Secret Intelligence in the European States System, 1918-1989 9780804788915

A collection of essays that analyze the recent evidence concerning the history of the European state system of the last

195 32 2MB

English Pages 256 [254] Year 2013

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Secret Intelligence in the European States System, 1918-1989
 9780804788915

Citation preview

SECRET INTELLIGENCE IN T H E EU R O P E A N S TAT E S S YS TEM, 1918 –1989

SECRET INTELLIGENCE I N T H E EU R O P E A N S TAT E S S YS TEM, 1918 –1989 Edited by Jonathan Haslam and Karina Urbach

Stanford University Press Stanford, California

Stanford University Press Stanford, California © 2014 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press. Printed in the United States of America on acid-free, archival-quality paper Library of Congress Cataloging-in-Publication Data   Secret intelligence in the European states system, 1918–1989 / edited by Jonathan Haslam and Karina Urbach.   pages cm   Includes bibliographical references and index.  ISBN 978-0-8047-8359-0 (cloth : alk. paper)   1. Intelligence service—Europe—History—20th century. 2. Secret service— Europe—History—20th century. 3. Europe—Foreign relations—1918–1945. 4. Europe—Foreign relations—1945– I. Haslam, Jonathan, editor of compilation. II. Urbach, Karina, editor of compilation. D424.S37 2014 327.124009'04—dc23 2013013384 ISBN 978-0-8047-8891-5 (electronic)

Contents

Contributors vii Introduction: The Role of Secret Intelligence in the International Relations of Europe in the Twentieth Century Jonathan Haslam and Karina Urbach

1

1 “Humint” by Default and the Problem of Trust: Soviet Intelligence, 1917–1941 Jonathan Haslam

12

2 Barbarossa and the Bomb: Two Cases of Soviet Intelligence in World War II David Holloway

36

3 Seeking a Scapegoat: Intelligence and Grand Strategy in France, 1919–1940 Stephen A. Schuker

81

4 French Intelligence About the East, 1945–1968 Georges-Henri Soutou

128

5 British Intelligence During the Cold War Richard J. Aldrich

149

6 The Stasi Confronts Western Strategies for Transformation, 1966–1975 Oliver Bange

170

vi

Contents

7 The West German Secret Services During the Cold War Holger Afflerbach Index

209 231

Contributors

is Professor of Central European History at the University of Leeds and the author of Falkenhayn: Politisches Denken und Handeln im Kaiserreich (Munich, 1994), a biography of the World War I chief of the German General Staff Erich von Falkenhayn, and of a study of the Triple Alliance, Der Dreibund: Europäische Grossmacht- und Allianzpolitik vor dem Ersten Weltkrieg (Vienna, 2002). He is an expert on World War I and is currently writing a book on the art of defeat. holger afflerbach

richard j . aldrich

is Professor of International Security at the University of Warwick. He has written widely on intelligence issues. His books include Intelligence and the War Against Japan: Britain, America and the Politics of Secret Service (Cambridge, 2000) and GCHQ: The Uncensored Story of Britain’s Most Secret Intelligence Agency (London, 2010). He co-edits the series Studies in Intelligence with Christopher Andrew. oliver bange studied at the LSE and Mannheim University. He is Privatdozent at the Military Research Council of the German Army (MGFA) at Potsdam. He has published several monographs on the subject of détente and on Anglo-German postwar relations, among them The EEC Crisis of 1963—Kennedy, Macmillan, de Gaulle and Adenauer in Conflict (Basingstoke, UK, 2000).

is a Fellow of the British Academy and Professor of the History of International Relations at Cambridge University, where he is also a Fellow of Corpus Christi College. His latest work is Russia’s Cold War: From jonathan hasl am

vii

viii

Contributors

the October Revolution to the Fall of the Wall (New Haven, CT, 2011), and he is currently researching and writing a history of the Soviet intelligence services. is Raymond A. Spruance Professor of International History at Stanford University. His book Stalin and the Bomb: The Soviet Union and Atomic Energy, 1939–1956 (New Haven, CT, 1994). Holloway also wrote The Soviet Union and the Arms Race (New Haven, CT, 1983) and co-authored The Reagan Strategic Defense Initiative: Technical, Political and Arms Control Assessment (Stanford, CA, 1984).

david holloway

stephen a . schuker

is William W. Corcoran Professor of History at the University of Virginia. He is the author of The End of French Predominance in Europe: The Financial Crisis of 1924 and the Adoption of the Dawes Plan (Chapel Hill, NC, 1976) and American “Reparations” to Germany, 1919–33: Implications for the Third-World Debt Crisis (Princeton, NJ, 1988), and the editor of Deutschland und Frankreich: vom Konflikt zur Aussöhnung: Die Gestaltung der westeuropäischen Sicherheit, 1914–1963 (Munich, 2000). georges - henri soutou

is Professor Emeritus at the University of Paris–Sorbonne (Paris IV). He works on the history of the First World War, FrancoGerman relations, and the Cold War. His major publications related to this book are L’alliance incertaine: Les rapports politico-strategiques franco-allemands, 1954–1996 (Paris, 1996) and La guerre de cinquante ans: Les relations Est-Ouest, 1943–1990 (Paris, 2001). is a Senior Research Fellow at the Institute of Historical Research, University of London. Her Cambridge PhD was published as Bismarck’s Favourite Englishman: Lord Odo Russell’s Mission to Berlin (New York, 1999). She has published extensively on Anglo-German diplomatic and social history (including a biography of Queen Victoria) and has completed a monograph titled “Go-Betweens for Hitler: German Aristocratic Families in Europe, 1900 to 1939,” to be published by Oxford University Press in 2014.

k arina urbach

Introduction The Role of Secret Intelligence in the International Relations of Europe in the Twentieth Century Jonathan Haslam and Karina Urbach

T

his book takes a look at the role of secret intelligence in the history of the European states system of the twentieth century. The reader may reasonably object that this has been done before. The last serious attempt, Knowing One’s Enemies: Intelligence Assessment Before the Two World Wars, edited by Harvard’s Ernest R. May, appeared some thirty years ago, however, and it was necessarily restricted in scope because of the absence of solid documentation for the Cold War. May later despaired that his book was scarcely ever cited, that the revolution in intelligence studies was largely selfcontained, and that it had not had much effect outside its own inner circle. If this is true, something can be said for another attempt, and one that reaches into the Cold War. Although limited in impact, three decades of further research have greatly advanced our understanding of intelligence history. Specialist journals have become well established; work of real value has appeared. Traditional perspectives have given ground, and judgments have been revised in the light of new evidence. Progress has nonetheless inevitably been limited. Direct access to the entire torso of secret intelligence, rather than choice cuts, is still well beyond reach. Yet even though access to information is incomplete, one is struck by just how much attention the history receives from major intelligence agencies: the SVR (Sluzhba Vneshney Razvedki, or Foreign Intelligence Service, the KGB’s successor operating abroad) and the CIA both have websites of some significance devoted to the subject. And both MI5 and MI6, neither noted

1

2

Introduction

for openness, have published official histories (that of the former is infinitely better than that of the latter).1 Whether this degree of attention arises from understanding the relevance of intelligence history to current and future needs or from a self-conscious preoccupation with institutional reputation is a moot point; possibly both. Intelligence releases at times appear to represent a continuation of secret warfare by other means. To the detriment of scholarship worldwide, the untrammeled capacity to withhold information is a crucial weapon in the armories of the great powers. How and when did all this start? The unclassified study of intelligence history originated, against all the odds, with committed amateurs on both sides of the Atlantic: notably the American journalist David Kahn and the British historian of French diplomacy Christopher Andrew, alike preoccupied with decryption. Both initially drew active encouragement from Sir Harry Hinsley after performing at his legendary seminar at St John’s College, Cambridge, during 1974–1975. And it was not long before others also took a keen interest, including May, who had long acted discreetly for the U.S. government in such matters. A curious but fruitful result was the coincidental appearance in 1984 of both The Missing Dimension: Governments and Intelligence Communities in the Twentieth Century, edited by Christopher Andrew and David Dilks in Britain, and Knowing One’s Enemies: Intelligence Assessment Before the Two World Wars, edited by Ernest May in the United States. Of the two, May’s was undoubtedly the richer and has had the longest shelf-life because of its coherence. Yet for all of the coverage that Knowing One’s Enemies undoubtedly obtained, and the mass of recruits subsequently taking up the subject of intelligence history, exposure of this missing dimension was never welcomed universally. For some, no sacrifice was entailed. May’s career had already peaked. Andrew’s had not, however. Arguably, it took longer to do so for the very reason that he championed intelligence history with undaunted energy and enthusiasm. The situation that arose was not unlike that in sixteenth-century Europe. In those turbulent days, undesirable public attention to the secrets contained in reasons of state was priggishly compared to tastelessly exposing the pudenda of a woman. But in our case the missing dimension has not so much aroused moral outrage so much as “sniffy” indifference among more orthodox historians. The critics may nonetheless have had a point. In common with reasons of state, the history of intelligence has undoubtedly suffered from excessive



Introduction

3

claims for its capacity to explain. And the ambitious academic is all too tempted to see the only sources available—few and far between—as necessarily good sources. All too often, intelligence sources have been seized upon with alacrity but uncritically by those without sufficient knowledge and attention to the wider diplomatic and sociocultural context, the exploration of which is itself a massive task, beyond the skills of most enthusiasts. Intelligence history has arguably failed to integrate itself into the mainstream by not taking diplomatic and military history sufficiently seriously on their own terms. The study of intelligence history is, after all, no less dependent on a direct understanding of foreign countries and access to their culture than is the history of international relations as a whole, and that requires competence in foreign languages and a firsthand understanding of how other societies work. These have been at a premium since the passing of the World War II generation of professors, with their linguistic expertise and life experience overseas. Youthful scholars are temperamentally averse to devoting sufficient time in appropriate training to acquire the tools that would ultimately give their research true meaning. Moreover, the general run of international relations history and, indeed, military history by comparison appears a good deal less interesting; staid, indeed; certainly less exotic, some—most social historians we know—would say boring. Simplistic reductions of diplomatic history to ready-made formulae are all too familiar among incautious political scientists indifferent to and ignorant of the texture of the past. The reduction of international relations history to intelligence history is equally distorting. The appropriate diplomatic backdrop against which intelligence history can be portrayed is vital—even if only in faint but accurate silhouette. By this alone can its value be judged. Without it, the intelligence product, once acquired with so much ingenuity and at times in great peril, may by virtue of the unusual status attached to it receive more importance than it would otherwise merit. If true in general, this danger is even more real in the case of the twentieth century, a fascinating but challenging era. For the historian, the inherent complexity of its international relations has been compounded by a torrent of once secret documentation on all sides. The accessibility of archival sources has expanded at an ever greater pace since the 1960s. The opening of top secret documents following the fall of Communism in Central and Eastern Europe in 1989 turned a swelling stream into a veritable flood.

4

Introduction

Undeterred because largely ignorant, many have been drawn to this recondite world despite rather than because of the novelist John le Carré and his much publicized moral misgivings, as though in search of an alternative vocation to be experienced vicariously through the safety of the text, rather than the precarious reality of life as a spy or the choice of life behind a desk in the backrooms of public service. Secret intelligence and its acquisition are, after all, more often humdrum prose than inspirational poetry. Ignorant of this core truth, isolated from the mainstream, and contrary to May’s initial fears, some enthusiasts, undaunted by its inherent limitations, have threatened to turn the once “missing dimension” into the dominant, over-determining dimension; indulging it excessively as though to compensate for former neglect, yet all too often basing assertion on thin evidence or unearthing reams of detail from available archives (one thinks in particular of the East German collection) of remote interest to anybody. An innovation such as intelligence history will undoubtedly attract more than a fair degree of attention among the public, not least because it sells books (except in today’s Russia). It also exerts a spellbinding effect on those naturally attracted to the excitement of thrillers and novelty for its own sake. Here, of course, academics are no different from anyone else. The glittering prospect of a sensationalist discovery somehow seems more likely in the field of secret intelligence than in the case of day-to-day diplomacy; but, as the forty-niners discovered in California, fools’ gold can be found in abundance by the impatient and untutored eye. The late Bill Odom, head of the National Security Agency under U.S. President Ronald Reagan, liked to tell the following joke: A balloonist had lost his way and lowered his balloon to try to reorient himself by taking a closer look at the ground. Still disoriented, he dropped down farther in order to ask someone on the ground. He spotted a man on a bicycle and yelled to him, “Where am I?” The cyclist stopped, looked up in puzzlement, and answered, “You’re in a balloon.” The balloonist replied, “But more precisely, where am I?” The cyclist answered, “In a basket under a balloon.” Frustrated by these evasive answers, he reversed this question and asked, “Where are you?” The cyclist replied, “On a bicycle.” Exasperated, the balloonist reacted, “You must be an intelligence officer.” The cyclist was now puzzled and asked, “How did you know?” The balloonist said, “You give accurate but useless answers to my questions.” The history of intelligence, therefore, must be the record of accurate and but useless answers to ambiguous questions about the adversary.2



Introduction

5

Beyond the world of academia and in the secret world, the separation of collection from analysis has proved a vital safeguard against temptation. That, of course, cannot be true of historians, who have both jobs to do: analysis is for them intricately bound up in narrative. Inadequate grounding in the larger context then makes objective historical explanation more difficult. It most certainly renders dismissal of intelligence reports by statesmen utterly incomprehensible. As a result some historians have been tempted to retreat for explanation into a deus ex machina or plain moral condemnation, which, for Western historians of Russia, is always the weapon of last resort. Here the most famous example of this kind, Stalin’s blindness to the forthcoming invasion of the Soviet Union in June 1941, immediately springs to mind. But more of that later. By the same token, the separation of the user from the analyst and the protection of the analyst from the consumer have proved no less important. The failure to abide by these rules of thumb frequently led to disaster. Not least for these reasons, historians who had previously neglected intelligence as a factor worthy of attention could be expected to play down its importance. As with all innovations, therefore, a balance must be struck between clinging to the dull rigors of tried and tested methods and abandoning the apparent drudgery of old-fashioned research in diplomatic history for a single key that promises to unlock all the arcana imperii, or at least the most important. Sir Harry Hinsley, a sage official historian of British intelligence during World War II, did indeed worry lest the ingenuous drawn to the romance of the secret world would all too easily be seduced by the intelligence community, who could leak secrets selectively for their own undisclosed and possibly nefarious purposes; at the very least black propaganda. For the dependent historian privileged with scraps of secret intelligence from the top table, it is difficult to bite the hand that feeds the hungry. Due to his wartime career as personal assistant to the head of the Government Code and Cypher School—the British code-breaking effort—Hinsley necessarily considered secret intelligence a factor critical to war and international relations. This opinion was, of course, subject to the important caveat that the value and justification of intelligence depend on the use that is made of its findings.3 Experienced diplomats, however, with no firsthand involvement in collection frequently found the product difficult to evaluate because of its uncertain provenance. If analysis without knowledge of provenance is problematic, then policy prescription without accurate analysis is folly.

6

Introduction

Memory of that notorious forgery the Zinoviev Letter, carelessly acquired for cash from a Russian émigré by MI6 in 1924 and falsely attested to by the Foreign Office’s much respected permanent undersecretary Sir Eyre Crowe, left an indelible imprint upon British diplomats who were privy to these events at close quarters.4 Robert Jervis alludes to another factor: “Foreign ministries, of course, carry out diplomacy, and so one might think that they would value intelligence more highly because of its closer links to this mission. But most diplomats have prided themselves on being generalists and have tended to believe, often correctly, that they can understand other countries better than can specialised intelligence officers.”5 Jervis also rightly points out that diplomats are naturally ill-disposed toward intelligence officers, whose activities sometimes lead to complications in relations with the host government; at its worst a breach in relations and withdrawal of the embassy. George Kennan needed no clandestine help to predict Stalin’s behavior, which certainly mattered, because no secret intelligence of this kind was available. Under President Harry Truman, he played a major role in spawning covert operations, which he soon very much had cause to regret. Reacting later against what he saw as excessive attention to secret intelligence, Kennan insisted “that the involvement of our government in the acquisition of secret intelligence, by espionage and other unavowed processes, while perhaps occasionally unavoidable, has had ascribed to it a degree of importance far greater than it deserves.”6 This sentiment was frequently echoed by Odom. Moreover, Gordon Barrass, a veteran of MI6 and formerly chief of the assessments staff at the British cabinet office, offers judicial counsel to the unwary doubtless borne of sobering experience. A firm advocate of “good intelligence,” without which “policy is all too easily shaped by fear, ignorance or optimism,” Barrass—by trade and therefore by inclination a strong proponent of human intelligence gathering—strongly implies that learning foreign languages and imbibing alien cultures—both traditionalist pursuits for the diplomatic historian—are vital to correct assessment of intelligence. And for good reason: “every society and bureaucracy has its own mantras, rhetoric, and conventions of political correctness.”7 These limitations undermining the value of intelligence received in-depth analysis from Richard Betts shortly before intelligence history took off. In a pioneering article on “Analysis, War and Decision: Why Intelligence Failures



Introduction

7

Are Inevitable,” Betts took us a dimension beyond the soul-searching after the coup that overthrew Salvador Allende in Chile in 1973. Betts had been intimately involved on the staff of the Senate Select Committee to Study Governmental Operations with Respect to Intelligence Activities under the chairmanship of Senator Frank Church (D–Idaho), investigating the misdemeanors of various U.S. administrations—the so-called “crown jewels”—in 1975. What Betts analyzes is “the inadequacy of intelligence,” a feature he finds intractable. Why does intelligence fail? In his view: “In the best-known cases of intelligence failure, the most crucial mistakes have seldom been made by collectors of raw information, occasionally by professionals who produce finished analyses, but most often by the decision makers who consume the product of intelligence services.” In short, “Policy premises constrict perception, and administrative workloads constrain reflection. Intelligence failure is political and psychological more often than organizational.” 8 This much is highlighted in the present book by Stephen Schuker in chapter 3 with respect to France in 1940 and by David Holloway in chapter 2 with respect to Russia in 1941. Jervis, who drew Betts to Columbia University, opened his own academic career applying psychology to the study of international relations. He echoes Sun Tzu to the effect that “to understand each other’s behavior, decisionmakers usually have to understand how their own state is acting and how others see them. Although this would seem easy, in fact it is not. States have powerful and idealized, if not self-serving, self-images; they follow double standards and rarely appreciate the extent to which they menace others’ interests.”9 The inner world of East Germany’s secret services outlined by Oliver Bange in chapter 6 certainly fits that pattern. Trenchantly expressing the utmost skepticism, Jervis argues that “The basic outlines of threat assessment . . . are rarely in the province of intelligence, even though one could argue that this should be its most important function.” They belong to the realm of policy, and policy is in the hands of others. Here the answers determine the selection of the evidence rather than vice versa. Thus “all too often . . . intelligence estimates tell us more about interests and foreign policy preferences of powerful groups than . . . about what the other side’s intentions and capabilities are.” Jervis thus safely concludes: “We should not expect too much of intelligence.”10 In chapter 7, Holger Afflerbach provides ready illustration from the history of the Gehlen Organization, the predecessor of the BND (Bundesnachrichtendienst, or Federal Intelligence Service) in West Germany.

8

Introduction

To say the least, the jury is evidently out as to how far secret intelligence is valuable or critical to the successful conduct of international relations, even among those formerly so intimately involved. From the operational point of view, what historians are doing is mirroring in the past the perennial dilemmas of the present. Thus what they have to say about the record should be of more than passing interest to present practitioners. It is not accidental that the first major work addressing the impact of secret intelligence on the history of international relations edited by May was funded by a rare figure unparalleled elsewhere: the enlightened patron of open-minded thinking heading assessments at the office of the Secretary of Defense, Andrew Marshall. The essays below address both political and military intelligence, and strategic rather than field intelligence. Gathering intelligence comes through two means: human intelligence (humint) and signals/communication intelligence (sigint/comint). The preference for one—comint—over the other— humint—is often a matter of educational strength and tradition (preeminence in mathematics, for example, in Sweden, Poland, and Russia) as well as practicality (such as direct access to the best technology, which can be denied the adversary: the United States stands head and shoulders above others in this regard). Though, as Jonathan Haslam demonstrates in chapter 1, in Stalin’s Russia, the vision of intelligence grew out of very recent revolutionary experience and operations against the counterrevolution, so humint always prevailed over sigint. Whatever is gathered has, of course, to be interpreted. Cryptography is of no value without cryptanalysis. This process is critical to correct application. Otherwise what is collected will be wasted or, worse still, misinterpreted. Critical to assessment, Barrass reminds us, are the ethnic, cultural, and ideological prisms through which intelligence is interpreted. In this respect, intelligence assessment is a process no different from assessment of information coming in from diplomatic or, indeed, open sources. We have to imitate the anthropologist. We are not, as political scientists and economists would have it, naturally “rational” beings whose behavior can be predicted according to an abstract logic, but beings who have to be tutored to reason in an often unreasoning world. We are looking at the role of intelligence in the history of international relations throughout the twentieth century in order to assess its impact over time. The problem of measuring impact is that our access to reliable sources is still highly restricted, and uneven between states and uneven over time. As



Introduction

9

U.S. Secretary of Defense Donald Rumsfeld reminded us: there are known unknowns and unknown unknowns. It is always hard to know what is being denied us and why (the unknown unknowns), though taking the long view with access to files more plentiful for the earliest period gives us a sense of what is missing (the known unknowns) than would have been obtained by focusing on a narrower and more recent period. Drawing lessons from a study such as this is no easy matter. We nonetheless believe that insights may be obtained for the current world by adopting a larger historical perspective. The essays here presented offer an array of insights horizontally across countries and vertically through time. But all agree on the importance of not overestimating the contribution to be expected from secret intelligence. This is easier to see when it proves impossible to penetrate the enemy camp for direct access to the intentions of the opponent. In chapter 4, Georges-Henri Soutou emphasizes the problem postwar France faced in this respect and then illustrates the necessary improvisation that followed from it. Absent knowledge of what the adversary intended, the searchlight was focused on outward signs of enemy movement in the field of battle, which became something of an art form at allied intelligence in West Berlin. The essays also concur on the damaging impact of high politics on the processes of gathering and analyzing the resulting product. But this was only one element, though crucial, produced by the larger material, as well as ideational context in which intelligence agencies had to operate. And here the dominant image of the massively invested U.S. intelligence services, or, indeed, that of the Russians in the Cold War, proves entirely misleading when observing even those of the great powers in Europe, where limited means required making choices between intelligence and defense capabilities as a whole, and between defense and diplomacy. As Richard Aldrich highlights for us in chapter 5, secret intelligence operates out of neither a social nor an economic vacuum. All too often, purist approaches to the subject modeled on postwar American behavior ignore that fundamental truth. In an unusually well-documented contribution on Russia in chapter 2, Holloway argues the reasons for Stalin’s unpreparedness in 1940–1941. This, it is fair to say, has long been an obsessive object of interest and study in Russia, because it made all the difference to the course of the war that followed from June 1941. Rather than rushing into moral judgments about the régime, Holloway instead takes a cool look at what information came in to Stalin and allows for the fact that not all the incoming intelligence data were consistent.

10

Introduction

And in respect of revelations about U.S. construction of the atomic bomb, Holloway shows that Stalin once again consciously distanced himself from the findings of the intelligence services in reaching a final judgment. Whereas in the former case, it nearly led to disaster; in respect of the latter, Stalin was undoubtedly correct. Bange, looking at the DDR, points out how intelligence gathering and analysis under dictatorship is hindered by mirror-imaging the adversary’s intentions, and thereby misdirecting an expensive intelligence effort to futile ends, while consuming precious, limited material and human means. The East German regime, here as elsewhere, undermined its own security through a lack of self-questioning and subordination to doctrine. For those dubious about the role that intelligence can play even in democracies, Schuker—a skeptic—has a great deal of interest to say in chapter 3. In his view, only when material military power stands in balance can intelligence make a difference; and in 1940, along with much else in Paris, French intelligence failed. France, of course, surrendered very early on; arguably due to deep-seated problems within French society that Hitler sensed instinctively. Was it therefore a surprise that French intelligence suffered as a consequence? Intelligence can never be better than the context that confines it and from which it operates. This is the point of departure for Afflerbach in chapter 7, examining the close relationship between the postwar problem of recreating the German state after defeat in war but with the limited resources, in terms of personnel, inherited in large part from the previous and now utterly discredited regime. Britain had no such disadvantages. But its relative success occurred against a background of tight secrecy. Thus our contributor here has artfully had to find something of a back door into intelligence by asking about money: “cuts and economies can illuminate what intelligence was provided and at what cost,” Aldrich persuasively argues in chapter 5. This approach allows him to survey the Anglo-American relationship from the vantage point of Britain’s diminishing resources in the face of the need to come up with a steady flow of results to sustain a crucial strategic alliance. Scant resources were, of course, the major problem for postwar France, as Soutou relates in chapter 4; particularly acute given, not only the magnitude of objectives originally laid down by General de Gaulle, but the poverty of the budget available even after he resumed power. The response was quite naturally to focus on the material and the measurable in terms of enemy



Introduction

11

capabilities rather than the ephemera that emerged when aiming at higher targets relating to intention. These various chapters all highlight the critical importance of the prevailing circumstances in which intelligence operates. It appears that the context of relevance for effectiveness is primarily but not entirely domestic. It includes the impact of technology on encryption and decryption; the form of government and nature of decision-making; ideological and cultural pressures on the collation and analysis of information; limitations imposed by financial stringency; and the deadweight of tradition (fighting the last war). Such elements may exert an unseen influence on daily practice, but their cumulative effect overall may prove critical in undermining efficiency. An understanding of them requires interdisciplinary expertise. The foreign context is arguably much more simple and no different from that in warfare: namely, the strength of the intelligence service relative to the adversary’s. If anything, this underlines a core truth that Machiavelli so well understood: the more formidable the adversary, the greater the importance of putting one’s own house in order. Notes 1. K. Jeffrey, MI6: The History of the Secret Intelligence Service, 1909–1949 (London, 2011); C. Andrew, The Defence of the Realm: The Authorized History of MI5 (London, 2010). 2. MS in the authors’ possession. 3. See F. H. Hinsley’s lecture on “The Influence of Ultra in the Second World War,” Intelligencer 14, no. 2 (Winter–Spring 2004): 463–465; reprinted but without the exchanges with the audience, in F. Hinsley and A. Stripp, Codebreakers: The Inside History of Bletchley Park (Oxford, 1993), 1–13. 4. The late former diplomat and historian E. H. Carr was one such. 5. R. Jervis, “Intelligence and Foreign Policy: A Review Essay,” International Security 11, no. 3 (Winter 1986–1987): 141–161. 6. G. Kennan, Around the Cragged Hill: A Personal and Political Philosophy (New York, 1993), 209. 7. G. Barrass, The Great Cold War: A Journey Through the Hall of Mirrors (Stanford, CA, 2009), 407. 8. Betts, “Analysis, War and Decision: Why Intelligence Failures Are Inevitable,” World Politics 31, 1 (October 1978): 61–89. 9. Jervis, “Intelligence and Foreign Policy”; and see also Jervis, Why Intelligence Fails: Lessons from the Iranian Revolution and the Iraq War (Ithaca, NY, 2010). 10. Jervis, “Intelligence and Foreign Policy.”

1

“Humint” by Default and the Problem of Trust Soviet Intelligence, 1917–1941 Jonathan Haslam As far as I am concerned, reliance upon secret intelligence also carries little conviction. —People’s Commissar for Foreign Affairs Maxim Litvinov, 11 April 1939

T

he Bolsheviks who seized power in Russia under Lenin on 7 November 1917 believed they had no need of a foreign intelligence network for the world socialist revolution to triumph, because hostile capitalist states would by definition disappear.1 Indeed, the Bolsheviks expected the focus of world revolution to move from backward Russia to advanced, industrialized Germany. The revolution was thus at this early stage deeply internationalist in outlook: hence the creation of the Communist International (Comintern) in March 1919. “From provincial Moscow, from half-Asiatic Russia, we will embark on the expansive route of European revolution,” Trotsky boasted. “It will lead us to a world revolution. Remember the millions of the German petite bourgeoisie, awaiting the moment for revenge. In them we will find a reserve army and bring up our cavalry with this army to the Rhine to advance further in the form of a revolutionary proletarian war. We will repeat the French revolution, but in the reverse geographical direction: the revolutionary armies will advance 12



“Humint” by Default and the Problem of Trust

13

not from the West to the East, but from the East to the West. The decisive moment has come. You can almost literally hear the steps of history.”2 Even the dour, skeptical Stalin crowed about moving “the centre of revolution from Moscow to Berlin.”3 Yet uprisings in Germany collapsed ignominiously, and the Bolsheviks hesitated to risk all on one throw of the die. Since the Bolsheviks expected a Europe-wide revolution within years, if not months, of defeating counterrevolutionary armies on Russian soil, the Soviet régime was taken by surprise and forced to improvise foreign intelligence at short notice when its plans faltered and fell through after the Polish defeat of the Red Army outside Warsaw in 1920 and the failures first of the Communist Märzaktion (March action) in Germany in 1921 and then of the “German October” in November 1923.4 The climate thus oscillated wildly between revolutionary optimism and deep despondency. Throughout, the counterrevolutionary emigration and its allies within Russia—a fifth column—remained a much feared (and exaggerated) focus of attention. The entire situation was regarded as fluid. There was no sense of permanence. This provides a critical clue as to the nature of the Soviet Union that emerged under Stalin from 1929 and to the story that unfolds: matters domestic necessarily overrode matters foreign. The Great Terror (1937–1939) that cost the Red Army over half its officer corps and much more besides proved the dreadful apotheosis of this perverse order of priorities. After Lenin’s death on 21 January 1924, this deterioration, which had become the focus of anxiety in his last letters, became ever more pronounced: hence the irresistible rise of Stalin, who, although Georgian, personified Russian provincialism. Hard though it may be to believe, as late as 1930 the Politburo still underscored its first priority as “exposing and penetrating centres of pernicious émigrés, independently of their location.” The kidnapping of the counterrevolutionary leader General Evgenii Miller from Paris seven years later, at a time when the Soviet Union was in alliance with France, underlined Stalin’s continued preoccupation, regardless of the cost in trust with the Popular Front régime. With war looming, on 10 May 1939, Pavel Sudoplatov was appointed deputy head of foreign intelligence within the GUGB/NKVD. He was astounded to be briefed by the new Commissar Lavrentii Beria and Stalin himself to the effect that the most important task that lay ahead was the liquidation of arch-rival in exile, Leon Trotsky.5 * * *

14

“Humint” by Default and the Problem of Trust

Having ejected the forces of General Anton Denikin from the Ukraine and thrown back Józef Piłsudski’s offensive from Poland in the summer of 1920, the Bolsheviks switched from counting on spontaneous uprisings abroad to aiding revolutions at the point of the bayonet. Poland was always crucial. It bridged Bolshevik Russia with the long-hoped for revolution in Weimar Germany. “We decided to use our armed forces,” Lenin told a conference of the Russian Communist Party, “in order to help sovietise Poland. Out of that arose the policy for the future as a whole.” This was not done through Party resolutions, but “we said to ourselves that we must make contact by means of bayonets—has the social revolution of the proletariat in Poland not matured?”6 It had not. When Mikhail Tukhachevskii marched on Warsaw in July–August 1920, Lenin stubbornly persisted in the extravagantly misplaced belief that success led by the Red Army was just around the corner. This was no accident. Such illusions were deeply embedded in this new régime. A fallacious confidence in the power of Soviet arms combined with willful misperception of conditions abroad was to continue, one way or another, throughout the life of the Soviet régime and at considerable cost to the efficient operation of its intelligence services. In 1947, Major-General Sir John Sinclair, who later headed MI6, wrote that “it was not generally realised that the controlling element in Russia had virtually no correct appreciation of developments in the outside world, and that they relied for their information on various channels who were successively bent on feeding their superiors with such information as they thought would be most acceptable. This inevitably led to the controlling element receiving a progressively exaggerated form of information on foreign affairs.”7 Faced with Lenin’s obstinate disregard for the reality za kordonom— abroad—Karl Radek, himself a Pole, pointed out that “we must refrain from the practice of using bayonets to sound out the international situation. Bayonets would be good if we needed to aid a particular revolution; but for seeing how the land lies in this or that country we have another weapon—Marxism, and for this we do not have to call upon Red Army soldiers.”8 Of course, Marxism is not an entirely unproblematic prism through which to view the world and, even if it were, one in any case needed to know what precisely the other side had in mind. After all, they were not Marxists. The only early gesture in the direction of reality had been the hurried establishment in April 1920 of a foreign department within the semi-autonomous Special Section (Osobii Otdel) of the Cheka, the Soviet secret police.9



“Humint” by Default and the Problem of Trust

15

Confronted with the sobering fact of “colossal defeat” in Poland, Lenin sounded the retreat but warned that “in spite of complete failure in this instance, our first defeat, we will time and again switch from a defensive policy to the offensive until we have finally beaten all of them.”10 The basic problem was not stated, however: those in Moscow knew little of how Poles really felt. In September, the Politburo thus laid out the case for reorganizing foreign intelligence along more effective lines: “We went to Warsaw blindly and suffered a catastrophe. Bearing in mind the complex international situation in which we find ourselves, the question of our intelligence service must be made the appropriate priority. Only a serious, properly constituted intelligence service will save us from blindly meeting the unexpected.”11 As a result, the Cheka finally acquired its foreign department—INO (Inostrannyi Otdel)—on 20 December 1920. Only with time and as revolution stubbornly failed to appear in the capitalist West did Moscow take a firm hand on foreign intelligence—both signals intelligence (sigint) and human intelligence (humint)—to enhance the conduct of diplomacy and the operations of Comintern. But the relationship between the world of intelligence and the world of the revolution was an awkward one. Given a dictator’s reliance on the secret police, it might be supposed that intelligence professionals would be favored, but Stalin “was his own intelligence boss”12 and “reacted to intelligence material with irritation,” senior officers recalled.13 This is scarcely surprising given the annoying way that material could all too frequently point to a reality that did not correspond with his own perceptions. At first the Bolshevik leadership carelessly trusted in the security of their ciphers. This is not unusual. A veteran of Government Communications Headquarters in the United Kingdom has pointed out that in cryptography the “inherent advantages of the defence are matched by its scope for human frailty and the greater intellectual challenge presented by the offence; states are always confident about the security of their own ciphers and find it hard to exclude laziness in their use.”14 Indeed, in 1921, when Britain was reorganizing the system for making and breaking codes and ciphers, professionals met with skepticism from users: diplomats in particular. One knowledgeable figure from the secret world complained bitterly of the “general apathy that exists in these matters and the disbelief in the powers of cryptographers.”15 Moscow was no different in this from London. It is therefore not particularly surprising to learn that Soviet ciphers took a long while to become safe; and

16

“Humint” by Default and the Problem of Trust

only became so after officials were repeatedly reminded of just how vulnerable they were. Whereas diplomats in Britain were the spoiled beneficiaries of effective decryption against Soviet Russia in particular, the People’s Commissariat of Foreign Affairs (Narkomindel) was forever nagging the leadership about the unsatisfactory state of cryptography and decryption. Neither for the first nor the last time, Commissar Chicherin protested to Lenin on 20 August 1920 that the “decryption of our ciphers . . . is entirely within the bounds of possibility.”16 This was not least because the existing staff were grossly overloaded. “An increase in the number of our cryptographers is now a task of primary importance,” Chicherin wrote to the commissar for finance, Nikolai Krestinsky, on 1 September 1920.17 The consequences of failing to act on this did not take long to make themselves felt. Indeed, throughout the negotiations between July 1920 and March 1921 that led to the de facto recognition of the Bolshevik government by the British, London was reading Soviet ciphers.18 And the Russians more than once became aware of that fact. The trouble was, in the fevered atmosphere of revolution and conspiracy, it was all too easy for Chekists and, indeed, for Lenin himself, to assume that treachery lay behind the problem. Cuts in government spending after the inauguration of the New Economic Policy only made matters worse. Given the persistent priority accorded to the threat of counterrevolution, where human intelligence was at a premium, it was only too easy to neglect cryptography as a sphere of activity marginal to the success of the revolution. Lenin had handed the job of setting up a reliable interdepartmental cryptographic service to the Cheka. In January 1921, the ruling collegium agreed to convene a meeting of all departments concerned: foreign affairs, military, foreign trade, and Cheka. Gleb Bokii, an ethnic Ukrainian, represented the Cheka. He had joined the Russian Social Democratic Labor Party in 1900 and had taken part in both the 1905 revolution and the 1917 coup. In the process, he had been arrested twelve times and exiled to Siberia twice. Bokii had trained as a hydrologist, but he was not appointed for his skills in mathematics. By all accounts a pathological, gloomy disciplinarian and a tough individualist, he had acquired a fearsome reputation hunting down counterrevolutionaries in Turkestan. Bokii was tasked with defining the functions of the new service and appointed to head a new eighth special department, mandated by the “small” Council of People’s Commissars, chaired by Lenin, on 5 May.19 This was named the Special Department (Spetsotdel or SPEKO).20 Bokii also



“Humint” by Default and the Problem of Trust

17

joined the Cheka’s ruling collegium in July at the initiative of its director, Felix Dzerzhinsky.21 The fact that a scourge of counterrevolutionaries was placed in such a position indicates the order of priorities. Bokii had also been a leading member of the Party underground for two decades in Saint Petersburg, however, creating and using various codes and ciphers.22 He resisted the creeping process of Stalinization, and as a result, his relations with the general secretary deteriorated.23 In the early 1930s, Stalin tried and failed to have him removed.24 The Spetsotdel had the unique privilege of autonomy. Bokii communicated directly with the Politburo and Sovnarkom (the Council of People’s Commissars) without intermediary. The department had six, later seven, sections of which cryptographic work in the strict sense occupied only three— the second, third, and fourth. Consisting of seven members and headed by F. Tikhomirov, the second section dealt with theoretical issues and the preparation of codes and ciphers. The third section, initially with only three members, was headed by the deputy chief of the department, Eikhmanis, and managed the process of delivering codes and ciphers to establishments abroad. The fourth and largest section, composed of eight members and run by Bokii’s assistant A. Gusev, had the job of “breaking foreign and anti-soviet ciphers and codes and deciphering documents.” It was the job of the fifth section to obtain foreign codes and ciphers.25 The department was no great success, however. Progress in decryption was disappointingly slow. A few code breakers remained from the tsarist service—V. I. Krivosh-Nemanich, I. A. Zybin, and I. M. Yamchenko, among others26 —but the Bolsheviks sorely lacked cryptographers who were competent in foreign languages—a problem also under the tsar, and one that rival powers did not face, except from injudicious choice. The absence of linguists was a dilemma also faced by the Inotdel (INO) as a whole. Now a part of the Cheka’s successor, the GPU (after the formation of the USSR, the OGPU), the INO became known as the INOGPU. But the OGPU did not yet have the unchallenged preeminence it was to acquire under Stalin. Relations between the Narkomindel at Kuznetsky Most and its “near neighbors” the INOGPU, next door at the Lubyanka, were ever uneasy. The Narkomindel depended on reliable intelligence, but it was inclined to go its own way when opportunity knocked. This inevitably led to disputes over turf, mediated in March 1923 by the indefatigable Party secretary Vyacheslav Molotov, who, as often happened, was asked to chair a Politburo committee to resolve matters.

18

“Humint” by Default and the Problem of Trust

Iosif Unshlikht, deputy head of the OGPU, complained that “Latterly, increasing instances of direct approaches to Narkomindel from an array of people with offers of a secret political character (for example, offers tendered to comrades Chicherin and Petrushevich) are leading to parallel activity alongside GPU structures that are specially devised for intelligence work; inevitably leading both to completely unnecessary expenditure on foreign currency and to negative consequences of a political nature for the organs of the Narkomindel.” The GPU asked the Politburo “to concentrate all varieties of intelligence work (diplomatic, political) with which Narkomindel from time to time comes into contact exclusively in the organs of the GPU.” And “when, in individual instances, representatives of Narkomindel are presented with this or that possibility in the field of intelligence work, the representative of Narkomindel has to agree on his moves in advance with the GPU or its organs on the spot.”27 Relations between the two departments were not improved as a result. Perhaps only those between the OGPU and the Razvedupravleniya—military intelligence—were worse; and they “always fought violently.”28 In a rather transparent bid to gain a foothold within the Narkomindel, Dzerzhinsky proposed that his first deputy, Vyacheslav Menzhinsky, join its ruling directorate—the collegium. This would have given him the right to challenge the views of the commissar and his deputies at Politburo meetings; something Chicherin and his first deputy, Maxim Litvinov, were unlikely to view with equanimity. In the end, Litvinov and Menzhinsky would attend the Politburo together, representing their different institutions. Secure in its authority and with little sure instinct for the realities of the world behind the lines, complacency continued in the Spetsotdel’s creation of secure codes and ciphers. Even when the British revealed what they had read in the traffic with publication of Foreign Secretary Lord Curzon’s ultimatum of 1923, the Bolsheviks still failed to sit up and take note. A diehard Tory government swept to power in Britain on the back of a red scare in October 1924. That owed something to the publication of a notorious forgery—the Zinoviev Letter. This letter purported to emanate from the head of Comintern (the Communist International), Grigorii Zinoviev, with instructions to the British Communist Party concerning subversion of the armed forces, an entirely plausible action given Comintern policy and practice.29 In Moscow on 20 November, Razvedupr (military intelligence: the Fourth Directorate of the general staff) established that the Zinoviev Letter had been



“Humint” by Default and the Problem of Trust

19

produced by a certain Pokrowski, who had sold it to the MI6 head of station in Riga for £500.30 Stanley Baldwin’s cabinet then reacted with fury when from the end of May 1925, the Russians fueled a revolution against the British in China, a key trading partner and major recipient of investment from the city. Throughout, the Government Code and Cypher School (GC&CS) in London continued to read Soviet ciphers until a police raid on 12 May 1927 on the Soviet trade delegation, Arcos, failed to produce evidence of Moscow’s interference in the affairs of other states. Using decrypted Russian communications as casus belli, the Baldwin government then broke off diplomatic relations on 27 May. Although it is generally doubted that Stalin and his supporters actually believed war to be immediately in the offing (it was a convenient stick with which to beat Trotsky), the Politburo nonetheless took the precaution of obliging all Soviet envoys “to destroy immediately all secret material not absolutely vital for current work whether of the embassy itself or the representatives without exception of Soviet and Party organs, including therewith OGPU, Razvedupr and Comintern. The same requirement is laid upon trade representatives in relation to materials in the trade missions.” All this was to be done within four days, as was the liquidation of “that part of the technical security apparatus that is not completely necessary for current work.”31 A fortnight later, the Politburo issued instructions for the removal of all INOGPU, Razvedupr, Comintern, Profintern (trades union international), and MOPR (international workers’ aid) personnel from the premises of Soviet missions.32 As a further measure, on 7 July, the Politburo also agreed to restrict Comintern contact with Soviet embassies to only five, regarded as essential: Berlin, Vienna, Stockholm, Shanghai, and Wuhan. There, one authorized agent “with his own code and cipher” would receive instructions from Osip Pyatnitsky (né Tarshis), the “old man” who ran its underground communications, logistics, and intelligence network (OMS). These instructions would be communicated orally only to a go-between who could not be identified by others as an official of the local Communist Party. Immediately afterward, all correspondence with Moscow would be destroyed. In addition correspondence to and from OGPU and Razvedupr would go via Narkomindel so that the contents would be known to it and its envoys.33 War may not have been imminent, but on 16 June, Commissar for Military and Naval Affairs Kliment Voroshilov nonetheless viewed it as a possibility in 1928. Doubtless on more expert advice, he proposed the creation of sabotage

20

“Humint” by Default and the Problem of Trust

detachments on enemy territory. Britain was to be the main target, as the leader of the offensive against the Soviet Union, followed by Poland and Romania, whose secret services were doing London’s bidding. The proposed target was Ireland. This was agreed at the Politburo on 23 June, subject to “maximum secrecy.” On 1 April 1929, Yakov Serebryanskii, formerly illegal rezident (station chief in CIA parlance) for INOGPU in Belgium and France, took charge of the Special Group—which thus became known as “Yasha’s Group”—that answered directly to Dzerzhinsky’s successor as chairman of OGPU, Menzhinsky.34 Its purpose was deep penetration of military-strategic sites in the event of war. This was the second priority, not put into effect until July 1930. The initial priority was conduct of sabotage and terrorist operations. The first of these was the kidnapping of the leading counterrevolutionary General Alexander Kutepov in Paris on 26 January 1930, for which Serebryanskii was awarded the Order of the Red Banner.35 Serebryanskii’s leadership of the group was confirmed on 13 July 1934 with the absorption of the OGPU into the NKVD. Chicherin, himself formerly an archivist in the old Foreign Ministry, rightly suspected that cryptographers who had formerly worked for the tsarist régime and were now working for the British were thoroughly familiar with “the very system” used in Moscow. So merely changing ciphers was inadequate.36 Indeed, work in London was led by Ernst (“Fetty”) Fetterlein, formerly a leading cryptanalyst in tsarist Russia, who in October 1925 took on board the young J. E. S. (“Josh”) Cooper, to break Russian traffic. Cooper recalled that the Russians used “book cyphers, mostly one part, recyphered with a 1000 group additive key.” From Simla, where more extensive traffic could be intercepted (to and from Moscow and Central and South Asia), the inspired Captain John Tiltman and his ranking superior Colonel Jeffrey succeeded in solving the basic Soviet code book.37 The Russians now mounted a formidable obstacle to encryption of their own secret shifrogrammy—secret telegrams—through the use of “one-time” tables (also known as the “one-time pad”). “Change ciphers every day,” the Politburo ordered, “check up on the cryptographic staff; send a special person with unlimited authority to put in place the strictest level of secrecy in cryptographic work, bearing in mind above all the deviousness [ob’ezd] of such countries as France, Italy, Warsaw, Tokyo, Berlin (which is listed for special reasons).” Moreover, “Unquestionably refrain from the practice of enciphering messages by telegram or radio on especially secret matters.” Instead messages would be enciphered and delivered by courier.38



“Humint” by Default and the Problem of Trust

21

These fundamental changes left cryptanalysts in London high and dry. “Russian high-grade was ‘written off’ by everyone after the Arcos disclosures and introduction of OTP [one-time pads],” Cooper, then a junior at the Russian section of GC&CS recalled. “Only Tiltman went so far as to read a few groups of ‘wrapover’ texts when a pad was used to a depth of two at the end of some messages. This was felt to be interesting but of little practical value. It might however, if persisted in, have led to discovery of re-use of pads; we knew from previous experience of their old diplomatic system that the Russians were capable of re-using additive tables.” Indeed, the experience of deciphering INO cables from the latter half of World War II came to demonstrate conclusively that Soviet confidence in the OTP was justified only where duplicates were never used. And the sheer effort of mass-producing OTPs by hand inevitably resulted in reuse of existing tables. Thus for the time being the Soviet régime had solved the serious problem of cipher security. But the British still gained access to lower-level traffic, notably in the Near East and South Asia. They also broke Comintern codes— thanks to the treachery of a Comintern agent—and read Moscow’s instructions to the Spanish Communists, among others, during the early phases of the civil war from July 1936 through to 1938. The Russians soon became aware of it. Commissar Genrikh Yagoda wrote to Party secretary Nikolai Yezhov that “English intelligence is intercepting and deciphering enciphered radiograms from Comintern’s executive committee sent from Moscow to Spain, Czechoslovakia and England, and also enciphered radiograms from fraternal parties directed to Moscow.” In view of the fact that the codes were “prepared by Komintern’s executive committee” and “agreed with the NKVD special department (comrade Bokii),” Yagoda suspected treachery from someone connected to the Comintern executive. He recommended that “a new, parallel cryptographic department” be set up, using new codes, but maintaining the old codes known to the British in order to purvey “political disinformation.”39 So the initial breakthrough by the British meant that ultimately they fell victim to deception. Partly because Bokii—now the equivalent of a lieutenant general40 — obstinately blocked technical innovation, no serious improvement in Soviet decryption of foreign encrypted communications occurred until the late 1940s. This barrier finally fell with Bokii’s arrest on 16 May 1937. The fact that the Russians had fallen behind their adversaries in this respect had become apparent by 1935; and the gap was widening. Ciphers were broken largely

22

“Humint” by Default and the Problem of Trust

through the theft of their keys. In the 1920s, OGPU station chiefs abroad were tasked with “drawing into secret collaboration people who had the means of carrying out instructions for obtaining information on the organization of cryptographic work by Western special services.”41 Within the department, decryption was done entirely by hand and the department had only the haziest idea about the use of machines for such purposes. This left Moscow well behind Berlin, possessed of Enigma machines, for example. Yet the removal of Bokii resulted in no immediate improvements, leaving the Russians substantially at a disadvantage when war came.42 Lack of personnel had also been a problem. Many more mathematicians and linguists were needed than the Russians possessed. By the summer of 1934, with the transformation of OGPU into the GUGB of the NKVD, what was now the ninth department amounted to one hundred staff.43 Razvedupr was even more pressed. That year, according to the Latvian Old Bolshevik Jan Berzin, 58 percent of foreign ciphered documents that could have yielded valuable information remained untouched because insufficient staff were available.44 By September 1939—renamed the seventh department—the Spetsotdel’s staff had more than doubled, to 230.45 But growth in numbers had yet to have a decisive impact on performance. Military ciphers were a particular problem, much complained of by the Red Army. Finally, they received what they had long asked for: on 16 July 1939, the eleventh department—the military decryption service—was established under the Fifth Directorate of the armed forces, tasked with breaking not only military codes but also the military intelligence communications of foreign powers. The NKVD, now under Lavrenty Beria, was ignominiously forced to hand over all the relevant materials at its disposal.46 Christopher Andrew and Oleg Gordievsky make a confident assertion: “The myth has developed that codebreaking coups are achieved simply by brilliant mathematicians. . . . In reality, most major breaks of high-grade code and cipher systems on which evidence is available were achieved with the help of at least partial information on those systems provided by espionage. Soviet codebreakers in the 1930s had vastly greater assistance from espionage than their Western counterparts.”47 In stark contrast, a Polish cryptanalyst explained how they broke into Enigma: “We did it partly by mathematics and partly by Verrat [treason]. But we could have done it all by mathematics.” It is not clear whether this referred to Enigma as a whole or just the input order.



