘rally against the foreigners’ reaction to the Taliban, effectively doomed them, in part because the Taliban were based in Pakistan and sat astride logistical support for the victorious warlords, whose combat effectiveness still depended largely on supplies now both dwindling in quantity and accessible to interdiction by Taliban fighters. In sum, the Soviets left an Afghanistan reduced to a loose confederation of autonomous micro states, each complete with its own ethnicity, language and leadership. The Soviet exit also removed the chief underlying common cause among Afghanistan’s micro states, which was ‘ejecting the foreigners’. The Taliban then moved in, claiming to be incorruptible, taking advantage of their years of

Imagined States and the Failure of COIN

233

patient social mobilization beyond Soviet reach in Pakistan, and leveraging their ability to interdict now dwindling supplies of food, arms and ammunition to the squabbling and venal post-DRA warlords. It should go without saying that any Afghan administrators the Soviets had trained during their occupation either fled or were captured and subsequently killed or imprisoned. For their part, the Taliban soon had to face the difficulty all revolutionaries do: once the incumbents are defeated, the revolutionaries must accommodate themselves to the far less romantic task of governing. In their case, the Taliban set about attempting to do so by means of a narrow interpretation of Sharia law, while awaiting what they believed would be the inevitable spread of their revolution to Central Asia and beyond. The result of their ‘government’ was two-­ fold. First, as they settled down, their marriages resulted in children, which made it difficult for them to avoid the temptation to graft. Opium poppy cultivation exploded, and with it a flow of drug money that made the Taliban rich. Second, their increasingly obvious corruption, and medieval ‘one-­size-fits-­all’ approach to justice rapidly alienated the survivors of the war. Far from recovering from the war and rebuilding, the Taliban seemed content to preside over Afghanistan’s devolution into a kind of gigantic criminal enterprise. The resentment Taliban hypocrisy bred was the single biggest cause of their rapid collapse and defeat during OEF.

Operation Enduring Freedom The groundwork for OEF, as it was called by the United States after the terror attacks of 11 September 2001, had been laid well before the attacks of that day. Details of the planning and execution of what became a punitive expedition against the Taliban in Afghanistan are best read elsewhere. Here, I want to focus on two key aspects of the operation, which is technically still in effect. First, we should ask whether as a result of US and allied efforts since 2001, there is today a significant body of Afghans whose first interest is rebuilding Afghanistan, as opposed to shoring up the power of one warlord or another over his rivals. Second, we should ask whether, as a result of OEF, the long-­term prospects for COIN success – as judged by the state of the art in terms of strategy – are likely to result in the formation of a state in Afghanistan. In other words, will Afghanistan be sufficiently free of insurgency that it can begin to undertake long-­term investment and contracts capable of rebuilding its infrastructure and a non-­ criminal economy?

234

War and State-Building in Afghanistan

OEF began on 7 October 2001. Its original stated aim was to oust the Taliban from power as punishment for their complicity in the terror attacks which caused the deaths of over 3,000 people in the United States, most of them civilians. In short, it was a kind of punitive expedition. The fact that as a punitive expedition it was so successful so quickly has spawned a number of excellent analyses,18 but the best point out three key features. First, the number of US combat forces was severely limited. Afghanistan is land-­locked, and surrounded by six other states, none of whom could easily abide a major US logistical effort. Thus, the first US personnel into the country were a mix of CIA special activities division teams, and US Special Operations Forces (SOF). This small footprint had two important consequences: (1) it rewarded collaboration with the Taliban’s local opposition groups (mainly, Ahmed Shah Massoud’s Northern Alliance); and (2) it prevented the traditional Afghan ‘rally against the foreigners’ effect. In terms of collaboration, this is doubtless the single most important cause of OEF’s early success, because the surviving opponents of the Taliban possessed intimate knowledge of Taliban forces’ tactics, resources, bases and leaders. Second, the Taliban had overrun Afghanistan by 1996 against forces that did not possess air support. Tactically and operationally, they were not accustomed to the difficulties of movement and concentration when an opposing force controlled the skies and possessed weapons such as the AC–130 Spectre gunship. With the help of local allies in the Northern Alliance, US SOF knew what each Taliban unit had, and where it was located. It was then a relatively simple task for highly skilled coalition forces to call in air strikes on Taliban heavy weapons and vehicles, or in support of local offensive operations. Third, and perhaps most importantly, the Taliban could not mimic the insurgency of the mujahideen against the Soviets because of all the ill-­will they had built up with their draconian Sharia law and naked venality. The Taliban sought to efface, rather than harness, the ethnic, cultural, topographical, historical and linguistic richness of the territory they controlled. Their rule had been established by a combination of force and bribery, and above all by an assertion that God was on their side. Tactical reversals appeared to give lie to that last claim, and their accumulating defeats made effective bribery impossible. Without social support the Taliban were forced to confront a massively fire-­supported Northern Alliance in pitched battles, a contest they were bound to lose (and did). On 9 November, NATO and Northern Alliance forces ousted the Taliban from the strategically vital city of Mazar-­i Sharif, and NATO and Northern Alliance objectives began to diverge. The Northern Alliance wanted to evict the

Imagined States and the Failure of COIN

235

Taliban from Afghanistan and take power, whereas the US and Britain were increasingly distracted by a need to find and capture (or kill) the Al-Qaeda leadership, in particular Osama bin Laden. On 12 November, the Taliban fled Kabul, and their control of Afghanistan began to unravel. In the ensuing weeks, one after another of Afghanistan’s major cities fell to allied forces as Taliban commanders and their security teams fled to make last stands in cave complexes near the border with Pakistan, or across the frontier proper into Pakistan itself, where they would remain relatively safe until 2009. By December, Afghanistan had no effective central government. Two developments are crucial. First, the United Nations hosted the Bonn Conference to arrange a transition government. The Taliban were not represented, but four of the opposition factions were. The idea was that the interim government would have a mandate to draft a constitution and prepare for elections. Second, the United Nations authorized the formation of the ISAF, on the grounds that unless and until security could be established in Kabul and other major centres of population, no progress could be made on reconstructing the Afghan state. The process and its logic should sound familiar by now. Effectively, an alien principle of legitimacy – elections – would be imposed on Afghanistan’s micro states as a way of selecting which warlord would have the right of first access to the international community’s wealth and security resources. In terms of security, a limited contingent of armed forces would be required to establish and maintain security within and between the country’s major cities and bases. What is remarkable is how little security was actually needed in 2002. Bear in mind that a great deal of the Taliban’s legitimacy had been based on the ancient principle of trial by combat (in which victory serves as perfect evidence of God’s favour). The root principle of Salafism – the return of Islam to the pinnacle of global civilization – was to be accomplished by enforcement of strict adherence to God’s laws (Sharia). The more conservative religious victors against the Soviet Union (most famously, Osama bin Laden) attributed mujahideen victory not to Stinger MANPADs but to God’s favour: it was God who directed that the mujahideen ‘David’ should defeat the Soviet ‘Goliath’ (a reversal of the 1948 Nekbah, in which a vastly outnumbered Israel was victorious over a coalition of Arab [and Islamic] armies; again, attributed by Salafists to God’s collective punishment for Muslim sin). So, the rapid defeat of the Taliban damaged their legitimacy, and their rapid exit made them appear to have received God’s judgement in the negative. Into this vacuum the international community promised to re-­establish the rule of law (nonSharia law) and rebuild Afghanistan’s eviscerated infrastructure and non-­criminal

236

War and State-Building in Afghanistan

economy. Straightaway, however, the effort was bedevilled by two problems. First, there was no public service network that could provide physical security or administer incoming aid. Instead, it had to be funnelled via foreign administrators through the warlords (eventually Hamid Karzai), and that meant that a great deal of what arrived would never reach the people, and by extension, hurt rather than help the COIN effort. Keep in mind that if this sounds like an accusation of corruption and venality, it should. But the system of familial loyalty in Afghanistan demands that the head of a clan take care of his family and relatives first. This must be true regardless of the skill sets of those who are hired under such circumstances. From the standpoint of managing a large multinational state, this sort of nepotism is lethal, but it works fine in a small-­scale state or province; especially when the aim of government is narrowed to war fighting and survival. This is true because war necessarily narrows the distribution of relevant skill sets. Once war ends, however, the need for a broader set of skills becomes an open question. Once the Taliban fled and Afghanistan’s roads, rail network and air bases came under ISAF control, there was no structural limit to the number of coalition forces that could be deployed to Afghanistan, and this created two looming problems. The first was the problem of supply: ISAF forces demand a constant stream of supplies and this creates operational and diplomatic bottlenecks and vulnerabilities. The second was that ISAF and aid personnel were ‘foreigners:’ the bigger the footprint, the greater the hazard. There was simply no avoiding this, because unlike war, development cannot be run by a few well-­informed special operations forces. It demanded contractors, heavy equipment and skill sets which Afghanistan, after 20 years of bitter war, no longer possessed. What most Afghans were good at was survival; and that meant a talent for accumulating cash, weapons and fighters. In the winter of 2006, for example, NATO introduced Provincial Reconstruction Teams (PRTs) into southern Afghanistan, but the effort did not go well. The Taliban had been returning since 2003, and by 2006 had fairly well established themselves across much of the territory of Afghanistan. From the spring until the next winter, the Taliban battled NATO forces in a series of engagements in which they were never completely defeated. The interaction began to look increasingly like the pattern of Soviet engagements had. A corrupt, semi-­competent government in Kabul, supported by ‘foreigners’, was able to mount limited offensives against insurgent (Taliban) positions that resulted in short-­term tactical victories, but did not alter the long-­term strategic picture. In 2007, ISAF forces went on the offensive, and achieved important victories that nevertheless did nothing to alter the basic strategic picture. Why is this?

Imagined States and the Failure of COIN

237

Some of the answers are well known: Pakistan proved either unwilling or unable (likely a bit of both) to interdict fighters across its frontier with Afghanistan. Many areas in Pakistan became functional sanctuary for the Taliban. But what is less well understood is that the interaction between ISAF and the Taliban favoured (and still favours) the Taliban, because all ISAF can do well is kill (they have proven more lethal to the Taliban and, for all the reports of collateral damage which are horrible, an order of magnitude more discriminate than the Soviets had ever been). The Taliban are about ideas, not corpses. They are modestly effective at killing Afghan security personnel and ISAF soldiers, but this is not how they plan to win. They plan to win the battle of ideas because their preferred principle of legitimacy, ‘God wills it’, is closer to Afghan experience than ISAF’s ‘democratic elections’. The international community’s reconstruction efforts have not resulted in creating job opportunities other than the traditional smuggling and fighting ones. Young men still have very few options conducive to the creation of a stable and functional non-­criminal economy. In short, the Taliban are killing ISAF and Afghan armed forces, and being out-­killed by a large margin in almost every engagement. But they don’t have to kill or even kill proportionately to win. They can win by simply waiting for ISAF to leave, and then going after government forces while playing off one warlord against another. They will have Pakistan as a logistical and intelligence base area as before, and once the skies are clear of ISAF air support, and relatively clear of drones, their forces will simply roll up whatever the ‘Afghan’ government puts in their way. The OEF’s transformation from a punitive expedition (a resounding success) quickly morphed into a preventive minimalist neo-­colonial enterprise (ISAF), in which a limited contingent of armed forces sought to secure key political and communications sites from destruction. ‘ISAF opened the floodgates to international governmental and non-­governmental organizations, many of which undertook at great expense and personal risk to aid Afghanistan’s peoples. But they were all foreigners.’ Unlike the Taliban, they don’t marry locally and everyone understands they aren’t staying. The ISAF’s COIN campaign has become progressively better over time (especially since 2007, when in the Pentagon, the ‘counterinsurgency’ faction won out, at least temporarily, over the ‘fight and win the nation’s wars’ faction19), but has suffered most from a lack of resources that reach the people of Afghanistan in sufficient quantity to give young males an alternative to criminal or insurgent activity. In order to get things done, resources must be funnelled through local warlords, and this means that too much of it went to increasing

238

War and State-Building in Afghanistan

each recipient’s security and wealth at the expense of the whole. Bottom-­up COIN is much more effective at denying the Taliban bases and interim political objectives, but has come at the political cost of intensifying Afghanistan’s political Balkanization. So at the end of the day, OEF and ISAF will leave an Afghanistan that is really a collection of rival micro states and not a single state with a capable supra-­regional government. One strong potential counterargument is that in terms of the original and most important of a state’s public servants – its security personnel – Afghanistan is increasingly well provisioned. By the end of 2012, two years before the planned draw-­down of ISAF combat forces, US General William Caldwell is expected to have increased the number of Afghan National Army (ANA) forces to over 350,000. Why not count the ANA as the nucleus of a real, as opposed to imagined, state? It is not a good argument. These forces are simply not loyal to something as abstract as the Afghan state (or to the extent they are, they are loyal in the negative sense of defending Afghanistan from foreigners). Instead, as I have argued elsewhere, they are loyal to their families, who in turn owe their fealty, security and economic prospects to local warlords. In other words, ISAF is busily training security personnel who at most will serve as a nascent army for one warlord or another (most often, and by design, Hamid Karzai). Even its ability to defend Afghanistan as a state, if called upon, depends on resources and skills (in particular calling in fires from air or artillery) that remain beyond it. Once ISAF leaves we should expect it to devolve to the control of Afghanistan’s provincial warlords rather than remain a resource of Kabul’s authority. The ANA, then, is illustrative of the nature of public servants currently working in Afghanistan: once the foreigners leave, they will revert to whichever warlord is responsible for securing their families.

Conclusions NATO and ISAF have undertaken COIN in Afghanistan in order to preserve a nascent Afghan state long enough for a legitimate central government to take over the task of governance. Success will be measured by the ability of the central government in Kabul to provide, at a minimum, physical security for Afghanistan’s peoples. In other words, ISAF’s COIN is aimed at supporting the creation and maintenance of a state that incorporates the diversity of Afghanistan’s ethnic, linguistic and religious groups. This will not happen because there is not now sufficient indigenous demand for a state in Afghanistan. Instead, ISAF’s

Imagined States and the Failure of COIN

239

commanders have become increasingly successful at countering Taliban (to cite but one group working to undermine the incumbent government) insurgents valley by valley, but not overall. The states system has proven remarkably durable over the nearly four centuries since we date its advent in the mid-­seventeenth century. I have followed Charles Tilly’s logic in arguing that so long as mutual predation seems likely, the state is apt to remain a dominant (if costly) form of political association. One important empirical change of the past three decades has been a dramatic decline in the likelihood of interstate war. This would appear to be good news. But my argument is that for an increasing number of the world’s people, one consequence is that in the absence of a credible threat of conquest, the costs of living in states have exceeded the benefits. And here is where the argument becomes somewhat novel. We assume an existential indigenous demand for states as axiomatic (think ‘national self-­determination’, for example). Everyone wants his or her own state, we imagine. But what if it were no longer a question of wanting a state of one’s own, but rather a desire to live in some alternative form of association? People who live in clans and tribes, civilizations and empires, are killed from time to time just as those who live in states. But many of these people appear to be uninterested in living in states. This may be an unintended and ironic consequence of the success of peace operations and norms entrepreneurs – as Joshua Goldstein and Steven Pinker argue respectively.20 Whether individuals benefit as such by choosing to live in polities other than states, the negative externalities many of these associations export have become a real and lethal concern for those who consider life within a state, and by extension a states system, normal and necessary. It is this lethality that leads to an interest in intervention, including military intervention, aimed at restoring the state. It is also this lethality, and the concern that like contagion, lack of desire to live in states will increase the likelihood of a return to world war and global thermonuclear war, that leads to imagined states; and an unwillingness to consider the benefits of polity heterogeneity as opposed to the costs only. We should care about imagined states – in this case Afghanistan – not only because they generate increasingly costly negative externalities such as narco trafficking and terrorism, but because there may be better ways of mitigating those externalities than interventions aimed at restoring an unwanted state. The nursery rhyme has it this way: Humpty Dumpty sat on a wall, Humpty Dumpty had a great fall,

240

War and State-Building in Afghanistan

All the King’s horses and all the King’s men Couldn’t put Humpty together again.

What I have asserted here is that instead of assuming our job is to put Humpty together again, we should be asking a better set of questions. In particular: What are the relative costs and benefits of rebuilding an Afghan state? Might there be a better (or at least less costly) way to preserve our citizens from terror or environmental degradation or organized crime than a punitive, preventive or pre-­emptive military intervention? The solution to the apparent dilemma of imagined states such as Afghanistan is as logical as it is unpalatable: either (1) abandon all hope of successful intervention21 and look to our own defences (invest resources that might have gone towards intervention in homeland security), or (2) re-­introduce a credible threat of major interstate war (this would create indigenous demand for statehood). Theory, in other words, needs to catch up to reality, which is increasingly taking the form of polity heterogeneity: a mix of different forms of political association including states, commonwealths, regional economic blocs, multinational corporations, intergovernmental organizations, international organizations, transnational religion, civilizations, clans, tribes, and so on. States, and a system of states, may be had, but perhaps only at the very high cost of re-­ introducing its chief rationale for dominance: war. The primary focus of this chapter has been the relationship between Afghanistan as an imagined state and COIN efforts there since 2001. Put bluntly, OEF has failed. This was recognized most recently at a summit in Chicago in which a new term, ‘Afghan good enough’ entered into vogue as a euphemism for the now common practice of achieving victory by fiat (in this case, by retroactively reducing the political objectives to match what has been achieved in theatre).22 I do not call attention to this process with either joy or recrimination: all states about to lose a war do this. The US, for example, did the same thing via the rhetorical mechanism of ‘peace with honor’ in Vietnam in 1973. But it will not do for us to pretend here that OEF succeeded by accepting ‘Afghan good enough’. We should call a failure a failure and set about learning from it. The policy implications of this argument should be by now clear. In so far as the large sums of blood and treasure spent to rehabilitate ‘Afghanistan’ since 2001 have come from self-­interested states (e.g. the United States in reaction to the terror attacks of 9/11), the best policy option remaining is to leave Afghanistan to its own devices, and divert the resources that might have been spent on COIN towards homeland security. Note, by ‘best’ I mean the lesser of two evils. Leaving

Imagined States and the Failure of COIN

241

Afghanistan to Afghans will have at least five costly consequences. First, it will mean escalated suffering for an already long-­suffering Afghan people (in particular its women). Second, and most significantly for intervening and donor states, it will look as though whoever is in charge has ‘lost’ Afghanistan. Third, it will mean an increase in the export of by-­products of organized criminal activity, including human and drug trafficking. Fourth, it will likely result in the disintegration of Afghanistan into micro states (cf. Yugoslavia in the late 1990s). Fifth and most ominously, if the Taliban and their allies again take over Afghanistan, it is possible that far from continuing to use Pakistan as a base area from which to manipulate Afghan politics, Afghanistan may become a base area for manipulating Pakistani politics. Given that Pakistan is known to possess nuclear weapons, this is a significant risk of ISAF’s departure. As hurtful as all this is, however, it must be considered in comparison to the extraordinary costs of continuing futile efforts to build a state that not enough Afghans want. If a state is truly wanted, all that is needed (logically) is to facilitate the conditions that increase indigenous demand for a state. We should explore ways other than war as a means of doing this. Indigenous demand will lead to the progressive creation and maintenance of a core of competent Afghan public servants, which will in turn lead to a state as such (likely not a democratic state, and likely a state which continues to host actors whose interests are inimical to those of NATO member states as well as others). Until or unless indigenous demand for a state spanning the territory we know as Afghanistan comes into force, Afghanistan – like Somalia and Democratic Republic of Congo (to cite but two contemporary examples) – will remain imagined, rather than real; and the consequences will continue to be lethal – albeit incrementally – for hundreds of thousands of the world’s citizens.

