1,935 798 773KB
Pages 129 Page size 252 x 378.36 pts Year 2010
ALLIANCE MANAGEMENT AND MAINTENANCE
To Shuchi and Mira
Alliance Management and Maintenance Restructuring NATO for the 21st Century
JOHN R. DENI Political Advisor to US Military Forces in Europe and Adjunct Lecturer at Heidelberg University
© John R. Deni 2007 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the publisher. John R. Deni has asserted his right under the Copyright, Designs and Patents Act, 1988, to be identified as the author of this work. The views expressed are those of the author and do not reflect the official policy or position of the US Department of Defense or the US Government. Published by Ashgate Publishing Limited Gower House Croft Road Aldershot Hampshire GU11 3HR England
Ashgate Publishing Company Suite 420 101 Cherry Street Burlington, VT 05401-4405 USA
Ashgate website: http://www.ashgate.com British Library Cataloguing in Publication Data Deni, John R. Alliance management and maintenance : restructuring NATO for the 21st century 1. North Atlantic Treaty Organization - Armed Forces Organization 2. North Atlantic Treaty Organization Administration 3. Peacekeeping forces 4. Security, International 5. Military doctrine I. Title 355'.031'091821 Library of Congress Cataloging-in-Publication Data Deni, John R. Alliance management and maintenance : restructuring NATO for the 21st century / by John R. Deni. p. cm. Includes bibliographical references and index. ISBN 978-0-7546-7039-1 1. North Atlantic Treaty Organization--Organization. 2. North Atlantic Treaty Organization--Military policy. 3. Security, International. 4. World politics--1989- I. Title. II. Title: Restructuring NATO for the 21st century. UA646.3.D45 2007 355'.031091821--dc22 2007000326 ISBN: 978-0-7546-7039-1 Cover Design: Pica Design, LLC | picadesign.com Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall.
Contents 1
Introduction
1
2
Managing Alliances Work on Alliance Management to Date Addressing Change in Alliances The Role of Organizational Studies in Analyzing NATO Resisting Change Adapting and Learning Rationing Activity
9 10 12 15 16 17 18
3
Assessing Military Doctrine and Structures Doctrine and its Role The Sources of Doctrine Intra-Alliance Bargaining Understanding the Process and Content of Doctrinal Change Expected Outcomes
19 19 21 23 26 27
4
The Development of the NRDCs The Alliance Structure Immediate Post-Cold War Changes Evidence of Doctrinal Change? Toward Real Changes in Doctrine and Force Structure MC 400 – Setting the Stage The Command Structure The Force Structure
29 29 33 34 38 40 42 45
5
Changing Threats and the Alliance’s Response Can Security Threats Explain NATO’s Doctrine and Structures? The Suboptimal Alliance Response Missed Opportunities – Defense Capabilities Missed Opportunities – Stability and Security in Albania Missed Opportunities – Bosnia
55 55 58 58 62 64
6
The Impact of Political Bargaining Bargaining Over Expeditionary Capabilities
67 67
Alliance Management and Maintenance
vi
Bargaining Among the HRF(L) Candidates The German-Netherlands Corps The Spanish Corps The Eurocorps The Turkish Corps The ARRC The Italian Corps Summary Bargaining Over Command Relationships Summary
71 71 73 74 75 76 77 78 79 84
7
The Alliance’s Response to Terrorism The Political Guidance MC 472 and MC 161 Implementation of the Military Concept against Terrorism Anti-Terrorism Consequence Management Counterterrorism Military Cooperation
85 86 89 90 90 92 94 99
8
Conclusions and Implications Addressing Doctrine and Structure Implications of this Study
Bibliography Index
101 102 106 109 119
Chapter 1
Introduction In 2002, NATO began implementing significant changes in its command and force structures.1 Gone would be the heavy, armor-based forces of the Cold War era, judged by some as too slow and too large for the challenges of the post-Cold War security environment. The new security landscape would be marked by nationalist, ethnic, and religious conflicts and unconventional, transnational threats such as terrorism. Although the alliance commitment to collective defense as embodied in Article 5 of the North Atlantic treaty would remain, the threat of a Soviet or Russian armored column spearheading an attack through the Fulda Gap in West Germany simply did not exist any longer. Even though the alliance’s force structure was beginning to undergo significant changes in 2002, the bi-polar world of the Cold War had actually come to an abrupt end over a decade earlier, following popular revolutions all over Central Europe in 1989, the fall of the Berlin Wall in November of that year, and German reunification less than a year later. In April 1991, the Warsaw Pact disbanded, and the Soviet Union followed suit eight months later on 26 December 1991. Given this seismic shift in the security environment in the late 1980s and the early 1990s, the alliance came to recognize that both the way that NATO would provide security for its members and the means that it would employ in that effort needed to change – more expeditionary forces and supporting elements were needed for rapid deployment to zones of conflict beyond alliance member state territory. In the place of Cold War era heavy forces geared primarily to territorial defense, NATO heads of state and government began to create a command and force structure comprised of lighter, faster, and more readily deployable headquarters and forces to be known as the NATO Rapid Deployment Corps (NRDCs). NATO’s actions in 2002 and 2003 surrounding the implementation of the NRDC force structure marked the beginning of a process of significant post-Cold War structural changes for the alliance. However, the process that brought the alliance to that point was significantly slower than what one might have expected of the most successful alliance in history and the outcome of that process failed to correspond to what NATO itself had identified as its requirements. These issues surrounding the development of the NRDCs and the doctrinal concepts underpinning them collectively represent a useful window for examining alliance management during a period of transition. Upon initial investigation, the changes implemented by the alliance beginning in 2002 appear to have been based on changes in the security environment. Those 1 “Command structure” and “force structure” are two distinct military concepts. The former refers to the organizational bodies that command and control military forces. The latter refers to the specific military forces or units that fall under the command structure. However, in practice within NATO the distinction between the two is sometimes blurred.
2
Alliance Management and Maintenance
changes were reflected in alliance doctrine, which is based on a strategy – or Strategic Concept – that was rewritten in 1991 to incorporate the realities of the end of the Cold War and then most recently updated again in 1999.2 Both the 1991 and the 1999 Strategic Concepts identified new, post-Cold War threats – such as ethnic and religious strife, terrorism, and proliferation – and prescribed in general terms the kinds of capabilities necessary to counter the new threats. Alliance doctrine draws on these important documents, as well as guidance from the member state governments. Like the Strategic Concepts, such guidance has, since the early 1990s, also emphasized the changed threat environment, directing alliance officials to craft new doctrine and structures in response. However, a threat-based interpretation of the way in which the alliance managed itself through a period of great transition is somewhat flawed for at least two reasons. First, if NATO’s doctrine and structures were driven by changes in the post-Cold War world, it seems counterintuitive that the alliance would spend scarce defense resources on more high readiness units than it needed instead of plowing that money into sustainment and deployment capabilities necessary to meet the new threats. To explain, NATO determined that it required three of the NRDCs to be at the highest levels of readiness – the so-called High Readiness Forces (HRFs) – and yet in the end the alliance selected six High Readiness Forces. Assuming a certain degree of fungibility in defense spending and assuming little growth in member state defense budgets, developing and maintaining six of these expensive HRFs meant that other, equally important defense capabilities such as sustainment and deployment went under-funded. This naturally raises the question of why the alliance and its member states essentially chose to shortchange these other capability areas and establish more high readiness NRDCs than were required. Perhaps the Strategic Concept and ministerial guidance did not account for all of the likely threats, leaving alliance members concerned that three high readiness forces would not be enough. However, a threat-based interpretation of doctrinal development could reasonably be expected to predict high levels of integration among the strategy that identifies the threats in the security environment, the doctrine that outlines how the alliance will handle the threats, and the command and force structures that confront and mitigate the threats, so this possible explanation seems unlikely.3 Alternatively, in deciding to approve the development of six high readiness units, perhaps the alliance was actually responding to imperatives other than threats. If so, it begs the questions of what those imperatives were and why the alliance was motivated to act upon them, ultimately at the expense of other priorities.
2 In recent years, that has been talk of a revisiting the alliance’s strategy. During the November 2006 summit in Riga, alliance leaders issued the Comprehensive Political Guidance (CPG), which is designed not to replace or update the Strategic Concept but rather to provide further guidance on its implementation. 3 Barry Posen argues that threat-based approaches predict high degrees of integration between force structure, doctrine, and strategy. The Sources of Military Doctrine: France, Britain, and Germany Between the World Wars (Ithaca, NY: Cornell University Press, 1984), p. 80.
Introduction
3
Second, if NATO was driven to revise its doctrine and structures because of the end of the Cold War and the development of new threats, it is unclear why the alliance took so long – from 1991 to 2002 – to conduct planning and then to finally begin implementation of such changes. New post-Cold War threats were evident a decade before in places like Bosnia, were acknowledged by NATO’s member states, and were reflected in the alliance’s 1991 Strategic Concept. Yet despite this, it was not until 2002 that NATO began to implement the structural changes. An objective observer might expect an alliance dedicated to mitigating threats to member state security to have acted or responded sooner, before those threats led to conflict as was the case in the Balkans on three separate occasions in the 1990s. Indeed, one might argue that member states and citizens of those member states should reasonably expect a slightly better response time from an organization billed as the most successful alliance in history. This book will attempt to answer these and other questions surrounding the NRDCs, the sources of NATO’s military doctrine, the level of integration between NATO’s doctrine and structures on the one hand and its strategy on the other, and the management of an alliance in transition. In doing so, this book will show that a threat-based perspective is necessary but insufficient for explaining NATO’s management of the changing environment of the post-Cold War era for a variety of reasons, including those noted above. Instead, we must employ other tools to gain insights into how the North Atlantic alliance managed to navigate the significant changes required of its force and command structures. To better understand both the outcome of the process NATO engaged in to change its doctrine and structures and the nature of that process, we must turn to what scholars have written in two distinct, but in this situation related, areas of political science – 1) alliance maintenance or management and 2) the sources of military doctrine. Relative to other areas of study in international relations, alliance maintenance has seen surprisingly little attention from scholars. Certainly several scholars have attempted to expand our understanding of alliances over the last several decades, such as in Stephen Walt’s important work, The Origins of Alliances. Nevertheless, the few studies of alliance maintenance or management that have been done have tended to focus on a very limited array of subjects not entirely applicable to understanding how an alliance with an integrated, institutionalized military manages or maintains itself over time, particularly in the face of significant change. For instance, many of these works have essentially consisted of quantitative analyses of whether alliance formation leads to war or of alliance duration and subsequent failure. Most other works on the subject of alliance management have been derivatives of the larger academic debate over collective goods theory and therefore focused almost exclusively on burdensharing. An important exception to these trends was Glenn Snyder’s 1997 book, Alliance Politics, in which the author focused specifically on alliance maintenance by examining intra-alliance bargaining. Snyder argued that bargaining power within an alliance depends on three factors – the allies’ dependence on the alliance to protect it from threats, the allies’ commitment to the alliance, and the allies’ comparative
Alliance Management and Maintenance
4
interest in the object of the bargaining.4 In general, a state’s bargaining power – when it has a higher chance of obtaining its preferred outcome – is greatest when its dependence is low, its commitment is loose, and its interests at stake are great. Although not precisely applicable in the case of NATO for reasons explained later in this book, this sort of construct can help provide a framework for assessing the process the alliance pursued in changing its structures and doctrine as well as the outcomes of that process. This framework can therefore provide a useful window through which we can observe alliance management in the face of change. With regard to what scholars have written on the sources of military doctrine, the existing body of scholarly literature is a rich one, comprised of writings by both academics and practitioners. Together, these scholars can be broken down into three broad categories in terms of how they define doctrine, what role it plays, and what its sources are, although the dividing lines between such groupings are certainly not hard and fast. The first group, comprised mostly of traditional academic works on military doctrine such as Barry Posen’s The Sources of Military Doctrine, tends to rely on a somewhat broad definition of doctrine and its role and tends to focus on structural or functional sources as the most important independent variables that shape doctrine. The second group is comprised mostly of what might be considered alternative academic works, such as Deborah Avant’s Political Institutions and Military Change. They are similar to the first group in their views on doctrine’s definition and its role, but their opinions differ significantly when it comes to doctrine’s sources, as the alternative academic works rely primarily on institutional characteristics, culture, or political factors as independent variables. The third group consists mostly of practitioners or practitioner-scholars, such as Harry Summers, and it is here where one can find a more useful construct – relative to the first two groups – for examining and understanding elements of both process and outcomes in NATO’s doctrinal changes. In comparison to the first two groups, this third group has a far more precise notion of what doctrine is and what it does, especially with regard to command and force structures, but when it comes to doctrine’s sources they borrow from the strengths of both of the first two groups, emphasizing structural variables as well as alternative variables like domestic political factors or institutional characteristics. With rare exception though, most who have written on doctrine and force structure have addressed individual states through their writings, not international military organizations or groups of states. It therefore seems appropriate to assess whether the literature on military doctrine would have anything to offer a book such as this. There are at least two ways in which to tackle this issue. First, one could argue that when it comes to certain characteristics, states and international military organizations share enough similarities that applying the literature on military doctrine does not entail all that great of a conceptual leap. After all, individual states and international military organizations are both system-level actors and both are usually hierarchical entities of collective action, characterized by a certain degree of organizational structure and beholden to similar bureaucratic pressures. More specifically, NATO has an integrated military bureaucracy similar in many ways 4
Glenn Snyder, Alliance Politics (Ithaca: Cornell University Press, 1997), p. 166.
Introduction
5
to that of most states. Likewise, states and international military organizations are both subject to the preferences and prejudices of individual political leaders. Finally, both states and international military organizations are subject to the forces of public opinion. Another way of answering whether the literature on military doctrine is applicable in the case of NATO is to argue more simply that the literature is relevant because of the very nature of an international military organization such as the North Atlantic alliance. NATO most clearly has a growing body of military doctrine; it has a command structure comprised of senior military officials; it has a force structure led by those officials; and it applies its doctrine, its leaders, and its forces in military operations. Because the proverbial shoe appears to fit, it makes sense to use the existing, available literature on doctrine to examine NATO more closely. Insights from scholars who have written on the sources of military doctrine as well as those that have studied bargaining within alliances will help us to understand how NATO managed the transition from Cold War-era military doctrine and structures to post-Cold War doctrine and structures that emphasized lighter, more deployable, expeditionary forces. More specifically, the analysis of the sources of NATO’s doctrine will be nested within the broader analysis of alliance management – in this sense, the former can be thought of as a case study used in the examination of the latter. In telling the story of that case study, this book will show how NATO’s attempts at doctrinal and structural change throughout the 1990s were less than optimal, leaving the alliance particularly challenged in fulfilling its own strategy with regard to out-of-area sustainment and deployment and leaving it ill-prepared for several security challenges during the first decade after the Soviet Union’s demise that its own strategy said it should have been ready to handle. By exploring the formulation of NATO military doctrine and hence alliance force structure, this book will accomplish several useful goals. First, this book will offer an additional, unique perspective and a qualitative approach to the study of alliance management and maintenance, a field of inquiry that has seen relatively little attention in comparison to topics such as alliance formation. Second, this book will add new empirical data to our understanding of NATO’s post-Cold War evolution – no major works in the field currently cover the issue of alliance doctrine and force structure revision up through the early part of the 21st century. Third, since it examines the development of doctrine and force structure of an international military organization, instead of those of states, this book will add another perspective to the ongoing debate over the sources of military doctrine and the factors that affect the integration of doctrine and strategy. And finally, this book will try to shed light on the continuing debate surrounding NATO’s role in security, how the alliance will fight, and whether NATO is properly structured to continue providing security for its member states. Using NATO and the NRDCs as a means to explore alliance maintenance and to examine the sources of military doctrine is a compelling task from both practical and theoretical perspectives for at least four reasons. First, both theory and history tell us that alliances are typically focused on safeguarding their member states from security threats. It would seem fairly obvious that in terms of NATO doctrine and structures, threats should drive the alliance’s actions and decisions. The case of the NRDCs
6
Alliance Management and Maintenance
therefore represents a good one for examining whether responding to security threats really does explain alliance management and the sources of doctrine. Second, increasing our understanding of how alliances work is important as such bi- or multilateral groupings – both formal and informal – have only grown in relevance since the end of the Cold War, especially through coalitions of willing alliance members. Today, NATO’s responsibilities and involvement in Afghanistan are clearly growing, as the alliance-led International Security Assistance Force (ISAF) takes the lead for all but a small array of reconstruction and military operations in Afghanistan.5 Elsewhere, the alliance has taken on a major training mission in Iraq, it supports the African Union in Darfur, and it continues to play a critical role in Balkans stability operations. Although it is too soon to call NATO a global alliance, its importance throughout the world appears only to be growing. A better understanding of alliance maintenance and management would seem to offer decision-makers key insights on how to successfully navigate change at a time when NATO’s relevance is on the rise. Third, nothing is as fundamental to a military entity as its military forces and the structures used to command them. Without those forces and command and control elements, a military entity, such as an alliance with an integrated military command or an actor-level military organization, cannot fulfill its most basic mission of promoting security. In NATO’s case, those forces and command structures are today the NRDCs, comprised of command and control structures and personnel and assigned combat arms, combat support (CS), and combat service support forces (CSS).6 For those interested in understanding more about how the alliance developed these critical tools, it makes sense to examine the ways in which NATO developed its command and force structures. Finally, within NATO’s growing body of military doctrine, that which deals with the NRDCs is of significant importance. The NRDCs are today the collective centerpiece of NATO’s military command and operational capabilities. In NATO’s conception, the NRDCs include command and control elements as well as actual combat forces assigned by each of the hosting countries and any other participating countries.7 In the event of a NATO combat operation, the command and control elements of the NRDCs are currently envisioned as becoming the headquarters that, together with senior leadership from one of the Regional 5 NATO assumed control of operations in eastern Afghanstan – completing the counterclockwise expansion of responsibility for first the north, then the west, and then the south – in early October 2006. 6 Combat Support, or CS, consists of fire support and operational assistance provided to combat elements. It may include artillery, air defense, aviation, engineer, military police, signal, and electronic warfare. In contrast, Combat Service Support, or CSS, is the support provided to sustain combat forces primarily in the fields of administration and logistics. It may include administrative services, chaplain service, civil affairs, food service, finance, legal service, maintenance, medical service, military police, supply, transportation, and other logistical services. 7 In some cases, such as with the Turkish NRDC, the forces assigned to that command are almost entirely Turkish. In other cases, such as with the Eurocorps, several countries have assigned forces.
Introduction
7
Commands (Brunnsum is home to the Northern Regional Command, Naples is home to the Southern Regional Command, and Lisbon is home to the Maritime Regional Command), would command alliance forces. The alliance forces would consist of those assigned to that particular NRDC and mobilized for the operation in question or whatever designated forces were identified at that time to fulfill the NATO Response Force (NRF) requirement.8 Initially, the High Readiness Forces (Land), or HRF(L)s – the NRDCs at the highest level of readiness – would fulfill the NRDC requirement for a period of six months each, with other NRDCs at lower levels of readiness following on as necessary for the duration of the conflict or crisis. Prior to the development of the NRDCs, NATO has had to rely on ad hoc command and force contributions from member states, which are usually offered for periods of no more than six to 12 months. Bringing military units and command elements together in combat that have never trained with each other before is a recipe for inefficient and ineffective military performance. The procedure for building these less-than-effective force structures, commonly referred to as the force generation process, is onerous and inefficient, and is recognized by people within NATO and by outside experts as broken. To make matters worse, if the operation in question lasts longer than the initial six to 12 months, the alliance must go through the same problematic force generation process again and again. By creating standing NRDCs, the alliance took an important step toward fixing the way it generates military forces for operational requirements. Understanding how the alliance created the NRDCs and the doctrine that underpins them is an important task for anyone interested in knowing more about whether and how NATO will continue to provide security to its member states. The next chapter will address how alliance management has been studied to date, and will consider whether past alliances or the field of organizational studies provide any utility in studying NATO’s evolution. The third chapter will set out the terms of the discussion presented in this book. It will allow the reader to come to grips with broad topics such as ‘military doctrine’ and ‘political bargaining’ in the context of NATO. In the language of political scientists, this chapter serves the purpose of laying out both the dependent variables – that which this book seeks to explain – and the independent variables – the explanatory tools this book will rely upon. Chapter 4 presents the key elements of NATO’s post-Cold War transition relevant to the subjects discussed in this book, focusing specifically on the changes in strategy, doctrine, and structures that resulted in the implementation of the NRDC-centered command and force structures. Chapter 5 examines whether security threats are adequate to explain alliance decisions and actions, especially in terms of alliance management. The sixth chapter examines the role of political bargaining in alliance management, taking the reader deeper into the political context of member state and alliance decision-making. The penultimate chapter projects the findings of the book forward by examining the alliance’s response to the US-led war on terrorism. The last chapter, Chapter 8, offers summary comments and conclusions on the way ahead for NATO. 8 The NATO Response Force (NRF) requirement, roughly one brigade comprised of about 3,500 soldiers, rotates among alliance members every six months.
8
Alliance Management and Maintenance
Finally, regarding methodology, with the exception of a small collection of primary documents, such as alliance strategies, doctrines, and other planning documents, the vast majority of research for this book consisted of interviews conducted either in person or telephonically between August 2002 and August 2006. All interviews were conducted in English, and most of the face-to-face interviews took place during any of several visits by the author to Washington, Brussels, and Mons. The interviewees ranged from mid-ranking military officers to ambassadors, and all had personal experience in dealing with one or more of the issues involving NATO doctrine and structure from the early 1990s through the first decade of the 21st century.
Chapter 2
Managing Alliances Alliances between political entities have been a fixture of international relations since antiquity. Thanks to the historian Thucydides, we know that as early as the 5th century BCE, the city-state of Athens formed a military alliance with other Greek city-states to create the Delian League. The purpose of that alliance was to protect members from both the threat of Persia and that of the Spartanled Peloponnesian League. Similar in some ways to NATO today, the Delian League was comprised of members who were each represented at a council, who pledged support to each other, and who contributed resources to the common defense. More recently, and especially since the onset of the modern international system of nation-states in 1648, alliances have continued to maintain their relevance. For the modern student of international affairs, and of alliances specifically, the major historical examples of the importance of alliances include the Bismarckian system of alliances in the second half of the 19th century, the alliances of the world wars, and the Cold War era bipolar alliances. Nevertheless, as others have noted, it is difficult to contemplate international affairs at all without invoking the notion of alliances.1 Today, for example, the NATO alliance remains as relevant as ever, despite the demise of its original reason for being. Even when alliances or other international organizations have been unwilling or unable to act, less formal allied arrangements, such as so-called coalitions of the willing, have filled the gaps when collective action was necessary. Hence, it is clear alliances remain a vital part of the international relations firmament. Why states engage in alliances has been the subject of significant study over time, particularly in the last 50 years. History, as many scholars of the last half century have documented, has shown that two or more states ally together for one or more of a small handful of reasons. These reasons include the need to balance against the power of or the threat posed by a feared third party or the desire to ally with another state in order to join what is perceived as the winning team, as in so-called ‘bandwagoning.’ In contrast to the study of alliance formation, a subject that has received less attention is that of alliance maintenance and management. In light of the continued, perhaps even growing, importance of alliances in international relations, this seems particularly inconsistent, and both scholars and practitioners need to understand why and how alliances manage themselves, especially during times of change. 1 George Liska, Nations in Alliance: The Limits of Interdependence (Baltimore, MD: The Johns Hopkins University Press), 1962.
10
Alliance Management and Maintenance
Work on Alliance Management to Date Many scholars have attempted to expand our understanding of alliances over the last couple of decades. Stephen Walt’s important work, The Origins of Alliances, represented a significant theoretical advance over more traditional interpretations on the formation of alliances in its emphasis on balancing against threats versus simply against power.2 However, his work dealt primarily with alliance formation, not management, and his empirical focus was the Middle East – two limiting factors for those interested in alliance maintenance in general. The few studies of alliance maintenance that have been done have tended to focus more on quantitative analysis of alliance duration and failure or on whether alliance formation leads to war, or were derivatives of the larger academic discussion on collective goods theory and therefore focused primarily on burdensharing.3 Such studies provide limited utility for a qualitative examination of alliance maintenance that entails iterative bargaining along established channels of interaction over the division of common alliance benefits and burdens, such as in the case of NATO’s decisions on command and force structures. In rarer instances, some scholars have attempted to uncover the key variables involved in alliance management. For example, Charles Kupchan examined possible explanations for institutional behavior generally and alliance cohesion more specifically. Kupchan posited four potential independent variables, generated from both systemic and domestic models of international politics, that might explain intra-alliance behavior:4 1. External Threat hypothesis – alliance behavior is a response to the rise and fall of external threats to security. 2. Alliance Security Dilemma hypothesis – alliance cohesion is a function of the coercive potential of the alliance leader and its ability to exact cooperative behavior from its weaker allies. 3. Collective Action hypothesis – alliance behavior is a public goods problem, defined by the avoidance of responsibility. 4. Domestic Politics hypothesis – alliance behavior is determined by political and economic factors at the domestic level.
2 Stephen M. Walt, The Origin of Alliances (Ithaca: Cornell University Press, 1987); for the more traditional realist or neorealist take, see for example Glenn Snyder, “The Security Dilemma in Alliance Politics,” World Politics (July 1984), pp. 461–95; and Hans J. Morgenthau, Politics Among Nations (New York: Alfred A. Knopf, 1956). 3 D. Scott Bennett, “Testing Alternative Models of Alliance Duration, 1816–1984,” American Journal of Political Science (July 1997), pp. 846–78; Mancur Olson and Richard Zeckhauser, “An Economic Review of Alliances,” The Review of Economics and Statistics (August 1966), pp. 266–79; or see Wallace J. Thies, Friendly Rivals: Bargaining and Burden-Shifting in NATO (Armonk, New York: M.E. Sharpe, Inc., 2003). 4 Charles A. Kupchan, “NATO and the Persian Gulf: Examining Intra-Alliance Behavior,” International Organization, Spring 1988.
Managing Alliances
11
Kupchan tests these competing explanations for the behavior of security institutions in one case study – NATO behavior toward the Persian Gulf region in the early 1980s – concluding that the Alliance Security Dilemma hypothesis best explained increased alliance cooperation regarding policy toward the Persian Gulf in 1980–82. Oran Young addressed how institutions, including those focused on security issues, are formed and how they change over time.5 Young accomplished this by examining behavior in the international arena at two different levels of analysis – the level of the state and the level of the organization. With specific regard to Young’s work on how security institutions change and manage that change over time, he relied on three explanations: 1. International security institutions can change when internal contradictions arise that cause mismatches between tasks and missions or ends and means. 2. Security institutions can change when there are shifts in the underlying structures of power. 3. Security institutions can change when exogenous factors such as technology have a significant impact on the institutions in question. Randolph Siverson and Harvey Starr addressed the behavior of security institutions by relying on actor-level domestic political phenomena such as bureaucratic politics theory. The authors specifically cited the impact that domestic political phenomena such as changes in government have on alliances.6 Ultimately however, these scholarly writings do not add up to a body of knowledge that can be used to describe or explain the case of NATO’s post-Cold War transformation and how the alliance managed itself through a period of significant change and transition to wind up with the force and command structures it did in the early part of the 21st century. For example, Kupchan relies on a single case study, so his conclusions are therefore of limited applicability.7 Moreover, he blurs a collection of actor-centric variables into one catchall ‘Domestic Politics’ hypothesis without devoting any attention to specific actor-level explanations such as organizational factors, bureaucratic imperatives, or electoral politics. Despite his intentions, Young ends up expending more effort discussing regime formation than he does regime change. Moreover, he does not address specifically the modalities of change within security-focused “negotiated” organizations, such as NATO. For their part, Siverson and Starr are more focused on government change and the subsequent stability of a government’s commitment toward an alliance – their goal is to determine the veracity of realist and neorealist claims that a state’s alliance ‘portfolio’ should be insensitive to the nature of its government, and are therefore less interested in building up a body of knowledge on alliance management per se. 5 Oran R. Young, “Regime Dynamics: The Rise and Fall of International Regimes,” International Organization, Spring 1982. 6 Randolph Siverson and Harvey Starr, “Regime Change and the Restructuring of Alliances,” American Journal of Political Science, February 1994. 7 Kupchan admits this toward the end of his article when he writes, “…a single case study cannot confirm or disconfirm the applicability of these hypotheses to other alliance debates….”
12
Alliance Management and Maintenance
Addressing Change in Alliances If the scholarly literature is not up to the job, the obvious question begged by a call to study the management of alliances is, in what manner might one accomplish such a task. When it comes to studying the management of alliances, focusing on alliance decision-making – especially in the realm of how it provides security for its members – in the face of change is a particularly compelling methodology for discovering root causes and identifying independent variables. From a relative perspective, the dayto-day management of an alliance like NATO is not of particular interest because much of it occurs on a sort of administrative or organizational auto-pilot. This is not to diminish the overall importance of this type of alliance “management,” but issues dealing with alliance ends, ways, and means during periods of change are far more appealing from both theoretical and practical perspectives. For example, the student of alliance management is far more interested in how the alliance responded to the end of the Cold War and the decisions that comprised that response than in how the alliance handled its supply requisition process, its personnel staffing efforts, or its other standard operating procedures. This study of alliance management will focus most broadly on the alliance response to the security environment following the end of the Cold War. A critical element of that story involves the transformation of the alliance from what was primarily a collective defense organization with elements of collective security to primarily a collective security organization with elements of collective defense. Certainly the alliance retains collective defense – as embodied in Article 5 – as a central element of its reason for being. Nonetheless, the alliance’s emphasis has shifted from traditional collective defense, epitomized by a reliance on territorial defense forces and organizations, toward more collective security functions, as seen in the shift toward more expeditionary forces and structures. This evolution has been well-documented by scholars. For example, David Yost examined how NATO’s raison d’être has evolved since the end of the Cold War and why NATO has remained relevant.8 What is less frequently acknowledged though is the fact that NATO is certainly not the first such organization to evolve from a collective defense alliance to one focused largely on collective security – the St. Petersburg Convention of 1873 and the Franco-Belgian Accord of 1920 were both collective defense alliances that evolved into collective security arrangements well before NATO’s own transformation. The St. Petersburg Convention, signed in St. Petersburg on 6 May 1873 by Emperor William I of Germany and Tsar Alexander II of Russia, provided for either party to come to the aid of the other with 200,000 men if either was attacked by another 8 Yost identifies eight explanations for NATO’s continued importance in the post-Cold War era – maintaining US engagement in European security; resolving intra-West European security dilemmas; reassuring Germany’s neighbors and allies; limiting the scope of nuclear proliferation in NATO Europe; promoting the denationalization of defense planning; providing a forum for the coordination of Western security policies; supplying economic benefits to all the allies; and encouraging and legitimizing democratic forms of government. David S. Yost, NATO Transformed: The Alliance’s New Roles in International Security (Washington, DC: US Institute of Peace), 1999.
Managing Alliances
13
European power. Austria was invited into the alliance in early June 1873 when the Tsar visited Vienna, but the Austrian Emperor Francis Joseph rejected this primarily because of his concerns over Balkan politics. Although no significant international disagreements stood between Germany and Russia forming an alliance, Russia and Austria were rivals in the Balkans, and Vienna was therefore reluctant to engage in an alliance with Russia that might complicate the pursuit of its interests in the Balkans. Instead, Russia and Austria signed a much more general convention on 6 June 1873 known as the Schönbrunn Convention. It involved merely a promise to take counsel and work out a military agreement if action became necessary. The Schönbrunn Convention, in contrast to the St. Petersburg Convention, was designed to preserve peace more broadly through collective security and did not involve a commitment to come to the defense of another signatory. William I of Germany accepted the Schönbrunn Convention when he traveled to Vienna in October 1873. Little more was heard of in regard to the St. Petersburg Convention and it was not entirely clear if it was still considered valid following Germany’s accession to the Schönbrunn Convention. In 1881, when discussions were opened toward the conclusion of a formal alliance between the three governments, there was some reference to the St. Petersburg Convention and at that time it officially was declared superseded by the Schönbrunn Convention. The negotiations in 1881 led to the signing of the Alliance of the Three Emperors, in which Austria, Germany, and Russia pledged to maintain neutrality should one of them get into a war with a fourth country. Hence, what began as an alliance of collective defense involving a commitment to come to the aid of an ally evolved over the course of several years into a more general collective security arrangement designed to keep the peace between all signatories. The Franco-Belgian Accord of 1920 is another example of an alliance that evolved from collective defense to collective security – or at least an example of a collective defense arrangement that was replaced by a collective security agreement.9 Following World War I, Belgium desired some sort of protection against the possible resurgence of Germany, but at the same time it wanted to avoid becoming a French satellite or getting dragged into a war by an allied country bent on enforcing the Treaty of Versailles. In part, Belgium’s approach was a reflection of the divide between the French-speaking Walloons and the Dutch-speaking Flemings – the former favored closer ties with France while the latter preferred a more arms-length policy with Paris. For its part, France wanted a way to quickly reinforce the Rhineland – to be occupied by allied troops for 15 years as per the terms of the Treaty of Versailles – if relations with Germany soured or if Germany showed signs of rearming. Additionally, France wanted to ensure Belgium would be on its side in the event of another war with Germany. Following initial discussions between French and Belgian officials, formal negotiations toward a technical military agreement began in April 1920. The first 9 For a detailed account of the negotiations leading to the Franco-Belgian Accord, and especially the political dynamics inside Belgium at the time, see Jonathan Helmreich, “The Negotiation of the Franco-Belgian Military Accord of 1920,” French Historical Studies, vol. 3, no. 3 (Spring 1964), pp. 360–378.
14
Alliance Management and Maintenance
article of the accord pertained specifically to the occupation of the Rhineland. It outlined the requirement for each country to provide forces for the occupation and for reinforcement should Germany exhibit or threaten aggression. It specified that, should the Germans re-arm, Belgium and France would each mobilize their armed forces. The second article of the Accord outlined what each country would do as the Rhineland occupation wound down, including the integration of defense measures along the Belgian, Luxembourg, and French frontiers. The third and final article of the Accord covered the annual defense cooperation talks to be conducted by the general staffs of each country. Following further negotiations between senior military officials through the summer of 1920 – and unsuccessful efforts by the Belgians to bring in the British – the Accord was signed by French and Belgian military leaders on 7 September 1920. Several days later, the Belgian and French prime ministers exchanged letters acknowledging the agreement and essentially ratifying it. In 1925, the Treaty of Locarno opened for signature. This collective security treaty required each of the major European powers (France, Italy, Germany and the United Kingdom) to agree to maintain the peace in Europe. Other small powers, such as Belgium, also signed. After an exchange of diplomatic communications and bilateral consultations, the Belgian and French governments concluded that Locarno superseded the Military Accord of 1920. In 1936, the Accord was officially abrogated – this was a necessary step, as the Accord itself contained no provisions for its termination or on its duration. If both the Franco-Belgian Accord and the St. Petersburg Convention preceded NATO as collective defense agreements that evolved into collective security arrangements, perhaps they hold lessons for NATO on how alliances manage themselves through periods of transition. Of course, what differs significantly between the cases of the Franco-Belgian Accord and the St. Petersburg Convention on the one hand and NATO on the other are that the former lacked integrated structures like what NATO would eventually develop. Neither the Franco-Belgian Accord nor the St. Petersburg Convention had multinational command structures or integrated military planning and organization along the lines that NATO has achieved in its nearly 60 years of existence. What both of these examples from the past do have in common though is the role played by changing threat perceptions. In the case of the first example, Russian concerns with regard to Vienna’s intentions in the Balkans as well as those of the Ottoman Empire prompted it to settle for the collective security of the St. Petersburg Convention – which included Austria – even though it might have preferred the terms of the Schönbrunn Convention. For their part, the Austrians were incentivized to sign on to the St. Petersburg Convention based on their own concerns over the Ottoman Empire as well as the coming to power of the Gladstone government in Britain, which was perceived as unfriendly toward Vienna. With regard to the second example, the Franco-Belgian Accord of 1920 was deemed superseded by Locarno in part because French and Belgian leaders viewed the collective security provided by the Locarno Treaty, which included both Britain and Germany, as a more effective means of guaranteeing their borders. Perhaps more importantly, Belgian leaders later formally abrogated the Accord in 1936 in order to return their country to a neutral status, believing it was the best way to avoid antagonizing a resurgent Germany, which had just re-militarized the
Managing Alliances
15
Rhineland without any negative repercussions. In the cases of the St. Petersburg Convention and the Franco-Belgian Accord, security threats played the key role in driving alliance management and evolution. Nevertheless, most scholarly studies addressing NATO’s transformation not only gloss over the point that some alliances of the past have transformed from collective defense to collective security – and hence could prove useful in analyzing NATO’s evolution – they also have tended to overlook the transformation of alliance command and force structures in the context of alliance maintenance or management. This is short-sighted, since how NATO dealt with fundamental change in its approach to security – from collective defense to collective security – provides a unique window through which to assess alliance management and maintenance through a turbulent time period. The Role of Organizational Studies in Analyzing NATO If an examination of change in NATO’s ways and means is a compelling method of studying alliance management and maintenance, the utility of employing organizational studies to shed light on that change must also be addressed. Given that NATO is an international military and political organization with both an integrated military structure (IMS) and a civilian secretariat, this makes sense. As an organization, NATO has changed significantly since its founding. Indeed, the Treaty of Washington establishing the alliance in 1949 offered only vague references to the administrative and political structures available to the alliance – in fact, no specific mention was made of the Secretary General. The treaty only referenced the establishment of a Council – which would later become the North Atlantic Council – that would set up subsidiary bodies as necessary.10 Nevertheless, in order to handle the duties of the alliance, the Council quickly expanded the array of entities supporting the alliance – by December 1949, only eight months since the treaty had been signed, NATO had established under itself seven working staffs, boards, or committees. It took just over another year before NATO established a working group to develop a budget for the growing NATO staff, and in May 1951 the alliance members approved the establishment of an international budget. By the end of that same year, NATO had a permanent, continuously sitting central body known as the Council Deputies, its own unified secretariat, and three major operating agencies.11 More than half a century later, the alliance’s structure has clearly grown, taking on a diverse array of functions. By 2001, NATO employed 1,750 people at its headquarters on a fulltime basis, was comprised of over a dozen major offices or divisions, and included 12 subordinate alliance policy committees, organizations, or agencies.12 10 “The North Atlantic Treaty (Washington, DC, 4 April 1949),” as reprinted in The NATO Handbook (1991), p. 529. 11 Robert S. Jordan, The NATO International Secretariat/Staff, 1952–57 (London: Oxford University Press, 1967), p. 25. 12 The NATO Handbook (2001). The employment figure does not include the 1,400 personnel or so that also work at the headquarters but who are members of national diplomatic
Alliance Management and Maintenance
16
Given the growth of the alliance – particularly in terms of its organizational size but also in terms of its responsibilities – it seems appropriate to ask how the study of organizations might contribute to an understanding of alliance management and maintenance. Organizations are established to accomplish objectives that individual entities cannot achieve alone.13 In order to achieve their goals, organizations combine inputs, processes, and outputs into standard operating procedures (SOPs). These SOPs are further combined into programs, which become elements in the organization’s repertoire of behaviors. These repertoires are necessary for an organization to control and synthesize the typically large, unwieldy sub-units of which it is comprised and to protect the organization from uncertain externalities.14 The individual humans that comprise those sub-units usually develop personal and professional interests in the well being of their sub-unit and hence in its repertoire; without the repertoire, the sub-unit is nothing but a group of people. From the organization’s perspective, the better the individuals of a given sub-unit perform their repertoires, the more effectively the organization can control and direct the efforts of all the sub-units. Over time, the performance and interests of individuals becomes institutionalized – in many cases, organizations themselves are personified through the behavior of individuals that seek to protect and promote the ‘equities’ of particular organizations. The institutionalization of behaviors results in organizations maintaining programs and even repertoires long after they have become outmoded or unnecessary. However, it is through the repeated use of programs and repertoires that organizations succeed. Based on the scholarly literature regarding international organizations, one might expect to see an organization like NATO engage in three types of behavior in order to manage the significant changes brought about by the end of the Cold War15: • • •
Resist change by ignoring inconsistencies between means and ends and by continuing to do what has been done before; Anticipate, adapt, and learn only when necessary; and, Ration organizational activity so as to perform only the most vital tasks first, followed by less important tasks.