“Humint” by Default and the Problem of Trust

23

Even if one dismisses the Polish boast as little more than that and accepts that “cribs” are invaluable to code-breaking, in the particular instance of Soviet intelligence, it is entirely misleading. From that premise, the untutored or unwary reader might be tempted to conclude that the Bolsheviks had the edge in decryption as a result of an extensive humint network, whereas entirely the reverse was true. Necessity dictated the need to rely upon human agency to compensate for technical deficiency. Dependence upon humint was therefore, not an option, but dire necessity. An epitaph on this sorry state of affairs was given by Commander Alastair Denniston, who headed GC&CS. Assertions by the former counselor at the Soviet embassy in Rome, Leon Helfand, that Moscow was reading “all our telegrams” carried “no weight” as far as Denniston was concerned. Helfand was referring to memoranda of conversations coming out of British embassies, which were not generally enciphered unless they contained comment from the ambassador. Indeed, “in peace time,” it was to be expected that “‘about 75% of the traffic” moved in this form “and was expected to be read by any Government who wished to read telegrams.” With regard to the remaining 25 percent of the traffic, which was encrypted, tables of codes and ciphers “were changed every three months.”48 Helfand had defected in July 1940. Although London had earlier suspected him to be an intelligence officer when he served in Paris, the debriefings appeared to indicate otherwise. In fact, the original suspicions appear true.49 If so, Helfand would not have been the only defector to mislead the authorities successfully in order to escape assassination by Stalin. Prior to hostilities, the British were not exactly path breakers in engineering effective decryption by electromechanical means. They had fallen back following the advances gained in Room 40 during World War I. GC&CS was technically weak, and it was hard to get the authorities interested in such matters. Radio interception in Britain was, for example, badly neglected. Britain was nonetheless much ahead of the Russians. Denniston was thus “doubtful if there were enough traffic available to the U.S.S.R. to break the cypher even if they possessed the modern mechanical methods we now use to break subtractor tables.”50 The evidence given by the OGPU defector Georgii Agabekov at the end of the 1920s alerted the British to the fact that the Russians were almost entirely reliant upon human intelligence purloining cipher keys and cribs in the form of original documents that could be matched to intercepted telegrams in order to compensate for their weakness in decryption.51

24

“Humint” by Default and the Problem of Trust

After the outbreak of war and on information supplied by the defecting NKVD officer Walter Krivitsky, in October 1939, Captain John King, a Foreign Office cipher clerk, was arrested, tried, and sentenced for selling keys to the Russians. Valentine Vivian, deputy chief of MI6, observed that this showed quite conclusively that the Soviets spent £20,000 in facilitating Pieck approach to F.O. circles and nearly as much again in suborning King to hand over F.O. telegrams in solved form for photo-reproduction and submission to Moscow. The Soviets were, we know, at the same time paying large sums to a traitor at Rome for the same purpose and, though a double-banking of such methods is understandable, since King or the Rome traitor might at any moment be transferred or found out, why should it have been necessary to suborn either King or the Rome traitor, if they were, and had been for years, getting all our telegrams by the much less chancy and complete methods of cryptography?52

Soviet decryption was scarcely more successful at the end of the 1930s than it had been at the beginning of the decade. But here, too, a breakthrough came courtesy of humint, enabling the Russians to read Japanese diplomatic traffic. At the beginning of the decade, the Spetsotdel had obtained the code book of the Japanese embassy thanks to the efforts of Krivitsky at the INO and KRO (OGPU counterintelligence). In 1938, after the code book was updated, the deputy NKVD station chief in Sofia, Bulgaria, Vasilii Pudin, who had some direct experience working against Tokyo in Outer Mongolia, bribed a senior Japanese diplomat to hand over the new diplomatic code.53 This proved of inestimable advantage since, unlike those of any other state, Japanese missions overseas communicated with one another as well as with the home government. Moscow was thus fully informed as to Japanese policy and arguments about its orientation through to the German invasion in June 1941 and thereafter. The development of human intelligence compensated in part for the failures of sigint, but not entirely. Its general limitations have been spelled out by Michael Herman: The identification and recruitment of potential agents takes a long time. Communications with their controllers are the most vulnerable points and have to be limited; their reporting is therefore slow, and usually excludes “real time” intelligence. Agents’ observations and recollections are subject to the usual human frailties; they are liable to go off the rails and their reliability cannot be counted on. The controller can never be completely sure that they are not



“Humint” by Default and the Problem of Trust

25

fabricating or distorting reports or, worse, acting as double agents, providing deceptive material and penetrating the intelligence service they claim to be working for. Humint has a reputation with its users for unreliable information, and its agencies are always torn between source protection and revealing enough to establish its credibility.54

But how much of this applied to the Soviet Union? On the one hand Moscow undoubtedly benefited from an extraordinary windfall stemming from the spirit of internationalism aroused by the October Revolution and assiduously propagated thereafter by the Comintern, ironically at the very time that the Soviet Union under Stalin’s merciless grip was turning in on itself. Artur Artuzov (né Fraucci) was thus appointed deputy head of the INO on 1 January 1930, moving up to head the department on 1 August 1931. It was no accident that Artuzov had made his name in counterintelligence, assisted by the former tsarist police chief Lieutenant General Vladimir Zhunkovskii in the formation of a fake counterrevolutionary organization, the Trust, which successfully lured many to their deaths, including the terrorist Boris Savinkov and the MI6 agent Sidney Reilly.55 Stalin now added to his opponents in the West the notable figure of Trotsky in exile, who threatened to undermine his rule from within. Once again domestic priorities took precedent over foreign ones. Pavel Sudoplatov, deputy head of the INO in 1939 insists that “Foreign secret intelligence until 1939 considered counterrevolutionary tasks abroad as the main direction of its activities.”56 Alexander Orlov (Leiba Fel’dbin), who managed to defect west without betraying the names of Soviet intelligence agents, wrote that “In their operative work the Soviet intelligence officers prefer to deal with men and women who are devoted to the ideals of communism or are in sympathy with the aims and policies of the Soviet Union, because such people give all they have and may be relied on.”57 Stalin also held this view. “In intelligence,” he wrote, “one must have several hundred people who are friends (more than we have as agents) prepared to undertake any of our tasks . . . friends give the highest class of intelligence.”58 This certainly proved the case with the key spies known as the Cambridge Five (see below). In a report to the “Centre”— NKVD INO headquarters—Arnold Deutsch (nom de guerre Stefan Lang) commented on the agents he recruited in England: “They all came to us on completing university at Oxford and Cambridge. They shared Communist convictions. It occurred under the influence of the extended revolutionary

26

“Humint” by Default and the Problem of Trust

movement which in recent years has seized certain circles among the English intelligentsia and the two fortresses of English intellectual life, Cambridge and Oxford, in particular.”59 Deutsch, a Slovakian Jew, was a chemical engineer and a man of initiative and intellect, though regarded as a “bumptious” figure by others in the service. He was assigned to Britain in February 1934, having briefly served in Paris, and registered for a degree in psychology at London University. With the encouragement of the illegal Russian station chief, Alexander Orlov, Deutsch used the recruitment of Kim Philby as the starting point for gathering in his immediate circle of friends. Deutsch summed it all up later: “The chances of candidates, in the final analysis, depends upon their provenance, completion of education at one of the aristocratic secondary schools and receipt of higher education at Cambridge or Oxford, references from those of high standing, reliable politics. No less important are parental connections and those connections that the students themselves establish among people in their own circle.” 60 Only the most capable were chosen. In July 1934, convinced of the arguments, the Centre set its London station chiefs the long-term task of penetrating British intelligence. For this purpose, a monthly budget of £500 was established.61 Philby was recruited in June 1934, followed by Donald Maclean (September 1934), Guy Burgess (January 1935), Antony Blunt (January 1937), and John Cairncross (May 1937). There were, in addition, nine others, including, so it is claimed, a similar group from Oxford; thus making a total, not of five, but of fourteen.62 They supplied invaluable material, including masses of top secret documents. This practice was not, however, an unmitigated boon. Communists were the object of close surveillance throughout the capitalist world, and those who had been Communists in their youth were bound to be uncovered sooner or later. Philby, however, had avoided joining the Communist Party as a student because he envisaged a career in the Foreign Office. If agents were by nature hard to control, the added danger was that the ideologically committed might place a higher value on zeal than on tradecraft even had they been trained, which, in most cases, they had not been. In urging his return to London after he was recalled to Moscow in September 1937, at the height of the Great Terror, Deutsch insisted that maintaining direct contact with the Cambridge Five was essential: All of our people are young and have so little special experience in our work. . . . Not receiving any news from us at all, they will necessarily



“Humint” by Default and the Problem of Trust

27

experience disillusionment: all of them work from conviction and with enthusiasm; and it is all too easy for it to occur to them that they have been rejected. . . . I want once again to point out the special makeup of our apparatus. All of them believe in us. They are sure that we are always there, forever and everywhere; that we fear nothing; that we will never leave anyone to the arbitration of fate; that we are above all, accurate, precise, and reliable. And the success of our work has in part been based up to now on the fact that we have never disappointed them.63

All Communists had long since been asked to stand ready to spy if in a position to do so. A circular in 1924 to all Communist Parties from secretary of the Comintern executive (IKKI), the Bulgarian Vasil Kolarov, stated: “Each member of the Communist Party is obliged to supply reports on a daily basis on the activity of those organs of state power in which he personally works or the activity of those of which he is aware through representative organs, whichever the Party requires of him. The leading Party organs will pass on reports immediately to the IKKI through the OMS.”64 This cut out Soviet state intelligence organizations, leading to disaster. The most extensive attempt at using Party members—as so-called worker correspondents—was in France during the late 1920s, which resulted in the arrest and imprisonment of Communist Party leaders. As a result, Moscow turned full circle and ruled out use of Party members for espionage work; 65 though this was at times observed in the breach, notably by military intelligence. The human factor proved critical in the case of the German Communist turned spy Richard Sorge, who had been recruited for Razvedupr, but was never trained as an intelligence officer. “He began working without any schooling in counterintelligence.”66 It was thus not Japanese military counterintelligence that uncovered his role, but the political police, who unraveled connections through known Communists with whom he was working, who had been knowingly recruited by Sorge’s superiors. It was then only a matter of time before Sorge’s arrest, interrogation, and execution.67 Razvedupr was located at 19 Znamenskaya, near Red Square. Relations between civilian and military intelligence organizations were never easy. Krivitsky—who had acquired jurisdiction in Western Europe over both intelligence organizations before his defection in 1937—recalled that the feud between OGPU and Razvedupr represented a “continual struggle by the Ogpu to control both organizations.” After the disastrous setback in Poland during the late summer of 1920, Soviet military intelligence was almost annexed

28

“Humint” by Default and the Problem of Trust

by the Cheka and saved only by Trotsky, the creator of the Red Army. It was run with amateurish exuberance from 1 April 1924 on by Berzin. It held one advantage over its civilian counterpart and rival, however. The head of the INO could see Stalin only in the company of his superiors, whereas the head of Razvedupr had the right of direct access. This effectively cut out the chief of the general staff and the commissar for military and naval affairs. Despite this privilege and an advantage in being very much larger than its civilian counterpart, the practices of the Fourth Directorate fell far short of efficiency. Famed as the leader of the Soviet military intelligence network known by the Germans as the Red Orchestra (Rote Kapelle), Leopold Trepper recalled that Berzin “in no way corresponded to the stereotyped image of the professional intelligence man.” At the height of the crisis in Anglo-Soviet relations in the mid-1920s, Razvedupr was found seriously wanting.68 Disaster was sooner or later unavoidable. The struggle eventually culminated in a takeover by the OGPU in May 1934, following a series of disasters for military intelligence. These calamities were the direct consequence of a network that, though seemingly efficient in closely coordinating Soviet intelligence throughout Western and Central Europe, actually meant any number of key figures within it could jeopardize the activities of others across the entire network if arrested and successfully interrogated. In addition, though told not to use local Communists—by explicit order of the Politburo on 8 December 1926—that was precisely what was done.69 Thus the recruitment and use of Sorge was but one careless instance among many, and with fatal consequences for him that were entirely predictable. In spring 1930, just in case standing orders were ignored, Berzin instructed the new Soviet station chief in Austria, the Bulgarian Ivan Vinarov (“Mart”), “to avoid any contact with members of Communist organizations and their leadership.” But as soon as he arrived in Vienna, Vinarov boldly proceeded to construct his network precisely from such people. This did not matter not just for Austria alone, because his remit also encompassed Poland, Czechoslovakia, Romania, Bulgaria, Yugoslavia, Turkey, and Greece; in other words, the entire French alliance system, with several additions. Operating through local post and telegraph offices, Vinarov’s network purloined enciphered communications and sent them on to Moscow with great success.70 A reprimand would thus have seemed ungracious in the circumstances. But the habit became somewhat ingrained. And it was out of the



“Humint” by Default and the Problem of Trust

29

Austrian network that catastrophe eventually transpired. In 1932, the police arrested two illegals; a third, though exposed, managed to escape. The Centre unwisely then sent him to work in Latvia, which had close relations with the Austrian police. Sure enough in the summer of 1933, he was arrested and almost the entire Soviet network broken open. In the course of this, a courier, Julius Trossin, came to the attention of the police. This was critical, because Trossin’s path crossed Soviet espionage operations in Germany, Romania, Finland, Estonia, and the United States. He was duly arrested in Hamburg on 6 July 1933. As a result the entire network was exposed and links to the Soviet station in Britain broken. Worse followed. In the summer of 1933 Razvedupr’s recruiting agent in Germany—who had links with illegal Soviet stations in a number of countries—was arrested. That October, the station chief, his assistants, and some others were arrested in Finland.71 This was just the kind of excuse OGPU had been waiting for. The Politburo appointed Artuzov, the head of the INO, deputy head of the Fourth Directorate. To ensure that previous mishaps were not repeated, the Politburo also insisted that groups of agents in military intelligence were not to know one another, and that each group communicate with the Centre independently. But this takeover had other consequences, no less disastrous for Moscow. It meant that supervising NKVD personnel in the field held information on both organizations within their grasp. The defection of Walter Krivitsky from his station in Holland showed how damaging this could prove, and had Alexander Orlov revealed all he knew to the British authorities, the entire network would have gone down. Indeed, this is what Moscow suspected. In February 1940, Beria, now commissar for internal affairs, ordered the closure of the London station on the grounds that everyone, including the Cambridge Five, were known to former intelligence officers and were therefore exposed as spies. Of course, those defections were a self-inflicted wound, a direct consequence of Stalin’s personality and his unreasoning fears. Stalin’s wave of terror against Bolshevik cadres throughout the Party and state was justified on grounds of the assassination of the Leningrad Party secretary Sergei Kirov in December 1934. It all but destroyed what had been painstakingly built up at the INO and Razvedupr since their inception. It hit code and cipher breaking as it did the rest of the services. In subsequent years, the Spetsotdel had to make do with recruits who had only secondary education or incomplete university education and lacked any special training.72

30

“Humint” by Default and the Problem of Trust

In the case of the key country for penetration, Nazi Germany, the head of the Berlin station, Boris Gordon, was recalled in May 1937. He was shot on 21 August, and this was also the fate of the head of the German section at the INO, along with many under him. Then almost all remaining members of Berlin station were liquidated in turn. Gordon’s replacement, Alexander Agayants, died on the operating table in December 1938. As a result, through the critical year 1939, only two operatives remained on station, with no station chief at all. A new rezident, Anayak Kobulov, arrived only that September, and he was incompetent. “From the summer of 1938 until the war broke out,” Trepper recalls, “we laid aside all intelligence work, properly speaking.” On 13 May 1939, after the transfer of the Berlin station’s chief, Vladimir Dekanozov, to become deputy commissar at the Narkomindel, the deputy head, Pavel Fitin, was appointed in his place. Fitin, the son of peasants, was a mere thirty-one-year-old, having had his career in the NKVD, which began only in August 1938, artificially accelerated by the relentless progress of the terror.73 Reporting on this disastrous state of affairs nearly two years later, at the beginning of 1941, Fitin wrote: “By the beginning of 1939, as a result of the exposure of the enemy’s leadership of the Foreign Department at that time, almost all station chiefs on the other side of the line were recalled and removed from work. The majority of them were then arrested, and the remainder were subjected to examination.”74 As a result: “There could be no question whatever of any kind of intelligence work on the other side of the line. The task ahead consisted of creating an apparatus of stations on the other side of the line, together with the establishment of the department’s own apparatus.” This was never entirely true, however. Information continued to flow to Moscow. There were now a total of fourteen INO agents in Britain. But Anatoly Gorsky, the only remaining officer, and a junior one at that, was left to manage them entirely on his own. And what did come in was necessarily treated with great suspicion. First of all, could Stalin assume that Orlov had not in fact revealed the names of such agents, which he was in a good position to do? Second, until the record of his debriefing was seen by Blunt, upon his joining MI5, it was uncertain whether Krivitsky had not known some of these names. Stalin was thus left with a bewildering dilemma. In February 1940, Beria closed the London office on the grounds that these agents, including the Cambridge Five, were known to the enemy. In Moscow, however, Gorsky was able to convince his superiors that this was not so and was returned to



“Humint” by Default and the Problem of Trust

31

London that November to reestablish contact. Suspicions were all too easily rearoused, and later in the war, when their British agents were unable to uncover those subverting the Soviet Union, Gorsky’s superiors reverted to their prior suspicions.75 Herman has pointed out that intercepts are often only a random sample. What matters is interpretation. Here those working under Stalin suffered another severe disadvantage for the period under discussion. “A major defect in our intelligence work was the weak arrangement for the analysis of information coming in from agents,” Sudoplatov recalls.76 A “special” office had been set up in 1926 for this purpose. And by the end of the 1930s, a “special” bureau under Kossa, later a noted economist, conducted analyses of incoming intelligence at the NKVD. But it remained an Achilles’ heel for the régime. There was no overarching analysis of all incoming data until the summer of 1943. According to historians of Soviet intelligence with archival access, agents’ reports were directed to Stalin himself. This was a structural impediment that could be remedied, as it eventually was. But the man at the top was critical to the efficient exploitation of intelligence for the conduct of foreign policy, and Stalin found trust to be an elusive quality. Even documents obtained by agents remained of dubious provenance. Yet throughout the interwar period, Stalin failed to upgrade the cryptographic apparatus to the point where he could liberate himself from networks of agents he found himself incapable of trusting. Part of the problem was that cryptography in the Soviet Union was confined to a narrow circle cut off from the outside world, where advances were rapidly being made by potential enemies. Part of the problem also was the lack of mathematical cryptanalysts, and the view taken that this was not a sphere for fundamental research on a plane with the sciences. Thus as the sheer quantity of cryptographic traffic worldwide grew toward the end of the 1930s, the difficulties deciphering it were also compounded by increasing complexity of the systems employed by the adversary. The Russians were still operating in the nineteenth century in this respect. Moreover, to those accustomed to assuming that the running of intelligence services in support of foreign policy is strictly subordinated to reasons of state, or, indeed, even to some higher ideological calling, such as Marxism-Leninism, what actually occurred in the Soviet Union between the wars offers a timely reminder that in a state restructured in tight subordination to a dictator at the mercy of morbid suspicion, no such rational assumptions are justified. Domestic considerations were uppermost in

32

“Humint” by Default and the Problem of Trust

Stalin’s mind. Moreover, having reindustrialized and rearmed Russia, he believed he had sufficient power to keep the outside world at bay; indeed, more than sufficient if he played his cards right. The German invasion on 22 June 1941 proved him entirely wrong. But that is a story that takes us beyond this chapter. Notes 1. Epigraph to this chapter: Litvinov to Maisky, 11 April 1939, in Dokumenty vneshnei politiki 1939, bk. 1 (Moscow, 1992), doc. 210. 2. Recorded at the time by Party secretary Vyacheslav Molotov. Quoted from the archives in V. Nikonov, Molotov: Molodost’ (Moscow, 2005), 704–705. 3. Stalin’s reply to Thalheim, 13 September 1923, quoted in P. Makarenko, “‘Nemetskii Oktyabr’: 1923 g. i sovetskaya vneshnyaya politika,” Voprosy Istorii, no. 3 (March 2012): 43. 4. Ibid. 5. P. Sudoplatov, Spetsoperatsii. Lubyanka i Kreml’ 1930–1950 gody (Moscow, 1998), 102–104. 6. “Iz doklada V. I. Lenina, ‘Politicheskii otchet TsK RKP (b)’ na konferentsii RKP (b), 22 September 1920,” in Russian State Archive of Socio-Political History, Politburo, TsK RKP (b)—VKP (b) i Komintern 1919–1943: Dokumenty (Moscow, 2004), 61. 7. Quoted in P. Cradock, Know Your Enemy: How the Joint Intelligence Committee Saw the World (London, 2002), 286. 8. Quoted in Haslam, “Comintern and Soviet Foreign Policy, 1919–1941,” in The Cambridge History of Russia (Cambridge, 2006), 3: 639. 9. I. Lander, Neglasnye Voiny: Istoriya spetsial’nykh sluzhb 1919–1945 (Odessa, 2007), 1: 180. 10. Politburo, 65. 11. Ocherki istorii Rossiskoi vneshnei razvedki (Moscow, 1997), 2: 10. 12. A. Orlov, “The Theory and Practice of Soviet Intelligence”: www.cia.gov/ library/center-for-the-study-of-intelligence/kent-csi/vol7no2/html1/v0. Orlov, a promising officer in Soviet intelligence, defected in the late 1930s when recalled to Moscow, for fear of being shot. 13. Sudoplatov, Spetsoperatsii, 196. 14. M. Herman, Intelligence Power in Peace and War (Cambridge, 1996), 69. 15. HW3/39, National Archives, Kew, UK. 16. Quoted in T. Soboleva, Tainost’ v istorii Rossii (Istoriya kriptograficheskoi sluzhby Rossii XVIII–nachala XXv.) (Moscow, 1994), 314. 17. ‘K istorii sozdaniya shifroval’noi sluzhby v MID Rossii’, Diplomaticheskii Vestnik, April 2001.



“Humint” by Default and the Problem of Trust

33

18. R. Ullman, Anglo-Soviet Relations, 1917–1921: The Anglo-Soviet Accord (Princeton, NJ, 1972). The decrypts are available now in HW12/10–HW12/21 at the National Archives, Kew, UK. 19. Soboleva, Tainost’, p. 320. 20. S. Lekarev and V. Pork, “Radioelektronnyi shchit i mech,” Nezavisimoe voennoe obozrenie, 26 January 2002. 21. Rossiya. XX Vek. Dokumenty Lubyanka. Organy VChK-OGPU-NKVD-NKGBMGB-MVD-KGB 1917–1991. Spravochnik, ed. A. Kokurin and N. Petrov (Moscow, 2003), 22. Also Lekarev and Pork, “Radioelektronnyi.” 22. T. Soboleva, Istoriya shifroval’nogo dela v Rossii (Moscow, 2002), 412. 23. Soboleva, Tainost’, 325. 24. Ibid., 354. 25. Ibid., 322. 26. Soboleva, Istoriya, 416. 27. Memo from Unshlikht to Stalin, 28 March 1923, in Rossiya. XX Vek. Dokumenty. Lubyanka. Stalin i VChk . . . Yanvar’ 1922–dekabr’ 1936, ed. V. Khaustov et al. (Moscow, 2003), 77–78. 28. L. Helfand, debriefed by Strang at the Foreign Office, July 1952, KV/2/2681, National Archives, Kew, UK. 29. For the latest research on the British side, see C. Andrew, The Defence of the Realm: The Authorized History of MI5 (London, 2010), 148–152. Here Andrew had changed his mind completely from a view he originally expressed in the Historical Journal, 1977 (20): 673–706, where he argued that the permanent undersecretary at the Foreign Office would never have lied about the letter’s provenance; a position sustained despite E. H. Carr’s insistence that it was wrong (Carr worked at the Foreign Office in those years). 30. Politburo TsK RKP(b)-VKP(b) i Evropa: resheniya ‘Osoboi papki’ 1923–1939, ed. G. Adibekov et al. (Moscow, 2001), 53, fn. 1. 31. Politburo protocol no. 102, 13 May 1927: Politburo TsK, doc. 79, 149. 32. Politburo protocol no. 84, 28 May 1927: ibid., doc. 273, 462. 33. Protocol no. 115, 7 July 1927: Politburo TsK RKP(b)-VKP(b) i Komintern. 1919– 1943. Dokumenty (Moscow, 2004), doc. 277. 34. For a summary biography of Serebryanskii in Russian: www.svr.gov.ru/history/serebrjanskij.htm. 35. A. Kolpakidi and D. Prokhorov, KGB: Spetsoperatsii sovetskoi razvedki (Moscow, 2000), 543–544. 36. Soboleva, Tainost’, 314. 37. C. McKay, “British SIGINT and the Bear, 1919–1941. Some Discoveries in the GC&CS Archive,” Royal Swedish Academy of War Sciences, KKrVAHT, no. 2, 1997. 38. Politburo protocol no. 84, 28 May 1927: Politburo TsK . . . i Evropa, 462. 39. Ibid., 731. 40. Rossiya. XX Vek, 64.

34

“Humint” by Default and the Problem of Trust

41. Lekarev and Pork, ‘Radioelektronnyi.” 42. Lander, Neglasnye, 423. 43. Rossiya. XX Vek, 60. 44. Soboleva, Istoriya, 423–424. 45. Ibid., 71. 46. Soboleva, Tainopis’, 357. 47. C. Andrew and O. Gordievsky, KGB: The Inside Story (London, 1990), 179. 48. Signed 23 June 1941: KV/2/2681, National Archives, Kew, UK. 49. I. Damaskin, Stalin i Razvedka (Moscow, 2004), 209. 50. Denniston memoir: HW3/32, National Archives, Kew, UK. 51. Georgii Agabekov’s memoir Zapiski chekista appeared in Berlin 1930. 52. KV/2/2681, National Archives, Kew, UK. 53. V. S. Antonov and V. Karpov, Tainye Informatory Kremlya-2. S nikh nachinalas’ razvedka (Moscow, 2003), 201. 54. Herman, Intelligence Power, 65. 55. For a summary account: “Stanovlenie vneshnei razvedki 1921–1925,” www.svr. gov.ru/history/stages.htm. 56. Sudoplatov, Spetsoperatsii. Lubyanka i Kreml’ 1930–1950 gody (Moscow, 1998) 57. A. Orlov, Handbook of Intelligence and Guerrilla Warfare (Ann Arbor, MI, 1963), 95. For more on the author, see A. Kolpakidi and D. Prokhorov, Vneshnyaya Razvedka Rossii (Moscow, 2001), 311. 58. Quoted from the archives by L. Shebarshin, Ruka Moskvy. Zapiski nachal’nika sovetskoi razvedki (Moscow, 1996), 189–190. 59. Quoted in Elita russkoi razvedki: Dela etikh lyudei sotavili by chest’ lyuboi razvedke mira (Moscow, 2005), 181. 60. Ibid., 182–183. 61. Lander, Neglasnye, 297. 62. Recruitment dates in Ocherki istorii, vols. 3 and 4. The reference to fourteen British turncoats is in ibid., vol. 14, 262. 63. Elita russkoi razvedki, 187. 64. Politburo . . . i Komintern, 275, fn. 1. 65. Orlov, Handbook, 96. 66. Testimony of an interrogator cited in A. Vitkovskii, “Zolotoe Ko’ltso ‘Ramzaya.’” 67. “B. I. Gudz’ ryadom s Zorge,” www.fsb.ru/history/autors/gudz3.html. 68. V. Kochik, “Sovetskaya voennaya razvedka: struktura i kadry (1924–1936),” Svobodnaya Mysl’, no. 8 (0000): 72–73. 69. Ibid., 74–75. 70. Ibid., 79–80. 71. A. Kolpakidi and D. Prokhorov, Imperiya GRU: Ocherki istorii rossiskoi voennoi razvedki (Moscow, 2000), 195–204. 72. I. Matskevich and E. Belov, 50 Let Institut kriptografii, svyazi i informatiki (Moscow, 1999).



“Humint” by Default and the Problem of Trust

35

73. A potted biography appears in A. Kolpakidi and D. Prokhorov, Vneshnyaya Razvedka Rossii (Moscow, 2001), 112–113. 74. Quoted in I. Damaskin, Stalin i razvedka (Moscow, 2004), 212. 75. For more on Gorsky, see Kolpakidi and Prokhorov, Vneshnyaya Razvedka, 202–203. 76. Sudoplatov, Spetsoperatsii, 197.

2

Barbarossa and the Bomb Two Cases of Soviet Intelligence in World War II David Holloway

H

ow important was intelligence in the conduct of international relations in the twentieth century? This chapter looks at two cases involving the Soviet Union. The first concerns the period leading up to the German invasion on June 22, 1941. Moscow received copious reports about German plans, but Stalin failed to bring the Red Army up to full combat readiness to meet the German attack. The second is atomic intelligence. Between 1940 and 1945, the Soviet Union obtained a great deal of information about nuclear research in Britain and the United States and was well informed about the progress of the Manhattan Project. Intelligence helped the Soviet Union build its own atomic bomb more quickly than it would otherwise have been able to do. Stalin’s response to the intelligence he received in 1941 is widely regarded as a terrible blunder. Soviet atomic espionage, on the other hand, was one of the most successful intelligence operations of the twentieth century. These are both cases in which the acquisition and interpretation of intelligence had significant consequences for international politics: the German invasion was the traumatic start to a long and brutal war, while the Soviet atomic project was a key element in the postwar rivalry between the Soviet Union and the United States. Stalin’s handling of intelligence in the months before the German attack is still a matter of intense controversy in Russia. What did he know about German plans and intentions, and what use did he make of the intelligence he received? What difference would it have made if he 36



Barbarossa and the Bomb

37

had concluded that Germany was about to launch an attack? Atomic intelligence too has been a subject of controversy. Did spies do more than physicists to build the Soviet bomb? What difference did it make to international politics that the Soviet Union developed the atomic bomb earlier than it would have done by its own efforts alone? The two cases are very different in substance, but they are alike in having been the subject of extensive literatures over the years. It has been known since 1941 that Stalin received warnings that Hitler was planning to attack the Soviet Union.1 Soviet atomic espionage has been no secret since 1946, when the defection of Igor Gouzenko, a cipher clerk at the Soviet embassy in Ottawa, was reported in the press.2 Since the collapse of the Soviet Union, a great deal of light has been shed on these cases by the publication in Russia of collections of newly declassified documents.3 The aim of this chapter is not to explore these two complex cases in detail, but rather to ask what it is we need to take into account when we seek to integrate the history of intelligence into the history of international relations. The chapter will focus on three questions. What intelligence did the Soviet Union acquire? How did it use that intelligence in formulating policy? What consequences did the intelligence, or the decisions informed by that intelligence, have for international relations? The focus is on the evaluation and use of intelligence rather than on its collection. The same organizations dealt with intelligence in both instances. Military intelligence was the responsibility of the General Staff of the Red Army; foreign intelligence came under the NKVD/NKGB.4 Both organizations had residencies abroad through which they maintained contact with secret agents, and both had been very badly affected by the Great Purge. Between 1937 and 1941, for example, five consecutive chiefs of Soviet military intelligence were arrested and shot. Soviet agent networks in Germany were almost completely destroyed by the combination of Nazi repression and Soviet purges.5 In spite of this, military intelligence and foreign intelligence produced a great deal of material on the issues discussed in this paper. Other organizations were involved too—Naval Intelligence, the People’s Commissariat of Foreign Affairs, the Comintern, and the Border Guards—but military intelligence and foreign intelligence were by far the most productive.6 Information from signals intelligence appears not to have played a significant role in either of these two cases.7 The most important link between the two cases was, of course, Stalin, who figures everywhere in this history, from the effect of his purges on the

38

Barbarossa and the Bomb

intelligence services and the armed forces in the late 1930s to his response to the bombing of Hiroshima and Nagasaki. By the late 1930s, he had disposed of his political opponents, both real and imagined, and created a political system that was “capable only of approving his decisions,” in the words of one of his biographers.8 It took great courage to disagree with “the Boss.”9 Stalin’s personality assumes therefore a key role in this history. Conspiracy and suspicion were second nature to him, and these qualities fused with the distrust and skepticism that are inherent in intelligence work to create a context for intelligence assessment and policy-making that was exceptionally complex and dangerous. This raises a fourth question for this chapter: what light do these cases throw on Stalin as a political leader? Soviet Intelligence and the German Invasion of June 1941 Hitler made no secret of his hatred of Bolshevism or of his ambition to restore and expand German power. In the mid-1930s, Stalin tried to contain Nazi Germany through collective security, but he reversed that policy on August 23, 1939, with the signing of the Nazi-Soviet Pact. When Germany invaded Poland on September 1, and Britain and France declared war in response, it appeared that the Soviet Union might enjoy a respite during which the capitalist powers would weaken one another through war. That at any rate was the hope that Stalin expressed to other Soviet leaders on September 7, 1939: “we see nothing wrong in their having a good hard fight and weakening each other.”10 In that same month the Soviet Union took over the eastern part of Poland and in the following year it annexed the Baltic States, Bessarabia, and northern Bukovina. Stalin used the Nazi-Soviet Pact to expand the territory of the Soviet Union, thereby creating a long border across which the Soviet Union and the Third Reich confronted each other directly. Germany’s rapid defeat of France in May and June 1940 produced a drastically new strategic situation in Europe, opening up the prospect that the war in the West might be brought to a quick end either by a German invasion of Britain or by a peace agreement between the two countries. Germany was emerging stronger, rather than weaker, from its war with the other capitalist powers. Soviet anxiety about this shift in the balance of power was deepened by the contrast between the brilliance of the Wehrmacht campaign in the West and the wretched performance of the Red Army in the Winter War



Barbarossa and the Bomb

39

with Finland. In July 1940, Hitler told his top commanders that he intended to attack the Soviet Union in the spring of 1941.11 On December 18, 1940, he signed Directive No. 21, ordering the German armed forces to be ready “to crush Soviet Russia in a speedy campaign even before the war with England is concluded.”12 This was to be a war of “conquest and annihilation.” Hitler’s purpose was not merely to exploit the resources possessed by the Soviet Union but also to dominate the living-space (Lebensraum) in the east and to destroy “Jewish Bolshevism.” It was an undertaking deeply rooted in Nazi ideology.13 The prospect of war with Germany was the central focus of Soviet military planning from the mid-1930s on.14 By early 1939, Soviet military intelligence was receiving reports that Hitler intended to use armed force to seize Ukraine, “the bread-basket of Europe.”15 The Nazi-Soviet Pact did not put an end to such warnings, and their number and specificity increased greatly after Germany’s stunning defeat of France. As its performance in the war with Finland showed, the Red Army was in a state of demoralization and disarray after the terrible purge Stalin had inflicted on it in 1937–1938. In 1939–1941 Stalin took steps to build up Soviet military power in response to the increasingly threatening strategic environment. The number of men under arms grew from under two million in January 1939 to over five million in June 1941.16 When it came to the point, however, Stalin refused to bring the Red Army up to full combat readiness. The Wehrmacht attacked on June 22, 1941, with a force of over three million men, and within three weeks it had advanced 300–600 kilometers into Soviet territory, putting in doubt the survival of the Soviet state. After the war Stalin forestalled critical discussion of the events of 1941 by presenting the Soviet retreat as part of a strategy of active defense.17 In the West, it was Churchill who set the tone: Stalin had committed a dreadful blunder in not heeding the warnings of Hitler’s attack.18 Khrushchev echoed this assessment in his “Secret Speech” to the Twentieth Congress of the Communist Party of the Soviet Union in 1956, when he took the failure to heed warnings of the German attack as one of the major points in his indictment of Stalin. “Everything was ignored,” Khrushchev declared, “warnings by certain Army commanders, statements by deserters from the enemy army, and even the open hostility of the enemy.”19 Assessments began to change in the 1970s. In Codeword Barbarossa, Barton Whaley took as his starting point the main conclusion of Roberta Wohlstetter’s classic 1962 study of Pearl Harbor: “it is only to be expected that the

40

Barbarossa and the Bomb

relevant signals, so clearly audible after an event, will be partially obscured before the event by surrounding noise.”20 In other words, only after an attack had taken place—or definitively not taken place—would it be possible to say with certainty which intelligence reports had been accurate and which had not. Whaley explored the ambiguities in the intelligence reaching Stalin. More important, he emphasized the role of Hitler’s campaign of strategic disinformation in deceiving Stalin as to German intentions. Wohlstetter’s thesis was inadequate, he claimed, because it treated intentional disinformation as part of the background “noise” rather than as a special type of signal, the aim of which was not to add to existing ambiguity but to create false certainty.21 The Soviet assessment of 1941 changed at about the same time. Nachal’nyi period voiny (The Initial Period of War), a study published by the Soviet Ministry of Defense in 1974, examined some of the complexities of surprise attack.22 Like Whaley’s book, it stressed the importance of the German disinformation campaign. It pointed also to the premises on which Stalin based his policy in the months leading up to the war: the belief that it would be possible to put off war until 1942, and the fear that Germany would seize on a provocative action by the Soviet Union as a pretext for war. These premises shaped the way in which Stalin construed the intelligence. It was not only Stalin’s misconceptions that were to blame, however, according to the Ministry of Defense study, for the Soviet High Command had based its plans on the false assumption that it would take two weeks after the enemy’s first strike for each side to complete the full deployment of its main forces.23 The problem of surprise attack was presented in this study as much more complex than Khrushchev’s critique of Stalin had portrayed it.24 The new evidence that has become available since the late 1980s has not led to a consensus on Stalin’s handling of intelligence. Dmitrii Volkogonov, who was able to use previously closed archives for his 1989 biography of Stalin, writes of Stalin’s “fateful miscalculations” and blames Stalin not only for failing to draw the right conclusions from the intelligence, but also—and more damningly—for creating a bureaucratic machine that was incapable of giving him independent advice.25 The compilers of 1941 god (The Year 1941), a two-volume collection of documents published in 1998, argue that Stalin had the information he needed to make the right decision: “the documents testify that the attack of 22 June was by no means a surprise attack, as Stalin later asserted. He, like the other top party, state, and military leaders, had at his disposal broad information about German intentions, obtained by



Barbarossa and the Bomb

41

intelligence and diplomatic organs.”26 Similarly, in a study of military intelligence from 1939 to 1943, Vladimir Lota concludes that the reports the Kremlin and the General Staff received from military intelligence in the first half of 1941 “were entirely sufficient for the leadership of the country to take measures to strengthen the security of the country and to bring the armed forces into a state adequate to the military threat hanging over the country.” Lota does add, however, that although Soviet leaders knew about the threat, they were misled by Hitler’s disinformation campaign.27 This is not, however, an entirely satisfactory position. It is not enough to say that among all the information Stalin had at his disposal was the information he needed to make the right decision. That ignores the problem of discriminating ex ante between the correct and the misleading information before events have established which is which. The intelligence has to be analyzed and interpreted, and that inevitably leads one to ask: who performs that function and what assumptions, preconceptions, and mind-sets do they bring to bear in the process? Not everyone subscribes to the view that Stalin had the information he needed. In his careful study of Soviet foreign policy in 1939–1941, Gabriel Gorodetsky emphasizes the indeterminate character of the intelligence Stalin received, arguing that “there was sufficient ambiguity in the vast intelligence offered to Stalin for him to be convinced that the attack might be deferred, or at best unleashed at a time of his own choosing, if he played his diplomatic cards well.”28 Geoffrey Roberts reaches a similar conclusion in Stalin’s Wars.29 This also leads one to ask: why was the intelligence analyzed and interpreted in the way it was? What Intelligence Did the Soviet Union Acquire? Between May 1939 and December 1940, the Intelligence Directorate of the Soviet General Staff received more than twenty-five reports from its residencies abroad that Germany had begun to plan for war against the Soviet Union; in the same period it prepared thirteen reports and summaries for the top leadership. The flow of intelligence grew rapidly as German preparations moved forward. Between January and June 1941, military intelligence received 267 reports from its residencies and brought 129 of these to the attention of the political and military leadership. Almost every day military intelligence reported to Stalin and other leaders on German intentions and capabilities.30 A similar increase took place in the activities of foreign intelligence.31

42

Barbarossa and the Bomb

There were several intelligence organizations in Moscow, but there was no central clearinghouse, no analytical unit to collate and evaluate the intelligence coming in from the different sources. There was an information department in the General Staff’s Intelligence Directorate and the NKGB had an information-analytical department, but these departments were apparently of poor quality, and their primary function appears to have been to assess the quality of the sources from which information was obtained rather than the information itself.32 As the editors of a collection of NKVD and NKGB documents from the eve of the war point out, only Stalin had the right to make a final evaluation of the intelligence material and to take decisions on the basis of it. “As the available documents testify,” they write, “at that time foreign intelligence had basically information functions; analytical work—an essential part of intelligence activity—was not carried out.”33 General F. I. Golikov, who was chief of military intelligence from July 1940 to November 1941, sent reports directly to Stalin and other leaders, but he had little personal contact with Stalin, with whom he met only twice during that time.34 P. M. Fitin, the chief of foreign intelligence, sent reports to Stalin through V. N. Merkulov, people’s commissar of state security from February 1941, and L. P. Beria, people’s commissar of internal affairs, with the latter serving as a significant filter.35 The key questions for Soviet intelligence were: Was Hitler planning to attack the Soviet Union? And if so, when, where, and with what forces? Numerous warnings of Hitler’s intention to attack reached Moscow in the twelve months before June 1941. Soviet intelligence did not have an agent on Hitler’s staff; nor did it have anyone in a position to hand over German military plans, but the volume of intelligence it received from agents who worked in the German government or were close to German military and political circles was nonetheless extensive. The intelligence reports fell into two broad categories. The first was information on German preparations for war, including the buildup of German forces along the Soviet border. This came from numerous reports, which left no doubt about the scale of the German buildup. The second consisted of reports on Hitler’s intentions. Most of these were unequivocal that Hitler was planning to attack the Soviet Union; others reported that German forces in the east were being built up for operations in the Balkans or for an attack on British forces in the Middle East.36 On June 11, 1941, according to Marshal G. K. Zhukov, then chief of the General Staff, in response to a request to bring the forces in the border military districts up to



Barbarossa and the Bomb

43

full combat readiness, Stalin went up to a map and pointed to the Middle East and said: “That’s where they [the Germans] are going.”37 In late June 1941, military intelligence drew up a list of intelligence reports it had received in the previous months. Why it was drawn up is not clear; nor is it clear what criteria were used in compiling it. It does not list all the reports received, but it does provide a picture of the kind of intelligence the Soviet Union was collecting. 38 The list names fifty-six reports about German preparations for war, all but two of which had been received since January. These reports left no doubt that Germany was planning to attack. Thirty-seven of them contained estimates of the date of a German attack, though the estimates were not precise or consistent. The earlier reports identified March or the spring as the likely time of the attack; in the later reports, the date was moved back to May or June. Sixteen of the intelligence reports on the list gave estimates of where the main German blow would fall, with ten of those identifying Ukraine as the primary target. The commonest explanation for Hitler’s intention to attack the Soviet Union was Germany’s need for food, coal, and oil. If that was indeed Hitler’s purpose, Ukraine was the obvious place for him to strike; one report said that he would aim also for Baku.39 One of the best sources Soviet military intelligence had was Rudolf von Scheliha (code name “Ariets”), a member of the German diplomatic corps who had good contacts in military and political circles in Berlin. Von Scheliha remained in touch with Ilse Stöbe (code name “Alta”), a journalist who served as a contact between the Berlin residency and a network of agents.40 On December 29, 1940, eleven days after Hitler signed Directive No. 21, the Soviet Berlin residency reported to Moscow that “Ariets” had informed “Alta” that Hitler had given the order to prepare for war with the Soviet Union.41 According to this report, war was to be declared in March 1941. When asked for confirmation, “Ariets” responded that he had it from a good source that this was not just a rumor, that Hitler had indeed signed an order. The Berlin residency sent this response to Moscow on January 4, 1941. In the same message, however, it reported that the Germans were counting on “putting the English on their knees in the spring and thus freeing their hands in the east.”42 The implication was that Hitler would attack the Soviet Union after dealing with the British. Directive No. 21, however, had explicitly ordered German forces to be ready to attack the Soviet Union before the war with Britain had been concluded.