Notes 1 On the relationship of civil service to the state as a form of political association, see Ivan Arreguín-Toft, ‘The Meaning of “State Failure”: Public Service, Public Servants, and the Contemporary Afghan State’, International Area Studies Review, vol. 15, no. 3 (2012), pp. 263–78. 2 Students of international politics will benefit from reflection on this point: if at root the state not only incorporates but enshrines violence, and the shearing away of other forms of power from violence is in some meaningful sense a masculinist artifact, then the very building blocks (the range of questions that can be asked and

242

War and State-Building in Afghanistan

answered) of international politics are unproductively gendered. For an introduction to this theme, see Carol Cohn, ‘Sex and Death in the World of Rational Defense Intellectuals’, Signs, vol. 12, no. 4 (Summer 1987), pp. 687–718; J. Ann Tickner, Gender in International Relations (New York: Columbia University Press, 1992), Chapter 2. 3 Thomas Hobbes, Leviathan (Revised Student Edition, Richard Tuck, ed.) (New York: Cambridge University Press, 1996). 4 Mis-­named because ‘failed’ implies that some person was unable to accomplish a thing she/he was trying to do; whereas if I am correct, an increasing number of the world’s people have ceased trying (due to lack of interest, not capacity) to create or maintain states. On failed states see, for example, Robert I. Rotberg, When States Fail: Causes and Consequences (Princeton, NJ: Princeton University Press, 2003); and Ashraf Ghani and Clare Lockhart, Fixing Failed States: A Framework for Rebuilding a Fractured World (New York: Oxford University Press, 2009). 5 Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1991), pp. 5–7. 6 Perhaps the best and most accessible intellectual forerunner of this line of social construction should go to Robert Kaplan. See Robert D. Kaplan, ‘The Coming Anarchy’, The New Republic, vol. 273, no. 2 (1994), pp. 44–76. In keeping with Hobbes’s thesis (reflected in the article’s title), Kaplan’s essay vividly unpacks Hobbes’s famous ‘nasty, brutish, and short’ description of life within a hypothetical state of nature. In Kaplan’s case, however, there is little hypothetical about the brutal quality of life in the African countries he surveys. The real (and unanswered) question, then as now, is what (if anything) to do about it. Kaplan assumes that the anarchy he describes in Africa is coming to the advanced-­industrial world. The piece may be read as an argument for intervention, or as support for isolation and homeland defence. 7 One crucial consequence was this cultural space’s conflation of lethality with power. The association has proven a useful one, but killing as power is not universally useful; and suffers particularly when directed at coercing groups whose legitimacy depends on sacrifice or martyrdom. Both the Roman experience with Christians, and the more contemporary experience of OECD states against Islamic militants (such as the Taliban in Afghanistan) stand as illustrative cases. 8 Charles Tilly, ‘Reflections on the History of European State-Making’, in Charles Tilly (ed.), The Formation of Nation-States in Western Europe (Princeton, NJ: Princeton University Press, 1975), p. 42. 9 See especially Joshua S. Goldstein, Winning the War on War: The Decline of Armed Conflict Worldwide (New York: Dutton, 2011); Steven Pinker, The Better Angels of Our Nature: Why Violence Has Declined (New York: Viking, 2011). 10 Note again the problem this question creates: so entrenched is the idea of the state as normal and necessary, that to ask whether one might be better served in one’s life

Imagined States and the Failure of COIN

243

aspirations by life within a tribal or clan association is tantamount to identifying oneself as either ignorant or backward. But to the extent that Tilly’s pithy association of the state with war is correct, this is tantamount to arguing it is backward to wish to live within a form of association not dedicated to preparing to engage in aggressive or defensive war. Certainly it is difficult to sustain a claim that for nomadic peoples a state is ideal; and once we begin questioning the universal necessity or benefit of the state as such, an entire range of subsidiary questions become accessible. 11 On this point, see James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven, CT: Yale University Press, 1999). 12 See Arreguín-Toft, ‘The Meaning of State Failure’. 13 For a brief summary version of the Soviet intervention with special reference to Soviet COIN, see Ivan Arreguín-Toft, How the Weak Win Wars: A Theory of Asymmetric Conflict (New York: Cambridge University Press, 2005), Chapter 8. 14 M. Hasan Kakar, Afghanistan: The Svoiet Invasion and the Afghan Response, 1979–1982 (Berkeley, CA: University of California Press, 1995), p. 15. 15 This argument seems preposterous until one thinks it through. The US Air Force came to US Defense Secretary McNamara in 1968 with a plan to further escalate the bombing of North Vietnam in order to coerce the North into, finally, ceasing to supply and support the Viet Cong. McNamara, upon looking over the target sets, demurred. His argument was that killing sufficient to coerce the North looked too much like genocide, and that the United States was not in the business of genocide. See, for example, Stanley Karnow, Vietnam: A History (New York: Harper Collins, 1983), p. 509. There is a deeper logic at work that is worth considering: what if your adversary says effectively, ‘go ahead and kill us, we won’t yield’, and then acts in ways that make this claim more than propaganda? This puts the actor attempting to use force in a dilemma: either accept genocide as a cost of victory, or give up the use of force. 16 Students of military history will instantly appreciate the similarities between Soviet forces in Afghanistan in the early 1980s and US forces in Korea in the winter of 1950. 17 Arreguín-Toft, How the Weak Win Wars, Chapter 8. 18 See, for example, Stephen Biddle, Afghanistan and the Future of Warfare: Implications for Army and Defense Policy (Carlisle, PA: The US Army War College, Strategic Studies Institute Monograph, 2002). 19 Debate still rages at high levels within the US military, however. See Elisabeth Bumiller, ‘West Point Divided on a Doctrine’s Fate’, The New York Times (27 May 2012). The logic of proponents and opponents of COIN as a viable military mission is simple and sound. On the contra side is the argument that insurgency and

244

War and State-Building in Afghanistan

terrorism are law enforcement problems and as such are not properly the province of the US military. On top of that, effective COIN takes at least seven years on average, and US public support even for ‘real’ wars hasn’t historically lasted longer than three. Therefore, COIN operations are existentially unwinnable. On the pro side is the argument that the distribution of wars has made major interstate conflict all but obsolete, the security, environmental and humanitarian costs of standing ‘idly by’ continue to rise, and if the military is going to remain relevant and useful it had best master the very difficult art of COIN. The problem from our perspective is that OEF appears to offer strong empirical support for both sides. 20 Goldstein, Winning the War on War; Pinker, The Better Angels of Our Nature. 21 One of the best arguments along these lines is that of Richard Falk. As Falk puts it, ‘Nonintervention is intolerable, but intervention remains impossible.’ See Richard Falk, ‘Hard Choices and Tragic Dilemmas’, The Nation, vol. 257, no. 21 (December 1993), p. 757. 22 See Helene Cooper and Thom Shanker, ‘US Redefines Afghan Success Before Conference’, The New York Times (17 May 2012).

9

Heart or Periphery? Afghanistan’s Complex Neighbourhood Relations1 Kristian Berg Harpviken

Introduction There is little controversy over the importance that Afghanistan’s neighbours have played in fuelling the conflict in the country. But once the attention turns to what it is that drives the neighbours’ commitment, how they are involved, and ultimately how to understand the relationship between Afghanistan and its neighbourhood, there is little consensus. Focusing on the fundamental security relationships in the larger neighbourhood, I argue that Afghanistan both historically and currently finds itself at the interface of three distinct regions: Central Asia, the Gulf region and South Asia. Each of these three regions is characterized by its own regional security dynamic that plays itself out in Afghanistan, a country that is peripheral, at most, to the security concerns in question. The continuity in the composition of the regional entities in question, even when its constituent states undergo dramatic political upheaval, is quite remarkable, and extends to the long historical lines in their relationship to Afghanistan. Writing in mid-2012, the international military presence is winding down, with all security transferred to the Afghan Government, but with an ambition to maintain a smaller international contingent well beyond that point. The criticality of the regional environment is becoming increasingly clear to all. In this light, it is interesting to examine US policy shifts over the long decade since the 2001 intervention. In September 2011, US policy took a new twist as Secretary of State, Hillary Clinton, launched ‘The New Silk Road’ as the key to stabilizing Afghanistan and the region at large. The vision is a reinvigoration of Afghanistan’s role as a connector of countries, focusing on the potential inherent in economic cooperation between South and Central Asia. While the

246

War and State-Building in Afghanistan

Map 6  Afghanistan and its Neighbourhood

vision was new, the intensifying attention to the regional was not. Already when President Barack Obama launched his new strategy on Afghanistan and Pakistan in Washington in late March 2009, there was a reinvigorated interest in the regional dimensions of the problem. Obama emphasized the importance of the region and his willingness to work with all parties (Iran included) to stabilize Afghanistan and to prevent Pakistan from imploding. The Obama initiative, while regional in outlook, was soon overtaken by what became known as ‘AfPak’, a focus on the interaction between Afghanistan and Pakistan, informed by the view that Pakistan was becoming a much greater liability to world peace than Afghanistan, and that many of the problems in Afghanistan was a direct result of Pakistan’s support to the insurgency. Over the two presidential periods preceding Obama, from 2001 to 2009, the Bush Administration pursued a more myopic view on the role of Afghanistan’s neighbourhood: it was one issue and one country at the time, in line with a foreign policy practice that favoured bilateralism over multilateralism, both at the regional and the global levels. Despite these reorientations in US policy over the critical long decade following 9/11, there is every reason to question the basic analysis that, counter-­intuitively, is common through the seemingly dramatic policy shifts. Fundamentally, we may delineate two opposing views on how to understand the relationship between Afghanistan and its larger neighbourhood. Some analysts emphasize states and their security relationships and see Afghanistan as an ‘insulator’, caught between different regional state systems, each with a strong

Afghanistan’s Complex Neighbourhood Relations

247

dynamic of their own. An alternative perspective – which is much more consistent with US policy formulations – sees Afghanistan as the ‘core’ of a larger conflict formation, nested together by a range of transnational networks. A variety of this perspective, informing the ‘New Silk Road’ vision, emphasizes economic interaction, with its rich historical precedents, but also places Afghanistan at the centre. This chapter privileges the former, codified by Barry Buzan and Ole Wæver in the Regional Security Complex approach, and will start by discussing this perspective.2 Next, the chapter will position Afghanistan, geopolitically and historically, with an emphasis on its more recent conflicts unfolding from the late 1970s onwards. It pursues the security dynamics of each of the core regions surrounding Afghanistan (South Asia, the Persian Gulf and Central Asia), taking a comparative and historical perspective, with emphasis on the period since the late 1970s. It concludes that each of Afghanistan’s three surrounding regions is characterized by deep security concerns that have little to do with Afghanistan. These concerns nonetheless inform their engagement in Afghanistan, which then comes to reflect conflicts and cleavages specific to the region. Before offering some overarching conclusions, the chapter discusses possible implications for the Afghan Government and its international allies. One possible implication is that for Afghanistan, it may be a more promising strategy to seek a unilateral non-­offensive or neutral status, rather than security integration with its neighbours. While this would necessitate a forum of Afghanistan’s neighbours in order to foster understanding for the Afghan position, it suggests a dramatic departure for mainstream policy proposals with their emphasis on an integrated regional approach.3

Regional Security Complexes The Regional Security Complexes (RSC) approach has its roots in work by Barry Buzan dating back to the early 1980s,4 and later refined in joint works by Buzan and Ole Wæver.5 A basic assumption is that geographic proximity matters for security, in the sense that states (or other referent actors) adjacent to each other do not have the option of disengagement, whereas distant actors do. In its 1998 definition, an RSC is: ‘a set of units whose major processes of securitization, desecuritization, or both are so interlinked that their security problems cannon reasonably be analyzed or resolved apart from one another’.6 In this perspective, geographic adjacency matters. The definition emphasizes interdependence; mutual security interdependence among the units that

248

War and State-Building in Afghanistan

constitute a RSC is distinctly more significant than that between any of these units and those outside it. Regions are defined as mutually exclusive; one cannot be a member of multiple RSCs. Accordingly, superpowers – and great powers – may play important roles in RSCs other than their own, but this does not make them into constituent members – they can always withdraw. Rather, the theory favours multilevel analysis, placing the regional level both in relation to a subordinate level of ‘units’, and a superordinate level of interregional and global dynamics. Buzan and Wæver’s insistence that RSC membership is mutually exclusive sets their definition apart from others,7 most notably the definition of RSCs provided by Lake and Morgan,8 who allow for multiple membership, hence also overlaps between RSCs. Buzan and Wæver see this as a confusion of levels, and as contrary to their ambition of formulating a theory of regional security that can be applied not only to one region at a time, but to the analysis of the entire global security structure. Hence, membership hinges on the following key criteria of inclusion: geographic proximity, relative durability of relationships and shared security priorities. Superpowers rarely meet these key criteria of inclusion, and when they do, it is only in the region in which they are geographically embedded (where they are also likely to be dominant). The first part of the definition entails an important modification from earlier definitions, emphasizing ‘securitization’, the constructivist assumption that what drives security dynamics is not threats in any objective sense, but the perceptions of and the dominant narratives about threats.9 In the 2003 book by Buzan and Wæver, the analysis of securitization plays a marginal role (although they themselves see it as essential to the theory). In the following, I will similarly presume that threats are, at least in part, socially constructed, without attempting to depict processes of securitization. It is also noticeable that whereas in early versions of the definition, the ‘state’ was the referent object for security, in later versions it has been replaced by ‘units’, which may encompass various types of non-­state actors as well as multilateral entities. Yet, the 2003 book, which presents a global map of RSCs, is overly state-­ centred. The authors acknowledge this as a possible weakness of their analysis – although not of the theory – but at the same time argue that states ‘remain de facto at the center of much of the structure of global security’.10 A key critique of the RSC approach would be that it fails to include the specific character of transnational terrorism. To this, Buzan and Wæver respond that even though transnational terrorism may rely on global networks and propagate a global agenda, its actors are firmly embedded in specific regional security agendas – hence the main effects of 9/11, for example, are seen in the dynamics of the

Afghanistan’s Complex Neighbourhood Relations

249

South Asian and Middle Eastern RSCs, as well as in the US engagement in those regions and with the great powers.11 Territoriality is resilient, and even new security threats – associated with globalizing processes – are best understood within an RSC framework. The RSCs are of different types. Mainly, they are distinguishable on two variables. First, there is the polarity question: are they unipolar (contains a great power), bipolar (dominated by rivalry between two units) or multipolar. Polarity has major implications for regional security dynamics. Secondly, and relatedly, there are variations in the crucial amity–enmity variable. Buzan and Wæver suggest a continuum in the maturing of an RSC, from conflict formation to security regime to security community.12 In a regional conflict formation, enmity dominates, there is no mutual trust and no institutions that can effectively constrain the use of force (Middle East, South Asia). In a regional security regime, mutual suspicion and fear of violence dominate politics, but are kept in check by working institutions (Southeast Asia). In a regional security community, there is a mutual understanding of security interdependence and the actors do not perceive the use of force as an option (European Union, North America). Not all regions fits neatly into one category; the Confederation of Independent States (CIS), for example, combines elements of formation and regime with the imprint of Russian struggle to establish hegemony. From the vantage point of the early twenty-­first century, Buzan and Wæver define Afghanistan as an insulator between distinct RSCs – South Asia, the Middle East and the CIS. For them, the buffer concept is reserved for a state (or a ‘mini-­complex’, set of states) within a RSC that keeps rival powers apart. An insulator is located at the boundary between complexes, keeping them apart – it is ‘a zone of indifference’.13 Other insulators in the present-­day global system are Burma, Turkey and the Sahel states. As an insulator, Afghanistan constitutes an entity of its own in a world divided into regions. Afghanistan ‘draws in neighbouring states, but its internal dynamics is strong enough to keep the larger dynamics separate’.14 This proposition is a rather bold one; not only does it imply that domestic dynamics was driving the war of the 1980s, when Afghanistan became one of the major battlegrounds of the so-­called Cold War, it also suggests that the virtual insulation is robust and is likely to endure. Buzan and Wæver’s proposition runs contrary to the widespread optimism of the immediate post-Cold War years – reinvigorated almost two decades later with the ‘New Silk Road’ initiative – when it was hoped that the newly independent Central Asian states would be linked to South Asia via an Afghan land bridge, an escape from their dependence on Russia. A peaceful Afghanistan would allow

250

War and State-Building in Afghanistan

the development of a new transportation infrastructure, including a gas pipeline linking Central Asia to the subcontinent.15 Writing now, two and a half decades after the collapse of the Soviet Union, and a very long decade after the intervention in Afghanistan, we can safely conclude that Afghanistan has not developed into the land bridge that was hoped for in the early 1990s. Following an RSC logic, we should not find this to surprising: the surrounding regions have a strong internal dynamics, and Afghanistan is virtually doomed to serve as a shock absorber where the tensions inherent in the surrounding regional complexes interact with existing domestic tensions to further deepen the conflict.

Positioning Afghanistan Geographically, Afghanistan is at the intersection of South Asia, Central Asia and the Middle East. External bureaucracies may place Afghanistan firmly in one of the three regions, in pursuit of clear geographic divisions, while area experts may argue that it has strong affinities with all of them. The shape of present-­day Afghanistan reflects the need, from the mid-1800s, for what is commonly referred to as a ‘buffer state’ between the British and Russian empires.16 From the south, the British were seeking to expand the territory under their control towards a defensible border, the so-­called ‘forward policy’.17 To the north, the Russians were seeking to solidify and expand their control over Central Asia, and tensions between the two empires intensified throughout the latter part of the nineteenth century. The buffer state started to materialize in the late 1870s, with the Gandamak Treaty, which concluded the Second Anglo-Afghan War (1878–9). In 1895, the establishment of the Wakhan Corridor, a narrow arm through a rather hostile mountain territory that extended Afghanistan to the Chinese border, brought the buffer logic to completion.18 Afghanistan maintained its status as a buffer state for much of the twentieth century. The deals of the late 1800s inherently made Afghan rulers dependent on the two neighbouring empires, on Britain in particular.19 In 1919, the Afghan King, receiving military support from the newly born Soviet Union, launched war against the British. The main outcome of the Third Anglo-Afghan War (1919) was that Afghanistan regained the right to formulate its own foreign policy. The pattern – since the turn of the century – where the northern neighbour, Russia, sought to counterbalance the British influence on Afghanistan, persisted. The Afghan response was a careful strategy of balancing between the two sides (bi-­tarafi in Dari), a form of non-­alignment that characterized the

Afghanistan’s Complex Neighbourhood Relations

251

country’s foreign policy well into the 1970s.20 Interestingly, in the period between the two wars, Afghanistan sought independence from the two major blocs by bringing in a third party, the Germans. During the Second World War, Afghanistan settled for neutrality. By 1947, the regional context changed dramatically, as the British withdrew from India, and Pakistan emerged as Afghanistan’s neighbour to the immediate south. The US stepped into the vacuum, cultivating an alliance with Pakistan. The Afghans insisted on bi-­tarafi.21 Access for Soviet and US assistance was carefully balanced. By the early 1970s, the balancing between superpowers started to falter. Daud Khan – who had been Afghanistan’s prime minister for a ten-­year period that ended in 1963, mainly as a result of his belligerent attitudes to Pakistan – took power in a bloodless coup in 1973. The 1973 coup was instigated by Daud, but relied on support from many within both the bureaucracy and the army. Importantly, the 1973 coup had the support of the Parcham faction within the People’s Democratic Party of Afghanistan (PDPA), which was more moderate than the rival Khalq faction, but also had close ties to the Soviet leadership.22 Parcham support was instrumental for Daud’s stabilization in the first years after the coup, and went hand in hand with an increasingly close reliance on the Soviets for development support, military competence and hardware. Even if Daud did try to keep the US aboard, the Afghan balance was tilted heavily in favour of the Soviets.23 Growing Islamist movements were also a concern for Daud. Soon after the coup, a group of Islamist leaders fled to Pakistan, where they were welcomed by Zulfiqar Ali Bhutto, who despite the ideological distance, saw the Afghan Islamists as the key to undermining Daud’s regime. The core of the Islamist movement based in Pakistan, unsuccessfully staged cross-­border attacks. They would most likely have remained a footnote in history, had it not been that after the 1979 Soviet intervention their ties to the Pakistani leadership and military intelligence became a key resource, as Pakistan selected the leaders of the exile-­ based resistance groups among them. Daud became increasingly worried about the Soviet tilt. He tried to limit the influence of the Parchamis in his own regime, and by 1977 he resorted to outright purges of communists of all brands (including the Khalq faction of PDPA and the China-­oriented Shula-­e Jawed).24 Members of the PDPA felt increasingly under pressure, and quietly started the planning of a coup, against the strong advice of Soviet representatives. Confrontation between Daud and the PDPA escalated, and by April 1978 the PDPA – relying on its supporters within the army – instigated a successful coup. Reluctantly, the Soviets backed the new regime financially and by sending in advisers. Reluctance morphed into deep concern as the Khalqi powerholders proved not only

252

War and State-Building in Afghanistan

politically impatient, but also more than willing to use forces of coercion to get reforms underway, which resulted in a number of spontaneous uprisings in various parts of the country.25 The year 1979 was dramatic in Afghanistan’s neighbourhood. In February, the Shah in Iran lost power to an alliance led by the revolutionary Islamist Khomeini, and the US lost its major platform in this part of the world. In Moscow, there was increasing concern that the communist revolution in neighbouring Afghanistan was going astray. By the end of the year, the Soviets intervened militarily in order to replace the Khalqi leadership in Afghanistan with Babrak Karmal, their own favourite candidate from the Parcham branch. The Soviets had mistakenly thought that their intervention would cause only modest international reactions.26 The rest of the world, and the US in particular, interpreted the invasion as an offensive manoeuvre and as a first step to expanding the Soviet influence in Asia, including securing access to the Arabian Sea. There were different factors weighing on the Soviet decision to intervene, but the motivations were overwhelmingly defensive rather than offensive. One key factor was the perceived need to install a more moderate leadership that could foster broader public support for reforms, in the place of the violent repression under the Khalqi leadership.27 Overnight, Afghanistan had become a major battleground of the Cold War, and it figured at the very centre of regional politics. From the north, the Soviets were directly involved with its military in a key role. Looking south, Pakistan, which received the larger share of refugees, filled the vacuum resulting from Iran turning its back on the US, and became the main bridgehead for support to the Afghan resistance movement. On the western edge, an unconsolidated Iranian regime was virtually overwhelmed by the war with Iraq (1980–8), and despite its sympathy with the cause of the Afghan resistance, restrained its active support to radical Shia groups, worried that it could otherwise lose Soviet sympathy in its struggle with Saddam Hussein. China, also bordering Afghanistan and a key actor in the larger neighbourhood, declined to engage, despite approaches by the Afghan Maoist movement, by no means a negligible force in Afghan politics in the late 1970s.28 For Afghanistan, the 1980s was a decade characterized by the confrontation between two superpowers, whose engagement transformed the Afghan political landscape and introduced new forms of organization and warfare. However, as borne out both by Iran’s reluctance to engage, and by Pakistan’s role as the broker for assistance to the resistance, neighbouring states – albeit informed by their own security concerns – played important roles in Afghanistan’s war.