Resisting Change In the context of the Cold War’s end, NATO’s survival was very much an issue in the late 1980s and early 1990s. The alliance could have been expected to cling to existing modes of behavior following the end of the Cold War and therefore to have or military delegations. 13 James D. Thompson, Organizations in Action (New Brunswick, NJ: Transaction Publishers, 2003, originally published by McGraw-Hill, 1967), p. 15. 14 Thompson, Organizations in Action, p. 24; Ernst Haas, When Knowledge is Power: Three Models of Change in International Organizations (Berkeley: University of California Press, 1990), p. 55. 15 Adapted from Thompson, Organizations in Action; Haas, When Knowledge is Power; Matthew Holden, Jr., “’Imperialism’ in Bureaucracy,” The American Political Science Review, vol. 60, no. 4 (December 1966), pp. 943–51; and Robert McCalla, “NATO’s Persistence After the Cold War,” International Organization, vol. 50, no. 3 (Summer 1996), pp. 445–75.
Managing Alliances
17
made little if any changes in its doctrine and force structure even though they may have been inappropriate for responding to the new security threats. Samuel Finer, a leading British scholar of public administration and organization, wrote about this phenomenon in 1962, noting that military organizations tend to be more dependent on standard operating procedures than their civilian counterparts and tend to retain those SOPs long after they have retained their usefulness.16 Other scholars, such as Barry Posen, have taken this analysis further, concluding that retention of outmoded programs and repertoires contributes to stagnant military doctrines. Adapting and Learning Sometimes, focusing solely on implementing well-known programs and repertoires as a means of managing uncertain externalities is not enough for an organization to succeed – it must therefore have the capacity to anticipate and adapt when new uncertain externalities cannot be overcome or mitigated by simply resisting change and resorting to familiar SOPs.17 Ernst Haas, a leading scholar of change in international organizations, wrote that when confronted with shifting externalities, organizations are likely to elicit one of three behaviors. First, they may adapt through a process he calls “incremental growth” – that is, the successive augmentation of an organization’s repertoire as member states or organizational leaders add new tasks to older ones without any change in the organization’s decision-making methodology.18 Second, Haas wrote that organizations might adapt through a process he calls “turbulent nongrowth,” whereby there are major changes in organizational decision-making, ends no longer cohere, and there are significant disconnects between ends and means. Finally, Haas noted that some organizations might change through learning, or what he terms “managed interdependence.” This entails a reexamination of the purposes of the organization in question, based on a knowledge-mediated decision-making dynamic. Haas did not address NATO – or for that matter any formal politicalmilitary alliances – among the 36 different international organizations studied in his book, When Knowledge is Power, but one can extrapolate inferences from his conclusions that can then be applied to NATO. Organizations, Haas noted, are most likely to engage in learning as a management technique for dealing with change when two conditions are met:19 • •
First, organizations must be led by a dominant power or coalition for long periods of time; and, Second, organizations must have a core anchoring concept on which all member can agree and around which develops a set of consensual precepts.
It is unclear whether NATO is equipped to employ learning as a method for managing change. On the one hand, the alliance has clearly had a dominant member since its 16 Samuel Finer, The Man on Horseback: The Role of the Military in Politics (New Brunswick, NJ: Transaction Publishers, 2002, original published by Pall Mall Press, 1962), p. 7. 17 Thompson, Organizations in Action, pp. 21–3. 18 Haas, When Knowledge is Power, p. 4. 19 Ibid., pp. 161–4.
Alliance Management and Maintenance
18
inception – the United States. On the other hand however, it is not clear whether the alliance’s central concept – presumably the Article 5 commitment to mutual defense – is as useful in terms of anchoring all its member states given the absence of the Soviet threat. Indeed, even if the alliance was characterized by these two points, this would not necessarily guarantee that NATO would exhibit learning versus some form of less efficient, less productive adaptation and management. Rationing Activity When implementing well-known repertoires fails and organizations prove incapable of adapting or learning, organizations might engage in rationing.20 That is, organizations under great stress from environmental change tend to focus on the most vital tasks and accomplish them first as a means of managing that change. In the context of NATO and its force and command structures, rationing might have led to a delay in implementing dramatic changes to doctrine and structures in the face of the collapse of the Soviet Union and the rise of a host of new, unconventional security threats. That is, again in the context of force and command structures, the alliance might have continued for some time to focus its energies and resources on territorial defense. As will be shown later in this book, there is some evidence that, as NATO managed the tremendous change brought on by the end of the Cold War in terms of its doctrine and structures, the alliance exhibited behavior similar to what some organizational studies theorists might have expected, including resistance to change, eventual adaptation, and rationing. However, what an approach based on organizational studies fails to appreciate fully, and what will be shown later in this book, is the degree to which much of this behavior was motivated by or was a byproduct of political bargaining and the complex array of interests and dependencies that underpin it. In short, what might at first appear to be the dynamics of organizational processes at work can be traced more specifically in terms of root causes to the interplay of member state interest as expressed and exercised through political bargaining. The next chapter will take the reader farther down this path by helping to specify the terms of the case study to be examined in the rest of the book.
20 Thompson, Organizations in Action, p. 23.
Chapter 3
Assessing Military Doctrine and Structures Before examining how and why NATO went from the end of the Cold War to the NRDCs, we first need to define the terms of the story ahead. Specifically, we require an understanding of what doctrine is and what role it plays in the development of military command and force structures. Only by understanding the nature of doctrine and its relationship to structures can one understand the specific elements of the case study, such as why NATO’s military doctrine and command and force structures changed as they did and why the changes took as long to implement as they did. By answering these questions – which form what academics might call the dependent variables of the story – we can then gain insights into broader subjects such as alliance management and the sources of military doctrine. Doctrine and its Role There is no precise agreement among academic scholars on how to define doctrine. Most modern academic works take their lead from Barry Posen’s The Sources of Military Doctrine, one of the best-known works on military doctrine. However, Posen devotes only limited attention to the actual definition of his dependent variable, military doctrine. He relegates most of his attention to the topic to an endnote where he states that there is little agreement among scholars on topics such as tactics, doctrine, or grand strategy. He then briefly stakes out where he thinks the differences lie between the three: In simplest terms, however, tactics is the study of how fights are fought. In my view, once one begins to ask questions about how battles are fought, one has entered the realm of military doctrine. When one begins to ask which wars will be fought, or if war should be fought, one has entered the realm of strategy.1
In Posen’s conceptualization, doctrine is one part of a broader planning hierarchy, resting between and linking tactics and strategy. Other key academic studies on the subject of military doctrine, including those by Kimberly Zisk Marten, Deborah Avant, Elizabeth Kier, Stuart Kaufman, and Stephen Peter Rosen, treat their subject matter in a similar manner insofar as they tend to favor Posen’s concepts on doctrine’s definition and its role. 1 Barry Posen, The Sources of Military Doctrine: France, Britain, and Germany Between the World Wars (Ithaca, NY: Cornell University Press, 1984), p. 245. Emphasis in the original.
20
Alliance Management and Maintenance
These concepts are only of limited value when examining NATO’s post-Cold War evolution for two reasons. First, some of the leading academic works commingle notions such as policy, strategy, and doctrine, which in NATO’s view are discrete concepts. Second, relying on the leading academic notions of doctrine’s role – simply a link between tactics and strategy – makes an examination of the linkages between NATO’s doctrine and its command and force structures very difficult. In order to enable a more complete understanding of the alliance’s changes during the 1990s, we need to employ a greater degree of specificity in conceptualizing doctrine and its role. One possible source for such specificity is the world of military practitioners, who have more precise notions of doctrine and its role in determining military structures. Military practitioners in the United States define doctrine as the fundamental principles – authoritative in nature but requiring judgment in application – by which the military forces or elements thereof guide their actions in support of objectives.2 Doctrine is a statement of officially sanctioned beliefs, warfighting principles, and terminology that describes and guides the proper use of combat forces in military operations. Practitioners believe that doctrine should provide information on the approach to achieving broader goals and missions, on what military organizations should look like and why, on what their lines of authority are, on what degree of control they have over their forces, and on how they are supported. For example, Field Manual (FM) 3-0, the definitive doctrinal document on land warfare operations for the US Army, points out that doctrine is the “expression” of how Army forces contribute to unified action in campaigns, major operations, battles, and engagements.3 It describes the Army’s approach and contributions to combat operations on land in joint operations by discussing the Army’s role in peace, conflict, and war; the fundamentals of full spectrum operations, battle command, and the operations process; the various types of Army operations; and information superiority and combat service support as enabling operations. NATO also defines doctrine as the fundamental principles, authoritative but requiring judgment in application, by which the military forces or elements thereof guide their actions in support of objectives.4 And like many of its member states, NATO’s doctrine fulfills the role of providing information on how alliance military forces will be organized, trained, commanded, and utilized. For example, NATO’s military doctrine regarding alliance command and force structure takes the form of three documents produced by the Military Committee (MC): 1. MC 400 contains the fundamental concepts behind the military implementation of the alliance strategy; 2. MC 324 covers the alliance command structure; and, 3. MC 317 deals with the alliance force structure. 2 Joint Publication 1–02, “DOD Dictionary of Military and Associated Terms,” as amended through 14 August 2002. 3 “FM 3–0: Operations,” Department of the Army, June 2001. 4 See the NATO Standardization Agency’s “Glossary of Terms and Definitions” (2002).
Assessing Military Doctrine and Structures
21
Each of these documents outlines how NATO will employ military means to achieve its objectives both within and outside its area of operations.5 For instance, MC 400, conceptually the broadest of the three documents, specifies four critical areas for the military implementation of NATO strategy: 1 2
3
4
The missions of allied military forces – Article 5 collective defense, nonArticle 5 crisis response, and consultation and cooperation; The critical capabilities necessary to fulfill the missions, including inter alia timely force availability, deployability and mobility, logistics sustainability, and survivability; The principles underpinning alliance military structures, which include the type and size of conflicts that NATO must be prepared to engage in, thereby establishing the central requirements for the alliance’s command and force structures; and, Additional arrangements supporting the missions of allied military forces, such as defense planning processes, common allied funding of certain functions or tasks, integrated air defense, and the stationing of forces away from home territory.
NATO’s doctrine on the military implementation of allied strategy is therefore a multifaceted set of guidelines that become the basis for military action at the strategic, operational, and tactical levels. As can be seen in just this one example of MC 400, NATO’s doctrinal concepts go beyond general notions of how battles are fought and they point directly to the types of command and force structures necessary to fulfill alliance strategy. For the purposes of this book, we will rely on a conceptualization of doctrine and its role that are closer to that of practitioners. One needs to recognize and take into consideration the link between command and force structures on the one hand and doctrine on the other if one hopes to have as complete an understanding of the alliance’s post-Cold War changes as possible. Otherwise, one might conclude that NATO had changed its doctrine in the early 1990s when it actually had not, which would subsequently color the analysis of alliance management. Relying on more specific concepts regarding doctrine’s definition and its role enables a more precise examination of alliance management and change. The Sources of Doctrine Among academic scholars there is a broad divide between two groups regarding doctrine’s sources. First, there are those that generally favor traditional structural sources or approaches – such as the structure of the international system – or functional approaches – such as the nature of organizations – for explaining doctrine. Second, there are those that favor alternative explanations that rely on somewhat more actor-specific political, cultural, or institutional factors. The analytical tools 5 NATO’s ‘area of operations’ is commonly understood to be the territory, airspace, and sea space of its member states.
22
Alliance Management and Maintenance
employed by those advocating the more traditional structural or functional approaches are generally considered by critics to be too blunt for accurately explaining what occurs in the real world – although parsimonious and straightforward, structural and functional arguments lack necessary refinement and nuance. In response to the limited utility of the traditional approaches, other academic scholars have developed alternative approaches, some of which rely upon institutional characteristics, culture, or political factors to explain cases where structural or functional explanations alone cannot do the job or where they prove inadequate for other reasons. The alternative approaches add significantly more powerful tools to aid in our hunt for the sources of military doctrine by offering a level of richness that goes beyond the parsimony of traditional works. These alternative approaches, with their emphasis on the particular context of specific actors, tend toward what Alexander George called actor-specific models of political behavior.6 Nonetheless, it is unclear whether the alternative explanations go far enough in providing the contextual richness necessary to have a complete understanding of doctrinal change within NATO. For example, some alternative approaches hold that political context is important for determining how electoral rules and constitutional structures create incentives for political actors. In NATO’s case, the political context during different points in the alliance’s history certainly has helped to shape its institutions – for instance, the intra-alliance political rivalry between the United States and France has been a leading factor behind American efforts to increase the responsibilities of the Defense Planning Committee (DPC), an alliance body on which the French are not voting members. However, the political context has also shaped more iterative interactions between the allies, not just those arguably less frequent interactions that led to development of or changes in NATO’s institutional characteristics. Political context – such as pressure to appease a restive electorate or the necessity of balancing competing budget requirements – shapes member state interests and perceptions as the allies conduct their regular meetings and coordination sessions week after week. In order to gain a better understanding of how politics influences doctrine in the real world every day, we can turn to practitioner views on the sources of military doctrine. Although they tend to be very actor-specific, we can nonetheless extrapolate from practitioner views some notions on the sources of military doctrine that augment and in some cases reinforce academic notions, thereby attaining the specificity necessary to examine the sources of NATO’s doctrinal change and observe the ways in which the alliance managed significant change. Practitioner views are helpful here because they push us in the direction of placing more reliance on areas traditionally overlooked by most academic works on the sources of military doctrine, such as specific political factors unique to countries and peoples at particular points in time. However, practitioners also validate some of the traditional approaches, albeit in a more actor-specific fashion, when they emphasize 6 Alexander L. George, Bridging the Gap: Theory & Practice in Foreign Policy (Washington, DC: US Institute of Peace Press, 1993). George argues for the development of such actor-specific behavioral models by academics so that decision-makers can shape better foreign policies. See pp. 125–31.
Assessing Military Doctrine and Structures
23
the role of threats and experience. For example, practitioners and practitionerscholars like Clausewitz draw on some of the same notions for determining doctrine as academic scholars – such as security threats and political factors – but practitioners tend to favor greater actor-level specificity than academic scholars. For an examination of NATO’s military doctrine, consideration of in-depth political factors and threat factors makes sense since the alliance is as much a political organization as a military one. Indeed, the story of how NATO military doctrine evolved through the 1990s and early part of the 21st century cannot be told without some reliance on both actor-specific political factors and security threats. Relying on both of these sets of factors to explain doctrine leads us toward a more actor-specific approach for building models of behavior with regard to doctrinal development, which both non-traditional academic scholars (implicitly) and practitioners (explicitly) view as important for explaining and hence understanding doctrinal development. However, the notion of ‘political factors’ as an independent variable is obviously broad, encapsulating everything from political culture to electoral politics to bureaucratic politics, to name a few – each of these is capable of commanding enough attention to fill numerous books. Moreover, each is typically used to explain a state’s behavior – in other words, these theories would have little to say about the behavior of an international military alliance such as NATO. What is needed then is a more specific understanding of what constitutes ‘political factors’ that simultaneously is appropriate for an in-depth examination of a military alliance. For this sort of understanding – and in order to properly model the political factors at work within an alliance – it is necessary to turn to the academic literature on intraalliance bargaining. Intra-Alliance Bargaining The literature on intra-alliance bargaining is generally subsumed by the broader literature on alliance formation and maintenance, a field that has seen surprisingly little attention from scholars relative to other areas of study in international relations. There is at least one relatively recent exception to this – Glenn Snyder’s Alliance Politics. Snyder addresses alliance maintenance and intra-alliance bargaining beyond the standard quantitative burden-sharing construct by examining various alliance relations among Austria, France, Germany, Russia, and the United Kingdom from 1879 to 1914.7 He argues that bargaining power within an alliance depends on three factors – the allies’ dependence on the alliance, their commitment to the alliance, and their comparative interest in the object of the bargaining.8 As mentioned earlier, a state’s bargaining power – when it has a higher chance of obtaining its preferred outcome – is greatest when its dependence is low, its commitment is loose, and its interests at stake are great.
7 Snyder, Alliance Politics. 8 Ibid., p. 166. This construct assumes that alliance members have an interest in maintaining the alliance because they receive benefits from it.
Alliance Management and Maintenance
24
Snyder defines dependence as a function of the net security benefit the ally receives from the alliance. Commitment is defined as an “arrangement of values that disposes one to act in a certain way,” arising from the nature of the contract or agreement between the allies, as well as a state’s interest in aiding an ally that exists apart from the promise spelled out in the agreement establishing the alliance.9 Finally, Snyder defines a country’s interest in an object of bargaining as the value it places on that object. In Snyder’s conceptualization, the interest component of the bargaining equation is what an ally stands to gain, while the commitment and dependence elements represent what an ally stands to lose. Of course, much of this is subjective, and Snyder acknowledges as much when he notes that assessments of interest, commitment, and dependence are all subject to various interpretations and perceptions (especially those of the decision-makers involved). Nonetheless, one can at least assume that policymakers have a good idea of the various interests, commitments, and dependency levels of their own countries. Using Snyder’s approach to model political bargaining within NATO in the post-Cold War is not without some challenges however. First, with regard to the utility of the dependence variable, since the end of the Cold War, the threats facing the allies have changed dramatically. Although all the allies now face new, post-Cold War threats, none of these pose the same existential challenges that the Soviet threat did. Terrorist plots and Balkans crises, for instance, do not compare to the threat of nuclear war or a Soviet armored assault through central Germany. Nonetheless, some of the allies are still clearly dependent on the alliance for critical security needs. For instance, Estonia, Latvia, and Lithuania are very much dependent on the alliance to provide fixed wing patrols of their airspace, something they view as vital given their proximity to Russia. So, although the dependence variable may not have the same compelling weight across the entire alliance that it may have had during the Cold War, it is still a factor to be considered – both in terms of an individual ally’s dependence on its alliance partners and the alliance’s dependence as a whole on the security contributions of individual members – in assessments of post-Cold War bargaining strength among the allies and between individual allies and the alliance as a whole. In contrast, the commitment variable is of far less applicability, since every member’s commitment is based on ratification of the Treaty of Washington. In theory at least, all the allies are equally committed to the alliance, but of course there are sometimes not-so-subtle differences between Atlanticists and Europeanists within the alliance. Nonetheless, for the purposes of this book, Snyder’s commitment variable is not very useful. Additionally, Snyder tends to assume unitary state actors, which could be problematic for studying an alliance comprised of diverse democracies, each with their own set of domestic political institutions and actors. He notes that his analysis, “exclude[s] domestic politics from the theory, largely on the grounds of parsimony.” However, he admits that his cases show, “that domestic factors sometimes make a difference.”10 Therefore, without offering an all-encompassing theory of the interaction between domestic 9 Ibid., p. 169. 10 Ibid., pp. 131–3.
Assessing Military Doctrine and Structures
25
politics and international relations in determining bargaining outcomes, he chooses to note internal political influences in the case studies when he deems them significant, although these instances appear somewhat arbitrary or at least open to interpretation. Of course, studies of the interaction between domestic politics and international relations are not necessarily new, even if those that deal specifically with NATO and security issues have been rare.11 Richard Eichenberg’s study of how domestic politics influenced bargaining over intermediate nuclear forces (INF) within NATO is one such rare example.12 He found, somewhat counter-intuitively, that a lack of domestic pressures led to disagreement between the allies on INF issues in 1977, whereas the presence of domestic pressures led to negotiated agreement between the allies on INF issues in 1979, 1983, and 1988. This was so because the lack of domestic pressures enabled allied leaders to express their personal, diverging preferences in 1977 and hence led to disagreement. In subsequent rounds of bargaining, the presence of such pressures pushed allies toward solution sets that benefited both them and their allies because the INF issue had become a crucial case for the well being of the alliance and no allied leader wanted to be seen as contributing to NATO’s demise. This sort of analysis is useful for providing an example of how one might factor domestic-level politics into allied decision-making, and it helps to augment what is generally a cursory treatment of this subject by Snyder. Simply put, Eichenberg’s study points to the domestic political rewards of international agreement, especially for leaders who might be facing tough times at home. If allied leaders are not particularly weak at home and hence are not in need of a political boost from an international agreement, then they may be less likely to move toward solutions in their allies’ win set, all else being equal. Moreover, this study indicates that when a particular alliance issue is perceived as a crucial test for alliance integrity, allied governments are likely to try to find common, acceptable solutions, even if their individual win sets are shrinking due to other domestic pressures, such as economic ones. On the other hand, if a particular alliance issue is not perceived as a crucial test for NATO’s integrity, allied governments may be less willing to migrate quickly toward each other’s win sets and hence arrive forthrightly at common solutions, all else being equal.
11 For a collection of studies on the interaction of domestic and international politics, see Peter B. Evans, Harold K. Jacobson, and Robert D. Putnam, eds, Double-Edged Diplomacy: International Bargaining and Domestic Politics (Berkeley: University of California Press, 1993). For a detailed study on the role of interest group politics on the policy choices of just one ally, the United States, and the immense complexity inherent in such an undertaking, see Kay L. Schlozman and John T. Tierney, Organized Interests and American Democracy (New York: Harper and Row, 1986). Likewise, for a study on the complexity involved in interest group competition in specific issue areas of US domestic policy, see John P. Heinz, Edward O. Laumann, Robert L. Nelson, and Robert H. Salisbury, The Hollow Core, Private Interests in National Policy Making (Cambridge, Massachusetts: Harvard University Press, 1993). 12 Richard C. Eichenberg, “Dual Track and Double Trouble: The Two-Level Politics of INF,” in Evans, Jacobson, and Putnam, Double-Edged Diplomacy, pp. 45–76.
26
Alliance Management and Maintenance
Understanding the Process and Content of Doctrinal Change Even though choosing the two approaches outlined above – one based on security threats and the other based on political bargaining – appears to be a reasonable approach to examining NATO’s doctrinal and structural changes and hence alliance management, one should be certain these are appropriate tools from an empirical standpoint. From this vantage point, certainly threat-based models are appropriate for assessing NATO’s military doctrine. As a military alliance concerned with collective defense, NATO is in the business of identifying and defending against threats. This has been the case since 1949 when the alliance was founded as a means of keeping the Soviet threat at bay.13 Immediately following World War II, military infrastructure and forces were dramatically reduced throughout Western Europe. But farther east, in parts of Europe occupied by the Soviet Army, Moscow chose to maintain its sizable wartime force structure well after the war had ended. The presence of this force on Western Europe’s doorstep plus the declared ideological aims of the Soviet Communist Party; direct threats to the sovereignty of Greece, Norway, and Turkey; the 1948 coup in Czechoslovakia; and the 1948 Berlin blockade all convinced leaders throughout Western Europe and North America that military alliance through the North Atlantic Treaty Organization was the only means by which to counter the Soviet threat.14 Throughout its subsequent existence, NATO’s central focus has been to provide for the security and defense of its member states – even today, more than a decade after the end of the Cold War, this remains the alliance’s core task.15 Similarly, consideration of political bargaining factors makes sense from an empirical point of view, because many perceive the alliance to be as much a political organization – subject to intra-alliance bargaining among its member states – as a military one. For example, Sean Kay noted that NATO has a very strong political nature, which stems from the period of the alliance’s founding – the need to emphasize the principles of democracy and peaceful relations; the need to address challenges from fragile economies and weak political systems; the US-driven need to ensure equitable burden-sharing; and a spirit of consultation and consensus.16 Kay argued that these factors show that NATO was formed around a broad political concept of shared security, rather than simply and solely a military response to the Soviet threat. Kay’s points are backed up in part by the alliance’s own stated fundamental task: to safeguard the security of its members through “political and military means.”17 Former Secretary-General of NATO Manfred Wörner argued this point emphatically as the Cold War came to an end, telling an audience in Turkey that NATO is “not a military coalition wedded to the status quo. [Rather, NATO is] a political alliance 13 Lord Ismay, NATO’s first Secretary General, said that NATO’s purpose was to keep the Russians out, the Americans in, and the Germans down. Even in this three-part conceptualization of the alliance’s raison d’etre, the central thread is threat and responses to it. 14 The NATO Handbook (1991), pp. 29–30. 15 Ibid., p. 30. 16 Sean Kay, NATO and the Future of European Security (Lanham, Maryland: Rowman and Littlefield, 1998), p. 32. 17 The NATO Handbook (1991), p. 30. Emphasis added.
Assessing Military Doctrine and Structures
27
18
striving to transform the status quo.” Additionally, as discussed earlier, many authors have documented the alliance’s history of intra-alliance bargaining over the sharing of collective burdens.19 Expected Outcomes Establishing the utility of employing two approaches to an analysis of allied changes to doctrine and structure, and hence of alliance management, begs the question of what one might expect to see in the end. At first glance, the story of the NRDCs and the path the alliance traveled in arriving at their implementation should be easily explainable by relying on security threats alone to explain the chain of events leading from the end of the Cold War to the NRDCs: 1. The Soviet Union dissolved and with it so too did the need for static, territorial defense doctrines and forces. 2. The post-Soviet threats facing the alliance were unpredictable in terms of location and scale. 3. The change in threat necessitated a change in doctrine and forces – emphasizing deployability, sustainability, and crisis response but retaining the ability to handle an array of conventional threats – to enable NATO to maintain the security of its member states. However, getting to the third point above was not as straightforward as a threatbased analysis alone might have expected, and in some respects – particularly with regard to sustainment and deployment capabilities – has yet to occur fully. For example, if responding to the threat required doctrine on the use of tanks, the procurement of three tanks, and the procurement of ships to transport them on, then a threat-based response would expect NATO to develop doctrine for and then procure three tanks and appropriate transport ships in a timely fashion, so that the alliance could meet the threats facing its members. To draw out the analogy further, this is not what happened – instead, NATO engaged in the equivalent of essentially ignoring the tank requirement at first and thereby allowing threats to grow and fester, buying other vehicles as an interim step, then developing doctrine for and buying six tanks, and finally side-stepping the issue of transport ships altogether. Obviously, festering threats undermine security, unarmored vehicles cannot do the job of tanks, six is more than the alliance said it needed to confront 18 Manfred Wörner, “The Future of the Alliance,” speech given at the University of Istanbul, Turkey, 18 September 1989. 19 See Jane Stromseth, The Origins of Flexible Response: NATO’s Debate over Strategy in the 1960s (London: Macmillan Press, 1988); Ivo Daalder, The Nature and Practice of Flexible Response: NATO Strategy and Theatre Nuclear Forces Since 1967 (New York: Columbia University Press, 1991); Snyder, Alliance Politics; Thies, Friendly Rivals; Robert McCalla, “NATO’s Persistence”; and Andrew Bennett, Joseph Lepgold, and Danny Unger, “Burden-Sharing in the Persian Gulf War,” International Organization, vol. 48, no. 1 (Winter 1994), pp. 39–75.
28
Alliance Management and Maintenance
the threat, and without transport ships the alliance cannot perform the out-of-area missions it says are vital to member security. Indeed, as will be shown, doctrinal integration with strategy were not hallmarks of NATO’s response – in terms of the content of the doctrine and structures the alliance ended up with, the alliance’s response to the collapse of the Soviet Union and the rise of new threats was less than optimal, at least from a threat-based perspective. As will be discussed in greater detail later, the suboptimal response of the alliance to the demise of the Soviet Union led to a waste of limited defense resources, left the alliance unable to sustain and deploy its forces over time and distance, and inhibited NATO from responding to security threats to its member states in the most efficient way possible in the 1990s. For example, had NATO responded in a more focused manner – along the lines of what a threat-based interpretation alone might have expected – its member states might have put more money into chronically under-funded long-range strategic lift accounts; it might have conducted a more efficient deployment and sustainment operation in Bosnia; it might not have collectively wasted precious defense budget resources on maintaining more forces in a higher state of readiness than its own strategy called for; and it might have been more effective in easing political, military, and economic instability in Albania in the mid-1990s. In order to understand alliance actions and their basis in the alliance’s doctrinal and structural response to the rise of new, post-Cold War threats, other explanatory tools beyond just the threat-based approach are required. Although necessary for part of the story, security threats are insufficient as a sole independent variable attempting to explain doctrine, structures, and the broader subject of alliance management. Political factors – expressed through member states interests and dependence on the alliance and exercised through intra-alliance bargaining – must be considered as well in order to understand both the process through which NATO changed its doctrine and structures and the actual doctrines and structures that resulted from this process. This book will show that both threat-based factors and political bargaining factors are necessary – and together sufficient. The next chapter will examine how NATO doctrine and force structures evolved from the end of the Cold War until the early twenty-first century. Then, Chapter 5 will examine this empirical information through the lenses of the theoretical frameworks outlined above. By examining in Chapter 4 the ‘how’ and ‘what’ of NATO’s move from strictly territorial defense to an alliance far more concerned with deployability, sustainability, crisis response, and out-of-area missions in addressing the post-Cold War world, this study will move on in Chapter 5 to shed light on the ‘why’ – the sources of doctrine.
Chapter 4
The Development of the NRDCs The Alliance Structure In the aftermath of World War II, Western European countries first became worried about the level of military power wielded by the Soviets in Eastern Europe in the late 1940s. The Soviet Union kept its troop levels at wartime strength following the defeat of Hitler’s Germany, while the Western Allies proceeded to demobilize.1 The only Western Ally in a position to provide military assistance – the United States – was unwilling to do so, choosing instead to focus on humanitarian and infrastructure assistance. For example, although the US-led Marshall Plan helped the Western European states to climb out of the economic devastation wrought by World War II, the Plan provided nothing in the way of increased military assistance. In the face of growing Soviet economic and political hegemony in Eastern and Central Europe, and in lieu of military assistance from Washington, Western European leaders turned to military consolidation of their own making as the best means to augment their security and protect the fragile economic recovery then underway. As a result, Britain, France, Belgium, the Netherlands, and Luxembourg formed the Western Union in early 1948. Shortly after its formation, the Western Union began to seek US involvement and support. American leaders expressed misgivings about military alliances in Western Europe because they believed such a move would encourage and require large military expenditures. A military build-up would siphon off scarce resources from economic recovery efforts. However, by mid-1948, the Soviet grip on Eastern Europe tightened, Berlin was under a Soviet blockade, and communists sought to make inroads in Greece, Turkey and elsewhere, and gradually, American leaders came to believe that only through a military alliance with US backing would Western Europe be guaranteed freedom from Soviet domination and coercion, allowing the post-war economic recovery effort to proceed unhindered. As a result, US leaders accepted the invitation to join in negotiations with West Europeans and Canada in December 1948 to create what would become the North Atlantic Treaty Organization. There were only limited references to alliance administrative and political structures in the 1949 Treaty of Washington. In fact, the only significant reference was to the establishment of what would become the North Atlantic Council (NAC), the body comprised of Heads of State and Government from each member state, but manned on a daily basis by Permanent Representatives at the alliance’s political 1 Total Soviet armed forces in Eastern Europe in 1948–49 numbered over 30 divisions, while combined US, British, and French forces amounted to no more than 10 divisions. See John Lewis Gaddis, Strategies of Containment (New York: Oxford University Press, 1982).
30
Alliance Management and Maintenance
headquarters in Brussels and which is the alliance’s highest decision-making authority.2 The Treaty authorized the NAC to establish subsidiary bodies, and in the months and years that followed the Treaty’s signing, the alliance grew in terms of organizational size and capability. Today, the alliance, a political-military entity separate from its member state representatives, is headed by a Secretary General and employs about 1,750 people at its headquarters on a fulltime basis. Additionally, the alliance’s secretariat has grown to include over a dozen major offices or divisions and an equal number of alliance policy committees, organizations, or agencies, plus many other bodies. One of the most important alliance bodies is the DPC. The DPC consists of Permanent Representatives, excluding France, and ostensibly deals with defense matters, though its purview has diminished and it now focuses primarily on force planning.3 Its primary task is to provide direction to the alliance’s military authorities. Within its areas of responsibility and competence, the DPC has the same authority as the NAC. The DPC and NAC rely on the Military Committee (MC) to help direct the alliance’s military authorities and thereby give effective direction to the integrated military structure. In permanent session, the MC is comprised of senior military representatives from the member states, but at least three times a year it meets at the level of the Chiefs of Defense (CHODs).4 The MC’s purpose is to provide advice on military policy and strategy to the alliance’s civilian political authorities, as well as to oversee the international military staff and the alliance’s strategic military commands.5 The MC is led by a Chairman, who is selected for a three-year term by the member state CHODs. The Chairman, traditionally a high-ranking non-American general or flag officer, and an ex-CHOD, acts exclusively in an international capacity, in contrast to the other members of the Military Committee, who act as national representatives of their respective Chief of Defense. Alliance decisions, whether they are made in the MC, the DPC, or the NAC, always reflect consensus among member states – if there is no consensus, no decisions are reached. In all committees, this may be achieved during an actual meeting or, if the matter does not require discussion, via a process known as the silence procedure, which functions as follows: The chairman of the committee or other body (for example, the Secretary General in the case of the NAC) presents issues – in the form of papers written by the staff in conjunction with member states – to the committee or body, along with a deadline. If no alliance member voices an objection before the deadline – and therefore ‘breaks silence’ – the paper and its 2 The NAC, chaired by the Secretary General, meets at least once per week, typically at NATO’s headquarters in Brussels. The NAC may also meet at the Ministerial or Head of State or Government level. 3 France withdrew from NATO’s integrated military structure in 1966 and is therefore not represented on the DPC. 4 The national military representatives that comprise the MC are typically senior general or flag officers; for instance, the US representative to the MC is a three-star general. 5 The alliance has two strategic military commands – Allied Command Operations (ACO) with headquarters in Mons, Belgium, and Allied Command Transformation (ACT) with headquarters in Norfolk, Virginia.