44

Barbarossa and the Bomb

The most accurate report Soviet intelligence received about the strategic concept of Operation Barbarossa came from the Berlin residency on February 28, 1941. On the basis of information from “Ariets,” “Alta” reported that knowledgeable military circles were certain that war with the Soviet Union would start in 1941, and that preparations should be well under way.43 Three army groups had been formed under Field Marshals Wilhelm Ritter von Leeb, Fedor von Bock, and Gerd von Rundstedt, to attack Leningrad, Moscow, and Kiev. The proposed date of the attack was May 20, and an enveloping strike would be delivered in the region of Pinsk by 120 divisions. Russian-speaking officers and NCOs had been deployed to the staffs; special armored trains were being built with the proper gauge for the Soviet railway lines. It is clear in retrospect that this information was largely correct: three army groups had been formed under the men named in the report, and the main German blow came north of the Pripet marshes.44 It was not clear at the time, however, that this information was accurate. Soviet intelligence was more or less consistently wrong in estimating that the main German attack would come not from the west, but from the southwest. In June 1941, the NKGB also drew up a list. This named twenty-five reports from two of its key agents in Berlin: “Korsikanets” (the Corsican) and “Starshina” (Sergeant-Major).45 “Korsikanets” was the code name of Arvid Harnack, an official in the German Economics Ministry; “Starshina” was Harro Shulze-Boyson, an officer on the Main Staff of the German Air Force. Both men were members of the Rote Kapelle spy ring. The overwhelming impression created by their intelligence reports, as summarized in the list, was that Germany was getting ready to attack. What is also striking about their reports, however, is the changing and contradictory information they provided. For example: on April 14, they reported that Germany would issue an ultimatum before attacking; on April 24, that the attack on the Soviet Union had been postponed in favor of an invasion of the Middle East; on April 30, that a final decision had been taken to start a war with the Soviet Union; on May 9, that there would be a “war of nerves” to demoralize the Soviet Union before an ultimatum was presented; on June 9, that an ultimatum was to be expected and that the decision to attack the Soviet Union had been put off until the middle of June; on June 11, that the decision to attack had been taken; and on June 16, that everything was ready for an attack.46 Stalin wrote on the last of these reports: “To Comrade Merkulov. You can send your ‘source’ from



Barbarossa and the Bomb

45

the German Aviation Staff to his f***ing mother. This is not a ‘source’ but a ‘disinformer’ [dezinformator].”47 The intelligence received by the Soviet Union was impressive in its volume, but three things lessened its value. The first was that the intelligence organizations classified as unreliable some agents who were later shown to have provided accurate information and trusted some whom they should not have trusted, thereby depriving themselves of important intelligence and opening themselves to disinformation. This is a common danger in the world of intelligence, but it was amplified by the suspiciousness and search for traitors that was an intrinsic feature of Stalinist politics. Richard Sorge (code name “Ramsay”) is a notable example of the first category. Sorge worked as a journalist in Tokyo where he had operated a ring since the mid-1930s. He was close to the German ambassador and supplied information about German and Japanese plans. After the war it became clear that he had been an important source of intelligence, but Golikov had treated his reports with suspicion, frequently classifying them as “doubtful.”48 An example of the second category is Orestes Berlings (code name “Litseist”), a Latvian journalist working in Berlin, whom German counterintelligence successfully planted on A. Z. Kobulov, the NKVD-NKGB station chief in Berlin from 1939 to 1941. Kobulov recruited Berlings as an agent, and German intelligence used Berlings to plant disinformation that Kobulov sent to Moscow.49 The second problem was that Soviet intelligence did not have an agent at the heart of the Nazi regime and had to rely on sources that might mix reliable information with gossip or speculation. There were agents close to German military and political circles, but it was precisely in such circles that accurate information could most easily become mixed with rumors and disinformation. This meant that accurate and inaccurate information was frequently to be found in the same report, as, for example, in the report from the Berlin residency on January 4, 1941, that stated that Hitler was planning to attack the Soviet Union but was counting on dealing with Britain first. The third difficulty Moscow faced was that the reports it received were sometimes contaminated by Germany’s campaign of deception, as well as by rumor and speculation. The German High Command issued directives on this campaign on February 15 and May 12, 1941.50 The major theme was that Germany was continuing to prepare for an invasion of Britain, and that Hitler would not attack the Soviet Union until Britain had given up the fight. Other recurrent stories, some disseminated by German intelligence, found their

46

Barbarossa and the Bomb

way into intelligence reports sent to Moscow: German preparations were to ensure defense against a possible Red Army offensive; Germany would issue an ultimatum to the Soviet Union and attack only if the ultimatum were not accepted; German troops were stationed in the east to put pressure on the Soviet Union so that it would give in to an ultimatum; there were splits in the German leadership about the advisability of going to war with the Soviet Union. None of this was true: in September 1940, Hitler had postponed indefinitely the idea of invading Britain; he did not fear a Soviet attack, for he regarded the Red Army with contempt; he gave no thought to the idea of an ultimatum, for he intended to destroy the Soviet Union, not to extract concessions from it; nor was there a serious challenge in the German leadership to Hitler’s determination to attack the Soviet Union. At the operational level, Germany took measures to convince Soviet intelligence that the main attack would come in the southwest, and those measures had the desired effect.51 How Did the Soviet Union Use the Intelligence It Acquired? Stalin had taken a close interest in foreign and military intelligence since the early 1920s, finding it useful in domestic political intrigues. This interest was not always beneficial to the intelligence organs. It is hardly surprising that Golikov, whose five predecessors as chief of military intelligence had been arrested and shot, confessed in 1965 to the historian Viktor Anfilov that he had told Stalin what he wanted to hear, because he was afraid of him: Stalin “had formed the opinion that Germany would not attack us before it had finished the war with England. Knowing his character, we framed our conclusions to meet his point of view.”52 Golikov had no experience in intelligence before his appointment and, in spite of his understandable eagerness to please, Stalin seems not to have thought highly of him, remarking in February 1941: “as an intelligence officer, he is inexperienced, naïve. An intelligence officer ought to be like the devil: trusting no one, not even himself.”53 Molotov complained many years later that on the eve of the war he had “spent half a day every day reading intelligence reports. What wasn’t there in them! What dates didn’t they name! And if we had not stood firm, the war could have started much earlier.”54 This last cryptic remark implies that if the Soviet leaders had believed the intelligence reports they were receiving, they might have mobilized the armed forces sooner and thereby precipitated war. Molotov commented: “I think you can’t rely on intelligence officers. You have to listen to them, but you have to check up on them. Intelligence people



Barbarossa and the Bomb

47

can push you into such a dangerous position that then you don’t know where you are. You can’t count the number of provocateurs there are here and there. Therefore you can’t rely on intelligence people without the most careful, constant checking and rechecking.”55 Molotov was aware, however, that suspicion and skepticism could be carried too far. “You can’t rely on individual testimony,” he said. “But if you are, so to say, too distrustful, it is very easy to go to the other extreme.”56 That of course was the crux of the matter. Suspicion and distrust could lead to solipsism: the only person one could believe was oneself and, as Stalin noted in his comment on Golikov, even that might be dangerous. The Soviet leaders were receiving a great volume of intelligence. How should they read it? What should they believe? There was overwhelming evidence in the intelligence reports to indicate that Hitler had decided on war against the Soviet Union and was preparing his forces for it. But the intelligence was not entirely consistent. Some reports suggested that Germany would not attack the Soviet Union until it had dealt with Britain. There was also evidence that Hitler’s aim was not to attack the Soviet Union but to win concessions by exerting military pressure. What has to be explained, then, is why Stalin chose not to act on the evidence of an impending German attack but to rely on the evidence that war could be postponed. Stalin’s policy appears to have been based on four crucial assumptions, for each of which supporting evidence could be found in the intelligence reports. The first was that Hitler understood that it would be a strategic error for Germany to find itself in a war on two fronts and would therefore not start a war with the Soviet Union before either defeating Britain or signing a peace treaty with it. In his most unequivocal statement of this premise, Stalin told the new graduates from the military academies on May 5, 1941: “In 1870, the Germans beat the French. Why? Because they fought on one front. The Germans suffered defeat in 1916–1917. Why? Because they fought on two fronts.”57 It was not that Stalin “trusted” Hitler, as is sometimes claimed; rather, he assumed that Hitler, for whom he had high regard as a political leader, would not make what he saw as a serious strategic mistake. Stalin’s reading of history apparently convinced him that German leaders, including Hitler, had drawn from their own history the lesson that a two-front war was to be avoided. The German deception campaign reinforced that conviction by taking elaborate measures to convince the Soviet Union that Germany was still planning to invade Britain.58

48

Barbarossa and the Bomb

The second assumption was that Britain was desperate to draw the Soviet Union into the war with Germany. This was not an unreasonable assumption from the point of view of either realpolitik or Marxism-Leninism, and it framed the way intelligence reports were assessed in Moscow. A prime example of this framing is Golikov’s report of March 20, 1941, to the top Party, state, and military leadership. This provided a summary of information received from agents in the period from July 1940 to March 1941. The first sentence of the report reads: “Most of the agents’ data concerning the possibilities of war with the USSR in the spring of 1941 come from Anglo-American sources whose mission today is without doubt to strive to worsen relations between the USSR and Germany” (emphasis added).59 Golikov goes on to say that, given the nature of fascism and of Hitler’s plans as laid out in Mein Kampf, some of the intelligence agents’ reports deserved serious attention. In the body of the report, he outlines various statements from intelligence reports without making much effort to assess their relative value. At the end, he offers two conclusions. The first was that the most likely date for the beginning of military action against the Soviet Union was the moment when Britain has been defeated or has concluded peace with Germany. The second was: “The rumors and documents that speak of the inevitability of war against the USSR this spring must be assessed as disinformation coming from British and even perhaps German intelligence.”60 The third assumption that shaped Stalin’s thinking was his expectation that an ultimatum demanding territorial or economic concessions would precede a German attack. Hitler never seriously considered making such an ultimatum. The idea of the ultimatum nevertheless found its way into the intelligence reports, whether inspired by the German deception campaign or merely the result of rumor.61 It dovetailed with Stalin’s belief that what Hitler wanted from the Soviet Union was food and fuel, and that the main attack would come from the southwest. In October 1940, Stalin had told People’s Commissar of Defense S. K. Timoshenko and Chief of the General Staff K. A. Meretskov that “Ukrainian grain and Donbass coal have particular importance for the Germans,” and had asked them to adjust the new operational-strategic war plan to reflect the likelihood that the main German blow would come from the southwest.62 The prospect of a German ultimatum lent support to the idea that war could be delayed by negotiation or concessions. Stalin—like Soviet intelligence—failed to grasp that what Hitler wanted was not concessions from the Soviet Union but the complete annihilation of the Soviet state.



Barbarossa and the Bomb

49

When Georgii Dimitrov, secretary of the Executive Committee of the Comintern, went to the Kremlin at 7 a.m. on June 22, Stalin complained that the Germans had not presented an ultimatum: “They attacked us without declaring any grievances, without demanding any negotiations; they attacked us viciously, like gangsters.”63 The fourth assumption was that the German leadership was divided on the issue of war with the Soviet Union. Golikov referred in his March 20 report to information suggesting that there were two currents of thought among Germans. The first was that the Soviet Union was weak and now was the time to finish it off; the second was that the Russian soldier was very effective in defense and war would therefore be risky, and that it would be better to maintain good relations with the Soviet Union.64 A report from “Starshina” on April 17, 1941, pointed to two groups in the German government: one, headed by Hermann Göring, chief of the Luftwaffe, was eager for war with the Soviet Union; the other, headed by the Foreign Minister, Joachim von Ribbentrop, was opposed to it.65 There were other reports of divisions in the German leadership, with the High Command favoring war.66 The German ambassador in Moscow, Werner von Schulenberg, who opposed the war, may have contributed to that impression by his attempts to persuade the Soviet government to keep sending signals to Berlin that it wanted good relations with Germany.67 These four assumptions appear to have played an important role in Stalin’s reading of the intelligence, and there was evidence for each of them in the intelligence materials. They reinforced Stalin’s own preferences: he expected war with Germany and did much to prepare for it, but he evidently hoped to put it off. He told Churchill in August 1942 that he had not needed warnings because he knew that war would come but hoped he could gain another six months.68 (Over the past twenty years or so, a school of thought has emerged in Russia that claims that Hitler’s attack on the Soviet Union forestalled a plan by Stalin to attack Germany and spread revolution to the rest of Europe. This is not the place to review the controversy, except to note that several authors, including Gorodetsky in Grand Delusion, have effectively rebutted this thesis.) 69 Taken together the four assumptions framed the intelligence in such a way as to underpin Stalin’s hope that war could be postponed. Moreover, they invalidated, or at least cast doubt on, much of the intelligence received by Moscow. Stalin’s policy would have made perfect sense if the premises on which he viewed the intelligence had proved to be true, but they were mistaken. Hitler’s

50

Barbarossa and the Bomb

strategic goal was to destroy the Soviet Union, and to do so without issuing an ultimatum or seeking concessions and without waiting until Britain had finally been dealt with. To argue that Stalin brought a particular set of assumptions to the evaluation of intelligence is not to absolve him of responsibility for the disaster of 1941, but it does help to explain why he interpreted the intelligence in the way that he did. In two important respects, however, his approach to the intelligence is surprising. In the first place, it is surely strange that Stalin should regard the war in the West in 1940–1941 as a second front, since Britain had no troops left on the continent. He certainly changed his mind after the German attack on the Soviet Union, when he began to press Britain very hard to open a “second front” in Europe. Second, it is striking that Stalin failed to grasp the ideological character of the war that Hitler was planning to launch. He thought that the desire for access to Soviet resources was the driving force behind Hitler’s policy. Apart from the brief reference to Mein Kampf in Golikov’s memorandum of March 20, 1941, there seems to be no discussion of Nazi ideology in the intelligence reports or in the contemporary assessments of Hitler’s intentions. This is surely surprising in view of the well-known and frequently expressed Nazi hatred of “Jewish Bolshevism.” What Were the Consequences for International Relations? Marshal A. M. Vasilevskii, deputy chief of the Operations Directorate of the Soviet General Staff in 1940–1941, many years later gave a clear statement of the choice Stalin made: There were reasons enough to try to delay the USSR’s entry into the war, and Stalin’s tough line not to permit what Germany might be able to use as a pretext for unleashing war was justified by the historic interests of the socialist motherland. His guilt consists in not seeing, in not catching, the limit beyond which such a policy became not only unnecessary but also dangerous [emphasis added]. That limit should have been crossed boldly, the Armed Forces should have been brought as quickly as possible into full combat readiness, mobilization carried out, the country turned into an armed camp.70

Stalin gave high priority in 1940–1941 to preparing the country for war, raising the number of troops, and increasing defense production. The situation in Europe was dangerous, especially after the defeat of France. The High Command initiated reforms designed to eliminate deficiencies that the Finnish War had exposed in the Red Army. It also studied the military campaigns



Barbarossa and the Bomb

51

of the first year of World War II to see what lessons should be drawn from them.71 In April and May 1941, the General Staff began the secret mobilization of over 800,000 reservists. In May, it decided to move additional divisions to the USSR’s western and southwestern borders and issued orders to speed up the construction of fortified regions along the new border. Between June 12 and 15, it ordered the western military districts to move their divisions closer to the border. From June 12, Timoshenko and Zhukov (who had become chief of the General Staff in January 1941) pressed Stalin every day to allow Soviet forces to be brought up to full combat readiness, but Stalin was reluctant to follow their advice.72 “Are you trying to push us to provoke war?” he asked them. “Now the main thing is not to provoke military clashes, the situation has heated up, we must be careful.”73 The order to bring the armed forces in the border military districts up to full combat readiness was not prepared until the evening of June 21 and not sent out until after midnight, just hours before the German attack.74 The result of Stalin’s policy was not total inaction, but rather a disastrously sluggish response to the German preparations for war. By June 22, 1941, the organizational restructuring and rearmament of the Red Army were under way but far from complete. Work on fortified regions along the new border had been started but not finished. Many of the divisions that had been ordered in May to move from the interior to the western military districts were still in transit. Divisions lacked their full complement of horses and transport and found it difficult to move equipment and supplies to their forward positions. (The state and collective farms were supposed to transfer horses and carts to the Red Army in the event of war, but doing that would greatly disrupt the harvest.) The tank and motorized divisions of the new mechanized corps were deployed far from each other and unable to inflict counterstrikes to crush the German forces that wedged themselves between them. Aircraft were lined up on airfields without camouflage and vulnerable to enemy strikes.75 Moreover, the Soviet Union had prepared for a main blow coming from the southwest and had assigned an additional twenty-five divisions to that front in May and June.76 Zhukov later acknowledged that the Soviet armed forces were too weak to have been able to stop German forces from penetrating deep into Soviet territory, but if the Red Army had been brought up to full readiness and had correctly anticipated where the main blow would come from, the Wehrmacht would have met stronger resistance.77 Stalin’s reluctance to move from caution to bold action had serious consequences.

52

Barbarossa and the Bomb

Stalin’s decision depended, not only on an assessment of Hitler’s intentions, but also on an evaluation of the consequences of making the wrong decision. If Hitler was not intending to attack, then caution was the appropriate policy and bringing the Red Army up to full readiness would be dangerous; as Stalin feared, it might precipitate the very attack he wanted to avoid. If Hitler was intending to attack, then bringing the Red Army up to full readiness was the correct policy and failure to do so in good time would give Germany a significant military advantage. Stalin had to weigh the consequences of both potential wrong decisions, of a false negative as well as a false positive. In making his choice, Stalin underestimated the consequences of not bringing the Red Army up to full readiness. He apparently believed that a German attack would be preceded by an ultimatum demanding territorial and economic concessions, and such an ultimatum could be used to delay the start of the war. Even if there were no ultimatum, the Soviet war plan assumed that the war would begin with border clashes, on the grounds that it would take several days for the two sides to bring their main forces to bear.78 Germany’s victories over Poland and France did not upset this assumption, for the Soviet leaders did not believe that a similar blitzkrieg could be launched against a state as large and powerful as the Soviet Union. It was not foreseen that Germany would be able to launch a sudden attack with such massive force. In his memoirs, Zhukov writes: “Neither the defense commissar [Timoshenko], nor I, nor my predecessors B. M. Shaposhnikov and K. A. Meretskov, nor the leaders of the General Staff expected that the enemy would concentrate such a mass of armored and motorized forces and commit them on the first day.”79 Stalin also overestimated the dangers of taking action to bring the Red Army up to full combat readiness. He was fearful of moving too early lest he provoke the very war he wished to avoid. This fear was linked to his belief that elements in the German leadership—the High Command in particular—were eager for war and might use Soviet measures to raise combat readiness as a pretext for launching an attack. The commanders of the western military districts were called to Moscow in June on Stalin’s orders to receive strict instructions from Timoshenko and Zhukov about the need for vigilance but also about the need to avoid doing anything that might provoke the Germans.80 The fear of provocation is reflected in the order to bring the forces in the western military districts up to full combat readiness, which was prepared late in the evening of June 21 in response to increasing signs of a German attack and sent out after midnight. This admonished Soviet forces “not to give



Barbarossa and the Bomb

53

way to provocative actions of any kind which might produce major complications.”81 The fear of provoking a German attack had no basis in fact. Hitler was set on war. He did not need to be provoked into it. Stalin’s fear that Soviet actions might provoke war was misplaced. It is of course possible that bringing Soviet forces up to full readiness might have provoked Hitler to attack earlier than he did, but presumably in that case, the Red Army would have been more ready to meet the attack than it was in fact as a result of avoiding actions that might appear provocative. Important though Stalin’s response to intelligence was in 1941, it cannot by itself explain the military situation the Soviet Union faced. Soviet military doctrine, as it developed in the 1920s and 1930s, provides a crucial part of the context. Soviet doctrine stressed the offensive; defense and defensive operations were relatively neglected as a result.82 The emphasis on the offensive sprang from a sense of revolutionary optimism and a belief in the potential of new weapons such as tanks and aircraft. Soviet doctrine assigned to the armed forces the dual mission of repulsing the enemy’s attack and simultaneously crushing him decisively in military operations carried out on his territory. The Red Army was therefore supposed to be ready to respond to a surprise attack by inflicting a crushing blow on the enemy.83 Military planning was based on the assumption that in the first hours and days of war, the Red Army would conduct holding operations in fortified regions constructed for that purpose and that it would be able, by virtue of its high combat readiness, to neutralize a surprise attack.84 It would then take the offensive in order to deliver a devastating response to the enemy. This military doctrine had three important implications for Soviet policy. First, it influenced the deployment of Soviet forces. The first strategic echelon of the Red Army was deployed close to the border. When the Soviet Union seized the eastern part of Poland in 1939, it abandoned the existing defensive line (the “Stalin line”) and began to build fortified regions along the new border, but these were not completed by the time of the German invasion. The decision to move the defensive line was apparently opposed by B. M. Shaposhnikov, the chief of the General Staff at the time, and is now widely seen to have been a mistake.85 Second, Soviet military doctrine put very heavy weight on the political and military leadership’s ability to understand the enemy’s intentions and to bring Soviet forces up to full readiness in good time. Stalin clearly failed on that score in 1941. Timoshenko and Zhukov feared that Germany’s

54

Barbarossa and the Bomb

mobilization would enable the Wehrmacht to launch a surprise preemptive attack as Soviet forces were being deployed. They were so alarmed that in May 1941, they prepared a plan for a preemptive strike. Stalin had said at the reception for graduates of the military academies on May 5 that “the Red Army is a modern army and a modern army is an offensive army.”86 Timoshenko and Zhukov appear to have taken encouragement from this; they proposed a shift from the existing plan for a Soviet counterattack to a Soviet preemptive strike in the event of war, with the goal of crushing German forces and gaining control of former Polish territory and East Prussia. The significance of this document, dated May 15, 1941, is a matter of some dispute. It would have involved the secret mobilization and concentration of the Red Army, and it would have taken time to get ready to implement it. In any event, it appears that Stalin rejected it categorically, and certainly he did not act on it.87 The proposal shows, however, just how concerned Timoshenko and Zhukov were about the way in which a war might begin. Third, the war plans based on Soviet military doctrine depended for their success on a correct understanding of the initial period of war. There was one Soviet military analyst, G. S. Isserson, who grasped that this had changed with the German attack on Poland. “No one,” he wrote in 1940, “can now say when mobilization, concentration, and deployment have taken place—actions that by the example of previous wars and in particular of the first imperialist war were marked by wholly defined time frames.”88 The mobilization and concentration of forces would now take place inconspicuously and gradually before the attack. Isserson’s was a lone voice, however, and his arguments were not taken into account. The assumption that mobilization would be completed after the initial clashes continued to shape Soviet thinking about the way in which war would begin. As Zhukov commented in his memoirs, neither he nor his predecessors thought that Germany could concentrate such a mass of forces and commit them on the first day of the war. The Red Army found itself in an impossible position in June 1941, for it was prepared neither for offense nor for defense. It was deployed forward in order to be able to take the offensive after blocking an enemy attack, but it was not brought up to full readiness in good time, and its plans were based on a misleading conception of the way in which the war would start. It was not deployed for defense in depth: it was not willing to give up territory in order to deny the enemy the ability to break through the defense, because such a breakthrough was not envisaged by the doctrine or the war plans.89 Its



Barbarossa and the Bomb

55

difficulties were further compounded by a factor that all too quickly became apparent: the sheer might of the German military machine. Soviet intelligence had focused on the size of the German forces and their deployment, as well as on the date of a possible attack, but perhaps the greatest surprise on June 22, 1941, resulted from “the faulty, uncritical assessment of Hitler’s military machine and of the power of the German army.”90 Neither Soviet intelligence nor the Soviet leadership had grasped just what a formidable fighting force the Wehrmacht was.91 Atomic Espionage Soviet scientists greeted the discovery of nuclear fission early in 1939 with as much excitement as their colleagues in other countries, and in 1939–1941, they did some pioneering work on nuclear chain reactions. Leningrad was the main center of this research, and Igor Kurchatov was the central figure in organizing work on nuclear fission. Soviet research paralleled the work being done in Germany, France, Britain, and the United States. Like their colleagues abroad, Soviet scientists were interested in the potential of nuclear fission both as a source of electric power and as an explosive. The German invasion put an almost complete stop to Soviet nuclear research, which was not resumed until 1943.92 In spite of the intensive work being done on nuclear chain reactions in various countries, in early 1940, the atomic bomb seemed to be a very remote possibility, and not one that could have a bearing on the war that had begun with the German invasion of Poland. The picture changed in March 1940, when two refugees from Nazism, the physicists Otto Frisch and Rudolf Peierls, who were working at the University of Birmingham, wrote a memorandum showing that the atomic bomb was much more feasible than anyone realized. The following month, in response to this memorandum, the British government set up a committee, known as the Maud Committee, to investigate the possibility of using atomic energy for practical purposes, both as a source of electric power and as a bomb.93 The Maud Committee, which consisted of six leading physicists, completed its work in July 1941. Its report made a cogent case that an atomic bomb could be built within two and a half to three years, before the war with Germany was likely to be over. On the basis of the Maud Report, Churchill decided that Britain should take steps to build an atomic bomb. The report

56

Barbarossa and the Bomb

also played a role in Washington’s decision to make the bomb. When Vannevar Bush, director of the Office of Scientific Research and Development, briefed President Roosevelt on the report on October 9, 1941, Roosevelt authorized Bush to speed up the American effort. This was a key decision, setting in motion the chain of events that would lead to the development of the bomb and its use against Hiroshima and Nagasaki. In McGeorge Bundy’s view, it stood alone among the decisions of World War II “in its speed, its loneliness, and the magnitude of its eventual consequences.”94 In June 1942, Roosevelt took the decision to proceed with the creation of the atomic industry needed to build the bomb. In the spring of 1943, scientists began to arrive at the new Los Alamos laboratory, where the bomb was to be designed and fabricated. What Nuclear Intelligence Did the Soviet Union Acquire? In August 1941, Klaus Fuchs, a refugee physicist from Nazi Germany who was working with Peierls at the University of Birmingham, contacted a member of the military attaché’s staff at the Soviet embassy in London and told him about his own research for the British atomic project. He also informed him that Werner Heisenberg was working in Leipzig on the uranium bomb, and that such a bomb could be immensely destructive.95 Not long afterward, an event of even greater importance from the point of view of intelligence took place. In late September or early October, A. V. Gor’skii, the NKGB station chief in London, sent a copy of the Maud Committee’s report to Moscow. He also sent the appendices to the report, as well as minutes of meetings of the Defense Services Panel of the British Cabinet Scientific Advisory Committee, where the report had been discussed in September. The source of this information was John Cairncross, the “fifth man” of the Cambridge Five, who worked at the time as private secretary to Lord Hankey, chairman of the Cabinet Scientific Advisory Committee.96 Fuchs continued to supply the Soviet Union with information about his own work on isotope separation, though Moscow was more interested in the scale of the British effort than in Fuchs’s theoretical calculations. Fuchs went to the United States in December 1943 as part of the British mission to the Manhattan Project. At first he worked in New York on calculations for the gaseous diffusion isotope separation plant and supplied information on that work to the Soviet Union. In August 1944, he was transferred to Los Alamos, where he joined the Theoretical Division. He was now in a position to provide detailed information about the design of the atomic bomb. Fuchs was



Barbarossa and the Bomb

57

undoubtedly the most important source Soviet intelligence had in the Manhattan Project, but it had other sources too.97 One of these informed Moscow about Enrico Fermi’s success in achieving a nuclear fission chain reaction in Chicago in December 1942. Other sources provided information on fundamental physical data, isotope separation, and nuclear reactors. The Soviet Union also received information about the organization and progress of the Manhattan Project: in February 1945, the NKGB reported that the United States might test the bomb in the next two or three months.98 Early in 1945, Kurchatov received reports about the spontaneous fission problem in plutonium, and about implosion as a means of detonating the plutonium bomb. This was an extremely important issue. In the summer of 1944, it was discovered at Los Alamos that one of the isotopes of plutonium has a very high rate of spontaneous fission. Unless the plutonium core could be compressed very rapidly, the chain reaction would begin before the material was fully compressed and would then fizzle. Designing the implosion mechanism for the plutonium bomb was the hardest problem Los Alamos had to solve. The first report to reach Moscow about this problem came from Theodore Hall, a young physicist working at Los Alamos, and the second from Klaus Fuchs. Fuchs’s report was the more detailed of the two: the data on the high spontaneous fission rate of plutonium were “exceptionally important,” Kurchatov wrote.99 In June, Fuchs passed on more information, including a full description of the bomb that was to be tested at Alamogordo. He provided a sketch of the bomb and its components, including the implosive lenses, and gave all the important dimensions. He also reported that the test would take place on July 10 (in fact it took place on July 16).100 The Soviet Union had very little intelligence about the German project, but in the summer of 1943, it concluded, from sources in Britain as well as Germany, that German nuclear research was moving forward slowly.101 In May 1945, a group of Soviet nuclear scientists traveled to Germany to learn what German scientists had achieved and to gather materials, equipment, and people for the Soviet project. The British and Americans had rounded up the leading German scientists to prevent them from falling into Soviet hands, but about 200 German scientists, engineers, and technicians did move to the Soviet Union to work in the nuclear project.102 The most important result of the mission to Germany, however, was that the Soviet group found more than 100 tons of uranium oxide, which according to Kurchatov speeded up the building of the first Soviet reactor by one year.103

58

Barbarossa and the Bomb

How Did the Soviet Union Use the Nuclear Intelligence It Acquired? No action was taken by Moscow when the Maud Report arrived. It was only a year later that the decision was taken to start nuclear research again. On September 28, 1942, Stalin signed an instruction to the Academy of Sciences to set up a special nuclear laboratory and to report by April of the following year on the possibility of making a uranium bomb or uranium fuel.104 This was a detailed document, which had been prepared by the physicist Abram Ioffe and S. V. Kaftanov, who oversaw science policy for the State Defense Committee, in response to evidence of German interest in the bomb. The Soviet Union had by this time had the Maud Report for about a year and had received other information about nuclear research in Britain and the United States, but neither Ioffe nor Kaftanov was aware of this.105 Stalin himself may not have known of the intelligence from Britain and the United States when he signed the order on September 28, for it was only nine days later, on October 6, 1942, that Lavrentii Beria sent him a detailed memorandum about foreign work on the bomb.106 In February 1943, the State Defense Committee issued another order, appointing Igor Kurchatov as scientific director of the nuclear project and extending the deadline for the report on the feasibility of the bomb to July 5, 1943.107 Kurchatov first saw the intelligence materials, including the Maud Report, in October and November 1942.108 In the following months, as he assembled his laboratory and organized research, he reviewed more materials and gained a clear picture of the main directions of work abroad. He understood that Britain and the United States had moved far ahead of the Soviet Union. He learned about the plutonium path to the bomb, which Soviet physicists— unlike their American, British, and German counterparts—had not seen as a possibility, and he understood that this would make it possible to circumvent the enormously difficult task of uranium isotope separation. He learned in July 1943 of Fermi’s success the previous December in achieving a chain reaction, and he paid great attention to reports of American and British work on heavy water and graphite-moderated reactors.109 By early 1943 it was clear to him that there were several ways to build an atomic bomb, and that it would take far less time to do so than those without access to the intelligence materials understood.110 Beria had not hurried to inform Stalin about the intelligence from Britain and the United States; the Soviet Union was fighting for its survival when the intelligence reached Moscow. The nuclear project authorized by Stalin



Barbarossa and the Bomb

59

in September 1942 was very small and got under way slowly. This is perhaps hardly surprising. Soviet forces were desperately trying to hold on to Stalingrad when Stalin signed the order to renew research. Under the circumstances, a bomb that might take years to build was bound to be a low priority. But there was more to it than that. According to one of the officers involved in atomic intelligence, “from the very beginning Beria suspected disinformation in these reports, thinking that the enemy was trying to draw us in this way into huge expenditures of resources and effort on work that had no future.”111 There were suspicions at the time in Moscow that the Cambridge Five were double agents, and that might also have affected the attitude to the material transmitted by Cairncross.112 Even after Stalingrad, nuclear research did not enjoy high priority. Kurchatov, who understood better than anyone the contrast between the Manhattan Project and the limited Soviet effort, urged more than once that a special committee of political leaders and scientists be created to direct the project, but his advice was not heeded until after Hiroshima.113 In September 1944, he wrote to Beria, who oversaw the intelligence and internal security apparatus of the Soviet state, that the Manhattan Project represented a concentration of scientific and engineering forces “on a scale unseen in the history of world science,” while the situation in the Soviet Union was completely unsatisfactory. He complained about the supply of raw materials and the lack of cooperation from research institutes and from industry. He appealed to Beria to give instructions to have the Soviet project organized “in a way that corresponds to the possibilities and significance of our great state in world culture.”114 Whether Kurchatov ever sent this letter is not clear. In December 1944, Beria took over general responsibility for the atomic project from Molotov; his earlier suspicions about the intelligence material had evidently been dispelled. This was an important change because Beria was an energetic manager and had the whole police system, including the labor camps, at his disposal, but it did not mark the transition from laboratory research to the industrial production of fissionable materials.115 That was still nine months off. On May 15, 1945, Beria and Kurchatov wrote to Stalin requesting permission to increase the number of staff positions in Kurchatov’s laboratory from 230 in 1944 to 855 in 1945. Most of these positions were for workers and support personnel. Since Kurchatov’s laboratory was the main center of nuclear research, these numbers indicate that the Soviet effort in the middle of 1945 was still quite small, about the same size as the German wartime project.116

60

Barbarossa and the Bomb

Kurchatov was given scientific and technical intelligence to evaluate; apparently he was also given intelligence about the organization of the Manhattan Project. He prepared lists of questions based on the material he read and these were transmitted by the intelligence agencies to their informants. Kurchatov did not see intelligence relating to the recruitment and handling of agents.117 That was the business of the intelligence agencies. Among scientists, only Kurchatov reviewed all the scientific and technical intelligence material, though I. K. Kikoin and Iu. B. Khariton were shown some reports. During the war Kurchatov used the material to commission studies from colleagues and to draw up plans for the expansion of the project. Kurchatov’s assessments of the intelligence material show great appreciation, not only of the value of the intelligence, but also of the quality of the physics. He understood how rich the material was, and he was frustrated by the authorities’ unwillingness to give the Soviet project the priority he thought it deserved. It was only in August 1945 that Stalin decided to give top priority to building a Soviet bomb. He held a series of meetings on August 17–20 to decide on the organization of the project and on August 20—two weeks to the day after Hiroshima—he signed a decree setting up the Special Committee on the Atomic Bomb. This committee, which was subordinate to the State Defense Committee, was given extraordinary powers and resources. Its task was to build a Soviet bomb as quickly as possible. Stalin made Beria chairman of the committee, whose membership comprised two other leading Party figures (G. M. Malenkov and N. A. Voznesenskii), three industrial managers (B. L. Vannikov, A. P. Zaveniagin, and M. G. Pervukhin), and two scientists (I. V. Kurchatov and P. L. Kapitsa). General V. A. Makhnev, who served as head of the committee’s secretariat, was also a member.118 The decree also established the First Chief Directorate of the Council of People’s Commissars to organize and manage the research institutes, design bureaus, and industrial enterprises—primarily the isotope separation plants and the production reactors—of the new atomic industry. What Were the Consequences of the Intelligence for the Soviet Atomic Project? Stalin’s decision to give the highest priority to building the bomb was a remarkable exercise of political will. The Soviet Union had been devastated by the war; now it had to take up the challenge of building a whole atomic industry. At the first meeting of the Special Committee on August 24, 1945,



Barbarossa and the Bomb

61

Kurchatov gave a report on the state of research on the uranium problem in the Soviet Union. He outlined a range of subjects: the various methods for separating uranium-235 (thermal diffusion, centrifuges, and gaseous diffusion); the designs of graphite reactors and heavy water reactors and the materials needed for them; research on the design of atomic bombs; defense against atomic bombs; the research done in Germany and the contribution of the equipment and the scientists brought from Germany to the Soviet Union.119 It was a major undertaking to turn the research program that existed in August 1945 into an industry that would produce the atomic bomb in the shortest possible time. Uranium deposits had to be found and mined; isotope separation plants had to be built; plutonium production reactors had to be designed and constructed; the separation of plutonium from spent uranium fuel had to be mastered; and the bomb itself had to be designed and tested. A period of frantic activity followed the August 20 decree: uranium prospecting was stepped up; factories and institutes were assigned to the First Chief Directorate; scientists, engineers, and managers were recruited to the project. The Manhattan Project had employed about 120,000 people at its peak. By 1951 the First Chief Directorate had 68,800 people working in its plants. An additional 259,000 were engaged in construction. In the Soviet Union, 109,500 people were involved in prospecting for uranium and mining and milling it; an additional 190,500 were employed in the same activities in Eastern Europe, particularly East Germany. Thus the total number of people working in the Soviet atomic project in 1951 was over 600,000, and many of those—especially those involved in construction and mining—were camp laborers.120 How helpful was intelligence in this undertaking? According to Boris Vannikov, people’s commissar of munitions and first head of the First Chief Directorate, Stalin told him during the meetings on August 17–20 that “an atomic explosion had taken place and therefore the atomic bomb was no longer a problem but a practical task.” Stalin said that Soviet scientists had at their disposal “significant literary [sic] materials on atomic energy but they were being used unsystematically and ineffectively.” (“Literary” here is presumably a euphemism for “intelligence.”) It would be wrong, Stalin said, to be carried away by these materials; one should adopt an attitude of “sensible caution.”121 There is of course a great difference between acquiring a theoretical understanding of the atomic bomb and building an industry that will produce the

62

Barbarossa and the Bomb

materials needed for the bomb, as well as the bomb itself. It is important to bear that in mind when assessing the importance of intelligence for the Soviet atomic project. Nevertheless, the work that Kurchatov and his colleagues had done in 1943–1945, which was to a great degree driven by the intelligence received from abroad, was of considerable help in the months after August 1945 when the atomic industry had to be planned. It provided a picture of the effort that was needed to build the bomb. It complemented the Smyth Report, the official account of the Manhattan Project, which was issued in Washington in August 1945.122 Soviet intelligence managed to obtain drafts of the Smyth Report before it was published. It was quickly translated into Russian and distributed in an edition of 30,000 copies; it played an important role in the Soviet project because it made it possible to inform those involved in the project of the scale of effort required and of some of the difficult challenges involved.123 The intelligence information could not be used in the same way without exposing the fact that the Soviet Union had acquired secret information from abroad. The intelligence, however, had its own specific value. The information obtained during the war made it possible for Kurchatov and his colleagues to restart research, and it gave them time to work on the issues that the intelligence identified. Moreover, the intelligence made its own important contributions—for example, by pointing to the spontaneous fission problem in plutonium-240, by showing that this problem could be overcome by the implosion method, and by providing a detailed description of the design of the plutonium bomb tested by the United States at Alamogordo and dropped on Nagasaki. None of this information is contained in the Smyth Report. The intelligence influenced technical choices made by the Soviet Union, most notably the design of the first Soviet explosive device, which was a copy of the first American plutonium bomb. It was tested at Semipalatinsk on August 29, 1949. Margaret Gowing, the official historian of the British project, cites an estimate of one year saved by British participation in the Manhattan Project, if the contribution of the Maud Report is taken into account; and Bundy notes that there would have been no bomb in 1945 without this, even if that estimate is cut in half.124 The common estimate is that intelligence saved the Soviet project one to two years. That seems reasonable. Two factors should nevertheless be borne in mind. First, everything in the intelligence had to be checked; nothing could be taken on trust. This was not just a matter of fearing that there might be disinformation in the intelligence. It was also a



Barbarossa and the Bomb

63

matter of standard scientific practice: theoretical calculations had to be scrutinized and physical experiments replicated. It was important therefore that the Soviet Union had a strong physics community. Second, the length of time it took the Soviet Union to build the bomb was determined above all by the availability of uranium and the production of fissionable materials. The availability of uranium was a great worry in 1945; and the production of fissionable materials proved to be difficult. The first plutonium production reactor had to surmount numerous engineering problems before it could be brought into operation. The gaseous diffusion isotope separation plant also ran into technical difficulties.125 Thus intelligence made an important contribution to the Soviet project, but that contribution has to be understood in the context of the creation of an atomic industry. What Were the Consequences for the End of the Pacific War? When and how did the atomic bomb enter into Stalin’s strategic calculations? Antony Beevor has argued that the desire to seize German uranium as well as nuclear scientists and research laboratories was a major reason for Stalin’s determination to take Berlin before the Western allies did. Beevor’s claim is founded on faulty evidence, but it is appropriate to ask when and how the atomic bomb first had an impact on Soviet policy.126 Tsuyoshi Hasegawa has claimed that the atomic bomb was a factor in Stalin’s calculations about the war against Japan as early as 1943.127 This claim seems implausible, and Hasegawa offers no evidence to support it. If Stalin had believed in 1943 that the atomic bomb was so important, he would have given greater priority to the Soviet nuclear project. The evidence we have indicates that the bomb did not enter Stalin’s strategic calculations until July or August 1945. The question is whether it entered those calculations during the Potsdam Conference, which lasted from July 17 to August 2, 1945, or only with the destruction of Hiroshima on August 6. Hasegawa argues that during the Potsdam Conference Stalin realized that the United States now had the bomb, that it would use it against Japan, and that the use of the bomb would lead Japan to surrender. The available evidence points to a different interpretation: that Stalin was taken by surprise by the bombing of Hiroshima.128 On July 10, 1945, a week before the opening of the Potsdam Conference, Merkulov wrote to Beria that several reliable sources had reported that the United States would conduct its first experimental test of an atomic bomb in July, probably on July 10.129 (The test took place on July 16.) Merkulov’s