Afghanistan’s Complex Neighbourhood Relations

253

As Afghanistan entered the 1990s, the superpower overlay was beginning to vanish. The Soviet Union withdrew its forces in February 1989, after several years of preparation. Soon after, the US started to wind down its engagement. By autumn 1991, the Soviet Union fell apart; by spring the following year, the Afghan communists under Dr. Najibullah’s leadership were forced to step down. In the absence of credible external support, the regime fragmented, as various factions formed alliances with their favourite groups within the resistance. The pivotal event was when the regime’s main militia commander, Abdul Rashid Dostum, struck a deal with groups within the resistance.29 By then, various Afghan groups increasingly identified themselves in ethnic terms, and with a fairly clear pattern of support from various neighbouring powers: Uzbekistan supporting Dostum’s forces (the so-­called Uzbek militia); Russia maintaining contact with various groups of northern origin, particularly Jamiat-­e Islami (dominantly Tajik); Iran throwing its weight between the rather newly formed umbrella of Shia and Hazara groups, the Hezb-­e Wahdat (Unity Party); and Pakistan keeping up its ties with the Pashtun-­dominated groups that had been set up in exile, privileging the Islamist Gulbuddin Hekmatyar.30 The vision of a liberated Afghanistan that would serve as a land bridge between the landlocked new Central Asian states and South Asia was soon overtaken by brutal ethnicized warfare, in which the various Afghan groups had different backers in the larger neighbourhood. Undoubtedly, in the absence of the superpower confrontation, the Afghan power dynamic was strong enough to reproduce itself in new formats, yet neighbouring states did significantly contribute to its exacerbation. It was against this background that the Taliban movement emerged in the autumn of 1994. It did not take long for Pakistan to throw itself behind this new actor – contributing money, arms, military advice – thereby playing an instrumental role in the Taliban’s sweeping success which brought it to take over the capital, Kabul, as early as autumn 1996.31 Entering Kabul, the Taliban also inherited Osama bin Laden and the Al-Qaeda leadership, who stayed behind when the ruling alliance evacuated. The international community’s isolation of the Taliban regime, only recognized by Pakistan, Saudi Arabia and the United Arab Emirates, served as further encouragement of the Taliban–Al-Qaeda alliance.32 This did not deter Pakistan from continuing its support, hopeful that the Taliban would eventually gain control and constitute the ‘friendly government’ in Kabul that was seen as pivotal in relation to the threat from the east, India. Other neighbouring states, for varying reasons, were sceptical of the Taliban, whose alliance with Al-Qaeda increasingly also meant the hosting of various resistance movements from the larger region. Yet, while continuing to support their respective

254

War and State-Building in Afghanistan

allies on the Afghan battleground, the engagement of the neighbours (except Pakistan) can at best be described as lukewarm, and testifies the extent to which other security concerns figured more prominently on their agendas. The terror attacks on 11 September 2001 led to the US war against the Taliban regime in Afghanistan. By this time, the various anti-Taliban forces in the country had joined the so-­called Northern Alliance, most prominently composed of the Tajik-­dominated Jamiat-­e Islami, the Uzbek-­dominated Jombush-­e Milli and the Hazara-­dominated Hezb-­e Wahdat. The alliance was one of necessity, but mutual trust was in short supply, following the grave atrocities of the 1990s, where all these groups had at one time or another fought each other, and committed grave atrocities to civilians on ethnic or religious grounds. Externally, the three main groups in question had the support of Russia, Uzbekistan, Iran (in that order) and increasingly of India. All neighbouring states declared their support of the US-led effort, although in hindsight, it seems clear that Pakistan pursued a dual track, in which maintaining a Taliban capacity went hand in hand with sharing intelligence with the interventionists. The US-led intervention in 2001, followed by a large-­scale military presence, partially served as a lid on the direct engagement of neighbouring states with military groups, although it is clear that old relationships were cultivated, in the expectation of the departure of the international forces.

South Asia and Afghanistan Within the South Asia region, the overwhelming security issue is the conflict between India and Pakistan, which has at regular intervals led to armed confrontations on the disputed borderline in Kashmir. India aspires to be the regional hegemon, and its economic success and relative political stability have strengthened it considerably in recent years at the cost of Pakistan – a trend that has been reinforced by US policies, as manifest in the US–Indian deal on nuclear energy. For Pakistan, which sees its security relationship with India as existential, Afghanistan has traditionally been seen to offer ‘strategic depth’ – an area to which Pakistan could withdraw and regroup its forces in the case of a confrontation with India. Pakistan has engaged deeply with various Afghan groups that it thought would serve its interests, as well as with militant groups who have been willing to confront its big neighbour to the east. There are outstanding issues of considerable concern between Pakistan and Afghanistan, the disputed border and the shared Pashtun ethnic population being the most

Afghanistan’s Complex Neighbourhood Relations

255

important, but India is the key security challenge. Since 2001, India has built a significant presence in Afghanistan, while the Pakistanis have been kept at arm’s length. Understandable as this is, it has not contributed to a sense of common security, and the potential for Indian–Pakistani tensions to continue to play out on Afghan territory appears considerable. While the India–Pakistan tussle dominates regional dynamic, the region also includes Bangladesh, Bhutan, the Maldives, Nepal and Sri Lanka. There are multiple regional organizations, the most important of which is the South Asian Association for Regional Security (SAARC), but the India–Pakistani rivalry has hampered its effectiveness. Rivalries between the superpowers – the United States, Soviet Union (later Russia) and China – have considerable impacts within the complex, but it never became a key scene of the Cold War. The basic dynamics of the South Asian complex was not altered with the end of the Cold War. Since 2001, however, one may see the seeds of change, as India’s economic success and relative political stability strengthen its position vis-à-­vis Pakistan, a trend that has been reinforced by the US tendency to favour India, for example through opening up for collaboration on nuclear energy. Several analysts point to a possible disintegration of the Pakistani state, which is facing multiple internal threats from ethno- and religio-­political groups, as the factor that is most likely to upset the balance in the region.33 Afghanistan is important to the South Asian complex in several ways, most of which have to do with India–Pakistan enmity. Most importantly, Pakistan – undoubtedly the weaker party in the equation – looks to Afghanistan for ‘strategic depth’. In the case of an escalation of tensions with India, possibly a military attack, Pakistan would rely on Afghanistan for regrouping its forces.34 A worst-­case scenario, as seen from Islamabad, is a close alliance between Kabul and Delhi, which the Pakistani security establishment believes would be used for active cross-­border stabilization. A different concern is with ethnic unrest in Pakistan, particularly among the Pashtun population, and – related – the unsettled status of the Afghan–Pak border.35 In pursuit of a friendly Afghan government, which does not stimulate ethnic unrest in Pakistan, the latter has quite consistently favoured the radical Islamic groups. The prospect of Pashtun mobilization transcending the border is certainly a main concern for the Pakistani security elite, and an effective agreement on the status of the border would be a positive step. As Frédéric Grare points out, discussing the often counter-­intuitive alliances between Pakistan and various radical Afghan movements, ‘the Indo-Pak dispute is still [post–2001] the main determinant of Pakistan’s Afghan policy’.36 As Nathan also points out, this

256

War and State-Building in Afghanistan

bipolar security dynamic dominates the stage in the South Asian region, effectively precluding other issues from entering the agenda, and preventing the emergence of effective common security institutions.37 Against this background, let us move back to the final months of 1979. Pakistan was under the rule of military dictator Zia ul Haq, who had just executed his democratically elected predecessor, Zulfiqar Ali Bhutto, and was internationally isolated. Seeing the strong international reactions to the Soviet intervention in Afghanistan, Pakistan grasped the opportunity of establishing itself as a broker of international support to the Afghan resistance (Iran was not a candidate, given the 1979 ‘Revolution’). The US was a primary source of money and arms, but its support for the resistance was complemented by substantial contributions from other sources, particularly Saudi Arabia. Furthermore, the US also provided significant financial assistance to Pakistan directly. Pakistan’s military intelligence agency, the ISI, took control over the distribution of arms, financial support and training of personnel.38 The net effect was a solidification of the military role in Pakistani politics.39 For Pakistan, the Soviet intervention in Afghanistan provided a sorely needed opportunity to re-­engage as a full member of the international community. Despite providing some technical-­military assistance to the PDPA regime and maintaining political ties just like it had with previous regimes, India had minimal visibility in the Afghan conflict during the 1980s. Opening up its military bases to Soviet use was out of the question.40 The main concern here was probably not to provoke a reaction from the US – which was closely tied in with Pakistan for its Afghan mission – and this could prove detrimental for an India that otherwise was steadily enhancing its relative power vis-à-­vis Pakistan. On its side, Pakistan had been under massive international pressure following General Zia’s military coup in 1977, and the pursuant execution of his predecessor, Zulfiqar Ali Bhutto. The Soviet intervention made Pakistan the only possible neighbouring country that would collaborate with the West (Iran, the main US ally, had fallen under Islamist rule in February 1979), and provided the opportunity to be welcomed back as a member of the international community. Even from 1992, when the PDPA regime fell, India did not scale up its presence in Afghanistan. Pakistan, on the other hand, was heavily engaged, brokering two failed peace agreements between the former resistance groups, but was increasingly frustrated by the inability of its main protégé, Gulbuddin Hekmatyar, to act constructively and contribute to a more Pakistan-­friendly government. Rather, it was the Pakistani-­sceptics of what would later become known as the Northern Alliance – Jamiat-­e Islami, Hezb-­e Wahdat and Jombush-­e

Afghanistan’s Complex Neighbourhood Relations

257

Milli – that had the upper hand in government. This frustrated the Pakistanis to the extent that when the Taliban emerged – apparently as a spontaneous protest to local warlord repression – Pakistan was quick to start supporting the new movement. The declining US interest in Afghanistan had served as an open invitation to Pakistan to further strengthen its grip.41 In the latter half of the 1990s, Pakistan’s influence in Afghanistan was at an all-­time high, despite serious concerns with the ongoing fighting in the north, as well as the Taliban’s disregard for advice from Islamabad. With increasing international attention to Kashmir, and pressure on Pakistan to prevent insurgent groups from training and operating on its territory, Afghanistan also became useful as a base for various groups engaged in Kashmir. In the process, networks between Afghan and Pakistani radicals, as well as with Al-Qaeda activists, were strengthened. The Pakistani intelligence worked closely with various radical groups, seeing them as a foreign policy tool.42 The Indians, of course, were well aware that many of the armed insurgents operating in Kashmir and elsewhere were trained in Afghanistan, under Pakistani guidance, and took the first steps to strengthen its ties with the de jure government who by the late 1990s controlled only some 10 per cent of the country’s territory. In the aftermath of the 2001 intervention, the India–Pakistan rivalry related to Afghanistan has taken on an entirely new form. Already in the spring of 2002, Indian engineers were active in exploring opportunities for regulating the Kunar River, which runs along the Pakistani border and is a main source of irrigation for the agriculturally super-­important Punjab plains, without any attempt at informing the Pakistanis. Indian visibility in Afghanistan went from low to high, as the country built up an extensive diplomatic presence, with a large embassy in the capital, and four consulates, including in the cities of Jalalabad and Kandahar, both of which are within close reach of the Pakistani border. India also acted quickly in becoming a significant contributor to aid programmes, with a multifaceted and highly visible effort that compared in scope only to the larger Western donors. India was warmly welcomed by the new regime. Many of the ministers, Karzai included, had lived in India, and they had good connections and positive sentiments for the Indians. The new Indian presence was of a scope that clearly signalled long-­term interest. Pakistan, on the other hand, had unwillingly had to accept the demolition of the Taliban regime. Given the overarching security orientation, it made things considerably worse to observe the installation of a new government cultivating its ties to India, while it was seeking to keep Pakistan at arm’s length. In the words of one Pakistani observer, a former military officer turned politician who this

258

War and State-Building in Afghanistan

author met in Peshawar in 2006: ‘Pakistan has lost everything’. By 2004, it was widely believed, also in Pakistani decision-­making circles, that India was both funding and arming insurgents in Baluchistan as well as in Waziristan, stirring the pot in both of the civil conflicts that Pakistan was engaged in.43 The stand-­off between India and Pakistan is a staple, but the balance of influence in Afghanistan has been turned upside down since 2001. Given the almost existential character that the relationship with Afghanistan has for Pakistan, the consequences, both for Afghanistan and for South Asia at large, may be substantial.44 Developments in Afghanistan also paralleled larger trends, no less worrying from a Pakistani perspective. Pakistan has always been the smaller brother in the relationship, and the loss of East Pakistan (now Bangladesh) in 1971 had made that even clearer. But the 1990s and early 2000s were characterized by considerable progress, economically and otherwise, in India, while Pakistan stagnated. In Islamabad, the fact that India was solidifying its position as the undisputable regional hegemon was not taken lightly. It made things no better that the US clearly pursued regional stability in South Asia through a further strengthening of India as a hegemon.45 Both countries are self-­declared nuclear powers that stand outside the non-­proliferation regime, yet it was the Indians that got a deal with the US, effectively permitting US transfer of dual-­purpose material and equipment. For the US, counterbalancing China was probably a key motive in striking the deal. Yet, seen from Islamabad – where China is considered a major ally, pivotal to Pakistan’s effort to balance Indian supremacy – the deal was a real blow. First placed on the agenda in June 2005, almost five years later, by early 2010, there had been no traces of a similar offer to Pakistan. While the relationship between the US and Pakistan has been fluctuating considerably, the first long decade following the 9/11 terror attacks are, in hindsight, a downward spiral. In 2001–2, Pakistan declared itself a loyal ally in the war on terror. By 2012, it is not only the uneven US treatment of India and Pakistan that has made things difficult. Key elements in the Pakistani state apparatus have continued to support the Taliban and other insurgent groups in Afghanistan, the extent of which has gradually become clearer. The hunting down of Osama bin Laden in Abbottabad, close to the main Military Academy, dealt another blow to US confidence in Pakistan and its security apparatus, regardless of whether there is evidence of complicity.46 The Pakistani Government, on its side, is under solid pressure from a US-hostile public, and is not being helped by drone attacks and military raids that at worst kill Pakistani soldiers, all of which is understood as a violation of sovereignty. It is against this dim background that Anatol Lieven, in the conclusion to his 2012 book, Pakistan: A Hard Country, prescribes a

Afghanistan’s Complex Neighbourhood Relations

259

multifaceted US engagement with Pakistan – the alternative is worse – and warns against US conservatives who believe India can be used to control Pakistan. If we go beyond the US influence, what about other security relationships with various powers at a lesser distance? For South Asia, the China factor is important. Pakistan and China are long-­standing allies, who exchange mutual diplomatic support on conflictual issues (from Kashmir to Tibet), collaborate economically (including the Karakorum Highway and the new Gwadar deep-­sea port at the Hormuz strait) and in the military domain. The two countries share the perception of India as a main threat, and for Pakistan the alliance with China is clearly aimed to counterbalance Indian hegemony. Yet, even though China shares borders with both India and Pakistan, has a superpower stature and pivotal security interests in eastern parts of Asia, it is hardly an integral part of the South Asian security dynamic. In relation to Afghanistan, an alliance has developed since the last part of the 1990s between India, Iran and Russia, joining together one key actor from each of the three surrounding regions. For this trio, the shared interest lies in stemming the tide of radical Islam and the Taliban, and they share a deep suspicion of Pakistan, including of its engagement with radical groups. In Afghanistan, this has resulted in what appears to be a loosely coordinated effort to support the so-­ called Northern Alliance, within which the three countries each have their favourite groups. The trio of India, Iran and Russia seems a marriage of convenience, but with interesting implications for present-­day Afghanistan, and even more so for a future one where the US-led alliance has withdrawn.

The Persian Gulf and Afghanistan Within the Persian Gulf, the security dynamic has traditionally been tripolar – between Iran, Iraq and Saudi Arabia. The 2003 US intervention in Iraq did affect the regional dynamic in fundamental ways, bringing about a massive US military presence and regime change in Iraq, and producing considerable Iranian influence on Iraqi politics. By 2014, the long-­standing Iran–Saudi rivalry persisted in the region, but seemingly with less direct influence on Afghan politics than was the case prior to 2001. The Saudis have become less engaged, while Iran’s main concern now seems to be with a player outside the region, the US, with pressure related to the nuclear issue mounting. Somewhat ironically, the US and Iran have common interests in Afghanistan, and Iran contributed significantly to make the 2001 Bonn Peace Agreement possible, only to be

260

War and State-Building in Afghanistan

rewarded by the US declaring Iran to be part of the ‘axis of evil’ a few weeks later. Against that background, ideas to build on common interests in Afghanistan to pilot a US–Iranian rapprochement face poor odds. Iran does want a stable neighbour to its east, also in order to stem direct threats such as the drugs trade and illegal migration. Yet, with a massive US presence in the country, and continued unease about the relationship with Saudi Arabia, Iran continues to hedge its bets by maintaining relationships to multiple parties. Beyond the region, this has fostered an alliance of three countries – Iran, India and Russia – that over the past few years has seen sufficiently common interests to join hands in supporting the anti-Taliban Alliance. The main tensions within the Persian Gulf are over the political-­ideological leadership in the Islamic part of the world, between Iran and Saudi Arabia – Shia- and Sunni-dominated respectively. The political-­ideological contention is not only a war of ideas, it does materialize as firm security threats: the Iran–Iraq War (1980–8); the Iraqi intervention in Kuwait (1991); as well as in more low-­ key conflicts throughout the region, exemplified more recently in the escalation of conflict in Yemen, where the Saudis see a strong Iranian hand that threatens their own stability, and in the 2011 Saudi-­assisted clampdown on ‘Arab Spring’ protesters in Bahrain. Identity issues are important, and are brought to the fore through the existence of population groups that span state boundaries, such as the Kurds, and the existence of religious minorities in many of the region’s states (such as the Shia in northern parts of Yemen). Control over energy resources, transport routes, and pricing of oil and gas are always high on the agenda. While there is a large set of issues specific to the Gulf sub-­complex, the interdependence with the Levant – particularly in the Israeli–Palestinian conflict – is considerable. Actors such as Syria play a key role in both sub-­complexes, yet it is most durably integrated in the Middle East. As Jubin Goodarzi has convincingly demonstrated, the Iranian–Syrian alliance has, despite massive ideological differences, been fairly stable over the past three decades, its foundations laid in the 1979–82 period.47 The future of this alliance remains uncertain in the wake of the Arab Spring and the unpredictable nature of Syria’s future polity. The implications for the balance of power within the Gulf could be large. The Gulf states identify with the Palestinian – and the Arab – cause, besides being a substantial source for various political and religious entities throughout the larger region and beyond. Nonetheless, the Persian Gulf region is here seen to encompass Iran, Iraq and Kuwait, as well as the countries on the Arabian Peninsula: Bahrain, Oman, Saudi Arabia, Qatar, the United Arab Emirates and Yemen. The overarching enmity between the three, now two dominant states,