The Development of the NRDCs
31
contents, be they taskings, conclusions, or recommendations, are approved. This process facilitates consensus by forcing resolution of issues during the drafting phase rather than allowing a more time-consuming process of voting on individual issues and then compiling them into a single approved document. In contrast to the alliance’s success in facilitating relatively forthright decisionmaking through such tools as the silence procedure, one problem the alliance has had great difficulty in overcoming during its history is that of inequitable burden-sharing. Alliances between states have always faced this problem, and NATO is no exception – indeed, an imbalance exists within the alliance in terms of resources allocated, usability of forces, and capabilities.6 In the simplest terms, one of the primary ways the alliance deals with this is to establish guidelines for defense expenditures – the European allies are encouraged to spend at least two percent of their gross domestic product (GDP) each on defense. In practice, the majority of the European allies routinely fall below two percent year after year. Burden sharing is not just an issue within the alliance in terms of finances. The allies also face challenges in equitably sharing risk, manpower, and materiel costs; here too the United States typically carries the largest load, although there are exceptions, particularly with regard to some alliance operations.7 Like all international organizations, NATO exists because its members want it to. If all members agreed to disband the alliance, NATO’s roughly 1,750 employees would find themselves without jobs. Nonetheless, and like some other international organizations, the alliance occasionally displays the characteristics of a unitary, independent actor on the international stage. There are numerous manifestations of this, but three examples should suffice to provide an idea of how this occurs in practice. First, when the Secretary General speaks in public, which occurs frequently as noted on the alliance’s website, he does so as a representative of the alliance per se and the head of the international secretariat, not as a representative of any one member state. Second, the papers drafted by the international staff for member state deliberation often reflect positions that the international staff has crafted based on its own expertise and views and not necessarily in accordance with the positions of 6 Usability, in the NATO context, refers to the forces deployed or available for prompt deployments to an alliance or other operation. During the Riga summit in November 2006, the alliance agreed to establish a goal of having 40 percent of each member states’ land forces deployable, with 8 percent of each member states’ land forces available at any given moment for actual deployment and indefinite sustainment out-of-area. By way of comparison, the United States Army is roughly 70 percent deployable, and during peacetime roughly 13 percent of US Army forces can be deployed on short notice and sustained indefinitely. Many of the European allies are having great difficulty achieving the far lower NATO goals. All allies are required to report these figures on a regular basis to the alliance. With regard to capabilities, the trans-Atlantic capabilities gap is well documented. For example, see David Gompert, Richard Kugler, and Martin Libicki, Mind the Gap: Promoting a Transatlantic Revolution in Military Affairs (Washington: National Defense University Press, 1999). 7 Two of the most significant exceptions today include the International Security Assistance Force (ISAF) in Afghanistan and the NATO Response Force (NRF), both of which are manned predominantly by European allies, while the US provides most of the strategic enablers such as airlift.
32
Alliance Management and Maintenance
individual member states. Sometimes the result is that the international staff ‘gets in front of’ the member states on particular issues while that issue is being worked – that is, the staff may stake out a preliminary position that is farther reaching than most member states may be willing to go. This is never the case, however, after a decision by the alliance has been made on the issue in question. And in perhaps the most important example, the alliance has its own integrated military structure, which includes allied command structures and force structures, plus all of the strategies, doctrines, and plans that underpin each. NATO is unique in this sense – no other alliance in modern history has had such an integrated military structure. NATO’s command and force structures – such as the Rapid Deployment Corps discussed at the outset of this book – stem from its doctrine as noted in the preceding chapter and ultimately from its strategy, which is embodied in the alliance’s Strategic Concept. The Strategic Concept articulates the purpose and fundamental security tasks of the alliance, and is akin to the grand strategy or security strategy of individual states.8 It provides strategic perspective, addresses security challenges and risks, defines NATO’s approach to security, and outlines the principles of alliance strategy and overall guidelines for the alliance’s military forces. The importance of the Strategic Concept cannot be overstated – as the NATO Handbook notes, the Strategic Concept, “is the authoritative statement of the alliance’s objectives and provides the highest level guidance on the political and military means to be used in achieving them.”9 Following the end of the Cold War, NATO revised its Strategic Concept in 1991 and for the first time released it to the public.10 The 1991 strategy recognized that although the collective defense commitment embodied in Article 5 of the Treaty of Washington was still valid, the approach to security in Europe had to change because the nature of the threats facing the continent had changed. In addition to the military dimension of security, the 1991 Strategic Concept noted that security in Europe had political, economic, social, and environmental elements.11 In response, NATO was to prepare to participate in the full range of crisis resolution efforts, from political measures to military measures.12 Most importantly, the 1991 Strategic Concept stated that NATO faced new, diverse, and “multi-directional” threats, all of which would compel the alliance to move away from the concept of forward, territorial-based defense.13 NATO would respond to the new security environment by enacting “significant modifications in the missions of the Allies’ military forces and in their posture.”14 Of special note, the allies would develop a limited but “militarily significant” proportion of ground, air, and sea immediate and rapid 8 In the United States, NATO’s Strategic Concept would be similar to the National Security Strategy in terms of scope and purpose. 9 The NATO Handbook (2001), p. 43. 10 For an in-depth look at the process of how NATO developed its 1991 Strategic Concept, see Rob de Wijk, NATO on the Brink of the New Millennium: The Battle for Consensus (London: Brassey’s, 1997), pp. 20–47. 11 NATO Strategic Concept, 1991, paragraph 24. 12 Ibid., paragraph 32. 13 Ibid., paragraph 39–40. 14 Ibid., paragraph 39.
The Development of the NRDCs
33
15
reaction elements. Clearly, the 1991 Strategic Concept laid the groundwork for a significant revision to NATO’s doctrine and structures. Immediate Post-Cold War Changes Some changes in NATO military structures occurred immediately after the end of the Cold War – these were almost entirely the result of changes made by member states in their national force structures. The demise of the Soviet Union and the Warsaw Pact led several alliance members, including the United States, to decrease their defense budgets and the size of their armed forces and/or their forces based in Europe. For example, after real defense expenditure increases amounting to 10 percent between 1980 and 1990, the European allies cut their defense budgets by 12 percent in real terms between 1990 and 1994, while the United States cut its defense budget by 18 percent over the same period.16 With regard to military manpower, the European allies implemented average troop strength cuts of 15 percent between 1990 and 1993. So dramatic were the cuts that some feared a general unraveling of the integrated military regime altogether.17 On an individual basis, some allies cut their forces even more dramatically. For example, Germany cut its overall troop strength by 27 percent and Spain by 23 percent between 1990 and 1993. The United States cut the size of its active duty army forces from three quarters of a million soldiers to just under half a million. Allies that based soldiers in Europe dramatically cut forces there as well. The United States cut its European-based forces by two thirds to roughly 110,000. British forces deployed in Germany fell by more than 50 percent, and Canada removed all of its forces except for those attached to NATO Airborne Warning and Control System (AWACS) crews. Such dramatic cuts in member defense budgets and manpower forced NATO planners to make adjustments in alliance command and force structure and related planning. NATO created three new categories for all its designated forces: reaction forces (which were further broken down into immediate reaction forces and rapid reaction forces), main defense forces, and augmentation forces. Among the reaction forces, the alliance’s new Immediate Reaction Force (Land) (IRF(L)) would be the first to respond in the event of an Article 5 threat to an alliance member. However, the IRF(L) was actually just a new name for the Allied Command Europe (ACE) Mobile Force (Land) (AMF(L)), established in 1960 as a small multinational force, the core of which is a land force consisting of a brigade-sized formation of roughly 5,000 soldiers. The IRF(L), and before it the AMF(L), was designed for short-notice deployment to any part of Allied Command Europe to act, in part, as a sort of tripwire. The fact that 15 Ibid., paragraph 46. 16 “Financial and Economic Data Relating to NATO Defence: Defence Expenditures of NATO Countries (1975–1997),” NATO Press Release, M-DPC-2(97)147, 2 December 1997. 17 “Financial and Economic Data Relating to NATO Defence,” 2 December 1997; Martin Smith, NATO in the First Decade After the Cold War (Dordrecht, The Netherlands: Kluwer Academic Publishers, 2000), p. 66.
34
Alliance Management and Maintenance
14 NATO countries contributed soldiers to the IRF(L) was an embodiment of the phrase, “an armed attack against one or more of them in Europe or North America shall be considered an attack against them all” from the original Treaty of Washington – if the IRF(L)was engaged in hostilities, chances were that soldiers from most if not all of the contributing allies, and hence most of the alliance, would be bloodied, virtually compelling a broad-based response on the part of the alliance. The Allied Command Europe (ACE) Rapid Reaction Corps (ARRC) reinforced the IRF(L) and was the major land component of the rapid reaction forces. The ARRC was established in October 1992, and consists of a multinational staff of 300 personnel (representing almost all NATO countries) based in Rheindahlen, Germany.18 During peacetime, it has operational control over only one division – a combined division consisting of one brigade each from Britain, Belgium, Germany, and the Netherlands and known as the Multinational Division-Central (MND[C]). In the event of hostilities, the ARRC was designed to have the ability to call upon any of ten other divisions, depending on availability and mission type, that NATO member states have assigned to it to reach its full strength of four divisions. In what was heralded as probably the most important change of the post-Cold War force structure, NATO decided that the main defense forces would include multinational corps.19 Previously, in the event of war in Central Europe, NATO had planned to field the vast majority of its forces in that region in eight national corps, aligned within established corps boundaries.20 Although the corps would be integrated further up the chain of command, each national corps commander would retain command authority over training, logistics, task organization, and mission assignments for his corps. The troops of each corps would come from a single country, and each corps would consist of two or more divisions of 15–20,000 soldiers each. These national corps would operate in their own areas, delineated by corps boundaries. Under the force structure changes announced in the early 1990s however, the alliance’s main defense forces would consist of not just national formations but also multinational corps, such as the Danish-German corps, the German-Netherlands corps, and two German-US corps. Finally, the augmentation forces rounded out the alliance force structure. These forces consisted of other national formations typically held at much lower level of readiness and availability. Augmentation forces would be used to reinforce other NATO forces. Evidence of Doctrinal Change? Despite appearances, NATO command and force structure changes immediately following the end of the Cold War did not reflect changes in doctrine – the alliance had not fundamentally altered how it would fight or how it would apply military 18 The ARRC was the land component of NATO’s rapid reaction forces. Air and naval elements were also part of the rapid reaction forces. 19 The NATO Handbook (2001), p. 258; also, interview with a NATO official, 16 May 2003. 20 Young, Multinational Land Formations And NATO, p. 7.
The Development of the NRDCs
35
means to achieve broader security goals. Instead, the alliance command and force structure changes were driven by NATO planners trying to catch up with what were often unilateral, often uncoordinated, cuts in force structure implemented by almost all of the allies at the end of the Cold War. According to those involved in allied planning at the time, the immediate post-Cold War years were marked by a precipitous alliance reorganization to stay ahead of the demands for a peace dividend. As a former senior ARRC staff member noted, “NATO had to hastily do something.”21 However, instead of doctrinal change driven by the new post-Cold War threats outlined in the 1991 Strategic Concept, that ‘something’ largely entailed less than substantive efforts to keep up with member states that had unilaterally gutted their defense budgets and infrastructures. Throughout most NATO countries, citizens expected real decreases in defense spending as the Cold War wound down, and politicians felt compelled to meet these expectations. In a poll conducted in Western Europe in October 1991, barely a majority (56 percent) of respondents identified military defense as very important, while those same respondents all viewed unemployment (97 percent), pollution (97 percent), terrorism (94 percent), energy supplies (92 percent), the poor (89 percent), social justice (82 percent), and assistance for the developing world (80 percent) as more important issues of the day.22 European politicians got the message – between 1985 and 1995, 10 of the 14 European member states of NATO saw real reductions in defense spending.23 On the other side of the Atlantic, US troop cuts were motivated in part by a desire to achieve greater allied burden sharing. In February 1992, the House and Senate passed House Concurrent Resolution 275, which expressed the sense of the Congress that the United States should reduce its military expenditures and use the savings from the reduction to reinvest in American economic and human resources. Months later, Congress passed the US National Defense Authorization Act for 1992 and 1993, which called for the administration to cut American troop strength in Europe to 100,000 by 1995.24 With regard to the force structure that emerged on the basis of major cuts to defense budgets and troop levels, doctrine and coordinated planning trailed decisions already made by individual allies to cut defense budgets. As a former senior NATO force planner noted, “the NATO force planning process was crumbling apart. … The political masters in alliance member states were pulling the rug out from under
21 Interview with a British military officer assigned to NATO, 9 July 2003. Emphasis in the interviewee’s spoken comments. 22 For the impact of domestic public opinion and expectations on defense spending in the post-Cold War period, see Stanley Sloan, “NATO’s Future in a New Europe,” International Affairs, July 1990; Julian Lindley-French, “Leading Alone Or Acting Together? The Transatlantic Security Agenda For The Next US Presidency,” Institute for Security Studies of WEU, September 2000; and De Wijk, NATO on the Brink of the New Millennium, p. 20. 23 Exceptions were Luxembourg, Norway, Portugal, and Turkey. 24 Also, see remarks by Rep. David Bonior on 16 April 1991 expressing the need to trim forces and infrastructure in Europe and to get the allies to share more of the common defense burden.
Alliance Management and Maintenance
36 25
the military.” A US General Accounting Office (GAO) report found that as of May 1992, after alliance members had already begun to cut military budgets and forces and two years after the 1990 London summit when the alliance had announced its intention to move toward the new three-tiered force structure described earlier, NATO still “had not written mission statements for the various forces, determined precisely how these forces would be used in a conflict, fully revised its command structure, or defined new readiness requirements.”26 In other cases, the same report found that some members had substantially cut conscription terms beyond what NATO had intended.27 So, although the alliance had identified new threats in its 1991 Strategic Concept and elsewhere, NATO and alliance member restructuring efforts – if wholesale cuts can be labeled as such – were not a response to new, postSoviet threats. Similarly, the newfound preference on the part of several allies for multinational force structures did not initially reflect a change in doctrine either. Instead, the allies resorted to multinational corps largely for economic reasons. For example, allies had so deeply cut their individual force levels, through what one observer termed “competitive disarmament,” that fielding purely national corps became too difficult from a financial and human resources perspective. Indeed much of the shift toward multinational corps consisted of not much more than slapping together smaller military units to comprise a corps on paper.28 Efforts to create truly multinational formations had fallen short because the allies refused to yield command authority to commanders of multinational corps in peacetime, favored stationing subordinate units in their parent country rather than under the assigned multinational headquarters, and refused to provide the commanders of multinational corps sufficient authority during non-Article 5 missions.29 Where rigid, national corps boundary lines previously existed in NATO plans, national division boundary lines just as hard and fast had sprung up. The lack of real integration and doctrine on multilateral corps operations, reflected in challenges associated with interoperability, command authority, transfer of authority, and corps combat service support, would continue to plague the alliance throughout the 1990s and into the 21st century. One scholar practitioner compared this situation regarding multinational corps to baking a cake without mixing the ingredients beforehand.30 More generally, another observer – a retired US military officer responsible for handling issues on allied military doctrine 25 Interview with a retired US military officer involved in alliance force planning in the early and mid 1990s, 24 October 2003. 26 “NATO: A Changing Alliance Faces New Challenges,” United States General Accounting Office (GAO/NSLAD-92-262), July 1992, p. 6. 27 Ibid., p. 16. 28 In the hierarchy of western military organization, squads are the smallest element (9–10 soldiers), followed by platoons (16–44), companies (62–190), battalions (300–1,000), brigades (3–5,000), divisions (10–15,000), corps (20–45,000), armies (50,000+), and army groups (two or more armies). The number of soldiers noted in each element is an average – the specific number of troops in each depends on its type (such as infantry or aviation). 29 For greater detail on these points, see Raymond A. Millen, Tweaking NATO: The Case for Integrated Multinational Divisions (Carlisle, PA: The Strategic Studies Institute, 2002). 30 Millen, Tweaking NATO, p. vi.
The Development of the NRDCs
37
– noted that throughout the 1990s, alliance doctrine “continued to address just Article 5 operations, even though policy was focused on terror and other new threats.”31 As for the development of the ARRC, it was not initially the result of threat-driven change nor did it ultimately entail significant doctrinal innovation. Instead, it mostly came about as a result of coordination between then Supreme Allied Commander, Europe, US General John Galvin, and the British Chief of Defense Staff, as the only way to save what was left of the British Army of the Rhine (BAOR).32 In this way, the British used the ARRC as the justification for maintaining forces on the continent when political pressures in London were leaning toward further cuts of the British Army’s troop strength in Germany.33 From the alliance’s perspective, the BAOR was a logical choice to lead the ARRC because the British had been left out of the matchmaking that resulted in the multinational corps formation process, primarily due to equipment compatibility issues. Later, when alliance force planners made the rounds of member state capitals to solicit contributions to join the British-led ARRC, many allies jumped at the opportunity, perceiving there to be significant political prestige attached to involvement with the ARRC, interoperability challenges notwithstanding by that point. The problem was that the allies contributed forces other than what the force planners requested. “Less than a handful of the nine to 10 divisions proposed by member states were truly capable of rapid reaction,” according to one of the alliance’s senior force planners at the time.34 The majority of member state offers lacked associated combat support and combat service support and consisted of heavy forces, which ran counter to the very purpose of the ARRC as a deployable, rapid reaction force. Additionally, until 2002 the ARRC was NATO’s only corps-size rapid reaction force. The alliance had no other similarly structured or ready force, despite requirements outlined in ministerial guidance and requests from the Supreme Allied Commander in Europe (SACEUR).35 Additionally, there were some within NATO that pursued the development of the ARRC because they felt that if the alliance could just develop the capability, doctrine for its use would come later.36 This put a poorly constructed cart clearly ahead of the horse – again, as with the multinational corps, the alliance had not fundamentally altered how it would 31 Interview with a retired US military officer formerly responsible for issues regarding allied doctrine, 24 January 2003. 32 Interview with a NATO official, 16 May 2003; an interview with a senior retired US military officer involved in alliance planning, 24 October 2003. 33 The British Army faced significant cuts across the board, from a troop strength of 156,000 to 116,000. See Michael Evans, “Almost half of army to be in Nato’s rapid reaction force,” London Times, 24 July 1991. 34 Interview with a senior retired US military officer involved in alliance planning, 24 October 2003; and correspondence with a former senior Congressional staff member, 11 May 2004. 35 Interview with a senior retired US military officer involved in alliance planning, 24 October 2003; and an interview with a senior Turkish military officer assigned to NATO, 3 October 2003. 36 Interview with a German military officer assigned to NATO, 1 August 2003; and an interview with a senior retired US military officer involved in alliance planning, 24 October 2003.
38
Alliance Management and Maintenance
fulfill its mission via military means since the ARRC was a very small percentage of the overall alliance force structure and it lacked significant CS and CSS necessary for deployability, sustainability, and rapid response. Moreover, although most in the alliance recognized that future security threats were likely to come from outside what is known as the Central Region of Germany and the Benelux, most were unwilling or unable to agree on what “out-of-area” meant and how the alliance would operationalize its response to these new threats.37 Nonetheless, the alliance’s military and civilian authorities struggled to stay ahead of and draw into the alliance context the sweeping, mostly cost-driven changes taking place in force structures at the national level. As a result, in April 1991 the Military Committee endorsed the force structure outlined in the previous section – comprised of Reaction Forces, Main Defense Forces, and Augmentation Forces – which represented a post-Cold War transition only insofar as the allies had come to a consensus on avoiding wholesale liquidation of national force structures. In May 1991, the Defense Planning Committee gave its blessing. Later that same year, in November 1991, the alliance completed the process of revising its strategy into what would become the Strategic Concept discussed earlier. However, taking the next step to move beyond the stop-gap measures of the immediate post-Cold War period from 1989–91 and match alliance ways and means to the ends described in the 1991 Strategic Concept – and addressing legitimate concerns over the utility of the stop-gap measures in light of continually falling defense budgets – would have to wait. Toward Real Changes in Doctrine and Force Structure Bringing the alliance’s stated goals in line with the rapid pace of unilateral reductions and other decisions taken by member states outside the context of NATO practically compelled the alliance to put the horse back in front of the cart and move in the direction of revised doctrine – the alliance was in danger of losing the integrity that is critical to assuring friends and dissuading foes. In September 1994, the allied Chiefs of Defense approved basic guidance for the development of the Long-Term Study (LTS), an effort to reexamine the military interpretation – in terms of command and force structures – of the alliance’s 1991 Strategic Concept. The basic guidance, also known as the Terms of Reference, outlined the parameters for the study, which included work on refining doctrinal requirements necessary to fulfill the strategy and on elaborating specific command and force structures. Finally, NATO had begun formally the process of doctrinal change. The roots of the LTS actually extend back to slightly earlier alliance decisions and pronouncements. For example, following its May 1993 meeting in Brussels, the Defense Planning Committee noted in its communiqué that the alliance would adapt its forces by developing a more mobile and flexible force structure so that it could respond to a wide range of potential contingencies. It further stated that reaction 37 It took at least eight months for the alliance to even agree on what constituted the “Euro-Atlantic area.” Interview with a US military officer, 13 March 2003. See De Wijk, NATO on the Brink of the New Millennium for a discussion of the out-of-area debate.
The Development of the NRDCs
39
forces would play a central role in this new structure and that preliminary steps toward development of this structure were already underway.38 Similarly, during the October 1993 informal meeting of defense ministers in Travemünde, Germany, the United States formally proposed the Combined Joint Task Force (CJTF) concept in order to push the alliance further down the road toward out-of-area operations and toward expeditionary doctrine and structures.39 Additionally, the CJTF held out the potential of drawing in alliance partners such as Poland or Sweden by allowing non-NATO personnel to participate, provided they contributed meaningful forces to a particular operation.40 The CJTF concept also provided a way to operationalize the budding European security identity within the context of NATO.41 Just a few months later, in January 1994, the alliance heads of state and government formally endorsed further study of the CJTF concept, tasking the NAC and the MC to examine how NATO’s procedures and structures might be adapted.42 Although work on the LTS and the CJTF were not initially linked in a formal sense, a good amount of cross-fertilization occurred throughout the mid1990s and especially after the LTS began addressing command structure changes in the late 1990s.43 Finally, at a meeting in May 1994, the DPC reinforced sentiments expressed in the 1991 Strategic Concept by concluding that the security challenges facing the alliance had become more diverse and more complex than those of the Cold War. Correspondingly, the DPC noted that the alliance would require forces and structures that could respond effectively to a broader range of contingencies, from collective defense to peacekeeping.44 The LTS was developed under the auspices of NATO’s Military Committee and was designed to outline the specific modalities of how the integrated military command would conform to the new strategy. Work on the LTS occurred through the forum of the Military Transitional Issues Working Group (MTIWG), which reported to the MC. The MTIWG was comprised of military officers of Colonel or Captain (navy) rank from the Military Representative (MilRep) offices of each of
38 Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 26 May 1993. 39 Anthony Cragg, “The Combined Joint Task Force concept: a key component of the Alliance’s adaptation,” NATO Review (July 1996), pp. 7–10. Also, correspondence with a NATO official, 22 April 2004. 40 This same logic applied to the French, who continue to remain outside the integrated military command even though they retain alliance membership. Correspondence with a NATO official, 22 April 2004. 41 Interview with a former DoD official responsible for US policy toward NATO, 22 September 2004. 42 Declaration of the Heads of State and Government, Ministerial Meeting of the North Atlantic Council/North Atlantic Cooperation Council, Brussels, 10–11 January 1994. 43 Interview with a retired US military officer assigned to NATO at the time, 6 April 2004. 44 Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 24 May 1994.
40
Alliance Management and Maintenance
the member states, plus nonvoting participants of similar rank from the alliance’s strategic commands (Mons and Norfolk) and observers from the international staff. MC 400 – Setting the Stage The first part of the LTS involved a review of the specific guidance for the implementation of the 1991 Strategic Concept. The existing doctrinal guidance on implementation of the alliance’s broad strategy took the form of the Military Committee’s Directive for Military Implementation of the Alliance’s Strategic Concept, first issued as “MC 400” in December 1992. By 1994, given the dramatic changes taking place in geopolitics at the time, it was recognized as outdated primarily because it failed to go beyond consideration of Article 5 missions and did not address the out-of-area concept. Therefore, the first stage of the LTS was to inform the development of a revised MC 400 document. Despite the necessity of updating the alliance’s doctrine regarding the military modalities for meeting NATO’s political ends, early work on the LTS was slow going for four main reasons. First, in January 1993 a new American administration arrived in the White House and promptly focused on the domestic issues that had brought it to office. Between inheriting what would become the infamous US intervention in Somalia, handling the issue of gays in the military, trying to ensure Russia’s transition to capitalism and democracy, focusing its attention on Bosnia and alliance enlargement, and the short, one-year tenure of Secretary of Defense Les Aspin, the Clinton administration’s early handling of foreign affairs and military policy faced significant challenges.45 Coupled with the necessity of turning around an economy in recession and addressing other pressing domestic issues such as healthcare, reform of NATO simply did not command enough attention. Without prejudice to the other allies, the lack of intensive senior US engagement on a topic as politically charged as organizational reform of NATO was bound to frustrate progress. As one former US ambassador to NATO put it with regard to this subject, “Rarely do things happen in NATO without US leadership.”46 Second, NATO finally began to focus more keenly on the festering civil war in the former Yugoslavia in the early-to-mid 1990s. This war increasingly required significant alliance attention as NATO went from providing support to the UN Protection Force (UNPROFOR) in Bosnia to planning for and leading the follow-on peacekeeping force, known as the Implementation Force (IFOR), under the terms of the Dayton peace accord.47 In late 1995, following the December 14 signing of 45 Correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; correspondence with a former DoD official, 5 August 2004; and Ronald Asmus, Opening NATO’s Door: How the Alliance Remade Itself for a New Era (New York: Columbia University Press, 2002), pp. 18–29. 46 Interview with a former US ambassador to NATO, 21 March 2003. 47 Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 29 November 1995. During this ministerial meeting, NATO defense ministers received updates from SACEUR on planning for IFOR, which had come to occupy a significant amount of attention from the NATO secretariat and integrated military command.
The Development of the NRDCs
41
the Dayton Accord, the alliance began deployment of the first enabling elements of IFOR. The assumption of responsibility for peacekeeping in Bosnia required significant alliance attention and focus for several reasons: •
• • •
It marked a political and cultural watershed as the first major operational deployment of land forces under NATO command beyond the territory of its member states. It required NATO to exercise out-of-area deployment and sustainment arrangements that it had never done in a real operation. Coordination was greatly complicated by the participation of numerous nonNATO countries, each with their own tactics, techniques, and procedures. IFOR marked NATO’s first foray into peace support operations.
Third, NATO was busily preparing to develop a program of engagement for the former members of the Warsaw Pact, some of which had begun to express interest in closer relations with western institutions. For NATO, this engagement primarily and initially took the form of the North Atlantic Cooperation Council (NACC), a consultative body consisting of all NATO members plus most of the countries of Central and Eastern Europe and the former Soviet Union. When the NACC proved to be not much more than a talking shop, NATO’s new partners in the east pushed for closer cooperation. The result was what would eventually become the Partnership for Peace (PfP) – a program, not a doctrine or strategy, that sought to promote military reform and mil-to-mil ties through the development of peacekeeping skills. In January 1994, NATO Heads of State and Government issued an invitation for non-member states to participate in the new Partnership for Peace, noting that participation would play a key role in the evolutionary process of alliance expansion.48 The alliance spent most of 1995 engaged in a process of studying the prospects for enlargement. The internal study was designed to determine how NATO would enlarge, the principles to guide the process, and the implications of expanded membership.49 Finally, the role of the French in developing the LTS caused some delay. Although development of the Strategic Concept had included France, the initial work on the LTS occurred under the auspices of the integrated military structure of which France was not a part. Nonetheless, those developing the LTS realized the importance of French participation, particularly since the alliance had already agreed that peacekeeping operations would include all member states.50 The decision on how to facilitate active French participation was placed on the NAC agenda in early 1995, where it remained without any action until 1996. This inaction helped to prevent the NAC from providing guidance to the MTIWG on the LTS’s development. As a result of deliberations over how to take French views into consideration, the NATO enlargement process, assumption of peacekeeping responsibility in Bosnia, and an American administration with its attention focused elsewhere, work
48 “Partnership for Peace: Invitation,” issued by the Heads of State and Government participating in the Meeting of the North Atlantic Council, Brussels, 10–11 January 1994. 49 Final Communiqué, North Atlantic Council in Ministerial Session, 30 May 1995. 50 France took part in the MTIWG’s proceedings as an observer.
42
Alliance Management and Maintenance
on revised alliance doctrine and structures proceeded slowly and deliberately, even though ministerial communiqués during this period routinely hailed the ongoing progress toward more flexible structures and forces capable of responding to the post-Cold War security environment.51 Despite the fact that the Military Committee had largely completed work on revising the outmoded doctrine contained in MC 400 by November 1995, it was not until June 1996 that NATO’s political authorities approved its revision as MC 400/1. The primary difference between MC 400 and 400/1 was that the former concentrated on mission elements for Article 5 collective defense, while the latter included doctrine addressing a broader array of missions including peace support operations, crisis management, and regional collective defense.52 Like its predecessor document, MC 400/1 was designed to translate the strategic guidance into something that could be implemented militarily. Specifically, MC 400/1 outlined the missions of alliance military forces, the operational capabilities necessary to accomplish these missions, and the principles behind alliance military structures.53 NATO had finally taken a decisive step toward changing how it fought, but the journey was not over yet because MC 400/1 only formed the first part of the doctrinal puzzle that would turn the alliance from a Cold War institution into one capable of responding to the numerous post-Soviet threats identified in the 1991 Strategic Concept. The Command Structure As the LTS moved toward its second phase – focusing on the command structure – the alliance began to realize pressure to speed up the pace of the MTIWG’s work prior to the accession of the Czech Republic, Hungary, and Poland that would occur in early 1999. Most allies recognized that a revised command structure would necessarily be a smaller one and would mean cuts in not just the number of commands but also the number of general and flag officer billets available among the NATO commands. Delaying completion of the command structure phase until after the three new members acceded would have only made more challenging the intra-alliance negotiations over what commands went where and who received the general and flag officer assignments at those commands.54 As one US military officer 51 See for example, paragraph 3 of the Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 24 May 1994; paragraphs 2 and 9 of the Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 30 May 1995; and paragraphs 14 and 17 of the Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 29 November 1995. These are emblematic of final communiqués issued throughout 1993, 1994, and 1995. 52 De Wijk, NATO on the Brink of the New Millennium, pp. 101–6. 53 Interview with a US defense official involved in NATO policy planning, 25 January 2003. 54 Interview with a US military officer assigned to NATO during this time period, 13 March 2003; interview with a retired US military officer assigned to NATO during this time period, 15 March 2004.
The Development of the NRDCs
43
involved in the command structure debate noted, “the three new members would have wanted a headquarters on their territory or one of their general officers in an important position,” and existing members of the alliance wanted to retain the plum assignments for themselves.55 Nevertheless, realigning the command structure took nearly two years of study, deliberation, and negotiation on the part of the MTIWG. This was primarily the case because the member states participating in the negotiations were deliberating over not just doctrine but tangible expressions of alliance support and commitment – the location of command structures.56 As one US military officer involved in the deliberations noted, “Things always got emotional whenever the topic was the location of command elements.”57 Another participant in the negotiations stated that, “convincing a country to give up something took lots of time and lots of hard talk.”58 Member states were all keenly interested in keeping NATO commands on their territory for reasons both intangible – a NATO command brought political prestige – and tangible – most members states would collocate their national military command structures with those of the NATO facilities in their country, thereby getting NATO to pay for at least part of the facility.59 Additionally, traditional rivalries, such as between Greece and Turkey or Spain and Portugal, prevented faster conclusion of an agreement among the member states. For example, SACLANT60 favored having all of the North Atlantic under its area of responsibility (AOR) for reasons of operational expediency. However, Spanish officials in Madrid did not want Portugal, which played a key role in SACLANT and was home to one of the subordinate SACLANT commands, to have control over any sea space around Spain’s Canary Islands.61 This problem was only resolved through an agreement to create a 50-mile buffer zone around
55 Interview with a US military officer assigned to NATO during this time period, 13 March 2003. 56 Interview with a retired US military officer assigned to NATO during this time period, 6 April 2004; interview with a retired US military officer assigned to NATO during this time period, 15 March 2004. 57 Interview with a retired US military officer assigned to NATO during this time period, 6 April 2004. 58 Interview with a retired British military officer assigned to NATO, 11 June 2004. 59 Interview with a retired US military officer assigned to NATO during this time period, 15 March 2004. Admittedly, the alliance funds only those facility elements that NATO will use. Nonetheless, this has historically led to a bias within alliance members toward, for example, the construction of large main operating bases instead of investments in deployable combat, CS, and CSS capabilities. See M.J. Cunningham, “The Main Operating Base: Castle or Coffin?” RUSI Journal (December 1987), pp. 35–36, as cited in Thies, Friendly Rivals, pp. 11–12. 60 Supreme Allied Command Atlantic (SACLANT) is known today as Allied Command Transformation (ACT) and no longer retains responsibility for any geographic space. 61 Interview with a retired US military officer assigned to NATO during this time period, 15 March 2004; interview with a retired US military officer assigned to NATO during this time period, 27 April 2004.
44
Alliance Management and Maintenance
the Canaries that was part of SHAPE’s AOR; all of the sea space around this buffer zone was part of SACLANT’s AOR. In 1997, NATO’s Military Committee submitted the revised command structure doctrine and plan – known as MC 324 – to the alliance’s Defense Ministers for approval. The new NATO command structure cut headquarters elements from 65 to about 20, and was based on a series of Joint Sub-Regional Commands (JSRCs) and Component Commands (CCs) geographically spread throughout the alliance. As with past NATO command structures, none of the JSRCs or CCs had assigned forces – that is, there were no permanently assigned military forces for any of the 11 JSRCs and CCs.62 NATO would continue to have no land forces of its own, aside from those committed to it by member states in times of conflict. The integrated military command of the alliance continued to be a military entity consisting of headquarters or command elements with little to command during peacetime. Despite the fact that the command structure revision was presented for approval in 1997, it would take another two years to complete implementation.63 Much of the delay during this period stemmed from deliberations over the socalled flags-to-post issue – the allocation of general and flag officers to NATO commands – which was always a contentious issue within NATO.64 Additionally, the conflict in Kosovo from late 1998 through the middle of 1999 played a part in the delay insofar as it demanded the time and attention of alliance personnel who would have otherwise been focused on wrapping up work on the command structure and related doctrine.65 The Kosovo conflict and on-going negotiations on the flags-to-post and other issues regarding the final form of the command structure necessarily postponed addressing the long overdue force structure changes through the LTS.
62 Exceptions to this include part of the NATO integrated air defense structure, some communications units, the Standing Naval Forces, and the alliance’s rapid reaction corps. But, the crews that staff these units are provided by member states, and theoretically those contributing countries could decide during a crisis to withdraw their personnel. 63 Implementation of the command structure entered its final phase in September 1999. For a description of some of the challenges facing NATO in reforming its command structure, see Thomas-Durell Young, Reforming NATO’s Military Structures: The Long-Term Study and its Implications for Land Forces (Carlisle, PA: Strategic Studies Institute, 1998). 64 Interview with a retired US military officer assigned to NATO during this time period, 15 March 2004; interview with a retired US military officer assigned to NATO during this time period, 6 April 2004; and interview with a retired US military officer assigned to NATO during this time period, 27 April 2004. For a discussion of one of the most contentious aspects of the flags-to-post issue – whether a European would take over AFSOUTH in Naples – see De Wijk, NATO on the Brink of the New Millennium, pp. 135–9. 65 Because of limited personnel and other resources, the Kosovo crisis caused personnel to be taken off of assignments dealing with doctrine and command and force structure issues and redirected to work issues regarding Kosovo. Simultaneously, the attention of senior leaders within the alliance also turned to issues regarding Kosovo and away from seemingly less important tasks such as doctrinal revision. Interview with a retired US military officer assigned to NATO during this time period, 27 April 2004.
The Development of the NRDCs
45
The Force Structure The third and final stage of the LTS – and the final step in NATO’s effort to truly change how it would fight in the post-Cold War environment – was the force structure review. NATO force structure was, and for the most part still is, based on the military units identified by member states that, while under full national control during peacetime, would be assigned to the alliance in wartime. For the duration of hostilities, those national forces would fall under the command and control of a NATO commander, such as from one of the JSRCs or CCs. Previously, most of those forces consisted of so-called heavy or armored units primarily devoted to territorial defense. Such units were characterized by their large, heavy equipment and, in the case of most of the European allies, their lack of deployable combat support and combat service support capabilities. The lack of deployable CS and CSS capabilities on the part of the European allies was largely a result of the Cold War experience. Many of the allied armies – particularly the American, Belgian, British, Canadian, Dutch, French, and German – expected to fight on the plains of Central Europe, in the so-called Central Region of Germany and the Benelux, and so most of the Cold War-era European allies lacked any historical imperative to develop significant deployment capabilities such as strategic lift because they did not calculate that they might have to conduct land operations very far from home.66 In fact, Cold War-era doctrine in West Germany called for deriving logistical support from the regions where the Bundeswehr soldiers were based – they were expected to live off the land.67 As NATO planners thought through the contingencies that the alliance might be required to respond to, it was not until 1997–98 that ministerial guidance began to call for doctrine to outline force structure requirements that would permit three corps-size operations at once – one within NATO territory, one adjacent to it, and one farther out-of-area.68 The shift in ministerial guidance was in part the result of the extensive demands made on alliance military capabilities – especially in the Balkans – through the mid to late 1990s.69 In practice, this would mean NATO could handle an on-going IFOR/SFOR operation, an emerging KFOR operation, and a main defense reinforcement capability if, for example, a recidivist Russia acted against Norway or Poland, or Syria or Iraq acted against Turkey. Military planners within NATO realized they needed to be able to meet all of these potential requirements with forces that would necessarily be 66 Interview with a NATO official involved in force planning, 16 May 2003; interview with a German military officer assigned to NATO, 1 August 2003; correspondence with a Dutch military officer assigned to NATO, 4 May 2004. 67 Interview with a Dutch military officer assigned to NATO, 1 August 2003. 68 Interview with a US defense official involved in policy planning, 25 January 2003, and interview with a US military officer assigned to NATO, 7 February 2003. Also see the Final Communiqué of the Ministerial Meetings of the Defense Planning Committee and the Nuclear Planning Group, 12 June 1997; and the Final Communiqué of the Ministerial Meeting of the Defense Planning Committee and the Nuclear Planning Group, 2 December 1997. 69 The NATO Handbook (1998), p. 245.