64

Barbarossa and the Bomb

memorandum said that this would be a plutonium bomb with an expected yield of five kilotons (in fact it was twenty kilotons). It is very likely that Beria gave this memorandum to Stalin before he left for Potsdam.130 The memorandum said nothing about the war or about the possible use of the bomb against Japan. Did this information affect the timing of Soviet entry into the war against Japan? At Yalta in February 1945, Stalin had reached agreement with Roosevelt and Churchill that two or three months after the end of the war in Europe, the Soviet Union would enter the war against Japan in return for important concessions in the Far East.131 In the middle of June, the Soviet Union planned to start operations against Japan between August 20 and 25, according to General S. M. Shtemenko, who as chief of the Operations Directorate of the General Staff was deeply involved in planning the war against Japan.132 That plan seems to have held firm for some weeks, but then the date of entry into the war was advanced to August 11. Stalin evidently wanted to move the date even further forward, for on July 16, the day he arrived in Potsdam, he telephoned Marshal A. M. Vasilevskii, commander in chief of Soviet forces in the Far East, to ask him to advance by ten days the date of entry into the war, to begin the offensive on August 1. Vasilevskii told Stalin that Soviet forces would not be ready by August 1; he asked that August 11 remain in effect as the date of entry, and Stalin agreed.133 This was a provisional date, because Stalin had not yet issued the order to begin operations against Japan. On the following day, July 17, Stalin told Truman that the Soviet Union would be ready to enter the war by the middle of August.134 At Potsdam, Truman approached Stalin after a plenary session on July 24, as the latter was about to leave the conference room. He casually mentioned that the United States “had a new weapon of unusual destructive force.” He did not say, however, that this was the atomic bomb. According to Truman’s memoirs, Stalin replied “he was glad to hear it and hoped we would make ‘good use of it against the Japanese.’” Stalin’s interpreter, who translated Truman’s remark, recalled that Stalin merely nodded his head.135 Molotov and Zhukov, who were both at Potsdam, later claimed that Stalin knew that Truman had the atomic bomb in mind; and A. A. Gromyko later recalled a conversation with Stalin at Potsdam about the atomic bomb.136 There is no contemporary documentation to confirm it, but it seems clear that Stalin understood what Truman was referring to. Molotov, Zhukov, and Gromyko all report that what concerned Stalin at Potsdam was that the United States



Barbarossa and the Bomb

65

and Britain would use the bomb to put pressure on the Soviet Union. General A. I. Antonov, the chief of the General Staff, informed Shtemenko at Potsdam that Stalin had told him that the Americans had a new bomb of great destructive power. According to Shtemenko, neither Stalin nor Antonov understood the significance of the bomb; at any rate, they did not give the General Staff new instructions for the war against Japan.137 After his conversation with Vasilevskii on July 16, Stalin appears to have taken no action at Potsdam to advance the date of the attack on Japan, in spite of the fact that Vasilevskii reported to him by phone every day during the conference.138 Stalin was eager to enter the war before Japan surrendered. There were two things to cause him anxiety. The first were the increasingly active Japanese approaches to Moscow, which suggested growing desperation on the part of the Japanese government to find a way out of the war. The second was Stalin’s chronic suspicion that the Western allies would conclude a compromise peace with Japan, thereby thwarting Soviet aims in the Far East and allowing Japan to remain a powerful military-political force. There is no evidence that before August 6, Stalin’s policy was driven by the fear that the United States would use the atomic bomb to end the war at a stroke; indeed, Shtemenko’s testimony suggests otherwise. If Stalin had understood just how destructive the atomic bomb was, and that it would soon be used against Japan, he would surely have applied further pressure on Vasilevskii to speed up preparations for war. That he did not do so suggests that he did not understand the impact the atomic bomb was about to have on the war and more generally on international politics. On August 3, Vasilevskii sent Stalin a report on the state of Soviet forces in the Far East; the best time to attack, he wrote, would be August 9–10. He asked that final instructions be given to him no later than August 5.139 Stalin accepted Vasilevskii’s advice and ordered that the attack be launched on August 10 at 2400 hours (Trans-Baikal time). On August 7, however, Stalin sent Vasilevskii a new order, advancing the attack on Manchuria by fortyeight hours.140 Soviet forces were now to begin their offensive on August 8, not August 10, at 2400 hours (Trans-Baikal time). Stalin’s order to Vasilevskii on August 7 contained no explanation of the change of date, but it seems likely that it was the atomic bombing of Hiroshima the day before that impelled him to speed up Soviet entry into the war. In order to secure the gains promised at Yalta, the Soviet Union had to enter the war. On the evening of August 8, three hours after the Soviet attack had begun, Stalin told Averell Harriman,

66

Barbarossa and the Bomb

the U.S. ambassador in Moscow, that “he thought the Japanese were at present looking for a pretext to replace the present government with one which would be qualified to undertake a surrender. The bomb might give them this pretext.”141 This suggests that Stalin believed that the bomb could precipitate a quick surrender by Japan. Hiroshima introduced a new element into international politics. The bomb was a now a factor in the balance of power. The evidence is indirect, but it seems clear that Hiroshima came as a shock to Stalin. He had known about the possibility of the atomic bomb since at least 1942, and he understood what Truman had in mind when he spoke at Potsdam of a weapon of unusual destructive power. But Hiroshima was a stunning demonstration of the power of the new weapon and also of the American willingness to use it. Stalin’s decision to launch a crash program to develop the bomb is eloquent testimony to the importance he ascribed to it once Hiroshima had demonstrated its power. Conclusion Four questions were posed at the beginning of this chapter. The first was: what intelligence did the Soviet Union acquire? In both cases, the Soviet Union obtained extensive information from agents abroad. It was impressive that Soviet intelligence networks continued to function at all after the enormous damage inflicted on them by the Great Purge of 1936–1938; it was even more remarkable that they were able to supply Moscow with a great deal of information about German preparations for war. In the atomic case, the Soviet intelligence services received very detailed information about the design of the bomb. There was, nevertheless, an important difference in the quality of the intelligence obtained in the two cases. The intelligence about German plans and preparations in 1941 was copious, but it was not of the highest quality. No copy of the Operation Barbarossa directive found its way to Moscow, for example; nor did operational plans for the war against the Soviet Union. Besides, as has been seen, the intelligence contained some reports that could be interpreted as indicating that Hitler was not planning to attack the Soviet Union. In the atomic case, by contrast, the Soviet Union obtained what were arguably the most important documents: the Maud Report, which made the case that the bomb could be built, and a detailed description of the design of the plutonium bomb. John Cairncross and Klaus Fuchs were, from the Soviet



Barbarossa and the Bomb

67

point of view, the right men in the right place at the right time. It was the equivalent of having a source on Hitler’s staff or on the staff of the German High Command. There may have been contradictory intelligence about the bomb and the Manhattan Project but there was evidently enough material of a very high quality to make a realistic assessment of the scientific and technical aspects of the Manhattan Project. There is an important difference in the nature of the intelligence in the two cases. In 1941, what was vital was strategic intelligence, that is, intelligence needed for the formulation of the Soviet Union’s policy and strategy as well as its war plans. In defining its own strategy, each state has to take account of the strategy of other states, and each state has to take account of the effect its own actions may have on the actions of a potential enemy. The most important questions for Soviet intelligence in 1941 had to do with Hitler’s intentions and the Wehrmacht’s plans and preparations for war. Stalin had not only to assess Hitler’s intentions but also to judge what impact his own actions might have on German actions. He feared that Hitler could be provoked into war by Soviet actions, and that fear helped to make him act cautiously rather than boldly. In the atomic case, it was not intentions that Kurchatov evaluated but scientific and technical information about which it was possible to make more or less conclusive judgments. It was this kind of intelligence that was so impressive in the atomic case. As far as can be judged from the existing sources, Soviet intelligence did not obtain much information about the decision to use the bomb in Japan or more generally about American or British thinking about the strategic significance of the bomb.142 It was the destruction of Hiroshima that brought home to Stalin how important the bomb was and galvanized him into action. These two cases suggest that, difficult though the collection of intelligence is, the assessment of its significance is harder still. The greatest challenge is to gain a broad understanding of how a potential enemy sees the world and what he plans to do. The great mistake in 1941 was Stalin’s inability to understand Hitler’s plan for war against the Soviet Union, and in particular its ideological dimension. That failure was reflected in the set of assumptions he brought to the construal of the intelligence. In the atomic case, the scientific and technical intelligence was outstandingly successful, but nothing seems to have been found before Hiroshima to throw light on the impact the bomb would have— or that the American and British believed the bomb would have—on the war in the Pacific and on international relations after the war.

68

Barbarossa and the Bomb

The second question was: how did the Soviet Union use the intelligence in formulating policy? Again there is a contrast between the two cases. In 1941, Moscow received copious intelligence about Germany’s forces and about Hitler’s plan to attack the Soviet Union. It was clear that the threat from Germany was growing, and in that sense Stalin had strategic warning. But the intelligence reports did not all point to the same conclusion and that made the careful analysis and assessment of the intelligence all the more important. It was precisely here that Soviet intelligence was weak. Neither military intelligence nor foreign intelligence had strong analytical capacity. Stalin’s baleful rule was an even more important factor. He had created a system in which it took courage to offer independent or dissenting advice. His own preferences and prejudices imposed themselves heavily on the way in which intelligence was assessed. Nevertheless, it is apparent that many intelligence reports reached Stalin in the form in which they arrived in Moscow. Although some filtering evidently took place, it was not enough to prevent him from receiving a great deal of information about German preparations for war and many reports on German intentions. Nor did everyone merely agree with Stalin’s judgments. Timoshenko and Zhukov in particular were willing to press Stalin to bring the Red Army up to full readiness. Atomic intelligence was different, partly no doubt because the atomic bomb had relatively low priority before August 1945 and partly because the intelligence was scientific and technical rather than political or military. Kurchatov analyzed the material he was given and drew conclusions about what the Soviet Union needed to do if it was to have a successful atomic project. At first, to his great frustration, the issue of the atomic bomb was not taken as seriously as he thought it should be, and he tried to convince members of the Soviet leadership of its importance. When the atomic bomb became a top priority project, he was in a very strong position to explain to others the necessary scope of the project and to give detailed advice on what needed to be done. Intelligence did not play a central role in the formulation of policy vis-à-vis the United States, but it did play an important role in the atomic project itself. The third question was: what consequences did intelligence have for international relations? In both of these cases, intelligence was clearly important, but its importance was not intrinsic to the information itself. The importance of intelligence was defined by the way it was used—how it was assessed and how it interacted with other factors to shape policy. In these two cases,



Barbarossa and the Bomb

69

intelligence played a less important role than one might imagine if one looked only at the information contained in the intelligence reports. Of course it is essential to ask what intelligence the Soviet Union obtained and how it acquired it. But it is even more important to ask how that intelligence was handled in Moscow, and how it was interpreted. That, after all, is the point at which intelligence and policy meet. Stalin’s response to the intelligence he was receiving about German plans to attack the Soviet Union was of great importance, but it did not spring from the intelligence itself; it has to be explained in terms of the assumptions he brought to the intelligence and the way in which he weighed the consequences of different courses of action. Thus a crucial if obvious point is that one cannot look only at intelligence collection; one has to look also at how the intelligence is assessed and interpreted and how it fits into the broader policy-making process. How much difference would it have made if Stalin had heeded the advice of Timoshenko and Zhukov to give the order to bring the Red Army up to full readiness before he in fact did so? Zhukov did not think that the Red Army could have stopped the German invasion. The war would not have gone according to the Soviet plan, but a stronger response to the initial German attack might nevertheless have enabled the Soviet Union to avoid the worst disasters of 1941 and 1942, to stabilize its defensive lines more quickly, and to win the war at a lower cost in Soviet lives. If the Red Army had pushed into Germany sooner than it did, it might have occupied the whole of Germany, with far-reaching consequences for the postwar world. If the Soviet Union had not obtained intelligence about British and American work on the bomb, would that have made a difference? The American atomic monopoly would have lasted longer, but the Soviet Union would still have been able to build a bomb by 1950 or 1951: it was the fact of the bomb rather than the wartime intelligence that convinced Stalin to turn the atomic project into a crash program. What would the consequences have been if the Soviet Union had not yet possessed the bomb in 1950? Would Stalin still have approved Kim Il Sung’s plan to reunite Korea by force? Like all counterfactuals, this is open to debate, though the establishment of the People’s Republic of China was certainly a more important factor in Stalin’s decision. The fourth question was: what light do these two cases throw on Stalin as a political leader? In 1941, when he played a central role, intelligence was handled in a disjointed and disorganized way. There was no central clearinghouse and no center for the analysis of intelligence. The brutality and

70

Barbarossa and the Bomb

suspiciousness of Stalinist politics were very damaging in the assessment of intelligence and the formation of policy. The complex intelligence Moscow was receiving in 1940–1941 needed careful and cool analysis, but that was difficult in the political atmosphere Stalin had created. Stalin’s responsibility for the disaster of 1941 cannot be dealt with merely by pointing to inconsistencies in the intelligence reaching Moscow. The answer has to be sought in the assumptions he brought to the interpretation of the intelligence, and also in the despotic character of his rule, which greatly discouraged independent analysis and advice. The case of atomic intelligence was somewhat different, partly because this was scientific and technical, rather than strategic, intelligence, but also because during the war, Stalin altered his style of leadership, allocating particular areas of responsibility to other leaders, a practice he continued after the war. Before Hiroshima, the atomic project was a relatively low-priority affair and did not impinge on Soviet foreign or military policy; it had no direct strategic significance. Kurchatov’s assessment of the intelligence was very straightforward and highly competent, but his reports did not have implications for the policy the Soviet Union should conduct toward other states, except in the relatively narrow sphere of atomic intelligence. After Hiroshima, the atomic project became a top priority for the state. Beria was eager for success and he protected his key scientists against the “anti-cosmopolitan” (i.e., anti-Semitic) campaign of the late 1940s and early 1950s and also against those philosophers and physicists who wanted to put physics on a stronger Marxist-Leninist footing. The scientists involved in the atomic project were thus shielded to some degree from the broader Stalinist regime, though they were well aware that if they did not succeed in building the bomb, they might find themselves severely punished for their failure.143 Finally, to what extent should the historian of international relations in the twentieth century take account of intelligence? The answer is surely: to the maximum degree possible. There is a natural affinity between the craft of intelligence and the historian’s craft. The most difficult aspect of intelligence work is to grasp what the other side is thinking, to understand the intentions and the mentality behind the information that has been collected. In this the intelligence analyst and the historian share a common task, though the historian works under much less pressure of time, and the historian’s work very rarely has the kind of consequences that the collection and evaluation of intelligence can have. At the same time, the historian has to treat intelligence



Barbarossa and the Bomb

71

with care, not imputing to it greater importance than it deserves, and placing it in the context of other factors that shape international politics.

Notes 1. In a radio broadcast on June 22, 1941, the day of the German invasion, Churchill claimed: “All this was no surprise to me. In fact I gave clear and precise warnings to Stalin of what was coming.” The text of the broadcast is available at http://www.ibiblio.org/pha/timeline/410622dwp.html. Churchill greatly exaggerated the clarity and precision of the warnings he had given Stalin. See Gabriel Gorodetsky, Grand Delusion: Stalin and the German Invasion of Russia (New Haven, CT: Yale University Press, 1999) chap. 8. 2. Igor Gouzenko, This Was My Choice (London: Eyre & Spottiswoode, 1948); Amy Knight, How the Cold War Began: The Igor Gouzenko Affair and the Hunt for Soviet Spies (New York: Carroll & Graf, 2005). 3. These collections are sometimes intended by their editors to make particular points—the intelligence services, for example, are eager to demonstrate that the catastrophe of 1941 was not an “intelligence failure,” that they provided Stalin with everything he needed to make the right decision—but they do provide a whole new basis for examining the role of intelligence in Soviet policy. 4. Military intelligence was the responsibility of the Intelligence Directorate (RU: Razvedyvatel’noe upravlenie) of the General Staff, which became the Chief Intelligence Directorate (GRU: Glavnoe razvedyvatel’noe upravlenie) in February 1942. Foreign intelligence was the responsibility of the Fifth Department of the Chief Directorate of State Security (GUGB: Glavnoe upravlenie gosudarstvennoi bezopasnosti) of the People’s Commissariat of Internal Affairs (NKVD: Narodnyi kommissariat gosudarstvennoi bezopasnosti). When the GUGB was converted into the People’s Commissariat of State Security (NKGB: Narodnyi kommissariat gosudarstvennoi bezopasnosti) in February 1941, the Fifth Department of the GUGB became the First Directorate of the NKGB; in July 1941, the NKGB was subordinated once more to the NKVD, but it became a separate people’s commissariat again in April 1943. On the organization of foreign and military intelligence on the eve of the war, see David E. Murphy, What Stalin Really Knew: The Enigma of Barbarossa (New Haven, CT: Yale University Press, 2005), 91–96, 253–255; see also V. M. Lur’e and V. Ia. Kochik, GRU: dela i liudi (Saint Petersburg: Neva, 2002), 71–97. 5. Christopher Andrew and Oleg Gordievsky, KGB: The Inside Story (New York: HarperCollins, 1990), 236–238. 6. Mikhail Mel’tiukhov, Upushchennyi shans Stalina (Moscow: Veche, 2000), 295–296. 7. The Soviet Union had some signals intelligence capability in 1939–1941, but it does not appear to have provided intelligence about German plans and intentions, though it does seem to have yielded information about German troop movements.

72

Barbarossa and the Bomb

See Murphy, What Stalin Really Knew, 255; Mel’tiukhov, Upushchennyi shans, 296–297; Andrew and Gordievsky, KGB, 174–174, 178–182, 225–226. 8. Dmitri Volkogonov, Triumf i tragediia: politicheskii portret I. V. Stalina (Moscow: Novosti, 1989), bk. 2, pt. 1, 125. 9. Volkogonov, Triumf i tragediia, bk. 2, pt. 1, 125–148. 10. The Diary of Georgi Dimitrov, 1933–1949, ed. Ivo Banac (New Haven, CT: Yale University Press, 2003), 115. 11. Chris Bellamy, Absolute War: Soviet Russia in the Second World War (London: Pan Books, 2007), 118–119. 12. “Direktiva No. 21 Verkhovnogo Komandovaniia vooruzhennymi solami Germanii (‘Operatsiia Barbarossa’),” December 18, 1940, in 1941 god (Moscow: Mezhdunarodnyi fond “Demokratiia,” 1998), bk. 1, 452. 13. Jürgen Förster, “Operation Barbarossa as a War of Conquest and Annihilation,” in Germany and the Second World War, vol. 4: The Attack on the Soviet Union, ed. Militärgeschichtliches Forschungsamt, Potsdam, Germany (Oxford: Clarendon Press, 1998), 481. 14. John Erickson, The Road to Stalingrad (London: Weidenfeld & Nicolson, 1974), 1–9. 15. “O voenno-politicheskikh tseliakh Germanii,” January 1939, in Voennaia razvedka informiruet (Moscow: Mezhdunarodnyi fond “Demokratiia,” 2008), 25–26. 16. V. A. Anfilov, Doroga k tragedii sorok pervogo goda (Moscow: Akopov, 1997), 104. For somewhat higher numbers, see Mel’tiukhov, Upushchennyi shans, 446. 17. I. V. Stalin, “Otvet t-shchu Razinu,” in Robert H. McNeal, ed., I. V. Stalin, Works, vol. 3: 1945–1953 (Stanford, CA: Stanford University Press, 1967), 29–34; see also Matthew Gallagher, The Soviet History of World War II: Myths, Memories, and Realities (New York: Praeger, 1963). 18. Winston S. Churchill, The Second World War, vol. 3 (London: Cassell, 1950), 316, 319–323. See also Barton Whaley, Codeword Barbarossa (Cambridge, MA: MIT Press, 1973), 3. 19. N. S. Khrushchev, Doklad na zakrytom zasedanii XX s’ezda KPSS “O kul’te lichnosti I ego posledstviiakh” (Moscow: Gospolitizdat, 1959), 34. It was the accumulation of power in the hands of one man that led to the catastrophe of 1941, Khrushchev argued. In the changing political climate of the Brezhnev years, A. M. Nekrich’s damning analysis of Stalin’s failure to heed the warnings on the eve of the war, 1941 22 iiunia (Moscow: Nauka, 1965), proved controversial. Nekrich was expelled from the Party and his book was withdrawn from libraries and bookshops. 20. Roberta Wohlstetter, Pearl Harbor: Warning and Decision (Stanford, CA: Stanford University Press, 1962), 397. 21. Whaley, Codeword Barbarossa, 244. 22. S. P. Ivanov, Nachal’nyi period voiny (Moscow: Voenizdat, 1974). 23. Ibid., 191–196 on the German disinformation campaign; 206, 209–213 on assumptions and plans.



Barbarossa and the Bomb

73

24. It is worth noting, in this context, that in the early 1980s, when the Soviet leadership was particularly alarmed by the possibility of a surprise nuclear attack (VRIaN: vnezapnoe raketno-iadernoe napadenie), a group of intelligence officers was given the task of analyzing the signs of Hitler’s preparation for his attack on the Soviet Union. They drew on more than a thousand forgotten KGB files to see what the experience of 1941 could tell them about the problem of surprise attack in the nuclear age. See V. S. Motov, “Agent A/201—’Breitenbakh,’ in Trudy obshchestva izucheniia istorii otechestvennykh spetssluzhb, vol. 2 (Moscow: Kuchkovo polie, 2006), 295. The results of their analysis appear not to have been published. 25. Volkogonov, Triumf i tragediia, bk. 2, pt. 1, 125–148. 26. 1941 god, bk. 1, 10. 27. Vladimir Lota, Sekretnyi front General’nogo shtaba (Moscow: Molodaia Gvardiia, 2005), 68. 28. Gorodetsky, Grand Delusion, 321. 29. Geoffrey Roberts, Stalin’s Wars: From World War to Cold War, 1939–1953 (New Haven, CT: Yale University Press, 2006), 68. 30. Lota, Sekretnyi front, 60. 31. See the documents in Organy gosudarstvennoi bezopasnosti SSSR v velikoi otechestvennoi voine: Sbornik dokumentov. Tom I: Nakanune in 2 books (Moscow: “Kniga i biznes,” 1995). 32. Mel’tiukhov, Upushchennyi shans, 319–320. Lota offers a less critical account of the analytical work of the GRU; see his Sekretnyi front, 115–116. 33. Organy gosudarstvennoi bezopasnosti SSSR, vol. 1, bk. 1, viii. 34. Lota, Sekretnyi front, 36. 35. On Beria as an “information filter,” see Lota, Sekretnyi front, 46–47; see also Murphy, What Stalin Really Knew, 93. According to Lota, “Beria’s ‘information filter’ completely or to a significant degree deprived Stalin of documentary information about the growing threat of war.” 36. See, e.g., the note from Merkulov to Stalin, Molotov, Timoshenko, and Beria on April 24, 1941 on a report from “Starshina,” “Zapiska NKGB SSSR I. V. Stalinu, V. M. Molotovu, S. K. Timoshenko i L. P. Beriia s preprovozhdeniem agenturnogo soobshcheniia iz Berlina,” in 1941 god, bk. 2, 108–110. 37. Voennaia razvedka, 753, n. 26. 38. This list (Perechen’ donesenii o voennoi podgotovke Germanii protiv SSSR (ianvar’–iiun’ 1941 g.)) can be found in Lota, Sekretnyi front, 220–234. Twenty of the reports came from agent “X,” Gerhard Kegel, deputy chief of the economic department of the German embassy in Moscow. Kegel was a Communist who had begun to work for Soviet military intelligence in 1933 and had joined the Nazi Party at Moscow’s suggestion. For a discussion of the list, see Lota, Sekretnyi front, 10–15, and Voennaia razvedka, 7. 39. The report that mentioned Baku was from the military intelligence station chief in Belgrade, on March 9, 1941. Lota, Sekretnyi front, 222.

74

Barbarossa and the Bomb

40. On Stöbe and von Scheliha, see Lota, Sekretny front, 75–94, and Murphy, What Stalin Really Knew, 14–15, 64–65. 41. “Donesenie Meteora iz Berlina nachal’niku Razvedupravleniia Genshtaba Krasnoi Armii 29 dekabria 1940 g.,” in Lota, Sekretnyi front, 163. 42. “Donesenie ‘Meteora’ iz Berlina o voennykh prigotovleniiakh Germanii,” January 4, 1941. Voennaia razvedka, 527. 43. “Iz agenturnogo soobshcheniia ‘Al’ty’ iz Berlina o napravleniiakh i date Germanskogo nastupleniia na SSSR.” Voennaia razvedka, 536–537. 44. The reference to Pinsk is curious, because it is in the Pripet marshes and the main German strike was somewhat to the north, along the Brest-Minsk axis. 45. For the NKGB’s list (Kalendar’ soobshchenii agentov Berlinskoi rezidentury NKGB SSSR ‘Korsikantsa’ i ‘Starshiny’ o podgotovke Germanii k voine s SSSR za period s 6 sentiabria 1940 g. po 16 iiunia 1941 g.), see 1941 god, bk. 2, 400–407. 46. Mel’tiukhov, Upushchennyi shans, 309–310. 47. Ibid., 310. 48. Lota, Sekretnyi front, 110–129. 49. Organy gosudarstvennoi bezopasnosti SSSR, vol. 1, bk. 2, 11. 50. On the German deception campaign, see Whaley, Codeword Barbarossa, 170–177, 247–266; Murphy, What Stalin Really Knew, 172–184; “Ukazanie shtaba operativnogo rukovodstva OKV o meropriiatiiakh po dezinformatsii,” 15 February 1941, 1941 god, bk. 1, 661–664, and “Ukazanie shtaba operativnogo rukovodstva OKV o meropriiatiiakh po provedeniiu vtoroi fazy dezinformatsii,” 12 May 1941, ibid., bk. 2, 195–196. 51. Voennaia razvedka, 524. 52. Anfilov, Doroga k tragedii, 193. 53. The Diary of Georgi Dimitrov, 149. Khrushchev recalled Stalin saying in 1951: “I’m finished. I trust no one, not even myself.” N. S. Khrushchev, Khrushchev Remembers, vol. 1 (Harmondsworth, UK: Penguin Books, 1977), 328. 54. Sto sorok besed s Molotovym: iz dnevnika F, Chueva (Moscow: Terra, 1991), 32. Molotov begins this passage by saying: “When I was chairman of the Council of People’s Commissars.” That was a position he held from 1930 until May 5, 1941, but the context of the conversation makes it clear he is speaking about the eve of the war. 55. Ibid., 31–32. 56. Ibid., 32. 57. “Vystuplenie general’nogo sekretaria TsK VKP(b) I. V. Stalina pered vypusknikami voennykh akademii RKKA v Kremle,” May 5, 1941, in 1941 god, vol. 2, 160. 58. See esp. the OKW directive of February 15, 1941, “Ukazanie shtaba operativnogo rukovodstva OKV,” 661–664. 59. “Doklad nachal’nika razvedupravleniia Genshtaba Krasnoi Armii general-leitenanta Golikova v NKO SSSR, SNK SSSR i TsK VKP(b) “Vyskazyvaniia, [orgmeropriiatiia] i varianty boevykh deistvii Germanskoi armii protiv SSSR,” March 20, 1941, in 1941 god, bk. 1, 776. 60. Ibid., 780.



Barbarossa and the Bomb

75

61. Voennaia razvedka, 753, n. 15; Murphy, What Stalin Really Knew, 177. 62. Volkogonov, Triumf i tragediia, bk. 2, pt. 1, 132–135; M. V. Zakharov, General’nyi shtab v predvoennye gody (Moscow: Voenizdat, 1989), 212–224. 63. The Diary of Georgi Dimitrov, 166. 64. “Doklad nachal’nika,” 777. 65. “Zapis’ ‘Stepanova’ o nekotorykh soobshcheniiakh ‘starshiny,’” in 1941 god, bk. 2, 90. 66. A. O. Chubarian, Kanun tragedii: Stalin i mezhdunarodnyi krizis sentiabr’1939– iiun’ 1941 goda (Moscow: Nauka, 2008), 456. 67. Ibid., 457. 68. Lota, Sekretnyi front, 66. 69. This controversy was initiated by the publication in 1985 of an article by Viktor Suvorov (Rezun), a former Soviet military intelligence officer who defected to the West in the late 1970s. He put forward his thesis in the Journal of the Royal United Services Institute in June 1985 and subsequently in Icebreaker: Who Started the Second World War? (London: Hamish Hamilton, 1990). The thesis has been analyzed and rejected by, among others, Gorodetsky in Grand Delusion and John Erickson in “Barbarossa, June 1941: Who Attacked Whom?” History Today 51, no. 7 (July 2001): 11–17 70. Quoted in Volkogonov, Triumf i tragediia, bk. 2, pt. 2, 142. 71. Erickson, The Road to Stalingrad, chaps. 1 and 2. 72. Anfilov, Doroga k tragedii, 201–207. 73. “Iz neopublikovannykh vospominanii marshala sovetskogo soiuza G. K. Zhukova,” in 1941 god, bk. 2, 501–502. 74. Erickson, The Road to Stalingrad, 110. 75. Anfilov, Doroga k tragedii, 201–204, 223. 76. V. A. Zolotarev, ed., Istoriia voennoi strategii Rossii (Moscow: Kuchkovo polie, 2000), 252. 77. “Iz neopublikovannykh vospominanii,” 502; Voennaia razvedka, 527. 78. G. K. Zhukov, Vospominaniia i razmyshleniia, 10th ed. (Moscow: Novosti, 1990), 1: 338. 79.Ibid., 324. 80. Anfilov, Doroga k tragedii, 205. 81. Erickson, The Road to Stalingrad, 109–110. 82. For a discussion of Soviet military doctrine in relation to 1941, see Cynthia Roberts, “Planning for War: The Red Army and the Catastrophe of 1941,” Europe-Asia Studies 47, no. 8 (1995): 1293–1326. 83. Anfilov, Doroga k tragedii, 157–158. 84. Ibid., 158. 85. Roberts, “Planning for war,” 1308. 86. “Vystuplenie general’nogo sekretaria,” in 1941 god, bk. 2, 162. 87. “Zapiska narkoma oborony SSSR I nachal’nika Genshtaba Krasnoi Armii predsedateliu SNK SSSR I.V. Stalinu s soobrazheniiami po planu strategicheskogo razvertivaniia vooruzhennykh sil Sovetskogo Soiuza na sluchai voiny s Germaniei I

76

Barbarossa and the Bomb

ee soiuznikami,” in 1941 god, bk. 2, 215–220. For commentary see, inter alia, ibid., 296; Anfilov, Doroga k tragedii, 166, 171–172; Bellamy, Absolute War, 106–118; Mel’tiukhov, Upushchennyi shans Stalina, 370–414; Gorodetsky, Grand Delusion, 238–240. 88. G. S. Isserson, Novye formy bor’by (Moscow: Voennoe izdatel’stvo, 1940), 28, quoted by Anfilov, Doroga k tragedii, 74. Isserson noted that of course mobilization and concentration could not be hidden completely, and he made a very prescient comment: “To one degree or another the concentration [of forces] will become known. From the threat of war to the start of war, however, there is always one more step. It gives rise to doubt as to whether a real military action is being prepared or just a threat. And while one side remains in doubt, the other, having firmly decided on action, continues its concentration, until finally a huge armed force has been deployed on the border. After that, it remains only to give the signal and war will immediately be discharged on its full scale.” Isserson was arrested soon after completing his book and spent fifteen years in prisons and camps. Anfilov, Doroga k tragedii, 74–75. 89. Roberts, “Planning for War,” 1296. 90. “Ot sostavitelei,” in 1941 god, bk. 1, 10. 91. On this more general problem, see John Erickson, “Threat Identification and Strategic Appraisal by the Soviet Union, 1930–1941,” in Ernest R. May, ed., Knowing One’s Enemies: Intelligence Assessment Before the Two World Wars (Princeton, NJ: Princeton University Press, 1984), 375–423. 92. David Holloway, Stalin and the Bomb: The Soviet Union and Atomic Energy, 1939–1956 (New Haven, CT: Yale University Press, 1994), 49–71. 93. Margaret Gowing, Britain and Atomic Energy (London: Macmillan, 1964), 40–89 provides a good account of the Frisch-Peierls memorandum and the Maud Committee; the Report and appendices are reprinted on 394–436. 94. McGeorge Bundy, Danger and Survival: Choices About the Bomb in the First Fifty Years (New York: Random House, 1988), 88. 95. “Rasshifrovka soobshcheniia rukovoditelia Londonskoi rezidentury v Razvedupravlenie Genshtaba KA o rabotakh po sozdaniiu atomnoi bomby v Anglii,” 10 August 1941, in L .D. Riabev, ed., Atomnyi proekt SSSR, vol. 1: 1938–1945, pt. 2 (Moscow: izd. MFTI, 2002), 434–435. (This invaluable series will be referred to below as AP SSSR.) On Klaus Fuchs, see “The Interrogation of Klaus Fuchs, 1950,” in R. Gerald Hughes, Peter Jackson, and Len Scott, Exploring Intelligence Archives: Enquiries into the Secret State (London: Routledge, 2008), 123–139. This contains the text of Fuchs’s interviews with Michael Perrin, an overview by Michael Goodman and a comparison by me of Fuchs’s confession with declassified Soviet documents. 96. On Gorskii’s reports see the two memoranda of the First Department of the USSR NKVD: “Spravka 1–go Upravleniia NKVD SSSR o soderzhanii poluchennoi iz Londona agenturnoi informatsii o ‘soveshchanii Komiteta po uranu,’” no earlier than September 9, 1941, and no later than October 3, and “Spravka 1–go Upravleniia NKVD SSSR o soderzhanii doklada ‘Uranovogo komiteta,’ podgotovlennaia po poluchennoi iz Londona agenturnoi informatsii,” no earlier than October 3, 1941, and no later than October 10, in AP SSSR, vol. 1, pt.1 (Moscow: Nauka, Fizmatlit, 1998), 239–242. For



Barbarossa and the Bomb

77

the material that was actually sent, see AP SSSR, vol. 2, pt. 6, (Moscow-Sarov: Nauka, VNII, 2006), 706–763. Donald Maclean is identified as the source in AP SSSR, vol. 1, pt. 1, 240, but Yuri Modin identifies Cairncross as the source in his My Five Cambridge Friends: Burgess, Maclean, Philby, Blunt and Cairncross (New York: Farrar Straus Giroux, 1994), 109–110. Cairncross’s identity as the source was confirmed to me by V. B. Barkovskii in a conversation in October 1999. He said that as long as Cairncross was alive, the KGB could not reveal his role. Barkovskii worked in scientific-technical intelligence in the NKGB residency at the London embassy in 1941. Cairncross died in 1995. 97. Pavel Sudoplatov, who headed the department responsible for processing atomic intelligence after the war, claimed that Niels Bohr, Enrico Fermi, Robert Oppenheimer, and Leo Szilard had knowingly passed atomic secrets to the Soviet Union during and after World War II. See Pavel Sudoplatov and Anatolii Sudoplatov with Jerold L. and Leona P. Schechter, Special Tasks: The Memoirs of an Unwanted Witness—a Soviet Spymaster (Boston: Little, Brown, 1994). The charges received a great deal of publicity at the time, but the new evidence strengthens even further the judgment that the claim is not true. 98. Merkulov to Beria, February 28, 1945 (“Pis’mo NKGB L. P. Berii ‘o khode rabot po sozdaniiu atomnoi bomby’ za rubezhom”), AP SSSR, vol. 1, pt. 2, 235. 99. See the intelligence assessments by Kurchatov on March 16 and April 7, 1945, “Zakliuchenie I. V. Kurchatova na razvedmaterialy (razdel ‘atomnaia bomba’), postupivshiie iz 1–go Upravleniia NKGB SSSR” and “Zakliuchenie I. V. Kurchatova na razvedmaterialy ob effektivnosti ‘iadernogo vzryvchatogo veshchestva,’ metodakh vzryva i dr., postupivshie iz 1–go Upravleniia NKGB SSSR),” in AP SSSR, vol. 1, pt. 2, 245–246, 261–264. On Hall, see Joseph Albright and Marcia Kunstel, Bombshell: The Secret Story of America’s Unknown Atomic Spy Conspiracy (New York: Times Books, 1997). On Fuchs, see Holloway, “Commentary: An Analysis of Sir Michael Perrin’s Interviews,” 133–139. 100. Holloway, “Commentary,” 135–136. V. N. Merkulov to L. P. Beria, 18 October 1945, “U istokov sovetskogo atomnogo proekta,” Voprosy istorii estestvoznania i tekhniki, 1992, no. 3: 126–129. See also the other detailed reports on the bomb, apparently not supplied by Fuchs, that were presented to the Technical Council of the Special Committee on the Atomic Bomb at the end of 1945 and in early 1946. AP SSSR, vol. 2, pt. 6, 763–800. 101. A report prepared by G. B. Ovakimian, a colonel in the NKGB, on the basis of foreign intelligence, noted on July 29, 1943, that there were no reports pointing to German success in the work on uranium and that the common view among British scientists was that the Germans were having great difficulties with both scientific resources and equipment. “Spravka G. B. Ovakimiana ‘O rabotakh po novomu istochniku energii—uranu’, podgotovlennaia na osnove razvedmaterialov, poluchennykh NKVD i NKGB SSSR za 1941–iiul’ 1943gg.,” in AP SSSR, vol. 1, pt. 2, 465. 102. The British interned the leading German physicists at Farm Hall, a country house near Cambridge, and held them incommunicado from June to December 1945.

78

Barbarossa and the Bomb

See Operation Epsilon: The Farm Hall Transcripts (Berkeley: University of California Press, 1993). 103. Holloway, Stalin and the Bomb, 108–112. 104. Molotov to Stalin September 27, 1942, “Zapiska zamestitelia predesedatelia GKO V. M. Molotova I. V. Stalinu o proektakh rasporiazhenii po vozobnovleniiu rabot v oblasti ispol’zovaniia atomnoi energii,” and “Rasporiazhenie GKO No. 2352ss ‘Ob organizatsii rabot po uranu,” September 28, 1942, in AP SSSR, vol. 1, pt. 1, 268–271. 105. Holloway, Stalin and the Bomb, 74–79, 84–86. 106. Beria to Stalin, October 6, 1942, and a memorandum by L. Fitin, the chief of Foreign Intelligence, “Spravka 1–go Upravleniia NKVD SSSR po materialu ‘Ispol’zovanie urana kak istochnik energii i kak vzryvchatogo veshchestva,” in AP SSSR, vol. 1, pt. 1, 271–274. A draft of this memorandum was ready on March 6 but not sent to Stalin at the time (AP SSSR, vol. 1, pt. 1, 244–245). When I wrote in Stalin and the Bomb, 84, that the memorandum had been sent on March 6, 1942, I did not know that that was a draft. Beria could have told Stalin about the intelligence before October 6, but after such a conversation, he would surely have sent the memorandum, which had already been drafted, to Stalin. 107. “Rasporiazhenie GKO No. GOKO-2872ss o dopolnitel’nykh meropriiatiiakh v organizatsii rabot po uranu,” in AP SSSR, vol. 1, pt. 1, 306–307. 108. Kurchatov’s assessment of these materials, including the Maud Report (“Dokladnaia zapiska I. V. Kurchatova V. M. Molotovu s analizom razvedmaterialov i predlozheniiami ob organizatsii rabot po sozdaniiu atomnogo oruzhiia v SSSR”) can be found in AP SSSR, vol. 1, pt. 1, 276—280. 109. Merkulov to M. G. Pervukhin, July 16, 1943 (“Pis’mo NKGB SSSR M. G. Pervukhinu o napravlenii razvedmaterialov”), in AP SSSR, vol. 1, pt. 1, 363. 110. “Zapiska zaveduiushchego Laboratoriei No. 2 I. V. Kurchatova zamestiteliu predesedatelia SNK SSSR M. G. Pervukhinu s analizom soderzhaniia razvedmaterialov i predlozheniiami k programme rabot,” March 7, 1943, in AP SSSR, vol. 1, pt. 1, 314–320. 111. “U istokov sovetskogo atomnogo proekta: rol’ razvedki,” Vestnik istorii estestvoznaniia i tekhniki, 1992, no. 105. 112. Christopher Andrew and Vasili Mitrokhin, The Sword and the Shield: The Mitrokhin Archive and the Secret History of the KGB (New York: Basic Books, 1999), 118–119. 113. See, e.g., his memorandum to Molotov on November 27, 1942 (“Dokladnaia zapiska”), in AP SSSR, vol. 1, pt. 1, 279. See also the memorandum from M. G. Pervukhin to Stalin on May 19, 1944 (“Zapiska M. G. Pervukhina I. V. Stalinu ‘O probleme urana’”), accompanying a detailed report by Kurchatov on the state of research, in AP SSSR, vol. 1, pt. 2, 73–74. 114. Kurchatov to Beria, September 29, 1944 (“Zapiska I. V. Kurchatova L. P. Berii”), in AP SSSR, vol. 1, pt. 2, 127. It is not clear whether the letter was in fact sent.



Barbarossa and the Bomb

79

115. L. D. Riabev, L. I. Kudinova, and N. S. Rabotnov, “K istorii sovetskogo atomnogo proekta (1938–1945),” in Nauka i obshchestvo: Istoriia sovetskogo atomnogo proekta (Moscow: Izdat, 1997), 35–36. 116. Ibid., 36. 117. See Vladimir Lota, GRU i atomnaia bomba (Moscow: Olma-Press, 2002). For extensive discussion of materials from the KGB archives, see John Earl Hynes, Harvey Klehr, and Alexander Vassiliev, Spies: The Rise and Fall of the KGB in America (New Haven, CT: Yale University Press, 2009) 33–143. 118. “Postanovlenie GOKO No. 9887ss/op ‘O Spetsial’nom komitete pri GOKO,’” in AP SSSR, vol. 1, pt. 2,11–14. 119. “Tezisy soobshcheniia I. V. Kurchatova na pervom zasedanii Spetsial’nogo komiteta o sostoianii rabot po uranu v SSSR,” in AP SSSR, vol. 2, pt. 1, 612–613. 120. The figure of 68,800 comes from “Spravka o predpiiatiiakh Pervogo glavnogo upravleniia, ikh moshchnostiakh i chislennosti rabotaiushchikh,” October 1, 1951, in AP SSSR, vol. 2, pt. 5 (Moscow-Sarov: Nauka, VNIIEF, 2005), 696; the figure of 259,000 comes from Appendix no. 1 to “Spravka o kapitolovlozheniiakh v atomnuiu promyshlennost’,” ibid., 645–646; the figures for prospecting, mining, and processing come from “O sostoianiiu rabot po razvitiiu atomnoi promyshlennosti,” 16 November 1951, ibid., 719. 121. B. L. Vannikov, “U istokov sozdaniia sovetskogo atomnogo oruzhiia,” in B. L. Vannikov: memuary, vospominaniia, stat’i (Moscow: TsNIIatominform, 1997), 90. 122. Henry DeWolf Smyth, Atomic Energy for Military Purposes: The Official Report on the Development of the Atomic Bomb Under the Auspices of the United States Government, 1940–1945 (1945; Stanford, CA: Stanford University Press, 1989). 123. For a discussion of the role of the Smyth Report in the Soviet project, see Michael D. Gordin, Red Cloud at Dawn: Truman, Stalin, and the End of the Atomic Monopoly (New York: Farrar, Straus & Giroux, 2009), 93–116. 124. Gowing, Britain and Atomic Energy, 267–268; Bundy, Danger and Survival, 107. 125. Arkadii Kruglov, The History of the Soviet Atomic Industry (London: Taylor & Francis, 2002), 51–62, 144–147. 126. Anthony Beevor, The Fall of Berlin 1945 (New York: Penguin Books, 2002), 138–139. Beevor claims that Stalin summoned Beria and leading nuclear scientists to his dacha in March 1942 to berate them for having neglected nuclear research, which then expanded “dramatically” over the next three years. There is no good evidence that such a meeting took place, and the Soviet project had relatively low priority until August 1945. 127. Tsuyoshi Hasegawa, Racing the Enemy: Stalin, Truman, and the Surrender of Japan (Cambridge, MA: Harvard University Press, 2005), 44. 128. Ibid., 155, 331–332; see also David Holloway, “Jockeying for Position in the Postwar World: Soviet Entry into the War with Japan in August 1945,” in Tsuyoshi Hasegawa, ed., The End of the Pacific War: Reappraisals (Stanford, CA: Stanford University Press, 2007), 145–188, where I discuss at greater length the argument set out here.

80

Barbarossa and the Bomb

129. Merkulov to Beria, July 10, 1945 (“Pis’mo NKGB L.P. Berii o podgotovke ispytaniia atomnoi bomby v SShA”), in AP SSSR, vol 1, pt. 2, 335–336. 130. Indirect confirmation is provided by the fact that the original of this memorandum was not found in Beria’s papers. No search was done in Stalin’s papers. AP SSSR, vol. 1, pt. 2, 335–336, n. 1. 131. The text of the accord can be found in U.S. Department of State, Foreign Relations of the United States: The Conferences at Malta and Yalta, 1945 (Washington, DC:GPO, 1955), 984. 132. S. M. Shtemenko, General’nyi shtab v gody voiny, vol. 1 (Moscow: Voenizdat, 1981), 416. 133. A. M. Vasilevskii, “Final,” Voenno-istoricheskii zhurnal, 1967, no. 6: 85; see also A. M. Vasilevskii, Delo vsei zhizni (Moscow: Politizdat, 1974), 513. 134. U.S. Department of State, Foreign Relations of the United States: The Conference of Berlin (Potsdam), 1945, vol. 2 (Washington, DC:GPO, 1945), 45. 135. Harry S. Truman, Memoirs, vol. 1: 1945, Year of Decisions (New York: Signet Books, 1965), 458; for Stalin’s interpreter’s statement, see V. G. Trukhanovskii, Angliiskoe iadernoe oruzhie (Moscow: Mezhdunarodnye otnosheniia, 1985), 23. 136. Sto sorok besed, 81; G. K. Zhukov, Vospominaniia i razmyshleniia, 10th ed., vol. 3, (Moscow: Novosti, 1990), 334; A. A. Gromyko, Pamiatnoe, 2nd ed., 1: 276. 137. Shtemenko, General’nyi shtab, 1: 425. 138. A. M. Vasilevskii, “Final,” Voenno-istoricheskii zhurnal, 1967, no. 6: 86. 139. The text of Vasilevskii’s report is in V. A. Zolotarev, ed., Sovetsko-iaponskaia voina 1945 goda: istoriia voenno-politicheskogo protivoborstva dvukh derzhav v 30–40e gody, published in the series Russkii Arkhiv: Velikaia Otechestvennaia as volumes 7 (1) and 7 (2) (Moscow: Terra, 1997 and 2000), 7 (1): 337–338. 140.For the Stavka directive of August 7, see ibid., 340–341. 141. George Kennan, cable to Washington, DC, August 8, 1945, Library of Congress, W. A. Harriman Papers, box 181. 142. Soviet military intelligence did learn about the Quebec Agreement of August 1943 reestablishing nuclear cooperation between the United States and Britain. V. I. Lota, GRU. Ispytanie voinoi (Moscow: Kuchkovo polie, 2010), 382. 143. Holloway, Stalin and the Bomb, 206–213.