Afghanistan’s Complex Neighbourhood Relations

261

with a fairly constant overlay of superpower politics, has effectively prevented the emergence of functioning regional security institutions.48 Mutual confidence is low, each country being preoccupied with its own security in the most direct sense. In the Buzan and Wæver scheme, the Persian Gulf belongs to the larger Middle Eastern security complex, which stretches from Morocco in the west to Iran in the east. Buzan and Wæver argues that within this large area, security interdependencies are of a scope that merits seeing it as one regional complex, but at the same time points to three distinct sub-­complexes: the Maghreb, the Levant and the Gulf.49 This conceptualization does illustrate the inherent problems with standardizing definitions in the analysis of international politics. While the larger Middle East may be sufficiently interconnected to be tagged as a security complex, it is at a considerably higher level of abstraction than many of the other complexes in the scheme. Several of the sub-­complexes – the Persian Gulf being a case in point – potentially have a higher degree of interconnectedness than many complexes elsewhere in the world. In the interest of comparability, we focus on the Persian Gulf sub-­complex, while keeping in mind the particular status of the larger Middle Eastern area on processes within the region. Within the Persian Gulf area, the two main actors in relation to Afghanistan are Iran and Saudi Arabia, whose battle for Islamic leadership has also been played out there.50 Saudi Arabia has a long-­standing relationship with Pakistan, and has used its financial strength to gain influence in many parts of the Islamic world. After the 1979 Revolution, Iran has actively promoted its potential as the main political force in the Islamic world, generally with limited success.51 Iran’s effectiveness has been hampered by the limited appeal of its radical Shia ideology, and the country’s engagement in conflicts of a nearly existential character – the war with Iraq in particular – has encouraged a good measure of pragmatism in relations with other states. As a major superpower, the US has fostered a close alliance with the Saudi rulers, focusing both on energy and security, with one of its primary ambitions being to stem Iranian influence. Undoubtedly, Iran’s standoff with the US, from 1979 onwards, has also affected its policies on Afghanistan. Iran has taken a special interest in Afghanistan’s Shia minority, who are in large measure identical to the Hazara ethnic group. The Hazara occupy Afghanistan’s mountainous central areas. Among the country’s larger ethnic groups, the Hazara are the only one that does not have a population with the same identity across the border (there are Hazara migrant communities in Pakistan and Iran, but interspersed with other dominant identity groups). For

262

War and State-Building in Afghanistan

this reason, the religious Shia identity – shared with Iran’s majority, a state-­ bearing identity after Khomeini came to power in 1979 – attained huge significance.52 When the Hazara found that the Afghan exile party structure, set up in Pakistan in the early war months, did not have room for them, the only option was to seek support in Iran.53 The reception was lukewarm, but radical Islamist groups did receive considerable support, and among the Hazara, the 1980s were characterized by internal struggles that probably cost many more lives than the communist versus resistance struggle. This has come in tandem with a social, political and economic revolution for many Hazara, who as a group now have an unprecedented representation in Afghan politics. Despite the establishment of the Unity Party – Hezb-­e-Wahdat – in 1989, internal tensions between Islamist and ethnicist projects ran high.54 Iran, however, while continuing to cultivate its ties with Shia groups, expanded the set of favoured groups already around 1992 to encompass all of the Farsi-­speaking groups (later known as the Northern Alliance). Iran’s support to Afghan groups have been overwhelmingly based on religious (Shia) or cultural cum linguistic (Farsiwan) ties, just as in the Middle East, but still reflects less an expression of solidarity than a pragmatic choice on the Iranian side on how to most effectively secure its own interest. While Afghanistan itself has never been the main concern in Iranian foreign policy, it has been important not to exacerbate insecurities by getting an adversary – in the form of the US, Pakistan, or an alliance of both – firmly positioned on its eastern border. Following the revolution in February 1979, Iran had few friends in the world. The US had lost its major ally in the region, and when the Soviets intervened in Afghanistan at the end of the year, the reactions were harsh. The Iranian Revolution, and particularly Khomeini’s ‘neither East nor West’ doctrine, contradicted the bipolar paradigm in world politics, in sharp contrast to the war in Afghanistan, which quickly evolved into a classic Cold War confrontation. With the revolution, the rivalry with Saudi Arabia over leadership among Muslim states was refreshed, inspired by Iran’s credentials as an ‘Islamic state’.55 At the same time, Iran’s principal support to the Iraqi opposition had led it into open war with Iraq (1980–8), and it became important for Tehran not to alienate the Soviets into supporting Iraq in that war.56 Hence, Iran maintained diplomatic relations with Kabul, and found itself forced to constrain its support for the Afghan resistance, despite ideological affinities, leaving the main supportive roles to the US, Pakistan and Saudi Arabia. The exception to the rule was the Afghan Shias, who were now permitted to set up offices in Iran. Rivalry with the Saudis would have invited a much more active Iranian role, but locked into an

Afghanistan’s Complex Neighbourhood Relations

263

all-­encompassing war with Iraq, as well as being on hostile terms with the US, Iran had to act pragmatically and constrain its engagement. The 1990s started with the US-led war in Iraq, supported by Saudi Arabia and other countries in the region. Iran, despite its longstanding rivalry with Iraq, did not welcome this display of superpower force in its neighborhood; neither did it welcome the Saudi collaboration in the war. To Iran, the relationship with Russia remained a priority.57 Even during the years of armed conflict in Tajikistan, Iran took great care not to alienate Russia, in spite of its own engagement with the Islamist opposition.58 In Afghanistan, the fall of the PDPA regime had given way to new rounds of warfare, this time between former partners in the resistance. Iran moved from the relatively narrow approach of the 1980s, where support to Islamist Shia groups was dominant, to a broader approach, building relationships with most of the non-Pashtun groups, and particularly the Tajik-­dominated Jamiat-­e Islami.59 In the latter half of the 1990s, Iran’s pattern of support coincided almost perfectly with that of India and Russia. Saudi Arabia, on the other hand, continued to work in close alignment with Pakistan, which by the mid-1990s meant support of the Taliban, a group that the Iranians detested (the Saudis were later to reassess its policy, as Taliban–Al-Qaeda ties grew stronger). Again, we see that patterns of conflict and cooperation endure. The policies pursued by Iran and Saudi Arabia, both significant actors in the Afghan conflict, largely reflect concerns within their own security complex (including its superpower overlay). The most dramatic episode between Iran and Taliban-­ruled Afghanistan occurred in the fall of 1998, with the Taliban’s killing of nine diplomats at Iran’s consulate in Mazar-­e Sharif. The Iranians were outraged, but also saw that this was an opportunity to get international sympathy.60 Troops were mobilized along the border, and the Taliban responded by regrouping their forces. On the home front, Iran emphasized the Taliban onslaught on non-Pashtun populations, both Shia and other. The underlying Iranian conviction, however, was that Saudi Arabia, Pakistan and the US were working closely together to consolidate Taliban power. The key to defusing the situation was not found in the Iran–Taliban relationship, but in addressing regional security dynamics, convincing the Iranians that the US and Saudi Arabia opposed the Taliban (and Pakistan’s support for it), manifested in the 1999 Security Council sanctions regime.61 The 2001 intervention in Afghanistan brought a semi-­permanent US presence to the neighbourhood. Most of the states in the region saw this as a mixed blessing, but it was particularly unwelcome for Iran, who after having collaborated extensively in facilitating the 2001 military campaign as well as in brokering the

264

War and State-Building in Afghanistan

Bonn Peace Agreement, found itself declared part of the ‘axis of evil’ by Washington in early 2002.62 Things got worse with the Iraq intervention in 2003; now Iran found itself surrounded by states hosting the forces of an increasingly hostile US. The debate on Iran’s nuclear programme loomed in the background, and the threats and accusations exchanged between Iran and Israel reached unprecedented levels. The real concern was that the US – or Israel, with US acceptance – would take the military campaign to Iran. For Iran’s power elite, it has been painfully evident that the US would very much like to see regime change in Iran, and that various actors in the region – both governments and non-­state actors – could be seen as useful instruments in bringing it about. Ultimately, as Barzegar points out, the objective is to redefine the regional security order, constraining Iran’s role in it as much as possible.63 Ironically, as the immediate US success in Iraq was replaced by massive resistance, the US inclination to move militarily against Iran was muted, while Tehran now found itself freed of both the Taliban and the Saddam regime. Iran gained considerable influence in post-2003 Iraqi politics through its careful manoeuvring with Shia groups of variegated ideological orientation.64 Saudi– Iranian relations remained tense, but the Saudis had a limited engagement on the Afghan scene in the aftermath of their Taliban adventure of the 1990s, which had turned out to be contrary to their own interests. The US–Afghan Declaration of Strategic Partnership, signed in Washington in May 2005, fed into Iranian concerns that the US could use its presence in Afghanistan to threaten Iran. Karzai was positive to sign a pact with Iran, but was instructed by his Washington counterparts not to sign such an agreement, which caused further concern in Tehran. Later negotiations on long-­term US military presence in Afghanistan added to Tehran’s unease. Iran kept its Afghan engagement up, both at the central level, where many of its old associates are in key positions, and in the areas close to the Iranian border, where it is heavily engaged in reconstruction and development. In relation to Obama’s interest in a regional approach to Afghanistan’s problems, it has been suggested that the commonality of interests between the US and Iran in Afghanistan can serve as a platform to foster confidence between the two countries.65 This, however, has proven a hard sell to the Iranians, who felt betrayed when their cooperation during the 2001 intervention, and the political negotiations that followed, were followed by the ‘axis-­of-evil’ categorization of Bush. The inclination since is to go for a comprehensive political process in which all outstanding issues are addressed, or nothing.66 In Afghanistan, there is a need for Iran to be present, in order to know what the US and its allies are

Afghanistan’s Complex Neighbourhood Relations

265

doing, potentially as an opening to a constructive dialogue with the US. With Iraq in turmoil, the tripolar rivalry of the Gulf complex is reduced to a bipolar one, and in post-Taliban Afghanistan, Saudi Arabia’s miscalculations are now converted into gains for Iran, which it has since worked hard to secure. Albeit the changing role of Iraq transforms the basic plot from a threesome to a twosome rivalry, there is considerable continuity to the security dynamic in the Persian Gulf. Iran–Saudi relations remain at the core, but always against the backdrop of strong superpower involvement. Even for Iran, there is a remarkable consistency in the regional security orientation between the Shah and the post-1979 Islamist regime, despite the turnaround in relations with the US.

Central Asia and Afghanistan The Central Asian states, having gained independence with the dissolution of the Soviet Union in 1991, all find themselves in a squeeze, on the one hand depending heavily on Russia as the heir to the Soviet Union and a regional hegemon, yet also seeking to strengthen their newly built independence. Of the five states, three – Tajikistan, Turkmenistan and Uzbekistan – border Afghanistan to their south, while Kazakhstan and Kirgizstan do not. Afghanistan to the south is a problem, seen as a source of drugs and unrest. While Afghanistan does not overshadow the preoccupation with regime stability and relations with Russia, it does affect both, not least through its role in hosting and fostering militant groups that are a threat also in Central Asia. Among the Central Asian states, there is no strong sense of common security, and cooperation is mainly coming about through the roles of Russia and China. Uzbekistan aspires to hegemonic status within Central Asia proper, but is challenged by Kazakhstan, which has similar ambitions. Also in the relationship between the five Soviet successor states, there is little that is converted into an engagement in Afghanistan. There are relationships with various Afghan groups, but despite the common ethnic background of populations on both sides of the border, some 70 years of effective separation under Soviet rule has had a deep impact on networks, and it has also fostered rather incompatible political cultures. The detrimental effects on Afghanistan of the security dynamics within Central Asia may be of a lesser scope than is the case for both South Asia and the Persian Gulf, yet there is significant concern over the presence of US and NATO troops, and over Afghan militarization, and in particular Russia does keep a close eye and maintain local connections.

266

War and State-Building in Afghanistan

Central Asian security dynamics, argues Roy Allison convincingly, is best characterized as a form of ‘virtual regionalism’: ‘To Central Asian rulers the central objective of such [i.e. security] coordination through macro-­regional frameworks has been to reinforce regime security.’67 In post-Soviet Central Asia, what we have now are states-­in-the-­making, with authoritarian rulers who heed an authoritarian Soviet legacy, and whose main security concern is with domestic threats to regime stability. As to extra-­territorial security threats, the key ones are transnational, in the form of radical groups that operate across state boundaries. These groups are chiefly of a radical Islamic character, as other political alternatives have been effectively muted.68 Several of the groups in question operate in more than one state, some of them beyond the Central Asian region, and some (i.e. Hizbut-Tahrir) have a global outreach. However, the preoccupation with transnational groups is also derived from the concern over regime stability. Similarly, each of the states in Central Asia proper – Kirgizstan, Kazakhstan, Tajikistan, Turkmenistan, Uzbekistan – have sought to maintain good links to Russia, who is effectively a guarantor for the regimes of the day, but at the same time also a regional hegemon from which is wanted a certain independence. There is considerable continuity between the Soviet and the post-Soviet eras, and each of the five states struggles to balance fostering independence from, yet continuing to uphold the still so crucial relationships with, Russia. In addition, both the US and China seek to strengthen their influence, and post-Soviet Central Asia can be said to have an overlay of superpower competition, with China, Russia and the US all seeking to expand their economic and political influence.69 Progress towards genuine regional cooperation has been hampered by civil war and unrest. Regional cooperation is hampered by a lack of mutual trust between the countries in the region, and by rivalry between Uzbekistan and Kazakhstan for becoming the dominant regional power.70 Ultimately, for regional cooperation to evolve, it is a precondition that there is a sense of common security challenges among the states in the region.71 This has barely been the case in the first two decades following 1991. For Buzan and Wæver, the five Central Asian states are part of the CIS regional complex, within which Russia has lost control over the states referred to as ‘Eastern Europe’ during the Cold War.72 Russia strives to maintain hegemony within a region that encompasses four distinct sub-­complexes – the Baltic states, the ‘western’ states (Belarus, Ukraine), the Caucasus and Central Asia – commonly referred to by Russian leaders as the ‘near abroad’. Each of the four sub-­regions has a strong security dynamic of their own, but what unites them is

Afghanistan’s Complex Neighbourhood Relations

267

that Russia constitutes a hegemon in the sub-­complex, which is also reflected in the way that alliances aiming to balance Russia’s dominance transcend boundaries between sub-­regions. Just as above, where we focused on the Persian Gulf as a sub-­complex of the Middle East, we will here be focusing on Central Asia as a sub-­complex of the CIS. There is a key difference between the two cases though, in that the former had three sub-­complexes with distinct memberships, whereas the latter is constituted by four sub-­complexes that share a common hegemon. It is also worth noting that several analysts within an RSC tradition talk about regional mini-­complexes in Central Asia – such as the conflict-­ridden Fergana valley, at the intersection of Kazakhstan, Tajikistan and Uzbekistan – even though we will not do so here.73 Afghanistan is not a key concern for the Central Asians, apart from being a place where transnational militants may find sanctuary and alliance partners, as they did when the Taliban were in power.74 This possible threat is most relevant to the states bordering Afghanistan – Tajikistan, Turkmenistan and Uzbekistan – but only the latter has engaged seriously to affect political developments there, through its support to the Uzbek warlord Rashid Dostum. Hence, at the most general level, the relationship with Afghanistan is remarkably stable, reflecting primarily a concern that challengers of domestic power can find support there. As such, it is a virtual blue copy of Russian concerns with radical Islamic movements in the so-­called ‘near abroad’, of which Central Asia is a key part.75 A different concern relating to Afghanistan is drugs, which are both sold in, and transited through, Central Asia, with considerable impacts. While this is a concern to the Central Asian states, it does seem that a preoccupation with security in its most traditional sense, as opposed to wider concepts of a human security type, largely keeps such issues off the security agenda. Prior to the 1991 implosion of the Soviet Union, the Central Asian republics did not have a foreign policy of their own. As discussed above, the Soviet Union ended up in Afghanistan not because it had aggressive geopolitical designs of gaining access to hot waters in the Indian Ocean, as the dominant interpretation in the 1980s had it, but because many of the power holders saw it as the only way of saving a communist ‘revolution’ in a neighbouring country from failure. The concern over a potential radicalizing impact on Muslims within Central Asia was not part of that original calculus. As Ahmed Rashid points out: ‘Tens of thousands of Central Asian conscripts fought for the Soviet army in Afghanistan and many were deeply affected by the mujahedin’s Islamic zeal.’76 One example is Juma Namangani, the later leader of the Islamic Movement of Uzbekistan, who served as a paratrooper in Afghanistan in the 1980s.77 The dimensions of such

268

War and State-Building in Afghanistan

phenomena among Soviet Central Asians are uncertain, yet it seems that we do here have a parallel to the so-­called blowback effect that hit the US in the 1990s, when many of its former allies in the Afghan war turned against it. While largely symbolic at the time, it is also worth noting that Afghan mujahideen, encouraged by Pakistani intelligence, conducted raids in Central Asia, which has had lasting impacts on mutual confidence. When the former Soviet republics gained independence in 1991 as a result of the dissolution of the Soviet Union, none of them had a developed policy on Afghanistan, and all of them had more immediate worries in the domestic transition. Nonetheless, when Tajikistan was thrown into civil war in 1992, major parts of the armed opposition found support in Afghanistan, where there were also refugee camps, which served as safe havens for the fighters. By 1997, when the Tajikistan peace agreement was signed, the Afghan Government, under immense pressure from the Taliban, turned around and backed the agreement, in return for Russian support, including access to the Kulyab air base in southern Tajikistan.78 Also, the Islamic Movement of Uzbekistan (IMU) had bases in Afghanistan in the 1990s, from where it launched attacks into Central Asia. Despite the concern with Afghanistan as a possible retreat for its opposition, however, the Central Asian states were primarily concerned with controlling domestic unrest and militancy, and with maintaining the international links – primarily with Russia – that they needed to succeed. The 1990s became an unruly period for Central Asia. The Russian presence was subdued, to the extent that by the end of the decade, there was reason to ask whether it had a privileged hegemon status. Other powers – China and the US, but also Iran, Turkey and others – sought to strengthen their influence in the area, exploiting Russia’s weakness to gain access to natural resources and new markets. Most of the challengers had limited success, China being the major exception. But as the Russian economy and military capacity started to recover just after the turn of the millennium, it also became evident that the political culture and networks of the Soviet era, combined with infrastructure, were effective forces in pulling the Central Asians in with Russia. Not surprisingly, the 2001 intervention that removed the Taliban from power in Afghanistan was supported by the Central Asians, who allowed use of its bases. But the Central Asian exposure to what MacFarlane has described as a ‘competitive pull’ from the three superpowers only intensified.79 Expectations of what to receive in return from the US for collaborating on Afghanistan were on the rise. US motivations in Central Asia were being questioned, and US insistence on democracy both in Afghanistan and the ‘greater Middle East’ has a bad

Afghanistan’s Complex Neighbourhood Relations

269

resonance with autocrat Central Asian leaders. Five years into the new millennium, US influence in Central Asia seemed to be waning. The Shanghai Cooperation Organization (SCO), spearheaded by China, with Russia and four Central Asian states (Turkmenistan is the exception, as it is in most regional initiatives) as members, has been gaining momentum, particularly in the new millennium.80 Starting out with a focus on border demarcation in the region, energy security has emerged as the main issue for the SCO, to the possible dismay of the US, and despite the great potential for Chinese–Russian conflict that goes with it. Although Afghanistan has now gained observer status in the SCO, the trend in Central Asia (and within SCO) is not to give further significance to Afghan affairs, despite the possible link with domestic threats. In 2005, the SCO called for the US to withdraw from bases in Central Asia, and by November the US withdrew from the Khanabad base in Uzbekistan (but maintained use of Manas airbase in Kirghizia). As the insurgency in Afghanistan and Pakistan intensified, it became increasingly clear that logistical access through the CIS would be helpful. Russia and the Central Asians kept the pressure on by signalling, yet frequently restraining, cooperation with the international forces fighting in Afghanistan. For Central Asia, the line of demarcation towards Afghanistan is even sharper than for the other two surrounding regions. This may seem counterintuitive, given the serious impacts of the Afghan situation on the region. Despite a common history, however, the long period of effective border control under the Soviet Union did have serious impacts on interaction at all levels. Since 1991, Russia has remained the key point of reference, greatly aided by Soviet infrastructure, political culture and networks, even though other powers have challenged Russian hegemony. Finally, there is also an obsession with domestic security, in the sense of securing regime survival. Yet, even though militant opposition groups have had close links to Afghanistan, the southern neighbour has not become a key security concern for the Central Asians. The inability of the five Central Asian republics to act in concert, both on Afghanistan and more generally, undermines their effectiveness, and has offered the opportunity for Russia to regain influence in the region.

Competing policy visions Nominally, there was broad regional support for the Afghan transition arrangement that was put in place after the overthrow of the Taliban in 2001.