46
Alliance Management and Maintenance
multinational because of still low European defense budgets, could deploy very quickly, could be supported and sustained far from home bases, and that trained together regularly.70 NATO’s experience in the Balkans raised concerns about the alliance’s ability to meet the defense needs of its members in the event of an Article 5 crisis. For example, if the ARRC was deployed to Bosnia to replace UNPROFOR, as actually happened, NATO then lacked one of its most important, most capable reinforcing elements in the event of an Article 5 crisis.71 NATO would therefore need at least one other ARRC-like, corps-size entity – otherwise, the alliance would rely on a reinforcement capability for an Article 5 crisis comprised of ad hoc forces that may have been at a lower level of readiness and that had likely never worked or trained together, obstacles that although not insurmountable were nevertheless certainly not something to be ignored or easily dismissed. Such a situation would have been fraught with even more risk if the force provider of last resort – the United States – had deployed its Germany-based V Corps to the Middle East or Northeast Asia, for example.
In order to accomplish three corps-size rapid response operations at once or nearly simultaneously, NATO required several more ARRC-like entities that would come to be called the NATO Rapid Deployment Corps (NRDCs). Three of the NRDCs would need to be in a higher state of readiness – High Readiness Forces (HRFs) – in order to meet alliance requirements for quick deployment
70 Interview with a US defense official involved in policy planning, 25 January 2003. 71 The NATO-led operation that replaced UNPROFOR was known as IFOR, or Implementation Force.
The Development of the NRDCs
47 72
to as many as three near simultaneous Article 5 or non-Article 5 crises. Two Forces at Lower Readiness (FLRs) would back up each of the three HRF(L)s – for a total of six FLRs – providing an A-B-C rotation scheme for each HRF(L), working as follows. First, an HRF(L) would deploy to the first crisis within 90 days for a period up to 180 days. During that time, the first backup FLR would prepare for deployment and would relieve the HRF(L) after the first 180 days. During the next 180 days, while the first FLR was deployed, the second FLR would prepare for deployment, ultimately relieving the first FLR at the end of its rotation. After another 180 days, the initial HRF(L) would then relieve the second FLR and the cycle would start anew for as long as the situation warranted. In this manner, there would always be one force on deployment, one force preparing for deployment, and one force at rest. The other two HRF(L)s and four FLRs would each follow the same rotation scheme for their respective crises. Further guidance to alliance military authorities on this subject was contained in two documents: MC 400/2 and MC 317/1. Both of these documents were initially the product of a working group that became the successor to the MTIWG. By the summer of 1999, issues regarding reform of NATO’s doctrine and its command and force structures were no longer deemed ‘transitional,’ and so the MTIWG was renamed the Military Committee Working Group – Strategic Issues (MCWG-SI). As with the MTIWG, the MCWG-SI was comprised of Colonel and Captain (navy) rank personnel from each of the MilRep offices and nonvoting participants from the strategic commands, plus observers from the international staff. The IMS staff members of the MCWG-SI would typically take the lead in drafting papers for the working group to consider. In standard practice for NATO (as well as for other types of international organizations), representatives would convey these working papers to their capitals and subsequently deliver their country’s response in a formal comment during one of the working group’s meetings.73 These sorts of formal presentations would comprise most of the working group’s meeting time. Serious give-and-take negotiation and compromise on the force structure only took place outside of the formal meetings, such as during the coffee breaks or well after the meetings adjourned. As one participant in the deliberations noted, “the reps (sic) all had instructions from their capitals, and they’d relay these dutifully during working group sessions. Any real progress happened on the margins, where we found room for negotiation.”74 Aside from the often stilted, pro forma working group sessions, another potential challenge to efficient completion of the work on the force structure lay in the tacit rivalry between the strategic NATO commands in Mons and Norfolk 72 There are maritime as well as land HRFs. The abbreviation “HRF(L)” denotes a land HRF. 73 Even though these sessions often took on a pro forma air, they were not necessarily without lighter moments. For example, following the comments of the Turkish representative to a particular proposal, the Greek representative announced that although he had yet to receive official guidance from his capital, he wanted to nonetheless disagree with everything the Turkish representative had just said. 74 Interview with a retired US military officer assigned to NATO during this time period, 6 April 2004.
48
Alliance Management and Maintenance
on the one hand and the international military staff on the other, the former often viewing the latter as interlopers when it came to military planning. “The strategic commanders weren’t always big fans of the international military staff,” said one participant in the deliberations. For example, he continued, “SACEUR thought the real military planning should be done by SHAPE, not the IMS.”75 Despite these potential challenges, work on the force structure moved forward in the form of MC 400/2 and MC 317/1. Two reasons why include the nature of the working group and some of the forceful personalities involved. The MCWGSI was one of the most productive and effective working groups within the Military Committee structure primarily because it attracted the cream of the crop from each of the MilRep offices, given the gravity of the subject it was tasked to address. “It was so intellectually demanding,” said one American participant.76 A British participant noted that the MCWG-SI was unusual in that, “We worked 10-to-12-hour days, working late into the night to prepare documents for the next day….Those expecting a cushy NATO staff assignment didn’t find it on this working group.”77 Additionally, the Chairman of the Military Committee, General Klaus Naumann of the German Armed Forces, played a major role in pushing through solutions to seemingly intractable problems surrounding force structure and residual command structure issues.78 On some of the most contentious questions regarding the awarding of flag officer billets to particular countries, Naumann pushed through solutions that the parties themselves might never have arrived at or might only have achieved after many more months of deliberation. According to one US participant in the deliberations, Naumann “relentlessly drove the train through all these potential stumbling blocks.”79 The process was also helped by the very strong IMS team leading the staffing process at the time and the influence of SACEUR General John Shalikashvili.80 MC 400/2, an update to MC 400/1, was meant to incorporate pending changes in NATO’s Strategic Concept (which was updated in 1999) and to outline broader concepts on how the Military Committee, Strategic Commanders, and NATO member states were to develop and revise concepts, directives, plans, structures, and procedures. Perhaps most importantly, it outlined specific essential operational capabilities (EOCs) that would ultimately point NATO in the direction of developing a force structure capable of both Article 5 and nonArticle 5 operations – indeed, the same EOCs were required for both types of operations, indicating that the distinction between Article 5 and non-Article 5
75 Interview with a retired US military officer assigned to NATO during this time period, 27 April 2004. 76 Ibid. 77 Interview with a retired British military officer assigned to NATO, 11 June 2004. 78 Interviews with retired US military officers assigned to NATO during this time period, 15 March 2004 and 27 April 2004. 79 Interview with a retired US military officer assigned to NATO during this time period, 27 April 2004. 80 Correspondence with a NATO staff member involved in policy planning, 20 August 2004; correspondence with a former DoD official, 5 August 2004.
The Development of the NRDCs
49
81
had begun to fade away. MC 400/2 was approved in February 2000, and it compelled the alliance to subsequently revise other doctrine regarding force structure. Building on the doctrinal guidance in MC 400/2, the NAC approved MC 317/1, an update to MC 317, in July 2002. The revised MC 317/1 provided detailed doctrinal guidance on the structure of allied forces by formally doing away with the old force structure consisting of Reaction Forces, Main Defense Forces, and Augmentation Forces and establishing in their place the requirements for the HRF(L), FLR, and Long-Term build-up forces. Although implementation had yet to begin, this represented the first significant, meaningful shift in alliance doctrine on force structure since the end of the Cold War, a shift that fundamentally changed how the alliance would fight in a way that was far more profound than the comparatively superficial changes of the early 1990s. As NATO arrived at agreement on these new force categories and related readiness requirements, the Military Committee also finalized its work on precisely how it would measure the fitness of potential HRF(L) candidates. These two efforts – coming up with the criteria for assessing the NRDCs and development of MC 317/1 – occurred in parallel following a twin-track approach adopted by the Chiefs of Defense in 2000. Despite the seeming illogic of drafting criteria for doctrine that itself was still under development, the CHODs adopted this approach in order to speed up the process of finalizing the force structure.82 NATO leaders wanted to avoid the kind of contentious, drawn-out negotiations on the force structure that had bogged down the alliance during deliberations over the command structure. Given their experience in negotiating the command structure, alliance leaders feared that achieving consensus on MC 317/1 and the NRDC assessment criteria could prove just as contentious, even prohibitively so. The MC, according to one American participant in the deliberations, “had already recognized that the process was taking too long, and to work these issues sequentially meant only more opportunities for delay.”83 But, they believed that at least one of the documents – MC 317/1 or the criteria – would make it through the negotiating process, and they assumed that as long as one of these made it, there was a chance that the NRDCs would become a reality in the near future. Eventually, SHAPE crafted a list of 65 criteria for assessing NRDC candidates. Several alliance members – more than the three initially required – nominated themselves as not just NRDC candidates, but NRDC candidates of the highest readiness level – the HRF(L)s. This was surprising because at first glance there were significant disincentives for alliance members to have HRF(L)s on their territory. For example, as noted earlier, HRF(L)s are not merely headquarters elements – they 81 Interview with a German military officer assigned to NATO, 1 August 2003; also, correspondence with a German military officer assigned to NATO, 22 April 2004. 82 Correspondence with a Germany military official assigned to NATO, 22 April 2004 and interview with a retired US military officer assigned to NATO during this time period, 27 April 2004. 83 Interview with a retired US military officer assigned to NATO during this time period, 27 April 2004.
50
Alliance Management and Maintenance
require participating states to ensure that they can provide the necessary military forces with deployable combat support and combat service support elements on a fulltime basis, necessitating a substantial initial financial investment. Additionally, the HRF(L) readiness requirements – deploying within 90 days – are expensive to maintain continuously. Both of these elements – deployable CS and CSS and heightened readiness – would be challenging for European allies to achieve, particularly when defense spending is falling, not rising, throughout Europe. Nonetheless, alliance members perceived there to be significant benefits in gaining the HRF(L) designation in comparison to that of FLR. Indeed, as one observer put it, member states were “beating down the door” to get the HRF designation.84 There were a variety of reasons for this. First, countries hosting an HRF(L) receive NATO exercise funding and access to NATO’s training and evaluation infrastructure – as mentioned, in an era of shrinking defense budgets, any additional outside funding for training and readiness is vital.85 Second, alliance members continued to perceive the existence of political prestige – not to mention the tangible benefits of NATO-funded construction – in having alliance infrastructure located on their territory.86 With the reduction in command elements as part of the revised command structure, there were fewer infrastructure pieces to go around. As one US government official noted, “the European allies know that if they have an HRF, they have credibility.”87 Third, the military organizations of many allies use NATO-generated requirements to justify their own national force structures and budget requests and therefore saw the HRF as a budget justifier.88 As one Polish officer at NATO, speaking of Polish forces but indicative of the situation with regard to many other allies, put it, “[our] top-heavy command structure needed to be utilized, otherwise they might have to be reduced.”89 Throughout the 1990s, as NATO placed less and less emphasis on requirements for large forces capable of territorial defense, several allied military officials, such as those of the Royal Netherlands army, began to perceive potentially devastating cuts heading their way in terms of manpower, budget resources, and political prestige. In order to counter this, many of those military organizations pushed strenuously for their country to contribute an HRF(L). Fourth, alliance members attached a negative connotation to the phrase ‘Forces at Lower Readiness,’ and therefore preferred having HRF(L)s because they seemed sexier. In the words of one Dutch officer at NATO, the European allies preferred 84 Interview with a US government official involved in policy planning, 25 January 2003. 85 Interview with a French military officer assigned to NATO, 3 October 2003; interview with a Polish military officer assigned to NATO, 3 October 2003. 86 Interview with a Dutch military officer assigned to NATO, 1 August 2003; interview with an Italian military officer assigned to NATO, 1 August 2003. 87 Interview with a US government official responsible for military planning, 19 February 2003. 88 Interview with a German military officer assigned to NATO, 1 August 2003; interview with a British military officer assigned to NATO, 10 July 2003; and an interview with a Dutch military officer assigned to NATO, 1 August 2003. 89 Interview with a Polish military officer assigned to NATO, 3 October 2003.
The Development of the NRDCs
51
being in the “major leagues” through having an HRF(L), rather than being relegated to the minor leagues of the FLR.90 This view was so widespread that NATO would later avoid referring to “FLRs” and would instead favor the more neutral “Graduated Readiness Forces” when referring to any of the NRDCs. Despite these potential benefits, after alliance members learned more regarding the stringent criteria for HRF(L) designation, one of the initial nine candidates dropped out – Greece – in favor of becoming a FLR. A second candidate, the United States, also dropped out for a variety of reasons, including the desire to promote the development of increased capabilities among the European allies.91 By 2002, the alliance had six clear HRF(L) candidates: 1. The Allied Command Europe (ACE) Rapid Reaction Corps (ARRC) in Rheindahlen, Germany; 2. The Eurocorps in Strasbourg, France (NRDC-EC); 3. The German-Netherlands Corps in Münster, Germany (NRDC-GNL); 4. III Corps Turkish Army in Istanbul, Turkey (NRDC-T); 5. A Spanish corps based in Valencia (NRDC-SP); and, 6. An Italian corps based in Milan (NRDC-IT). Each of the six candidates began the thorough HRF(L) certification process in 2002. An alliance entity known as the Deployable Headquarters Task Force (DHQTF) conducted the certifications. Led by officials from SHAPE, the DHQTF focused primarily on determining if the HRF candidates had the appropriate equipment, personnel, units, documentation, and procedures. They did not assess military proficiency through the observation of military exercises, for example. The DHQTF evaluations were very formal and were designed to be transparent, equitable, and strictly objective. Reports of DHQTF assessments were submitted to both the relevant headquarters commander and SACEUR. Subsequently, SACEUR reviewed the results and relayed them, and his recommendation on whether to endorse the candidate as an HRF(L), to NATO ambassadors. By late 2002, each of the six candidates had gone through the certification process and had received the HRF(L) designation for initial operating capability (IOC). By the end of 2003, having gone through another assessment, each achieved certification of full operational capability (FOC). The review process and the criteria for HRF(L) certification were intended to be rigorous, but some within NATO have questioned whether each of the HRFs are at the same level of combat service and combat service support capability. According to one military official at NATO, the issue now confronting the alliance is, “the robustness of the [HRF(L)] headquarters – can they really deploy and sustain themselves?”92 Another member of the NATO 90 Interview with a Dutch military officer assigned to NATO, 1 August 2003. 91 Interview with a US military officer assigned to the European Command, 14 November 2002; and an interview with a former DoD official responsible for NATO policy, 22 September 2004. The reason cited here and other reasons for the US withdrawal are outlined in detail in the next chapter. 92 Interview with a NATO military official involved in policy and force planning, 11 June 2004.
52
Alliance Management and Maintenance
international staff put it more bluntly: “Some of us think [the HRF(L) host nations] are playing shell games – they don’t really have everything they say they do.”93 As NATO has inherited greater responsibility in Afghanistan, and as the HRF(L)s have begun to deploy there, these concerns have turned out to be accurate.94 In the case of the HRF(L)s that are sourced primarily from a single country, such as NRDC-T or NRDC-IT, shortfalls in subordinate combat units or shortfalls in equipment necessary to outfit the corps and its combat units can usually be made up by drawing on the larger national base of the respective lead nation’s military forces. For example, if NRDC-T needed communications equipment, it would first seek to tap into whatever was available from elsewhere within the Turkish military. The countries that host the HRF(L)s, such as Turkey, do this because none of them want to be embarrassed by fielding an under-manned or under-equipped force. However, the headquarters elements of the HRF(L)s that command the subordinate combat units are routinely multinational and hence comprised of personnel drawn from many allies, and NATO countries have become notorious for not filling their allotted positions in the HRF(L) headquarters.95 When NRDC-IT based in Milan, prepared to take over command of ISAF from NRDC-T in 2005, it conducted a mission rehearsal exercise (MRE) prior to the official transfer of authority. An MRE is a standard event among NATO militaries – it allows a military force to rehearse scenarios it is likely to encounter during a military operation. The MRE is no mere administrative requirement – it and the other preparatory exercises conducted prior to deployment are absolutely critical to ensuring the proper functioning of headquarters and subordinate units, and Western militaries go to great lengths to develop meaningful and realistic rehearsal exercises. Unfortunately, because of the aforementioned shortfalls in headquarters staffing, NRDC-IT had less than 50 percent of its required manpower during its MRE.96 Obviously, with less than half the command and control team in place, this would negatively impact its ability to function properly on the ground. In addition to gaps in staffing, shortfalls also occasionally exist in terms of necessary equipment, despite efforts to tap into national force pools. In no other area has this been more evident than in rotary wing assets. In March 2005, as a Dutch attack helicopter unit ended its tour of duty and redeployed from Afghanistan to the Netherlands, it took its AH-64 Apache helicopters with it. The Dutch contribution – the only attack helicopters assigned to ISAF at the time – amounted to less than 50 percent of the ISAF combined joint statement of requirements (CJSOR), but NRDC-T that took over leadership of ISAF in February 2005 lacked the necessary assets to backfill even this limited Dutch contribution. Unable to draw attack helicopter assets from other Turkish national resources, NRDC-T referred the issue 93 Interview with a NATO official involved in military planning, 31 March 2004. 94 Interview with a US military officer assigned to ISAF, 4 October 2006. 95 Additionally, because the corps tend to be land-centric, they typically lack the ability to conduct joint operations, as their manning documents are not structured for ‘jointness.’ 96 Each HRF(L) has a ‘peacetime establishment’ manning level and a ‘crisis establishment’ manning level. The former is what is necessary when the unit is in garrison, the latter when the unit goes to war. The crisis establishment for the Italian HRF(L) was roughly 600 personnel. For the ISAF VIII MRE, the Italian HRF(L) was staffed at only 45–50 percent of this 600– person requirement.
The Development of the NRDCs
53
to its higher NATO headquarters in Brunnsum, which in turn submitted it to SHAPE for force generation. For four months out of its six-month rotation, ISAF VII, led by NRDC-T, lacked rotary wing attack aviation in the Kabul area. Later that year, when NRDC-IT took over from the Turks in August 2005 as part of ISAF VIII, again an attack helicopter shortfall existed. Only with the added pressure of the Afghan elections occurring on their watch was NRDC-IT able to tap into Italian national assets to close some of the attack helicopter gap. Despite shortcomings in implementation, and even though the result of the alliance’s force structure revamp – six HRF(L)s vice three – did not correspond to NATO’s own requirements, by late 2003 the alliance had achieved, in part, what its own strategies had called for in the early 1990s – the development of doctrine and some of the structures necessary to address a changed security environment and the capability to fight in a different way.97
97 Of course, the missing parts remain the deployment and sustainment capabilities.
This page intentionally left blank
Chapter 5
Changing Threats and the Alliance’s Response Threats alone cannot explain NATO’s post-Cold War behavior with regard to doctrine and structures. Otherwise, the alliance and its member states would not have spent money on capabilities they did not need and they would have acted sooner to achieve the necessary capabilities. It is quite clear – as evidenced by alliance initiatives to attain sustainment and deployment capabilities – that NATO knew what doctrines and structures it should have pursued in the wake of the Soviet Union’s demise. Achieving such capabilities might have resulted in more effective and efficient alliance responses to the security threats of the 1990s. Instead, political bargaining got in the way of a threat-based answer to the end of the Cold War, resulting in an alliance response that was far from optimal. Bargaining behavior was clearly evident – indeed, dominant as an explanatory tool – in how the alliance moved toward an expeditionary doctrine and structure, how it chose which candidates to designate as HRF(L)s, and how it structured command relationships. Can Security Threats Explain NATO’s Doctrine and Structures? At first glance, the decision to change NATO’s doctrine and subsequently its force structure was based purely on changed security threats – the demise of the Soviet Union and the development of other types of threats to the security of NATO-member states demanded a new way of providing security for the alliance and a new way of fighting. Such an interpretation presumes that the nature of the threats confronting the members of the alliance changed and therefore the alliance’s response had to change through the adjustment of doctrine and the force structure used to implement that doctrine. Indeed, both NATO’s 1991 and 1999 Strategic Concepts recognized and identified a new array of threats and stated that the alliance would need to change in order to address the new security environment. NATO’s own internal threat assessments had concluded the same regarding the fundamentally changed security environment. As early as 1988, NATO analysts had questioned the ability of the Soviet Union both to move its forces quickly through Eastern Europe and to make effective use of Warsaw Pact units at its disposal. By the end of 1989, both political and military analysts had concluded that the slow demise of the Soviet threat permitted the alliance to make dramatic changes to its force structure. The first effort at a formal study assessing the impact of the changed threat environment on NATO, known as the Wittman paper after its primary author on the
56
Alliance Management and Maintenance
international military staff, German Colonel Dr. Klaus Wittman, was completed in early 1990. One of the most compelling aspects of the Wittman paper, which the Military Committee took formal responsibility for in May 1990, was its observation that the nature of the threats facing the alliance had fundamentally changed – indeed, the paper predicted the eventual folding of the Warsaw Pact as a functioning military alliance. The authors of the paper noted that although the residual military capability of the Soviet Union would remain a threat, NATO would increasingly have to turn its attention to more unpredictable threats such as ethnic strife, religious fundamentalism, and terrorism. The authors concluded that in response, although readiness levels might decrease due to a reduced concern with a surprise Soviet attack through Central Europe, more mobile and flexible forces would be required, signaling the need to abandon the doctrine and command and force structures that had sustained NATO through much of the Cold War. Hence, changed security threats triggered the alliance to begin assessing and updating its doctrine and command and force structures. However, relying solely on security threats to explain the sources of NATO’s doctrine, and ultimately its command and force structures, is problematic for two reasons. First, the shift in doctrine and the related changes in command and force structures were implemented well after such changes became necessary as outlined above and in subsequent alliance revisions to NATO strategy. Certainly some of this can be explained away by considering how uncertain the European security situation looked in 1990 or 1991 and the fact that Russian forces were not completely withdrawn from Central Europe until September 1994. Analysts and policymakers alike were unsure how the political landscape would evolve. Nevertheless, it took the alliance over a decade, from 1991 until 2003, before it completed all of the doctrinal and command and force structure changes that would enable it to respond effectively and efficiently to the non-traditional challenges identified in the 1991 Strategic Concept and confirmed in the 1999 Strategic Concept. The notion that NATO was more likely to face diverse, multidirectional, transnational threats such as ethnic conflict and terrorism and that it therefore needed more flexible, deployable forces was recognized publicly by the alliance on numerous occasions. The alliance knew it faced novel threats for which its heavy, largely nondeployable, and territorial defense-based doctrine and command and force structures were a poor response, and yet the development of revised military doctrine and the actual completion of necessary changes in command and force structure did not occur for several more years. Certainly the development of new weapons systems and capabilities can take many years, but there are also many commercial, off-the-shelf technologies and existing weapons platforms that the alliance could have leveraged in order to strengthen its deployment and sustainment capabilities and hence make itself more useful for addressing post-Cold War threats to member state security. In essence, the command and force structure changes implemented in 2002 and 2003 were based upon a doctrine that sought to fulfill requirements identified in the 1991 Strategic Concept and arguably as early as 1989 or 1990. It seems unusual that a threat-based interpretation would have expected NATO – an alliance that has as its stated reason of being the defense of member state security – to have waited so long to adjust to the new security threats.
Changing Threats and the Alliance’s Response
57
Some might argue that this is just asking too much of an alliance of sovereign states. After all, getting the alliance to turn on a dime is simply not practical. Nevertheless, one can examine the events surrounding NATO’s founding at the outset of the Cold War – an international security watershed certainly as important in its implications as the demise of the Soviet Union and the corresponding end of the Cold War – for an example of how the Western allies had previously responded to dramatic changes in the security landscape. Walter LaFeber, a leading historian of American foreign policy, has argued that by 1946 many in the West had already begun to realize that the Soviet Union represented a significant threat to European security.1 The Soviets had taken aggressive action toward Manchuria, Iran, Turkey, and Europe, and Stalin issued a provocative speech in February of that year essentially calling for the Soviet Union to maintain a wartime footing militarily. In response to these and other provocations, by early 1948, the Brussels Treaty was signed and by the middle of that same year negotiations on the Treaty of Washington had begun. In sum, just three years after the end of World War II, the rising Soviet threat and the onset of the Cold War compelled the Western powers to agree to an unprecedented treaty aligning themselves together and pledging mutual defense. Compared to the beginning of the Cold War and NATO’s founding, efforts to adjust doctrine and command and force structures after the end of the Cold War took far longer. The second reason – and arguably the more important one – why relying solely on changed security threats to explain NATO’s post-Cold War doctrinal and command and force structure changes is inaccurate is that NATO’s force structure did not correspond to the actual threats facing the alliance or provide the alliance with all the capabilities it needed for the post-Cold War era. After NATO finally began to implement the necessary force structure changes in 2002, it did not choose to establish the three land High Readiness Force headquarters that were deemed necessary as noted in the alliance’s strategic planning documents and in ministerial guidance. Instead, the alliance chose to establish as many as six HRF(L)s. One wonders whether there might today be eight HRF(L)s, had the Greeks and the Americans decided to push forward with their initial candidacies – it seemed unlikely that NATO would have turned them down. If NATO were responding solely to threats in the external environment, it would have established only the three HRF(L)s that were required to fulfill its doctrine and strategy, directing excess capacity and resources toward the procurement of other necessary defense capabilities such as strategic lift, intelligence, or communications. Something other than threats must have led it to establish six HRF(L)s. Even using solely academic notions of doctrine, threat-based explanations fall short of telling the whole story. For example, using the most commonly accepted academic definition of doctrine – how battles are fought – it is clear that NATO did not change the way in which it would fight battles until the development of the NRDCs was complete. Its strategy (in the form of the Strategic Concepts of 1991 and 1999) identified new threats and called for new ways of fighting, but the alliance’s 1 Walter LaFeber, America, Russia, and Cold War: 1945–1984, fifth edition (New York: Knopf, 1985), pp. 29–40.
58
Alliance Management and Maintenance
early efforts at adjusting how it would fight were largely superficial efforts to patch over dramatic, budget-driven defense cuts throughout the alliance. For instance, and as noted in the previous chapter, the ARRC lacked the deployable CS and CSS necessary to fulfill post-Cold War requirements, and the multinational corps were so in name only. A traditional academic approach would have expected significant doctrinal innovation and strong integration of doctrine with strategy during times of crisis, several of which occurred in the early 1990s, including the attack on the Russian parliament in August 1991, the collapse of the Soviet Union in December 1991, and Serbian wars against Slovenia, Croatia, and Bosnia. Yet there was no significant impact on alliance doctrine, particularly in terms of integration, beyond the largely surface changes noted previously. The demise of the Soviet Union and the Balkans conflicts certainly had an impact on NATO strategy, as seen in changes in the 1991 and 1999 Strategic Concepts, but the alliance did not respond to crises as a traditional academic approach might have expected. Indeed, in at least one instance, the opposite occurred – the onset of the Kosovo war actually contributed to the slowing down of the process of doctrinal change. Nonetheless, changed security threats did play a role. Certainly, threats help to explain why NATO initiated a doctrinal and command and force structure review. Additionally, the response that NATO ultimately chose by 2003 in terms of doctrine and structures bore some relationship to the threats it identified as early as the late 1980s insofar as the new doctrine and structures emphasized non-territorial, expeditionary capabilities used in defense of out-of-area interests against nontraditional threats. However, the response was clearly suboptimal from a threat-based perspective because it ultimately resulted in an inefficient allocation of military resources among the allies and forced the alliance to choose among less than ideal alternatives for responding to some of the major Euro-Atlantic security challenges of the 1990s. The Suboptimal Alliance Response NATO’s doctrinal and structural responses to the demise of old threats and the development of new ones was suboptimal as evidenced in several instances, preventing the alliance from acting as effectively and efficiently as it might have during the 1990s. These instances include the alliance’s continuing lack of certain critical defense capabilities, its response to the crisis in Albania, and its response to Serb-initiated wars in the former Yugoslavia. Missed Opportunities – Defense Capabilities Perhaps the most significant manifestation of NATO’s suboptimal doctrinal and structural response is the inefficiency with which the alliance and its member states allocated increasingly scarce defense procurement funds. This missed opportunity is the most profound of those considered in this section because it arguably has the longest-lasting implications for alliance capabilities and hence member state
Changing Threats and the Alliance’s Response
59
security. Because of long lead-times necessary for the development and procurement of some military capabilities, decisions made today on doctrine and force structure have implications that can last a decade or longer. When the alliance chose to accept all six High Ready Force (Land) candidates, it chose three more than it needed, according to its own analysis of the capabilities necessary for dealing with the threat environment. The alliance’s threat analysis and the requirements for an appropriate response were outlined in part of alliance doctrine that called for NATO to be able to conduct three corps-size operations at once, requiring three HRF(L)s. The requirement for three HRF(L)s therefore was not simply chosen out of thin air – it was based on an objective, military assessment of the security threats confronting the alliance. However, subsequent doctrine and plans would enshrine all six candidates – every country that volunteered an HRF(L) candidate was allowed to field one.2 This means that three of those HRF(L) host nations were paying to develop allied capabilities that NATO did not need. For example, Spain and Italy in particular were known to have spent the equivalent of tens of millions of US dollars on upgrading their capabilities in order to obtain the HRF(L) designation. It is not just the initial fielding of excess HRF(L)s that represented a waste of scarce defense resources – the costs of maintaining HRF(L) readiness is also quite high. The ARRC alone requires 50 exercises every year to maintain proficiency and readiness, consuming thousands of man-hours and significant financial resources annually. Instead of wasting money on needless capabilities, alliance members could have – and, if the alliance’s own initiatives since the early 1990s are any guide, should have – made more efficient use of their limited defense budgets by spending money on other, more useful capabilities. What should the alliance have spent money on instead? Although some argue that a capabilities gap has existed since the alliance’s inception, the most significant manifestations of this gap only became readily apparent in the 1990s as a result of two phenomena – NATO’s decision to take on out-of-area missions and the inability of the European allies to maintain pace with American defense expenditures in terms of research and development and procurement. As the alliance moved to take on out-of-area missions, most of the force structure of the European allies immediately became obsolete since it was primarily devoted to a strategy of territorial defense and consisted largely of heavy armor divisions and conscripted infantry. Only the United States had invested significantly in such capabilities as long-range lift during the Cold War, which it would have used to reinforce its allies in Europe in the event of a Soviet 2 Of course officially the alliance has no authority to command its sovereign member states to take particular actions, or to not do something. In this case, when at least six allies came forward with offers to develop HRFs, the alliance could not have told member states that they could not do this – after all, the alliance is comprised of sovereign states that have agreed to collective defense, not to surrendering their own decision-making authority in areas such as defense procurement to a supra-national body. However, if the alliance or if enough allies individually disagreed with the decision to accept and certify six HRFs, roadblocks (administrative, political, or otherwise) could have been erected to cause delay and possibly reconsideration of the decision. In other words, NATO and individual allies could have limited the number of HRFs.
60
Alliance Management and Maintenance
attack and which it subsequently found vitally necessary for NATO out-of-area deployments. Most European allies, on the other hand, planned to fight a third world war in Central and Western Europe, and so deployability and sustainability were not driving factors in the development of or dominant characteristics of their doctrine and military structures. In addition, other significant shortcomings in European capabilities existed in such areas as advanced information technology and precision munitions. By the mid-1990s, long-time alliance observers argued that significant inconsistencies existed between NATO’s stated goal of confronting new, postSoviet threats to alliance security, including out-of-area threats, and the alliance’s military capabilities.3 Force improvements and adjustments to command structure were recognized as areas that clearly needed attention if the alliance was to continue to provide security for its members in the context of the 1991 Strategic Concept. Specifically, precision-guided munitions, air surveillance, and intelligence gathering capabilities were areas in which the United States carried the lion’s share of the burden during the alliance’s efforts against Bosnian Serb forces in 1994. By the time of the war in Kosovo, the transatlantic gap in capabilities had moved from visible to glaringly obvious. Of the 23,000 bombs dropped in the Kosovo campaign, 7 percent were precision munitions dropped by European allies while roughly 25 percent were precision munitions dropped by the United States. Of the 1,045 allied aircraft involved in the Kosovo campaign, 770 were American, and of the 27,000 support sorties flown by allied aircraft, American forces flew roughly 70 percent. Of the sorties that required precision strike or had to occur in adverse weather conditions, the vast majority were American.4 Plainly stated, experts and analysts saw that the Kosovo war highlighted a capabilities gap that had existed since at least the 1991 Gulf War, wherein the European allies were unable to keep up with the United States in terms of advanced, specialized military forces capable of projecting power over distance and time. But the realization that the alliance – and especially the European allies – needed to develop capabilities in areas such as strategic lift was not simply the view of outside experts. As early as 1993, NATO officials recognized that the alliance lacked critical capabilities in areas such as intelligence gathering, communications, and strategic airlift that had to be addressed if NATO hoped to have success against post-Cold War threats.5 Indeed, the alliance met at the level of the DPC in late 1992 to discuss its shortcomings in surveillance and command 3 For example, see Richard Kugler, US-West European Cooperation in Out-of-Area Military Operations: Problems and Prospects (Washington, DC: RAND Corporation, 1994). 4 Data in this paragraph comes from John E. Peters, David Shlapak, and Timothy Liston, “Allied Power Projection Capabilities,” in Persian Gulf Security: Improving Allied Military Contributions, (Washington, DC: RAND Corporation, 2001), pp. 71–104. 5 “NATO Intel,” Aviation Week & Space Technology, vol. 138, no. 3 (18 January 1993), p. 19. The article quotes NATO’s Supreme Allied Commander in Europe, General John Shalikashvili, who argued for a reorganization of NATO’s intelligence gathering capabilities in order to confront the new types of conflicts the alliance would face in the post-Cold War world.
Changing Threats and the Alliance’s Response
61
systems, which were necessary for new conflict management and rapid reaction missions, particularly for operations in areas in or near North Africa and the Middle East where ethnic strife and economic deprivation threatened to create significant instability along the alliance’s southern flank. By the late 1990s, the challenges associated with the transatlantic capabilities gap had become so pronounced – and was so obvious in Kosovo – that the alliance formalized a coordinated, collective response. In two widely hailed alliance programs – first the Defense Capabilities Initiative (DCI) and later the Prague Capabilities Commitments (PCC) – NATO not only identified the specific shortcomings, but it formally declared its collective decision to act on acquiring the necessary, missing capabilities. These initiatives were viewed as absolutely vital to ensuring alliance cohesion, which some argued was at risk as the gap between American and European capabilities and technology threatened to undermine the allies’ ability to operate together across the spectrum of conflict. By pushing the allies farther apart in terms of military doctrine and capability, such an eventuality could have threatened one of the alliance’s greatest accomplishments – the denationalization of Western European defense. Faced with this growing problem, at the April 1999 Washington alliance summit – which occurred during the Kosovo war – NATO announced it was launching the DCI, “to ensure that all Allies not only remain interoperable, but that they also improve and update their capabilities to face the new security challenges,” according to Secretary General Lord Robertson.6 The DCI was designed to improve alliance capabilities in five overlapping areas: • •
• • •
Mobility and Deployability – the ability to deploy forces quickly to where they are needed; Sustainability – the ability to maintain and supply forces far from their home bases and to ensure that sufficient fresh forces are available for long-duration operations; Effective Engagement – the ability to successfully engage an adversary in all types of operations, from high to low intensity; Survivability – the ability to protect forces and infrastructure against current and future threats; and, Interoperable Communications – ensuring that command, control, and information systems are compatible with each other.
These areas quickly became broad categories into which 58 more specific capabilities were sub-divided. The demands of trying to coordinate cooperation across all of the 58 specific capabilities, plus “spotty performance” on the part of many allies in actually acquiring the necessary capabilities, led the alliance to redouble its efforts at the Prague summit of November 2002.7 There, individual NATO member states made
6 7
“NATO’s Defense Capabilities Initiative,” NATO Factsheet (9 August 2000). Interview with a member of the NATO international staff, 21 March 2005.
Alliance Management and Maintenance
62
specific commitments – although admittedly only political in nature – to improve their military capabilities across eight areas: • • • • • • • •
Chemical, biological, radiological, and nuclear defense; Intelligence, surveillance, and target acquisition; Air-to-ground surveillance; Command, control, and communications; Combat effectiveness, including precision guided munitions and suppression of enemy air defenses; Strategic air and sea lift; Air-to-air refueling; and, Deployable combat support and combat service support units.