3

Seeking a Scapegoat Intelligence and Grand Strategy in France, 1919–1940 Stephen A. Schuker

I Colonel Maurice Gauché, chief of French Army intelligence from 1935 to 1940, addressed the Centre des hautes études militaires as the war clouds gathered in 1939. “The study of past campaigns shows that a well-informed commander is rarely an unhappy commander,” said Gauché, “provided of course that the two opposing forces enjoy comparable resources.” In retrospect, Gauché wished he had added a qualification: “When the imbalance of power between the two belligerents is too great, intelligence becomes useless owing to the absence of sufficient and appropriate means to exploit it.” That, Gauché contends, describes precisely what happened in 1940.1 Gauché’s colleagues and subordinates within the French intelligence community at the end of the Third Republic vigorously endorse both his approach and his conclusions. Captain Paul Paillole, deputy chief of counterintelligence, Major Henri Navarre, deputy chief of foreign intelligence gathering (Service de renseignements; SR), and Major Gustave Bertrand, the leading light in signals intelligence, enlarge upon the manifold obstacles to intelligence collection and dissemination within a military culture that tended not to value their work highly. Yet they too emphasize accomplishments over many years and believe that the chief responsibility for France’s collapse in May–June 1940 lay elsewhere.2 The working-level intelligence unit, under the leadership of the much-admired Colonel Louis Rivet from 1935 onward, claimed many solid accomplishments, notwithstanding its meager budget,

81

82

Seeking a Scapegoat

shortage of personnel, and limited access to military and civilian decisionmakers, all emphasized by its relegation to symbolically shabby headquarters on the avenue de Tourville, outside the War Ministry itself.3 Moreover, Colonel Ulrich Liss, head of the Wehrmacht’s Fremde Heere West from 1937 to 1943, evinced high regard for the work of his French colleagues under the arduous circumstances prevailing in 1939–1940.4 France’s military intelligence agency, the Deuxième Bureau de l’État-major général (Second Bureau of the General Staff), grossly exaggerated the number of Wehrmacht infantry and tank units in September 1939, Liss concedes in a postmortem analysis, partly because it confused paramilitary forces with fully trained divisions. Yet every intelligence agency likes to build a “security factor” into its estimates.5 French human agents, Liss confirms, carefully followed the transfer of German forces from east to west after the Polish campaign. They monitored the probable order of battle from the Netherlands border in the north to Baden in the south. By early May, the 2ème Bureau guessed total Wehrmacht strength at 137 divisions, just one more than the actual number; it correctly ascertained how many belonged to Army Groups A, B, and C, as well as to the reserve. It could not pinpoint the exact location of the ten Panzer divisions.6 The Germans had made it hard for the French by minimizing radio communication and training the Panzers well back from the front so that neither prisoners nor planning papers could fall into hostile hands. Still, if Gauché rated as something of an alarmist, he certainly got the trajectory of the German buildup right. As early as June 1938, he had predicted that the Wehrmacht could mobilize 300 divisions by the end of 1942, at a time when his phlegmatic Whitehall counterparts still imagined that finance and raw-material constraints would limit the adversary to half of that.7 In 1940, the Germans could have launched an invasion through any of eight routes dispersed along a 600-mile front.8 With a far smaller population, the French command had to make hard choices about distribution of its troops. The French commander in chief, General Maurice Gamelin, nurtured a true obsession about mounting a defense in Belgium, where an encounter would take place at prepared positions along a line half the length of the one in northern France.9 Nonetheless, several additional factors combined to shape his thinking. The 1914–1918 devastation of France’s ten northern departments, where two-thirds of the country’s heavy industry lay, rendered it psychologically inconceivable for anyone in a representative democracy to contemplate a repetition.10 What’s more, the French government saw its chief support in the



Seeking a Scapegoat

83

British alliance, and British security interests focused narrowly on the Low Countries, the likely jumping-off point for an air or sea attack on the home islands. The 2ème Bureau had no input on the making of grand strategy; it functioned perforce within a framework crafted at a higher level. The French had recruited one important agent, Hans-Thilo Schmidt, in the Cipher Office of the German General Staff. And they received parsimonious disguised warnings about the projected date for the German attack and its repeated postponement from the Dutch. Yet the Netherlands government, artlessly hoping that the country’s usefulness as a neutral would spare it from invasion as in World War I, chose to disbelieve what Colonel Hans Oster, deputy chief of the Abwehr, told the Dutch military attaché in Berlin. Hence it never revealed its source.11 Nor did Belgium, stubbornly guarding its neutrality, provide wholehearted intelligence cooperation. When a German courier plane carrying invasion plans made a forced landing in Belgian Limbourg in January 1940, the king’s adviser, General Raoul van Overstraeten, turned over, not the raw documents, but merely a two-page extract to Gamelin’s representative. The anti-French van Overstraeten sought to instrumentalize the Mechelen incident as part of the ongoing Belgian campaign to extract maximum promises from the Western allies while evading commitments of their own.12 The 2ème Bureau could hardly read Hitler’s mind or get authentic information about his disputes with his generals during the following weeks. It had to reason inductively from agent reports about the order of battle. And it faced ongoing pressure from the Armée de l’Air to minimize observation flights for fear of losing good pilots condemned to fly rattletrap reconnaissance planes.13 In any event, we now know that the Wehrmacht did not adopt the socalled sickle-cut strategy immediately after compromise of its earlier plans. Nor did it begin with a programmatic commitment to blitzkrieg. Karl-Heinz Frieser shows that Generals Franz Halder, Erich von Manstein, and others fought long and hard about how exactly to fashion their multi-pronged attack. The decision to place the center of gravity in the Ardennes emerged in late February through a process of evolution from three intermediate plans. Subordinate commanders continued to make operational adjustments thereafter.14 No dramatic single decision took place for the 2ème Bureau to ferret out. Good intelligence work consists of sifting through a massive volume of information, disinformation, rumor, and propaganda and trying to separate the wheat from the chaff. As Colonel Liss found out when he later had to

84

Seeking a Scapegoat

brief Hitler, overreporting keeps decision-makers in an unproductive state of alarm. Normally, the intelligence staff cannot conclusively tell where an attack will center until air reconnaissance, prisoner interrogation, captured documents, and signals intelligence fill out the picture. In May 1940, French reconnaissance floundered because the Luftwaffe achieved early and complete mastery of the air. The Panzer corps advanced so quickly that the defenders took few prisoners; no significant documents about the German drive westward fell into 2ème Bureau hands until the night of 16/17 May; and a blackout of Enigma intercepts after 1 May (on which more below) sapped the quality of signals intelligence. Under the circumstances, Liss concludes, Gauché’s 2ème Bureau and the SR had done all they humanly could. They had “nothing to fear from the judgment of history.”15 II Scholars who have reviewed the matter since documentary material became available tend not to endorse that favorable evaluation. Martin Alexander, the leading defender of General Gamelin’s reputation as commander in chief, criticizes the interwar intelligence establishment as atomized, uncoordinated, and bereft of an Office of Net Assessment that could appraise disparate strands of raw intelligence accurately and bring the result to bear on policy decisions. Ministers and the top army brass paid little heed to intelligence before 1938, Alexander concedes. The fault, however, lay not merely with the chronic disorganization of French government, but also in a bureaucratic disconnect between the secretariat of the Conseil supérieur de la défense nationale, which produced studies that senior decision-makers were (theoretically) supposed to read, and intelligence gatherers and processors ranked lower on the totem pole.16 Douglas Porch expresses irritation with the intelligence insiders’ retrospective pretensions to clairvoyance and their attribution of blame to policy-makers for ignoring their predictions.17 Olivier Forcade adds that the officers who compiled the daily intelligence brief both at GHQ and at Northeast Front headquarters under General Joseph Georges threw in everything under formalized rubrics. Rather like stockbrokers giving advice in a volatile market, they preferred to avoid definite predictions lest future developments prove them wrong. More often than not, that left it to Operations (the 3ème Bureau) to sort out what to forward up the chain of command.18



Seeking a Scapegoat

85

In a notably balanced account, Peter Jackson surveys the whole corpus of evidentiary material that became available before the return of Secret Service archives from Russia.19 He credits French intelligence agencies with numerous tactical successes. These include warning of Hitler’s major aggressive moves in advance and charting the Wehrmacht’s evolving order of battle. But Jackson deplores the 2ème Bureau’s willingness to manipulate conclusions and accept worst-case scenarios to meet the shifting political requirements of the High Command. He likewise regrets its inability to establish credibility among civilian policy-makers, particularly at the Quai d’Orsay. Jackson also pinpoints a number of “egregious miscalculations.”20 Influenced by the stereotype of Teutonic efficiency and anxious to help their military chiefs wake up the somnolent civilians, the 2ème Bureau staff overestimated the productive capacity of the German armament industries under Marshal Göring’s Four-Year Plan. This led them to overstate the number of divisions that Hitler could mobilize during the Munich crisis (although given the pathetic weakness of the French air force that hardly made a significant difference). Later on, however, they exaggerated the strain that mobilization would impose upon the German economy and thereby indulged their masters’ fantasy that the superior resources of the French and British empires would prove decisive in the end.21 Among recent analysts, Ernest R. May mounts the most devastating critique of French intelligence. His powerfully written indictment has gone far to shift the contemporary focus of debate over France’s fall away from strategy and policy. May ascribes the country’s defeat in May–June 1940 almost entirely to an intelligence failure.22 He doesn’t dispute the weaknesses of French military doctrine, training, or higher command. Yet he deems those second-order phenomena. The air forces on the opposing sides, he contends (against the weight of received wisdom), were “closely matched.” And the French and British forces at the Gembloux Gap in Belgium had “more and better tanks and a huge advantage in artillery and munitions” over the Wehrmacht. May also casts aside the issue of national cohesion: France purportedly stood in no greater danger of moral collapse than Germany. By contrast, May asserts, Allied intelligence services “performed abominably” in every way. They missed the mark in overestimating the numbers and quality of troops, tanks, and aircraft on the opposing side. Not only did they fail to predict the fast-moving German offensive through the Ardennes; they compounded the error by not recognizing for several days where the

86

Seeking a Scapegoat

enemy center of gravity lay. Thus, the attackers achieved greater surprise at Sedan than at Pearl Harbor or in Operation Barbarossa, where the defending forces at least acknowledged their mistake promptly. May imagines a different outcome if French reconnaissance had noticed the “horrendous traffic jams” in the Ardennes and sent in “squadrons of bombers.” In any event, France’s best mechanized formations as well as the whole British Expeditionary Force found themselves stranded in Belgium while the Panzer forces under Heinz Guderian and Erwin Rommel, protected by infantry-support dive bombers, cut through French reserve formations like butter on their drive to the sea. A sometime adviser to U.S. intelligence agencies, May wrote his broadside partly for heuristic purposes. He presses home the point that General Kurt von Tippelskirch, head of the German General Staff intelligence directorate, enjoyed respect as a qualified troop commander. In contrast with French Army practice, German operational planners worked in fruitful coordination with their intelligence counterparts.23 Owing to decentralization, similar coordination of intelligence and operations remained scarcely imaginable in France. Between September 1939 and January 1940, General Gamelin dispersed his intelligence support staff into half a dozen locations. A corporal’s guard remained with him at the Château de Vincennes; others, including the German specialists under Major Paul Baril, attended General Georges at La Ferté sous Jouarre, the Northeast Front headquarters forty miles east of Paris. When the war began, the 2ème Bureau took the title of the 5ème Bureau. Its administrative and evaluation sections fell under the control of General Louis Colson, army chief of staff, who took up residence at Command Post Victor, twenty-five miles away in another direction. Most of the cryptographers, including the Polish Enigma decoders, set up shop not far from P.C. Victor, although a residue remained in Paris. General Aimé Doumenc, major general of the armies, whose logistic and administrative assignments overlapped with Colson’s, organized his own intelligence staff in yet another Paris suburb starting in January 1940. Daladier kept a sixth group by his side at the War Ministry. Lacking secure telephones, the various commands communicated by motorcycle dispatch rider, and Gamelin and Georges, who cordially loathed each other, communicated personally as little as possible.24 These confused arrangements serve as a metaphor for the larger disorganization and personal feuds that rendered both the military and civilian establishments dysfunctional. In December 1939, the loyal republican admiral François



Seeking a Scapegoat

87

Darlan, who owed his position as naval chief to the Popular Front, wrote that “the country has the impression of being out of control.” On a lightning visit to Paris a few weeks later, Marshal Pétain described the tangle of responsibilities within the High Command as “complete anarchy.”25 May’s thesis deserves attentive scrutiny. How accurately does he assess the comparative strength of the contending parties? A later part of this essay will recapitulate the position as it appeared to eyewitnesses and senior officers who later trawled the archives. Suffice it to say at this point that experts considered the northern Ardennes eminently penetrable; the team representing the Germans got through all obstacles in prewar exercises. No level playing field existed north of the Maginot Line. The Luftwaffe controlled the skies from beginning to end. If one counts aircraft that could get off the ground rather than those in the paper inventory, the Luftwaffe outnumbered the Armée de l’Air in the relevant sector seven to one in bombers and two to one in fighters. French reconnaissance planes, though outnumbered only four to three, generally lacked the speed to carry out daylight observation. The French, to be sure, possessed a competitive number of tanks. Those tanks, however, were improperly distributed, had limited range, and could not refuel under aerial attack. A parliamentary committee reported in March 1940 that the 2nd and 9th armies around the Sedan sector lacked the requisite equipment and field fortifications to defend themselves, but the High Command did not react. Olivier Forcade, who has used the secret service archives repatriated from Moscow, acknowledges among a welter of other problems that an intelligence breakdown occurred. But he presents the issue as more complicated than it appears in May’s account. The 2ème Bureau gave a dozen warnings of imminent attack from November 1939 onward, including four in April 1940. When nothing happened in the field, it lost credibility. Gauché’s team could not localize all the Panzer divisions, but a radio intercept on 3 April suggested the presence of some near Trier, opposite Luxembourg. Major Baril of the Northeast Front 2ème Bureau shortly picked up at least two more agent reports of forthcoming action on a broad front including the Ardennes. Given the imprecision of these indications, however, Gauché suspected disinformation. An undercover agent in the Abwehr reported to Major Paul Paillole at the Section de renseignements (Intelligence Section, SR) on 12 April that the Germans were studying roads, bridges, topography, and forward defenses in the Sedan-Abbeville sector. This led to a debate at Northeast Front headquarters on how long it would take enemy armored

88

Seeking a Scapegoat

divisions to reach the Meuse River. Baril thought they could do it in two days; the head of Operations remained unconvinced. The Polish cryptographers under Major Bertrand, working with three Enigma machines, had managed to decode five thousand messages relating to France beginning in October 1939. But few initiates knew enough about the provenance of those messages to value the results appropriately. Gauché and Baril didn’t learn where the information originated until March 1940. Colonel Rivet of SR, surreptitiously bypassing Gauché, obtained six appointments to brief Gamelin or his adjutant between 19 March and 28 April 1940. As an old alumnus of Joffre’s World War I staff, Gamelin had not previously set much store by intelligence. Except when under duress, he preferred to read literature rather than to micromanage the military machine.26 Hence this degree of access for an intelligence officer marked a new precedent. The Germans changed their codes on 1 May, however, so signals intelligence dried up until 21 May—too late to have decisive effect.27 Forcade also explains why French intelligence could not bounce back after the initial German onslaught. First, the obsolescent reconnaissance aircraft of the Armée de l’Air could not function in the face of Luftwaffe fighter supremacy. Second, SR could not evacuate the intricate web of substations and agents that it had established—at The Hague, Liège, Brussels, and, most important, Lille—before the Wehrmacht overran those points. The French 2nd and 9th Armies had never developed significant intelligence capacity of their own. They had rather depended on the SR branch substation at Charleville-Mezières. Thus loss of the latter crippled the eyes and ears of the armies assigned to the most exposed salient. Although the SR continued to track the Panzers through radio intercepts, that effort faced formidable obstacles. In summary, Forcade is less censorious of intelligence gathering than of “the quasi-cultural and intellectual underestimation of military intelligence by the High Command.” Would better intelligence have turned the tide? The bulk of the evidence suggests not. Based on his idiosyncratic reading of the World War I experience, Gamelin calculated in September 1939 that French and German soldiers, if they met in commensurate numbers on a prepared front, would fight to a draw. A blockade and oil shortages might eventually produce a collapse of the Nazi economy, although this would take years, not months.28 Yet the hubristic notion that French soldiers could equal the Germans man for man ignored sociology and statistics. In Germany, the army acted as “the school of the



Seeking a Scapegoat

89

nation.” From the wars of unification onward, German soldiers proportionately outfought all opponents, even when numerically inferior and outclassed in equipment and logistics. German battle effectiveness derived from high morale, superior unit cohesion, a hierarchy of social values that attracted the cream of society to the officer corps, and a system of operational flexibility that encouraged initiative in the ranks.29 German militarism also led to tragic excess and atrocities, of course. But those who call the 1940 result a “damn near-run thing” neglect the unquantifiable element in fighting power.30 III Whatever weight one accords to May’s thesis, it has given a fillip to the study of French intelligence in its own terms. Admiral Pierre Lacoste and a distinguished roster of collaborators have sought to identify a particular national “culture” of intelligence. They find that culture rooted in the vicissitudes of history and the evolving nature of French society. The institutions and practices of intelligence long bore the stigmata of the inferiority feelings that pervaded French institutions following Prussia’s victory in 1870–1871. They also reflected the ideological divisions that separated Right and Left at the time of the Dreyfus affair, leaving a residue of civilian distrust of the military. Other considerations that made the consumers of intelligence wary from the Third to the early Fifth Republics include the rise and persistence of a Communist Party with divided loyalties; the mutual incomprehension among the followers of Vichy, Giraud, de Gaulle, and the internal resistance after the 1940 collapse; and the ever-present corporate rivalries among different ministries. Those factors have led to a widespread conviction that all intelligence is political, as well as to a proliferation of competing agencies working in isolation.31 The French tradition of intelligence collection goes back to the 1620s, when Cardinal Richelieu established a cabinet noir to intercept and decipher the correspondence of King Louis XIII’s noble enemies. During the eighteenth century, those services, lodged alternately in the king’s household or in the Foreign or War ministries, comprised both intelligence collection and counterintelligence. Louis XV supervised a secret du roi that carried out personal diplomacy behind the Foreign Ministry’s back. The notion of a double-tracked foreign policy thus enjoys a distinguished paternity in France, where the civil liberties tradition never struck roots as deep as in Anglo-American societies.

90

Seeking a Scapegoat

Police surveillance of the citizenry became a pervasive feature of French life under every regime from the nineteenth to the twenty-first centuries.32 By the early twentieth century, the Quai d’Orsay, the Sûreté générale attached to the Interior Ministry, the Paris police, the PTT (communications services), and the Army and Navy had each developed its own intelligence operation. Those agencies intercepted letters, cables, and telephone messages as new technologies arose. They shadowed foreigners, radicals, and other suspects as a matter of course.33 Military and naval attachés were stationed in the leading embassies abroad to collect open-source information.34 The embassies used secret funds to hire informants, some authentic but others purveyors of idle gossip, on the side. As the rotary press made newspapers affordable for the working classes, the Quai d’Orsay disbursed secret funds to the French and foreign press in order to influence public opinion. Foreign embassies in France (except for the straight-laced Americans) also routinely paid off the press. Since journalists accepted bribes from all comers, even the most discerning observer could not reliably tell who had paid whom to print what.35 The spread of international telegraphy led to a race between code designers and code breakers in the late nineteenth century, analogous to the contemporaneous strivings to develop thicker armored plate and armor-piercing shells.36 The reputation of the existing military intelligence unit suffered badly after its falsifications against Captain Dreyfus came to light. The Army thereafter lost control of domestic counterintelligence to the Sûreté générale, though it clawed back oversight over espionage on foreign soil a decade later. It gradually expanded other functions and in 1903 formally created a Cipher Bureau. The prewar decade witnessed great technical advances in wireless telegraphy and the installation of transmission links between the allies in Paris, London, and Saint Petersburg. All the same, rivalry between the military, the Foreign Ministry, and the PTT for oversight of telephones and other cryptographic functions persisted. Not until January 1914 did the military coding section and the cryptanalysis bureau agree to merge their efforts under the direct aegis of the war minister. When hostilities broke out, Paris could boast military intelligence capabilities more sophisticated than those of Berlin, but it had nothing to rival “Room 40” in London or the comparable facilities in Saint Petersburg. Most generals continued to place greater reliance on human agents than on signals intelligence.37 The Quai d’Orsay, meanwhile, made its own strides in signals intelligence. As early as the 1890s, it broke the diplomatic codes of England, Germany,



Seeking a Scapegoat

91

Turkey, Italy, and Spain. Yet the aristocrats who ran the Foreign Ministry declined to cooperate with functionaries at the less prestigious ministries in the game, and the cabinet, which lasted a mere nine months on average in the prewar Third Republic, traditionally exercised little oversight in matters of war or diplomacy. Much duplication of effort resulted. The Quai d’Orsay failed to inform the military about the secret 1902 accord with Italy until 1909, and the War Ministry told the diplomats very little about ongoing staff talks with the British in 1906.38 The French had learned the general outlines of the Schlieffen Plan in 1905 and had specifically warned the Brussels government that “neutrality” would not save their land from invasion.39 The 2ème Bureau recruited a substantial network of agents within the Reich and even penetrated the German Army. It found it easy to cultivate agents among bilingual nationals who had grown up in Alsace-Lorraine. It also set up counterintelligence units in the military regions and staffed border surveillance posts to monitor German mobilization plans. Despite all those measures, the French Army remained surprisingly unprepared when the Deutsches Heer thrust through Belgium in August 1914. The difficulty lay not in intelligence collection, but in intraservice rivalry and the refusal of the French commander in chief, General Joseph Joffre, to pay due attention to SR reports. The leading generals had squabbled so incessantly over strategy and personnel assignments during the two Moroccan crises that War Minister Adolphe Messimy felt obliged in 1911 to appoint a chief of staff who could crack heads. A solid logistics and supply expert, Joffre possessed the administrative skills to restore harmony. But few of his fellow generals considered him an imaginative thinker. The Army’s most brilliant strategist, who reasoned inductively from evidence of railroad building that the Germans might move through Belgium, found himself squeezed out. A true believer in the fashionable doctrine of the offensive, Joffre focused single-mindedly on his own Plan XVII for a two-pronged advance into Lorraine and the Luxembourg Ardennes. He waved away reports about Berlin’s battle plans that cast doubt upon his preconceived views. The 2ème Bureau obtained unimpeachable proof in April 1914 that the adversary would use reserves in the front line, bolstering its initial advance and rendering Plan XVII more risky. But that did not deter Joffre, who weakened the defenses on the northern frontier to pursue the holy grail of an offensive à l’outrance (all-out attack) in the east.40 In a different military culture, the failure of Plan XVII and the country’s narrow escape from disaster at the Battle of the Marne might have led

92

Seeking a Scapegoat

to instant respect for the intelligence function. Progress did take place, but slowly. Joffre himself remained a skeptic. In 1916, he ignored reports from the SR Belfort outpost about a German concentration before the Battle of the Somme. Military intelligence did not come into its own until Ferdinand Foch became commander in chief in 1918. By that time, the SR had an agent, Police Inspector Woegele, working within the German General Staff. And in a coup sometimes called “the radiogram of victory,” cryptanalysts provided advance notice that Erich Ludendorff planned to launch an offensive in June 1918. Additional intercepts over the next month allowed Foch to counterattack at Château Thierry before the Reichsheer had prepared its new position.41 A crime of passion by a former prime minister’s wife, followed by a spectacular murder trial, publicly revealed in July 1914 that the French routinely read diplomatic intercepts from three separate nations. Most foreign offices promptly changed their codes. During the first weeks of the world war, the French obtained no signals intelligence at all. Despite their limited numbers, however, French cryptanalysts achieved notable tactical results later on. In October 1914, they broke an important German military code. They steadily improved at reading code and radio traffic, interrogating prisoners, employing spies, and carrying out balloon, zeppelin, and fixed-wing aircraft reconnaissance.42 On balance, French signals intelligence outperformed its German counterpart during the war. The Quai d’Orsay files suggest that the ministry gained much high-level political intelligence through a team of gifted agents (including Emile Haguenin, André François-Poncet, and René Massigli) that it stationed in a press and propaganda bureau in Berne, Switzerland. IV While the maneuvers of rival secret services on opposite sides of the trenches went on behind closed doors, fantasies of widespread domestic spying raised anxiety among the French public throughout the war. Interior Minister Louis Malvy made a judgment call not to round up the thousands of radicals and syndicalists whose names appeared on “Carnet B” (the famous list of suspects) at the start of hostilities. Since the working class ignored the ideology of the Second International and rallied to the colors, this decision worked out well.43 In the postwar period, however, an abundant spy literature served to sustain the popular fear of a nefarious fifth column with its tentacles reaching everywhere.44



Seeking a Scapegoat

93

In early 1917, the prime minister, ignoring grumbling from the SR, strengthened the powers of the Sûreté over revolutionary propaganda and subversion. The Sûreté enlarged its human intelligence staff and its monitoring of telephones, telegraph, and mail. It ferreted out a certain number of small-bore spies, as well as exposing politicians such as Joseph Caillaux (and even ex-premier Aristide Briand) who had carried out negotiations behind the backs of their ministerial colleagues. All the same, one has the impression that the Sûreté wasted resources by continuing to trail politicians and other subjects of wartime interest right through the Depression. The Interior Ministry archives hold stacks of rotting folders attesting to that effort. The files reveal, for example, that Georges Mandel (a.k.a. Jéroboam Rothschild) patronized a newspaper kiosk, visited his mother, and took a walk in the Bois de Boulogne, and that Édouard Herriot declaimed at a human-rights rally that peace was France and France was peace. According to rumors bandied about at the Chambre des Députés bar, the first act of every interior minister was to burn his own file.45 To be sure, the Sûreté collected piles of scurrilous material, but the details reached public consciousness only when it served someone’s interest to broadcast them in the course of a political scandal.46 During the 1920s, the Interior Ministry sought to justify its “humint” budget by discovering a new domestic role. It found one by tracking the Soviet “menace.” Several hundred thousand White Russian émigrés had settled in Paris, and both the OGPU (later NKVD and KGB) and the Comintern centered their West European activities in the city. Major Guy Schlesser of SR calls the capital the “paradise of spies.”47 At first the Soviet Union employed enthusiasts from the Parti communiste français (PCF) for industrial spying, although it sent in OGPU and Comintern professionals as well. The Interior Ministry failed to suppress most of this activity. Jacques Duclos, who masterminded the PCF spying, remained exempt from prosecution thanks to his parliamentary immunity, and Soviet agents carried out kidnappings or assassinations of defectors and White Russian leaders without being brought to justice.48 A shadowy Office of Nationalities, another hothouse product of the war, also sought to keep tabs on foreigners. That agency likewise proved relatively ineffective.49 Outraged SR operatives believed that the Sûreté lost its focus on counterespionage because its agents had too many ancillary duties to fulfill. Until 1936, the 136 “special commissioners” reported to individual prefects around the country, who assigned them to all manner of police and political

94

Seeking a Scapegoat

work. The wide regional distribution of special commissioners made little sense. The German, Italian, and Soviet embassies and spy agencies naturally centered their efforts on Paris. The border regions remained underserved because 45 percent of the workers who built the fortifications came from outside the country. During the late 1930s, despite the tightening of border controls under Daladier, the influx of tens of thousands of refugees, a certain proportion of them Communists or Nazi operatives, strained the capacities of the Sûreté and the Paris police to the breaking point. Between August 1937 and February 1939, the SR brought order out of chaos through a succession of administrative arrangements with the Interior Ministry. A central intelligence bureau (Bureau de centralisation des renseignements) took shape. But the SR’s Major Schlesser recalls bitterly that he had embarked upon a labor of Sisyphus.50 The Quai d’Orsay vetoed any oversight of foreign embassies or scrutiny of propaganda. Many spies were caught only after the fact. A stenographer who sold the minutes of the Senate Foreign Affairs Commission to the Germans, for example, was only unmasked when the damage was done. Édouard Pfeiffer, who doubled as a Soviet agent and Prime Minister Édouard Daladier’s chef de cabinet (principal private secretary), remained undetected (despite his notorious friendship with fellow Homintern initiate Guy Burgess) until the Venona intercepts revealed his activities long after World War II.51 V To judge from surviving indications, the French Foreign Ministry and Army both failed to sustain the trajectory of progress in refining intelligence capability that they had attained in World War I. Coming on top of the usual internecine rivalries, the chronic postwar budget deficit inclined policy-makers to make cuts where they could. Premier Raymond Poincaré stopped the distribution of diplomatic decrypts to the armed forces in 1922, with patently negative results.52 The valuable reports from agent “Daniel,” for example, do not seem to have reached the military during the 1923 Ruhr occupation. One of the self-styled ministers in Hans Adam Dorten’s Rhineland separatist movement served as an informer for the Heimatdienst, but the French discovered this only years later when the former “minister” went into business with Baden’s counterintelligence chief.53 Paul Tirard, president of the Rhineland High Commission, who reported to the Quai, appears to have been



Seeking a Scapegoat

95

completely blindsided in early 1924 by the operation mounted by the Bavarian government to murder the separatists in the Palatinate.54 After years of suffering from the Quai d’Orsay’s refusal to share information on a regular basis, the 2ème Bureau pressed for creation of an interministerial intelligence committee. That institution, however, which met for brief periods in 1937–1938, did not lead to much. Open-source reports from the military, naval, and air attachés passed through the Foreign Ministry, and the diplomats digested them, but analyses generated by the 2ème Bureau did not reach the Quai on a regular basis and were often discounted when they did. The Quai even put up bureaucratic resistance to the stationing of SR representatives at consulates in the Reich until 1938. And the next year, Alexis Léger, secretary-general of the Foreign Ministry, proposed to establish his own secret service, though the project was overtaken by events.55 In contrast with the pre-1914 period, the Quai d’Orsay also neglected cipher security. Although the SR placed telephone taps on the German, Soviet, Italian, and British embassies in 1936, the Foreign Ministry failed to crack high-grade German diplomatic or military codes at any point before 1940.56 It is puzzling that signals intelligence did not keep up with advances abroad. Not only could the British read French diplomatic ciphers freely from 1920 onward, but the French had so little consciousness of this that their ambassador in London, the comte de St. Aulaire, cabled the details of his plot to dislodge the British foreign secretary in 1923, quite unaware that the decrypt would reach Lord Curzon’s red box the following morning.57 By the mid-1930s, Hermann Göring’s Forschungsamt was also reading French diplomatic traffic freely. By the time World War II broke out, code security had been breached at a minimum of fifteen French embassies.58 French military intelligence maintained better security than that, but like other parts of the military machine, it was hobbled by atomization and chronic budget constraints. The Navy and Air Force jealously guarded their own intelligence operations. The 2ème Bureau, which metamorphosed into the 5ème Bureau upon the outbreak of war, comprised eight sections. The two lead sections, SR (the Service de renseignements) and SCR (Counterespionage) handled covert intelligence gathering. The latter unit also supervised agents abroad. Section D held responsibility for codes and ciphers. Other sections dealt with wiretaps, radio and chemicals, sabotage, and administration. The whole operation ran on a shoestring, however, and the score of officers assigned to it perceived it as a professional dead end. Between 1934 and 1939,

96

Seeking a Scapegoat

Colonel Rivet, the head of the SR-SCR, was summoned to advise the prime minister only four times, and to brief Gamelin on a mere handful of additional occasions before late March 1940.59 SCR reached cooperative arrangements with the Surveillance du territoire at the Interior Ministry and the Paris police between 1937 and 1939. Nevertheless, all three agencies continued to suffer from political interference. Even though the German and Soviet embassies ran many spies directly, the Quai d’Orsay shied away from confrontation. At one point, Premier Daladier threatened to fire the entire SCR staff when they sought to expel Otto Abetz, later ambassador to Vichy, who masterminded Nazi propaganda in Paris. The SCR maintained a chain of border listening posts at Lille, Metz, Belfort, and Marseille (as well as in North Africa to detect colonial subversion). Each major border post was staffed by fifteen to twenty officers. Those branches in turn recruited some 1,500 paid agents who monitored Wehrmacht movements and penetrated Abwehr substations. They also debriefed thousands of French travelers, called “honorable correspondents,” who had spent some time in the Reich. Yet the size of this apparatus caused problems of its own. The difficulty lay in picking out the crucial nugget from a profusion of contradictory reports and catching the attention of the High Command despite several intermediate layers of bureaucracy. The chief flaw in French intelligence derived from an organizational schema that rendered decision-makers skeptical of its work product. Above the SR-SCE operation on the avenue de Tourville stood the 2ème Bureau on the rue de l’Université, headed successively by Colonels Louis Koeltz and Maurice Gauché, and attached directly to the General Staff. The most important unit there, the Section des armées étrangères, synthesized the information coming from the SR-SCR, its naval equivalent, the military and air attachés, and allied intelligence agencies insofar as they would cooperate. It produced daily, weekly, and triennial intelligence briefs that went to the general staff, the service ministers, and the secretariat of the highest-level defense committee (the Conseil supérieur de la défense nationale and successor organizations). Unfortunately, the daily bulletins lacked context. The larger synthetic reports, while broadly accurate, homogenized information to the level of cliché. No wonder that the self-satisfied General Gamelin, as well as civilian ministers—even Daladier, who had served in intelligence himself during World War I—gave more credence to private sources and rumors than to papers coming up the chain of command.



Seeking a Scapegoat

97

In addition, some representatives abroad proved more insightful than did others. Pierre de Margerie, ambassador to Berlin up to 1931, reported too hopefully, and his successor, the able but lubricious André François-Poncet, believed for several years that France might somehow reach a modus vivendi with the Nazi regime through cleverness. General Georges Renondeau, the percipient military attaché from August 1932 to November 1938, crafted dispatches that always hit the target, and the assistant air attaché, Captain Paul Stehlin, produced astonishingly good information by dint of personal contacts in the Luftwaffe. However, Renondeau’s fatuous successor, General Henri Didelet, told Gamelin and his other pals at GHQ what they wanted to hear, namely, that the Germans would not be ready for war until 1942.60 Admittedly, the intelligence establishment did not contain economists, scientists, or industrial experts. For cultural reasons, the limited number of able specialists in those fields did not choose military careers. Lack of technical expertise reinforced the tendency to underestimate how long it would take German war industries to ramp up production. During the Weygand era (1930–1934), the intelligence staff produced alarming reports on how fast the Reich could rearm, partly because the commander in chief needed documentation to persuade insouciant war ministers and pusillanimous parliamentarians not to slash the budgets even more drastically than they did. During the Gamelin years (1935–1940), intelligence reports overstated Wehrmacht readiness by counting reserves and auxiliaries as almost equivalent to fully trained troops, a venial error perhaps given the historical aptitude of the German soldier for combat. The French, like the British, were impressed by the artfully staged Nazi demonstrations of air supremacy and exaggerated the number of first-line planes that the Reich could bring to bear. However, it is a myth that General Joseph Vuillemin warned against action during the Munich crisis and expressed anxiety about going to war in August 1939 because he overestimated the Luftwaffe. His fears stemmed rather from all-too-accurate appraisal of catastrophic weakness in the Armée de l’Air.61 One can identify no specific point from 1936 onward when clearer appreciation of enemy capabilities on the ground or in the air would have broadened the realistic options open to French policy-makers. Of the hundreds of agents inside Germany, only three provided truly decisive information. Chief among them stood Hans-Thilo Schmidt, the disgruntled brother of a general, who held a job in air signals intelligence (Göring’s Forschungsamt). Schmidt, known to the SR as Asché, turned over blueprints

98

Seeking a Scapegoat

for a prototype Enigma machine in 1932. Ironically, success in obtaining the Enigma blueprints, although later crucial for the Anglo-American war effort, did not do the French much good. Major Bertrand of D-Section turned the plans over to Polish cryptologists, and a group of Polish mathematicians figured out how to read the increasingly complex German codes between 1933 and 1939. Polish intelligence, however, withheld news of this accomplishment from the French and British until July 1939, presumably because the Warsaw governing authorities, obsessed by their “two enemies” theory, were not sure for a long time where their political interests lay. Bertrand’s Polish team decoded more than 8,000 intercepts between October 1939 and April 1940, and, as noted above, Gamelin belatedly recognized their value. But Enigma failed the code-breakers when the Germans added an extra key, making the messages unintelligible, during the three-week period surrounding the May 1940 blitzkrieg.62 VI Today the leading textbooks put intelligence failure front and center as the principal explanation for the fall of France. “The failure of the French to predict the locus of the German invasion,” students learn, “must rank as a failure of intelligence as dramatic as the American failure to predict Pearl Harbor or the Israeli failure to predict the Egyptian attack in 1973.”63 The diffusion of this interpretation marks a true historiographical revolution. But the latest interpretations are not necessarily the best. For the first several decades after World War II, most analysts looked elsewhere for the essential factors contributing to the debacle.64 Marc Bloch set the tone in his classic L’étrange défaite (Strange Defeat), based on personal observations during the 1940 collapse. The traditional debate focused on the respective importance of outdated military doctrine and inadequate rearmament on the one hand, and institutional decadence on the other. Bloch found himself torn between fury at the “utter incompetence of the High Command” and his sense that an entire generation had failed.65 Historians who blame the military fault particularly the doctrine of the “continuous front.” In World War I, the defense held the advantage. It could bring up reserves to plug, or colmater, gaps in the trenches. That doctrine lost its logic once the adversary could combine air superiority and tank mobility as a force multiplier to concentrate strength and punch through the weakest



Seeking a Scapegoat

99

point in the defense line. The geriatric composition of the French officer corps, with generals well along in their sixties and colonels ten years senior to their Wehrmacht counterparts, militated against strategic adaptability and tactical flexibility. In any event, the continuous front was doomed in May 1940 when Gamelin unilaterally denuded the reserve to send the cream of his army streaking north toward Breda in the Netherlands.66 As Bradford Lee demonstrates, one cannot properly analyze strategy and armaments in separate compartments. Each ineluctably shapes the requirements of the other.67 Owing to the late but severe Depression and the compression of its military budget, France fell behind Germany in the production not merely of modern bombers and fighters, but also of tanks, anti-tank guns, anti-aircraft guns, mobile artillery, and all manner of other war matériel. When Daladier’s cabinet abandoned the comfortable illusions of the Popular Front in autumn 1938 and launched a substantial rearmament program, the country’s atomized war industries, hobbled by an antediluvian machine-tool park and insufficient capital to introduce assemblyline techniques, could not ramp up high-quality production without what turned out to be a fateful lag.68 Those who support the deliquescence theory hold that in the 1930s, French political and social institutions became increasingly dysfunctional.69 The political class in the Third Republic nurtured a visceral suspicion, amounting almost to paranoia, of the “man on horseback.” Following the dictum of the Radical-Socialist philosopher Alain, the proper latter-day Jacobin favored parliamentary supremacy as the best guarantee of democracy. This doctrine did not change when other advanced nations strengthened governance from the center to meet the requirements of a complex industrial society. The Third Republic operated through a series of weak coalition cabinets including ministers from antagonistic political parties. Reshuffles took place with kaleidoscopic frequency. The president of the Republic enjoyed little more than ceremonial powers, and those who sought to vest greater power in the executive—for example, Alexandre Millerand in 1924 and André Tardieu in the early 1930s—met with frustration and ostracism.70 Stanley Hoffmann has described France during this era as a “stalemate society.” The Chamber of Deputies could always muster a negative majority against change. Yet only in extremis (and often not even then) could it unite for structural reform.71 The deputies played an important linking role between the administration and their constituencies, but they received such paltry salaries that a startling

100

Seeking a Scapegoat

number felt compelled to mount fiddles on the side. Corruption ran rampant. Frequent scandals, although usually begun for reasons of political intrigue, undermined public trust.72 Backbenchers retained a local focus and rarely enjoyed the luxury of concentrating on national issues. The same national politicians frequently turned up in succeeding ministries, leading some political scientists to dispute the theory of ministerial instability.73 Nevertheless, the cabinet rarely functioned effectively as a coordinating body. No one had rationalized French government in the way that Sir Maurice Hankey had imposed orderly management on Whitehall during and after World War I. Except in times of emergency or under strong personalities such as Poincaré, Tardieu, or Doumergue, prime ministers had to negotiate almost as equals with other ministers. Each ministry thus ran semi-autonomously. Owing to the characteristic social institution of la brouille—life-long estrangements that often derived from obscure schoolboy or professional quarrels—high civil servants in rival ministries or generals with different patrons cooperated only under compulsion. The fissiparous culture of the governing elite found expression in the intelligence community, of course, but also right through the military establishment. The Vichy regime staged the Riom trial in 1942–1943 in order to discredit the Third Republic. Many of those who testified, hoping to save their own skins, indulged in exaggeration and caricature. Still, the representation of class conflict in the 1930s, and of the verbal violence that accompanied it, rested on a fundament of truth. In the undeclared civil war that raged during the Popular Front years, public opinion suffered from what the later Resistance leader Georges Bidault called “an atomized, incoherent, and fanaticized press.” Left-wing newspapers routinely described conservatives as “fascist,” while bourgeois publications voiced apprehension that wildcat strikes and limited nationalizations would pave the way for genuine insurrection. Lacking adequate revenue from advertising and subscriptions, some 80 percent of the newspapers, according to Daladier’s estimate, received under-the-table subventions from someone.74 The Soviet, Italian, and German governments distributed funds most lavishly. The German embassy, the Dienststelle Ribbentrop, and Goebbels’s Propaganda Ministry suborned complaisant journalists to create an image of Nazi Germany as an unstoppable dynamo.75 Prudent readers did not always credit the propaganda, but they grew skeptical of what their own leaders said. Reestablishing national solidarity on the 1914 model became a daunting task once the war broke out.