270

War and State-Building in Afghanistan

Afghanistan’s neighbours took part in the treaty talks in Bonn, yet the resulting treaty did little to ensure constructive regional engagement, and worse, the intervening forces and the Afghan Government did little to pursue a regional compact in the years that followed. Afghanistan and its six neighbours met in December 2002, resulting in the ‘Kabul Declaration on Good-­neighborly Relations’. Despite this initiative, the Afghan Government – just like the US – dealt with the region one country at a time. In the case of Pakistan, presumed to be the main troublemaker, the instinct in Kabul seemed to be isolation rather than constructive engagement, inviting its arch-­rival India to play a significant role in Afghanistan’s reconstruction, and relying on a long-­term alliance with the US for securing itself against external threats. By 2011–12, transition – the international withdrawal and the full transfer of security responsibility to Afghans – had become the main preoccupation. Still, the dominant instinct in Kabul was to solicit US security guarantees, in a virtual attempt to disconnect the country from a troublesome neighbourhood through erecting military defences. President Karzai himself cultivated a vision of Afghanistan as an ally of the US on a par with NATO countries – a so-­called ‘major non-NATO ally’ – a wish likely to be fulfilled.81 The vision is to find its form in the ‘Strategic Partnership’ agreement between the US and Afghanistan, which will regulate post-­transition US military presence. The challenge, of course, is that none of the neighbours are particularly enthusiastic about a US military presence in Afghanistan. The vision of an Afghanistan with protection from a distant superpower, linked to the regional compact entailed in seeing Afghanistan as the ‘Heart of Asia’, informs the so-­called Istanbul process. Since 2009, this has become the main publicly known arena (there may be others) where Afghanistan’s neighbours have been meeting to discuss the neighbourhood architecture of the future. Turkey, who plays on its historic ties as well as its non-­confrontational military stance within the NATO operation is projecting itself as a future major actor in Afghanistan, and has been hosting a series of events. There has been close coordination with the US, and a number of other countries, including Germany and Norway, have played an important role in keeping the talks moving. By early 2014, the Istanbul process is still in existence, but progress seems minimal, as other and more pressing issues have diluted the agenda, and led to problems with securing the participation and commitment that is needed. From the Afghan Government’s perspective, an alternative vision would have been to pursue some sort of a neutrality status for Afghanistan. To comfort the concerns of its neighbours, this could take the form of a unilateral Afghan

Afghanistan’s Complex Neighbourhood Relations

271

guarantee of non-­aggression towards its neighbours. When this idea was floated publicly by former UN Special Representative of the Secretary-General, Kai Eide, he was firmly reprimanded by the President, who made it clear that this was not perceived as a viable option.82 Various forms of neutrality statuses have been relatively successfully pursued by a number of other smaller countries whose geopolitical location is complicated, including Finland, Switzerland and Turkmenistan.83 In the Afghan capital, neutrality has seemingly been taken to mean neutralization, in the sense of a serious infringement on state sovereignty, and there has been little thinking into whether there are neutrality formulas that could fit Afghanistan’s complicated situation. Neutrality, assuming the mode of analysis that relates Afghanistan’s problems with its neighbours to problems within their respective security complexes, is no silver bullet. Yet, it would be worth exploring more seriously, as a main concern of neighbouring countries now is the way a long-­term international presence in Afghanistan will affect their other security concerns, and if it would undermine the possibility of justifying engagement in Afghanistan with the threats it poses.

Conclusion While the dominant international debate on security has undergone a deep transformation – encapsulated in the notion of human security – the security discourses within all of the three regions surrounding Afghanistan remain traditional. Some of the concerns of neighbouring states, such as drugs and migration, relate to a broader security agenda, yet these issues are not conceptualized mainly as security related, and while important, they do not tend to dominate the agenda. Most regimes in the region are preoccupied with regime survival, in the face of internal threats, and with threats from other states within their respective regions, which they often see as existential. It is therefore not surprising that in none of the three regions do we see anything reminiscent of a security community, where the states within a region band together around common interests, based on mutual trust and embedded in strong regional institutions. Rather we see deep tensions, distrust and weak multilateral institutions. This creates a double-­bind for external actors wanting to contribute to genuine regionalism. Is the trick to affect attitudes to security so that regional collaborations can build on better foundations? Or is it to bring about concrete collaboration that will contribute to enhance trust through the power of example? Given the authoritarian orientation of many of the regimes in question,

272

War and State-Building in Afghanistan

which is at the root of the preoccupation with regime survival, it is hard to foresee genuine attitude change without political change. A related point is that security relationships seem remarkably robust. Geography matters, and states that are located side-­by-side develop relations – positive or negative – that are not easily escaped from. While distant powers – as we saw with the US and Russia in Afghanistan from the early 1990s – have the luxury of being able to disengage, those embedded in a long-­standing regional relationship cannot choose this route. The composition of regional security complexes are remarkably stable: the Persian Gulf region has changed the most, following intervention and regime change in Iraq; Central Asia is partially transformed as a result of the collapse of the Soviet Union (yet old patterns of interaction persist also there); and the India–Pakistan rivalry continues to overshadow everything else in South Asia. From this follows an important realization for distant interveners: they lack the ability to bring about quick change, just as they lack the ability for long-­term sustained engagement. For states in the region, one obvious strategy is to wait the distant interveners out. Post–2001 strategies of many of Afghanistan’s neighbours can easily be read in this light. Pakistan has declared its support for the war in Afghanistan (and at home), but continued to cultivate ties with Afghan insurgent groups. The Russians keep a very low profile, but continue to entertain their networks. There is no silver bullet that can bring about regional security cooperation, but it is clear that unless external actors succeed in triggering genuine regional initiatives, they will not be successful. More fundamentally, what is it about the security dynamics in Afghanistan’s larger neighbourhood that serves to virtually block any movement towards a concerted effort for strengthening peace and security in Afghanistan? Taking Buzan and Wæver’s thinking on Regional Security Complexes as a point of departure, this chapter focuses on the dynamic within each of the three regions that border on Afghanistan: South Asia, the Persian Gulf and Central Asia. The narratives of each region give considerable support to the view that the three regions surrounding Afghanistan are distinct security regional complexes. We have also examined the security concerns in the surrounding states and regions that relate to Afghanistan in a most direct sense – transnational militants, migration, drugs – which are important, yet do not have the same direct, existential and enduring character as the key security issues within each complex. Seen from the regions that surround it, Afghanistan remains at best a secondary concern. One implication is that the post-2001 Afghan Government has had limited room for manoeuvre in affecting its regional neighbourhood. Not only is

Afghanistan’s Complex Neighbourhood Relations

273

the security policies of neighbouring countries anything but Afghanistan-­ centric, the country also falls short on institutional and human capacity, and it has little to offer its neighbours. Only if the neighbouring states develop a common understanding of how Afghan instability threatens their security, and come to share the view that the best cure is a stable Afghanistan, can we expect the emergence of a concerted effort among its neighbours. But an RSC-guided analysis is also less than satisfactory in several ways. Perhaps most importantly, it tends to underplay the weight of the transnational networks that span Afghanistan’s borders and which ties it in with the security dynamic of its neighbours. The alternative perspective offered by Barnett Rubin and associates, under the heading Regional Conflict Formations (RCFs), emphasizes such networks, hence placing Afghanistan at the core of a larger regional formation rather than seeing it as an insulator between distinct regions.84 The problem with the Rubinian RCF perspective, however, is that it presumes that the relationships between states match transnational patterns of non-­state interaction. But political decision-­makers in the neighbourhood, even if they acknowledge non-­state and transnational threats, think of security in conventional terms, focusing on relations between states. Patterns of stately interaction – amiable or not – also define the pattern for state-­based responses. This is why, even if transnational threats may be becoming increasingly important, states in the Afghan neighbourhood remain preoccupied with region-­specific threats. While Buzan and Wæver’s RSC approach can be criticized for dealing with the transnational primarily from the perspective of states, it does help us understand the critical question of why states in the neighbourhood, despite common transnational threats, are loath to engage in constructive cooperation. A number of other criticisms of the RSC approach are easier to tackle. One is that such an approach risks overplaying the strength of a region’s borders. Rather than seeing Afghanistan as distinct from South Asia, one may talk about degrees of inclusion, seeing the country as somewhat peripheral, yet a member (as it would also be in Central Asia and the Persian Gulf regions). Undoubtedly, Afghanistan’s importance in the dynamic of the surrounding regions may be considerable, but its importance varies fundamentally over time, and it is never the main issue. A second critique would be that states in various regions may form strong ties. Cases in point would be the India–Iran–Russia alliance in Afghanistan or the Iran–Pakistan enmity. Both of these, however, have proved unstable, and are better understood as tactical alliances – or enmities – driven by deep-­seated relationships within the respective regions. There is also the question

274

War and State-Building in Afghanistan

of the role of superpowers and major powers. In the case above, we have seen that superpower intervention – as underlined by the US withdrawal from Afghanistan by the early 1990s – is inherently unreliable: distant powers come and go at their own discretion, neighbours stay put. In essence, the application of an RSC approach – where the dominant trend is to see the country as the core of a larger (conflictual) region – to the case of Afghanistan highlights its analytical value. The approach helps us distinguish between fundamental security relationships and fluctuating alliances. It gives primacy to the regional setup, but roots this in the insistence on agency of states and other key actors. It may come across as deterministic, even defeatist. If the rather unhelpful engagement of Afghanistan’s neighbours in the country’s conflict over the past 30 years is motivated by deep-­seated conflicts in their respective regions – as in the case of Pakistan’s existential fear of India – then it seems to be a long and complicated way to resolution that presumes the building of functioning security communities in the surrounding regions. Yet, an RSC perspective may also inform a more optimistic perspective: If neighbours’ involvement is not first and foremost inspired by issues that have to do with Afghanistan, then there must be ways in which to ensure that their conflicts are not played out on Afghan ground. In the case of Pakistan and India, for example, could the existential enmity between the two manifest itself in less destructive ways, possibly also on a different playing field? It may well be worth investing more brainpower in exploring an alternative path, guided by the idea of a unilateral Afghan guarantee of non-­aggression towards its neighbours.

Notes 1 This chapter is a revised and expanded version of my 2010 article ‘Caught in the Middle: Regional Perspectives on Afghanistan’, Comparative Social Research, vol. 27, pp. 277–305. My research for the chapter has been financed by the Norwegian Ministry of Foreign Affairs (MFA) as part of the project ‘Afghanistan in a Neighbourhood Perspective’, and by the Research Council of Norway as part of ‘State Failure and Regional Insecurity’, a collaborative strategic institute programme involving the Peace Research Institute Oslo (PRIO) and the Norwegian Institute for International Affairs (NUPI). Halvor Berggrav, Jonas Gräns and Farrid Shamsuddin have provided excellent research assistance. I have benefited greatly from discussions with Shahrbanou Tadjbakhsh, my main collaborator in the MFA-funded project. Readers interested in a more comprehensive discussion, rooted in a similar

Afghanistan’s Complex Neighbourhood Relations

275

analytical framework, should consult Kristian Berg Harpviken and Shahrbanou Tadjbaksh, Afghanistan Caught in Regional Insecurity (London: Hurst, forthcoming). 2 Barry Buzan and Ole Wæver, Regions and Powers: The Structure of International Security (Cambridge: Cambridge University Press, 2003). 3 Harpviken and Tadjbaksh, Afghanistan Caught in Regional Insecurity. 4 Barry Buzan, People, States and Fear: An Agenda for International Security Studies in the Post-Cold War Era (Brighton: Harvester Press, 1983). 5 Barry Buzan and Ole Wæver, ‘Liberalism and Security: The Contradictions of the Liberal Leviathan’, Copri Working Paper 23–1998 (Copenhagen: Copenhagen Peace Research Institute, 1998); Buzan and Wæver, Regions and Powers. 6 Buzan and Wæver, ‘Liberalism and Security’, p. 201 7 Buzan and Wæver, Regions and Powers, p. 48 8 See David A. Lake, ‘Regional Security Complexes: A Systems Approach’, in D.A. Lake and P.M. Morgan (eds.), Regional Orders: Building Security in a New World (University Park, PA: The Pennsylvania State University Press, 1997). 9 Buzan, People, States and Fear. 10 Buzan and Wæver, Regions and Powers, p. 476. 11 Buzan and Wæver, Regions and Powers, pp. 466–8. 12 Buzan and Wæver, Regions and Powers, pp. 471–2. 13 Buzan and Wæver, Regions and Powers, p. 483. 14 Buzan and Wæver, Regions and Powers, p. 111. 15 Ahmed Rashid, Taliban: Islam, Oil and the New Great Game in Central Asia (London: I.B. Tauris, 2000). 16 Michael Greenfield Parthem, ‘The Buffer System in International Relations’, Journal of Conflict Resolution, vol. 27, no. 1 (1983), pp. 3–26. 17 David B. Jenkins, ‘The History of Afghanistan as a Buffer State’, in John Chay and Thomas E. Ross (eds), Buffer States in World Politics (Boulder, CO: Westview Press, 1986), p. 178. 18 William C. Rowe, ‘The Wakhan Corridor: Endgame of the Great Game’, in A.C. Diener and J. Hagen (eds), Borderlines and Borderlands: Political Oddities at the Edge of the Nation-State (Lanham, MD: Rowman & Littlefield, 2010). 19 Ashraf Ghani, ‘The Afghan State and its Adaptation to the Environment of Central and Southwest Asia’, in H. Malik (ed.), Soviet–American Relations with Pakistan, Iran and Afghanistan (New York: St. Martin’s Press, 1990), pp. 310–32, see especially p. 323. 20 Ludwig W. Adamec, Afghanistan’s Foreign Affairs to the Mid-Twentieth Century: Relations with the U.S.S.R., Germany and Britain (Tucson, AZ: University of Arizona Press, 1991). 21 Louis Dupree, Afghanistan (1973, second ed., Princeton, NJ: Princeton University Press, 1980), p. 511; Richard S. Newell, Politics of Afghanistan (Ithaca, NY: Cornell University Press, 1972).

276

War and State-Building in Afghanistan

22 Lawrence Ziring, ‘Buffer States on the Rim of Asia: Pakistan, Afghanistan, Iran and the Superpowers’, in H. Malik (ed.), Soviet–American Relations with Pakistan, Iran and Afghanistan (Basingstoke: Macmillan, 1987). 23 Odd Arne Westad, The Global Cold War: Third World Interventions and the Making of Our Times (Cambridge: Cambridge University Press, 2005), pp. 299–300. 24 Dupree, Afghanistan, pp. 770–1; Anthony Hyman, Afghanistan under Soviet Domination, 1964–91 (1982, 3rd ed., London: Macmillan, 1992), pp. 76–7. 25 M. Nazif Shahrani, ‘Introduction: Marxist “Revolution” and Islamic Resistance in Afghanistan’, in M.N. Shahrani and R.L. Canfield (eds), Revolutions and Rebellions in Afghanistan (Berkeley, CA: Institute of International Studies, University of California, 1984), pp. 3–57, see especially pp. 11–25. 26 Henry S. Bradsher, Afghan Communism and Soviet Intervention (Oxford: Oxford University Press, 1999), pp. 87–9. 27 G.M. Kornienko, ‘The Afghan Endeavour: Perplexities of the Military Incursion and Withdrawal’, Journal of South Asian and Middle Eastern Studies, vol. 17, no. 2 (1994), pp. 2–17; Odd Arne Westad, ‘Prelude to the Invasion: The Soviet Union and the Afghan Communists, 1978–1979’, International History Review, vol. 16, no. 1 (1994), pp. 49–69. 28 Hafizullah Emadi, ‘China’s Politics and Developments in Afghanistan’, Journal of Asian and African Studies, vol. 28, nos 1/2 (1993), pp. 107–17. 29 Barnett R. Rubin, The Fragmentation of Afghanistan: State Formation and Collapse in the International System (1995, reprint, New Haven, CT: Yale University Press, 2002), pp. 269–71. 30 Olivier Roy, Afghanistan: From Holy War to Civil War, The Inaugural Leon B. Poullada Memorial Lecture Series (Princeton, NJ: Darwin Press, 1995). 31 Kristian Berg Harpviken, ‘Transcending Traditionalism: The Emergence of NonState Military Formations in Afghanistan’, Journal of Peace Research, vol. 34, no. 3 (1997), pp. 271–87. 32 Alex Strick van Linschoten and Felix Kuehn, An Enemy We Created: The Myth of the Taliban/Al-Qaeda Merger in Afghanistan, 1970–2010 (London: Hurst, 2012), pp. 159–88. 33 Stephen Cohen, The Idea of Pakistan (Washington, DC: Brookings Institution Press, 2004). 34 Marvin G. Weinbaum, ‘Afghanistan and its Neighbors: An Ever Dangerous Neighborhood’, Special Report 162 (Washington, DC: United States Institute of Peace, 2006), p. 6. 35 Barnett R. Rubin and Abubaker Siddique, ‘Resolving the Pakistan–Afghanistan Stalemate’, Special Report 176 (Washington, DC: United States Institute of Peace, 2006). 36 Frédéric Grare, Pakistan and the Afghan Conflict, 1979–1985: With an Afterword Covering Events from 1985–2001(Karachi: Oxford University Press, 2003), p. 196.

Afghanistan’s Complex Neighbourhood Relations

277

37 Laurie Nathan, ‘Power, Security and Regional Conflict Management in Southern Africa and South Asia’, Comparative Social Research, vol. 27 (2010), pp. 309–32. 38 Mohammad Yousaf and Mark Adkin, The Bear Trap: Afghanistan’s Untold Story (Lahore: Jang Publishers, 1992), pp. 97–9. 39 Shuja Nawaz, Crossed Swords: Pakistan, its Army, and the Wars Within (Oxford: Oxford University Press, 2008). 40 Brahma Chellany, ‘Fighting Terrorism in Southern Asia’, International Security, vol. 26, no. 3 (2001), pp. 94–116. 41 Richard Mackenzie, ‘The United States and the Taliban’, in W. Maley (ed.), Fundamentalism Reborn? Afghanistan and the Taliban (New York: New York University Press, 1998), p. 96. 42 Olivier Roy, ‘The Transnational Dimension of Radical Islamic Movements’, in D. Groves (ed.), Talibanisation: Extremism and Regional Instability in South and Central Asia (Berlin: Stiftung Wissenschaft und Politik, 2001). 43 Rahimullah Yusufzai, ‘A New Proxy War?’, Newsline (June 2006). 44 Shahrbanou Tadjbaksh, ‘South Asia and Afghanistan: The Robust India–Pakistan Rivalry’, PRIO Paper (Oslo: Peace Research Institute Oslo, 2011). 45 Anatol Lieven, Pakistan: A Hard Country (London: Allen Lane, 2011). 46 Peter Bergen, Manhunt: From 9/11 to Abbottabad – the Ten-Year Search for Osama Bin Laden (London: The Bodley Head, 2012). 47 Jubin M. Goodarzi, Syria and Iran: Diplomatic Alliance and Power Politics in the Middle East (London: I.B. Tauris, 2009). 48 Jacqueline S. Ismael and Tareq Y. Ismael, ‘Globalization and the Arab World in Middle East Politics: Regional Dynamics in Historical Perspective’, Arab Studies Quarterly, vol. 21 (1999), pp. 129–44. 49 Buzan and Wæver, Regions and Powers, pp. 188–89. 50 Anwar-­ul-Haq Ahady, ‘Saudi Arabia, Iran and the Conflict in Afghanistan’, in W. Maley (ed.) Fundamentalism Reborn? Afghanistan and the Taliban (New York: New York University Press, 1998), pp. 117–34; Mushahid Hussain, ‘Regional and Arab Approach to the Afghan Issue’, Paper presented at a conference hosted by the Writers Union of Free Afghanistan (WUFA), Peshawar (31 August 1991). 51 Sharam Chubin and Charles Tripp, ‘Iran–Saudi Arabia Relations and Regional Order’, Adelphi Paper 304 (London: The International Institute for Strategic Studies, 1996). 52 Jan-Heeren Grevemeyer, ‘Ethnicity and National Liberation: The Afghan Hazara between Resistance and Civil War’, in Jean-Pierre Digard (ed.), Le Fait Ethnique en Iran et en Afghanistan (Paris: CNRS, 1988), pp. 211–18. 53 Kristian Berg Harpviken, ‘The Hazara of Afghanistan: The Thorny Path Towards Political Unity: 1978–1992’, in T. Atabaki and J. O’Kane (eds), Post-Soviet Central Asia (London: I.B. Tauris, 1998), pp. 185, 189.