The allies identified three means through which increased capabilities in these areas were to be achieved – multinational efforts with lead states; role specialization whereby states would focus on a few niche capabilities; and reprioritization whereby states would give up on some procurement programs in order to concentrate resources on others. With regard to the multinational projects, Germany was tapped to head the consortium on strategic airlift, Norway on strategic sealift, Spain on air-to-air refueling, and the Netherlands on precision-guided munitions. Since the time of the Prague summit, the allies have met with only limited success in trying to acquire the capabilities they knew they needed a decade ago. In trying to ease the burden on members, the alliance decided to allow allies to count on-going national projects toward PCC goals, which are reported every six months. However, this concession was of little use – one US military staff officer assigned to the alliance noted that NATO’s structure “based as it is on heavy, immobile forces, can’t meet the needs or requirements (of the alliance).”8 More recently, the Spanish have indicated their interest in giving up the air-to-air refueling lead, largely for budgetary reasons.9 Had the Spanish and other allies not pursued HRFs, there is no guarantee that the funds used to develop and maintain the HRFs would have been plowed into such areas as air-to-air refueling, but at least the allies would have had at their disposal additional funds in an era of shrinking defense budgets. Missed Opportunities – Stability and Security in Albania A second example of how the alliance’s suboptimal doctrinal and structural response to the end of the Cold War might have had significant implications for European security can be found in the case of political instability in Albania during the mid1990s. Albania’s post-communist transition was rife with political and economic strife. The challenges facing Albanian society reached a pinnacle in late 1996 and early 1997 when, following the collapse of several pyramid investment schemes, hundreds of thousands of Albanians lost their life savings. In late January 1997, riots broke out and government buildings were destroyed in numerous locations 8 9
Interview with a US staff officer assigned to NATO headquarters, 7 February 2003. Interview with a member of the US delegation to NATO, 21 March 2005.
Changing Threats and the Alliance’s Response
63
throughout the country. The riots continued through February and into March of that year – on March 1, the prime minister resigned and the next day a state of emergency was declared. The security situation in Albania deteriorated rapidly, as security services dissolved and most of the country fell into the hands of rebel groups and criminal gangs. By the thousands, refugees began pouring across the Adriatic toward Italy, and foreign countries started to evacuate their citizens from Albania. Between January and March, more than 360 people were killed and 3,500 were wounded as armed conflict broke out throughout the small, impoverished country. By the end of March 1997, the UN Security Council had approved the dispatch of a multinational military force to Albania to oversee the distribution of international humanitarian aid and to impose order. Despite the fact that instability in Albania directly threatened stability in neighboring Yugoslavia and Macedonia – each with significant ethnic Albanian populations – NATO chose not to become engaged in the mission. The military forces of Britain, France, and Germany were reportedly stretched thin by peacekeeping operations in Bosnia,10 and the ARRC had just completed a year-long deployment to Bosnia and was therefore unavailable for deployment so soon after returning home.11 With the alliance formally passing on the chance to stem further instability in the Balkans, it was left to Italy – awash with refugees and facing a domestic political crisis of its own over its Albania policy – to cobble together an ad hoc peacekeeping force. Italian forces comprised roughly 3,000 of the total 7,000 committed to what was called Operation Alba, with the remaining troop contributions coming from Austria, Denmark, France, Greece, Portugal, Romania, Slovenia, Spain, and Turkey. The mission was to permit the distribution of humanitarian aid, but Operation Alba was also designed to promote a political solution to the crisis and thereby prevent an all-out civil war. Aside from the problems associated with any military operation conducted by an ad hoc collection of forces – especially in terms of how long such forces take to assemble, negotiating command and control issues, and working through differing tactics, techniques, and procedures – a major challenge inherent in this operation was one of impartiality. Because Italy – which had invaded Albania during World War II and which had complex cultural and historical ties to Albania – led the operation and comprised such a large proportion of the troops (and because Greece, which borders Albania and has historical claims to Albanian territory, contributed forces), the peacekeeping force was subject to accusations that it lacked impartiality.12 This perceived lack of impartiality risked spurring greater instability within Albania, as the peacekeepers themselves might have become the object of attacks by the warring parties. 10 William Drozdiak, “Europeans Reject Call For Help in Albania; NATO Reluctant to Intervene Despite Threat,” The Washington Post, 14 March 1997, p. A19. 11 Among Western militaries, a yearlong deployment in a military operation is typically followed by a period of at least several months during which the unit rests, recovers, and reconstitutes itself. This then is typically followed by a period of another several months of training to make the unit fully ready for deployment again. 12 Marcella Favretto, with Tasos Kokkinides, “Anarchy in Albania: Collapse of European Collective Security?” British American Security Information Council Occasional Papers on International Security Policy, no. 21, June 1997.
64
Alliance Management and Maintenance
In any case, the last troops associated with Operation Alba left Albania on 12 August 1997, roughly two weeks after elections were held there. During their brief stay, the military forces of Operation Alba had a reputation for not doing much beyond taking over major seaports and airports or staying in their barracks, and indeed criminal elements retained control of sizeable areas of Albania throughout the operation. Since security and government authority throughout the country were still not firmly established as Operation Alba came to an end, Albania has since struggled through political and economic reforms. Taken together, the lack of doctrine and necessary command and force structures cannot alone explain why NATO did not become more engaged in the crisis in Albania – political will, or the lack thereof, played a key role in addition to other factors. However, if NATO and its member states had had the appropriate doctrine and structures in place to deal with the kind of threat represented by Albania’s descent into chaos during 1996 and 1997 – again, the kind of post-Soviet threats identified in the alliance’s 1991 strategy – member state decision-makers might have been less reluctant to become involved through NATO and an international response, led by NATO, might have been swifter, more effective, longer lasting, and less susceptible to claims that it lacked impartiality. Instead, some allied governments declared themselves tapped out, a time-consuming, inefficient ad hoc approach was adopted, and the mission fell largely to the Italians, who were not necessarily trusted by the Albanians and who were motivated primarily by their desire to shut down refugee points of embarkation versus a serious concern over helping the government in Tirana to establish control and long-term stability. Missed Opportunities – Bosnia A final example of how a more optimal doctrinal and structural alliance response to the end of the Cold War might have had profound implications is the story of how NATO responded to the Serb-led war against Bosnia. In June 1991, Croatia and Slovenia each declared independence from Yugoslavia, resulting in a military response from Belgrade. Because Slovenia is a more ethnically homogeneous state and because it does not border Serbia, the armed conflict between Belgrade and Ljubljana was relatively short-lived. In contrast, the war between Serbia and Croatia was far more intense and lasted significantly longer. Despite the conflict raging around (and in some cases in) Bosnia – 10,000 people had died in Yugoslav conflicts by January 1992 – Bosnian Croats and Muslims formed a tactical alliance to get the Bosnian government to declare independence from Yugoslavia as well in February 1992, overcoming Bosnian Serb resistance.13 In response, Bosnian Serbs declared their own republic. Over the next several months, the situation in Bosnia deteriorated significantly as Bosnian Serbs, Croats, and Muslims all fought each other and refugees from the crisis begin to flood into Western Europe. By this point, the United Nations had begun to engage seriously, moving beyond an arms embargo by instituting 13 Bosnia was the most ethnically diverse of the Yugoslav republics, with 43 percent Muslims, 31 percent Serbs, and 17 percent Croats.
Changing Threats and the Alliance’s Response
65
sanctions against Serbia and declaring so-called safe areas for Bosnian Muslims in Bihac, Gorazde, Sarajevo, Srebrenica, Tuzla, and Zepa. In December 1992, in the midst of continued war in Bosnia, the Secretary General of the United Nations, Boutros Boutros-Ghali, sent his counterpart at NATO, Manfred Wörner, a letter requesting NATO assistance in carrying out UN resolutions. From the outbreak of war in the former Yugoslavia through the mid1990s, several cease-fires and peace negotiations had been concluded and failed for a variety of reasons, but often because many of the agreements lacked the threat and application of military force. Wörner had not been an advocate of out-of-area allied operations when he first became Secretary General of NATO, but over time, and particularly as a result of the deteriorating situation in the former Yugoslavia, he gradually viewed out-of-area peacekeeping missions as within NATO’s scope. Indeed, the crises in the Balkans had a similar effect on many within the alliance organization and on many of the allies. On 17 December 1992, the NAC moved to strengthen its relationship with the UN by declaring itself ready, on a case-by-case basis, to carry out peacekeeping operations under the authority of the UN Security Council. This made great sense from the perspective of the alliance’s 1991 strategy, which called for NATO to maintain “an overall capability to manage successfully crises affecting the security of its members,” and which noted that the, “Allies could, further, be called upon to contribute to global stability and peace by providing forces for United Nations missions.”14 However, as NATO’s involvement in the peacekeeping mission in Bosnia unfolded, it took place in what one scholar has termed ‘a conceptual vacuum.’15 NATO lacked a doctrine for out-of-area operations, which caused problems with regard to formulation of both political objectives and the military means to achieve them. The internal transformation of NATO lagged behind the external transformation – although its strategy had shifted, alliance doctrine and structures had not, so that by the mid-1990s (and as seen above in the case of Albania) there were significant doubts about the effectiveness of NATO in handling the new tasks it had assigned itself in the post-Cold War strategy.16 Indeed, during the allied bombing of Serb positions in August and September 1995, the alliance had to rely on US capabilities; Washington had to bring into the theater the necessary special operations forces and command, control, and communications capabilities. Later, during the actual deployment of NATO forces to Bosnia beginning in December 1995 and the beginning of the IFOR mission, there were significant challenges. For example, there was a lack of doctrine to deal with the deployment itself – non-NATO and Partnership for Peace countries were to participate with the alliance in a real-world operation for the first time and there was little NATO guidance 14 NATO Strategic Concept, 1991, paragraphs 19, 41. 15 De Wijk, NATO on the Brink of the New Millennium, p. 59. 16 For more on this in the context of Bosnia, see De Wijk, NATO on the Brink, pp. 117–9. De Wijk concludes, “by the end of 1995 only partial transformation had occurred” because “political controversy over the course to be steered stood in the way.” In fact, he terms many of NATO’s strategic-level shifts during the early to mid-1990s “hollow concepts” that gave little direction to the transformation process.
66
Alliance Management and Maintenance
on how to proceed. Additionally, NATO had no in-place ability to deploy forward its strategic command, control, communications, and intelligence capabilities – it had to rely extensively on ad hoc means provided primarily by the United States. Finally, in terms of other combat service support, there was a lack of organic psychological operations (PSYOP) and civil-military cooperation (CIMIC) capabilities in the ARRC, capabilities that should have been resident in a force structure prepared to handle peacekeeping missions.17 In the early and mid 1990s, the alliance clearly did not have in place the necessary doctrine and structures to conduct military missions that addressed the nontraditional, multi-dimensional threats outlined in its own 1991 strategy. If it had had these doctrines and structures, it might have responded more rapidly and effectively to the deteriorating situation in the former Yugoslavia. However, unconventional military tactics on the part of the Serbs – such as holding 300 UN observers as hostages in May 1995 – frustrated and restricted the use of NATO’s air power. NATO’s doctrines and forces, dominated by heavy, sophisticated weapons systems, were not well suited for a conflict in which the enemy operated in lightly armed militias broken into small groups that melted easily into the civilian population. As in the case of the Albanian crisis discussed above, appropriate doctrine and structures may not have been a panacea, but if the alliance had had in place the kinds of doctrine and structures necessary to respond to these threats – the sort of postSoviet threats that the alliance itself identified in its 1991 strategy – its member states might have had the appropriate forces and alliance operations in the Bosnian theater might have been more efficiently implemented with fewer of the problems noted above. Unfortunately, by the mid-1990s there remained only a limited potential for NATO to fulfill its mission in the kinds of non-conventional conflicts identified in its strategy due to the absence of the necessary doctrine and structures. As seen through the examples cited above, although changed security threats did play a role in the alliance’s doctrinal and structural response to the end of the Cold War, they do not tell the whole story – threats are necessary but not sufficient for explaining alliance behavior. Other explanatory tools are necessary to understand the process and content of NATO’s post-Cold War doctrinal and structural changes – in the case of the NATO Rapid Deployment Corps, this includes both the timing of the review and the nature and shape of the resulting doctrine and command and force structures. Indeed, a closer examination shows that other explanatory tools are necessary to shed light on why it was not until 2002 that NATO actually began to implement the NRDC-based force structure and on why the alliance wasted scarce resources by implementing a force structure that did not correspond to its own requirements. More specifically, one set of alternative factors appears to offer the necessary explanatory power needed to augment the threat-based explanation – political bargaining factors.
17 For more detail on these challenges, see Larry Wentz, ed., Lessons from Bosnia: The IFOR Experience (Washington: Institute for National Strategic Studies, 1997).
Chapter 6
The Impact of Political Bargaining The alliance’s doctrinal and structural decisions – or lack thereof – in the 1990s were clearly the result of something other than a response to a rational assessment of the threats facing the alliance. To understand both the process and the content of the doctrinal and command and force structure review that took place from roughly 1991 to 2003, one must consider the role of political bargaining. Intra-alliance bargaining among member states is motivated by the desire to avoid burdens and garner benefits and is characterized by the horse-trading that takes place whenever shared responsibilities or limited resources are to be divided up among numerous parties pursuing a collective good. From the available evidence, it appears as if political bargaining played a very significant role in determining both the process and the results of the doctrine and command and force structure review that ultimately resulted in the NRDCs and that led to a suboptimal response in terms of the alliance’s ability to handle post-Cold War threats efficiently and effectively. The next several sections will outline how relative levels of interest and, to a lesser degree, dependence shaped bargaining over the burdens to be shared and the benefits to be won, which in turn had a decisive impact on the process and the outcome of NATO’s post-Cold War doctrine and structure review. Bargaining Over Expeditionary Capabilities To address post-Cold War threats, NATO recognized that it needed expeditionary capabilities. The alliance’s own strategy acknowledged this requirement, and yet efforts to develop the necessary doctrine and force structure were subject to significant bargaining among the allies, especially between Germany on the one hand and the United States on the other. Germany had serious reservations, based on historical, political, and economic factors, about the direction the alliance appeared to be headed when the out-of-area topic was first broached.1 Berlin’s interest in this issue was rooted in post-war German politics and constitutional tradition. Prior to unification there was a political consensus in the former West Germany that the constitution, known as the Basic Law, forbade any military action that was not considered self-defense. In 1992, the German government agreed to permit the Bundeswehr to participate in NATO and WEU monitoring
1 Interview with a German military officer assigned to NATO, 1 August 2003; interview with a retired senior US military officer involved in NATO planning, 24 October 2003; interview with a British military officer assigned to NATO, 9 July 2003.
68
Alliance Management and Maintenance
missions in the former Yugoslavia. Domestic opposition to this controversial decision among members of the Social Democratic Party and the Liberal Democratic Party prompted the pursuit of legal action through the German constitutional court. In July 1994, the court ruled that Germany could employ its military forces out-of-area so long as deployments occurred in the context of collective security, also noting that NATO or UN-led deployments fell into this category. Nonetheless, there was still great political resistance within Germany to doing much beyond so-called ‘checkbook diplomacy’ – Germans would fund operations, and provide moral support, but many believed Bundeswehr soldiers should not conduct operations beyond the defense of Western Europe. Developing doctrine that emphasized expeditionary missions and that downplayed territorial defense meant focusing on expeditionary-type forces, not the augmentation forces that underpin territorial defense planning and that characterized large portions of Germany’s military. Augmentation forces in NATO’s context have traditionally consisted of the least deployable, least trained forces, which are most often comprised of conscripted personnel. These forces are necessary for territorial defense, especially when confronting a numerically superior Soviet enemy. However, conscripted augmentation forces of this sort are of little value for a post-Cold War NATO alliance because they can lack the necessary training, commitment, and deployability that professional, expeditionary forces typically have. German political and military leaders were reluctant to give up the requirement for augmentation forces for two reasons. First, the allied requirement for augmentation forces provided a significant source of the justification for Germany’s system of conscription and therefore was a major source of the military’s institutional strength.2 Second, and more importantly, giving up on the system of conscription would fly in the face of decades of established practice and conventional wisdom in post-World War II Germany, which held that a system of national conscription was the most effective way of preventing a militarist class from forming, which many in Germany viewed as one of the factors contributing to the rise of Nazism. The Germans were not alone in this though – other allies applied the brakes in order to delay the move toward a doctrine and force structure that would have enabled the alliance to field more expeditionary forces.3 When staff members of the IMS drafted a proposal in 1997 for a notional doctrine on NATO’s revised force structure that included requirements for more expeditionary forces at higher levels of readiness, several allies quickly squashed the idea. The problem was not necessarily that the IMS had proposed doctrine that did not make sense from a policy perspective – indeed, an educated reading of guidance approved by NATO to that point in time would likely have led an informed observer to the same conclusions on the future shape of the alliance’s forces. Rather, the reason why some member states objected was because many of them were unwilling or unable to secure the kind of 2 Germany is the only major European NATO member that continues to rely on conscripts. For example, France, Italy, the Netherlands, Spain, and the United Kingdom either do not use conscripts or are in the process of phasing conscription out. 3 Interview with a British military officer assigned to NATO, 9 July 2003.
The Impact of Political Bargaining
69
financial resources that would have enabled such a doctrine to be enacted or fulfilled.4 Developing forces at higher levels of readiness requires significant defense budget investments in rapid power projection and deployment capabilities, and throughout the 1990s most European defense budgets simply could not meet such demands. NATO’s own figures on member state defense expenditures as a percentage of gross domestic product (GDP) show that in European NATO, defense spending fell from 3.2 percent of GDP between 1985 and 1989, to 2.6 percent between 1990 and 1994, to 2.1 percent between 1995 and 1999.5 Most of European NATO faced two major economic hurdles in the 1990s, which contributed in part to lower defense budgets. The first was sputtering national economies marked by low growth rates and greater demands on generous social welfare and pension systems. The second was the significant pressure to restrain discretionary budget expenses in the run up to further economic integration within the European Union. These factors contributed significantly to lower European defense budgets – spending for logistical support was particularly hard hit. According to a former senior US military officer assigned to NATO, the alliance was getting worse “at doing stuff beyond putting combat troops on the ground.”6 From a bargaining perspective then, Germany and several other European allies initially had a moderately strong interest in avoiding a doctrine that relied on expeditionary capabilities, even though political leaders had implicitly agreed to this through NATO’s strategy. On the other hand, the United States, the alliance’s leading member, had significant interest in encouraging a move toward expeditionary doctrines and forces as a means of promoting more equitable burden-sharing in the alliance – it wanted the allies to contribute more readily to NATO interests beyond alliance territory. The experience of the alliance in the 1999 Kosovo war – in which American logistical support, command and control assets, satellite communications, precision guided munitions, and airframes accounted for the vast majority of the allied effort – had a particularly strong influence in spurring Washington to find ways to more equitably share the costs of the common defense and thereby to lighten the operational load on US forces and capabilities.7 Therefore, Washington declined to nominate its own HRF(L) and ultimately supported the creation of more HRF(L)s than were demanded by NATO’s own strategy, guidance, and doctrine because it was thought this would require more allies to invest more heavily in deployable CS and CSS, something the United States had long favored, especially as the alliance has looked to conduct 4 Interview with a retired US military officer, 27 April 2004. 5 “Financial and Economic Data Relating to NATO Defence: Defence Expenditures of NATO Countries (1980–2003),” Table 3, NATO Press Release, 1 December 2003. 6 Interview with a retired senior US military officer formerly assigned to NATO, 24 October 2003. 7 Interview with a US military officer assigned to the European Command, 14 November 2002; an interview with a US defense official, 25 January 2003; an interview with a US defense official, 19 February 2003; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; correspondence with a former DoD official, 5 August 2004; and interview with a former DoD official responsible for NATO policy, 22 September 2004.
70
Alliance Management and Maintenance
more missions outside the Central Region.8 In the words of one US military officer involved in NATO policy and planning, Washington wanted the European allies to “pick up the slack with regard to mobility and interoperability.”9 Gradually, the European allies shifted from opposing expeditionary doctrines and structures on historical, political, and economic grounds. There are several reasons for why they eventually become supporters of expeditionary capabilities, especially as embodied in the NATO Rapid Deployment Corps.10 First of all, the challenges of instability emanating from the Balkans was very real for the European allies, many of whom found themselves awash with refugees from the region or found themselves under increasing political pressure domestically to stem the violence in Bosnia on human rights grounds. These domestic economic and political dynamics forced them to realize that in order to create conditions of peace and stability, NATO would have to project power outside of its own member state territory, and power projection requires expeditionary capabilities. Second, NATO strategy demanded an expeditionary doctrine and capability, and the alliance’s credibility gap only grew larger with each passing year as the allies put off hard decisions on developing expeditionary forces to meet the needs of their own alliance strategy. Third, the European allies had great interest in acquiring the prestige that came with developing the NATO Rapid Deployment Corps. Hosting an NRDC would confer a level of military capability and accomplishment that would augment international as well as domestic perceptions of their country and its armed forces – it meant “being at the top of the NATO heap.”11 Fourth, the European allies still desired a close relationship with US military forces, and hosting an NRDC brought with it the possibility of having US staff officers assigned to that NRDC, which itself was a way to garner greater prestige and a means of potentially influencing the US military. In sum, the allies’ interest in avoiding the manifold costs associated with expeditionary capabilities was outweighed – and only overcome after a significant period of time as they saw events unfold in Bosnia and elsewhere – by their stronger interest in promoting peace in the Balkans, their interest in a credible and viable alliance, their interest in international military prestige, and their interest in influencing and belonging to the ‘big leagues’ with the United States. 8 Interview with a US military officer assigned to the European Command, 14 November 2002, and interview with a US government official responsible for military planning, 19 February 2003. 9 Interview with a US military officer assigned to the European Command, 14 November 2002. 10 Interview with a US military officer assigned to the European Command, 14 November 2002; an interview with a US defense official, 25 January 2003; an interview with a US government official responsible for military planning, 19 February 2003; an interview with a US military officer, 10 April 2003; an interview with a British officer assigned to NATO, 9 July 2003; an interview with a German military official assigned to NATO, 1 August 2003; an interview with a Polish officer assigned to NATO, 3 October 2003; an interview with a senior Turkish officer assigned to NATO, 3 October 2003; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004. 11 Interview with a British military officer, 9 July 2003.
The Impact of Political Bargaining
71
However, the allies only partially embraced the commitment to expeditionary capabilities, continuing to the present day to underfund deployability and sustainability. A threat-based approach likely would have expected the alliance to embrace expeditionary capabilities more quickly and more fully. Instead, the alliance did so only after a great deal of time had passed, during which crises in southeastern Europe worsened, and only in part, side-stepping necessary investments in deployment and sustainment, the things necessary to get the NRDCs to crisis locations and support them once they arrived. Even though the alliance had at least begun heading down the path toward greater deployability and sustainability of the sort necessary to implement its own strategy, there were still plenty of unresolved issues facing member states in this regard. Subsequent deliberation would focus not on the most appropriate response to the threats facing the alliance, but rather would revolve around allies trying to obtain as much individual benefit as possible while doing only what was minimally necessary to fulfill alliance strategy. Bargaining Among the HRF(L) Candidates The German-Netherlands Corps In trying to obtain an HRF(L), the interests and dependence of the Dutch, and hence their bargaining position, was shaped in part by domestic political and economic factors. Dutch officials maintained that unless the Netherlands were at least party to a land HRF, their military budget was likely to decrease substantially.12 During the period in which the candidate countries were making their pitches for the HRF(L) designation, there was strong reason to believe that the Dutch military was engaged in a serious battle for its institutional reputation as a result of the Srebrenica incident and would be in a poor position to secure a strong budget allocation in the absence of a NATO requirement.13 In early April 2002, the Dutch Institute for War Documentation completed five years of research into and analysis of the Srebrenica incident by releasing a 6,000-page report that concluded that the Dutch military had, “consciously tried to withhold information from the investigation in the Srebrenica affair and, particularly, the de-briefing of [the Dutch soldiers involved in the incident].” So politically devastating was the report, that a week after the report’s release, the Dutch cabinet of Prime Minister 12 Interview with a US defense official involved in policy planning, 25 January 2003; interview with a Dutch military officer assigned to NATO, 1 August 2003; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; and correspondence with a former DoD official, 5 August 2004. 13 Srebrenica was a Muslim enclave in eastern Bosnia under siege in 1995 by the Serbs. A contingent of roughly 400 Dutch soldiers, operating under the UN-mandated Protection Force (UNPROFOR), was charged with protecting the 30–40,000 Bosnian Muslims in Srebrenica. As Serbian forces attacked Srebrenica in July 1995, the Dutch soldiers escorted women and children out of the city, leaving behind roughly 7,500 Muslim men who were subsequently massacred by the attacking Serbs.
72
Alliance Management and Maintenance
Wim Kok resigned, announcing that the government assumed full responsibility for the actions taken by the Dutch army and government in 1995. More generally, the fact that defense budgets, and subsequently national force structures, had already been slashed following the end of the Cold War lent even greater weight to fears of further cuts claimed by the Dutch. By the late 1990s, most European members of NATO were spending less than two percent of their gross domestic product (GDP) on defense, and indeed the Royal Netherlands army faced particularly dire resourcing circumstances for its 32,000 active duty personnel. For Germany’s part, its bargaining position was strengthened by its strong interest in righting the perceived wrong suffered when Britain was named the ARRC’s lead.14 German officials were upset by the alliance decision to designate Britain the framework country for the ARRC in the mid-1990s – after all, the ARRC headquarters and the units that comprised the ARRC were based primarily in Germany, and so to have another country leading the ARRC was something of an affront to German pride. German officials felt they were owed an opportunity to participate in the high-visibility world of the HRF(L)s, and so they argued strenuously for designation of the German-Netherlands Corps as an HRF(L). Additionally, and like the Dutch, defense budget considerations also drove German interests. Without the requirement generated by membership in one of the HRF(L)s, the German military would have had a harder time justifying its defense budget. Germany is one of NATO’s largest members and had traditionally contributed substantially to alliance force structure. The transformation of NATO’s force structure away from territorial defense threatened to leave those member state militaries that were comprised of territorial defense forces, such as Germany’s, without a rationale for their existence. German officials wanted to ensure that if NATO was going to dramatically change its force structure, Germany would have a prominent role in that new structure. From the alliance’s perspective, NATO had both a moderate interest in securing German participation in an HRF(L) and a limited level of dependence on Germany as well, which brings significant political weight to the NATO table as well as potentially significant military resources.15 In this case, the alliance truly needed German participation – to leave such a major ally outside the HRF(L) structure would have relegated the German military to second-class status within the alliance and would have risked undermining political support for the alliance within Germany, something that NATO could ill afford in the face of already declining German defense budgets. In this situation then, the alliance was weakened in its bargaining 14 Information in this paragraph comes from an interview with a US defense official involved in policy planning, 25 January 2003; an interview with a US government official responsible for military planning, 19 February 2003; an interview with a former US military officer assigned to NATO, 23 October 2003; an interview with a German military official assigned to NATO, 1 August 2003; an interview with a Dutch military official assigned to NATO, 1 August 2003; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; and correspondence with a former DoD official, 5 August 2004. 15 Correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; and correspondence with a former DoD official, 5 August 2004.
The Impact of Political Bargaining
73
position by its dependence on Germany for its military contribution to allied security. Meanwhile, Germany and the Netherlands saw their bargaining positions greatly strengthened by their strong interests in the HRF(L) designation, driven by the need for prestige and by domestic budgetary factors, resulting in the designation of the German-Netherlands Corps as an HRF(L). The Spanish Corps As with Germany, which received the HRF(L) designation in part by exploiting the dependence of NATO on a robust German contribution to alliance defense, intraalliance bargaining over locating an HRF(L) in Spain was shaped in part by the dependence of NATO on retaining Spain’s commitment to the integrated military command (IMC).16 Related to this, Madrid’s case for an HRF(L) was also motivated and strengthened by its strong interest in obtaining tangible rewards for the Spanish decision to rejoin the IMC.17 When Spain joined NATO in 1982, it did not join the IMC. This decision was affirmed just a few years later when, in 1986, a national referendum supported by the Spanish Socialist party of Prime Minister Javier Solana resulted in Spain remaining outside the IMC. A decade later, in November 1996, following conservative José Maria Aznar’s electoral defeat of the Socialists, Madrid reversed course and took the first steps toward joining the IMC.18 Despite the gradual lessening of anti-US and anti-NATO sentiment among the Spanish populace in the decade since the 1986 referendum, Aznar’s young government needed some sort of concession to compensate for the change in policy. Indeed, the resolution on integration into NATO’s military structures that passed the Spanish parliament directed the government to seek such an accommodation, noting, “Spain should have command and operational responsibilities in accordance with its military contribution and political weight.”19 To fail in obtaining an HRF or other ‘command and operational responsibilities’ might risk a political backlash against the decision to return to the IMC, against NATO, and of course against the Aznar government. In order to save themselves from these domestic political troubles, Spanish leaders argued for the HRF designation. From NATO’s perspective, the alliance was eager to cement the military relationship by retaining Spanish membership in the IMC, something that might have led to a French decision to rejoin as well. Additionally, Spanish membership in the IMC made common alliance defense structures all the more robust and a Spanish HRF(L) would act as a means for the alliance to promote greater integration of 16 Interview with a US military officer involved in the negotiations, 27 April 2004. 17 Interview with a retired US military officer previously assigned to NATO, 15 March 2004; and an interview with a US government official responsible for military planning, 19 February 2003. 18 At the July 1997 NATO summit, the alliance confirmed that Spain would join the IMC as soon as NATO had completed its command structure review. 19 Quoted in Antonia Marquina, “Spain and the ‘Europeanization’ of Security,” in Carl Lankowski and Simon Serfaty, eds., Europeanizing Security? NATO and an Integrating Europe, American Institute of Contemporary German Studies, The Johns Hopkins University, Research Report No. 9, 1999.
74
Alliance Management and Maintenance
and reform within the Spanish military.20 Specifically, the alliance hoped a Spanish HRF(L) would promote increased interoperability in command and control elements between Spanish forces and those of other alliance members. Indeed, in making their case for the HRF(L) designation, the Spanish outlined plans, which they have largely followed through on since then, to restructure their entire land force around their Valencia-based land HRF.21 The alliance’s dependence on the promise of Spanish defense reform and dependence on Spain for its continued participation in the IMC – coupled with Madrid’s interest in gaining tangible rewards for its decision to join the IMC – led to a decision to designate both of Spain’s rapid deployment candidates – one land and the other maritime – as HRFs. The Eurocorps Intra-alliance bargaining also played a major role in the Eurocorp’s designation as an HRF(L). Since the alliance’s founding, there has been a divide, sometimes more pronounced than others, between NATO’s Atlanticists, such as the Netherlands, Norway, the United Kingdom, and the United States, on the one hand, and NATO’s Europeanists, such as Belgium, France, and Greece, on the other. The former tend to favor trans-Atlantic security ties as the primary means of maintaining peace and stability in Europe, while the latter tend to favor European solutions to European challenges. The development of the Eurocorps, which began as the product of a summit between French President François Mitterrand and German Chancellor Helmut Kohl in 1992 and is based on a French-German brigade formed in 1991, was seen by some Atlanticists as a Europeanist effort to strengthen the hand of those such as the French who want to forge solely European solutions to security issues on the continent. During alliance deliberation over the rapid deployment corps, the more Atlanticist allies, especially the United States and the United Kingdom, favored designating the Eurocorps as an HRF(L) as a means of avoiding any weakening of NATO relative to the European Union’s growing role in security and defense policy and those that were promoting it, such as the French.22 Many within the alliance feared that if the Eurocorps were not incorporated into the alliance’s force structure, both it and the accompanying significant French force contribution would occur and evolve outside of NATO. By casting Eurocorps in the context of NATO – and designating the Eurocorps as one of the HRF(L)s – the Atlanticists hoped to ensure that NATO remained the primary international organization dealing with European security, and thereby undercut French efforts to strengthen the EU at NATO’s expense. An America military officer involved in NATO force planning summarized this view by
20 Interview with a member of the NATO international staff involved in force planning, 16 May 2003. 21 Ibid. 22 Interview with a US military officer assigned to the European Command, 14 November 2002; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; and correspondence with a former DoD official, 5 August 2004.
The Impact of Political Bargaining
75
noting that the Atlanticists in the alliance, “didn’t want the Eurocorps developing outside the NATO umbrella.”23 However, the French saw benefit in designating the Eurocorps an HRF(L) also, and not because they viewed the issue from the perspective of the AtlanticistEuropeanist rivalry. Instead, they favored its designation as an HRF(L) because this would provide France with a means of ensuring at least some of the French armed forces – that is, the units committed to the Eurocorps – could take part in the rigorous training, evaluation, and deployment cycle that comes with being an HRF(L).24 It is very expensive from a defense budgeting perspective to engage in the kinds of training necessary to maintain a world-class military, which is certainly something the French see as necessary for their vital interests. In order to maintain those worldclass capabilities, Paris hoped to take advantage of the training opportunities – funded in part by NATO common funding – available as part of the HRF training cycle. In this case, the bargaining win sets of ostensibly opposing allies or groupings of allies coincided significantly – the interests of France in securing access to advanced training opportunities and the interests of the alliance in keeping the Eurocorps within NATO’s embrace led to a decision that all parties favored, but for vastly different reasons, and hence the Eurocorps was designated an HRF(L). The Turkish Corps With regard to the III Corps of the Turkish Army, there were primarily three reasons for why the alliance decided to designate it as an HRF(L).25 First, many in the alliance had long favored further integration of Turkey, NATO’s only majority Muslim country, with that of the other European allies, at least in terms of defense. The United States has favored this because it has hoped to promote integration in general as well as for other operational reasons given Turkey’s strategic location, while many European allies have favored this because they hope to strengthen concepts regarding civilian control of the military and respect for human rights within the Turkish armed forces. Second, the Turks have long had a strong commitment to spending significant sums of money on defense, something that many in the alliance were eager to promote by awarding Ankara an HRF(L) designation. Third, those within NATO pushing for an HRF(L) in Turkey hoped to spur not just continued robust defense spending but also defense reform within the Turkish military, making it less focused on territorial defense and more interested in expeditionary warfare. 23 Interview with a US military officer assigned to the European Command, 14 November 2002. 24 Interview with a French military officer assigned to NATO, 3 October 2003. Belgium, France, Germany, Luxembourg, and Spain all contribute to the Eurocorps, but France, which hosts the Eurocorps headquarters in Strasbourg, is viewed as the first among equals. 25 Interview with a US military officer assigned to the European Command, 14 November 2002; an interview with a US defense official, 25 January 2003; an interview with a US government official responsible for military planning, 19 February 2003; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; and correspondence with a former DoD official, 5 August 2004.
76
Alliance Management and Maintenance
The Turks had a strong interest in strengthening alliance capabilities in Southwest Asia, basing many of their arguments for an HRF(L) on the military and political threats in their region. Ankara argued that future threats to European security were likely to emanate from areas contiguous to or near Turkey, and that the alliance needed to beef up its capabilities in the region to conduct both Article 5 and nonArticle 5 missions throughout the region. The Turks pointed out that if they did not have an HRF(L) on their territory, Turkey might have to wait for as long as three months for the most capable alliance reinforcements to arrive.26 Having an HRF(L) in close proximity, according to Ankara’s logic, would enable the alliance to meet security challenges in the region more quickly. As a US military officer present for the working group deliberations noted, “Through all of it, Turkey was the most difficult to work with – they wanted to ensure everybody knew they lived in the roughest neighborhood.”27 However, despite trying to drum up the threat, the consensus view among observers and those involved in alliance decision-making at the time is that Turkey gained an HRF(L) “for mostly political reasons” as one former US defense official acknowledged, indicating that the alliance did not necessarily subscribe to the same assessment of Turkey’s security situation as Ankara.28 In any case, as was the case with regard to the Eurocorps, the win sets for both the alliance and an HRF(L) candidate coincided but for different reasons. There were strong incentives from the perspectives of both the alliance and the Turks to create an HRF(L) in Turkey, resulting in the designation of the Turkish II Corps as an HRF(L). The ARRC The Allied Rapid Reaction Corps was, in many ways, the model high readiness force, and so it only seemed reasonable to designate it as an HRF(L). Additionally, several allies, including the Americans, British, Germans, and Dutch, believed it was only logical that the ARRC become an HRF(L) because it “had a relatively positive track record,” as one US military officer involved in NATO force planning said.29 Nonetheless, the ARRC still went through the same nomination and review processes as the other candidates. Unlike those other candidates though, the ARRC was favored by the alliance in part because it was the only one of the six candidates that had significant British participation. As with Germany above, the alliance had a strong interest in retaining Britain’s military commitment to NATO, something that might have been undercut had the ARRC not been designated an HRF(L).30 26 Interview with a senior Turkish military officer assigned to NATO, 3 October 2003. 27 Interview with a US military officer involved in the negotiations, 27 April 2004. 28 Interview with a former DoD official responsible for US policy toward NATO, 22 September 2004. 29 Interview with a US military officer assigned to the European Command, 14 November 2002. 30 Interview with a US military officer assigned to the European Command, 14 November 2002; an interview with a US defense official involved in policy planning, 25 January 2003; an interview with a US government official responsible for military planning, 19 February 2003; an interview with a former US military officer assigned to NATO, 23 October 2003; and an interview with a German military official assigned to NATO, 1 August 2003.