Seeking a Scapegoat

101

Social cohesion is a multifaceted phenomenon that defies exact measurement. The World War I bloodletting continued to wreak psychological havoc, not only on the survivors, but also on the younger generation. France’s public school teachers drove home the lesson that true patriotism rested not on oldfashioned values, but on a commitment to international solidarity and peace. In the 1930s, some 82 percent of French schoolteachers held membership in the pro-pacifist, Socialist-affiliated Syndicat national des instituteurs.76 Doubtless most conscripts in the “B-Series” reserve divisions fought bravely by their own lights in 1940. The question remains whether an anti-militarist current in the wider society shaped the attitude with which they had approached one-year basic training in earlier years. Gamelin, for one, thought that it had. After the Sedan front collapsed, he complained that the citizen-soldiers did not believe in the war: “Inclined to criticize incessantly all those holding a smidgen of authority, encouraged to enjoy the easy life as a mark of civilization, today’s conscript never received between the wars the moral and patriotic education that would have prepared him for the main show.”77 VII Strategy, armaments, economic mobilization, and morale of the nation in arms are interrelated in complex ways. The point to observe is that the purely military and the wider structural explanations of France’s deliquescence are not contradictory, but complementary. Clausewitz famously observed that success in warfare depends upon a “wondrous trinity”—the persistent and creative interplay among the government, the army, and the people.78 The theorist of war Martin van Creveld contends that this trinitarian analysis no longer applies in the dawning era of counterinsurgency and low-intensity warfare.79 But military professionals unanimously hold that it provides a serviceable framework for analysis in the period, roughly from 1792 to 1945, when organized nations, armies, and peoples contended against each other. Where does intelligence fit in the Clausewitzian equation? Intelligence can serve as one force multiplier among others, particularly in determining tactical outcomes. During World War II, the information that the Allies obtained from “Ultra” and “Magic” also made a difference on the strategic level. Still, academic specialists caution against the illusion that intelligence professionals can ever wield thaumaturgical powers. Richard Betts, for example, stresses that intelligence failures are inevitable. Perfecting norms and procedures

102

Seeking a Scapegoat

for analysis cannot prevent them. Breakdowns can occur in amassing data, resolving ambiguities, communicating conclusions to decision-makers operating within a political context, and persuading them of its relevance. Indeed, the consumers of intelligence are more likely to err than collectors of raw information or those who generate finished analysis.80 The nub of the difficulty, as Michael Handel observes in his classic study of the politics of intelligence, is that even in those rare cases where the data point in a single direction, facts do not speak for themselves. They must win acceptance among government actors with a range of ideological preferences, biases deriving from temperament or experience, conflicting bureaucratic agendas, and loyalties to diverse domestic constituencies. In other words, when intelligence assessments are applied to operational decisions in the real world, they habitually become politicized. Leaders scrutinize intelligence data for whatever proves politically or bureaucratically useful. As an added complication, decision-makers atop large bureaucracies, whether military or civilian, operate under tight time constraints. They do not have the leisure to dispassionately assess reports that run counter to what they think they know. Hence, as Robert Jervis notes, decision-makers hardly ever accept unwelcome news without agonizing inner struggle.81 These generalizations, while not crafted with the 1940 debacle in mind, help to situate the problems of the SR and the 2ème Bureau between the wars within a larger context. They illuminate why those agencies, aside from the difficulties of gathering and correlating evidence without robust input from signals intelligence, experienced such difficulty in obtaining a regular hearing at the highest levels. Whatever the flaws of the atomized intelligence services during those decades, they reflected a larger absence of coordination—incoherence is not too strong a word—within labile coalition governments and in the military establishment. The governing elite, moreover, confronted such overwhelming problems of demography, military defense, and foreign policy after World War I that net assessment of enemy capabilities, at least until a new conflict became imminent, featured as a second-order problem. Small countries adjacent to powerful, ideologically alien, and irremediably hostile powers sometimes face such circumstances. Individuals can choose suicide, “internal emigration,” or territorial emigration. Nation-states cannot. The public in such conditions may embrace magical thinking or resort to psychological denial. Leaders preoccupied with managing a parliamentary



Seeking a Scapegoat

103

coalition necessarily employ the intelligence product within a larger political environment. VIII France’s strategic dilemmas remained insoluble no matter how fine the foreign reporting. This is a thrice-told tale, but fundamental nonetheless. The French people emerged from World War I with a widespread conviction that a second such experience would finish the country off. The demographic facts were grim. Owing to the nation’s low birthrate during the previous half-century, France already had a skewed population distribution in 1914.82 The million and a half men who died or emerged horribly disabled from the war left a hole in the labor force and the marriage market greater than those in Germany or in England. In 1921, France had 37,000,000 inhabitants, Germany 64,000,000. The German population was growing three times faster proportionately than the French. By 1938, after incorporating Austria and the Sudetenland, Germany would have twice the population of France. And France could scarcely aspire to match the potential industrial power of the Reich once that country, drawing on the superiority of its scientific and technical education, reconstituted its manufacturing base. Notwithstanding reasonable economic growth in the 1920s, France remained an inward-looking, predominantly agricultural society. The United States and Great Britain had provided the margin of victory in 1918. Many who cherished the heritage of la grande nation deplored this inconvenient truth. Behind the bunting and the blather, the French did not much like the AngloSaxons. Aside from economically backward allies in Eastern Europe, in fact, France effectively stood alone to contend with the erstwhile foe. A great many Germans thought of the Weimar Republic, in the words of Chancellor Heinrich Brüning, as a type of “mandate or colonial regime,” shoved down their throats by the victors. The “fulfillment” chancellor, Joseph Wirth, saw eye to eye with General von Seeckt that “Poland must be finished off.” Even Foreign Minister Gustav Stresemann, the purported apostle of reconciliation, nurtured the most expansive revisionist aims.83 Moreover, the unambiguously pro-republican parties never enjoyed a working majority in the Reichstag after 1920. When a bitter and revanchist Germany recovered, France could never hope to contain the Reich through its own resources alone. That perception

104

Seeking a Scapegoat

was a fact of life, an element in the miasma suffusing every security debate in the Chamber of Deputies for twenty years. At the 1919 Paris peace conference, Clemenceau traded a permanent occupation of the Rhineland for a fifteen-year, three-stage occupation that would end in 1935, precisely when the annual army draftee intake would fall by half owing to the low birthrate during the war. In exchange, Clemenceau hoped to obtain an Anglo-American guarantee pact. President Woodrow Wilson did not submit the pact to the U.S. Senate, however, after that body rejected the Versailles treaty. France negotiated for a substitute guarantee pact with Britain under various rubrics from 1921 to 1924; but the British refused to be drawn. France tried to narrow the gap with Germany industrially by obtaining coal, coke, and capital on reparation account; the Germans refused to pay in a meaningful way. The French occupation of the Ruhr in 1923 failed to coerce the Germans into compliance. Some French generals wanted to separate the Rhineland by force from the Reich, but that would alienate Britain further and require a permanent occupation that would worsen the chronic labor shortage. The treaty required the Reich to disarm, yet the Germans fulfilled the disarmament stipulations grudgingly, if at all. The SR kept close track of German violations, but had no influence on the outcome. In 1927, the Inter-Allied Military Control Commission, having lost British support, withdrew without completing its mission.84 Franco-British intelligence cooperation did not resume for eleven years once the Control Commission had folded its tents.85 In 1925, Britain at length consented to a mutual guarantee of the Rhineland at Locarno. The Cartel des Gauches government of the day, overwhelmed by pacifist sentiment and pressure for a reduction to one-year military service, agreed with “eyes wide shut” because it perceived no other option. The fine print of the pact made clear that, if Germany attacked its neighbors to the east, France could not assist them and retain the promise of British intervention on its side. Therefore, while the mass media continued to play up France’s Eastern alliances with Poland and the Little Entente, the French high command gradually ceased to consider them a central element in planning.86 Aristide Briand, French foreign minister from 1925 to 1932, read neither diplomatic dispatches nor intelligence reports. Still, he had a golden tongue, a conciliatory temperament, and a gift for self-deception. He imagined that he could gradually reconcile Germany to a less revisionist course. He came to feel a quasi-religious mission to establish international peace, “Gesta Dei



Seeking a Scapegoat

105

per Francos,” as he explained the concept to his staff.87 The majority of the traumatized French public joined him in praying that wishing could make it so. When the world monetary regime broke down in 1931, however, it became apparent that Briand’s policy had failed. Prime Minister James Ramsay MacDonald of England now opined that France had caused the war, and the American secretary of state sought to intimidate Pierre Laval into forcing retrocession of the Polish Corridor to the Reich as a contribution to peace.88 The French army had to decide in 1927–1928 how to dispose its forces as the Rhineland occupation neared its end. With the franc stabilized after fourteen years of monetary disruption, the authorities anticipated having more money than men during the “hollow years” to come. They decided to build the Maginot Line, a string of forts that would allow a continuous field of fire from the Swiss border to the hinge of Luxembourg. If one proceeds from the French shibboleth of the “nation in arms,” the decision embodied a perverse logic. The small standing army could provide cover, couverture, in the Maginot Line while the reserves had time to mobilize. At first, planners expected to prepare a defense in depth within fortified regions, but that concept eventually fell by the wayside. The expectation that Belgium would coordinate the line north of Montmédy with its own fortifications went glimmering when the Brussels government returned to neutrality.89 In any event, the choice to build a concrete barrier excluded any role for a rapid deployment force. It committed France irrevocably to a defensive strategy. In 1936, Defense Minister Daladier reminded the planning body tasked with organizing the nation in wartime: “France has a defensive policy commanded by ethnological, political, and psychological considerations.”90 To complicate the conundrum, France sought to uphold the gold standard after Britain and the United States abandoned it during the years 1931–1933. That required deflation, fiscal stringency, and compression of the military budget. Despite the best efforts of General Weygand, the left-wing ministries in power from 1932 to 1934 (most notoriously, those of Herriot, Joseph PaulBoncour, and Daladier himself) cut funds for the military by almost onequarter. One could hardly economize on food, uniforms, and barracks for the draftees. The Army had already made a commitment to build the Maginot Line, even though the high water table made it prohibitively expensive to extend the line from Montmédy west to the sea. A single practical solution emerged: to cancel field maneuvers and to slash the budget for new equipment to the vanishing point, just as the Reich pushed rearmament into high gear.91

106

Seeking a Scapegoat

Open sources allowed anyone who paid close attention to follow the progression of events. The French intelligence community paid close attention to the course of rearmament in the Reich from 1930 until Hitler openly threw off the strictures of Versailles five years later. Yet even the casual newspaper reader at the time of the Geneva Disarmament Conference of 1932–1933 could hardly escape the conclusion that Britain and America would do nothing. France effectively stood alone.92 The 2ème Bureau marshaled the evidence perfectly well. Once the 1932 elections gave the Left a strong majority in the Chamber, the political class did not think it feasible to react. A subtle combination of political and psychological impulses—parliamentary stalemate, pacifist sentiment, financial penury, the voters’ wish to defend the franc, perceived threats to regime stability from both the Right and the Left—produced creeping paralysis. Some observers began to use the word “decadence”; others thought that the term lacked precision. The Popular Front that came to power under Léon Blum in 1936 privileged domestic reform over everything else. Daladier, who took over the defense portfolio again, slowly began to promote rearmament to the extent compatible with the social and economic goals of the Popular Front coalition. He apparently felt twinges of guilt for letting the Army run down too far during his earlier incumbency at the rue St.-Dominique. He shortly instructed his staff to start compiling a record of what he had done to promote the national defense.93 Despite the statistical labors of Robert Frankenstein, the details of credits voted, actual expenditures, and tangible results remain shrouded in obscurity. It is certain that the Reich spent almost twice as great a percentage of a much larger national income on the military in 1936–1938, and that France fell ever further behind in almost every category of weaponry in 1939.94 IX General Gamelin accurately informed the cabinet in November 1935 that Hitler planned to remilitarize the Rhineland and that without general mobilization and allied support­—neither realistic possibilities—France could do nothing to stop him. That judgment followed quite logically from the fine print of Mobilization Plan D-bis, adopted seven months earlier. Shortly afterward, the political leadership abandoned the notion of a Franco-Italian alliance that Foreign Minister Pierre Laval and General Gamelin had earlier



Seeking a Scapegoat

107

pursued. Given the frenetic character of British opposition to Mussolini’s attack on Abyssinia in fall 1935, risk-averse logic made it necessary to privilege good relations with England over the will-o’-the-wisp of rapprochement with the mercurial Italian dictator.95 All the same, this Hobson’s choice marked the last possible moment when France, on strictly geographical grounds, could have gone to the aid of its eastern allies, even if the mobilization plan had not already excluded the idea. French intelligence could read Austrian diplomatic traffic and gave precise advance warning of the Anschluß in February 1938; Daladier and the High Command pondered the matter and decided not to react. When the Czech president Edvard Beneš came to Paris in May 1938 to discuss the forthcoming German demand for the Sudetenland, Daladier warned him plainly that France could do very little for Czechoslovakia. The Czech intelligence chief, General František Moravec, subsequently alleged that he had passed on information suggesting that France could assault the Siegfried Line successfully. Such a claim shows just how out of touch the Czechs were with reality. The French partial mobilization of September 1938 resulted, as expected, in chaos. Moreover, as General Vuillemin pointed out, France possessed no more than fifty first-line aircraft at the time, and hardly any anti-aircraft guns.96 Once the Munich conference exposed the country’s military weakness, France could not conjure up any coherent policy in Eastern Europe. After several years of indecision, some politicians persuaded themselves that they could turn the Franco-Soviet non-aggression pact of 1935 into a military convention. In August 1939, the High Command dispatched General Doumenc along with a British mission to negotiate.97 This was always a fool’s errand, not only because Poland would never admit Russian troops, but also because Stalin had already striven for months to conclude a Nazi-Soviet entente. A German diplomat in Moscow had kept the U.S. embassy secretly apprised of developments, but the Americans could not tell the French because of the Quai d’Orsay’s well-known “sieve-like qualities.”98 The 2ème Bureau did not learn in real time of Stalin’s explanation that Russia’s interest lay in pushing the Nazis and the Western capitalists into a mutually destructive conflict in order to pave the way for world revolution.99 But, to its credit, French Intelligence had expressed skepticism since 1935 about the Kremlin’s real designs. Characteristically, the politicians did not consider this an intelligence function. Daladier told Gamelin pointedly “that the Deuxième would do better to devote itself to gathering military information.”100

108

Seeking a Scapegoat

X In a nutshell, German effective military superiority over France loomed so large from 1936 onward that no improvement in net assessment of German capabilities or intentions by the intelligence staff could have reversed the tide. The “gravediggers of France,” as the journalist “Pertinax” (André Géraud) called them, operated at higher levels. French deficiencies included a pitifully inadequate air force, defective doctrine in the use of tanks, primitive military communications, the breakdown of industrial mobilization, a command structure that made strategy turn on political influence, and, most important of all, incoherence in governance. Space allows only brief synthesis of the literature on each of those subjects. France did not start to design a competitive air force until late 1938. We do not know whether Pierre Cot, air minister in 1933–1934 and again from 1936–1938, enrolled as a Soviet agent during his tenure at the Boulevard Victor or signed up with the KGB afterward. His policies, at any rate, set aircraft construction so far back that the country could never catch up.101 Cot placed primary emphasis on creating a strategic bomber fleet of manifestly obsolescent design that he imagined could take part in an “anti-fascist” coalition. His scheme to build multiple-mission “BCR” aircraft produced lumbering machines that did nothing well. The atomized French airframe industry required consolidation before it could adopt mass production, but Cot wreaked havoc through a high-handed nationalization program. He also refused to shut down the assembly lines for the introduction of new prototypes, purportedly because he feared unemployment. Finally, in the name of democratization, he purged the officer corps and packed the civilian ranks of the Air Ministry with Communists. Many would later turn up in the Resistance—but only after the USSR switched sides in 1941.102 When Guy La Chambre succeeded Cot, he valiantly tried to stop the rot. Cot had fudged production figures so grotesquely that British had described him as “a remarkable liar.” La Chambre could not admit the full truth, however, without undermining the credibility of France as an ally.103 Very slowly, the ministry licensed manufacturers to fabricate aircraft capable of infantry support and reconnaissance under modern conditions. Still, the industry lacked the mass-production facilities, skilled workers, aluminum, and other raw materials needed to turn out sophisticated bombers. La Chambre and Daladier realized that they could only make good those deficiencies by



Seeking a Scapegoat

109

obtaining help from the United States. Ambassador William Bullitt thought that the Allies needed a minimum of 10,000 aircraft to win the war. Daladier agreed. Eventually, the Allies contracted for 1,600 planes. Alas, less than 200 reached the front before the end of the Battle of France, although the few that arrived indeed outperformed domestic models.104 On 10 May 1940, France could theoretically boast 1,368 first-line aircraft in the Métropole, compared with 3,500 Luftwaffe planes. But the disproportion (disregarding the superior speed, payload, robustness, and maneuverability of German machines in almost every category) loomed larger than those figures suggest. The buildup came too late for the Armée de l’Air to elaborate an effective strategic doctrine. And it had to divide its 1,000 modern planes among four zones of operation as well as to divert several fighter squadrons for the defense of cities and airfields that lacked anti-aircraft protection. Moreover, the German air fleet included a greater number of fighters and bombers; almost all modern French bombers were still undergoing shakedown training in the southeast. Finally, while French factories churned out another 1,000 planes to replace losses over the next month, just half were ready to fly when delivered. After the first week in June, French aircraft effectively disappeared from the skies.105 French inferiority in tank warfare derived more from doctrine than equipment, yet also played a noteworthy role in the defeat. Admittedly, the French Army as an institution did not encourage thinking “outside the box.” Yet the decision to award priority to motorization of the combat-ready infantry and light cavalry units and to postpone creation of independently maneuverable tank divisions on the German model represented a justifiable, indeed logical, political choice in the middle 1930s. The intelligence services charted the progress of German tank development accurately. Major Schlesser of SCR arranged to translate General Heinz Guderian’s 1937 book Achtung – Panzer! into French and directed that copies be placed in regimental libraries. When Lieutenant Colonel Charles de Gaulle authored a comparable scheme, General Julien Dufieux, the leading tank expert, calculated that implementing it would require doubling the army budget, as well as a shift to a long-service army that ran counter to French republican tradition. In addition, a defensive army could use heavy tanks only for a hypothetical counterattack. Characteristically, de Gaulle made light of logistic and cost constraints. Yet without expensive reinforcement, most French bridges could not hold heavy tanks. What’s more, France had no synthetic oil industry that could replace imports from the Caribbean were German submarines to dominate the Atlantic.106

110

Seeking a Scapegoat

By dint of single-minded focus on tank production, France entered the lists with 3,020 tanks, slightly more than the first wave that the Wehrmacht could field. French tanks also boasted heavier armor, although they moved more slowly than their German equivalents. The achievement, however, came at grievous cost. Of 102 infantry divisions that took the field, only twenty-two received the requisite complement of anti-tank guns, anti-aircraft guns, mortars, heavy machine guns, and armored tractors. What’s more, the three light mechanized divisions (DLMs) and the three heavy reserve armored divisions (DCRs) existing at the outbreak of hostilities comprised together only 960 tanks. The other tanks were scattered about the infantry in battalion-sized units and largely wasted. While the DLMs performed brilliantly in screening and reconnaissance missions at the Gembloux Gap in Belgium, the DCRs, which were supposed to await an opportunity to counterattack, dissolved instead. The hastily formed DCRs had not received sufficient training to coordinate their logistic and repair functions. The B-1 bis and the H-39 tanks in the DCRs had sacrificed cruising range for thicker armor, and they could not refuel under air attack. When they ran dry, the crews of those superb fighting machines had to abandon them.107 Communications figured as a third area in which the military machine broke down. The lack of scale and scope may perhaps explain why technical innovation stagnated in so many branches of French industry during the prewar decade. The French radio industry notoriously failed to keep up with advances elsewhere, and few engineering officers underwent training in that specialty. Tank designers in the 1930s did not originally plan for radios, and retrofitting, when it began in 1939, proved awkward. For example, on the most advanced cavalry tank, the Somua S-35, the after-market radio stood atop the shell casing ejection port and therefore failed as soon as the tank went into battle. The absence of short-range radios linking tank units and supporting aircraft made their coordination on the blitzkrieg model impossible. An antediluvian transmission system also hindered command and control on a higher level. Skeptical of newfangled radio telephones and assuming that the continuous front would hold, French technicians laid overland telephone wires to connect headquarters staffs with the front. Once Panzer units overwhelmed forward defenses and cut the wires, senior commanders could no longer control the battlefield in real time.108 Backwardness in telecommunications stands as a synecdoche for the larger inability of the French economy to sustain a modern war. Third Republic



Seeking a Scapegoat

111

politicians, mostly products of the standard literary education of the day, lacked the intellectual framework for understanding industrial organization. With utmost reluctance, Daladier agreed at the start of the war to appoint the eminent efficiency expert Raoul Dautry as minister of armaments and scientific research. Dautry surrounded himself with a galaxy of Polytechnic graduates, who labored night and day to stimulate industrial mobilization. They encountered steady obstruction both from the military, who did not grasp the need to furlough the necessary engineers and skilled workers, and from the premier himself, who gave deceptive production reports to parliament and demanded that the armaments minister cover his misstatements. By mid-December 1939, Dautry confided to intimates that the effort had been “completely bungled” and that France would lose the war.109 Faced with stagnating output, missing components, and a horrifying percentage of defects in matériel, both the politicians and the High Command looked for scapegoats. They convinced themselves that the Communists were sabotaging production. Admittedly, the surly and belligerent metal workers, angered by frozen wages and longer hours, did not display the same dedication that British labor mustered in the national crisis. Yet only a few incidents of organized sabotage took place. The governing elites did not wish to face the unpalatable truth. Atomized and undercapitalized industries, handicapped by old production techniques and bereft of a rational scheme for allocating raw materials, simply could not mobilize sufficiently to sustain a modern war.110 A mountain of evidence contradicts the legend that no one thought the Wehrmacht could get through the Ardennes. Every graduate of Saint-Cyr, the French military academy, knew Marshal Foch’s dictum by heart: all terrains are penetrable if not vigorously defended. The northern Ardennes, unlike the thickly wooded areas further south, contains several valleys that facilitate passage by armored units with bridging equipment. In the 1937 and 1938 war games, the side playing the enemy reached the slow-flowing Meuse River more quickly than the defending infantry, lacking trucks, could bring up reinforcements. General André Corap, head of the 9th Army, called it “idiocy” to think that the Germans could not get through. Even General Gamelin confessed to the British Chief of the Imperial General Staff (CIGS), soon after the war started, that Hitler might attack there, although he did nothing to prepare for such an eventuality.111 General André-Gaston Prételat, who surveyed the area for the Conseil supérieur de la guerre following the Munich crisis, found no anti-tank obstacles, no barbed wire, and scattered bunkers that would never

112

Seeking a Scapegoat

withstand serious assault, even when complete. A parliamentary commission headed by Pierre Taittinger again reconnoitered the border from Montmédy to Valenciennes in March 1940, and the junior officers from Georges’s Operations Bureau who drafted its report emphasized the “terrifying flimsiness” of the defenses. Gamelin’s camarilla brushed this report aside, however, and the well-connected General Charles Huntziger, who headed the 2nd Army west of Sedan, replied ironically that he needed no lessons from a former corporal. Yet other senior officers took the situation more seriously. Many experts had warned against expecting good performance from the Army’s leasttrained and worst equipped “B-Series” inactive reserve divisions, five of which happened to hold the Sedan salient at the juncture of the 2nd and 9th armies. The twenty-one B divisions had originally taken shape as a type of regional guard. The older reservists barely remembered basic training, and their morale had deteriorated badly when separated from their families during the drôle de guerre. Even better-led troops learned to keep their heads down and just went through the motions. Corap begged repeatedly for more resources, but he lacked political clout. Thus his sector remained the Army’s “poor relation”—last in line for seasoned formations, money for obstacles, and new equipment right down to the day of the German attack. In the first week of May, a former École de Guerre professor told Georges that the failure to strengthen the central front seemed illogical. Georges agreed, but threw up his hands: given Gamelin’s perverse insistence on sending the motorized forces north to Breda and the power of Prételat’s Paris friends to warehouse superfluous divisions behind the Maginot Line in the east, he could do “very little.”112 When political influence shapes the disposition of forces, even in part, optimal strategy suffers.113 Incoherent leadership at the top intensified all the other problems. Colonel Paul de Villelume, chief liaison officer between the General Staff and the Quai d’Orsay between 1935 and March 1940, and afterward director of Paul Reynaud’s military cabinet, paints a horrifying picture in his diary of the paralyzing fears, administrative chaos, frivolous rivalries, and disregard of elementary logic at the highest levels of French government during the “phony war.” Both Raymond de Sainte-Suzanne and Roland de Margerie provide concordant eyewitness accounts of chicanery and intrigue at the Quai d’Orsay that reinforce the impression of a government spiraling out of control.114 The two main figures in the cabinet, Daladier and Paul Reynaud, the man who replaced him as premier in March 1940, hated each other with a



Seeking a Scapegoat

113

passion. The American ambassador ruled out reconciliation: “The difference in their policies is slight and they are both able men, but the lady love of each hates the lady love of the other, and . . . venom distilled in a horizontal position is always fatal.”115 The respective mistresses drew the subordinates of both men into their cabals so that the conduct of public business came to revolve around personalities. No one thought that France could help Poland in August 1939, but the War Committee saw no point in waiting passively for a German attack. Thereafter, however, the cabinet deadlocked bitterly on whether to pursue vigorous action or to seek a compromise peace. Formal coordination of policy and strategy became impossible, and Daladier made key decisions at informal late-night gabfests with a few subordinates at which no minutes were kept.116 Daladier stood between the political camps, and his own mood oscillated wildly. Always a weaker character than he affected to be in public, the premier told his staff following the Polish defeat that if the Germans invaded France he would “blow his brains out.” To Ambassador Bullitt he offered a variant: the bombardment of France would be so terrible that the people would “drive him from office and probably kill him.”117 Secretary-general Alexis Léger of the Quai d’Orsay had also abandoned hope; he confessed to Bullitt on 30 September 1939 that “the game was lost.”118 One could not, of course, make such admissions publicly and hope to sustain the trust of a fractious and disoriented parliament. The right-wing parties longed for a display of energy—at least against the Soviet Union. The need to manage parliamentary sentiment partly explains why phantasmagoric stratagems for peripheral actions received such public attention. Schemes for a landing at Salonika in Greece to create a Balkan front, an attack on Russia by way of Scandinavia to assist Finland, or an air strike into the Caucasus to bomb the oil fields and raise the tribes were successively aired and discarded.119 Oddly enough, no one thought to revise the plan that would send millions of citizens fleeing south to “sister cities,” clogging the roads of northern France in the event of a Nazi attack. Daladier’s cabinet lost its footing in mid-March 1940 after Finland signed a humiliating armistice with the Soviet Union. Yet Paul Reynaud, who succeeded him as premier, won the Chamber’s approval by a single vote, with his own party voting against him. Reynaud aimed to refashion national union by constructing a monster cabinet of thirty-five ministers and undersecretaries. In fact, bringing together so many people with diametrically opposing views proved a recipe for immobility. Seeking to model himself on Clemenceau,

114

Seeking a Scapegoat

Reynaud imagined that he could boost efficiency through a seven-member war cabinet. That inner group also remained irremediably split, however, with Paul Baudouin, secretary of the war cabinet, numbering from the outset among the greatest pessimists.120 Gamelin had always maintained that he had no obligation to reveal the details of operations to his civilian superiors, but he flatly declined to cooperate with Reynaud, whom he denounced as “a swine . . . utterly unworthy of confidence.”121 At length Reynaud resolved to dismiss Gamelin, but Daladier, who retained the war ministry, blocked him. Deeply frustrated, Reynaud sought to resign on the very eve of the German attack. He agreed to reshuffle the cabinet instead only after a fervent appeal by the president of the Republic. Still, interministerial conflicts did not abate once the shooting war began. Under the pressure of events, Reynaud engineered a more successful shuffle of personnel on 18 May. He took the War Ministry himself and relegated Daladier to the Foreign Ministry, recalled General Weygand from Syria to replace the hapless Gamelin, and forced through the dismissal of the disloyal Léger at the Quai d’Orsay.122 But new men brought new quarrels. Reynaud had numerous talents, but even his admirers conceded that his courage sometimes wavered.123 Within a couple of weeks, he had embarked on an unedifying dispute with Weygand on the respective responsibilities of the Army and political class for the debacle.124 The question whether the SR had warned forcefully enough of a German strike through the Ardennes thus falls into a wider perspective. The strategists didn’t need to wait for confirmation from the SR that the center of gravity of the German attack ran through the Ardennes. Colonel de Villelume alerted Reynaud and Daladier within thirty-six hours to the fact that the Luftwaffe was not attacking French columns in Belgium seriously and rather seemed to be drawing them in, but he elicited no immediate reaction.125 Forty-two days would elapse between the moment when the Wehrmacht crossed the borders and the date that the French government, from its bolt-hole in Bordeaux, threw in the towel. But the latter part of the battle involved much political play-acting. Five days on, troops of the B-divisions on the Meuse were running for their lives, and diplomats were burning archives in the Quai d’Orsay courtyard. The outcome did not reflect an intelligence failure, but rather the collapse of a nation. As the British liaison officer, Sir Edward Spears, observed, it was as though behind the façade France had been “eaten away by white ants.”126



Seeking a Scapegoat

115

XI Although the 2ème Bureau played no more than a walk-on part in the debacle, the institution of independent military intelligence paradoxically would not survive. The leading officers of the SR-SCR mounted secret resistance as part of Vichy’s Armistice Army. They fled to join General Henri Giraud in Algiers shortly after the Allied landings in November 1942. Yet they found themselves squeezed out in late 1943 when General de Gaulle’s minions, chief among them Jacques Soustelle, secured the upper hand in North Africa and imposed a strict loyalty test. In London, meanwhile, de Gaulle had set up a Bureau central de renseignement et d’action (BCRA) under André Dewavrin, an engineering officer with no intelligence background, who took the name Passy. The BCRA carried out several undifferentiated missions at once: intelligence gathering, subversive action, controlling the interior resistance, and propaganda. Gradually the BCRA came to privilege its political role.127 Since de Gaulle had a single overriding objective—to seize control of the governmental machine after Liberation—other functions declined in importance. After hard-fought turf wars, a politically dominated Service de documentation extérieure et de contre-espionnage (SDECE) emerged in late 1945. That organization, staffed principally by worthy resistance veterans, took root under the auspices of the president of the council. When a scandal erupted over the disappearance of public funds, the Constituent Assembly added supervision by an interministerial committee. A domestic political police, the Direction de la surveillance du territoire (DST), survived at the Interior Ministry. The armed forces did not obtain permission to reconstitute professional military intelligence on the Third Republic model until 1966. The tradition of politicized intelligence remains embedded in French culture. Several denizens of the Elysée Palace under the Fifth Republic have maintained their own secret du roi.128 Notes 1. General Maurice-Henri Gauché, Le Deuxième Bureau au travail, 1935–1940 (Paris: Amiot-Dumont, 1953). 2. Paul Paillole, Services spéciaux, 1935–1945 (Paris: R. Laffont, 1975), trans. as Fighting the Nazis: French Military Intelligence and Counterintelligence, 1935–1945 (New York: Enigma, 2003); Paillole, Notre espion chez Hitler (Paris: R. Laffont, 1985); Paillole, L’homme des services secrets: Entretiens avec Alain-Gilles Minella (Paris: Julliard,

116

Seeking a Scapegoat

1995); General Henri Navarre et al., Le Service des renseignements, 1871–1944 (Paris: Plon, 1978), esp. 15–122); Gustave Bertrand, Enigma; ou, La plus grande énigme de la guerre 1939–1945 (Paris: Plon, 1973). Paillole also compiled a dozen cartons of testimony by fellow intelligence operatives, now located in Fonds privé Paillole, 1K 545, Service historique de l’Armée de Terre, Vincennes [SHAT], ably exploited in Olivier Forcade, “Le Renseignement face à l’Allemagne au printemps 1940 et au début de la campagne de France,” in Christine Levisse-Touzé, ed., La campagne de 1940: Actes du colloque, 16 au 18 novembre 2000 (Paris: Tallandier, 2001), 126–155. 3. General Louis Rivet, “Le camp allemand dans la fièvre des alertes (1939–1940),” Revue de défense nationale 5 (July 1949): 33–48; “Etions-nous renseignés en mai 1940?” pts. 1 and 2, Revue de défense nationale 6 (June 1950): 636–648 (July 1950): 24–39. Rivet’s son-in-law, Georges Castellan, drawing on privileged access, adduced further evidence for that view in “La Wehrmacht vue de la France,” pts. 1 and 2, Revue historique de l’Armée 5, no. 2 (1949): 35–48, no. 3 (1949): 39–56, and in Le réarmement clandestin du Reich, 1930–1935, vu par le 2e Bureau de l’État-major français (Paris: Plon, 1954). 4. Col. Ulrich Liss, “Die Tätigkeit des französischen 2. Bureau im Westfeldzug 1939/40,” Wehrwissenschaftliche Rundschau 10 (1960): 267–278; for Liss’s praise of French intelligence culture compared with that of the Wehrmacht, see his “Erfahrungen und Gedanken zum 1c-Wesen,” ibid. 7 (1957): 616–627. For an early evaluation that cites 2ème Bureau mistakes but mainly emphasizes Gamelin’s “inconceivable contempt” for their work, see Gerd Brausch, “Sedan 1940. Deuxième Bureau und strategischer Überraschung,” Militärgeschichtliche Mitteilungen, no. 2 (1967,): 13–92. 5. Liss, “Tätigkeit,” 271. 6. Some dispute continues on when French intelligence located the German armored forces. Major Navarre of the SR German section claims that a Luxembourg gendarme reported before midnight on 9 May that Panzer Group von Kleist had crossed the frontier. Evidently Major Baril of Georges’s Northeast Front GHQ became convinced on 11 May that the German center of gravity would roll through the Ardennes. Those bulletins reached decision-makers only with a lag. A subsidiary difficulty derived from the fact that the 2ème Bureau suspected the existence of twelve rather than ten Panzer divisions. The latter two were still embryonic, however, and the tanks designated for them ended up as reserves for the other ten, most of which failed to attain full complement. See Navarre, Le Service, 110–113, and François Delpla, ed., Les papiers secrets du Général Doumenc, 1939–1940 (Paris: Orban, 1991), 200–201. 7. See M.I.3 report to deputy director of military intelligence on conversation with Gauché, “Estimated Strength of the German Army,” 15 June 1938, WO 190/634, National Archives of Great Britain (hereafter PRO). The M.I.3 skeptics could not believe that the Germans could field in excess of 49 regular divisions four years hence, with 110 more in the reserve and the Landwehr. As it turned out, the Germans had 259 divisions under arms at the end of 1942 and 280 in February 1943—just slightly fewer than Gauché had prophesied.



Seeking a Scapegoat

117

8. For specifics, see Bradford A. Lee, “Strategy, Arms, and the Collapse of France,” in Richard Langhorne, ed., Diplomacy and Intelligence during the Second World War (Cambridge: Cambridge University Press, 1985), 53. 9. On the development of Gamelin’s preoccupations as he wrestled with the strategic issues after the Belgians returned to neutrality in 1936, see Les relations militaires franco-belges, de mars 1936 au 10 mai 1940, ed. Jean Vanwelkenhuysen et al. (Paris: CNRS, 1968); also Martin S. Alexander, The Republic in Danger: General Maurice Gamelin and the Politics of French Defence, 1933–1940 (New York: Cambridge University Press, 1992). WO 106/1685 (General Strategy); WO 106/1687 (prewar discussions with Gamelin); WO 106/1776 (conversations with Daladier and Gamelin), all Directorate of Military Operations files at PRO, provide a useful sequence of documents. 10. On the horror of the World War I experience in the devastated districts, see Richard Cobb, French and Germans, Germans and French: A Personal Interpretation of France Under Two Occupations (Hanover, NH: University Press of New England, 1983). Judith M. Hughes, To the Maginot Line: The Politics of French Military Preparation in the 1920’s (Cambridge, MA: Harvard University Press, 1971), demonstrates that the military’s adoption of a defensive strategy reflected the deepest yearnings of the French people after the demographic holocaust. Gamelin justifies the logic of a prepared defense on the Namur-Antwerp line in Servir, vol. 1: Les armées françaises de 1940 (Paris: Plon, 1946), 83–111. 11. Paillole, Notre espion chez Hitler; J. Vanwelkenhuysen, “Die Niederlände und der ‘Alarm’ im Januar 1940,” Vierteljahrshefte für Zeitgeschichte 8, no. 1 (1960): 17–36; Vanwelkenhuysen, “Die Krise vom Januar 1940,” Wehrwissenschaftliche Rundschau 9 (1955): 66–90. 12. Les relations militaires Franco-Belges, 102–106; Karl-Heinz Frieser, BlitzkriegLegende. Der Westfeldzug (Munich: Oldenbourg, 1996), 75–76, 102–103. Neither General Raoul van Overstraeten nor King Leopold III had ever looked on France with benevolence: see the former’s Au service de la Belgique: Dans l’étau (Paris: Plon, 1960); and Jean Stengers, Léopold III et le gouvernement: Les deux politiques belges de 1940 (Brussels: Racine, 2002). 13. Gauché, Deuxième Bureau, 232. The best air reconnaissance about the Siegfried Line came from a volunteer recruited by the SR field office in Belfort who flew civilian monoplanes. See Pierre Croissant, L’espion de la ligne Siegfried: Armand Chouffet, photographe aérien. Le renseignement français en Suisse (Paris: Lavauzelle, 2005). 14. Frieser, Blitzkrieg-Legende, 71–135. 15. Liss, “Tätigkeit,” 278. 16. Martin Alexander, “Did the Deuxième Bureau Work? The Role of Intelligence in French Defense Policy and Strategy, 1919–1939,” Intelligence and National Security 6 (1991): 293–333, esp. 303, 311; further elucidation in Alexander, Republic in Danger. 17. Douglas Porch, “French Military Intelligence and the Fall of France,” Intelligence and National Security 4 (1989): 28–58; Porch, The French Secret Services (New York: Farrar, Straus & Giroux, 1995), 163–173. 18. Forcade, “Le Renseignement face à l’Allemagne,” esp. 132–134, 137–140.

118

Seeking a Scapegoat

19. Peter Jackson, France and the Nazi Menace: Intelligence and Policy Making, 1934– 1939 (Oxford: Oxford University Press, 2000). 20. Ibid., 388. 21. Admittedly, controversy on this question persists. Adam Tooze’s highly regarded recent synthesis delineates the disjointed nature of the German armaments buildup, but does not place its weaknesses in comparative perspective. See The Wages of Destruction: The Making and Breaking of the Nazi Economy (London: Allen Lane, 2006), 326–395. 22. Ernest R. May, Strange Victory: Hitler’s Conquest of France (New York: Hill & Wang, 2000), quotations from 448–464. 23. Ibid., 20–21, 240–253. Tippelskirch’s later recollection that the French masses felt “saved” (erlöst) by the rapid victory of “Monsieur le soldat allemand” prompts doubt whether he had the skeptical temperament and cultural awareness that an intelligence chief should possess. Still, biography is beside the point. Kurt von Tippelskirch, Der Weltkrieg (Bonn: Athenäum, 1954), 94. 24. Porch, French Secret Service, 158–162; Alexander, Gamelin, 99–100. Gamelin evidently split GHQ because he and Daladier did not want Georges, who stood higher in the estimation of Finance Minister Reynaud, controlling promotions. Georges complained that influence determined advancement in the higher ranks. “If all these politicians would just f*** off,” he said disgustedly, “things would go along much better.” Élisabeth du Réau, “Haut commandant et pouvoir politique,” in Les armées françaises pendant la Seconde guerre mondiale, 1939–1945 (Paris: École nationale supérieure de techniques avancées, 1985), 67–82, scatological quotation on 78; also Paul de Villelume, Journal d’une défaite, août 1939–juin 1940 (Paris: Fayard, 1976), 192 (16 February 1940). 25. Réau, “Haut commandant,” 78–82. 26. Note Roland de Margerie’s unflattering description of the daily routine at General Headquarters during the drôle de guerre in Margerie, Journal 1939–1940 (Paris: Grasset, 2010), 64–65. 27. Forcade, “Le Renseignement face à l’Allemagne,” citing numerous sources; further details in Paillole, Services spéciaux, 173–199; Gauché, Deuxième Bureau, 214–228; Porch, French Secret Services, 171–173; Navarre, Service de renseignements, 113; Bertrand, Enigma, 79–80. 28. “Note concernant la conduite générale de la guerre,” 8 September 1939, Documents diplomatiques français (3 sept.–31 déc. 1939) (Brussels: Peter Lang, 2002), no. 26. Gamelin had begun pushing this theory fifteen months earlier: see Ambassador Bullitt to President Roosevelt, 13 June 1938, in William C. Bullitt Papers, Yale University, box 71/1794. 29. Martin van Crefeld, Fighting Power: German and U.S. Army Performance, 1939– 1945 (Westport, CT: Greenwood, 1982), 3–41, 163–175; van Crefeld, The Culture of War (New York: Presidio, 2008), 353–374; Trevor N. Dupuy, A Genius for War: The German Army and General Staff, 1807–1945 (Englewood Cliffs, NJ: Prentice-Hall, 1977). Even the BEF commander, General Lord Gort, reporting to the cabinet after the Dunkirk evacuation, ruefully called German generalship “extremely good,” junior leadership



Seeking a Scapegoat

119

“first rate,” and the foe’s individual battlefield performance “remarkable.” W.M. (40) 151, Conf. Annex, 1 June 1940, PRO. 30. MacGregor Knox and Williamson Murray, The Dynamics of Military Revolution, 1300–2050 (New York: Cambridge University Press, 2007), 169, 185, cite the Duke of Wellington’s dictum. On the heinous consequences of militarism, see Isabel V. Hull, Absolute Destruction: Military Culture and the Practices of War in Imperial Germany (Ithaca, NY: Cornell University Press, 2005). 31. Pierre Lacoste, ed., Le renseignement à la française (Paris: Economica, 1998). 32. Éugène Vaillé, Le cabinet noir (Paris: Presses universitaires de France, 1950); Gilles Perrault, Le Secret du roi, 3 vols. (Paris: Fayard, 1992–96); Olivier Forcade, “Considération sur le renseignement, la défense nationale et l’État secret en France aux XIXe et XXe siècles,” Revue historique des armées 247 (2007): 4–12; Christopher Andrew, “France and the German Menace,” in Ernest R. May, ed., Knowing One’s Enemies: Intelligence Assessment Before the Two World Wars (Princeton, NJ: Princeton University Press, 1984), 127–149; Andrew, “Déchiffrement et diplomatie. Le Cabinet noir du Quai d’Orsay sous la IIIe République,” Relations internationales 5 (1976), 37–64. 33. Jean-Jacques Becker, Le Carnet B: Les pouvoirs publics et l’antimilitarisme avant la guerre de 1914 (Paris: Klincksieck, 1973). 34. Maurice Vaïsse, “L’évolution de la function d’attaché militaire en France au XXe siècle,” Relations internationales 32 (1982): 507–524. 35. A. Raffalovitch, ed., L’abominable vénalité de la presse (Paris: Librairie du travail, 1931). 36. David Paull Nickles, Under the Wire: How the Telegraph Changed Diplomacy (Cambridge, MA: Harvard University Press, 2003). 37. Gérald Arboit, “L’émergence d’une cryptographie militaire en France,” Centre français de recherche sur le renseignement, Note historique no. 15, http://www.cf2r. org/fr/notes-historiques/lemergence-dune-cryptographie-militaire-en-france.php. 38. Andrew, “France and the German Menace,” in May, Knowing One’s Enemies, 127–149. 39. Admittedly, to speak of a unique Schlieffen Plan involves simplification. German strategy continued to evolve between 1905 and 1914, posing ongoing problems for French intelligence. See Terence Zuber, Inventing the Schlieffen Plan: German War Planning, 1871–1914 (New York: Oxford University Press, 2002). 40. Jan Karl Tanenbaum, “French Estimates of Germany’s Operational War Plans,” in May, Knowing One’s Enemies, 150–171. 41. Paul Paillole, “Les Services spéciaux de la Défense nationale pendant la guerre 1914–1918,” Anciens des Services spéciaux de la Défense nationale, Bulletin no. 59 (1968), http://www.aassdn.org/hsvEXhis02.html; Sophie de Lastours, 1914–1918: La France gagne la guerre des codes secrets (Paris: Tallandier, 1998). 42. Andrew, “France and the German Menace,” 130, 144–145; Porch, French Secret Services, 55–114; David Kahn, The Codebreakers: The Story of Secret Writing (New York: Macmillan, 1967), esp. 299–347; Edward Berenson, The Trial of Madame Caillaux

120

Seeking a Scapegoat

(Berkeley: University of California Press, 1992); Raphaële Ulrich-Pier, René Massigli: Une vie de diplomate (Paris: Peter Lang, 2006), 1: 37–53. 43. Malvy successfully fought a treason charge in a highly politicized trial in 1918, but he was convicted of criminal negligence. Jean-Yves Le Naour, L’affaire Malvy: Le Dreyfus de la Grande guerre (Paris: Hachette, 2007). 44. Michael B. Miller, Shanghai on the Metro: Spies, Intrigue, and the French Between the Wars (Berkeley: University of California Press, 1994). 45. Direction de la sûreté générale, “Notes Jean, 1918–1936,” F7/12951–12961, Archives nationales. 46. Paul Jankowski, Shades of Indignation: Political Scandals in France, Past and Present (New York: Berghahn, 2008). 47. Guy Schlesser, “Le Contre-Espionnage entre 1936 et 1940,” Anciens des Services spéciaux de la Défense nationale, Bulletin no. 9, http://www.aassdn.org/xlda10091. htm. 48. See extensive treatment based on French materials repatriated from Russia in Forcade, La République secrète, chaps. 5 and 7; also Annie Kriegel and Stéphane Courtois, Eugen Fried: Le grand secret du PCF (Paris: Seuil, 1997), esp. 118–287, and Thierry Wolton, Le grand recrutement (Paris: Grasset, 1993), who have used Soviet archives. Wolton shows in La France sous influence (Paris: Grasset, 1997) that the USSR built upon the Communist Party role in the post-1941 Resistance to increase its penetration of the state apparatus after the Liberation. 49. Georges-Henri Soutou, ed., Recherches sur la France et le problème des nationalités pendant la Première guerre mondiale (Paris: Sorbonne, 1995); Soutou, “Un exemple d’influence. Le renseignement français et le problème des nationalités: le cas de l’Office central des Nationalités,” in Lacoste, Le renseignement, 127–138. 50. Schlesser, “Le Contre-Espionnage entre 1936 et 1940.” 51. Wolton, Le grand recrutement, 209; Wolton, La France sous influence, 165; Porch, French Secret Service, 134; also Roger Faligot and Rémi Kauffer, Histoire mondiale du renseignement, 1870–1939 (Paris: Laffont, 1993). Goronwy Rees, who had fallen under Burgess’s spell, provides a lurid description of Pfeiffer’s “hair-raising” exploits, but the French security services remained indifferent. See Rees’s Sketches in Autobiography, ed. John Harris (Cardiff: University of Wales Press, 2001), 248, 256–257. 52. Anthony Adamthwaite, “French Military Intelligence and the Coming of War, 1935–1939,” in Christopher Andrew and Jeremy Noakes, eds., Intelligence and International Relations, 1900–1945 (Exeter, UK: Exeter University, 1987), 202. 53. “Note du Département a/s Dr. Liebing,” 30 March 1931, Ministère des affaires étrangères, Paris, série Z (Europe)—Rive Gauche du Rhin 45. 54. Stephen A. Schuker, “Bayern und der rheinische Separatismus 1923–1924,” Jahrbuch des Historischen Kollegs 1997 (Munich: Oldenbourg, 1997), 75–111. 55. Jackson, France and the Nazi Menace, 30–32, 39, 368. 56. Ibid., 21. 57. Stephen A. Schuker, The End of French Predominance in Europe (Chapel Hill: University of North Carolina Press, 1976), 197; Keith Jeffery and Alan Sharp, “Lord



Seeking a Scapegoat

121

Curzon and Secret Intelligence,” in Andrew and Noakes, Intelligence and International Relations, 120–121. Astoundingly, British code breakers decrypted some fifty important French embassy telegrams per month for twenty years without the French catching on. See Government Code and Cypher School, “Decrypts of Intercepted Diplomatic Communications,” HW 12/4–259, PRO. 58. Porch, French Secret Service, 153. 59. Extensive discussion in Paillole, Services spéciaux; Navarre, Le Service; also Jackson, France and the Nazi Menace, 21; Adamthwaite, “French Military Intelligence,” 203. 60. Gauché, Deuxième Bureau, 96–97, 180–182. 61. Arnaud Tessier, “Le général Vuillemin. Un haut responsable militaire face au danger allemand,” Revue historique des armées 2 (1987): 105–126; transcripts of Comité de guerre meetings in Thierry Sarmant and Ségolène Garçon, eds., Gouvernement et haut commandement au déclin de la IIIe République (Paris: Comité des travaux historiques et scientifiques, 2009); see also Vuillemin’s memoranda between 26 September 1938 and 26 August 1939 in Fonds Édouard Daladier, 4DA5 Dr 7 sdr b, Fondation nationale des sciences politiques. 62. Jean Stengers, “Enigma, the French, the Poles and the British, 1931–1940,” in Christopher Andrew and David Dilks, eds., The Missing Dimension: Governments and Intelligence in the Twentieth Century (London: Macmillan, 1984), 126–137; Jan S. Ciechanowski and Jacek Tebinka, “Cryptographic Cooperation—Enigma,” in Tessa Stirling et al., Intelligence Cooperation Between Poland and Great Britain During World War II (London: Valentine Mitchell, 2005), 1:443–462. 63. Julian Jackson, The Fall of France: The Nazi Invasion of 1940 (New York: Oxford University Press, 2003), 219. 64. For recapitulation of the literature, see John Cairns,” Recent Historians and the ‘Strange Defeat’ of 1940,” Journal of Modern History 46, no. 1 (March 1974): 60–85; Peter Jackson, “Recent Journeys Along the Road Back to France, 1940,” Historical Journal 39, no. 2 (June 1996), 497–510; and Joel Blatt, ed., The French Defeat of 1940: Reassessments (Providence, RI: Berghahn, 1998). 65. Marc Bloch, Strange Defeat: A Statement of Evidence Written in 1940 (New York: Norton, 1968), 25, 126–176. 66. Henry Dutailly carries the criticism a step farther. He contends that building the Maginot Line up to the hinge of the Meuse only made sense within a defensiveoffensive strategy. It could provide shelter for mobilization, but implicitly required that France take the offensive in Belgium. See his “Faiblesses et potentialités de l’armée de terre (1939–1940),” in Les armées françaises pendant la Seconde guerre mondiale (cited n. 24 above), 23–35. 67. Lee, “Strategy, Arms,” 43–65. 68. Robert Frank [Frankenstein], Le prix du réarmement français (1935–1939) (Paris: Sorbonne, 1982). Frank demonstrates that in 1939, Germany devoted 30 percent of national income to military expenses, compared with 23 percent in France (35). Daladier and Finance Minister Paul Reynaud balanced on the horns of an unenviable

122

Seeking a Scapegoat

dilemma: they needed a “liberal mobilization” to reassure businessmen, yet that ruled out creation of state-sponsored industry with the scale and scope to introduce transatlantic methods of mass production (71–88). 69. For contrasting approaches to this phenomenon, one emphasizing failed governance, the other social disintegration, see J.-B. Duroselle, La décadence, 1932–1939 (Paris: Imprimerie nationale, 1979), and Eugen Weber, The Hollow Years: France in the 1930s (New York: Norton, 1994). 70. Rudolph Binion, Defeated Leaders: The Political Fate of Caillaux, Jouvenel, and Tardieu (New York: Columbia University Press, 1960); François Monnet, Refaire la République: André Tardieu, une dérive réactionnaire, 1876–1945 (Paris: Fayard, 1993). 71. See Stanley Hoffmann, ed., In Search of France: The Economy, Society, and Political System in the Twentieth Century (New York: Harper & Row, 1965), 1–117; Georges Gurvich, “Social Structure of Pre-War France,” American Journal of Sociology 48, no. 5 (March 1943): 535–554. 72. Paul Jankowski, Stavisky: A Confidence Man in the Republic of Virtue (Ithaca, NY: Cornell University Press, 2002); Jankowski, Shades of Indignation. During the 1934 Stavisky scandal, 160 deputies were said to have taken bribes, a not implausible estimate given the absence of other income streams. Neville to Hilda Chamberlain, 3 February 1934, in The Neville Chamberlain Diary Letters, ed. Robert Self (Aldershot, UK: Ashgate, 2005), 4: 52. 73. Auguste Soulier, L’instabilité ministérielle sous la Troisième République, 1871–1938 (Paris: Sirey, 1939); Jacques Ollé-Laprune, La stabilité des ministres sous la Troisième République, 1879–1940 (Paris: Librairie générale de droit et de jurisprudence, 1962). 74. Duroselle, La Décadence, 203–209; quotations from 204–205; Weber, Hollow Years, 127–138. 75. Alfred Kupferman, “Diplomatie parallèle: Le rôle de la Ribbentrop-Dienststelle dans les tentatives d’action sur l’opinion française, 1934–1939,” Relations internationales, no. 3 (1974), 72–95. Journalists who took handouts from the Nazis did not prosper after 1944, but those who cultivated Soviet sources did. See, e.g., Denis Maréchal, Geneviève Tabouis: Les dernières nouvelles de demain (Paris: Nouveau monde, 2003). 76. Mona L. Siegel, The Moral Disarmament of France: Education, Pacifism, and Patriotism, 1914–1940 (Cambridge: Cambridge University Press, 2004); Norman Ingram, The Politics of Dissent: Pacifism in France 1919–1939 (Oxford: Clarendon Press, 1991). 77. Gamelin to minister of national defense, 18 May 1940, in his memoirs, Servir, vol. 3: La Guerre (septembre 1939–19 mai 1940) (Paris: Plon, 1947), 421–426; for the context, see Pierre Le Goyet, Le mystère Gamelin (Paris: Presses de la Cité, 1957), 332–337, 350–351. 78. Claus von Clausewitz, On War, ed. Michael Howard and Peter Paret (Princeton, NJ: Princeton University Press, 1984), 89. Perhaps lexicographers can determine why the editors translate wunderliche Dreifaltigkeit as “fascinating trinity.” 79. Martin van Creveld, The Transformation of War: The Most Radical Reinterpretation of Armed Conflict Since Clausewitz (New York: Free Press, 1991).