278

War and State-Building in Afghanistan

54 Niamatullah Ibrahimi, ‘The Dissipation of Political Capital among Afghanistan’s Hazaras: 2001–2009’, CSRC Working Paper 51 (London: Crisis State Research Centre, London School of Economics and Political Science, 2009). 55 Ahady, ‘Saudi Arabia, Iran and the Conflict in Afghanistan’, pp. 125–31. 56 Shireen T. Hunter, ‘The Soviet Union and the Islamic Republic of Iran’, in H. Malik (ed.), Soviet–American Relations with Pakistan, Iran and Afghanistan (Basingstoke: Macmillan, 1987), pp. 244–66, see especially pp. 257–9. 57 Joseph McMillan, ‘Saudi Arabia and Iraq: Oil, Religion and an Enduring Rivalry’, Special Report 157 (Washington DC: United States Institute of Peace, 2006). 58 Rashid, Taliban, p. 200. 59 Harpviken, ‘The Hazara of Afghanistan’, in Atabaki and Kane (eds), Post-Soviet Central Asia, p. 194. 60 Barnett R. Rubin and Sarah Batmanglich, The US and Iran in Afghanistan: Policy Gone Awry (Cambridge, MA: Center for International Studies, Massachusetts Institute of Technology, October 2008). 61 ‘UN Security Council Resolution 1267 (New York: UN Security Council, 1999). 62 Trita Parsi, Treacherous Alliance: The Secret Dealing of Israel, Iran and the US (New Haven, CT: Yale University Press, 2007), pp. 223–37. 63 Kayhan Barzegar, ‘Iran’s Foreign Policy Strategy after Saddam’, Washington Quarterly, vol. 33, no. 1 (2010), pp. 173–89, see especially p. 176. 64 Vali Nasr, The Shia Revival: How Conflicts within Islam will Shape the Future (New York: W.W. Norton, 2006), pp. 241–2. 65 Rubin and Batmanglich, ‘The US and Iran in Afghanistan’. 66 Barzegar, ‘Iran’s Foreign Policy Strategy after Saddam’, pp. 173–89. 67 Roy Allison, ‘Virtual Regionalism, Regional Structures and Regime Security in Central Asia’, Central Asian Studies, vol. 27, no. 2 (2008), pp. 185–202. The quotation is from p. 198. 68 Martha Olcott, Central Asia’s Second Chance (Washington, DC: Carnegie Endowment for Peace, 2005). 69 Shahrbanou Tadjbaksh, ‘Central Asia and Afghanistan: Insulation on the Silk Road, between Eurasia and the Heart of Asia’, PRIO Paper (Oslo: Peace Research Institute Oslo, 2012). 70 Olivier Roy, The New Central Asia: The Creation of Nations (London: I.B. Tauris, 2000), p. 193. 71 Roy Allison, ‘Regionalism, Regional Structures and Security Management in Central Asia’, International Affairs, vol. 80, no. 3 (2004), pp. 463–83, see especially p. 473. 72 Buzan and Wæver, Regions and Powers, pp. 397–439. 73 For example, Lena Jonson, ‘Russian Peacekeeping and Tajikistan’. Paper presented at Russia and International Peacekeeping, Oslo (13–14 November 1995). 74 Tadjbaksh, ‘Central Asia and Afghanistan’.

Afghanistan’s Complex Neighbourhood Relations

279

75 Rashid, Taliban, p. 162. 76 Ahmed Rashid, ‘Talibanisation and Central Asia’, in D. Groves (ed.) Talibanisation: Extremism and Regional Instability in South and Central Asia (Berlin: Stiftung Wissenschaft und Politik, 2001), p. 34. 77 Vitaly V. Naumkin, ‘Militant Islam in Central Asia: The Case of the Islamic Movement of Uzbekistan’, Berkeley Program in Soviet and Post-Soviet Studies Working Paper (Berkeley, CA: UC Berkeley, 2003), p. 22. 78 Anthony Hyman, ‘Russia, Central Asia and the Taliban’, in W. Maley (ed.), Fundamentalism Reborn? Afghanistan and the Taliban (New York: New York University Press, 1998), pp. 114–15. 79 Neil S. MacFarlane, ‘The United States and Regionalism in Central Asia’, International Affairs, vol. 80 (2004), pp. 447–61, see especially p. 460. 80 Nicklas Norling and Niklas Swanström, ‘The Shanghai Cooperation Organization, Trade, and the Roles of Iran, India and Pakistan’, Central Asian Survey, vol. 26 (2007), pp. 429–44. 81 ‘US to designate Afghanistan major non-NATO ally: Officials’, Reuters (1 May 2012) [http://www.reuters.com/article/2012/05/01/us-­afghanistan-obama-­natoidUSBRE8401DL20120501]. 82 Kai Eide, Power Struggle over Afghanistan: An Inside Look at What Went Wrong – and What We Can Do to Repair the Damage (New York: Skyhorse Publishing, 2012). 83 Hanspeter Neuhold, ‘Permanent Neutrality in Contemporary International Relations: A Comparative Perspective’, Irish Studies in International Affairs, vol. 1, no. 3 (1983), pp. 13–26. 84 See also Kristian Berg Harpviken, ‘Caught in the Middle? Regional Perspectives on Afghanistan’, Comparative Social Research, vol. 27 (2010), pp. 277–305; Reinoud Leenders,’Strong States in a Troubled Region: Anatomies of a Middle Eastern Regional Conflict Formation’, Comparative Social Research, vol. 27 (2010), pp. 171–95; Ståle Ulriksen, ‘Webs of War: Managing Regional Conflict Formations in West Africa and Central Africa’, Comparative Social Research, vol. 27 (2010), pp. 355–80.

Select Bibliography Archival sources British Library, London, United Kingdom Afghan Rebellion, L/PS/10/1285–95 Arthur Balfour Papers, Add MSS 49749 Broughton de Gyfford MSS, F 213/8 Charles Wood Papers F 78 John Lawrence Papers, F 90/30 Kabul Diary, 5 August 1887, L/PS/7/50. Kabul Diary, 11 July 1896, L/PS/7/87 Lord Halifax Papers, C 152/5 Lord Lansdowne Papers, D 558, vol. 25 Lord Lytton Papers, E 218/141 Memorandum by H.M. Durand, 21 January 1887, in Viceroy’s Private Secretary to Bradford, 18 April 1887, L/PS/7/49 Memorandum by India Office, 24 May 1916, ‘Situation at Kabul’, L/PS/10/593 Memorandum by Roberts, ‘The dangers to which a reverse would expose us’, 27 January 1891, L/MIL/17/14/80 Minute by Hirtzel, 26 August 1919, L/PS/10/576 Minute by TWH (Holderness), 25 October 1915, L/PS/10/459 Peshawar Diary, 31 May 1885, L/PS/7/44 Report on the Second Afghan War, Captain Oliver for Major General MacGregor, Simla, 1885–6, L/MIL/17/14/29/1 Report of the Tribal Control and Defence Committee, 1931, L/PS/12/3171 Ridgeway, J.W., ‘Strength of Afghan Troops in Turkestan and Badakshan’, September 1886, L/PS/18/A70 The Military Resources of Afghanistan, 23 May 1887, L/PS/18/A78

Public Records Office/The National Archives, Kew, Surrey, UK Chamier to Trenchard, 6 August 1925, RAF India to Air Ministry, 14 October 1925, AIR 5/608 Garnet Wolseley, ‘South African Journal, 1879/80’, 24 October October, WO 147/7

282

Select Bibliography

National Archives of India, New Delhi, India Kabul Diary, 37, 25 September 1895, Section F, November 1898, 5, 147–82 Strength and Distribution and Armament of the Afghan Army, Foreign Department, Secret-F, February 1893, nos. 224–9

Articles, books and chapters in edited volumes Adamec, Ludwig W., Afghanistan’s Foreign Affairs to the Mid-Twentieth Century: Relations with the U.S.S.R., Germany and Britain (Tucson, AZ: University of Arizona Press, 1991). Agabekov, George, O.G.P.U.: The Russian Secret Terror (New York: Brentano’s, 1931). Alam, Muzaffar, ‘Trade, State Policy and Regional Change: Aspects of Mughal-Uzbek Commercial Relations, c. 1550–1750’, Journal of the Economic and Social History of the Orient, vol. 37, no. 3 (1994), pp. 202–27. Alder, Garry, ‘Big Game Hunting in Central Asia’, Journal of Imperial and Commonwealth History, vol. 9, no. 3 (1981), pp. 318–30. Ali, M. Athar, Mughal India: Studies in Polity, Ideas, Society, and Culture (New Delhi: Oxford University Press, 2006). Arreguín-Toft, Ivan, ‘How the Weak Win Wars: A Theory of Asymmetric Conflict’, International Security, vol. 26, no. 1 (2001), pp. 93–128. Arreguín-Toft, Ivan, How the Weak Win Wars: A Theory of Asymmetric Conflict (New York: Cambridge University Press, 2005). Arreguín-Toft, Ivan, ‘The Meaning of “State Failure”: Public Service, Public Servants, and the Contemporary Afghan State’, International Area Studies Review, vol. 15, no. 3 (2012), pp. 263–78. Baev, Pavel K., ‘Sad Wisdom of Hindsight: Soviet Intervention in Afghanistan, 1979–1989’, in Cheryl Benard, Ole Kværnø, Peter Dahl Thruelsen and Kristen Cordell (eds), Afghanistan: State and Society, Great Power Politics and the Way Ahead (Santa Monica, CA: Rand, 2008). Baumann, Robert F., Russian-Soviet Unconventional Wars in the Caucasus, Central Asia, and Afghanistan, Leavenworth Papers, no. 20 (Fort Leavenworth, KS: Combat Studies Institute Press, 1993). Bellew, H.W., ‘A New Afghan Question’, Journal of the United Service Institution of India, 47 (1881), pp. 49–97. Bennett, Huw, ‘The Other Side of the COIN: Minimum and Exemplary Force in British Army Counterinsurgency in Kenya’, Small Wars & Insurgencies, vol. 18, no. 4 (2007), pp. 638–64. Bennett, Huw, ‘Minimum Force in British Counterinsurgency’, Small Wars & Insurgencies, vol. 21, no. 3 (2010), pp. 459–75.

Select Bibliography

283

Biddle, Stephen, Afghanistan and the Future of Warfare: Implications for Army and Defense Policy (Carlisle, PA: The US Army War College, Strategic Studies Institute Monograph, 2002). Billon, Philippe Le, ‘The Political Ecology of War: Natural Resources and Armed Conflicts’, Political Geography, vol. 20, no. 5 (2001), pp. 561–84. Boulding, K.E., Conflict and Defense: A General Theory (New York: Harper & Row, 1962). Buhaug, Halvard, ‘Dude, Where’s My Conflict? LSG, Relative Strength, and the Location of Civil War’, Conflict Management and Peace Science, vol. 53, no. 4 (2010), pp. 544–69. Buhaug, Halvard and Scott Gates, ‘The Geography of Civil War’, Journal of Peace Research, vol. 39, no. 4 (2002), pp. 417–33. Buhaug, Halvard and Jan Ketil Rød, ‘Local Determinants of African Civil Wars, 1970–2001’, Political Geography, vol. 25, no. 3 (2006), pp. 315–35. Buhaug, Halvard, Scott Gates and Päivi Lujala, ‘Geography, Rebel Capability, and the Duration of Civil Conflict’, Journal of Conflict Resolution, vol. 53, no. 4 (2009), pp. 544–69. Butler, Christopher and Scott Gates, ‘Asymmetry, Parity, and (Civil) War: Can International Theories of Power Help Us Understand Civil War?’, International Interactions, vol. 35 (2009), pp. 330–40. Butler, Christopher and Scott Gates, ‘African Range Wars: Climate, Conflict, and Property Rights’, Journal of Peace Research, vol. 49, no. 1 (2012), pp. 23–34. Buzan, Barry, People, States and Fear: An Agenda for International Security Studies in the Post-Cold War Era (Brighton: Harvester Press, 1983). Buzan, Barry and Ole Wæver, Regions and Powers: The Structure of International Security (Cambridge: Cambridge University Press, 2003). Charters, David A., ‘British Intelligence in the Palestine Campaign, 1945–47’, in Ian Beckett (ed.), Modern Counter-Insurgency (Farnham: Ashgate, 2007), pp. 3–28. Chima, Jugdep S., ‘Controlling the Sunni Insurgency in Iraq: “Political” and “Military” Strategies from Successful Counterinsurgency in Punjab-India’, Small Wars & Insurgencies, vol. 18, no. 4 (2007), pp. 615–37. Collier, Paul and Anke Hoeffler, ‘Greed and Grievance in Civil War’, Oxford Economic Papers, vol. 56, no. 4 (2004), pp. 563–95. Dale, Stephen F., The Garden of the Eight Paradises: Babur and the Culture of Empire in Central Asia, Afghanistan and India (1483–1530) (Leiden: E.J. Brill, 2004). Durand, H.M., The Life of Major-General Sir Henry Marion Durand, vol. 1 (London: W.H. Allen, 1883). Duthie, John Lowe, ‘Failure and Success: John Jacob’s Quetta Project, 1856–76’, Asian Affairs, vol. 10, no. 3 (1979), pp. 272–91. Fearon, James D., ‘Why Do Some Civil Wars Last So Much Longer Than Others?’, Journal of Peace Research, vol. 41, no. 3 (2004), pp. 275–301. Ferris, John, ‘Lord Salisbury, Secret Intelligence, and British Policy toward Russia and Central Asia, 1874–1878’, in John Robert Ferris, Intelligence and Strategy (London: Routledge, 2005), pp. 8–44.

284

Select Bibliography

Ferris, John, ‘The Internationalism of Islam: The British Perception of a Muslim Menace, 1840–1951’, Intelligence and National Security, vol. 24, no. 1 (2009), pp. 57–77. Ferris, John, ‘Small Wars and Great Games—The British Army and Hybrid Warfare, 1700–1970’, in Williamson Murray and Peter Mansoor (eds), Hybrid Warfare: Fighting Complex Opponents from the Ancient World to the Present (Cambridge: Cambridge University Press, 2012), pp. 199–224. French, David, ‘The British Army and the Empire, 1856–1956’, in Greg Kennedy (ed.), Imperial Defence: The Old World Order: 1856–1956 (London: Routledge, 2008), pp. 91–110. French, David, The British Way in Counter-­insurgency, 1945–1967 (Oxford: Oxford University Press, 2011). Galeotti, Mark, Afghanistan: The Soviet Union’s Last War (London: Frank Cass, 1995). Gates, Scott and Kaushik Roy, Unconventional Warfare in South Asia: Shadow Warriors and Counterinsurgency (Farnham: Ashgate, 2014). Ghani, Ashraf, ‘The Afghan State and its Adaptation to the Environment of Central and Southwest Asia’, in H. Malik (ed.), Soviet–American Relations with Pakistan, Iran and Afghanistan (New York: St. Martin’s Press, 1990), pp. 310–32. Ghani, Ashraf and Clare Lockhart, Fixing Failed States: A Framework for Rebuilding a Fractured World (New York: Oxford University Press, 2009). Gillard, David, The Struggle for Asia, 1828–1914 (London: Methuen, 1977). Giradet, Edward, Afghanistan: Soviet War (London: Croom Helm, 1985). Gleditsch, Kristian S., ‘Transnational Dimensions of Civil War’, Journal of Peace Research, vol. 44, no. 3 (2007), pp. 293–309. Goldstein, Joshua S., Winning the War on War: The Decline of Armed Conflict Worldwide (New York: Dutton, 2011). Gommans, Jos J.L., The Rise of the Indo-Afghan Empire, c. 1710–1780 (Leiden: E.J. Brill, 1995). Grare, Frédéric, Pakistan and the Afghan Conflict, 1979–1985: With an Afterword Covering Events from 1985–2001 (Karachi: Oxford University Press, 2003). Grasselli, Gabriella, British and American Responses to the Soviet Invasion of Afghanistan (Aldershot: Dartmouth Publishing, 1996). Grau, Lester W., The Bear Went Over the Mountain: Soviet Combat Tactics in Afghanistan (London: Frank Cass, 1998). Gray, Colin S., ‘Why Strategy is Difficult’, Joint Forces Quarterly, issue 22 (Summer 1999), pp. 80–6. Gregorian, Raffi, ‘ “Jungle Bashing” in Malaya: Towards a Formal Tactical Doctrine’, in Ian Beckett (ed.), Modern Counter-Insurgency (Farnham: Ashgate, 2007), pp. 29–50. Gregorian, Vartan, The Emergence of Modern Afghanistan: Politics of Reform and Modernization, 1880–1946 (Stanford, CA: Stanford University Press, 1969). Gupta, Hari Ram, Life and Work of Mohan Lal Kashmiri, 1812–1877 (Lahore: Minerva Book Shop, 1943). Gwynn, C.W., Imperial Policing (London: Macmillan, 1934).

Select Bibliography

285

Habib, Irfan, An Atlas of the Mughal Empire: Political and Economic Maps with Detailed Notes, Bibliography and Index (1982, reprint, Delhi: Oxford University Press, 1986). Habib, Irfan (ed.), Akbar and his India (1997, reprint, New Delhi: Oxford University Press, 2002). Haidar, Mansura, Medieval Central Asia: Polity, Economy and Military Organization (Fourteenth to Sixteenth Centuries) (New Delhi: Manohar, 2004). Harpviken, Kristian Berg, ‘Transcending Traditionalism: The Emergence of Non-State Military Formations in Afghanistan’, Journal of Peace Research, vol. 34, no. 3 (1997), pp. 271–87. Harpviken, Kristian Berg, Social Networks and Migration in Wartime Afghanistan (London: Palgrave Macmillan, 2009). Harpviken, Kristian Berg, ‘Caught in the Middle? Regional Perspectives on Afghanistan’, Comparative Social Research, vol. 27 (2010), pp. 277–305. Herbst, J., States and Power in Africa: Comparative Lessons in Authority and Control (Princeton, NJ: Princeton University Press, 2000). Hirshleifer, Jack, ‘The Paradox of Power’, Economics and Politics, vol. 3 (1991), pp. 177–200. Hirshleifer, Jack, The Dark Side of the Force: Economic Foundations of Conflict Theory (Cambridge: Cambridge University Press, 2001). Howell, Evelyn, Mizh: A Monograph of Government’s Relations with the Mahsud Tribe (1931, reprint, Karachi: Oxford in Asia Historical Reprint Series, 1979). Ingram, Edward, The Beginning of the Great Game in Asia: 1828–1834 (Oxford: Clarendon Press, 1979). Johnson, Rob, The Afghan Way of War, How and Why they Fight (New York: Open University Press, 2011). Kakar, Hasan Kawun, Government and Society in Afghanistan: The Reign of Amir Abd’al Rahman Khan (Austin, TX: University of Texas Press, 1979). Kakar, M. Hasan, Afghanistan: The Soviet Invasion and the Afghan Response, 1979–1982 (Berkeley, CA: University of California Press, 1995). Khan, Iqtidar Alam, The Political Biography of a Mughal Noble: Munim Khan-­i-Khanan, 1497–1575 (New Delhi: Munshiram Manoharlal, 1991). Kozlov, S. (ed.), Spetsnaz GRU–2: Voina ne okonchena, istoriya prodolzhaetsya (Spetsnaz GRU–2: The War is Not Over, History Continues) (Moscow: Russkaya Panorama, 2002). Lal, Mohan, Life of the Amir Dost Mohammaed Khan of Kabul, 2 vols (1846, reprint, New Delhi: Asian Educational Services, 2004). Lambrick, H.T., John Jacob of Jacobabad (Karachi: Oxford University Press, 1960). Lane-Poole, Stanley, The Emperor Babur (n.d., reprint, New Delhi: Sunita Publications, 1988). Leenders, Reinoud, ‘Strong States in a Troubled Region: Anatomies of a Middle Eastern Regional Conflict Formation’, Comparative Social Research, vol. 27 (2010), pp. 171–95. Lujala, Päivi, ‘The Spoils of Nature: Armed Conflict and Rebel Access to Natural Resources’, Journal of Peace Research, vol. 47, no. 1 (2010), pp. 15–29.

286

Select Bibliography

Macgregor, Douglas A., Breaking the Phalanx (New York: Praeger, 1997). MacGregor, Lady (ed.), The Life and Opinions of Major-General Sir Charles Metcalfe MacGregor, vol. 2 (Edinburgh: William Blackwood and Sons, 1888). Mack, Andrew, ‘Why Big Nations Lose Small Wars: The Politics of Asymmetric Conflict’, World Politics, vol. 27, no. 2 (1975), pp. 175–200. MacMunn, G.F., Vignettes from Indian War (1901, reprint, New Delhi: Low Price Publications, 1993). MacMunn, G.F., The Armies of India (1911, reprint, New Delhi: Heritage Publishers, 1991). MacMunn, G.F., Afghanistan from Darius to Amanullah (1929, reprint, Lahore: Sang-­eMeel Publications, 2002). Macrory, Patrick, Kabul Catastrophe: The Story of the Disastrous Retreat from Kabul, 1842 (Oxford: Oxford University Press, 1986). Maley, W. (ed.) Fundamentalism Reborn? Afghanistan and the Taliban (New York: New York University Press, 1998). Maley, W., The Afghanistan Wars (2002, reprint, London: Palgrave Macmillan, 2009). Martineau, John, The Life and Correspondence of the Right Hon. Sir Bartle Frere (London: John Murray, 1895). Mearsheimer, John, Liddell Hart and the Weight of History (Ithaca, NY: Cornell University Press, 1988). Melton, Stephen L., The Clausewitz Delusion: How the American Army Screwed Up the Wars in Iraq and Afghanistan (A Way Forward) (Minneapolis, MN: Zenith Press, 2009). Moreman, T.R., ‘ “Small Wars” and “Imperial Policing”: The British Army and the Theory and Practice of Colonial Warfare in the British Empire, 1919–1939’, Journal of Strategic Studies, vol. 19, no. 4 (1996), pp. 105–31. Moreman, T.R., The Army in India and the Development of Frontier Warfare, 1849–1947 (London: Macmillan, 1998). Nathan, Laurie, ‘Power, Security and Regional Conflict Management in Southern Africa and South Asia’, Comparative Social Research, vol. 27 (2010), pp. 309–32. Nawid, Senzil, ‘The State, the Clergy, and British Imperial Rule in Afghanistan during the 19th and Early 20th Centuries’, International Journal of Middle Eastern Studies, vol. 29, no. 4 (1997), pp. 581–605. Neuhold, Hanspeter, ‘Permanent Neutrality in Contemporary International Relations: A Comparative Perspective’, Irish Studies in International Affairs, vol. 1, no. 3 (1983), pp. 13–26. Noelle, Christine, State and Tribe in Nineteenth-Century Afghanistan: The Reign of Amir Dost Muhammad Khan (1826–1863) (1997, reprint, London: Routledge, 2008). Paris, Roland, At War’s End: Building Peace after Civil Conflict (Cambridge: Cambridge University Press, 2004). Parsi, Trita, Treacherous Alliance: The Secret Dealing of Israel, Iran and the US (New Haven, CT: Yale University Press, 2007).