The Impact of Political Bargaining
77
Many allies realized that, as was the case at the ARRC’s founding, without a force justification on the European landmass, the British Army was likely to suffer dramatic budget and manpower cuts. Units of the British Army of the Rhine (later known as British Forces Germany) simply had nowhere to go if it were removed from its bases in Germany, as the necessary military basing infrastructure to house and train it does not exist in Britain. As a result, any cuts in the end strength of Britain’s land forces based in Germany probably would have resulted in a cut in total British Army end strength. Indeed, this concern shaped the interests of the British as London argued for the HRF(L) designation. Additionally, the alliance had a moderate level of dependence on the British for their contribution to NATO’s defense capabilities – the British are among the few allies that can support the deployment and sustainment requirements of global operations. Not allowing the ARRC to become one of the HRF(L)s likely would have resulted in the dissolution of an established alliance capability led by one of the alliance’s most capable military forces. As NATO faced an increasingly growing transatlantic capabilities gap, failure to designate the ARRC as an HRF(L) would have been a step in the wrong direction. Therefore, once again the combination of strong interests on the part of the nominating ally and a moderate level of dependence on the part of the alliance led to the ARRC’s designation as an HRF(L). The Italian Corps Similar to Spain’s case in some respects, the alliance’s bargaining position toward the Italian candidacy was shaped by NATO’s moderate dependence on a reformed, expeditionary Italian contribution to alliance military capabilities, especially in the area of command and control.31 Indeed, in order to upgrade its expeditionary combat service support capabilities and qualify as an HRF(L) host for a corps headquarters based in Milan, the Italian government spent the equivalent of tens of millions of US dollars to upgrade its military tactical satellite communications capabilities. The bargaining position of the Italian government was shaped primarily by its strong desire to achieve political prestige – hosting an HRF(L) would permit the Italians to remain one of the premier European allies, along with the British, French, and Germans, and it would therefore allow Italy to continue playing a major role in European security affairs.32 This search for prestige has been a consistent theme in Italian foreign and security policy, as Rome has attempted to increase its political and military standing and influence in Europe and elsewhere in an effort to become more of an equal partner with not just the major European powers like Germany and France but also with its major transatlantic ally.
31 An interview with a US military officer assigned to the European Command, 14 November 2002; an interview with a US defense official involved in policy planning, 25 January 2003; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; and correspondence with a former DoD official, 5 August 2004. 32 Interview with an Italian military officer assigned to NATO, 1 August 2003.
78
Alliance Management and Maintenance
Italy’s interest in pursuing the HRF(L) designation was also strongly shaped by budgetary considerations. Like its British, German, and Dutch allies, Rome – and particularly the Italian defense establishment – was eager to find additional justifications for the Italian defense budget, and the NATO HRF(L) designation proved a useful tool in this regard. Rome argued that the lack of an HRF(L) on Italian soil would mean a significant long-term cut in the Italian military budget.33 Like some of the other candidates, the interest in prestige and budget justifiers were the primary factors driving Italy’s relatively strong bargaining position. In contrast, the alliance’s bargaining position was somewhat weakened by its dependence on as robust a contribution as possible to allied defense on the part of Italy. Summary Based on its doctrine, the alliance clearly had an interest in limiting the number of HRF(L)s to three. However, in several cases noted above, the alliance’s bargaining strength was undercut by its dependence on the military contributions of some key member states to allied defense and security, particularly as NATO moved toward more expeditionary missions and requirements. In two cases – the Eurocorps and the III Turkish Corps – the bargaining win sets of the proposing allies and the alliance actually coincided to some degree. More frequently though, it seems that the alliance was simply unable to overcome the strong interests of individual members in pursuing prestige and budget justifiers. Evidence suggests that the alliance as a whole – and certainly the United States, the alliance’s key member – viewed the creation of HRF(L)s and expeditionary capabilities in general as critical to alliance integrity, enabling it to go out of area instead of out of business. Taken together, these factors explain much of the reason behind why, after assessing each of the HRF(L) candidates, the alliance designated six of the NRDCs as high-readiness units even though NATO only required three. In nominating themselves as potential HRF(L) hosts, the candidate countries had staked some of their credibility on their claim of being able to deploy the necessary force elements in the required timeframes. To turn down a country that had nominated part of its armed forces as an HRF(L) would have been the equivalent of saying that that particular country was not as capable as it thought it was. This would have been a devastating political blow to the country in question, something that would have certainly become public knowledge and that might have soured public opinion toward NATO in the country in question. As one US military officer involved in NATO policy asked rhetorically, after member states had gone to all the trouble of nominating themselves as HRF(L) host countries and upgrading their forces as necessary, “was NATO really willing to tell one of these guys that they can’t play?”34 The Military Committee saw how seriously the member states were taking the HRF(L) nomination and certification process – with countries such as Italy, Spain, and Turkey spending the equivalent of 33 Interview with a US government official involved in policy planning, 25 January 2003. 34 Interview with a US military officer, 14 November 2002.
The Impact of Political Bargaining
79
a quarter billion dollars each to upgrade capabilities – and realized there was no way from a political perspective that only three could be chosen – the stakes had become too large to turn down a candidate.35 And so, as member states bargained over which candidates would earn the HRF(L) designation, according to a member of the NATO international staff, “six became palatable.”36 The decision to designate all six candidates as HRF(L)s flew in the face of not only NATO’s own strategy and doctrine, but also of alliance requirements to develop more sustainment and deployment capabilities – the combat support and combat service support necessary to deploy and sustain the HRF(L)s – as a threatbased interpretation might have expected. It should be noted that the alliance was never faced with a single decision point of either designating all six candidates as HRF(L)s or funding sustainment and deployment capabilities. Therefore, it would be disingenuous to view these alliance options as necessarily related or as two alternative courses of action facing the alliance at a specific point in time. Nonetheless, over time the alliance’s decision to designate all six clearly had implications for its sustainment and deployment capabilities insofar as less defense funds would be available for the latter if the largest and most capable allies were devoting their resources to creating and fielding prestigious high readiness forces. The allies’ individual interests in garnering the prestige associated with the HRF(L) designation and the alliance’s interest in pushing at least some portion of its command and force structures away from territorial defense outweighed NATO’s collective interest in fielding sufficient sustainment and deployment capabilities and creating other capabilities that it genuinely needs. In many allied capitals the HRF(L)s were perceived as ‘sexier’ than sustainment capabilities – alliance members saw greater prestige in rapid reaction forces than in air-to-air refueling or strategic airlift, for example, and this had a decisive impact in shaping the allies’ conceptualization of their interests in the bargaining over how many HRF(L)s NATO would designate. Bargaining Over Command Relationships Like NATO’s force structure, intra-alliance bargaining also characterized both the process and outcome of the command structure review. As noted in the previous chapter, SACLANT favored having all of the North Atlantic under its area of responsibility (AOR) for reasons of operational expediency. However, the Spanish government knew its public would not look favorably on having SACLANT assume control of the sea space surrounding Spain’s Canary Islands. This was because Portugal, Spain’s historic rival on the Iberian Peninsula, played a key role as host to a SACLANT subordinate command known as the Iberian Atlantic Command (IBERLANT). The Portuguese had also provided IBERLANT’s commander, a vice admiral who led a Lisbon-based combined staff of roughly 300 military personnel primarily from Portugal, the United States, and the United Kingdom. 35 Interview with a member of the NATO international staff involved in force planning, 16 May 2003; interview with a British military officer assigned to NATO, 9 July 2003. 36 Interview with a member of the NATO international staff involved in force planning, 16 May 2003.
80
Alliance Management and Maintenance
IBERLANT encompassed the area extending from the northern border of Portugal southward to the Tropic of Cancer and approximately 1,150 kilometers westward from the Straits of Gibraltar. Spain’s Canary Islands, as well as Portugal’s Madeira and Azore islands, were part of this massive seaspace. When Spain joined the alliance in 1982, initial expectations were that Madrid would be given a role in IBERLANT, but Portuguese officials soon made it clear that they were unwilling to cede command responsibilities to the Spanish, even on a rotating basis. To the relief of both sides, the Spanish people voted in a 1986 referendum to remain outside the IMC, and so issues surrounding Spain’s role in the NATO command structure were avoided. However, when Spain decided in 1996 to join the IMC, negotiations on who would have command over the seaspace around the Canaries became a major stumbling block to completion of work on the alliance command structure.37 Back and forth negotiations over how to address Spanish political sensitivities while simultaneously maintaining operational efficiency occupied significant time and effort on the part of negotiators. Eventually, agreement was reached to create a relatively small 50-mile buffer zone around the Canaries that was part of SHAPE’s area of responsibility; the rest of the area formerly overseen by IBERLANT remained under SACLANT’s area of responsibility. The outcome here was driven not by threats but by interests and dependencies that were shaped primarily by political factors, in this case by national honor and prestige. Had threats been the driving consideration then it is unlikely that a carved-out buffer zone – an island of SHAPE responsibility in a far larger SACLANT area of responsibility – would have been the result because this solution unnecessarily complicated the command structure and did not represent an efficient use of alliance resources. A rivalry between Spain and the United Kingdom, centered on Gibraltar, had a similar impact on shaping the bargaining that occurred over command relationships.38 Taken from Moorish forces by Spain in 1462, Gibraltar was later captured by an Anglo-Dutch fleet in 1704 during the War of the Spanish Succession. Spain formally ceded the territory to Great Britain in 1713, but Madrid occasionally launched military attacks to regain Gibraltar throughout the remainder of the 18th century. As the British Empire grew, and as the United Kingdom became a global power with global interests, so too did the importance to London of retaining foreign military bases such as Gibraltar. Nonetheless, Madrid continued to exert pressure wherever possible to regain Gibraltar – for example, the border crossing point from mainland Spain into Gibraltar was closed from 1967 to 1985.
37 Interview with a retired US military officer previously assigned to NATO, 15 March 2004; interview with a US military officer involved in the negotiations, 27 April 2004; and correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004. 38 Interview with a US military officer involved in the negotiations, 27 April 2004; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004; an interview with a retired British military officer assigned to NATO, 11 June 2004.
The Impact of Political Bargaining
81
More recently, Madrid offered Gibraltar autonomous-region status within Spain, and London presented a plan to maintain joint Anglo-Spanish sovereignty, but Gibraltarians have consistently, and fervently, rejected efforts to change the status quo – the most recent effort to change the status of Gibraltar’s sovereignty was voted down by a resounding 99 percent in late 2002. As for Gibraltar’s military role, the British garrison was withdrawn in the early 1990s, but the Royal Navy has continued to use Gibraltar as an important overseas base, especially for power projection into the Mediterranean. Because of continuing Spanish sensitivities toward the issue, any sort of doctrinal or structural arrangement that involved Spanish military officers taking part in a combined NATO command involving British officers and dealing with Gibraltar was impossible. Just as any kind of shared responsibility for Gibraltar’s security was rejected, so too were plans for either party to have such responsibility alone. The two sides were at loggerheads over the issue – indeed, the rivalry was so heated on this point, as a retired British officer involved in the negotiations noted, “the Spanish did not even want to allow the use of the phrase ‘Strait of Gibraltar’ in official NATO documents.”39 Finally, after much negotiation, a compromise was reached that essentially took Gibraltar off the negotiating table – London and Madrid agreed that the British base at Gibraltar would not have any NATO function whatsoever. In this case, the strong interests of both Britain and Spain led to a bargaining outcome in which neither got what it wanted but neither was forced to give the other side what it wanted either. As above, had threats been the deciding factor, it is likely that Gibraltar – a strategic location by any measure and certainly potentially useful in allowing the alliance to control the western Mediterranean – would have figured into alliance command structures. The strong interests of both Spain and the United Kingdom led to a negotiated solution that was different from what a threat-based solution might have yielded. Similar issues arose between Greece and Turkey over control of the Aegean Sea in the revised doctrine and structures, especially since there were allied military commands located at Larissa, Greece and Izmir, Turkey that could each lay claim to command and control of allied forces in the Aegean.40 The rivalry between Greece and Turkey is well-documented and has often been very confrontational – the most recent flare-up occurred in 1996 when Greek and Turkish naval forces almost started a war over an uninhabited Aegean island. The NATO alliance has typically been cited as a tool used to dampen the often-heated relationship between Athens and Ankara. The dispute over the command and control of the Aegean proved particularly challenging, anchored in years of disagreement between the two rivals.41 Indeed since 39 Interview with a retired British military officer assigned to NATO, 11 June 2004. 40 Interview with a retired US military officer previously assigned to NATO, 15 March 2004; interview with a US military officer involved in the negotiations, 27 April 2004; and correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004. 41 The Greek-Turkish rivalry also extended to issues as seemingly minor as terminology. The Greek name for the narrow waterway connecting the Black Sea to the Sea of Marmara is the Bosporus strait. However, the Turks use another name – Boğaziçi – and the two sides could not agree on how to refer to the strait in official NATO documents, requiring additional
82
Alliance Management and Maintenance
the end of the Cold War, there had been no functioning sub-regional joint command in the eastern Mediterranean, which was highly unusual given NATO’s interest in and operations in the nearby Balkans. Efforts to establish such a command in Larissa faced opposition from Turkey, while the existing NATO facility at Izmir suffered from funding problems linked to Greek obstructionism. Following lengthy back-andforth negotiations, and some creative diplomacy on the part of the alliance, Greece and Turkey agreed to drop their objections to each others’ headquarters based on the understanding that during peacetime, the commands at Larissa and Izmir would have no clearly established areas of responsibility. For peacetime military operations in the Aegean, such as exercises, the NATO regional command in Naples – the higher headquarters for both Larissa and Izmir – would be in charge. Additionally, it was agreed to employ equal numbers of alliance personnel at Larissa and Izmir and that the leadership teams at each command would be mirror images of each other from a national perspective – the command at Larissa is led by a Greek general, with an American general as deputy commander and a Turkish general as chief of staff, while the command at Izmir is led by a Turkish general, with an American general as deputy commander and a Greek general as chief of staff. Even in reaching the compromise, the Greeks still felt it necessary to characterize the agreement as a victory over the Turks, indicating the rivalry was alive and well: The result of the new arrangements is that Greece…is on a completely equal footing with the other Mediterranean member-states who each also have one headquarters, in Verona, Madrid, and Izmir. Thus Turkey, which hitherto operated three headquarters, will now have only one.42
A threat-based solution to this issue almost certainly would have involved a more clear-cut division of the area of responsibility than what the allies agreed to. Operationally, drawing dividing lines during peacetime is likely to be far easier than trying to define them during a crisis. Instead of a solution based on threat factors, the negotiated solution that the Greeks and Turks arrived at was similar to the case of Gibraltar discussed above – neither side got exactly what it wanted, but neither was forced to capitulate entirely to the other side. The bargaining that occurred in this situation reflected certainly strong interests on the part of both Greece and Turkey in the object of the negotiation – military authority over the Aegean. So strong were the interests of the parties involved that neither side had a clear bargaining advantage, and this became reflected in the negotiated solution. In addition to command relationships, the precise allocation of senior military officers to particular command structures was subject to significant intra-alliance bargaining and played a major role in delaying the finalization of the command and force structure arrangements. Most of the HRF(L) host nations wanted the United States to fill the key positions on their HRF headquarters staff, such as operational
efforts at negotiation. Interview with a retired British military officer assigned to NATO, 11 June 2004. 42 “Greece Gains Equality in New NATO Military Structure,” press release by the Greek Embassy to the United States, December 1997.
The Impact of Political Bargaining
83
43
or intelligence slots. There were primarily two reasons for this. First, the United States is viewed as having the most capable operations and intelligence assets in the alliance, and it was believed that Washington would contribute or make available its operational and intelligence assets to any HRF(L) in which Americans occupied the lead operations or intelligence billets. Second, and more generally, the HRF(L) candidates believed they would have greater influence with the US military – the most important military force in the alliance – by leveraging the formal and informal lines of communication created when an American staff officer occupies a billet in an allied command. If other countries also wanted to contribute to a particular HRF(L) headquarters, the host nation would ask them to wait until it received an answer from Washington on which billets US officers would fill. Meanwhile, because Washington wanted the European allies to fulfill as many of the requirements as possible in the interest of burden-sharing, the United States would wait to see what the host country and other allies were willing to contribute before assigning any US officers. This rather circular bargaining process – driven primarily by American interest in promoting burden-sharing, a limited degree of American dependence on Europe to provide some minimal contribution to collective security, European interest in prestige, and a limited degree of European dependence on American advanced technology and capabilities – greatly slowed down efforts to conclude a command and force structure. A rational response to threats likely would have led to a more efficient allocation of alliance resources. Similar bargaining considerations affected member states efforts to retain or gain command positions for themselves. The so-called flags-to-post issue – deciding which country’s flag officers (generals or admirals) would be assigned to lead specific allied commands – delayed final approval and implementation of NATO’s command structure changes as embodied in MC 324 from 1997 until late 1999.44 The governments of the member states perceived significant political prestige, especially relative to each other, in having their flag officers holding senior positions within the NATO military command structure. Moreover, many within the alliance believed that having one of their general or flag officers in a NATO billet in a foreign country provides a means of influencing the foreign country’s military and government. On a related point, many of the member states vying for command structures and flag officer billets also realized that such efforts would only become more difficult in an enlarged alliance. To the extent they were able to overcome some of the factors causing delays in the negotiations as noted above, members felt some incentive to wrap up as many of the command structure and flags-to-post issues as they could prior to March 1999, when three new countries would join the alliance – “there was 43 Interview with a US government official, 19 February 2003. 44 Interview with a US military officer assigned to the European Command, 14 November 2002; an interview with a US government official responsible for military planning, 19 February 2003; an interview with a US military officer assigned to the European Command, 10 April 2003; correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004.
Alliance Management and Maintenance
84
pressure to finish it before new countries came in and further complicated things,” according to a retired US military formerly assigned to SACLANT and involved in the deliberations.45 Waiting until after the 1999 enlargement round would have required dividing command structures and flag officer billets among 19 instead of just 16 – and dividing such benefits among 16 was difficult enough given the drawdown in alliance commands from 65 to 20. In this way, the pre-1999 members were able to retain for themselves as many of the command structures – and flag officer billets to staff them – as possible. Clearly, threats were not motivating this behavior. Instead, political bargaining – shaped by strong member interest in obtaining and maximizing their prestige – among members and between members and the alliance explains NATO’s behavior here. Summary It is clear that security threats are necessary but not sufficient for explaining NATO’s post-Cold War doctrine and hence its structures. Relying on a threat-based interpretation to explain alliance management during this period of change leaves significant gaps, especially in terms of exposing what the alliance should have devoted its resources to but did not. In examining both the process and outcomes of NATO’s doctrinal reviews, political bargaining factors help to explain parts of the story that changes in the threat environment cannot, including why the alliance misspent so much time and money before achieving in 2003 some of the doctrine and structures it needed to fulfill its strategy of 1991. The outcomes of the bargaining that occurred among members of the alliance were determined by varying levels of interest and dependence on the part of the allies and the alliance as a whole. By examining some of the elements that combined to shape interest and dependence – particularly the interest on the part of many allies in garnering prestige, tangible rewards to placate domestic audiences, and defense budget justifications, and the moderate dependence of the alliance on the contributions of key allies to collective defense – it is possible to gain a clearer understanding of the bargaining that occurred following the end of the Cold War and that dramatically impacted alliance doctrine and structures. This chapter showed that together, political factors and security threats account for NATO’s post-Cold War doctrinal and structural changes far better than threats alone, which in turn sheds light on how the alliance manages itself through a period of great change. The next chapter will test these findings on alliance management and maintenance by examining how NATO has responded to the most recent watershed in alliance security, the terrorist attacks of 11 September 2001.
45 Interview with a retired US military officer, 15 March 2004.
Chapter 7
The Alliance’s Response to Terrorism The preceding chapters have shown that alliance management and maintenance, even during periods of great change in the security environment, are subject to the influence of political factors – particularly domestic political factors such as electoral concerns, economic or budgetary issues, or the need for national prestige and honor – that manifest themselves through intra-alliance bargaining and that outstrip security threats in terms of explanatory and descriptive power. These findings can be further tested by examining how NATO has managed itself through a more recent watershed period in alliance security – the aftermath of the September 11 attacks. In the immediate aftermath of the September 11 terrorist attacks, NATO responded with a number of supportive measures. Foremost among these was the invocation of Article 5, when NATO declared the attack on the United States to be an attack on the entire alliance. This was particularly noteworthy for the alliance, as it was the first time that Article 5 had ever been invoked. In addition, the alliance took a number of other steps. For example, NATO accelerated efforts to enhance intelligence sharing. Although many of the allies had shared intelligence for decades, the ‘sharing’ that occurred was not systematic, it occurred on only a very limited, often bilateral basis, and it remained focused primarily on military operations to the exclusion of law enforcement needs.1 Moreover, the intelligence that was shared usually consisted of processed intelligence products, not the raw materials that underpin such analyses. Improving processes here was recognized in the immediate aftermath of the 9/11 attacks by then-Secretary General Lord Robertson as an area for improvement, one in which artificial barriers needed to come down.2 Additionally, the alliance agreed to provide blanket overflight clearance and access to ports and airfields on alliance territory, to offer assistance to states threatened as a result of their support for US efforts, to deploy alliance standing naval forces to the Eastern Mediterranean, and to deploy alliance AWACS aircraft to the United States.3
1 For a useful overview of this subject, see “The Changing Face of Intelligence: NATO Advanced Research Workshop – Report,” The Pluscarden Programme for the Study of Global Terrorism and Intelligence, St. Antony’s College, Oxford, 9–10 December 2005. 2 Lord George Robertson, “An Attack on Us All: NATO’s Response to Terrorism,” speech delivered at the National Press Club, Washington, DC, 10 October 2001. Accessed at . 3 “September 11 – 18 Months On: NATO’s contribution to the fight against terrorism,” NATO Factsheet, 1 April 2003.
86
Alliance Management and Maintenance
Following these immediate post-9/11 efforts, the alliance began Operation Active Endeavour in October 2001. Active Endeavour entails the deployment of some of the alliance’s standing naval forces to the Eastern Mediterranean to conduct maritime interdiction and monitoring against ships suspected of carrying contraband. In March 2003, the operation expanded to providing escort of some non-military ships through the Strait of Gibraltar.4 Although all of these actions were substantial, the alliance recognized that it could, and should, do more against a threat that knew no boundaries and often had no return address. Following the December 2001 foreign ministerial, the North Atlantic Council released a communiqué in which it stated that in order to better defend against the threat of terrorism the alliance would examine ways to adapt and enhance its military capabilities.5 The Council directed the Military Committee to begin developing an alliance strategy for dealing with terrorism. This represented the first indication by the alliance that the changed threat environment demanded a corresponding change in the alliance’s doctrine and force structure, something far more substantive from an alliance management perspective than ordering AWACS to North America since it would involve changing fundamental aspects of how NATO does what it does. This realization and the accompanying directives appear to have been driven by threat-based factors, just like NATO’s initial reaction to the end of the Cold War. In both cases, the alliance – at least at the political level – quickly recognized a changed threat environment and subsequently directed actions to remedy perceived deficiencies in how the alliance provides security for its member states. The Political Guidance Before work on a revised doctrine and force structure for combating terrorism could be started, the alliance’s political leadership decided it needed to provide the MC political guidance that would set the parameters for the strategy. The political guidance was developed by NATO’s international staff during the first half of 2002. At that time, just months after the 9/11 attacks, the authors of the guidance correctly read the mood of the alliance as one in which members were galvanized and ready for bold steps. The authors therefore incorporated some far-reaching provisions and language in the guidance. For example, the international staff proposed, and the alliance endorsed, language in the guidance that NATO should simply be prepared to deploy ‘as and where required,’ effectively dispensing with the diplomatic phrasing traditionally used to refer to the out-of-area concept and steamrolling past the reluctance on the part of some member states to interpret that concept broadly.6 This decision was reaffirmed during the May 2002 Reykjavik Defense Ministerial, when
4 “Operation Active Endeavour,” NATO Briefing, December 2003. 5 “NATO’s Response to Terrorism,” statement issued at the Ministerial Meeting of the North Atlantic Council held at NATO Headquarters, Brussels, 6 December 2001. 6 Interview with a NATO official involved in alliance counterterrorism policy, 10 February 2004.
The Alliance’s Response to Terrorism
87
ministers declared that, “NATO must be able to field forces that can move quickly to wherever they are needed.”7 Specifically, the political guidance provided by the NAC consisted of five main principles. First, the NAC stated that NATO’s actions against terrorism must have a sound legal basis and must conform to relevant provisions of the United Nations Charter. Second, the alliance’s actions should help deter, defend, disrupt, and protect against terrorist attacks or threat of attacks regardless of whether they are directed against populations, territory, infrastructure, or forces of any member. Third, if requested, the alliance should be prepared to act in support of the international community’s efforts against terrorism. Fourth, the alliance should be capable of providing assistance to national authorities following a terrorist attack, especially those attacks that involve the use of chemical, biological, radiological, or nuclear weapons. Finally, the alliance should operate on the assumption that it is preferable to deter or prevent terrorist attacks rather than to deal with their consequences and the attendant loss of life and property, and so NATO should be prepared to deploy as and where required to meet the needs of particular circumstances and, if necessary, to preempt. The political guidance represented a melding of some of the key foreign and security policy tenets of the allies at the time and was therefore a product of some type of intra-alliance bargaining. The outcome of the bargaining appears to have been one in which most allies got what they wanted though, which could have set the allies up for confrontation and disagreement further down the road. For example, many of the European allies, but particularly Germany and several of the smaller allies, favored a United Nations seal of approval on any alliance military action – preferably in the form of a Security Council resolution authorizing such action, but at least in terms of conformity with principles of the UN Charter. This sort of international, multilateral validation is appealing to Europeans for several reasons, but especially because it provides everyone with a voice.8 For Germany specifically, the primacy of the multilateral approach was validated during the closing years of the Cold War, when engaging former adversaries yielded extremely positive results. So accepted has the concept become in Germany and like-minded states that it is viewed as not only the primary tool for achieving foreign policy goals but as a worthwhile end in itself. Other European governments, such as the French, take a more pragmatic view, preferring multilateral mechanisms because they can provide international legitimacy and domestic political cover when governments might be otherwise reluctant to commit to collective military action. In another example, the concept of preemption – spelled out in the last of the five principles – was the subject of significant focus from the Bush administration following 9/11. Just days after the attacks, President Bush addressed a joint 7 Final Communiqué, Ministerial Meeting of the North Atlantic Council Held In Reykjavik on 14 May 2002. 8 For a discussion on the views of various European governments toward multilateralism, see Joachim Krause, “Multilateralism: Behind European Views,” The Washington Quarterly, vol. 27, no. 2 (Spring 2004), pp. 43–59.
88
Alliance Management and Maintenance
session of the US Congress in which he announced that the United States would consider not just terrorists enemies but also the states that harbor them: “From this day forward, any nation that continues to harbor or support terrorism will be regarded by the United States as a hostile regime.”9 This represented the first rhetorical manifestation of what would become a clear broadening of the preemptive war concept – one long accepted under international law and predicated on the existence of a known threat – to include preventive war as well. The extent of the broadening in administration thinking was revealed further in June 2002, when President Bush announced during a West Point commencement speech: If we wait for threats to fully materialize, we will have waited too long… the war on terror will not be won on the defensive. We must take the battle to the enemy, disrupt his plans, and confront the worst threats before they emerge… our security will require all Americans to be forward-looking and resolute, to be ready for preemptive action when necessary to defend our liberty and to defend our lives.10
Although he used the word ‘preemptive,’ what President Bush was speaking of was clearly preventive war – that is, war waged before threats materialize. By the fall of 2002, this broadened conceptualization of preemptive war had worked its way into the US National Security Strategy (NSS). In contrast to past strategies, in which preemption was perhaps an unstated principle, the new NSS explicitly noted the necessity of preventive preemption when warranted by particularly dire circumstances: We must adapt the concept of imminent threat to the capabilities and objectives of today’s adversaries.… The greater the threat, the greater is the risk of inaction – and the more compelling the case for taking anticipatory action to defend ourselves, even if uncertainty remains as to the time and place of the enemy’s attack.11
Even though the NAC-approved principles represented a sort of compromise between those in NATO that wanted to tie alliance military action to UN approval and those that wanted to push the alliance toward a more proactive, even aggressive stance that included preemptive action, the last of the principles proved to be too much for the alliance to swallow in the long run. So forward leaning was the formulation of this last point that many months later, after the member states had more time to digest the full implications of their political guidance, the alliance as a whole would judge this language a little too ambitious and would therefore pull back from such a bold
9 George W. Bush, “Address to a Joint Session of Congress and the American People,” Washington, DC, 20 September 2001. 10 George W. Bush, “President Bush Delivers Graduation Speech at West Point,” West Point, New York, 1 June 2002. 11 The National Security Strategy of the United States 17 September 2002.
The Alliance’s Response to Terrorism
89
12
statement in the Military Committee’s strategy. In particular, one NATO official involved in developing the alliance’s counterterrorism strategy and doctrine noted, “the French, Belgians, Greeks, and some others have tried to walk away from [it] because they only subsequently realized how far the alliance had gone in the heat of the moment.”13 MC 472 and MC 161 With the political guidance in hand, the Military Committee set about developing the strategy, known as the Military Concept for Defense Against Terrorism, or MC 472. It outlined in broad terms how the alliance would respond to the threats posed by terrorism. The strategy referenced a NATO terrorism threat assessment, known as MC 161,14 which concluded: •
• •
Sources of terrorism include religious extremism or economic, political, social, or demographic crises stemming from unresolved conflicts or emerging ideologies; State sponsorship of terrorism remains a potential threat even though it appears to be declining; and, Terrorist groups are expected to continue using conventional weapons and may seek to obtain and use weapons of mass destruction (WMD).
To respond to this threat, the Military Concept identified four broad roles for the alliance, and outlined some of the concrete alliance actions that might support each:15 1. Anti-Terrorism – defensive measures to reduce vulnerabilities to attack. Alliance actions here might include intelligence sharing. 2. Consequence Management – employed after an attack has taken place to mitigate the destructive effects. Tangible steps here might include robust force generation in the immediate aftermath of an attack or disaster relief coordination. 3. Counterterrorism – offensive military action in which NATO either plays a supporting role or is in the lead. In either case, the goal is to reduce terrorists’ abilities to carry out attacks. Alliance actions might include back-filling national requirements, deploying forces in support of coalition operations, or providing host nation support or logistics assistance. 12 Interview with a NATO official involved in alliance counterterrorism policy, 10 February 2004; and interview with a NATO official involved in alliance counterterrorism policy, 21 March 2005. 13 Interview with a NATO official involved in alliance counterterrorism policy, 10 February 2004. 14 The NATO terrorism threat assessment is updated at least annually, and has now become part of the broader alliance threat assessment process. 15 “NATO’s Military Concept for Defense Against Terrorism,” NATO Factsheet, updated 15 December 2003.
Alliance Management and Maintenance
90
4. Military Cooperation – efforts to synergize military and civilian operations. Examples might include working with police, customs, border security, and other internal security services, as well as international organizations such as the United Nations. In laying out these roles, the Concept established, if sometimes only implicitly, the requirements for new alliance doctrine and additional or modified force structures. For instance, the Concept called for identification of procedures for measuring force protection requirements, for clear arrangements on interacting with civilian authorities, and for streamlining alliance decision-making in crisis situations – all of these are doctrinal requirements. Likewise, the Concept stated that the alliance would require better intelligence collection capabilities, forces that are more deployable and at a higher state of readiness, more guided munitions in order to minimize the risk of collateral damage, and improved WMD defense capabilities – all of these are command or force structure requirements. Clearly, the alliance not only recognized that it needed modified or additional doctrine and structures, but it has also identified in a general sense what those doctrines and structures should consist of. The Concept was approved by NAC ambassadors and then endorsed at the NATO summit meeting of heads of state and government held in November 2002 in Prague. Implementation of the Military Concept against Terrorism Although NATO’s response to terrorism is in some ways still a work in progress, the alliance’s record of accomplishing the goals it set out for itself is, so far, a mixed one. Some successes seem evident – such as the alliance’s work in Afghanistan and its development of the NATO Response Force (NRF). However, even in these instances of apparent success the alliance continues to face major hurdles in achieving what it set out to do. In other contexts the alliance has yet to take up its own call for new doctrine, command arrangements, and force structure. NATO has recognized that a new threat exists, but has been hampered in its response to that threat by the lack of appropriate doctrine and subsequent processes and structures.16 Why this is the case – that is, why the alliance has not achieved greater success – reaffirms many of the insights into alliance maintenance and management outlined earlier in this book. In some cases, there are significant similarities to alliance actions (or inaction) from the mid- and late-1990s – as then, political bargaining has intervened, and along with it varying levels of dependence and interest, accounting for much of the explanation for NATO’s response to the fight against terrorism that geopolitical threats alone cannot. Anti-Terrorism The alliance has addressed anti-terrorism along two distinct lines of operation – force protection and intelligence sharing, although admittedly there are force 16 Interview with a NATO official involved in alliance counterterrorism policy, 21 March 2005.
The Alliance’s Response to Terrorism
91
protection implications in each of the four broad roles the alliance addresses through MC 472. Prior to the terrorist attacks of 9/11, the alliance had no force protection doctrine. Through its Allied Command Transformation, NATO developed a draft doctrine on this subject, but as of late 2006 the document had yet to be approved. Despite this, NATO has conducted combined force protection experiments to validate methodologies and technologies, begging the question of whether the cart – structures and forces – was not in front of the horse – the doctrine necessary to drive structures and forces. On intelligence cooperation, the alliance has only very recently achieved some success. This is ironic considering calls for increased intelligence sharing have become somewhat routine in allied summit statements.17 Nonetheless, such requests have been followed with only minimal action, and although some improvements have been made, there is significant room for further coordination and sharing of intelligence. Bilaterally, intelligence cooperation between the United States and many of its European allies dramatically improved following 9/11. The alliance hoped to achieve similar success in a multilateral context when it established the Terrorism Threat Intelligence Unit (TTIU) in 2003, but the mission of the TTIU was somewhat limited, as it was directed to analyze terrorist threats to the NATO organization. Following the Istanbul summit in 2004, allies renewed efforts to increase intelligence sharing, but progress was slow and difficult to achieve for two reasons. First, although allies remain willing to share intelligence bilaterally, they are often less than comfortable sharing information in a multilateral forum of 26 countries. Although all the allies subscribe to the same standards of classified handling methodologies, in the words of one NATO official, “at 26, secrets become less secret.”18 Second, with increased emphasis since 9/11 on the need for greater human intelligence, or HUMINT, allies have been reluctant to share information because they fear revealing sensitive sources and methods. As a result, it was not until late 2006 that the alliance had any sort of integrated intelligence analysis staff – in October of that year, the alliance opened its Intelligence Fusion Center (IFC) at the Royal Air Force Base Molesworth in the United Kingdom. However, at the time it was formally opened, the IFC did not even have half of its full complement of personnel. Moreover, for intelligence to be ‘fused,’ there must exist both a will and an ability, and on both scores NATO still has significant progress to make – in some cases member state officials have proven less than willing to share intelligence even in operations such as that in Afghanistan, and in other cases differing national computer systems or other intelligence hardware have proven incapable of communicating with each other.19 NATO is likely to address successfully these challenges over time, but in the meantime alliance forces in Afghanistan and elsewhere may suffer the practical consequences. 17 See the Prague Summit Declaration, 21 November 2002; the Istanbul Summit Communiqué, 28 June 2004; and the Riga Summit Declaration, 28 November 2006. 18 Interview with a member of the NATO international staff who specializes in anti- and counter-terrorism issues 27 July 2006. 19 Sebastian Spencer, “NATO Scrambling to Improve Intel Sharing, But Obstacles Remain,” Inside the Pentagon, 19 October 2006.
92
Alliance Management and Maintenance
Consequence Management The alliance identified streamlined force generation efforts as one area in which it could contribute to consequence management; however this remains a significant hurdle for NATO as noted earlier in this book. More directly, the alliance has made no headway since 9/11 in terms of improving the way in which it generates forces. This, despite the fact that the Secretary General of NATO, Jaap de Hoop Scheffer, has been one of the most vocal critics of NATO’s force generation process, noting that, “There is an obvious disconnect between our longer-term force planning system and the way we generate forces for a particular operation.”20 Here, broadly speaking, the United States is in many ways a victim of its own success. During the 1990s, Washington expended significant diplomatic energy to push the alliance into accepting out-of-area responsibilities and missions, but for several reasons most European allies resisted this, as noted earlier in this book. In the end though, the United States largely achieved its objective of getting the alliance to take on greater security responsibilities – to include military missions – outside of the territory of its member states. Alliance operations in Bosnia, Kosovo, Afghanistan, Darfur, Iraq, and elsewhere are testament to this. It would appear as if the more reluctant European allies lost the bargaining game that was the out-of-area debate, and certainly to some degree this is true. However, a more nuanced interpretation of events would hold that the European allies simply made a tactical retreat and relocated their line of resistance. To explain further, instead of objecting to a particular out-of-area mission supported by Washington simply because it is not within NATO’s traditional area of operations, the most reluctant European allies can afford to engage in a kind of disingenuous cooperation – they can acquiesce to American desires to push the alliance into the mission, but they can each simultaneously avoid a national commitment of forces thanks to the nature of the force generation process. By essentially taking advantage of the force generation process when their support of the mission in question may not go beyond the rhetorical, some allies thereby give the lie to allied ‘consensus,’ whether intentional or not. The alliance’s enlargement ironically has made this problem worse, as individual European allies – who now think of themselves as one of 23 European allies with military forces, vice a mere 13 prior to the two rounds of enlargement – feel even less compelled to provide for the common defense and are even more apt to freeride on the collective security (and force contributions) provided by other allies. The result of course is that NATO’s political appetite appears much larger than its capability to commit forces, but the risk in this sort of analysis is that it can conflate force generation problems with force availability issues – in analyzing member state decisions, one must be careful to separate instances of shaky will from instances in which the deployable forces necessary to fulfill the willpower simply do not exist. This can be a tricky exercise, since member states are sometimes less than completely transparent about what forces are available for alliance action and 20 Jaap de Hoop Scheffer, “NATO’s Istanbul Summit: New Mission, New Means,” speech at the Royal United Services Institute, London, 18 June 2004.