Seeking a Scapegoat

123

80. Richard K. Betts, Enemies of Intelligence: Knowledge and Power in American National Security (New York: Columbia University Press, 2008), esp. 19–52. 81. Michael I. Handel, War, Strategy, and Intelligence (London: Frank Cass, 1989); Robert Jervis, “Strategic Intelligence and Effective Policy,” in Stuart Farson, David Stafford, and Wesley Wark, eds., Security and Intelligence: New Perspectives for the 1990s (London: Frank Cass, 1991), 165–181. 82. Colin L. Dyer, Population and Society in Twentieth Century France (Sevenoaks, Kent, UK: Hodder & Stoughton, 1978). 83. Stephen A. Schuker, “Ambivalent Exile: Heinrich Brüning and America’s Good War,” in Christoph Buchheim et al., eds., Zerrissene Zwischenkriegszeit. Wirtschaftshistorische Beiträge (Baden-Baden: Nomos, 1994), 350; Heinrich August Winkler, Weimar, 1918–1933: Die Geschichte der ersten deutschen Demokratie (Munich: C. H. Beck, 1993), 169–170; Hans W. Gatzke, Stresemann and the Rearmament of Germany (Baltimore, 1954), 113–115. 84. General Charles Nollet, Une expérience de désarmament. Cinq ans de contrôle militaire en Allemagne (Paris: Gallimard, 1932); Forcade, La République secrète, chap. 4; J. H. Morgan, Assize of Arms: The Disarmament of Germany and her Rearmament (1919–1939) (New York: Oxford University Press, 1946); Michael Salewski, Entwaffnung und Militärkontrolle in Deutschland 1919–1927 (Munich: Oldenbourg, 1966). Cf. Richard J. Shuster, German Disarmament after World War I: The Diplomacy of International Arms Inspection, 1920–1931 (London: Routledge, 2006), who sees the operation as successful. 85. Peter Jackson and Joseph Maiolo, “Strategic Intelligence, Counter-Intelligence, and Alliance Diplomacy in Anglo-French Relations before the Second World War,” Militärgeschichtliche Zeitschrift 65, no. 2 (2006): 426. 86. Officers assigned to train the eastern allies naturally took their mission seriously. See, e.g., Richard F. Crane, A French Conscience in Prague: Louis Eugène Faucher and the Abandonment of Czechoslovakia (Boulder, CO: East European Monographs, distributed by Columbia University Press, 1996). 87. Laurence Badel, “Les réalisations concrètes,” in Jacques Bariéty, ed., Aristide Briand, la Société des nations et l’Europe, 1919–1932 (Strasbourg: Presses universitaires de Strasbourg, 2007), 298–99; for Briand’s working habits, consult Jules Laroche, Au Quai-d’Orsay avec Briand et Poincaré 1913–1926 (Paris: Hachette, 1957) . 88. Stephen Schuker, “Les États-Unis, la France, et l’Europe, 1919–1932,” in Bariéty, ed., Briand, 383–396. 89. J. E. Kaufmann and H. W. Kaufmann, Fortress France: The Maginot Line and French Defenses in World War II (Westport, CT: Praeger, 2006). 90. Daladier before the Comité permanent de la défense nationale, 26 June 1936, cited by du Réau, “Haut commandant,” 68. 91. Stephen A. Schuker, “France and the Remilitarization of the Rhineland, 1936,” French Historical Studies 14 (Spring 1986): 299–338; Lee, “Strategy, Arms,” 63–67; Frédéric Guelton, ed., Le “ journal” du général Weygand, 1929–1935 (Montpellier: Université Montpellier III, 1998).

124

Seeking a Scapegoat

92. Castellan, Le réarmement clandestin du Reich; Edward W. Bennett, German Rearmament and the West, 1932–1933 (Princeton, NJ: Princeton University Press, 1979). 93. “Principales réalisations du président. Étude demandée par le secrétariat particulier de M. Daladier,” 5 February 1937, 5N 581 Dr. 3, SHAT. 94. Frank, Le prix du réarmement, 34–36. 95. Schuker, “Remilitarization”; Henry Dutailly, Les problèmes de l’armée de terre française, 1939–1940 (Paris: Imprimerie nationale, 1980), 93–100; General P.-É. Tournoux, Haut commandement, gouvernement et défense des frontières du Nord et de l’Est, 1919–1939 (Paris: Nouvelle éditions latines, 1960), 337; Robert J. Young, In Command of France: French Foreign Policy and Military Planning, 1933–1940 (Cambridge, MA: Harvard University Press, 1978), 76–129. 96. Jackson, France and the Nazi Menace, 245–297; Porch, French Secret Service, 146; also Daladier’s “notes manuscrites” in 1DA7 Dr 6 sdr a, and “Documents généraux, Munich,” in 2DA1 Dr 3, Fonds Daladier. 97. Patrice Buffotot, “The French High Command and the Franco-Soviet Alliance, 1933–1939,” Journal of Strategic Studies 5 (December 1982): 46–59; Dutailly, Les problèmes, 45–50, 56–58; Delpla, ed., Les papiers secrets du Général Doumenc, 35–133. 98. Charles Bohlen to Ambassador Bullitt, n.d. [August 1939], William C. Bullitt Papers, Yale University, box 114/464. 99. Experts don’t agree whether Stalin first gave this explanation at a secret Politburo meeting on 19 August 1939, before signing of the Molotov-Ribbentrop pact. The latest evidence suggests that the copy of the speech found in the Russian archives may be authentic. In any event, Stalin said much the same thing to the Comintern chief Georgii Dimitrov on 7 September, and a Commissariat for Foreign Affairs official echoed the line when briefing a Czech delegation on 26 October 1939. See Albert Weeks, Stalin’s Other War, 1939–1941 (Lanham MD: Rowman & Littlefield, 2002), 171– 173; Ivo Banac, ed., The Diary of Georgi Dimitrov (New Haven, CT: Yale University Press, 2003), 115–116; V. L. Doroshenko et al., “Ne mif: Rech’ Stalina 19 avgusta 1939 goda,” Voprosy istorii, 2005, no. 8: 3–20; and Pravda Viktora Suvorova 3, ed. Dmitri Khmel’nitskii (Moscow: IAuza, 2007), 375–379. 100. Buffotot, “Franco-Soviet Alliance,” 549; Élisabeth du Réau, “Le renseignement et l’élaboration de la décision diplomatique et militaire: Le cas de la France, 1933–1940,” Relations internationales, no. 78 (1994): 241–260, esp. 254–256. 101. Fear and discouragement penetrated so far down the ranks that in late 1936 the head of Air Force intelligence said plaintively to a German embassy air adviser, “Do what you want in the East, but just leave us alone.” Hans Ritter memorandum, forwarded by Capt. Malcolm Christie to Robert Vansittart, 9 August 1940, FO 371/24312, PRO. 102. Sabine Jansen, Pierre Cot: Un antifasciste radical (Paris: Fayard, 1992), 138– 346; Thierry Vivier, La politique aéronautique militaire de la France, 1933–1939 (Paris: L’Harmattan, 1997), 105–413; Pierre Péan, Vie et morts de Jean Moulin (Paris: Fayard, 1998), 51–390; Herrick Chapman, State Capitalism and Working-Class Radicalism in the French Aircraft Industry (Berkeley: University of California Press, 1991), 71–147.



Seeking a Scapegoat

125

103. See the many distressing encounters of Foreign Office and Air Ministry officials with the French over the period from 1937 to 1939 in FO 371/20694, 21621, 22915–916; also Secretary of State for Air Sir Kingsley Wood’s frustrating showdown with La Chambre on 4 March 1939, AIR 19/59, PRO. 104. John McVickar Haight, American Aid to France, 1938–1940 (New York: Atheneum, 1970). 105. The Armée de l’Air became so disorganized that it stopped keeping records in early June; hence minor discrepancies occur in the numbers. I follow Pierre Buffotot and Jacques Ogier, “L’Armée de l’Air pendant la bataille de France: Essai de bilan numérique,” Revue historique des armées, no. 3 (1975): 88–117. See Patrick Facon, L’Armée de l’Air dans la tourmente: La bataille de France, 1939–1940 (Paris: Economica, 1997); Charles Christienne and Pierre Lissarrague, A History of French Military Aviation (Washington, DC: Smithsonian Institution Press, 1986), 267–344; Vivier, Politique aéronautique, 597–617. 106. Lee, “Arms, Strategy,” 60–63; Robert Nayberg, “La problématique du ravitaillement de la France en carburant dans l’Entre-deux-guerres,” Revue historique des armées, no. 4 (1979): 5–23; Gérard Saint-Martin, L’arme blindée française, mai–juin 1940 (Paris: Economica 1998), 1–108. Porch, French Secret Service, 146, discusses translation of Guderian’s book Achtung – Panzer! (Stuttgart: Union Deutsche Verlagsgesellschaft, 1937). 107. See quantitative comparisons of the respective forces in Saint-Martin, L’arme blindée, 317–333. 108. Pascal Griset, “Les industries d’armament: L’exemple des transmissions,” in Levisse-Touzé, ed., La campagne de 1940, 330–345. 109. Rémi Baudouï, Raoul Dautry: Le technocrate de la République (Paris: Balland, 1992), 187–220. 110. Talbot Imlay, “Mind the Gap: The Perception and Reality of Communist Sabotage of French War Production during the Phoney War, 1939–1940,” Past & Present 189 (2005): 179–224; Philippe Buton, “Le Parti, la guerre et la revolution, 1939–40,” Communisme 32–34 (1993): 41–68. 111. Gamelin conversation with the British CIGS, 6 October 1939, in WO 106/1684, PRO. 112. Bruno Chaix, “Les Ardennes en mai 1940: Mythes et réalités,” Revue historique des Armées, no. 2 (2000): 3–12; Pierre Lyet, La bataille de France (Paris: Payot, 1947); Lyet, “Souvenirs et témoignages, 1939–1940,” Revue historique de l’Armée 17, no. 1 (1961): 81–105; Lyet, “À propos de Sedan 1940,” Revue historique de l’Armée 18, no. 4 (1962): 89–109; Claude Paillat, Dossiers secrets de la France contemporaine, vol. 4, pt. 2, La guerre immobile (Paris, 1984), 328–386; P.-É. Tournoux, “Pouvait-on prévoir l’attaque allemande des Ardennes de mai 1940?” Revue historique de l’Armée, no. 2 (1971), 130– 141; Jean Vidalenc, “Les divisions de série ‘B’ dans l’armée française pendant la campagne de France 1939–1940,” Revue historique des armées, no. 4 (1980): 106–126. 113. Gamelin never explained the logic behind sticking with the Breda plan, which required denuding his strategic reserve, after learning in mid-April that the Dutch

126

Seeking a Scapegoat

planned to evacuate North Brabant. The idea of conserving a base for a counterattack against the Reich in the remote future hardly suffices. The Dutch forces he hoped to relieve amounted to little more than a militia. As Field Marshal Bernard Montgomery would later say, the Dutch had “no army of any sort that could begin to fight in the field even against naked savages” (Montgomery to CIGS, 13 August 1950, PREM 8/1154, PRO). On the vagaries of the decision-making process, see Don W. Alexander, “Repercussions of the Breda Variant,” French Historical Studies 8 (1974): 459–88. 114. Villelume, Journal d’une défaite; Raymond de Sainte-Suzanne, Une politique étrangère: Le Quai d’Orsay et Saint-John Perse à l’épreuve d’un regard, novembre 1938– juin 1940 (Paris: Viviane Hamy, 2000); Margerie, Journal 1939–1940. 115. Bullitt to R. Walton Moore, 18 April 1940, William C. Bullitt Papers, Yale University, box 58/1442. 116. Élisabeth du Réau, Édouard Daladier (Paris: Fayard, 1993), 355–400; Jean-Pierre Guichard, Paul Reynaud: Un homme d’État dans la tourmente, septembre 1939–juin 1940 (Paris: L’Harmattan, 2008). 117. Villelume, Journal d’une défaite, 47 (28 September 1939); Bullitt to Roosevelt, 16 September 1939, William C. Bullitt Papers, Yale University, box 72/1804. 118. Foreign Relations of the United States, 1939 (Washington, D.C.: GPO, 1950), 1: 460. 119. Georges-Henri Soutou, “Introduction,” in Levisse-Touzé, ed., La campagne de 1940, 21–37. 120. Édouard Bonnefous, Histoire politique de la Troisième République, vol. 7: La course vers l’abîme: La fin de la IIIe République (1938–1940) (Paris: Presses universitaires de France, 1967), 7: 148–155; The Private Diaries of Paul Baudouin: Undersecretary of State and Secretary of the War Cabinet in M. Paul Reynaud’s Administration, April 1940 to May 1940; Foreign Minister Under Marshal Pétain, June 1940 to January 1941, trans. Sir Charles Petrie (London: Eyre & Spottiswoode, 1948), chap. 3. 121. See Villelume, Journal, 196, 331 (17 February and 11 May 1940), on Gamelin’s refusal to share operational details with the civilians; Delpla, ed., Les papiers secrets du Général Doumenc, 209–210, for Gamelin’s comment on Reynaud. 122. For Reynaud’s tortuous struggle to dislodge the sinuous Léger, see Renaud Metz, Alexis Léger dit Saint-John Perse (Paris: Flammarion, 2008), 572–617. 123. See the many examples provided by Roland de Margerie, head of his civil cabinet from March through June, in Margerie, Journal 1939–1940, esp. 235–45. 124. Jacques Weygand, Mon père Weygand (Paris: Flammarion, 1970), 260–281, 340–373. 125. Villelume, Journal, 333 (12 May 1940); cf. Navarre, Service de renseignements, 113. 126. Hugh Dalton memorandum, 28 June 1940, CAB 127/204, PRO. Others also used the termite metaphor. See e.g., Francois de Mauriac, quoted July 1940 in FO 371/24312. 127. Colonel Passy [André Dewavrin], Mémoires du chef des services secrets de la France Libre, ed. J.-L. Crémieux-Brilhac (Paris: Odile Jacob, 2000).



Seeking a Scapegoat

127

128. Sébastien Laurent, “The Free French Intelligence Services: Intelligence and the Politics of Legitimacy,” Intelligence and National Security 15 (Winter 2000): 19–41; Pierre Péan, L’homme de l’ombre: Éléments d’enquête autour de Jacques Foccart (Paris: Fayard, 1990); Guy Birenbaum, Le cabinet noir: Au cœur du système Yves Bertrand (Paris: Les Arènes, 2008).

4

French Intelligence About the East, 1945–1968 Georges-Henri Soutou

F

or the international historian, what is important is not so much the nuts and bolts of intelligence, but what is told to the decisionmakers, and how they use it. France started in 1946 with a very ambitious, all-encompassing civilian secret service. It covered political and economic intelligence as well as military intelligence, as in the prewar era. This was the Service de documentation extérieure et de contre-espionnage, SDECE. But that service failed in around 1954 with the disaster of defeat at Dien Bien Phu in Indochina. The military staffs then discreetly regained control of much of the intelligence process. They realized that it was impossible to penetrate the Soviet system to assess Moscow’s intentions. Instead, they tried to develop a system of objective, early indicators for strategic early warning. The new system rested on both formal and informal cooperation between the several agencies concerned: the Deuxième Bureau of the General Staff, the SDECE, the Quai d’Orsay, and the Secrétariat général de la défense nationale (SGDN), which collated, analyzed and disseminated intelligence inside government. The same kind of cooperation could be found at embassy level, between military attachés, SDECE representatives, and diplomats. The Failure of Overambitious Intelligence The first “intelligence research plans” after 1946 were all-encompassing and global, which meant that they were completely beyond the means at the 128



French Intelligence About the East, 1945–1968

129

disposal of those responsible. This was in keeping with the ambitious plan of General Charles de Gaulle in 1943, when he created, under Jacques Soustelle, the DGER (Direction générale des études et recherches). It was meant to be a great intelligence service covering political and economic as well as military matters. Created in 1946, the new SDECE, put directly under the chief of the executive, was supposed to pursue that line. That ambitious concept of intelligence was translated into the first documents planning the general intelligence effort for all government agencies concerned. The first category was called the National Defense Intelligence Plan (Plan de renseignement de défense nationale). It was prepared at government level between the EMGDN (État-major général de la défense nationale, predecessor of the SGDN), under the prime minister, and the Combined General Staff of the Armed Forces (État-major combiné des forces armées) under the Ministry of Defense, particularly the various “2èmes Bureaux” (Foreign Armies). They were very ambitious and all-encompassing documents, covering politics, social, and economic problems, military matters, for all major countries and regions of the world. The first appeared in 1946; the second, in 1948.1 In 1946, there were two main areas of concern still pre–Cold War: the cohesion of the “Union française,” both inside and against foreign encroachments; and a resurrection of the German threat. But quickly the onset of Cold War, the Brussels Pact, and the preparation for the Atlantic Alliance led progressively to more realistic planning. For instance, the National Defense Intelligence Plan for 1948 took the Cold War into account. The objectives were to assess the possibility of world war in the near future, the possibility of Soviet moves toward Europe and Africa, and the Middle East (toward the oil fields and the Suez Canal), through local uprisings, supported or not by Soviet military force. But they also turned a keen eye on Britain and the United States: where exactly would they decide to defend themselves (a moot point in 1948)? Would they expect to occupy parts of the Union française? The main thrust of the document was thus still the problem of upholding French sovereignty against all-comers. A new National Defense Intelligence Plan was devised in 1952. It focused more clearly on the Soviet Union as the presumed adversary, “aiming at world hegemony through the combination of its ideological and political dynamism and of its economic and military power.” Yet the British and Americans, allies, and the neutral powers in Europe were still extensively covered.

130

French Intelligence About the East, 1945–1968

The huge, encyclopedic plan, with ten annexes, had some interesting features. Four threats were highlighted: “American imperialism,” “British imperialism,” “Soviet imperialism,” and “Chinese imperialism.” The British and American push for the reconstruction of Germany was also targeted, even though it had been suggested that they stop gathering intelligence about their allies, and content themselves with open sources.2 All kinds of information were also sought: political, economic, military, social.3 The Soviet part of the 1952 plan was comprehensive and constituted the main orientation of the intelligence effort. It covered the USSR’s politics, ideology, economy, and military; its exploitation of conflicting interests between Western countries; of the latter’s domestic problems; and Soviet geopolitical priorities (Europe, the Middle East with its oil, and Asia).4 That first period does not seem to have been a great success. For instance, on 6 April 1945, the DGER produced a report on the Polish situation that was very supportive of the Polish Communist leaders.5 Generally speaking, the French secret service seems to have had very little awareness of what was going on in Eastern Europe. The only interesting reports in 1945 and 1946 describe the views of the Vatican and probably originated from exchanges with Cardinal Éugène Tisserant, who was in charge of special information at the Holy See.6 The SDECE quickly built up networks in Eastern Europe. But we have little knowledge of their effectiveness.7 A long (60-page) study in May 1947, entitled “Social and cultural components of world situation,” was apparently a major effort disseminated at a high level in the government. It was an indigestible compilation concentrating on the trade unions and religious matters. There were some gems, for instance, about the influence of the dollar on the Vatican and its pro-German sentiment. The conclusion was that the probable way out of the East-West confrontation would be the adoption, worldwide, of some kind of syncretic ideology akin to the British Labour Party’s program.8 Another comprehensive report, on 20 August 20 1948, concerning “the evolution of world political situation” during July, was marginally more significant. But it represented only the “opinions of specialists and well-placed observers.” It concluded, for instance, that Washington, paralyzed by the looming presidential elections, had decided to tread very gingerly during the Berlin crisis and was ready to discard the London agreements about Germany. The conclusion was that the United States was unable to take any major initiative. All that can be said is that those views were not prophetic.9



French Intelligence About the East, 1945–1968

131

The year 1948 produced several reports from the SDECE. They were usually circulated to the Presidency of the Republic, to the Presidency of the Council, to the Quai d’Orsay, the EMGDN (later SGDN), and the 2ème Bureau of the EMA (Army General Staff). This is the usual top distribution list for that kind of information. The reports covered the Cominform (Communist Information Bureau, which replaced the dissolved Comintern in 1947) and related matters, but were of little value, not resting on any clear hard evidence, and ascribing a mythical importance to the Cominform as an organization, even for the preparation of several regional groupings of people’s democracies and of sophisticated war plans.10 Quai d’Orsay studies were much more to the point.11 Given this dearth of hard information about the USSR, the French ambassador in London, René Massigli, was grateful when Sir William Strang, permanent undersecretary at the British Foreign Office, showed him an MI6 report on 29 May 1948 stating that the indicators of Soviet behavior suggested that Moscow was not contemplating a war in the short term (an important question mark in the West at the time). The evidence cited was: no restrictions on passenger traffic; no change in the rhythm of construction of railroads; continuing effort to improve the living conditions of the civilian population; the civilian part of the economy continued to work and develop in the same way; buildings that military administrations would want to keep in case of mobilization continued to be returned to the civilian sector.12 Absent real human information, such relatively obvious indicators would also be heavily relied upon by the French a few years later. As to signals interception, it should first be noted that after the war, the cryptographic systems of the French armed forces relied on Enigma-type machines, whereas the diplomatic service continued to use manual codes that were easily penetrated from all sides. From 1949 to the end of the 1950s, the Quai d’Orsay switched to the most advanced Enigma system, the German Admiralty’s Geheimschreiber with ten wheels. But that could be read by London, Washington, and Moscow. Only from 1960 on were more secure electronic systems introduced.13 The SDECE also conducted signals interception and decoding. Apparently they were able in 1948 to break the code of a least one satellite embassy.14 But the most amusing instance was a (likely as not human) interception by the Quai (or the SDECE) of a U.S. interception of a cable from the French embassy in Moscow (1 December 1948). It concerned indirect Soviet overtures

132

French Intelligence About the East, 1945–1968

to the French about the Berlin crisis. The Russians told the French that they could help stop the rebuilding of a German state if they did not align themselves with the British and the Americans. Having become aware of the U.S. interception, and certainly taking into account a probable Soviet one, the Quai used this conduit to convey to its ambassador in Moscow an affirmation of steel-hearted resolve that was far from the real doubts felt in Paris about the militant American stance in the crisis, but was evidently meant for American and Soviet consumption.15

The Limits of the SDECE The SDECE had been created as a great political and general intelligence service, more or less on the model of the British Secret Intelligence Service (SIS, or MI6). It did not reach that standard. It suffered grievous failures: for instance, in Poland in 1949, where two French diplomatic employees were arrested by the Poles and charged with being SDECE agents.16 Certainly, there were also achievements: those that can be documented concern diplomatic mail interceptions and Germany;17 but that was a far remove from the crucial and impenetrable Soviet target. East German economic material and foreign trade statistics can be found, including low-level documents originating from the Soviet military administration in Germany (SMAD). A 91-page report from 23 January 1950 on the economic and social evolution of the Soviet Zone from 1945 to 1949 was certainly not without value, for example.18 But that scarcely amounted to consistent, high-level penetration. That said, it should be kept in mind that access to the overall production of the SDECE is limited to instances where sporadic reports are found in, for instance, the Quai d’Orsay archives. And the presence in those archives of such documents diminishes even more after 1948, probably due to the much stricter secrecy rules and classification system for the circulation of documents introduced that year at the request of London and Washington, after the Brussels treaty and in preparation for the Atlantic alliance.19 There were indeed some good SDECE documents: such as a 150-page report on Soviet leader Nikita Khrushchev, drafted in preparation for the French leadership on the eve of Khrushchev’s trip to France in March 1960. It is a perceptive portrait, readable still today; but one of the best parts comes evidently from observations gathered by the Americans during the September



French Intelligence About the East, 1945–1968

133

1959 trip of the Soviet leader to the United States.20 One might also note a 60-page report dated 31 January 1958 on Soviet penetration of the Middle East, which stressed the acceleration and successes of Moscow’s policy in the region—much more to the point than the sometimes more optimistic Quai d’Orsay assessment.21 Nonetheless, the SDECE probably tried to cover too much, with too few resources. Of course, the wars in Indochina (1946–1954) and then Algeria (1954–1962) took priority and absorbed most of its efforts, to the detriment of the Soviet target.22 And it is difficult, of course, as we have seen, to trace information coming from the SDECE, and hence to evaluate its value today. That information was of two kinds: (a) Short, precise “intelligence notes” (notes or fiches de renseignement) about a specific event gathered from secret information: normally, these are not found in the archives; even at the time, they were usually not left with the addressees (at the Foreign Office, the General Staff, the Prime Minister’s Office at Matignon, the SGDN . . . ) but only shown to them for reading on the spot, and then taken away. (One rare source with frequent allusions to such information coming from the SDECE is the diary of Vincent Auriol, president of the Republic from 1947 to 1953.)23 A close reading of Foreign Office or General Staff or SGDN documents may sometimes give a discreet indication that a secret renseignement is being used or alluded to. That usually means that it came through SDECE, probably through such short notes. Of course, some such SDECE fiches can be found in the files: for example, a report on 28 June 1956 on a COMECON meeting in Berlin from 18 to 25 May, explaining how harmonization of the national economic plans was discussed (the document leads us to conclude that a well-placed Hungarian source had given the information). Another one from 16 July 16 1956, on the effects of destalinization on the coordination between Soviet bloc economies, rests on press accounts, but alludes also to a source, referred to of course only by a number.24 There were also intercepts of diplomatic mail, but they were just shown to the concerned official and taken away. Normally, they do not show up in accessible archives. But it may be surmised that there were probably very few, if any, intercepts from Soviet bloc diplomatic mail. (b) The second kind of documents emanating from SDECE were more developed synthetic reports, which are more frequently to be found in

134

French Intelligence About the East, 1945–1968

the files. For instance, there is a series of reports about the economic situation in Soviet bloc countries, from 1950 to 1955: particularly notable is a 73-page report about the satellite countries in the spring of 1950 and COMECON (the Soviet bloc Council for Mutual Economic Assistance). This is based both on open sources and secret intelligence and is more a series of vignettes on each different country than a real synthetic description, but the division of labor for the production of economic goods among those countries and the resurrection of armaments industries as early as 1949 are clearly documented. At least, compared to current French production at the time about those topics, it was a useful description. There is also an interesting report from the SDECE about a reorganization of COMECON in 1955, with a division of labor and systems for coordinating national plans.25 But those reports did not, of course, penetrate hard intelligence targets. Some valuable details can nonetheless be found: (a) During the Suez and Hungary crises in 1956, the General Staff was very critical of SDECE’s performance: it had been unable to provide any information about Soviet troop movements, even in satellite countries. And there were false rumors particularly concerning the overflight over Turkey by Soviet aircraft at the time of the Anglo-French landing at Port Said, which caused a considerable alarm in many quarters. The 2ème Bureau of the General Staff concluded that in sending this spurious information to others than the organs qualified for evaluating intelligence, the SDECE had contributed to the disinformation campaign launched by Moscow after the Franco-British landing.26 (b) There are other instances of very critical evaluation of SDECE reports, notably by the Centre d’évaluation du renseignement (CER), a branch of the SGDN’s Intelligence Department, which commented on a SDECE report of 2 June 1959 on Yugoslav foreign and domestic policy: “this note amounts only to a contact report with a Yugoslav personality, perhaps well situated in government and political circles, but whose sincerity and objectivity may be questioned.”27 There was a similarly negative reaction in the same quarter to another SDECE report in May 1973 about the purported liquidation of “Titoist heresy” by Moscow: “no value whatsoever.”28



French Intelligence About the East, 1945–1968

135

The Role of the Direction de la surveillance du territoire (DST) Based on the files, there would seem to have been more cooperation between the SDECE and other departments than is usually assumed. East European emigrants living in France sometimes supplied useful information, which could be relayed to the SDECE. And evidently targeting the French Communist Party (PCF) also produced material useful for fighting Soviet foreign intelligence. We know now that as early as 1952, the DST department of France’s National Police was able to trace Soviet payments to the PCF, and that the Post Office opened mail between PCF headquarters and local organizations.29

More Focused Military Intelligence Notwithstanding the 1946 reorganization of intelligence in a civilian framework, the military recovered its pre-1939 control of foreign intelligence after World War II. In addition to the national defense intelligence plans (and not to be confused with them), the intelligence sections (2èmes Bureaux) of the different Army, Navy and Air Force staffs and the Intelligence Division of the General Staff of the Armed Forces (EMFA) came up with specifically military “plans for the gathering of intelligence” (plans de recherche de renseignement). These were much more focused and, it would seem, more realistic. In December 1946, for example, a research plan was devised dealing with “The USSR as Anti-Colonial Power and the Muslim World,” which was very thorough with respect to all aspects of Soviet policy toward Islam, both domestic and foreign.30 In October 1948, the EMGDN admitted little reliable information about the Soviet order of battle and Soviet military strength: not even overall troop numbers and percentages of the three different types of divisions according to their state of readiness. In order to overcome such a major failure, it instead concentrated on indicators of an impending Soviet attack that could be more easily observed: communications, infrastructures, and industrial basis.31 The EMFA managed to achieve an increasingly accurate view of Soviet forces in the European theater, essentially through reports from military attachés in the Soviet bloc. It also had the benefit of reports from the Mission française de liaison with the Soviet command in Germany, based in Potsdam.

136

French Intelligence About the East, 1945–1968

Enjoying a special status, EMFA representatives could roam around more or less freely and, with their British and American counterparts, gathered probably 80 percent of the useful hard information about the Soviet forces in Germany.32 On top of that, evidently, there was electronic intelligence from listening to the transmissions of Soviet forces. Apparently (but the evidence is elusive and we cannot be sure) this kind of intelligence was based only on French material. “Soviet Forces in Occupied Europe” are described in a 30-page printed study from the EMFA Intelligence Division in January 1952, which includes annexes of great interest, even if its authors admit that they know less about the subject than they had about the Wehrmacht in 1939. This report noted the steady Soviet bloc progress since 1950 in both numbers (an increment of 15 percent in Germany) and quality (particularly of airfields, aviation, supply depots, and probably motor vehicles and heavy tanks). It assessed overall troop levels in East Germany and Czechoslovakia at around half a million men, with twenty-five mechanized divisions (85 percent on a war footing), 5,500 tanks and self-propelled guns (including about 600 heavy tanks), and 1,500 aircraft (including 500 jets). Level of training and readiness was considered to be good, essentially because of the very high quality of Soviet wireless procedures during maneuvers. The overall prospects of a Soviet offensive in Germany were favorably assessed. But it was not known whether the logistics were at hand for a general offensive in Germany. It could only be said that rail and sea communications links with Germany could support more divisions than that theater could actually absorb. However, in the Balkans and along the Danube, it was noted, there were far too few divisions to launch an attack on northern Italy.33 The Reorientation of 1954 In line with those more practical developments, there was a switch in 1954 to a less sweeping and more realistic plan de renseignement, no longer specifically “for national defense” (meaning that it concerned only the armed forces, not all government agencies). It was clearly a retreat from the ambitious 1946 scheme of an all-encompassing civilian intelligence organization. The military’s prewar role was in part restored, perhaps linked to the postmortem on the Dien Bien Phu disaster, which had also been a failure both of intelligence and of governmental political-military coordination.



French Intelligence About the East, 1945–1968

137

The focus was more unambiguously directed at the Eastern bloc: the military and economic capabilities of the Soviet Union and its satellites, and their potential as a threat to Western Europe and the Union française.34 It was now seen as infeasible to penetrate the Soviet decision process: apart from the economies of the satellite countries, for which some occasional informants could be found, it was not possible to get secret intelligence on political or military questions. It was decided to emphasize intelligence about infrastructures and means of transport and communication, which were easier to observe and would give clues about Soviet logistical and hence offensive capabilities, if not intentions. The same approach was restated constantly in the instructions from Paris during the 1960s and 1970s. For instance, in June 1968, French military attachés in Warsaw were instructed to concentrate on logistics, using first open documentation, than making in situ observations (with pictures) of seven rail marshaling yards at the border with the USSR, of waterways, roads, military depots, and civilian fuel storage areas.35 In November 1972, the chief of the Logistics Division of the Centre d’étude du renseignement (part of the SGDN) advised the military attachés in Moscow to concentrate on road traffic (by regularly traveling on them) and on the Transiberian rail link with the Far East and China.36 Meanwhile they did not neglect scientific and technical intelligence. The “Plan de renseignements scientifiques et techniques” issued in March 1953 by the Intelligence Division of the EMFA was directed against all relevant countries, not specifically the Eastern bloc. A rather general document, it stressed communications (including encryption), missiles, nuclear energy, and aeronautics. That plan was only a general frame of reference, followed by more precise and detailed plans, but was nonetheless classified.37 During the 1960s, the fields of observation were enlarged to what could be called “alert indicators” (troop movements and state of readiness, transports, wireless transmissions, sudden changes of codes and electronic countermeasures . . . ), which might reveal signs of an impending Soviet offensive, without having access to the innermost Soviet decision-making. The general philosophy was to make the best intelligence use of what could be accessed, without waiting for a most improbable penetration of the Soviet target.38 It is well known that in 1966, after the fracas of the Ben Barka scandal, the French secret service was reshuffled: it no longer depended on the Office of the Prime Minister at Matignon, but was put (as before the war) under the

138

French Intelligence About the East, 1945–1968

authority of the Ministry of Defense. It is conceivable that the military took the opportunity of the SECE scandal to regain control over it. The Role of the Military Attachés The role of the military attachés became more important after the 1954 reorganization and with the growing role of the SGDN in the French intelligence system. Their work also became easier after the 1956 thaw in the Soviet Union: the military attachés in Moscow could then get the necessary authorization for many trips by car toward Kiev, the north, other directions, and even down to Tbilisi, a 4,500–km round-trip from Moscow, through Sokhumi, and back (along the “military road of the Caucasus” and through Ordjonikidze, Mineralny Vody, and Rostov). In their reports, they described roads, bridges, railways, airfields, and so on.39 In July 1957, Lieutenant-Colonel Guibaud traveled, a first since the war, on the Transiberian Railway. He noticed the scarcity of transport links and of military airfields, and the general underdevelopment of the country.40 Those observations led to a reassessment of Soviet war-making capabilities: they were now considered less threatening than had been judged earlier, for instance, by General Augustin Guillaume, military attaché in Moscow in 1946–1948, who after his return published a book titled Pourquoi l’Armée rouge a vaincu,41 in which he stressed the power of Soviet armed forces. This was to play an important role during the Suez crisis of November 1956, as we shall see. French military attachés in Warsaw were instructed in June 1968 to observe Polish domestic policies (including the role of the armed forces), foreign policy, and economic matters related to defense (including the budget, general economic development, COMECON, and armaments).42 They were asked to check consistently on confidential information received in Paris (apparently mostly through the DST, which had good access to Poles living in France), such as reports on low morale and lack of discipline in the Polish Army (December 1956), with military service time reduced to eighteen months (July 1961), Soviet troop movements in Poland, and Soviet officers in Polish railway stations (August and September 1961, at the time of the Berlin Wall crisis). Or, following information from a reliable source about Poland’s role in the SovietYugoslav reconciliation in May 1962, the military attachés were to observe Polish leaders’ attitude to Belgrade and to check on the presence of a Yugoslav observer at COMECON.43



French Intelligence About the East, 1945–1968

139

Instructions can be found for the attachés in Belgrade for 1973 (four pages of detailed instructions covering all problems and soft spots of Yugoslavia at the time) and 1979: apart from the very great importance granted, as usual, to road and railway reconnaissance by the attachés (their reports were stated in 1965 by the CER to be the best from all military missions in Eastern Europe), they were to study economics, armaments, and to evaluate Tito’s succession and Yugoslavia’s role in the Balkan and in the non-aligned movement. And the state of relations with Moscow was always followed with the keenest interest (e.g., in a note of 18 December 1963 about Soviet-Yugoslav military collaboration). A great percentage of useful observations derived from the “open” activities of military attachés once they were authorized to roam about countries of the Soviet bloc (within limits, of course). Why were they allowed that relative freedom after 1956? Perhaps the Soviet authorities were not averse to letting them understand that notwithstanding Khrushchev’s braggadocio, the USSR was incapable of waging a world war, aside from a dissuasive nuclear spasm. The Collation and Evaluation of Intelligence At the same time, and probably also because the SDECE did not live up to its sweeping charter, the whole system of collating and evaluating intelligence was progressively organized and located at the EMPDN/SGDN. Intelligence from all sources (open, secret, military, diplomatic) was collated and evaluated at the EMGDN, until 1962, then at its successor department, the SGDN (Secrétariat général de la défense nationale), which was under the authority of the prime minister. During the Fourth Republic, until 1958, the product disseminated apparently consisted entirely of so-called notes de renseignement prepared by the SGDN’s Service d’information générale, each covering a single topic, without periodicity or system, which were distributed to the usual list: the Elysée, Matignon, the Quai d’Orsay, the Defense and Services ministries, the chiefs of staff, and the director of the SDECE. These notes de renseignement were meant for the information of the military attachés abroad in particular. Many concerned political issues. Thus, for instance, a 25-page note de renseignement of 14 June 1950 perceptively analyzed the USSR’s ambiguous policy regarding Islam: it both excluded Muslims from social life inside the Soviet Union, not without some success, and professed freedom of belief, using that purported liberty in propaganda directed at Islamic countries in order to turn them against the West.44

140

French Intelligence About the East, 1945–1968

Most of the political notes de renseignement relied on open information. One from 26 April 1956 on Soviet-Yugoslav relations, resting evidently only on open information, was quite perceptive about the warming of those relations since 1953, for example, but did not add anything to the Quai d’Orsay’s assessment.45 It was more a tool of information for the military attachés, to keep them abreast of events, than real analysis. Occasionally, an informant could be found: for instance, from a so-called “People for Peace Congress” in Vienna in December 1952, and particularly for a very interesting note de renseignement of 21 April 1956 about the Twentieth Party Congress in February and Khrushchev’s secret speech, which was described by the SGDN in its basic outline long before its publication by CIA in June, with the (true) information about a letter on Stalin’s crimes to be read at Party cells, but which was quickly stopped because of opposition from many Party members.46 Many documents concerned the economy of the Soviet bloc, which was evidently a priority target, covered about once a week, but essentially from Soviet official data, and hence rather sanguine about the growth rate of the economy (estimated at around 10 percent in 1957), although they did not miss the imbalances and other problems (e.g., in a note de renseignement of 15 October 1957).47 There were both short notices on specific topics, and more synthetic, longer documents: for instance, a 9-page note about the economy of the USSR and its satellites on 13 November 1950, which was rather sanguine about the Soviet bloc’s prospects, taking issue with the Americans for their more pessimistic evaluation. This did not even make use of the long document from the SDECE on the same subject issued four months before.48 Much better though was a note de renseignement on 20 October 1955 about economic relations between Moscow and the satellites, stressing the fact that even after Stalin’s death, and under the guise of a purported liberalization, Moscow was still calling the tune.49 A note de renseignement about COMECON on 25 November1957, using both open and secret intelligence, stressed information according to which Poland was resisting full integration into COMECON and managed to retain a modicum of economic and tariff freedom of action, which is particularly interesting in view of the period covered by the document.50 Apparently, the Fifth Republic reorganized the system after 1958: henceforth, analysis was done by the CER. The product became much better, and less redundant with Quai d’Orsay’s own product. It was more rationally