Select Bibliography

287

Paschall, Rod, ‘Marxist Counterinsurgencies’, in Ian Beckett (ed.), Modern CounterInsurgency (Farnham: Ashgate, 2007), pp. 453–66. Paul, T.V., Asymmetric Conflicts: War Initiation by Weaker Powers (Cambridge: Cambridge University Press, 1994). Pelly, Captain Lewis (ed.), The Views and Opinions of Brigadier-General John Jacob, C.B. (London: Smith, Elder & Co., 1858). Pinker, Steven, The Better Angels of Our Nature: Why Violence Has Declined (New York: Viking, 2011). Porch, Douglas, Counterinsurgency: Exposing the Myths of the New Way of War (Cambridge: Cambridge University Press, 2013). Rashid, Ahmed, Taliban: Islam, Oil and the New Great Game in Central Asia (London: I.B. Tauris, 2000). Richards, John F., The Mughal Empire, The New Cambridge History of India, I:5 (1993, reprint, New Delhi: Foundation Books, 2002). Rid, Thomas, ‘The Nineteenth Century Origins of Counterinsurgency Doctrine’, Journal of Strategic Studies, vol. 33, no. 5 (2010), pp. 727–58. Roberts, Field-Marshal Lord, Forty-One Years in India, From Subaltern to Commander-­ in-Chief (London: Richard Bentley & Son, 1897). Roy, Kaushik, ‘Horses, Guns, and Governments: A Comparative Study of the Military Transition in the Manchu, Mughal, and Ottoman and Safavid Empires, circa 1400 to circa 1750’, International Area Studies Review, vol. 15, no. 2 (2012), pp. 99–121. Roy, Kaushik, The Army in British India: From Colonial Warfare to Total War, 1857–1947 (London: Bloomsbury Academic, 2013). Roy, Olivier, Afghanistan: From Holy War to Civil War (Princeton, NJ: Darwin Press, 1995). Rubin, Barnett R., The Fragmentation of Afghanistan: State Formation and Collapse in the International System (1995, reprint, New Haven, CT: Yale University Press, 2002). Saideman, Stephen M., ‘Discrimination in International Relations: Analyzing External Support of Ethnic Groups’, Journal of Peace Research, vol. 39, no. 1 (2002), pp. 27–50. Salehyan, Idean, ‘Transnational Rebels – Neighboring States as Sanctuary for Rebel Groups’, World Politics, vol. 59, no. 2 (2007), pp. 217–42. Sarkar, Jadunath, A Short History of Aurangzeb (1930, reprint, New Delhi: Orient Longman, 1979). Scott, James C., Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven, CT: Yale University Press, 1999). Shafer, Michael D., Deadly Paradigms (Princeton, NJ: Princeton University Press, 1988). Singhal, D.P., India and Afghanistan, 1876–1907: A Study in Diplomatic Relations (Melbourne: University of Queensland Press, 1963). Skaperdas, S., ‘Contest Success Functions’, Economic Theory, vol. 7, no. 2 (1996), pp. 283–90. Skaperdas, S., ‘On the Formation of Alliances in Conflict and Contests’, Public Choice, vol. 96, nos 1/2 (1998), pp. 25–42.

288

Select Bibliography

Slantchev, Branislav L., ‘The Power to Hurt: Costly Conflict with Completely Informed States’, American Political Science Review, vol. 97 (2003), pp. 123–33. Smith, Rupert, The Utility of Force (New York: Penguin, 2007). Smith, Vincent A., Akbar the Great Mogul: 1542–1605 (n.d., reprint, New Delhi: S. Chand, 1962). Stam, Allan C., Win, Lose, or Draw: Domestic Politics and the Crucible of War (Ann Arbor, MI: University of Michigan Press, 1996). Statiev, Alex, Social Conflict and Soviet Counterinsurgency in the Western Borderlands, 1944–50 (Cambridge: Cambridge University Press, 2009). Steinbach, Henry, The Country of the Sikhs (n.d., reprint, New Delhi: KLM Book House, 1977). Tanner, Stephen, Afghanistan: A Military History from Alexander the Great to the Fall of the Taliban (New York: Da Capo Press, 2002). Thornton, Thomas Henry, Colonel Sir Robert Sandeman: His Life and Work on our Indian Frontier (London: John Murray, 1895). Tripodi, Christian, ‘Peacemaking through Bribes or Cultural Empathy? The Political Officer and Britain’s Strategy Towards the North-West Frontier, 1901–1945’, Journal of Strategic Studies, vol. 31, no. 1 (2008), pp. 123–51. Tripodi, Christian, ‘Grand Strategy and the Graveyard of Assumptions: Britain and Afghanistan, 1839–1919’, Journal of Strategic Studies, vol. 33, no. 5 (2010), pp. 701–25. Ulriksen, Ståle, ‘Webs of War: Managing Regional Conflict Formations in West Africa and Central Africa’, Comparative Social Research, vol. 27 (2010), pp. 355–80. Vivekanandan, Jayashree, Interrogating International Relations: India’s Strategic Practice and the Return of History (London: Routledge, 2011). Warren, Alan, Waziristan, the Faqir of Ipi, and the Indian Army: The North West Frontier Revolt of 1936–37 (New York: Oxford University Press, 2000). Westad, Odd Arne, The Global Cold War: Third World Interventions and the Making of Our Times (Cambridge: Cambridge University Press, 2005). Wink, Andre, Akbar (Oxford: Oneworld, 2009). Yapp, M.E., Strategies of British India, Britain, Iran and Afghanistan, 1798–1850 (Oxford: Clarendon Press, 1980). Yousaf, Mohammad and Mark Adkin, The Bear Trap: Afghanistan’s Untold Story (Lahore: Jang Publishers, 1992).

Internet sources Boyd, John R., The Essence of Winning and Losing (1986/2010) [http://tobeortodo.com/ wp-­content/uploads/2011/11/essence_of_winning_losing.pdf]. Burke, Crispin, ‘Like It or Not, Small Wars Will Always Be Around’, World Politics Review (24 January 2012) [http://www.worldpoliticsreview.com/articles/11238/like-­it-or-­ not-small-­wars-will-­always-be-­around; accessed 25 July 2012].

Select Bibliography

289

Capstick, Michael, ‘Reviewing Canada’s Afghan Mission’, Policy Options (April 2008) [http://www.irpp.org/po/; accessed 15 February 2013]. Chiarelli, Peter W. and Stephen M. Smith, ‘Learning from Our Modern Wars: The Imperatives of Preparing for a Dangerous Future’, Military Review (September/October 2007) [http://www.army.mil/professionalWriting/volumes/ volume6/january_2008/1_08_1.html; accessed 22 February 2013]. Ellis, David C. and James Sisco, ‘Implementing COIN Doctrine in the Absence of a Legitimate State’, Small Wars Journal (13 October 2010) [http://smallwarsjournal. com/jrnl/art/implementing-­coin-doctrine-­in-the-­absence-of-a-legitimate-­state; accessed 25 July 2012]. Finel, Bernard, ‘A Substitute for Victory’, Foreign Affairs (8 April 2010) [http://www. foreignaffairs.com/articles/66189/bernard-­finel/a-­substitute-for-­victory; accessed 25 July 2012]. Grau, Lester, The Soviet–Afghan War: Superpower Mired in the Mountains [http://www. smallwars.quantico.usmc.mil/search/LessonsLearned/afghanistan/miredinmount. asp; accessed December 2010]. Hamid, Mustafa, ahzan al-­umm al-­kabira (A Mother’s Deep Sorrow), The Harmony Programme Database at CTC West Point, Doc. No. AFGP–2002–600092 [http://www. ctc.usma.edu/wp-­content/uploads/2010/08/AFGP–2002–600092-Orig-Meta.pdf]. Hamid, Mustafa, al-­hamaqa al-­kubra, aw al-­harb al-­ma’iz, Book 4 of tharthara fi saqaf al- alam. Downloaded from MAFA [Mustafa Hamid’s blog], 12 October 2009 [http:// mafa.maktoobblog.com/]. Heuser, Beatrice, ‘Ends, Ways, Means: Clausewitz and Other Prophets’, Infinity Journal Special Edition, Clausewitz and Contemporary Conflict (February 2012), p. 18 [http:// www.infinityjournal.com/article/48/Ends_Ways_Means_Clausewitz_and_Other_ Prophets; accessed 25 January 2012]. Korolev, V., Uroki voiny v Afganistane 1979–1989 godov [http://www.sdrvdv.org/ node/159; accessed December 2010]. Manea, O. and J. Nagl, ‘COIN is not Dead’, Small Wars Journal, 6 February 2012 [http:// smallwarsjournal.com/jrnl/art/coin-­is-not-­dead; accessed 25 July 2012]. Smith, Jonathan, ‘We Own the Night’, Small Wars Journal (22 February 2012) [http:// smallwarsjournal.com/jrnl/art/we-­own-the-­night; accessed 25 July 2012]. ‘US to designate Afghanistan major non-NATO ally: Officials’, Reuters (1 May 2012) [http://www.reuters.com/article/2012/05/01/us-­afghanistan-obama-­natoidUSBRE8401DL20120501]. Wilder, Andrew and Gordon Stuart, ‘Money Can’t Buy America Love’, Foreign Policy (1 December 2009) [http://www.foreignpolicy.com/articles/2009/12/01/money_cant_ buy_america_love?hidecomments=yes; accessed 25 July 2012].

290

Select Bibliography

Magazines and newspapers Afghan Jehad: Quarterly Magazine of the Cultural Council of the Afghanistan Resistance, Al-Jihad, Mujahideen, New York Times, Newsline, Time Magazine, Washington Post.

Translated non-English sources Babur-Nama (Memoirs of Babur), tr. from the original Turki text of Zahiruddin Muhammad Babur by A.S. Beveridge, 2 vols (n.d., reprint, New Delhi: Saeed International, 1989). Fazl, Abul, The Akbar Nama, tr. from the Persian by H. Beveridge, 3 vols (1902–39, reprint, New Delhi: Saeed International, 1989). Ghobar, Ghulam Mohammed, Afghanistan der Masir-­i Tarikh (Afghanistan on the Highway of History) (Kabul: Book Publishing Institute, 1967). Khan, Inayat, The Shah Jahan Nama, an abridged history of the Mughal Emperor Shah Jahan, compiled by his Royal Librarian, ed. and completed by W.E. Begley and Z.A. Desai (New Delhi: Oxford University Press, 1990). Khan, Saqi Mustad, Maasir-­i-Alamgiri, tr. into English and annotated by Jadunath Sarkar (1947, reprint, Calcutta: The Asiatic Society, 1990). Khan, Zain, Tabaqat-­i-Baburi, tr. S. Hasan Askari and annotated by B.P. Ambastha (Delhi: Idarah-­i-Adabiyat-­i-Delli, 1982). Mohammed, Kayz, Saraj al-Tawarikh (The Lamp of History) (Kabul: Afghan Government Publications, 1915). Nagar, Ishwardas, Futuhat-­i Alamgiri, tr. and ed. by Tasneem Ahmad (New Delhi: Idarah-­i- Adabiyat-­i-Delli, 1978). The Empire of the Great Mogul, De Laet’s Description of India and Fragment of Indian History, tr. by J.S. Hoyland and annotated by S.N. Banerjee (1928, reprint, New Delhi: Munshiram Manoharlal, 1974). The History of India as Told by its Own Historians: The Muslim Period, The Posthumous Papers of the Late H.M. Elliot, ed. and continued by John Dawson, 8 vols (1876–1877, reprint, Delhi: Low Price Publications, 2001). The Jahangirnama: Memoirs of Jahangir, Emperor of India, tr., ed. and annotated by Wheeler M. Thackston (New York: Oxford University Press in association with the Smithsonian Institution, 1999). The Tezkereh Al Vakiyat or Private Memoirs of the Mughal Emperor Humayun, written in Persian Language by Jouher, tr. by Major Charles Stewart (London: Oriental Translation Fund, 1833).

Glossary Ahadis Soldiers recruited and paid directly by the Mughal Emperor Amir Ruler of Afghanistan. Technically the status of this title was lower than Sultan Bakshi One in charge of management of the Mughal Army Ban Hand-­held rockets also known as hawais. It was originally introduced from China to India Durbar Court of a Muslim ruler Dushman A term meaning ‘enemy’ in Persian, used by Soviet troops to refer to the mujahideen Faujdar A Mughal official in charge of maintaining law and order in the district. He is equivalent to a modern police officer Gazi A religious warrior who is willing to become a shahid (martyr) for the cause of Islam Jagir A grant of land to a person in lieu of salary for maintaining cavalry contingent. The Safavids termed jagir as tuyul Jagirdar One who holds the jagir Jihad Holy war in the name of Islam Jihadi One who conducts jihad Kandak Battalion in the Afghan Army Khalsa Theocratic kingdom of the Sikhs Khan A wealthy man in a village who aspires to leadership of the tribe or a section of it in Afghanistan Khassadar Irregular village levy for policing and they operate under the village headmen Khel Section of a tribe Kishlak Villager Kizilbash Also known as Qizilbash. The term refers to a Persian soldier who wore red hats and settled in Afghanistan in the Safavid era Kos A unit for measuring distance. One kos is equivalent to 1.5 miles Lakh One lakh is 100,000 Lashkar An armed tribal war band Madrasa Islamic seminary Malik Head of a clan or tribe Mansabdar A Mughal official who was granted a mansab (an imperial rank) in lieu of his position in the military and civilian bureaucracy Mujahid/Mujahideen Holy warriors of Islam Mullah Muslim village priest

292

Glossary

Pir A religious person who heads a mystic order Pushtunwali Cultural code of conduct of the Pathans/Pashtuns Qawm Islamic community Raj Literal meaning realm; it refers to the Government of British India Raja A Hindu ruler Sangar Mountaintop defensive position created by the Afghans from rocks and boulders Sharia Islamic law Shura Council Silladar An individual who owns several horses on which he mounts bargirs (men without horses). A silladar then offers his service to the highest bidder Sirdar Tribal chief Sowar Cavalryman Suba Mughal province Subadar Governor of the suba Thana Police post Thanadar Chief of police post. He is subordinate to the faujdar Topkhana Artillery Department Tuyul Somewhat equivalent to jagir; a piece of land granted to an individual or a group in return for maintaining a cavalry contingent for service Ulema Islamic clergy Wazir Principal minister in the durbar

Index 40th Army (Soviet) 115–19, 122, 171 1973 coup 251 Abbas II, Shah 65–6, 69 Abdur Rahman 161–2, 165–6 Administrative Code 168 administrators/public servants 222, 223, 225, 227, 233 Afghan Air Force 171 Afghan-Arabs 131–46 Afghan Army 1860s and 1870s 157–61 1881–1901 161–6 1919 conflict 167 artillery 160, 165, 167–8 cavalry regiments 164–5 conscription 163 desertion 154 deterioration of 167–9 dispersed formations 161 early history 151–7 ethnic composition 162–3 expansion of 163 infantry 163–4 loyalty 154 officers 162 reservists 166 revenue extraction 156 tactics 155, 161 training 119, 160 uniforms 164 weaknesses 166 weapons 158, 159, 160, 164 see also Afghan National Army; Democratic Republic of Afghanistan Army Afghan communists 123 Afghan National Army (ANA) 175–82 administration 180 air support 177 civilian contractors 181 cohesion 177

competence of 180–1 counterinsurgency 147–50 desertion 177 ‘Embedded Partnering’ 149 ethnic composition 181 expansion of 149, 175, 177, 179, 238 leadership 178, 179 literacy training 178 loyalty 238 officers 180 professionalism 178 recruitment and retention 178 reduction of 182 ‘rule of three’ 150 sabotage of orders 181 social value of military service 176 training 176, 180, 182 unit structure 36 weaknesses 182 see also Afghan Army; Democratic Republic of Afghanistan Army Afghan National Police (ANP) 10–11, 175 Afghan National Security Forces (ANSF) 175 Afghan–Soviet War see Soviet–Afghan War Afghan Turkestan 43 Afghanistan 1500–1700 borders 44 culture 122–3 ethnic groups 16 geography 43 Government/Taliban controlled areas, 2007 34 independence 92 origin of Afghans 43–5 tribes 4–5, 46 see also neighbourhood relations ‘Afghanization’ policy 10 ‘AfPak’ policy 246 Afridi Uprising (1672) 69–70 aid 236

294

Index

air warfare 101–5, 123, 141, 234 Airport Project 143 Akbar, Emperor 54–61 Al-Qaeda 131, 137, 142, 235, 253 Algerian War 121 Alikozais 4 Allison, Roy 266 Amanallah, Amir 91–2, 93, 167, 169 Ambala campaign (1863) 97 Amin, Hafizullah 170 Amin Khan, Muhammad 69–70 amirs (rulers), British management of 94 ANA see Afghan National Army Anderson, Benedict 223–4 Anglo-Afghan Wars 5, 31–2, 87–8, 92, 167, 250 Annan, Kofi 207 ANP see Afghan National Police ANSF see Afghan National Security Forces artillery, Afghan Army 160, 165, 167–8 Aryans 45 assassinations 87 asymmetric warfare 21–41 Anglo-Afghan Wars 31–2 counterinsurgency literature 36–7 geography 25–7 the Great Game 30–1 hurt vs. hide 22–5 insurgents’ willingness to fight 27–8 loss-of-strength gradient 25–6 Mughals 30–1 resolve 28 resource allocation 24 Soviet–Afghan War 32–4 time preferences 28 United States war in Afghanistan 34–7 willingness to fight 27–8 Auckland, Lord 86 Aurangzeb, Emperor 69, 70–1 Ayyub, Muhammad 90 Babur, Zahir-ud-din Muhammad 46–50, 52 Badakshan 3, 56, 63, 64 Bahadur Khan 55 Balkh 3, 65 Baluch Nazhari 54 Baluchistan/Baluchs 6, 96, 100 bargaining theory 37

BATs see brigade advisory teams Bellew, H.W. 43–5 Bengal Army 101 Bennett, Huw 8 Berdal, Mats 208–9 Bhagu 69 Bhira 47–8 Bhutto, Zulfiqar Ali 251 bin Laden, Osama 131, 133, 134, 142, 143, 235, 258 bodyguards 5, 154, 165 Bolsheviks 92 see also communists Bombay Army 100 Bombay School, frontier management 95–6 bombing campaigns see air warfare Bonn Conference 235 Boyd, John 35 brigade advisory teams (BATs) 149 Britain Afghan Wars 5, 31–2, 87–8, 92, 167, 250 Afghanistan counterinsurgency 5–9, 79–112 air power 101–5 First Anglo-Afghan War 5, 87–8 ‘forward policy’ 250 invasions of Afghanistan 86–90 Kenya counterinsurgency 8 Palestine counterinsurgency 8 ‘small war’ doctrine 31 Third Anglo-Afghan War 31–2, 92, 167, 250 withdrawal from Afghanistan 94 British Army counterinsurgency 7 language skills 150 Pashtuns serving in 159–60 and Pashtunwali 150 weapons 158 British Empire counterinsurgency 79–112 North-West Frontier 13–14, 79–112 power and perception 82–5 specialized units 85 British Government of India (GOI) air policing 102–5 Baluchistan pacification policy 6 ‘Forward School’ policy 89