The Alliance’s Response to Terrorism
93
because when they do provide such information to the alliance it is often dated. Excluding force availability issues, the evidence seems to suggest that when the interests of the member states in solving a particular crisis with the application of military force are strong – and hence when political cohesion among the allies is strong and their bargaining win sets coincide more closely – force generation is easier than when the interests of the member states are somewhat weak – and hence political cohesion is also weak and bargaining win sets are more disparate. Where possible, the alliance has attempted to address some of its force generation challenges. For example, NATO has sought to change the way in which it funds operations, reasoning that paying for certain activities from common budgets may enable more allies to contribute forces. In the alliance’s current ‘costs lie where they fall’ system, the allies each foot the bill for their own forces. This is particularly problematic in an era of shrinking defense budgets – as noted earlier in this book, NATO’s own figures show that members are spending less and less, as a percentage of GDP.21 More recently, of the 23 European allies with military forces in 2004, 17 spent less than 2 percent of their GDP on defense, even though all agreed to this target. Of the six allies – Bulgaria, France, Greece, Romania, Turkey, and the United Kingdom – that did spend 2 percent or more each on defense, Bulgaria (60%), France (57%), Greece (77%), Romania (50%), and Turkey (49%) each spent significant percentages of their defense budgets on personnel.22 That leaves the United Kingdom as the only European ally that spends more than 2 percent of GDP on defense and that does not devote a majority (or near majority, as in the case of Turkey) of its budget to personnel costs. As a broad generalization, it is difficult to fund military operations in far away lands if the defense budget simply is not large enough or if most of that defense budget is devoted to personnel costs – for most allies, the cost of strategic lift alone is simply too great to allow them to participate in far away missions. Efforts by 15 NATO allies to collectively procure C-17 strategic lift aircraft from Boeing should help to address some of the alliance’s shortcomings in this area, at least in terms of the operational implications.23 By coming together in a consortium through NATO, the allies hope to leverage their limited financial means to acquire capabilities – estimated at roughly three or four C-17s, the first of which should be delivered sometime in 2007 – that they could not afford individually.24 Additionally, and among other proposals for modifying how the alliance pays for 21 “NATO-Russia Compendium of Financial and Economic Data Relating to Defence: Defence Expenditures of NRC Countries (1985–2005),” Tables 3 and 5, NATO Press Release, 9 December 2005. 22 In contrast, the United States spent 34 percent of its defense budget on personnel costs. 23 The C-17 consortium is comprised of Bulgaria, the Czech Republic, Denmark, Estonia, Hungary, Italy, Latvia, Lithuania, the Netherlands, Norway, Poland, Romania, the Slovak Republic, Slovenia and the United States. The consortium also includes Sweden, which is not a member of NATO. 24 Press conference with NATO Secretary General Jaap de Hoop Scheffer and the Assistant Secretary General for Defence Investment, Marshall S. Billingslea on NATO moving
94
Alliance Management and Maintenance
operations, in November 2006 NATO Secretary General de Hoop Scheffer called for common funding of certain costs – particularly strategic lift – of short term NRF deployments.25 However, this subject was not discussed formally during the 2006 summit in Riga, reflecting the lack of a bargaining consensus among the allies on the efficacy of this proposal. In addition to efforts in financial management and collective acquisition, which address the operational implications of force generation challenges, the alliance has pursued other steps to ameliorate more specifically the force generation process. For example, the Deputy Supreme Allied Commander Europe (DSACEUR) has been tasked with leading and managing the force generation process on a day-to-day basis, placing responsibility for leadership on this topic within the office of a single individual. Additionally, NATO instituted a Global Force Generation Conference effort in order to streamline the generation of forces for operations in the Balkans and Afghanistan. The first of these annual conferences occurred in November 2004; they have been viewed as useful for at least providing alliance military officials with greater insight into available national forces. However, neither these conferences nor the appointment of the DSACEUR as the primary lead for force generation efforts has addressed the core challenge in generating forces, at least in the current system – generating the political willpower necessary for national force contributions. Commonly funding significant portions of operations for which the alliances seeks to generate forces would be an important step in addressing this challenge when the lack of willpower is rooted in a lack of funding and related domestic budget battles, but it seems the force generation process – and any contribution it might make to consequence management – will continue to be held hostage to political bargaining within the alliance. Counterterrorism With regard to the alliance’s stated goal of reducing terrorists’ abilities to carry out attacks, NATO has made major progress in terms of its role in Afghanistan.26 There, the alliance’s counterterrorism role has two key elements. First, the alliance leads the International Security Assistance Force (ISAF), consisting of roughly 32,000 troops whose mission is to assist the Afghan government in extending security and stability throughout the South Asian country. Second, the alliance has developed a security cooperation program with Afghanistan that provides substantial training and education on defense reform, defense institution-building, and the military aspects of security sector reform. On both counts, the alliance has generally seen success. Starting with a small military presence initially in Kabul, the alliance expanded its responsibility to the north in 2004, to the west in 2005, to the south in mid-2006, to acquire C-17 Strategic Airlift Aircraft, 12 September 2006. 25 Jaap de Hoop Scheffer, “Global NATO: Overdue or Overstretch?” speech at the Security and Defence Agenda Conference, Brussels, Belgium, 6 November 2006. 26 ISAF, under NATO leadership, assumed responsibility for countering Taliban forces throughout Afghanistan by November 2006.
The Alliance’s Response to Terrorism
95
and to the east in late 2006. Thanks in part to ISAF and the more than two dozen NATO-led Provincial Reconstruction Teams (PRTs), large swaths of Afghanistan, especially in the north and west, are now mostly peaceful and stable; average annual economic growth since 2003 has been nine percent; and the GDP has grown from roughly $4.7 billion in 2003 to an estimated $7.2 billion in 2006. The PRTs, usually a combination of international military and civilian personnel based in provincial areas of Afghanistan, are responsible for supporting and facilitating the extension of the authority of the Afghan central government, security sector reform, and reconstruction. More specifically, the purpose of the NATO military personnel assigned to each PRT is to participate in patrolling, conduct liaison with the local population, and both train and operate alongside local Afghan security forces. With regard to building the military capacity of the Afghans, former SACEUR Gen. James Jones termed the reform of the Afghan National Army (ANA) the most successful pillar in the west’s reconstruction effort in Afghanistan. With NATO’s assistance, the ANA was, by the end of 2006, roughly half way to its goal of 70,000 soldiers. Moreover, by that same time the ANA had begun to successfully engage and defeat Taliban forces on the battlefield, albeit with allied assistance. The alliance accomplished these things in large part by deploying small, embedded training teams – so-called Operational Mentoring and Liaison Teams (OMLTs) – that worked sideby-side with Afghan army forces during military operations and for the purpose of training. Despite these successes though, the alliance’s efforts in Afghanistan remain plagued by major shortcomings – indeed, following the resurgence in Taliban activity throughout 2006, some observers continue to hold that the entire mission is at risk of failure.27 The greatest challenge has been simply fulfilling the military statement of requirements developed by NATO and agreed to by member state military representatives. Prior to any alliance mission, a combined joint statement of requirements (CJSOR) is developed that outlines the mission essential elements in terms of the specific number and types of units and equipment. Through each of the four stages of its expansion throughout Afghanistan, NATO countries have routinely come up short in fulfilling the CJSOR for ISAF. In September 2006, just after NATO conducted Stage 3 of its expansion in Afghanistan to the south and was preparing for Stage 4 to the east, the Secretary General called attention to the still unfulfilled Stage 3 CJSOR in what amounted to a highly unusual public browbeating.28 As the leaders of allied governments gathered in Latvia for the Riga summit in November 2006, the CJSOR for Stage 4 remained unfulfilled as well, at roughly 85 percent of the stated requirement, and Secretary General de Hoop Scheffer again called for 27 Ronald D. Asmus and Richard C. Holbrooke, “Re-Reinventing NATO,” paper presented at the conference on “Transforming NATO in a New Global Era,” held in conjunction with the NATO summit in Riga, Latvia, 27–29 November 2006. 28 Joint Press Conference by the Secretary General of NATO and the Minister of Defense of Slovenia, Portoroz, Slovenia, 28 September 2006.
96
Alliance Management and Maintenance
member states to make good on their commitments.29 Usually, the elements left unfulfilled include combat support, such as rotary lift assets, logistics support, and reconnaissance forces.30 The challenge of generating the forces to fulfill a CJSOR runs a familiar pattern of a classic collective action problem. Following the release of the CJSOR, member states make decisions on what forces or assets they will contribute. Typically, the United States waits to announce its contribution until the Europeans announce theirs, in the hopes of soliciting as many European forces as possible. When those announcements are made, Washington routinely is dissatisfied with the European contribution – mostly because it fails to completely fulfill the CJSOR – and so it engages in arm-twisting. As the deadline for contributing forces approaches, and when behind-the-scenes cajoling or public browbeating fails to elicit the required forces, the United States will usually offer some military assets in order to bridge at least some of the gap and prevent truly debilitating CJSOR shortfalls from endangering the mission at hand. The European allies know this is likely to happen, and so they are incentivized to hold out for as long as possible before contributing any more of their own assets. In this scenario, the bargaining on the part of the allies is of a very traditional nature in which allies attempt to pass the buck and avoid commitments, versus the style of bargaining discussed in most of this book wherein allies actively seek alliance responsibilities and designations for the associated prestige or other perceived benefits. Nevertheless, the factors that explain the outcome of pursuit-of-benefits bargaining – primarily interest and, to a lesser degree, dependence – clearly play a role in burden avoidance bargaining. For example, Washington is dependent upon the European allies to shoulder as much of the burden as possible in Afghanistan, primarily due to the overstretched nature of the American military. But if the most critical elements of the CJSOR remain unfulfilled, the mission may not get off the ground at all, resulting in even fewer alliance force contributions. In order to avoid disaster, and retain whatever European contributions were offered, the United States is usually willing to close the CJSOR gap, even if it must do so with forces that are typically never deployed and instead devoted to training missions.31 To do otherwise would place an even greater burden on US forces, which would subsequently become responsible for the entire mission, not merely closing the CJSOR gap.
29 Press briefing on NATO’s Riga Summit by NATO Secretary General Jaap de Hoop Scheffer, Brussels, Belgium, 24 November 2006. 30 The CJSOR challenges are nothing new to NATO – similar problems have also existed for many years in the context of alliance operations in the Balkans. See Gregory G. Johnson, “Examining the SFOR Experience,” NATO Review, no. 4, Winter 2004. 31 Rick Emert, “Hohenfels Unit Heading to Afghanistan to Help with Electiontime Security,” European Stars & Stripes, 4 September 2004; Seth Robson, “Op For Prepares for Afghanistan Deployment,” European Stars & Stripes, 13 June 2006. It was highly unusual – and controversial within US military circles – to deploy so-called Op For, or opposing forces, and the order to do so reflected both the overstretched state of the American military and the crisis within the alliance’s force generation efforts.
The Alliance’s Response to Terrorism
97
The second major challenge facing the alliance in Afghanistan continues to be that posed by national caveats. If member states are reluctant to forfeit all decisionmaking authority when they assign forces to NATO operations or if they fear the political consequences of having their forces involved in particular military activities or operations, they employ caveats to limit how and where their forces may be used. In these cases, the strong interests of some allies in protecting their sovereignty through national caveats strengthens their bargaining position, while the strong dependence of the alliance as a whole on the force contributions of individual member states for mission success weakens NATO’s relative position. These restrictions seriously complicate NATO’s ability to operate in Afghanistan, particularly in crises, because at a minimum they greatly lengthen the amount of coordination or approval time necessary before units can be injected into hostilities. In the worst cases, national caveats can create chaos for tactical commanders on the ground, when they are unable to move and employ forces at will, as the situation requires.32 To some extent, complete removal of all caveats is simply unrealistic in the context of NATO operations – the forces contributed to such operations are those of sovereign states, and no state is simply going to turn over assets without caring about how they are used. Nonetheless, by late 2006, as the alliance expanded its presence across all of Afghanistan, the caveat problem became debilitating in the context of ISAF combat operations in the southern part of the country, especially as military action between NATO forces – principally British, Canadian, and Dutch forces – and those of the Taliban spiked in the last half of that year. As allied leaders such as then-SACEUR Jones called for increased allied force contributions for the south, many member states refused to provide additional forces or refused to lift caveats on extant forces in Afghanistan that would permit, for example, the movement of units in Kabul to locations in the south of the country.33 This was the case despite repeated indications from the countries bearing the brunt of the action in the south that their political commitment to the mission was threatened by increased casualties.34 Some progress was made as a result of the Riga summit in November 2006, but only insofar as allies agreed to allow the deployment of their ISAF contributions anywhere in Afghanistan “in extremis” situations. Aside from these ill-defined crisis situations, there were few indications that allies had given NATO commanders on the ground in Afghanistan much additional flexibility on the movement of forces within the
32 National caveats played a role in the uneven response of NATO-led KFOR to an eruption of ethnic violence in March 2004 during which Albanian mobs burned Serb-owned houses to the ground. In some cases, NATO troops remained in their barracks while the violence flared. Ian Traynor, “Western Policy in Kosovo Attacked,” The Guardian (London), 23 April 2004, p. 17. 33 Michael Abramowitz and Pamela Constable, “Bush to Pursue Fresh NATO Commitments,” The Washington Post, 28 November 2006, p. A01. 34 David Pugliese, “Afghan Mission Shakes Support for Canadian Leaders,” Defense News, 18 September 2006; Molly Moore and John Ward Anderson, “NATO Faces Growing Hurdle As Call for Troops Falls Short,” The Washington Post, 18 September 2006, p. A10.
98
Alliance Management and Maintenance
area of operations, something that many informed observers continued to view as a major problem.35 Although it does not directly reduce the ability of terrorists to conduct attacks, the NATO Response Force (NRF) is viewed as a critical tool in the alliance’s counterterrorism arsenal.36 Created as a result of the November 2002 Prague summit, the NRF became fully operational, at least on paper, following the exercise Steadfast Jaguar conducted in Cape Verde in late summer 2006 and the Riga summit in November of the same year. Its purpose is to provide the alliance with a robust high readiness joint and combined force that is able to deploy quickly and conduct the full range of potential NATO missions. Additionally, it was hoped the NRF would act as a catalyst for capability development within the alliance. Declared initially operational in October 2004, the NRF needed an additional two years of development and preparation before the alliance could declare it fully capable of conducting operations across its mission spectrum. As noted earlier in this section, the creation of the NRF has been a significant success for the alliance, providing it for the first time with a truly rapid-reaction on-call emergency response force. However, the NRF’s development has been fraught with some of the same challenges NATO has faced in Afghanistan, especially in terms of fulfilling the NRF’s CJSOR. As in the case of ISAF, for every new NRF rotation, alliance leadership must twist arms, publicly embarrass, and otherwise cajole members into providing just the minimum level of forces. This problem reached a pinnacle in early 2006 when, just months before SACEUR Gen. James Jones was to declare the NRF fully operational, the alliance revealed that only 70 to 75 percent of the CJSOR for the NRF had been fulfilled.37 Resorting again to the airwaves, SACEUR stated he was not confident that the NRF would reach full operational capability (FOC) by October 2006, the intended deadline. Even up until the days and hours before the Riga summit in late November 2006, the CJSOR remained unfulfilled and hence the FOC declaration remained at risk.38 NATO leaders specifically cited shortfalls in fixed-wing transport, rotary wing assets, intelligence, and logistics units. Part of the reluctance of many European allies to commit forces to the NRF is based on their unwillingness or inability to devote sufficient financial resources to the types of forces necessary to conduct robust, high-intensity anti- or counterterrorism operations.39 For example, most of the European allies lack significant military 35 Sheryl Gay Stolberg and Judy Dempsey, “NATO Rebuffs Bush on Troops Restrictions in Afghanistan,” The International Herald Tribune, 29 November 2006. 36 Along with counterterrorism, the NRF’s missions include crisis response, peace enforcement, forced entry, embargo operations, maritime interdiction, non-combatant evacuations, and humanitarian relief operations. 37 “SACEUR Not Optimistic About NRF FOC Schedule,” SHAPE News Summary, 13 February 2006; Inci Kucukaksoy, “SAC-T Visits JWC,” The Three Swords: The Magazine of the Joint Warfare Centre, no. 4, 4 April 2006, pp. 14–15. 38 Paul Ames, “NATO Seeks Last-Minute Contributions to Meet Deadline for Elite New Force,” The Associated Press, 27 November 2006. 39 Interview with a member of the NATO international staff responsible for force planning, 16 May 2003; interview with a British military officer assigned to NATO, 9 July 2003; interview with a retired US military officer, 27 April 2004; and “NRF – NATO’s
The Alliance’s Response to Terrorism
99
intelligence capabilities, especially advanced signals intelligence (SIGINT), relying instead on the United States.40 As the former NATO Secretary General Lord George Robertson was fond of pointing out, much of the force structure of the alliance remains of little value for out-of-area operations, since it consists primarily of heavy armor divisions and conscripted infantry. Redirecting alliance force structure toward more advanced out-of-area capabilities requires the expenditure of significant funds, something several European allies have been reluctant to pursue as noted earlier. The development of HRF(L)s addressed in this book have helped several of the allies to acquire the forces necessary to fulfill most of the NRF CJSOR, but there are additional requirements beyond what the HRF(L)s offer. Admittedly, in some cases, modernization programs – particularly those centered on weapons development – require significant time. Nonetheless, some of the more advanced capabilities necessary for interoperable high-intensity operations, such as C4ISR41 capabilities, are available commercially and could have been procured by European member states relatively quickly. Although the alliance in general has a strong interest in trying to achieve greater capabilities in order to fulfill its strategy, NATO’s collective bargaining position relative to many individual European members is weakened by the former’s strong dependence on member state contributions in general – the European member states lack the same level of dependence and so are in a somewhat stronger bargaining position. In the case of the NRF’s FOC declaration, the brinksmanship of the force generation process lasted up until the very hours before the Riga summit. Luckily for the alliance, last minute contributions the day of the summit enabled Gen. Jones to recommend to the North Atlantic Council that the NRF be declared fully operational. Unfortunately, however, the NRF rotations last only six months, and so twice a year the alliance will need to go through the same process of soliciting, and bargaining over, CJSOR commitments. Military Cooperation Two documents have been critical to defining the policies and practices of NATO with regard to civil-military cooperation (CIMIC). The first, MC 411, was originally written in 1997 and revised in mid-2001, and it established an alliance policy on CIMIC. The second, Allied Joint Publication 9 (AJP-9) was promulgated in mid-2003 and provides more detailed guidelines for the planning and execution of CIMIC operations. Both initially grew out of the alliance’s experience in the Expeditionary Capability,” briefing presented by Colonel Brad Mason, SHAPE J5/J9 NRF Coordination Group, at the NATO and Multinational Concept Development and Experimentation Conference (Prague), 5 November 2003. 40 Pierre Claude Nolin, General Rapporteur, “Interoperability: The Need for Transatlantic Harmonisation,” annual report of the NATO Parliamentary Assembly’s Science and Technology Committee, November 2006. The report notes that, “In the SIGINT area, no operational systems exist in Europe.” 41 Command, Control, Communications, Computers, Intelligence, Surveillance, and Reconnaissance.
100
Alliance Management and Maintenance
Balkans and the need recognized at the time for formal CIMIC guidance for alliance commanders on the ground. By 2006, following lessons learned in Afghanistan and elsewhere, allied military leaders realized they needed to provide commanders even more detailed planning guidance, particularly with regard to institutional cooperation between NATO and the UN, the European Union, and non-governmental organizations.42 Early that year, the international military staff requested additional political guidance from the NAC in order to accomplish its objective. The NAC in turn tasked the international staff to develop a paper on the subject of civil-military cooperation that would inform its discussion and perhaps form the basis for the political guidance requested by the military staff. As of mid-2006, the international staff was still crafting the paper on civilmilitary cooperation, and it remains unclear when that work will be complete. Some of the more controversial issues under consideration by the alliance include under what authority NATO can coordinate with civilian organizations, the demarcation between military responsibilities and political responsibilities, and whether NATO should even have CIMIC tools in-house or whether it should rely on the UN or other organizations that some believe are perhaps better equipped or structured to cooperate with civilian entities. Indeed, some member states believe NATO should not be very engaged at all in this area, and so the topic seems for now to be the subject of political bargaining between, on the one hand, member states that see a more expansive role for NATO in working with civilian entities and, on the other hand, member states that would prefer to see the EU or the UN handling such functions. In summary, in each of the functional areas where the alliance identified potentially new or enhanced capabilities it could acquire or develop for the fight against terrorism – anti-terrorism, consequence management, counterterrorism, and military cooperation – political bargaining has so far contributed significantly to shaping NATO’s response, in some cases preventing a more efficient, effective role for the alliance. In fact, indications to date are that political bargaining derailed what began as a threat-driven response to the attacks of 9/11. As with NATO’s response to the end of the Cold War in the 1990s and as noted earlier in this chapter, the alliance’s response to 9/11 indicated that even though NATO recognized a changed threat environment and subsequently saw the need for changes to its doctrine and structures, political bargaining over what to do next played a decisive role in the process and content of NATO’s contribution to the fight against terrorism. It seems that once the drama and emotional intensity of the immediate post-9/11 environment wore off, patterns of alliance maintenance similar to what was seen in the case of the HRF(L)s returned to the fore as political bargaining outpaced a response to security threats as the driving force in alliance management.
42 Interview with a member of the NATO international staff, 27 July 2006.
Chapter 8
Conclusions and Implications This book’s primary purpose was to shed light on the subject of alliance management and maintenance by examining how NATO confronted change in the security environment through the modification of its doctrine and its command and force structures. The analysis contained in the preceding chapters points to the importance of political bargaining, and the competing interests and dependencies that underpin it, as the most important determinant or independent variable. The role of security threats in shaping or guiding alliance management was secondary – important in terms of sparking the need for change, but then eclipsed as an explanatory variable by political bargaining. Behavior that would ostensibly point toward the importance of organizational processes – such as resisting change or rationing – was shown, upon a more detailed investigation, to be actually the byproduct of the bargaining process among member states or between the alliance on the one hand and a specific member state on the other. These findings were reaffirmed through an examination of NATO’s response to date to the 9/11 attacks. More specifically, the case outlined in the preceding chapters indicates that threat-based factors are a necessary but insufficient tool for explaining the sources of military doctrine and hence for explaining alliance management and maintenance. Changes in the security threats facing the alliance helped to explain why NATO initiated a doctrinal and command and force structure review, and they explain part of the doctrinal and structural outcomes that NATO ultimately chose and implemented beginning in 2002–2003, insofar as the new doctrine and structures emphasized the capability to respond to non-traditional threats. However, there are serious shortcomings associated with relying solely on security threats in trying to understand the case of the NRDCs. A more complete understanding of the process and outcome of NATO’s post-Cold War doctrinal and structural changes requires the use of political bargaining factors. In this case, NATO’s ability to muster the necessary doctrinal and structural changes in the aftermath of a major watershed in international security was decisively delayed and shaped by political bargaining that was determined by varying levels of dependence and interest on the part of NATO and its member states If this pattern of behavior holds true in the future, it seems possible to expect that in response to the next major watershed in international security, the alliance will manage the change by recognizing the strategic necessity of modifying how it conducts business but will move in hesitating steps through the short term in actually implementing doctrinal and structural changes, resulting in at least temporarily poor integration of doctrine, structures, and strategy. As was seen in the alliance’s response to Balkans crises in the 1990s, we might expect the alliance to employ ad
102
Alliance Management and Maintenance
hoc or stopgap solutions or to simply go without the necessary capabilities as has been the case with regard to deployment and sustainment. To some degree, these expectations have been evident in the alliance’s response to terrorism and the attacks of 9/11. Despite NATO’s recognition that it needed additional or modified doctrine and structures in a variety of areas such as consequence management, the implementation of the necessary changes has been delayed and shaped by varying member state interests and dependencies. For example, the alliance’s actions in Afghanistan to date have been impressive, but major progress remains in the areas of fulfilling the CJSOR and eliminating operationally significant national caveats before NATO’s leadership of ISAF can be termed a success. Of course, NATO’s activities in response to terrorism are still a work in progress, and so this remains an avenue for further research to verify or clarify the preliminary observations presented in this book. Addressing Doctrine and Structure As was shown through this book, a study of alliance doctrine and structure is most compelling – and forms the basis for further research in the context of NATO’s response to the 9/11 terrorist attacks – for several reasons. First, nothing is as fundamental to NATO as its military structure. Without the forces and command and control elements that comprise that structure, NATO cannot fulfill its most basic mission of promoting security. Today, those NATO forces and command structures are the NRDCs, as evidenced by alliance member reliance on them to fulfill NRF commitments and ISAF deployments. Second, trying to determine when and how the alliance changed its doctrine and structures to end up with the NRDCs offers a unique window into alliance management and maintenance during a period of change. Finally, understanding how alliances manage or maintain themselves, particularly through periods of change, benefits practitioners and academics alike. They need to understand how and why NATO doctrine and structures changed during the first decade after the end of the Cold War in order to know whether the alliance is a more effective and efficient organization better prepared to deal with post-Soviet threats. In using doctrinal and structural change as a window for examining alliance management, this book contributes to furthering our understanding of how alliances deal with change over time. Chapter 2 pointed out that although alliances have been around for millennia, little is known or understood about the processes such organizations employ for managing or maintaining themselves. Most studies on alliances have been qualitative efforts largely focused on alliance formation or quantitative studies limited to questions of alliance duration. Separately, most scholarly works on NATO’s transformation since the end of the Cold War have failed to consider such change in the context of the broader pursuit for insights into how alliances in general manage themselves. This book sought to bridge that divide by examining alliance management and maintenance through a focused study on how NATO modified its doctrine and structures during a period of change in the security environment. Changes to doctrine and the structures that stem from it acted as the vehicle for a broader examination of alliance management.
Conclusions and Implications
103
This book pointed out in Chapter 3 that practitioners and academics view military doctrine somewhat differently. Particularly since the mid-1980s, academic scholars have developed a varied array of studies that provide a basic conceptualization of military doctrine, its characteristics, and its role. Most have tended to favor Barry Posen’s definition of doctrine – how battles are fought, a concept linking tactics and strategy – and few have linked doctrine to command and force structures. To understand how NATO doctrine and force structure evolved as they did over the course of the 1990s, more specific notions of doctrine, its characteristics, and its role are necessary than are available from most academic sources. For this level of precision, one must borrow from practitioner notions on doctrine. Practitioners have more precise notions on what doctrine is and what it does, particularly with regard to its role in the development of command and force structures and how those elements must be linked to both doctrine and grand strategy. They believe that doctrine should provide information on the approach to achieving broader goals and missions, on what military organizations should look like and why, on what their lines of authority are, on what degree of control they have over their forces, and on how they are supported. These more precise concepts provide a level of richness that goes beyond typical academic understandings of doctrine and its role and tie structures more directly to grand strategy. Such precision is necessary in order to understand and explain NATO’s post-Cold War evolution. By borrowing these more specific concepts on doctrine’s definition and its role from practitioners to augment the more general conceptualizations employed by most academic scholars, this book was able to arrive at a more complete explanation of changes in NATO doctrine and structures since the end of the Cold War because it employed a fuller understanding of what NATO’s doctrine really consists of. The differences, although somewhat more nuanced, also extend to practitioner and academic views on the sources of doctrine. Traditional academic approaches tend to focus on structural or functional explanations, taking Posen’s lead with reliance on the structure of the international system or the nature of organizations in explaining the sources of military doctrine. Other scholars have found that such traditional approaches on the sources of military doctrine lack the level of specificity and richness necessary to explain doctrine. Their non-traditional, alternative approaches, such as those relying on institutional, political, or cultural explanations, make an important contribution by helping to explain instances of doctrinal variation among what may appear to be similar states or organizations. However, even with the alternative academic approaches, the level of precision necessary to understand and explain the sources of NATO’s military doctrine may not be enough because they do not necessarily arm us with enough or the right type of context to know what was really occurring within NATO and between alliance members. For this level of precision, a more in-depth approach is necessary if we hope to explain NATO’s post-Cold War changes. For the necessary level of precision, this study again turned to practitioners. They tend to draw on some of the same notions as academic scholars – such as threat and political context – but practitioners tend to favor greater actor-level specificity than academics, largely because they are unconcerned with building theories. Actor-specific experience, threats, and politics permeate practitioner notions on the
Alliance Management and Maintenance
104
sources of military doctrine, adding significantly more powerful tools for explaining the sources of military doctrine, although necessarily reducing parsimony. For an examination of NATO’s military doctrine, consideration of in-depth political factors and threat factors makes sense since the alliance is as much a political organization as a military one. However, the notion of ‘political factors’ as an independent variable is obviously broad. In order to get a better grasp on this independent variable, this study relied on intra-alliance bargaining models. Such models emphasize member state interest, commitment, and dependence, and the domestic factors that shape each, in determining relative bargaining strength and the ability and willingness of bargaining states to achieve solutions in their own win sets and those of their negotiating partners. The third chapter also established the expectations for post-Cold War alliance doctrine and structures. An explanation based solely on security threats would have expected the alliance to perceive a significantly changed threat environment; to declare that the alliance has a critical role to play in responding to the new threats; to move forthrightly to address the new threat environment by instituting changes in its doctrine and structures; to shift alliance doctrine and force structure away from defending against large Soviet tank divisions; and to develop alliance forces that are more deployable and expeditionary, capable of handling traditional threats to allied territory as well as transnational threats and international crises. Given NATO’s history as an alliance established to counter the Soviet military threat to Western Europe, the case of NATO’s post-Cold War doctrine would seem easily explainable through an examination of security threats. However, as was discussed in Chapter 5, a threat-based explanation for the sources of NATO’s doctrine is flawed on two counts: •
•
The shift in doctrine and the related changes in command and force structures were implemented well after such changes became necessary as outlined above and in subsequent alliance revisions to NATO strategy – the command and force structure changes implemented in 2002 and 2003 were based upon a doctrine that sought to fulfill requirements identified in the 1991 Strategic Concept and arguably as early as 1989 or 1990. After NATO finally began to implement the necessary force structure changes in 2002, it did not choose to establish the three land High Readiness Force headquarters that were deemed necessary – instead, the alliance chose to establish as many as six HRF(L)s, resulting in an inefficient allocation of alliance resources, which could have been devoted to sustainment and deployment capabilities.
In order to accurately assess NATO’s doctrinal and structural changes in the postCold War era, one must also consider political bargaining factors. NATO is very much a political entity: the alliance’s own stated fundamental task is to safeguard the security of its members through “political and military means.”1 An approach centered on political bargaining would expect the alliance to utilize established 1
The NATO Handbook (1991), p. 30.
Conclusions and Implications
105
channels of communication to bargain over defense burdens and benefits; the ally with the strongest bargaining position to achieve its desired outcome; the allies with greater interest in an object of intra-alliance bargaining to have a stronger bargaining position than those with a weaker interest; and the allies with less dependence on the alliance for security to have a stronger bargaining position than those with a greater dependence. In examining the development of the NRDCs in Chapters 4, 5, and 6, it is clear that there were serious implications for the delay in fully addressing alliance doctrinal and structural shortcomings following the end of the Cold War and for NATO’s decision to designate six of the NRDCs as High Readiness Forces (Land) versus just three. For example, throughout the 1990s NATO lacked the ability to adequately deploy and sustain forces in accordance with its own strategy. Instead of spending limited defense budget resources on unnecessary capabilities – such as creating and then maintaining three more HRF(L)s than NATO needed – alliance members could have spent more money on other capabilities more beneficial, such as those capabilities identified by the alliance itself in two separate initiatives, including deployment and sustainment. This might have allowed the alliance to close the growing credibility gap between its strategy and its ability to fulfill that strategy. Additionally, if the alliance had had the appropriate doctrine and structures in place to deal with the kind of threat represented by Albania’s descent into chaos during 1996 and 1997, NATO might have become directly involved and the West’s response might have been swifter, more effective, longer lasting, and less susceptible to accusations that it lacked impartiality. Finally, if NATO had had the doctrine and structures that its own strategies called for, it might have responded more effectively and efficiently to the deteriorating situation in the former Yugoslavia. In taking as much time as it did to implement the NRDCs and in implementing six instead of three, the alliance was obviously reacting to imperatives other than security threats. This is not to say that security threats as an explanatory variable are useless – in fact, they are an important, necessary part of the story. They help to explain why NATO initiated a doctrinal review in the first place and they explain part of the outcome of that review insofar as they point toward the type of deployable, expeditionary doctrine and structures that NATO partially achieved with the implementation of the NRDCs. However, relying solely on security threats to explain how the alliance managed change during this time period overlooks key elements of how NATO went from its 1991 strategy to the 2003 implementation of the NRDCs and HRF(L)s. Political bargaining – and the interests and dependencies that define it – explains the parts of the story that security threats alone cannot, telling us why the alliance took so long to do what it said it would do and why the alliance ended up with a force structure that was at least slightly off the mark. Without considering political bargaining factors, reliance on security threats alone as an explanatory tool would not enable us to understand NATO’s post-Cold War doctrinal and structural changes nor how the alliance finds itself in its current state of having to publicly browbeat members into fulfilling requirements for ISAF or to lease commercial aircraft from third parties in order to deploy.
106
Alliance Management and Maintenance
Implications of this Study This study has found that political factors decisively shaped NATO’s doctrinal response to the end of the Cold War, but this is not necessarily a novel conclusion. Allied military officials and policymakers understand that most alliance actions occur only in the context of politics. Indeed, during the course of conducting interviews for this book, many interviewees began the discussion by stating, “First, you should know that NATO is driven by politics,” or something to that effect. For longtime observers of NATO, the fact that the alliance’s doctrine would be based on political factors is not particularly surprising, at least for practitioners. Nonetheless, conclusions drawn in this book clearly have implications for both foreign policy practitioners and for the academic community. For the former, the fact that political factors played such a major role, despite the rhetoric of a need to respond to the changing threat environment, speaks to an increasingly problematic disjoint between the alliance’s public stance and its private deliberations and decisions, or between its goals and the means with which it pursues them. In this case, NATO’s public stance was evident primarily through documents such as the 1991 and 1999 Strategic Concepts, official communiqués of ministerial meetings throughout the 90s, or other formal communications and outreach products available to the public. In most all of these, the alliance appeared from a strategic perspective to be driven by the requirements of a changed threat environment, and NATO routinely articulated the steps it was taking in response to the new, multifaceted threats facing member states. But behind the scenes, the process and content of the doctrinal and structural reviews became mired in intra-alliance bargaining. This is not to imply any sort of duplicity, or to argue that the alliance should be completely open in airing its sensitive internal deliberations, but it points to a growing problem of credibility. The disjoint between rhetoric and reality, especially on issues as important as how the alliance will fight and what tools it will employ in fighting, is of particular importance because it threatens to undermine alliance credibility and hence legitimacy. An entity like NATO will never fully escape politics, and it is unlikely that the alliance would unveil its classified deliberations to public scrutiny, but policymakers must work where they can – for example, by tempering rhetoric – to ensure that alliance statements do not outpace realities on the ground and that these elements do not undermine alliance credibility and legitimacy. Otherwise, an alliance without the necessary credibility and legitimacy will eventually fail at providing effective, efficient responses to genuine security concerns facing member states. Second, this book confirms for the practitioner the usefulness of treating NATO as a political organization first and a military alliance second. Juggling the political requirements of 26 sovereign states requires foresight and flexibility. Most importantly, it requires knowledge of the various levels of interest and dependence at play. These factors – and the electoral, budgetary, or other national politics that comprise them – underpin the success or failure of the horse-trading that frequently takes place away from NATO’s formal negotiating settings – more work often gets done over coffee than at the usually pro forma meetings of member state representatives. The bargaining win sets of 26 sovereign member states are very unlikely to align perfectly
Conclusions and Implications
107
on any given issue facing the alliance – as a former US ambassador to NATO put it, “The lack of coterminous interests makes it remarkable that NATO is able to make any headway at all.”2 Therefore, the successful practitioner or policymaker needs to focus on understanding the political motivations – both those that compel behavior and those that deter or limit behavior – of each of the member states and then on determining courses of action that succeed in fulfilling the win sets of most of the allies. The goal here is to achieve that most elegant of political solutions in which all member states arrive at consensus and are satisfied with the outcome, regardless of their individual motivations. Third, this study also shows the necessity of improving alliance efficiency and effectiveness if NATO is to remain something more than just a talking shop for European security. Presently, the major advantage of NATO over such bodies as the Organization for Security and Cooperation in Europe (OSCE), the Group of Eight (G8), the European Union (EU), or the UN Security Council is that NATO has an integrated military command capable of planning and leading collective, offensive military action. But if something as fundamental to the IMC as its very doctrine and the means with which it implements that doctrine are so easily the object of political winds, member states may be less willing to commit blood and treasure to the achievement of alliance objectives. After all, it is only natural for member states to seek assurances that their contributions and sacrifices are made in the name of legitimate security threats, not for the sake of political expediency. As noted throughout this book, burden-sharing is already a major challenge within the alliance, especially evident now as member states struggle to fill NATO requirements for ISAF. Any lessening of member state commitment to the alliance’s military capabilities would throw NATO into even greater imbalance with regard to how member states share the costs and risks of providing for their common security. Furthering the current imbalance, which exists in terms of resources allocated, risk, usability of forces, and capabilities, would only doom the alliance to eventual collapse. In order to forestall this, assuming the alliance continues to provide collective benefits that members are unable or unwilling to provide on their own, addressing alliance effectiveness and efficiency is a necessity. In addition to the implications for practitioners noted above, this study entails implications for the academic community. First, this study offers new empirical insights into NATO’s post-Cold War development, a topic of continued, strong interest among academics and policymakers alike. Other works have examined alliance adaptation from the early 1990s through the latter part of the same decade. However, no major work has traced internal alliance adaptation, and especially its attendant doctrinal and structural changes, from the early 1990s through the early part of the 21st century. This study fills that empirical gap in the literature, and it does so by relying mainly on primary sources such as interviews, alliance documents, and personal correspondence. Second, this study has academic implications insofar as it forms a basis for further work on the related topic of alliance maintenance. Most major studies on alliance theory deal with their formation, not with what it takes to maintain them. 2
Interview with a former US ambassador to NATO, 21 March 2003.
108
Alliance Management and Maintenance
With regard to the latter, this study offers a data point for those wishing to examine how NATO adapted itself to deal with the changes posed by the end of the Cold War, changes that some observers thought would mean the end of the alliance. It does this in part by presenting a story of how alliance members balanced often competing levels of interest and dependence in their pursuit of their win set. The way in which the member states accomplished this speaks to issues regarding alliance management during periods of significant change in the international environment, offering an avenue of further investigation for those researching alliance maintenance. Finally, this book entails implications for the academic community by making a novel contribution to the body of scholarly literature on the sources of military doctrine in general and the integration of military doctrine with grand strategy more specifically. For example, this study found that two sets of factors were both necessary and together sufficient to explain the process and outcomes of NATO’s doctrinal and structural changes in the post-Cold War period. Threat factors by themselves were seen to be necessary but not sufficient, requiring augmentation by political bargaining factors in order to explain – and perhaps to predict, in the case of NATO’s developing role in the fight against terrorism – the process and outcome of the alliance’s doctrinal review and subsequently of the command and force structure reviews. Without political bargaining factors, a complete understanding and explanation of NATO’s doctrinal and structural changes is impossible. Additionally, this book makes a novel contribution by choosing to survey the military doctrine of not one state, but rather a collection of states that have pooled some of their military capabilities. NATO clearly has a military doctrine; it therefore makes sense to consider it one of the data points available for those studying doctrine’s sources and its integration with strategy, despite the greater complexity of analyzing the doctrine of an alliance versus that of a single state. For both academics and practitioners, the story of NATO’s post-Cold War evolution is still unraveling. As it does so, policymakers will continue to grapple with the consequences of decisions made over the last 15 years regarding alliance doctrine and structures. If NATO is to continue providing security for its members and a common voice for the Western democracies, policymakers must ensure that the alliance’s ways and means match its ends.
Bibliography Air Force Basic Doctrine. (2003), US Air Force. “Allied Command Structures in the New NATO.” (1997), Washington, DC: Institute for National Strategic Studies. Ames, Paul. “NATO Seeks Last-Minute Contributions to Meet Deadline for Elite New Force.” The Associated Press, 27 November 2006. Ando, Salvo. “Preparing the Ground for an Alliance Peacekeeping Role.” NATO Review, April 1993, pp. 4–9. Appathurai, James. “Closing the Capabilities Gap.” NATO Review (Autumn 2002), pp. 2–4. Asmus, Ronald D. (2002), Opening NATO’s Door: How the Alliance Remade Itself for a New Era. (New York: Columbia University Press). Asmus, Ronald D. and Richard C. Holbrooke, “Re-Reinventing NATO,” paper presented at the conference on “Transforming NATO in a New Global Era,” held in conjunction with the NATO summit in Riga, Latvia 27–29 November 2006. Atkinson, Rick. “Air Assault Set Stage for Broader Role.” The Washington Post, 15 November 1995, p. A01. Avant, Deborah. (2004), Political Institutions and Military Change: Lessons From Peripheral Wars. (Ithaca: Cornell University Press). Bennett, Andrew, Joseph Lepgold, and Danny Unger. “Burden-Sharing in the Persian Gulf War.” International Organization, vol. 48, no. 1 (Winter 1994), pp. 39–75. Bennett, D. Scott. “Testing Alternative Models of Alliance Duration, 1816–1984.” American Journal of Political Science (July 1997), pp. 846–78. Boland, Frank. “Force Planning the new NATO.” NATO Review (Autumn 1998), pp. 32–35. “Bosnian Operations Stretch NATO Thin,” editorial, Aviation Week & Space Technology, vol. 140, no. 15 (11 April 1994), p. 70. Brodie, Bernard. (1959), Strategy in the Missile Age. (Princeton, N.J.: Princeton University Press). Bush, George W. “Address to a Joint Session of Congress and the American People,” Washington, DC 20 September 2001. Bush, George W. “President Bush Delivers Graduation Speech at West Point,” West Point, New York 1 June 2002. Correspondence with a Dutch military officer assigned to NATO, 4 May 2004. Correspondence with a former DoD official involved in NATO policy, 5 August 2004.
110
Alliance Management and Maintenance
Correspondence with a former senior Congressional staff member involved in NATO policy, 11 May 2004. Correspondence with a German military officer assigned to NATO, 22 April 2004. Correspondence with a NATO official, 22 April 2004. Correspondence with a NATO staff member involved in policy planning, 20 August 2004. Covault, Craig. “NATO Seeks Upgrades, Presses Air/Sea Blockade.” Aviation Week & Space Technology, vol. 137, no. 23 (7 December 1992), p. 62. Covault, Craig. “Growing Instability Clouds European Military Outlook.” Aviation Week & Space Technology, vol. 138, no. 11 (15 March 1993), p. 55. Cristman, Daniel W. “NATO’s Military Future.” Joint Forces Quarterly. Spring 1996, pp. 76–81. Cragg, Anthony. “The Combined Joint Task Force concept: a key component of the Alliance’s adaptation.” NATO Review (July 1996), pp. 7–10. Cunningham, M.J. “The Main Operating Base: Castle or Coffin?” RUSI Journal (December 1987), pp. 35–36. Cyert, Richard and James March. (1963), A Behavioral Theory of the Firm. (Englewood Cliffs, NJ: Prentice Hall). Daalder, Ivo. (1991), The Nature and Practice of Flexible Response: NATO Strategy and Theatre Nuclear Forces Since 1967. (New York: Columbia University Press). de Hoop Scheffer, Jaap. “Global NATO: Overdue or Overstretch?” speech at the Security and Defence Agenda Conference. Brussels, Belgium, 6 November 2006 de Wijk, Rob. (1997), NATO on the Brink of the New Millennium: The Battle for Consensus. (London: Brassey’s). Declaration of the Heads of State and Government, Ministerial Meeting of the North Atlantic Council/North Atlantic Cooperation Council, Brussels, 10–11 January 1994. Design for Military Operations: The British Military Doctrine, prepared under the direction of the Chief of the UK General Staff, 1996. Diedrich, John. “‘Paper Armies’ Not Enough.” Colorado Springs Gazette, 9 October 2003, p. 1. “DPC Discussion of Capabilities and Risk.” Telegram from US Mission NATO to Washington, 3 May 1968. Declassified on 29 August 2001. Drew, Dennis and Don Snow. (1988), Making Strategy: An Introduction to National Security Processes and Problems. (Maxwell AFB, Alabama: Air University Press). Drozdiak, William. “Europeans Reject Call For Help in Albania; NATO Reluctant to Intervene Despite Threat.” The Washington Post, 14 March 1997, p. A19. Duffield, John S. “International Regimes and Alliance Behavior: Explaining NATO Conventional Force Levels.” International Organization, vol. 46 (Autumn 1992), pp. 819–55. Eichenberg, Richard C. “Dual Track and Double Trouble: The Two-Level Politics of INF,” in Peter B. Evans, Harold K. Jacobson, and Robert D. Putnam, eds. (1993), Double-Edged Diplomacy: International Bargaining and Domestic Politics. (Berkeley: University of California Press, pp. 45–76).
Bibliography
111
Emert, Rick. “Hohenfels Unit Heading to Afghanistan to Help with Electiontime Security.” European Stars & Stripes, 4 September 2004. “Establishment of an Italian Framework HRF(L) HQ.” Briefing prepared by the Force Planning Department, General Planning Division, Italian Army Staff, April 2003. Evans, Michael. “Almost half of army to be in Nato’s rapid reaction force.” London Times, 24 July 1991. Evans, Peter B., Harold K. Jacobson, and Robert D. Putnam, eds. (1993), DoubleEdged Diplomacy: International Bargaining and Domestic Politics. (Berkeley: University of California Press). Favretto, Marcella and Tasos Kokkinides. “Anarchy in Albania: Collapse of European Collective Security?” British American Security Information Council Occasional Papers on International Security Policy, No. 21, June 1997. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 12–13 December 1991. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 26 May 1993. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 24 May 1994. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 29 November 1995. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 30 May 1995. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 8 June 1995. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 29 November 1995. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 12 June 1997. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 2 December 1997. Final Communiqué, Defense Planning Committee and Nuclear Planning Group, 11 June 1998. Final Communiqué, North Atlantic Council, 16 December 1997. Final Communiqué, North Atlantic Council in Ministerial Session, 2 December 1993. Final Communiqué, North Atlantic Council in Ministerial Session, 10 June 1993. Final Communiqué, North Atlantic Council in Ministerial Session, 30 May 1995. Final Communiqué, North Atlantic Council in Ministerial Session, 5 December 1995. “Financial and Economic Data Relating to NATO Defence: Defence Expenditures of NATO Countries (1975–1997).” NATO Press Release, 2 December 1997. “Financial and Economic Data Relating to NATO Defence: Defence Expenditures of NATO Countries (1980–2003),” NATO Press Release, 1 December 2003. Firlie, Michael E. “A New Approach: NATO Standing Combined Joint Task Forces.” Joint Forces Quarterly, Autumn/Winter 1999–2000, pp. 32–35.
112
Alliance Management and Maintenance
“FM 3-0: Operations.” Department of the Army, June 2001. Fox, William and Annette Fox. (1967), NATO and the Range of American Choice. (New York: Columbia University Press). Galvin, John R. “From Immediate Defense Towards Long-Term Stability.” NATO Review, December 1991, pp. 14–18. George, Alexander L. (1993), Bridging the Gap: Theory & Practice in Foreign Policy. (Washington, DC: US Institute of Peace Press). “Glossary of Terms and Definitions.” The NATO Standardization Agency, 2002. Gompert, David, Richard Kugler, and Martin Libicki. (1999), Mind the Gap: Promoting a Transatlantic Revolution in Military Affairs. (Washington: National Defense University Press). “Greece Gains Equality in New NATO Military Structure.” Press release by the Greek Embassy to the United States, December 1997. “Greece, Turkey Bury Old NATO Dispute Over Command Structure.” Turkish Daily News, 7 December 1997. Heinz, John P., Edward O. Laumann, Robert L. Nelson, and Robert H. Salisbury. (1993), The Hollow Core, Private Interests in National Policy Making. (Cambridge, Massachusetts: Harvard University Press). Helmreich, Jonathan, “The Negotiation of the Franco-Belgian Military Accord of 1920,” French Historical Studies, Spring 1964, pp. 360–378. Herrly, Peter F. “The Plight of Joint Doctrine After Kosovo,” Joint Forces Quarterly. Summer 1999, pp. 99–104. Holley, Jr., I. B. (1953), Ideas and Weapons: Exploitation of the Aerial Weapon by the United States during World War I. (New Haven: Yale University Press). Reprinted, Washington, D.C.: Office of Air Force History, 1983. Interview with a British military officer assigned to NATO, 9 July 2003. Interview with a Dutch military officer assigned to NATO, 1 August 2003. Interview with a former DoD official responsible for NATO policy, 22 September 2004. Interview with a former US ambassador to NATO, 21 March 2003. Interview with a French military officer assigned to NATO, 3 October 2003. Interview with a German military officer assigned to NATO, 1 August 2003. Interview with a member of the NATO international staff involved in force planning, 16 May 2003. Interview with a member of the NATO international staff, 21 March 2005. Interview with a member of the US delegation to NATO, 21 March 2005. Interview with a NATO military official involved in policy and force planning, 11 June 2004. Interview with a NATO official involved in alliance counterterrorism policy, 10 February 2004. Interview with a NATO official involved in alliance counterterrorism policy, 21 March 2005. Interview with a NATO official involved in force planning, 16 May 2003. Interview with a NATO official involved in military planning, 31 March 2004. Interview with a NATO official involved in policy planning, 2 February 2004. Interview with a Polish military officer assigned to NATO, 3 October 2003.
Bibliography
113
Interview with a retired British military officer assigned to NATO, 11 June 2004. Interview with a retired US military officer assigned to NATO, 27 April 2004. Interview with a retired US military officer formerly assigned to NATO, 6 April 2004. Interview with a retired US military officer formerly responsible for issues regarding allied doctrine, 24 January 2003. Interview with a retired US military officer previously assigned to NATO, 15 March 2004. Interview with a senior retired US military officer involved in alliance planning, 24 October 2003. Interview with a senior Turkish officer assigned to NATO, 3 October 2003. Interview with a US defense official involved in NATO policy planning, 25 January 2003. Interview with a US government official responsible for military planning, 19 February 2003. Interview with a US military officer assigned to ISAF, 4 October 2006. Interview with a US military officer assigned to NATO, 7 February 2003. Interview with a US military officer assigned to the European Command, 14 November 2002. Interview with a US military officer assigned to the US Mission, NATO, 14 March 2003. Interview with a US military officer assigned to the US European Command, 10 April 2003. Interview with an Italian military officer assigned to NATO, 1 August 2003. Joffe, Ellis. (1987), The Chinese Army After Mao. (Cambridge: Harvard University Press). Johnson, Gregory G. “Examining the SFOR Experience.” NATO Review, no. 4, Winter 2004. “Joint Doctrine Capstone and Keystone Primer,” US Department of Defense, 10 September 2001. Joint Press Conference by the Secretary General of NATO and the Minister of Defense of Slovenia, Portoroz, Slovenia, 28 September 2006. Joint Publication 1-02, “DOD Dictionary of Military and Associated Terms,” as amended through 14 August 2002. Joulwan, George A. “SHAPE and IFOR: Adapting to the Needs of Tomorrow.” NATO Review, March 1996, pp. 6–9. Kaufman, Stuart. “Organizational Politics and Change in Soviet Military Policy,” World Politics, April 1994, pp. 355–82. Kay, Sean. (1998), NATO and the Future of European Security. (Lanham, Maryland: Rowman and Littlefield). Kier, Elizabeth. (1997), Imagining War: French and British Military Doctrine Between the Wars. (Princeton, NJ: Princeton University Press). Krause, Joachim, “Multilateralism: Behind European Views,” The Washington Quarterly, vol. 27, no. 2, Spring 2004, pp. 43–59 Kriendler, John. “NATO’s Changing Role: Opportunities and Constraints for Peacekeeping.” NATO Review, June 1993, pp. 16–21.
114
Alliance Management and Maintenance
Kucukaksoy, Inci. “SAC-T Visits JWC.” The Three Swords: The Magazine of the Joint Warfare Centre, no. 4, 4 April 2006, pp. 14–15. Kugler, Richard. (1994), US-West European Cooperation in Out-of-Area Military Operations: Problems and Prospects. (Washington, DC: RAND Corporation). LaFeber, Walter. (1985), America, Russia, and Cold War: 1945–1984. (New York: Knopf). Land Warfare Doctrine 1: The Fundamentals of Land Warfare. (2002), Australian Army. Lax, Mark R. (1997), Australian Air Power Post 2010. (Montgomery, Alabama: USAF War College). Legge, Michael. “The Making of NATO’s New Strategy.” NATO Review, December 1991, pp. 9–14. Lenorovitz, Jeffrey M., Craig Covault, and Carole A. Shifrin. “European Forces Seeking Better Standoff Capability.” Aviation Week & Space Technology, vol. 140, no. 3, 17 January 1994, p. 59. Lind, William S. “Military Doctrine, Force Structure, and the Defense DecisionMaking Process.” Air University Review, May–June 1979. Lindley-French, Julian. “Leading Alone Or Acting Together? The Transatlantic Security Agenda For The Next US Presidency.” Institute for Security Studies of WEU, September 2000. Liska, George. (1962), Nations in Alliance: The Limits of Interdependence. (Baltimore, MD: The Johns Hopkins University Press). Marquina, Antonia. “Spain and the ‘Europeanization’ of Security.” As cited in Carl Lankowski and Simon Serfaty, eds, Europeanizing Security? NATO and an Integrating Europe, American Institute of Contemporary German Studies, The Johns Hopkins University, Research Report No. 9, 1999. Marshall, A.W. (1966), Determinants of NATO Force Posture. (Santa Monica, California: RAND). Marten, Kimberly Zisk. (1993), Engaging the Enemy: Organization Theory and Soviet Military Innovation, 1955–1991. (Princeton: Princeton University Press). McCalla, Robert. “NATO’s Persistence After the Cold War.” International Organization Summer 1996, pp. 445–75. Mearsheimer, John J. “Back to the Future: Instability in Europe After the Cold War.” International Security, vol. 15, Summer 1990, pp. 5–56. (2001), The Tragedy of Great Power Politics. (New York: W.W. Norton). Millen, Raymond A. (2002), Tweaking NATO: The Case for Integrated Multinational Divisions. (Carlisle, PA: The Strategic Studies Institute). Morgenthau, Hans J. (1956), Politics Among Nations. (New York: Alfred A. Knopf). The National Security Strategy of the United States 17 September 2002. “NATO: A Changing Alliance Faces New Challenges.” United States General Accounting Office (GAO/NSLAD-92-262), July 1992. “NATO’s Defense Capabilities Initiative,” NATO Factsheet, 9 August 2000. “NATO Intel,” Aviation Week & Space Technology, vol. 138, no. 3, 18 January 1993, p. 19.
Bibliography
115
“NATO’s Military Concept for Defense Against Terrorism,” NATO Factsheet, updated 15 December 2003. “NATO’s Response to Terrorism,” statement issued at the Ministerial Meeting of the North Atlantic Council held at NATO Headquarters, Brussels, 6 December 2001. NATO Strategic Concept, 1991. NATO Strategic Concept, 1999. “NATO Strategy and Force Structure.” Memorandum from the Office of the Secretary of Defense to the President, 16 January 1968. Declassified on 17 December 1999. NATO Update, week of 8–14 November 2000. Nauman, Klaus. “NATO’s New Military Command Structure.” NATO Review, Spring 1998, pp. 10–14. “New Capabilities: Transforming NATO Forces,” policy paper of The Atlantic Council of the United States, September 2002. Olson, Mancur and Richard Zeckhauser. “An Economic Review of Alliances.” The Review of Economics and Statistics, August 1966, pp. 266–79. “Operation Active Endeavour,” NATO Briefing, December 2003. “Partnership for Peace: Invitation.” Issued by the Heads of State and Government participating in the Meeting of the North Atlantic Council, Brussels, 10–11 January 1994. Peters, John E., David Shlapak, and Timothy Liston. “Allied Power Projection Capabilities” in (2001), Persian Gulf Security: Improving Allied Military Contributions. (Washington, DC: RAND Corporation, pp. 71–104). Posen, Barry. (1984), The Sources of Military Doctrine: France, Britain, and Germany Between the World Wars. (Ithaca, NY: Cornell University Press). Press briefing on NATO’s Riga Summit by NATO Secretary General Jaap de Hoop Scheffer, Brussels, Belgium 24 November 2006. Press conference with NATO Secretary General Jaap de Hoop Scheffer and the Assistant Secretary General for Defence Investment, Marshall S. Billingslea on NATO moving to acquire C-17 Strategic Airlift Aircraft 12 September 2006. Richburg, Keith B. “Gibraltar Votes to Remain British.” The Washington Post, 8 November 2002, p. A26. Robson, Seth. “Op For Prepares for Afghanistan Deployment.” European Stars & Stripes, 13 June 2006. Rosen, Stephen Peter. “New Ways of War: Understanding Military Innovation,” International Security, Summer 1988, pp. 134–68. Schlozman, Kay L. and John T. Tierney. (1986), Organized Interests and American Democracy. (New York: Harper and Row). “September 11 – 18 Months On: NATO’s contribution to the fight against terrorism,” NATO Factsheet, 1 April 2003. Shaping the Future of the Canadian Forces: A Strategy for 2020. National Defence Department of Canada, June 1999. Simon, Herbert. (1982), Models of Bounded Rationality, Volume 2, Behavioral Economics and Business Organization. (Cambridge, Massachusetts: MIT Press).
116
Alliance Management and Maintenance
Sloan, Elinor. “DCI: Responding to the US-led Revolution in Military Affairs.” NATO Review, Spring-Summer 2000, pp. 4–7. Sloan, Stanley. “NATO’s Future in a New Europe.” International Affairs, July 1990, pp. 495–511. “NATO: July 1997 Madrid Summit Outcome.” CRS Report for Congress, 14 July 1997. Smith, Martin. (2000), NATO in the First Decade After the Cold War. (Dordrecht, The Netherlands: Kluwer Academic Publishers). Smith, Roland. “A Changing NATO.” NATO Review, June 1997, pp. 8–11. Snyder, Glenn. “The Security Dilemma in Alliance Politics.” World Politics, July 1984, pp. 461–95. Snyder, Glenn. (1997), Alliance Politics. (Ithaca: Cornell University Press). Spencer, Sebastian, “NATO Scrambling to Improve Intel Sharing, But Obstacles Remain,” Inside the Pentagon, 19 October 2006. Stromseth, Jane. (1988), The Origins of Flexible Response: NATO’s Debate over Strategy in the 1960s. (London: Macmillan Press). Summer, Harry G. “Military Doctrine: Blueprint for Force Planning,” Strategic Review, vol. 20, no. 2, Spring 1992, pp. 9–22. The NATO Handbook, 1991. The NATO Handbook, 1998. The NATO Handbook, 2001. Thies, Wallace J. (2003), Friendly Rivals: Bargaining and Burden-Shifting in NATO. (Armonk, New York: M.E. Sharpe, Inc.). Traynor, Ian. “Western Policy in Kosovo Attacked.” The Guardian (London), 23 April 2004, p. 17. Ulrich, Marybeth Peterson. “NATO Since Kosovo: The Impact of the War in Kosovo on the Euro-Atlantic Security Community.” Paper presented at the 41st Annual Convention of the International Studies Association held in Los Angeles, California, 14–18 March 2000. Ulriksen, Ståle. “Military Europe: Capabilities and Constraints.” Paper presented at the Fra Konvent Til Konstitusjon: Europas Utvikling Og Norsk Forskning conference at the Centre For European Studies held in Kristiansand, Norway, 28–29 August 2003. US Department of State Fact Sheet, “NATO: Coalition Contributions to the War on Terrorism,” Bureau of European and Eurasian Affairs, Washington, DC, 31 October 2002. Venturoni, Guido. “The Military Preparedness of the Alliance’s Forces in the light of various Defence Reviews and Reforms.” Presentation at the International NATO Conference in Budapest, Hungary, 8 November 2001. Villacres, Edward J. and Christopher Bassford. “Reclaiming the Clausewitzian Trinity.” Parameters, Autumn 1995, pp. 9–19. Vittori, Jay M. “Making the Case for Multinational Military Doctrine.” Joint Forces Quarterly, Spring 1998. Walt, Stephen M. (1987), The Origins of Alliances. (Ithaca: Cornell University Press). Waltz, Kenneth N. (1979), Theory of International Politics. (Reading, MA: AddisonWesley).
Bibliography
117
Wentz, Larry, ed. (1997), Lessons from Bosnia: The IFOR Experience. (Washington: Institute for National Strategic Studies). Wilson, James Q. “Innovation in Organization: Notes Toward a Theory,” in James D. Thompson, (1966), Approaches to Organizational Design. (Pittsburgh: University of Pittsburgh Press). Wörner, Manfred. “Shaping the Alliance for the Future.” NATO Review, February 1994, pp. 3–6. “The Future of the Alliance,” speech given at the University of Istanbul, Turkey, 18 September 1989. Yost, David S. (1999), NATO Transformed: The Alliance’s New Roles in International Security. (Washington, DC: US Institute of Peace). Young, Thomas-Durell. (1997), Multinational Land Formations And NATO: Reforming Practices And Structures. (Carlisle, PA: Strategic Studies Institute). (1998), Reforming NATO’s Military Structures: The Long-Term Study and its Implications for Land Forces. (Carlisle, PA: Strategic Studies Institute).
This page intentionally left blank
Index Afghanistan 6, 31, 52, 90, 91, 92, 94, 95, 96, 97, 98, 100, 102, 111, 115 Afghan National Army (ANA) 95 Kabul 53, 94, 97 African Union 6 Alba, Operation 63, 64 Albania 28, 58, 62, 63, 64, 65, 105, 110, 111 Allied Command Europe (ACE) Mobile Force (Land) (AMF(L)) 33 Allied Command Europe (ACE) Rapid Reaction Corps (ARRC) 34, 35, 37, 46, 51, 58, 59, 63, 66, 72, 76, 77 Allied Command Operations (ACO) 30 Allied Command Transformation (ACT) 30, 43 Asmus, Ronald D. 40, 95, 109 Aspin, Les (former US Secretary of Defense) 40 Athens 9, 81 Austria 13, 23, 63 Avant, Deborah 4, 19, 109 Aznar, José Maria (former Spanish prime minister) 73
Canary Islands 43, 79, 80 Central Europe 1, 29, 34, 45, 56 Chiefs of Defense (CHODs) 30, 37 Civil-Military Cooperation (CIMIC) 66, 99, 100 Clausewitz, Karl von 23 Clinton, William J. (former US President) 40 Combat Service Support 6, 20, 36, 37, 43, 45, 50, 51, 58, 62, 66, 69, 77, 79 Combat Support 6, 37, 43, 45, 50, 58, 62, 69, 79, 96 Combined Joint Task Force (CJTF) 39, 110, 111 Command Structure 42, 109, 112, 115 Component Command (CC) 44 Congress 35, 88, 109, 116 Counterterrorism 89, 94 Croatia 58, 64 Czech Republic 42, 93
Balkans 3, 6, 13, 45, 46, 58, 63, 65, 70, 82, 94, 96, 100, 101 Basic Law (Germany) 67 Belgium 12, 13, 14, 29, 30, 34, 45, 74, 75, 94, 96, 110, 112, 115 Benelux (Belgium, the Netherlands, and Luxembourg) 38, 45 Bennett, D. Scott 3, 10, 27, 109 Bosnia 3, 28, 40, 41, 46, 58, 60, 63, 64, 65, 66, 70, 71, 92, 109, 117 British Army of the Rhine (BAOR) 37, 77 Brussels Treaty 57 Bulgaria 93 Bundeswehr (German Armed Forces) 45, 67
Danish-German Corps 34 Darfur 6, 92 Dayton Accord 41 Defense Capabilities Initiative (DCI) 61, 116 Defense Planning Committee (DPC) 22, 38, 39, 40, 42, 45, 111 De Hoop Scheffer, Jaap (Secretary General of NATO) 92, 93, 94, 95, 96, 110, 115 Delian League 9 Denmark 34, 63, 93 Deployability 61 Deployable Headquarters Task Force (DHQTF) 51 Deputy Supreme Allied Commander Europe (DSACEUR) 94 DPC (Defense Planning Committee) 22, 30, 33, 39, 60, 110
C4ISR (Command, Control, Communications, Computers, Intelligence, Surveillance, and Reconnaissance) 99
Eastern Europe 29, 41, 55 Eichenberg, Richard 25, 110 Eurocorps 6, 51, 74, 75 European Union (EU) 69, 74, 100, 107
120
Alliance Management and Maintenance
Finer, Samuel 17 Forces at Lower Readiness (FLR) 47, 50 Force Structure 38, 45, 114, 115 France 2, 13, 14, 19, 22, 23, 29, 30, 35, 39, 41, 45, 50, 51, 63, 68, 73, 74, 75, 77, 87, 89, 93, 112, 113, 114, 115 Franco-Belgian Accord 12, 13, 14 Fulda Gap 1 Galvin, John (former SACEUR) 37 George, Alexander 22 German-Netherlands Corps 34 Germany 1, 2, 12, 13, 14, 19, 23, 29, 33, 34, 37, 38, 39, 45, 46, 48, 49, 50, 51, 56, 62, 63, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 87, 110, 112, 114, 115 Basic Law 67 Bundeswehr (German Armed Forces) 45, 67 Gibraltar 80, 81, 82, 86, 115 Graduated Readiness Force (GRF) 51 Greece 9, 26, 29, 43, 47, 51, 63, 74, 81, 82, 93, 112 Haas, Ernst 16, 17 High Readiness Force (HRF) 2, 7, 46, 57, 104, 105 Holbrooke, Richard C. 95, 109 Human Intelligence (HUMINT) 91 Hungary 42, 93, 116 Iberian Atlantic Command (IBERLANT) 79, 80 IFOR 40, 41, 45, 46, 65, 66, 113, 117 Immediate Reaction Force (Land) (IRF(L)) 33, 34 Implementation Force (IFOR) 40, 41, 45, 46, 65, 66, 113, 117 Intelligence Fusion Center 91 International Military Staff (IMS) 15, 47, 48, 68 International Security Assistance Force 6, 31, 52, 94, 95, 97, 98, 102, 105, 107, 113 Intra-Alliance Bargaining 67, 74 Iran 57 Iraq 6, 45, 92 Italy 14, 50, 51, 52, 53, 59, 63, 68, 77, 78, 93, 111, 113 Joint Sub-Regional Command (JSRC) 44 Jones, James L. (former SACEUR) 95, 98
Kaufman, Stuart 19, 113 Kay, Sean 25, 26, 113, 115 KFOR 45, 97 Kier, Elizabeth 19, 113 Kohl, Helmut (former German chancellor) 74 Kok, Wim (former Dutch prime minister) 72 Kosovo 44, 58, 60, 61, 69, 92, 97, 112, 116 Kosovo Force (KFOR) 45, 97 Liska, George 9, 114 Long-Term Study (LTS) 38, 39, 40, 41, 42, 44, 45 Luxembourg 14, 29, 35, 75 Macedonia 63 Marshall Plan 29 Marten, Kimberly Zisk 19, 114 MC 161 (NATO Terrorism Threat Assessment) 89 MC 317 (Alliance Force Structure) 20, 47, 48, 49 MC 324 (Alliance Command Structure) 20, 44, 83 MC 400 (Military Implementation of the Alliance’s Strategic Concept) 20, 21, 40, 42, 47, 48, 49 MC 472 (Military Concept for Defence Against Terrorism) 89, 91 Military Committee (MC) 20, 30, 38, 39, 40, 42, 44, 47, 48, 49, 56, 78, 86, 89 Military Committee Working Group – Strategic Issues (MCWG-SI) 47, 48 Military Representative (MilRep) 39, 47, 48 Military Transitional Issues Working Group (MTIWG) 39, 41, 42, 43, 47 Mitterrand, François (former French president) 74 Molesworth Air Force Base (UK) 91 Mons 8, 30, 40, 47 Multinational Division Central (MND(C)) 34 National Security Strategy (NSS) (US) 88 NATO Secretary General 15, 26, 30, 31, 61, 65, 85, 92, 93, 94, 95, 96, 99, 113, 115 Strategic Concept 2, 3, 32, 35, 36, 38, 39, 40, 41, 42, 48, 55, 56, 57, 58, 60, 65, 104, 106, 115
Index NATO Rapid Deployment Corps (NRDC) 1, 6, 7, 49, 51, 52, 66, 70 ARRC (Britain) 34, 35, 37, 46, 51, 58, 59, 63, 66, 72, 76, 77 NRDC-EC (Eurocorps) 51 NRDC-GNL (Germany-Netherlands) 51 NRDC-IT (Italy) 51, 52, 53 NRDC-SP (Spain) 51 NRDC-T (Turkey) 51, 52 NATO Response Force 7, 31, 90, 94, 98, 99, 102 Naumann, Klaus (former Chairman of the Military Committee) 48 Netherlands, The 13, 29, 33, 34, 45, 50, 51, 52, 62, 68, 71, 72, 74, 76, 78, 80, 93, 97, 109, 112, 116 Norfolk 30, 40, 47 North Atlantic Cooperation Council 39, 41, 110 North Atlantic Council (NAC) 15, 29, 30, 39, 41, 49, 65, 86, 87, 88, 90, 99, 100, 110, 111, 115 North Atlantic Treaty (also see Treaty of Washington) 1 Norway 26, 35, 45, 62, 74, 93, 116 NRDC Eurocorps 6, 51, 74, 75 German-Netherlands Corps 34 Olson, Mancur 10, 115 Operational Mentoring and Liaison Teams (OMLT) 95 Organizational Studies 15 Partnership for Peace (PfP) 41, 65, 115 Peloponnesian League 9 Persia 9 Poland 39, 42, 45, 50, 70, 93, 112 Portugal 35, 43, 63, 79, 80 Posen, Barry 2, 4, 17, 19, 58, 103, 115 Prague Capabilities Commitments (PCC) 61, 62 Provincial Reconstruction Teams (PRT) 95 Psychological Operation (PSYOP) 66 Regional Commands RCC Brunnsum 7, 53 RCC Lisbon 7, 79 RCC Naples 7, 44, 82 Riga 2, 31, 91, 94, 95, 96, 97, 98, 99, 109, 115
121
Robertson, George (former Secretary General of NATO) 61, 85, 99 Romania 63, 93 Rosen, Stephen Peter 19, 115 Russia 1, 12, 13, 23, 40, 45, 56, 57, 58, 93, 114 Secretary General 15, 26, 30, 31, 61, 65, 85, 92, 93, 94, 95, 96, 99, 113, 115 SFOR 45, 96, 113 Shalikashvili, John (former SACEUR) 48, 60 Signals Intelligence (SIGINT) 99 Slovenia 58, 63, 64, 93, 95, 113 Snyder, Glenn 3, 4, 10, 23, 24, 25, 27, 116 Solana, Javier (former Spanish prime minister) 73 Soviet Union 1, 5, 18, 27, 28, 29, 33, 41, 55, 56, 57, 58 Spain 33, 43, 51, 59, 62, 63, 68, 73, 75, 77, 78, 79, 80, 81, 114 Stabilization Force (SFOR) 45, 96, 113 Steadfast Jaguar (NRF Exercise) 98 Strategic Concept 2, 3, 32, 35, 36, 38, 39, 40, 41, 42, 48, 55, 56, 57, 58, 60, 65, 104, 106, 115 Summers, Harry 4 Supreme Allied Commander Atlantic (SACLANT) 43, 79, 80, 84 Supreme Allied Commander Europe (SACEUR) 37, 40, 48, 51, 95, 97, 98 Supreme Headquarters Allied Powers Europe (SHAPE) 44, 48, 49, 51, 53, 80, 98, 99, 113 Sustainability 61 Sweden 39, 93 Syria 45 Thies, Wallace 10, 27, 43, 116 Thucydides 9 Treaty of Washington 15, 29, 32, 34, 57 Article 5 1, 12, 17, 18, 21, 32, 33, 36, 40, 42, 46, 47, 48, 76, 85 Turkey 6, 26, 27, 29, 35, 37, 43, 45, 47, 51, 52, 57, 63, 70, 75, 76, 78, 81, 82, 93, 112, 113, 117 Union of Soviet Socialist Republics (USSR) 1, 5, 18, 27, 28, 29, 33, 41, 55, 56, 57, 58 United Kingdom 14, 17, 23, 29, 33, 35, 37, 43, 45, 48, 50, 63, 67, 68, 70, 74, 76, 77, 78, 79, 80, 81, 82, 91, 93, 97, 98, 110, 111, 112, 113, 115
122
Alliance Management and Maintenance
United Nations (UN) 40, 63, 64, 65, 66, 68, 71, 87, 88, 90, 100, 107 UN Protection Force (UNPROFOR) 40, 46, 71 V Corps (US) 46 Walt, Stephen 3, 10, 116 Warsaw Pact 1, 33, 41, 55, 56 Western Europe 26, 29, 35, 60, 61, 64, 68, 104
Western European Union 35, 67, 114 Western Union (forerunner to NATO) 29 Wittman, Klaus 55 Wörner, Manfred (former Secretary General of NATO) 26, 27, 65, 117 Yugoslavia 40, 58, 63, 64, 65, 66, 68, 105 Zeckhauser, Richard 10, 115