Index North-West Frontier 13–14, 79–112 power and perception 82–5 Russian threat to Afghanistan 6–7 Brodsky, Joseph 123 Brzezinski, Zbigniew 124 ‘buffer state’, Afghanistan as 250 Building Peace after War (Berdal) 208–9 Burnes, Alexander 87 Bush Administration 246 Bust Fort 67 ‘butcher and bolt’ strategy 95, 98 Buzan, Barry 247–50, 261, 266 Caldwell, William 238 Callwell, Charles 35 Capstick, Mike 35–6 Cavagnari, Louis 89 cavalry 68, 151, 152, 153, 164–5 Central Asian states 265–9 Charters, David A. 8 China 252, 258, 259 CIS see Confederation of Independent States civil servants 222, 223, 225, 227, 233 civil war 51, 58, 63, 82, 88, 93, 94, 114, 123, 126, 169, 174, 178, 203, 206, 207, 222, 227, 266, 268 civilian casualties 11, 119 civilian contractors 181 clan rivalries 50 Clinton, Hillary 245 Close Frontier policy 88, 90–4 COIN see counterinsurgency Cold War 124–5, 255 collaboration and control 203–4 collective value, fighting effort 24 communists 123 see also Bolsheviks competitive control theory 203 compounds and towers 153 Confederation of Independent States (CIS) 249, 266–7 conflict resolution, political bargaining rules 210–11 conscription 163, 172 contest success functions (CSF) model 24 contraband resources 26, 28 counterinsurgency (COIN) Afghan National Army 147–50

295

air policing 102–5 Britain/British India 7, 8–9, 79–112 ‘COIN lite’ 205–6 criticism of 198–206 definition 194–8 enemy-centric approach 199, 202–3, 206 governance 196–7, 209 imagined states 221–44 Indian Army 12–13 legitimacy objective/issues 195–8, 200–2, 209, 210 Malayan model 7 military role 197–8, 203–6 Mughals 43–78 new doctrine 35–6 North-West Frontier 79–112 population-centric approach 197, 200–2, 206 Soviet Union 7, 9–10, 81–2, 113–30 stability 197 stakeholder-centric approach 206–11 success criteria 195 theories 12, 81, 191–220 three pillars of 35–6 United States 7–8, 35–6, 191–220 victory criteria 195 Cox, H.V. 102 criticism of counterinsurgency 198–206 cross-border raids 118 CSF see contest success functions cultural understanding of insurgents 13 Dalhousie, Marquis of 82–3 Daoud, President 169 Dara Shikoh, Prince 67 Daud, Mohammed 121, 227, 228–9, 251 Daulat Khan 49, 66 de Laet, Johannes 59 de Waal, Alex 209–11 Defense Strategic Guidance, US 191 Democratic Republic of Afghanistan Army (DRAA) 135–6, 140, 169–75 dispersed formations, Afghan Army 161 doctrines see military doctrines Dost Mohammed 5, 17, 86, 88, 156, 157, 158 Dostum, Abdul Rashid 253 DRAA see Democratic Republic of Afghanistan Army

296

Index

drugs 267 see also opium/poppy cultivation Durand Line 91 Durrani, Ahmad Shah 151, 154 Durrani, Timur 155–6 Durranis 155–6, 162 ‘Embedded Partnering’, Afghan National Army 149 Enduring Freedom, Operation 34, 221, 233–8 enemy-centric counterinsurgency 199, 202–3, 206 ethnic groups, Afghanistan 16 ethnic transplantation 52–3 external fighters (foreign fighters) 131–46 ‘failed states’ 209, 223 Fall, Bernard 198 fanaticism 83 FATA see Federally Administered Tribal Areas Fath Khan, Abul 57–8 Federally Administered Tribal Areas (FATA) 80 fighting capability, geography and 25–7 fighting effort 24 First Anglo-Afghan War (1839–42) 5, 87–8 First World War 91 FM 3–24 (United States counterinsurgency doctrine) 191–220 counterinsurgency definition 194–8 criticism of 198–206 insurgency definition 193–4 legitimacy objective/issues 195–8 success criteria 195 foreign fighters 131–46 forts 53, 56, 153 ‘forward policy’ 89, 250 Frere, Bartle 83, 157 frontier management 95–101 see also North-West Frontier Galula, David 5, 197 garrisons, Mughals 60–1 Gentile, Gian P. 202 geography 25–7, 43 Germany 91

Ghazi, Mirza 62, 63 ghazis (religious warriors) 159 Ghilzais 154, 162 Giradet, Edward 171 GOI see British Government of India Gorbachev, Mikhail 124–5 Gorka, Sebastian L. v. 199, 207 governance, counterinsurgency and 196–7, 209 governmental models 209 Grare, Frédéric 255 Grau, Lester 118 the Great Game 30–1, 44 guerrilla warfare 23, 80–2, 115, 116, 120 gunpowder 22, 54, 64, 65, 67, 83, 160 Gurmsir 58–9 Habibullah 91 Hakim, Mirza 55–6 Halifax, Lord 94 Hamid, Mustafa 137, 143 Haqqani, Jalaluddin 133, 138, 139 Hazaras 47, 49, 165, 261–2 ‘hearts and minds’ policy 119, 148 Heather, Randall W. 7 hejira (exodus/migration) 84 Hekmatyar, Gulbuddin 136, 138, 139 helicopters 32 Helmand 9, 150 Hemu, General 54–5 Hensman, Howard 159, 161 Herat-Farah Lowlands 43 Hindu Beg 47, 48 Hindu Kush 43 Hindustan 49, 51 Hirshleifer, Jack 24, 29 Hirtzel, Arthur 84, 103 Hobbes, Thomas 223 horses 14, 43, 47, 48–9, 52, 64, 65, 68, 164, 165 Howell, Evelyn 79 Humayun, Nasir-ud-din 49, 50–3 IEDs see improvised explosive devices imagined states, counterinsurgency failure 221–44 imperial anthropology 100 improvised explosive devices (IEDs) 148, 180

Index IMU see Islamic Movement of Uzbekistan independence of Afghanistan 92 India Afghan relations 254–9, 270 Close Frontier policy 88, 90–4 ‘Forward School’ policy 89 horse trade 48–9 North-West Frontier 13–14, 79–112 Northern Alliance support 259 trade 14, 48–9 see also British Empire; British Government of India Indian Army 12–13, 83, 103, 104 indiscriminate violence 204 Indus, River 56 Indus tribes 57 infantry, Afghan Army 163–4 infrastructural power 2 institutional models 209 insulator states 249 insurgency/insurgents cultural understanding of 13 definition 193–4 geographical advantages 25–7 hurt vs. hide 22–5 negotiations with 206–7 resource access 26 revenue extraction 26, 27–8 tactical adaptation 22–5 targeting of insurgent leaders 204 terrorism comparison 194 time preferences 28 willingness to fight 27–8 Inter-Services Intelligence (ISI), Pakistan 136, 139, 142, 256 International Security and Assistance Force (ISAF) formation of 235 ‘rule of three’ 150 support of Afghan National Army 149–50 tactical adaptation 35 Taliban conflict 25, 221, 236–8 ‘transition’ concept 182 withdrawal from Afghanistan 37 Iran 252, 256, 259–65 Iranian Plateau 43 Iraq 34, 37, 102, 259, 262–3, 265 irregular forces 81, 151, 152–3

297

ISAF see International Security and Assistance Force ISI see Inter-Services Intelligence Islam 3, 83–4, 123 Islamic Movement of Uzbekistan (IMU) 268 Islamism 251, 266 Ismailis 173–4 Israel 264 Israelites 43–5 Istanbul process 270 Italy 94 Jacob, John 95–6 Jagdara 57 Jahan, Shah 63–9 Jahangir, Mughal Emperor 61–3 Jaji, Battle of (1987) 133 Jalalabad, Battle of (1989) 134, 135–9 Jalali, Ali A. 149 Jani Beg, Mirza 54 jihad/jihadis (holy war/warriors) 34, 84, 131–46 Kabul ‘Kabul Declaration on Goodneighborly Relations’ initiative 270 Kabul–Kandahar axis 54 Lahore to Kabul road 59–61 Mughal period 47, 48, 52, 54, 56, 68 Kahmard Fort 63–4 Kalmatic Baluch tribe 54 Kalyvas, Stathis 203–4, 206, 211 Kamal, Babrak 229 Kamil Khan 69 Kamran, Mirza 50, 51, 52 Kandahar 51, 55, 59, 61, 65 Siege of 62–3, 66–8 Kara Nokars (cavalry) 152 Karimullah, Colonel 177 Karmal, President 170 Karzai, Hamid 150, 238, 264, 270 Kashmir 257 Kazakhstan 265–9 Kenya 8 KGB 119 Khaibar Pass 59, 68 Khalil Beg 63–4 Khalil, General 173

298 Khalqi leadership 251–2 khans (wealthy villagers) 165, 166 Khomeini, Ayatollah Ruhollah 262 Khost 64, 168 Battle of 139–43 Khyber Pass 59, 68 Khyber Rifles 98 Kikuyus 8 Kilcullen, David 199, 203, 207 kill-or-capture missions 204–5 Kirgizstan 265–9 Kizilbash 154, 155 Kohna Nokars (cavalry) 153 Kremlin 119 Kryuchkov, Vladimir 125 Kunar River 257 Kutsenko, General 172–3 Laet, Johannes de 59 Lahore 59 Lawrence, John 83, 88 legitimacy 195–8, 200–2, 209, 210, 235 Leviathan (Hobbes) 223 liberal counterinsurgency theory 81 Lieven, Anatol 258–9 literacy training (of ANA) 178 Litvinov, Maxim 93 livestock 70 local power structures 209 Lodhis 49–50 The Logic of Violence in Civil War (Kalyvas) 203–4 looting 154–5, 174, 175, 232 loss-of-strength gradient (LSG) 25–6 McChrystal, Stanley 149 madrasas (Islamic seminaries) 83 Maffey, John 104 Mahabat Khan 70 Malayan insurgency 7 Mangal, Gulab 182 Maoist counterinsurgency theory 81 ‘martial races’ concept 100 Massoud, Ahmad Shah 136, 139, 173 matchlock men 65, 66 Mau Mau insurgency 8 mentoring 150 mercenaries 51, 52 micro states 232

Index Middle East 102, 261 see also Iran; Iraq; Israel; Saudi Arabia; Syria military doctrines role of 192–3 Soviet Union 230–1 United States 191–220 military force, critique of 203–6 military service, social value 176 military training teams (MTTs) 149 ‘mission creep’ 231 Mission Without End? Peacekeeping in the African Political Marketplace (de Waal) 209–11 Mohammadzai 4 Mongols 45 Moshtarak, Operation 150 mountain warfare 35 MTTS see military training teams Muazzam Khan 70 Mughals 3–4, 14–15, 30, 59–61, 67, 68 Akbar, Emperor 54–61 asymmetric warfare 30–1 Babur, Zahir-ud-din Muhammad 46–50, 52 garrisons 60–1 Humayun, Nasir-ud-din 49, 50–3 interventions in Afghanistan 43–78 1504–1556 46–53 1556–1707 53–72 Jahan, Shah 63–9 Jahangir 61–3 Mughal–Uzbek treaty 65 Muhammadzai 4 mujahideen (holy warriors) 9, 10, 173 conflict with Najibullah 134–5 infighting 232 Jaji, Battle of 133 Jalalabad, Battle of 135–9 Khost, Battle of 139–43 looting 174, 175, 232 National Commanders’ Shura 138–9, 141 prisoners of war 136, 140 Soviet–Afghan war 32–4, 117, 119, 124 success of 230 tactics 231–2 US support 33 see also Taliban

Index Murad, Prince 64–5 Muslims, fear of 83–4 see also Islam; Shias Naderi, Sayyed Mansur 174 Nadir Khan 94 Nahmardi 54 Najibullah, Mohammad 10, 134–5, 138 Nathan, L. 255–6 the nation see the state National Commanders’ Shura (NCS) 138–9, 141 National Military Academy of Afghanistan (NMAA) 178 National Reconciliation plan, Najibullah’s 138 NATO see North Atlantic Treaty Organization natural gas pipeline project 15 natural resources, insurgents’ use of 27–8 Nazar Muhammad 64, 65 NCS see National Commanders’ Shura negotiations with insurgents 206–7 neighbourhood relations 245–79 Central Asia 265–9 India 254–9 Iran 259–65 map 246 Pakistan 254–9 Persian Gulf states 259–65 policy visions 269–71 regional security complexes 247–50 Saudi Arabia 259–65 South Asia 254–9 nepotism 236 neutrality status, Afghanistan 270–1 ‘New Silk Road’ policy 245, 247 night-raids, negative impact of 204–5 Nilab 69 NMAA see National Academy of Afghanistan nomads 14 North Atlantic Treaty Organization (NATO) 236 North-West Frontier air policing 101–5 British counterinsurgency 79–112 British expenditure 99

299

frontier management 13–14, 95–101 subsidies to tribal leaders 99 Northern Alliance 234–5, 254, 256–7, 259 Norwegian Armed Forces 193 Obama Administration 38, 246 Observe, Orient, Decide and Act (OODA) 35 OEF see Operation Enduring Freedom OMLTs see operational mentoring and liaison teams OODA see Observe, Orient, Decide and Act Operation Enduring Freedom (OEF) 34, 221, 233–8 Operation Moshtarak 150 operational mentoring and liaison teams (OMLTs) 11, 149 opium/poppy cultivation 68, 117, 233 Owen, William F. 204 Oxus, River 3 Pakistan Afghan relations 254–9 ‘AfPak’ policy 246 army 32 China alliance 259 ethnic unrest 255 Inter-Services Intelligence 136, 139, 142, 256 Taliban support 237, 253, 257, 258 US relations 125, 251 Palestine 8 Panipat, First Battle of (1526) 50 Panipat, Second Battle of (1556) 54–5 Panjsher Valley 115 ‘paradox of power’ model (Hirshleifer) 24 Parchamis 251 Pashtuns in British Army 159–60 British management of 96–101 founding of dynasty 151 Pathan etymology 46 personal courage 155 social code of 122, 123, 150, 205 unrest in Pakistan 255 Pashtunwali (code of conduct) 122, 123, 150, 205 Pathans see Pashtuns

300

Index

‘patrimonial marketplace’ model 210–11 peace-building literature 207–11 People’s Democratic Party of Afghanistan (PDPA) 169, 228, 251, 256 Percox, David A. 8 Persian Gulf states 259–65 Persians 30, 55, 61–3, 65–6 Petreaus, David 192 policy visions 269–71 political bargaining, conflict resolution 210–11 Popalzais 4 poppy cultivation 68, 117, 233 see also drugs population-centric counterinsurgency 197, 200–2, 206 population transfers 33, 52–3 Powindahs (trading nomads) 14 prisoner of war (POW) policies 136, 140 public servants 222, 223, 225, 227, 233 Punjab 13, 17, 153 Punjab School, frontier management 96, 97 qala (tower or compound) 153 qawm militia 158 Qizilbash 4–5, 66, 67 Quetta 96 racism 101 RAF see Royal Air Force Rahman, Abdul 2, 90, 91 Raj see British Government of India Rashid, Ahmed 267 Rawlinson, Field Marshal 35 Reading, Lord 102–3 Reagan, Ronald 124, 125 rebel groups see insurgency/insurgents refugees 9, 33, 117, 118, 122, 252 regional security complexes (RSC) 247–50 rentier state, Afghanistan as 1–2, 4, 16 resources endowment model 24 insurgent access to 26, 27–8 Revolution in Military Affairs (RMA) 114 Rid, Thomas 5 roads 49, 59–61, 116, 226 Roberts, Lord 83, 100 Rohillas 49 Royal Air Force (RAF) 92–3, 104, 105

royal bodyguard 165 RSC see regional security complexes Rubin, Barnett R. 1–2, 15, 16 ‘rule of three’ 150 Russia British India and 6–7 and Central Asia 269 hegemonic goal 266–7 influence/interventions in Afghanistan 90, 91 Northern Alliance support 259 see also Soviet Union Sadiq Khan 54 Sadullah Khan 67 Safavids 30, 44, 50, 65 Salafism 235 Salang highway 116 Salisbury, Lord 89 Samarkhand 14 Sandeman, Colonel 96, 97 Sarbuland Khan 71 Saudi Arabia 142, 143, 259–65 Saur Revolution 169–70 sawara-i kushada (feudal cavalry) 151 SCO see Shanghai Cooperation Organization ‘scorched earth’ policy 153 SCUD missiles 141 Scythians 45 Seistan 51 Selim Khan 53 Shaibani Khan 46–7 Shamsher Khan 69 Shanghai Cooperation Organization (SCO) 269 Sher Ali 88–9, 157, 158, 159, 160 Sher Shah (Sher Khan) 51, 52–3 Shias 261–2, 263 Shibar/Shibartu Pass 52 Shuja, Shah 86–7 Shujat Khan 71 shura (council) 138–9, 141 SIEDs see suicide IEDs Sikhs 153 sirdars (tribal chiefs) 50, 71–2 small unit patrols 7, 8 ‘small war’ doctrine 31 SOFs see special operations forces

Index Somaliland 102 South Asia region 254–9 Soviet Union 40th Army 115–19, 122, 171 counterinsurgency 7, 9–10, 81–2 interventions in Afghanistan 13, 92–5, 113–30, 170–4, 227–33, 252 military culture 122 military doctrine 230–1 see also Russia Soviet–Afghan War (1979–89) 113–30, 170–4, 229–33 Afghan-Arabs’ role 132–3 bombing campaigns 123 and Cold War 124–5 communication threats 231 Soviet strategy 117–20 Soviet tactics 114–16 Soviet withdrawal 134–5 spring offensives 232 sowars (cavalrymen) 164–5 special operations forces (SOFs) 205, 234 Spetsnaz groups 115 stakeholder-centric counterinsurgency 206–11 state ‘buffer state’, Afghanistan as 250 definition 222, 224 evolution of 225–6 imagined states 221–44 insulator states 249 micro states 232 priority of in Afghanistan 225–7 state-building literature 207–11 stereotypes 100, 101 STINGER portable surface-to-air missiles 116 strategic innovation 22–5, 35 subsidies to Afghan chiefs/tribes 88–9, 96–9 suicide IEDs (SIEDs) 148 surface-to-air missiles 116 ‘surge’ of troops 36–7, 148 Suri Empire 52 Suri, Sikander 53 Swat 57 Syria 260 tactical adaptation 22–5, 35 Tahmasp, Shah 51

301

Tajikistan/Tajiks 265–9 Taliban 10, 21, 120, 123, 175, 205, 263 Al-Qaeda alliance 253 battle of ideas 237 benefits of fighting 25 control of Afghanistan 33–4, 232 drug production/trade 233 emergence of 253 ill-feeling against in Afghanistan 234 ISAF interactions 236–8, 254 legitimacy of 235 looting 154–5, 174, 175 see also mujahideen Taraki, Nur Muhammad 121, 169, 170, 229 taxation 157 technological innovation 22 Tenai, General 174 terror/terrorism 8, 22, 23, 99, 166, 170, 194, 221, 227, 233, 234, 239, 240, 248, 254, 258 terrorists 8, 12, 116, 117, 124, 131, 194, 204 Third Anglo-Afghan War (1919) 31, 92, 167, 250 Tirah Campaign (1897–98) 97–8 Tirmiz 3 Todar Mal 58 tolls 72, 154 Torkhar Mountain 140, 143 towers and compounds 153 trade 14–15, 48–9, 59–61, 72 traditional Afghan culture 122–3 transition arrangements/concept, post-2001 182, 269–71 Trenchard, Hugh 103 Tribal Control and Defence Committee 100 Tripodi, Christian 6, 13 troop ’surges’ 36–7, 148 Turkestan 43 Turkey 91, 270 Turki horses 68 Turkmenistan 265–9 Turkoman Plain 43 uniforms, Afghan Army 164 United Kingdom (UK) see Britain; British Empire

302 United States (US) Afghan Declaration of Strategic Partnership 264 Afghanistan policy 10, 245–7, 264 and Central Asia 268–9 counterinsurgency 7–8, 35–6, 191–220 Defense Strategic Guidance 191 FM 3–24 doctrine 191–220 hierarchical structure of military 35 India relations 258 interventions in Afghanistan 13, 34–7, 233–8, 254 Iran relations 264 Iraq war 263 mujahideen support 33, 124 Operation Enduring Freedom 233–8 Pakistan relations 258–9 special operations forces 234 USSR see Soviet Union Uzbekistan 253, 265–9 Uzbeks 44, 50, 55, 58, 63, 64, 65 victory criteria, counterinsurgency 195 Vietnam War 121 Village Defence Militia 36 Waever, Ole 247–50, 261, 266 Wahabis 83

Index Wakhan Corridor 250 war on terror 258 warlords/warlordism 3, 30, 36, 122, 148, 169, 174, 178, 233, 235, 236, 237, 238, 257, 267 Waziristan 31, 32, 92, 97, 99 weapons Afghan Army 158, 159, 160, 164 British Army 158 ‘high-tech’ arms 32, 34 Mughal artillery 67 SCUD missiles 141 STINGER portable surface-to-air missiles 116 Western counterinsurgency theories 12 Wigram, Kenneth 79 willingness to fight, insurgents 27–8 Wolseley, Garnett 80 Wood, Charles 97 Ya’qub, Muhammad 89 Yate, Colonel 164–5 Yemen 260 Yusufzais 45, 48, 57–8, 69 Zain Khan 57, 58 Zameri, Brigadier General 179–80 Zhawar, Battles of (1985 and 1986) 32 Zia-ul-Hak, Muhammad 256