French Intelligence About the East, 1945–1968

141

organized, more systematic, and more regularly disseminated through two series of bulletins: (1) The CER prepared a general, synthetic Bulletin mensuel: Pays du Bloc soviétique et Yougoslavie meant for the various usual authorities, but also for French military attachés abroad. 51 Each bulletin in the years 1961–1965 was organized in the same way: • A one- or two-page introduction, not uninteresting, but not basically different from information available in Politique étrangère, the journal of the quasi-official “Centre de politique étrangère.” • A first part was devoted to the USSR, both domestic and foreign policy, and including the economy. • A second part was devoted to the armed forces of the Soviet Bloc, with five different paragraphs (East Germany and Czechoslovakia, Hungary and Bulgaria, Poland, Albania, the USSR). • A third part described the domestic policies of the satellite countries and Yugoslavia. Parts one and three were evidently compiled mostly from the press, with no input from the Quai d’Orsay or SDECE. Part two was the only one where secret intelligence was evidently used; it was mostly about troop movements and maneuvers, with only a few details of the Warsaw Pact order of battle. That information came evidently mostly from the military attachés and from the MFL in Potsdam. It was not useless: the bulletin for August 1961 gives an adequate description of Warsaw Pact military preparations at the time, noting the fact that all Soviet bloc countries’ armies were switching to a war footing, with an analysis of the Soviet–East German troop buildup around Berlin. The September issue indicates the arrival in the theater of a missile unit. And we know now that there were indeed such movements.52 As for the Cuban Missile Crisis of 1962, an analysis in the October bulletin stresses that the apparent lack of concern for secrecy when introducing the Soviet IRBM showed that the real aim was to force Washington to a global deal, including Berlin. That piece of analysis was most probably sound, but does not seem to rest on any hard intelligence (not even the one garnered by the SDECE in Cuba about the arrival of Soviet missiles).53 It was a purely political analysis; that type of analysis could be of course well off the mark, as shows the fact that the Sino-Soviet

142

French Intelligence About the East, 1945–1968

rift was not addressed before the March 1963 issue, and then without any sense of its finality. (2) There were also more detailed collections of Bulletins particuliers de renseignement by country. In the case of the 1969–1978 collection for the USSR,54 these are organized very much like the Fifth Republic Notes de renseignement described above, but are more systematic, more precise, and much better. Here finds interesting studies of the growing problems and imbalances of the Soviet economy and its increasing reliance on Western capital and know-how (15 November 1973); about Soviet logistics and military problems in the Far Eastern theater (1 March 1971, 11 October 1973); and about the reorganization of Soviet theater commands for the Middle East (24 January 1973). Contrary to most reports, the last two rely also on secret intelligence from French sources, confirming Allied information. The Bulletin particulier de renseignement concerning Poland for 1964–1965 and 1970–197955 was prepared along the same lines as the Soviet bulletin. The same applies to the Bulletin particulier for Yugoslavia for 1969–1977 (issued monthly). That bulletin is particularly interesting, mostly from open sources, but with some secret information (e.g., in December 1972 about overflight rights for Russia for forty military transport aircraft flying war matériel to Syria). It is also rich in detail about the crisis in Croatia in 1971–1972 and possible cases of “external interference,” without deciding whether such interference (there is an allusion to Italy) was real or Soviet disinformation.56

Informal Cooperation There seems to have been a certain amount of informal, ad hoc cooperation between various French intelligence agencies, but once again, we do not have access to most of the product. The SDECE, for instance, provided intelligence from allied agencies (dubbed TOTEM) to the SGDN. A rare instance is a SDECE report from 31 August 1967 about the trip to Rome that July of Boris Ponomarev, secretary of the Central Committee in charge of Communist parties in the capitalist world. The information had been provided by the Italian intelligence service and described Ponomarev’s declarations to the Italian Communist leaders about the need of Communist countries for



French Intelligence About the East, 1945–1968

143

peace, because of the expense incurred in supporting Vietnam and the Arab countries. He called for an international Communist conference in order to provide support for Soviet peace initiatives, despite Chinese opposition, and to escape from the quandary of either helping liberation movements, but then facing brutal American reactions, or not helping them, and thus renouncing the USSR’s historic role and great power status.57 There were frequent exchanges with the Quai d’Orsay: some studies (as in October 1955 about economic relations between Moscow and its satellites) were prepared at the request of the Foreign Ministry and in collaboration with it.58 Generally speaking, many documents were sent by the Quai, apart from direct contacts and the fact that traditionally the number two of the SGDN was a diplomat. There were many instances of such collaboration for Yugoslavia, which is particularly well covered: there are exchanges with both the DST and the SDECE to check or enlarge on various items of information. Those exchanges go in both directions. For instance, in April 1965, the DST evidently asks the CER to confirm an item of information about Yugoslavia in order to check the credibility of its own source.59 At embassy level, there were frequent free exchanges between diplomats, military attachés, and the SDECE representative. For instance, in the summer of 1958 I traveled with my father (then a diplomat serving in Moscow) by car down to Tbilisi and back. There were at the time no roadmaps on sale, only a small, rather imprecise atlas of the Soviet Union, and I marveled at how he always seemed very sure of where to go and what to expect—until, that is, I recently found in his papers a copy of the itinerary map drawn by the military attachés for the same trip in 1956, which one can see at the military archives in Vincennes! Needless to say, it was not just about tourism: all embassy members were expected to write reports after such trips. There was a sort of informal intelligence community between the military attachés, the diplomats, and the members of the CER of the SGDN. The latter undertook regularly journeys in the Soviet bloc countries to compare their analysis with that of the SDECE representatives, the military attachés, and the diplomats. A two-week trip through the USSR in August 1965 was undertaken by the CER’s Major de Rochegonde, who spoke Russian, and was shown around by the military attachés. He traveled a lot, including to Leningrad and Central Asia, and made excellent observations about the general backwardness of the country.60 Another mission to Moscow and Leningrad was undertaken in November 1972 by the head of the Economics and Logistics Department of the CER,

144

French Intelligence About the East, 1945–1968

Lieutenant-Colonel Verzat. He concluded that Soviet logistics (infrastructure and roads and railways) were still the most useful target, limited in scope maybe but sound. He decided to reorganize the military attachés in Moscow to let them concentrate on such targets.61 Other interesting instances of informal cooperation were missions of CER officers to Belgrade in May 1966 and June 1971. In both cases, the CER officer in charge of Yugoslavia at the SGDN came to discuss the first draft of a comprehensive study of Yugoslavia with the military attachés, the diplomats, and the SDECE station chief, who all suggested alterations or corrections.62 Intelligence Successes There were some intelligence successes. For instance, during the Suez Crisis of 1956, although the SDECE’s performance was, as we have seen, abysmal, and warned of a Soviet military intervention that never occurred, the 2ème Bureau concluded quite early on that Soviet military help to Nasser since the previous year had, in fact, been limited, and judged that “an intervention that would be more than just indirect or diplomatic was most unlikely.”63 During August 1968, after the Soviet occupation of Prague, the West was much exercised about the possibility of further Soviet interventions against Tito and Ceauşescu (when President Johnson alluded to the “dogs of war”). But an assessment by the CER on 12 September 1968, on the basis of the Soviet order of battle and from secret information (evidently Yugoslav), concluded that there was no sign of a Soviet intervention against Yugoslavia or Romania: the invasion of Czechoslovakia was defensive, to save socialism there, not a starting point for a new phase of Soviet expansion, which was an urgent and much disputed topic in the West at the time.64 Conclusion What we know is only the tip of the iceberg, of course (although I suspect that the iceberg may have been rather shallow). For instance, we know very little about French intelligence cooperation with the Western allies. Apparently, there were regular meetings of the U.S., British, and Canadian intelligence service attachés in Warsaw, for instance, but in the 1960s, the French did not take part. I found only two pieces of evidence for TOTEM exchanges between the SDECE and its NATO counterparts in the files, although there were certainly



French Intelligence About the East, 1945–1968

145

many more. There is also some evidence of military intelligence coming from NATO, but not enough to draw the overall picture, which remains very obscure, particularly without more precise information about those exchanges. What we can gather from the various bulletins and notes that can be found in the archives, and which are not raw intelligence but are intelligence estimates after analysis, is that a sort of permanently adjusted, pragmatic kind of Sovietology evolved. The target was not really penetrated; if its overall aims were understood, its immediate and secret political and strategic objectives could not be forecast. The most that could be achieved was ongoing evaluation of the economic and logistical capabilities of the Soviet bloc, plus some alert indicators (troop movements, electronic emissions). The end result was not that bad: the French intelligence system went far beyond the relatively ineffective SDECE (which apparently prioritized France’s colonial wars and then Africa), and it reached sound conclusions in all the major crises discussed—Suez 1956, Berlin 1961, Cuba 1962, and Prague 1968—and it did not exaggerate the Soviet threat. But the system actually functioned outside the framework designed in 1946, and it discreetly returned quite soon to a modernized version of the pre-1939 intelligence system—or such, at least, is my tentative conclusion. The effectiveness of the French intelligence system in reaching the upper echelons of the French government is another matter. Neither the SDECE nor the SGDN enjoyed very high status in the corridors of power—in any case, it was certainly not comparable to that of the CIA or the National Security Council in the United States, or the Joint Intelligence Committee (JIC) of the British Cabinet Office. Probably the most useful aspect of French intelligence was the many-faceted informal cooperation among representatives of various agencies that has been noted. It is probably not helpful to try to make the study of the French intelligence system fit into the framework of the British or American ones. Notes 1. Service historique de la défense (SHD), 9Q1 9; 23 March 1948, EMGDN (Présidence du Conseil). 2. SHD, 9Q1 9, 16 October 1950, note from the EMPDN. 3. SHD, 9Q1 9, 27 May, 3 October 1952, and 8 December 1952. 4. SGDN, Ministère des affaires étrangères (MAE), Service des pactes, box 235. 5. MAE, series in reorganization.

146

French Intelligence About the East, 1945–1968

6. Report from 11 January 1945, MAE, series in reorganization; report from 22 January 1946, MAE, Cabinet de Georges Bidault, vol. 6. On Cardinal Tisserant and the French secret service, see Constantin Melnik, Un espion dans le siècle (Paris: Plon, 1994). 7. Tom Bower, The Red Web: MI6 and the KGB Master Coup (London: Aurum, 1989). 8. MAE, série Y, 1944–1949, vol. 2. 9. Ibid. 10. Ibid. 11. See, e.g., Georges-Henri Soutou, “La politique française envers la Yougoslavie, 1945–1956,” Relations internationales, no. 104 (Winter 2000): 433–454. 12. MAE, Papiers Massigli, vol. 94. 13. Secretary-general Chauvel to Bonnet, ambassador in Washington, about the fact that the codes and ciphers team at the Quai d’Orsay was the object of suspicion, 17 March 1948, MAE, Papiers Bonnet, vol. 1. Georges-Henri Soutou, “La mécanisation du Chiffre au Quai d’Orsay, ou les aléas d’un système technique (1948–1958),” in Michèle Merger and Dominique Barjot, eds., Les entreprises et leurs réseaux: Hommes, capitaux, techniques et pouvoirs, XIXe–XXe siècles (Paris: Presses de l’Université de Paris–Sorbonne, 1998). 14. On the Franco-American bilateral aid agreement, 30 June 1948, MAE, série Y, 1944–1949, vol. 230. 15. Ibid., vol. 348. 16. Dariusz Jarosz and Maria Pasztor, Conflits brûlants de la Guerre froide: Les relations franco-polonaises de 1945 à 1954 (Paris: Lavauzelle, 2005). 17. Pierre Jardin, “Französicher Nachrichtendienst in Deutschland in den ersten Jahren des Kalten Krieges,” in Wolfgang Krieger and Jürgen Weber, eds., Spionage für den Frieden? Nachrichtendienste in Deutschland während des Kalten Krieges (Munich: Olzog, 1997). When I was working on my book L’alliance incertaine: Les rapports politico-stratégiques franco-allemands, 1954–1996 (Paris: Fayard, 1996), I was under the impression that Paris knew a lot about what was happening in Bonn in the de Gaulle years. This does not seem to have been the case under the Fourth Republic: see Georges-Henri Soutou, “Unter der Lupe. Die Reise Adenauers nach Moskau in französischer Sicht,” in Adenauers Moskaubesuch 1955: Eine Reise im internationalen Kontext, Rhöndorfer Gespräche, 22 (Bonn: Bouvier, 2007). 18. MAE, Service des pactes, box 52. 19. This came to light in the preparation of the 1948 volumes of Documents diplomatiques français. 20. SHD, 10 T 444, 9 March 1960. 21. Ibid., 31 January 1958. 22. Claude Faure, Aux services de la République du BCRA à la DGSE (Paris: Fayard, 2004); the best overall presentation today, although within the limits of an (un)official history. 23. Vincent Auriol, Journal du Septennat (Paris: Armand Colin, 1970–), particularly the volumes for 1947 and 1948.



French Intelligence About the East, 1945–1968

147

24. SHD, 9Q3 14. 25. Ibid., 4 July 1950 and 17 December 1955. 26. Note by the chief of the Second Division of the EMFA, 10 November 1956, SHD, Département Terre, Fonds Ely, 233 K 48. 27. SHD, 9Q3 72, June 11, 1959. 28. SHD, 9Q3 72/4. 29. Yves Bertrand, Je ne sais rien . . . mais je dirai (presque) tout (Paris: Plon, 2007), 154. 30. SHD, 9Q1 9. 31. SHD, 9Q1 9, 13 October 1948. 32. Apparently the records of the MFL were destroyed after 1991; but I was able to get some information. 33. SHD, 9Q3 67. 34. SHD, 9Q1 9, 29 June 1954. 35. SHD, 9Q3 29. 36. Note, 16 November 1972, SHD, 9Q3 49/1. 37. SHD, 10Q1. 38. Note from the Intelligence Division of the EMA on 6 May 1968, SHD, 93 Q 67/4. See Frédéric Guelton, “Les chefs militaires français et la réalité de la menace militaire soviétique, 1946–1950,” in Jean Delmas and Jean Kessler, eds., Renseignement et propagande pendant la guerre froide (1947–1953) (Brussels: Éditions Complexe, 1999), 257–276. 39. SHD, 10 T 444. 40. Ibid. 41. Guillaume, Pourquoi l’Armée rouge a vaincu (Paris: Julliard, 1948). 42. SHD, 9Q3 29. 43. Ibid. 44. SHD, 9Q3 49/2. 45. Ibid. Soutou, “La politique française envers la Yougoslavie.” 46. SHD, 9Q3 49/2. 47. Ibid. 48. Ibid., 14. 49. Ibid. 50. Ibid. 51. Ibid., 3/3. 52. Vojtech Mastny, Sven S. Holtsmark, and Andreas Wenger, eds., War Plans and Alliances in the Cold War: Threat Perceptions in the East and West (London: Routledge, 2006). Matthias Uhl, Krieg um Berlin? Die sowjetische Militär- und Sicherheitspolitik in der zweiten Berlin-Krise 1958 bis 1962 (Munich: Oldenbourg, 2008). 53. L’Europe et la crise de Cuba, ed. Maurice Vaïsse (Paris: Armand Colin, 1993). 54. SHD, 9Q3 49/3. 55. Ibid., 29 and 30. 56. Ibid., 72/3. See also Erich Schmidt-Eenboom, Der Schattenkrieger: Klaus Kinkel und der BND (Düsseldorf: ECON, 1995) on the BND’s role in Croatia in 1971.

148

French Intelligence About the East, 1945–1968

57. SHD, 9Q3 67/5. 58. See the correspondence between the Quai and the SGDN on the study in SHD, 9Q3 14. 59. SHD, 9Q3 72/4. 60. SHD, 9Q3 49/1. 61. Verzat, report on 16 November 1972, SHD, 9Q3 49/1. 62. SHD, 9Q3 72/2. 63. Ibid. SHD, 10 T 791, 26 September 1956. 64. SHD, 9Q3 72/4.

5

British Intelligence During the Cold War Richard J. Aldrich

Introduction Historians of intelligence rarely think about money.1 Yet if one were to choose two or three words to describe British intelligence during the Cold War, one of them would be “impecunious.” Intelligence practitioners, once they had achieved middle rank, thought about budgets much of the time. This was because postwar Britain underwent periodic financial crises, prompting severe cutbacks in all aspects of foreign and defense policy, including intelligence. Amid much argument, intelligence was protected against the worst of the cuts in the 1950s. However, money was always tight, and retrenchment had arrived for all by the mid-1960s. The mid-1970s brought further economies and the end of the Cold War inevitably brought further cuts. The prevailing sense of impoverishment among Whitehall’s secret servants was increased by an unhappy comparison. Britain’s closest intelligence partner was the United States, which enjoyed—in the eyes of London—almost unlimited resources. A perennial question was exactly how much Britain needed to contribute to the overall Western pool of intelligence to justify the generous transatlantic flow of intelligence. Intelligence chiefs did not hesitate to play upon the American factor in their battles with Treasury officials.2 For intelligence historians researching Britain’s Cold War, the minutiae of budget details have hardly been a top priority.3 Simply establishing a broad outline of events has been the major challenge, since the first decade of the Cold War still remains something of a historical Rubicon, and few

149

150

British Intelligence During the Cold War

departments have crossed this confidently in terms of the declassification of intelligence material. A quantity of Cold War intelligence files are available in the British archives, but they are the mere tip the of iceberg compared to what actually exists—and even more have been destroyed. Identifying reliable sources for intelligence activities beyond the Suez is still a major problem.4 By contrast, general commentaries on American intelligence and the Cold War are more plentiful, and for some years, they have addressed more contemporary subjects, such as the Vietnam War.5 Perhaps the greatest challenge for all intelligence historians examining events after 1945 has been to establish the chains of causation that link this subject to the wider fabric of Cold War history. Arguably, the Cold War represents a more slippery problem than World War II. The task of demonstrating the impact of intelligence on military strategy and operations in wartime, with its obvious lines on a map, is often easier. Showing how it affected the miasma of Cold War struggles, which were—more often than not—political and psychological, is less straightforward. Precisely for this reason, we might be advised to pay a little more attention to the battles over budgets. This is because, faced with repeated rounds of economies and savings, intelligence chiefs were sometimes asked the same general questions about value and relevance that now confront historians. They were required to cast up a summary of costs and benefits. As Leo Pliatzky once explained, “getting and spending” is an exercise in social choice that reveals the value that is attached to policies, programs, and organizations (Pliatzky 1982, 3). When we come to ask if British intelligence mattered and did it make a difference, we are likely to find some of the answers in the arguments over money.6 Spending may not be an accurate guide to notions of value during the first decade of the Cold War. Officials frankly admitted that prior to 1952, control of the British intelligence budget was rudimentary and no one really knew what the overall commitment was, since perhaps as much as half of it lay outside the secret vote.7 Thereafter, things gradually became better organized, partly due to the inexorable rise of Burke Trend, a senior Treasury official (later cabinet secretary) who took an avuncular interest in all things secret. Indeed, Trend has rightly been described as “something of a shop steward for the secret servants of the state,” and, together with Dick White, he did much to develop new processes in the 1960s and 1970s (Hennessy 1989, 213). Officials were soon able to ask what the British intelligence community was paid. This was asked with increasing frequency, since technology was pushing costs up.



British Intelligence During the Cold War

151

In 1957, the United States calculated that the overall price of its intelligence budget was approaching a billion dollars a year (Aldrich 2001, 521). The UK intelligence community was also seen as increasingly expensive, with costs at Government Communication Headquarters (GCHQ) accelerating fastest. Indeed, a mini-crisis occurred in 1962, when it looked as though spending on Britain’s overall sigint budget would soon overtake the entire cost of the Foreign Office with its panoply of embassies and diplomats. The total sigint budget agreed for 1963–1964 was £20,520,000 while the total for the Foreign Service was £21,816,000. The current round of economies required all overseas departments to find 10 percent savings, but GCHQ was engaged in special pleading for a 10 percent increase. The possibility that the servant might be paid more than its master triggered a wide-ranging review of GCHQ and its activities, led by the philosophy don and sometime counterintelligence officer, Sir Stuart Hampshire (Aldrich 2011). The Hampshire Review, which was Treasury-driven, was required to ask difficult questions that are not so very far from those that historians also face. How might the value and importance of intelligence actually be measured? What were the risks of cutting, set against the rewards of continuing to allow an upward spiral of costs and activity? The Treasury asked GCHQ to specify what it had achieved with several previous tranches of new money, allocated to it in five-year blocs, under a scheme called “Measures to Improve” (MTI). One of GCHQ’s main targets had been Soviet high-grade diplomatic communications, but this prize had eluded it, and so GCHQ squirmed a little before finally providing a full account of its endeavors. Matters were made more difficult because of the elusive nature of Britain’s sigint budget. As defense officials later explained, the sigint budget covered both the total costs of GCHQ and those elements in the Ministry of Defence (MoD) involved in strategic sigint activities. They added that the sigint budget “has no separate existence but is distributed for security reasons among several departments, the MoD votes bearing a large proportion of it.” In reality, even the total sigint budget of £20 million did not capture all of the expenditure in this secretive area.8 Predictably, this effort to measure the value of public expenditure in the realm of sigint proved difficult. Craftily, GCHQ fell back on wider arguments about Anglo-American relationship, an even bigger issue quite beyond the capacity of the Treasury to measure. Hampshire spent more than a month at the U.S. National Security Agency (NSA) in December 1962 and January 1963 investigating this aspect. Toward the end of the review, more senior figures,

152

British Intelligence During the Cold War

including Burke Trend and Norman Brook, concluded that they simply had to accept that, in addition to sustaining the Anglo-American relationship, intelligence also provided a range of functions that Britain simply could not do without. The most important of these was warning of both nuclear and conventional attack, contributing directly to international stability. GCHQ did not answer all the questions, but it still received most of their additional money. Moreover, while it proved frustratingly difficult to weigh and measure what things were worth, Trend’s functional analysis of British intelligence—asking what “services” the intelligence services provided—was not a bad approach. Accordingly, that perspective informs the subsequent structure of this essay (Aldrich 2011).

Warning The most important service that British intelligence provided was “warning.” Arguably, one of the main drivers behind the revolution in intelligence during the mid-twentieth century had been the increase in strategic mobility. Strategic war warning was of particular interest to Britain for several reasons. First, all western European countries chose to lean heavily on nuclear weapons for their defense, recognizing that they were cheaper than conventional forces. As U.S. President Eisenhower famously remarked, they provided more “bang for your buck.” Accordingly, European countries, which spent less per head of population than the United States on defense, were trapped by the problem of nuclear addiction. Put simply, low conventional defense spending locked them into a tripwire strategy in which nuclear weapons would be used early if war broke out, since Europe could not be defended for long with conventional forces. Lurking underneath was a further British interest in theater nuclear counterforce targeting, since in any such conflict Europe would be the main battleground. British intelligence spent an inordinate amount of time trying to locate Soviet theater nuclear forces in the hope of being able to blunt this capability should war break out.9 Second, at the conventional level, the intelligence “tripwire” was important to Britain, inasmuch as a British general was commander in chief of NATO’s Northern Army Group (NORTHAG), which defended the northern half of Germany. NORTHAG consisted of four national corps, which were not in their battle locations in peacetime and needed to be assembled for war. The British-led Northern Army Group comprised widely scattered Belgian,



British Intelligence During the Cold War

153

Dutch, and German components. This left NORTHAG horribly dependent on warning. In reality, anything less that forty-eight hours’ notice of a Soviet conventional attack would have left the British commander in chief all but paralyzed. Moreover, for reasons of security, NORTHAG and the British Army of the Rhine (BAOR) were not privy to the highest grades of intelligence during peacetime, typically the more sensitive forms of sigint. Thus on the eve of war, as NORTHAG HQ sped toward its temporary survival location in the forests of Aachen, it would have been bombarded with new types of intelligence that it was ill-equipped to handle. As with so much of Britain’s Cold War military infrastructure, the only reliable plan for “hot war” was that it should not happen at all (Aldrich 2008). Although the importance of strategic warning ensured that intelligence was given great emphasis in the strategic calculations of both East and West, its practical performance was poor. The year 1968 was a moment of spectacular failure. The British Joint Intelligence Committee (JIC) insisted that the USSR would not invade Czechoslovakia, despite a substantial volume of hard intelligence to the contrary (Hughes 2004). This was an analytical blunder in which a Foreign Office chairman of the JIC overruled important empirical evidence from defense sources. No less disturbing was false warning of a possible Soviet invasion of Romania in 1969 triggered by a sudden period of Soviet radio silence. Sir Dick White, the first British Cabinet Office intelligence coordinator, was incensed and ordered a major review. This led to the beefing up of the management of the Defence Intelligence Staff, presided over by Admiral Louis le Bailly.10 Worse was to come. The surprise attack that the Arabs inflicted upon Israel in October 1973 only served to provoke new levels of British anxiety about intelligence, warning, and communications. This episode witnessed excellent Arab deception and poor Israeli intelligence analysis, combined with indecisiveness about whether to mobilize or not. British observers were not slow to realize that the same problems would be faced in northern Europe, compounded by the need for complex inter-Allied consultation on the eve of war: NATO is confident of and requires 48 hours warning of an attack in Europe because of the scale of preparations necessary for even a limited attack. However, in a situation similar to the Central Front, the Israelis’ vaunted intelligence system noted all the indicators and failed to construe them correctly. NATO might do no better, and a close and careful reappraisal of our intelligence collection methods and analysis procedures is indicated.11

154

British Intelligence During the Cold War

There was increasing appreciation that their large standing forces made a surprise attack an attractive option for the Warsaw Pact nations. What would happen if they chose the advantage of complete surprise? Remarkably, it was conceded that although NATO procedures worked reasonably well during exercises based on adequate warning time, they had never been tested against a surprise attack. In late 1981, Douglas Nicoll, a senior GCHQ official, conducted an inquiry into what was increasingly regarded as the imperfect pattern of warning hitherto provided by the Joint Intelligence Committee (Goodman 2007a). Only months after he had completed his investigation, Britain was surprised again by the Argentinean invasion of the Falklands in early 1982. This triggered a second inquiry by Nicoll. Although he concluded that the Falklands attack had been “unwarnable” because of the haste with which the Argentineans had acted, Nicoll thought that Britain did not get good value for money as regards warning overall (Goodman 2007a, Freedman 2006). This was confirmed yet again in 1989, when the end of the Cold War was anticipated at some distance by enterprising journalists such as Timothy Garton Ash, but completely missed by the secret services. The end of the Cold War did not mean the end of intelligence mishaps, since the same leviathans failed to foresee the Iraqi invasion of Kuwait in 1991. Strategic warning was something that the Cold War adversaries could not afford to do without, but its performance was patchy at best. Stability Michael Herman has suggested that while the intelligence activities of each individual country are conducted in a competitive environment, the collective result that accrues can sometimes be equivalent to greater transparency in the international system as a whole. In other words, the cumulative effect of many agencies shining their torches into the dark corners is to illuminate the wider landscape. The international system is a thus a more reassuring place for the policy-makers who have access to this material (Herman 1995, 157–182, 362–376). The irony of intelligence adversaries working “cooperatively” and collectively was not lost on the practitioners themselves. When Nikita Khrushchev visited the United States in September 1959, Allen Dulles, the director of Central Intelligence, made a reference to double agents and suggested to him cheekily: “Maybe we should pool our efforts.” Khrushchev



British Intelligence During the Cold War

155

enjoyed the joke and replied: “Yes. We should buy our intelligence data together and save money. We’d have to pay the people only once!” (“The Elemental Force,” Time, 28 September 1959).12 Although Britain was not a direct participant in the bilateral Strategic Arms Limitation Talks (SALT), it was strongly interested in the success of arms control. It was also a signatory of the Partial Test Ban Treaty in 1963 and the Non-Proliferation Treaty (NPT) of 1968. Satellite imagery, dominated by the United States, was an essential component of the major Cold War arms control initiatives, but Britain’s contribution, based on both sigint and seismology, was not unimportant. Seismology was delivered in association with a number of Commonwealth partners.13 Indeed, one could argue that it is no coincidence that complex arms control treaties and increasingly sophisticated technical collection systems emerged at around the same time. The phrase “national technical means” first gained wide currency in the arms control talks of the 1960s (Goodman 2007b, Richelson 2006). In additional to supporting the overall treaty framework, British intelligence also monitored countries that were outside the NPT, including India, Israel, Libya, North Korea, Pakistan, and South Africa. Arguably, the NPT regime was successful, since the large numbers of nuclear states predicted for the late twentieth century did not transpire. Success was due in part to the ability to track developments over several decades and to apply discreet diplomatic pressure to aspiring nuclear states. Similar arguments can be made about intelligence in the realm of chemical and biological weapons. However, these processes were not without tensions. Politicians and diplomats, anxious to proceed with arms-control treaties, were often reluctant to listen to evidence from the intelligence agencies about discrepancies and cheating. This confirms the broader thesis that while intelligence may support arms control, its main impetus is normally political will, not technical or strategic context (Sheehan 1988). Yet, while intelligence products bolstered stability, the collection operations themselves were a source of risk. As James Reeve, a senior British diplomat recalls, the evolution of new initiatives toward the Soviet Union or its satellites would involve slow, laborious consultations with Western allies, with NATO, and with ambassadors for eventual approval by the foreign secretary and Cabinet. However: “Months of careful preparation could be overturned by an incident such as the shooting down of the American U-2 spy plane over Russia at the time of the summit with Nikita Khrushchev in Paris,

156

British Intelligence During the Cold War

or Commander Crabb’s fatal underwater exploits around the hull of a visiting Russian warship.” The hair-raising activities involved the collection of both imagery and short-range sigint and were a perennial source of risk (Reeve 1999, 123). A good example of calculated risk was British spy-flight activity in the early 1950s, when top-secret missions had preceded the U-2 program and penetrated the Soviet Union (Lashmar 1996). The U.S. Air Force and the RAF had developed a joint plan for overflights of the Soviet Union in order to obtain radar photographs of key targets. The U.S. State Department was still hesitant about American aircraft intruding over the Soviet Union, hence the plan was to lend four American RB-45C multi-jet-engined bombers to the RAF. Squadron Leader John Crampton led a team of RAF officers who trained at Barksdale Air Force Base in Louisiana. They were then stationed at RAF Sculthorpe for their first missions, which received direct approval from the newly elected Winston Churchill. On the night of 17 April 1952, three RB-45Cs in RAF colors—but without serial numbers—began their long missions, undertaking in-flight refueling as they went. One aircraft covered the Baltic States, Poland, and parts of East Germany. The second aircraft reconnoitered Byelorussia. The third aircraft, flown by Crampton and Sanders, took a long route over the Ukraine and the Black Sea. Two years later, in April 1954, the three missions were repeated. Once more, Crampton followed a long southerly route over the Ukraine, penetrating even further into Russia. In contrast to the previous visit, the Russians were ready for them. In retrospect, Crampton believes he was lucky not to be shot down. Over Kiev, they were met by an intense anti-aircraft barrage. They only survived because the Soviet gunners on the ground miscalculated their speed and fired just ahead of them. Hot shards of shrapnel hissed around them. Once they were back over Western Europe, they discovered that their refueling equipment was damaged, and so they came down early in West Germany (Lashmar 1996, 60–75; Peebles 2000). The potential for a shoot-down of one of these flights, perhaps over Moscow, raises interesting questions about how the risks were calculated. After the Buster Crabb frogman incident of 1956, and again after the Gary Powers U-2 shoot-down of 1960, procedures for the clearance of British “special intelligence activities,” and indeed of American missions from British bases, were progressively tightened. Typically, such activities were suspended during East-West summits or general elections.14



British Intelligence During the Cold War

157

Anthropology On the front line, British commanders in Germany were always anxious to expand their own considerable peacetime collection and analysis resources. One of their early successes was the Scientific and Technical Intelligence Bureau, which was perhaps the most important intelligence unit within the Intelligence Division of Control Commission Germany. This unit was responsible for interrogating returning Germans POWs and civilians who had been in the East. Because the USSR kept German POWs for more than a decade after the end of the war and used them extensively on military and industrial projects, they provided an invaluable source of intelligence on a vast range of Soviet projects, including atomic programs. Indeed, during the 1950s, these activities were probably as valuable as the material provided by SIS and its German equivalent, the BND (Maddrell 2006). The pressure on networks of traditional agents increased the value of another collection asset, the British Commander-in-Chief’s Mission to the Soviet Forces in Germany (BRIXMIS). This was initiated on 16 September 1946 under the Robertson-Malinin Agreement. The agreement called for an exchange of liaison missions, effectively teams of roving military attachés, whose task was to encourage good relations between the two occupation zones. Later, parallel agreements for smaller missions were concluded by the Soviet, French, and American commanders in Germany. These missions remained in being until October 1990 (Geraghty 1996, Gibson 2005). Inevitably, the ability of the BRIXMIS missions to move around the GDR relatively freely was quickly turned to address the task of gathering intelligence on Warsaw Pact forces. The BRIXMIS mission proved to be the premier source of technical intelligence about new Soviet equipment, examples of which were often “liberated” and spirited back to the West. Quite often this meant copies of field manuals, but it could even include examples of live artillery shells stolen from training areas. They also provided invaluable material on the order of battle and troop movements that could be cross-checked with sigint. Their missions were subjected to continual hostility from both Soviet Forces and the East German Stasi during this period, which even extended to the ramming of mission cars by Soviet trucks. BRIXMIS suffered no deaths, but their French and American counterparts were not so lucky. They also worked closely with signals intelligence and atomic intelligence units, placing specialist monitoring equipment in the GDR on behalf of British national agencies. While much has

158

British Intelligence During the Cold War

been made of the semi-covert activities of BRIXMIS, its greatest value remained direct liaison between the British and Soviet commanders in Germany. They provided an important means of defusing tensions and also of directly testing the temperature of East-West military relations (Wylde 1993, Holbrook 2005). The main problem with most British intelligence sources is that they would have degraded on the brink of war. BRIXMIS would have been rounded up at the outbreak of hostilities. Sigint was an uncertain prospect, because Soviet forces were thought to work on the basis of radio silence in the early phases of war, and forward collection sites in Berlin would have quickly disappeared. Aerial reconnaissance would have struggled to follow fast-moving Soviet armored thrusts. As a result, British commanders in Germany were thrown back on remarkably primitive techniques. In the first days of war in Central Europe they planned to use the Special Air Service as human spotters. Particular units were held on short readiness times, and the unofficial motto of soldiers performing this perilous role was “wait and fly—dig and die.”15 These units were to give priority to sightings of “nuclear units, formation HQs, armour, and bridging and ferrying equipment.” Their main task was to provide target intelligence for Honest John missile systems and heavy artillery.16 Some of these special units were based at Padeborn and were equipped with high-frequency Morse transmitters to provide long-range and, hopefully, continued communications. Because of the expected rapid rate of advance by Warsaw Pact forces, there was no need for these units to practice exotic skills to penetrate the enemy front line. Instead, the drill was to move forward quickly, usually by any available soft transport and meet the delaying force. Much of this activity was focused on what commanders referred to as the “demolition belt.” These were zones some way east of the Rhine where it was hoped that bottlenecks would occur among aggressor forces some twentyfour hours after the Warsaw Pact armies had attacked. All the component national corps of NORTHAG recognized the value of stay-behind operations. Both the SIS and the SAS were also involved in planning for strategic demolitions. This is one of a number of intersections between intelligence and special operations of which we know little.17 Alliances and Sustaining a Global Role In the first postwar decade, many aspects of the Anglo-American alliance continued more or less unbroken as a complex legacy of co-belligerency



British Intelligence During the Cold War

159

against the Axis. Most important was the signals intelligence alliance known as UKUSA, concluded in the 1940s. This joined Britain, the United States, and the “Old Dominions” in a highly secure sigint pact. British influence over the nascent sigint communities in countries such as Canada, Australia, and New Zealand, together with a legacy of imperial bases, gave GCHQ substantial advantages in the early years of the Cold War. All this was expressed in a series of agreements, in many cases bilateral rather than multilateral, that were concluded between these countries in the late 1940s and early 1950s. Further letters and memoranda continued to be signed and added, providing a complex latticework of understandings and working practices, although the main round of negotiations was complete by 1953. Contrary to popular belief, the overall result was not 100 percent exchange between these parties, but was nevertheless a remarkably effective exchange mechanism (Andrew 2001). Forced economies can illuminate what intelligence was provided and at what cost. In 1975, a major British review under Roy Mason identified the Defence Intelligence Staff (DIS) for significant reductions. One of the most vocal supporters of DIS was Derek Tonkin, a senior diplomat in the Permanent Under Secretary’s Department. His diplomatic colleagues had attacked defense intelligence activity in areas of the world where British forces were never likely to be deployed. However, Tonkin defended it in terms of the high value of the Anglo-American intelligence relationship. DIS, he argued, helped to offset an intelligence imbalance in terms of what Britain offered and what it received from the United States. In the mid-1970s, this was especially apparent with the onset of a new generation of U.S. sigint satellites. Tonkin also explained that while British defense cutbacks had been reluctantly accepted by Washington, the Americans had nevertheless urged that the UK intelligence effort should not be trimmed, since they regarded this as Britain’s preeminent contribution. Finally, and perhaps most intriguingly, Tonkin offered the observation that the transatlantic intelligence relationship was about rather more than intelligence. It had now come to constitute some of the wider substance of the mainstream political relationship and thus offered London a remarkable window on political thinking in Washington (Aldrich 2006a). Closely associated with membership of a worldwide intelligence alliance was global reach. Although perhaps as much as 60 percent of the British intelligence effort was focused on the Soviet bloc, this was lower than the U.S. percentage. British intelligence enjoyed a more diffuse focus and spent more time promoting British national interests and managing colonial and

160

British Intelligence During the Cold War

postcolonial crises. No less than nuclear weapons, this global intelligence capacity distinguished Britain as a significant actor, since few states attempted to gather intelligence on a worldwide basis during the second half of the twentieth century. As Herman has argued, states are able to deploy intelligence power in the same way as they might deploy military or economic power, and Britain is perhaps an example of this (Herman 1995, 379–385). In the operational realm, intelligence was often twinned with counterinsurgency—which lay at the intersection of Cold War and decolonization. During the late 1940s and early 1950s, during insurgencies in Palestine, then Malaya and Kenya, intelligence officers climbed the learning curve, albeit slowly. Good intelligence had the potential to allow the application of more selective military force and therefore encourage better cooperation from the local population. Surrendered guerrillas in particular played a major part in the Malayan campaign and began a British tradition of using turned guerrillas or “pseudo gangs” against the enemy, a technique that would be employed later in Kenya and Cyprus and copied again by the Selous Scouts in Rhodesia (Kitson 1986, Daly 1982). The approach taken in Malaya was not the only available model. French intelligence operatives in Algeria and Indochina routinely used torture, as did the Portuguese in Africa.18 The experience of applying intelligence to low-intensity operations became a British specialty. The archive-based history of British intelligence during the Cold War has only just begun to escape the 1950s, and some of these episodes have yet to be explored. However, David Easter has shown how important sigint was to “Confrontasi”—the secret war against Indonesia in Borneo and Sarawak between 1963 and 1966. Clive Jones has revealed the importance of covert action and mercenaries to British victories in the Yemen during the same decade (Easter 2004, Jones 2004). Global intelligence capability was expensive, and sometimes the tail began to wag the dog. We tend to think of both intelligence and special operations as playing a role in supporting foreign policy. However, as Britain attempted to maintain a global presence, foreign policy sometimes supported intelligence. Particular locations, such as Cyprus, were so important to the collection of sigint that the UKUSA relationship actually helped to shape the international politics of the region. In 1975, faced with yet another financial crisis, the British cabinet formally decided to withdraw from Britain’s bases in Cyprus. Within weeks, Washington told London that this decision was not acceptable, and that the British must stay. The reason was simple. The sigint bases that



British Intelligence During the Cold War

161

allowed America to listen in on the Middle East were quite indispensable, all the more so since their bases had just been closed down in Turkey. In the early twenty-first century, more than thirty years after the British Cabinet’s “decision” to withdraw from Cyprus, the British military presence was still there and the sigint bases have grown in size (O’Malley and Craig 1999). The island of Diego Garcia in the Indian Ocean offers another example. Indeed, this represented one of the more significant intelligence-base deals of the Cold War. In the 1960s, Britain created a new colonial territory by separating a range of small islands in the Indian Ocean from Mauritius. It then removed the population and turned the islands over to de facto American control. Although presented publicly as an effort to improve the security of the Indian Ocean, its real purpose was to project American power in the Gulf Region. In part, this base deal was driven by the loss of the National Security Agency’s massive sigint site at Kagnew in Ethiopia, which had housed some 5,000 staff at its height. Anxious about being accused of neocolonialism, the United States wished to stay at a distance from the creation of the new colony. Washington therefore sought to hide its financial contribution to Britain by offering it as a rebate on the British purchase of Polaris nuclear weapons. Mauritius was paid more than £10 million to resettle the indigenous population, but in reality spent very little on resettlement. The eviction of the islanders is now the subject of a long-standing international legal case (Vine 2004). The cases of Cyprus and Diego Garcia underlined the extent to which intelligence was drawing alongside the main departments of state as a significant actor in its own right in both Britain and the United States. Security and Countersubversion Although this essay is concerned mostly with the value of intelligence, rather than security services, the issues are closely related.19 Moreover, one might argue that it is important to look at both intelligence and security activity together, if only to confirm the uniqueness of the Anglo-American paradigm. Again, money is revealing. Only Britain, the United States, and perhaps a few other countries were spending the majority of their intelligence budgets on collecting foreign intelligence during the Cold War. Reportedly, shortly after the Cold War drew to a close, Britain’s formal intelligence budget amounted to some £1.1 billion, of which GCHQ accounted for £850 million and MI5 and MI6 received about £125 million each. Even allowing for the cost of Special

162

British Intelligence During the Cold War

Branch activity by local constabularies and intelligence activity by the British Army in Northern Ireland, internal security was probably no more than 20 percent of the overall UK intelligence budget. By contrast, the majority of intelligence communities during the Cold War, and indeed after, were predominantly security intelligence communities, which prioritized the surveillance of their own populations (Aitken 2000, 5). For the British, the initial trigger for security anxieties was a series of revelations about agents connected to atomic espionage. The atomic dimension added greatly to both public anxiety and private concern by officials. Igor Gouzenko, a defecting Soviet cipher clerk, brought the atomic espionage of the British scientist Alan Nunn May to public attention in 1946, and anxiety was heightened by the Klaus Fuchs case in 1950. Shortly afterward, this was followed by the defection of Guy Burgess and Donald Maclean. In the first decade of the Cold War, the best material on Soviet espionage in both Britain and the United States was coming from a stream of intercepted Soviet communications traffic called “Venona.” Although the breakthroughs this produced were small, and achieved only with painstaking effort, they contained revelations. For the small number of people indoctrinated, they conjured up the real possibility of a vast army of Soviet spies, since most of the Venona material remained undeciphered. Partly as a response, in 1950, the United States made it clear to Britain that the introduction of positive vetting, which involved background checks for all those with access to sensitive information, was a condition of continued cooperation. Pressure from Washington ensured that this was introduced as almost the last act of a departing British Labour Government in 1951 (Hennessy 2002, 90–98). Oleg Gordievsky and Vassili Mitrokhin each brought a treasure trove of information to the West. Remarkably, while still in place as the KGB head of station in London, Oleg Gordievsky briefed Prime Minister Margaret Thatcher before her meeting with Mikhail Gorbachev in December 1984. Although the recruitment of KGB officers proved more useful in the narrow world of “spy against spy,” it also had periodic reverberations upon wider policy. Newspaper headlines during the Cold War tended to suggest that Britain was “open house” for Soviet agents. There is no doubt that some British security practices during the first decade of the Cold War were weak. However, Britain suffered no more damaging security penetrations than the United States, despite that fact that it fielded a less substantial security apparatus. Peter Hennessy has argued that Britain performed well in the realm of



British Intelligence During the Cold War

163

security and counterintelligence. Not only did Britain score some notable successes, it refused to be psychologically overwhelmed by the scale of the Soviet subversive threat. It avoided the excesses of McCarthyism that had begun to emerge in the United States by the 1950s. In aggregate terms, the British response was proportionate to the threat (Hennessy 2002). At the same time, the cost for individuals touched by Britain’s minimalist security state could be high. Much concern focused on vetting. The number of individuals vetted was considerable and included policy-makers in Whitehall, researchers at Aldermaston, and even nongovernment staff in the arms industry. During the 1960s and 1970s, fear of Soviet penetration in both Washington and Whitehall was intense. Security measures extended to background checks on BBC employees, who were sometimes barred from promotion for fringe political activities. If deemed suspect, they had a “Christmas Tree” stamp placed on the covers of their personnel files. Controversially, potential ministers could also be blacklisted. Betty Boothroyd, the former speaker of the House of Commons, has revealed that MI5 had once asked her to gather information on four Labour MPs. Reportedly, MI5 also maintained files on both Jack Straw and Peter Mandelson as a result of their youthful political activities—alas, many of these interesting documents have now been destroyed (Boothroyd 2001, 80, 119; Nelson 2011, 109; Hollingsworth and Fielding, 100–102. The most disturbing domestic security activity was not undertaken by MI5, but by the Foreign Office’s Information Research Department (IRD). Set up in the late 1940s to counter the propaganda activities of the Soviet Cominform, the IRD was originally conceived of as a continuation of the wartime Political Warfare Executive, a quarrelsome sister organization to the Special Operations Executive (SOE). However, the IRD shifted from a strictly anticommunist agenda to a broader remit that included anti-Nasser and anti-Sukarno propaganda.20 Again, budgets and resources are revealing. At the height of its activities, the IRD numbered some 500 staff and was the largest Foreign Office department. The IRD’s agenda rolled on further to include domestic countersubversion, producing a range of anticommunist books under innocuous-seeming imprints, including Ampersand Ltd, that seem to have been designed for the home market as well as foreign consumption. By the 1960s, some of Britain’s outward-looking agencies had begun to turn inwards, not unlike the CIA and NSA in the United States (Lashmar and Oliver 1993).

164

British Intelligence During the Cold War

Conclusion In real terms, Britain has probably spent more on intelligence in the first decade of the twenty-first century than it did at any point during the Cold War. This is partly in response to “the new terrorism” and partly in response to some “old new threats,” such as proliferation and organized crime, which were considered novel in the 1990s. Yet the very idea of “newness” presents a problem, since it is often used as a convenient excuse to forget old lessons learned at some cost in previous decades. Admittedly, hindsight is easy— while foresight is difficult. Yet it is hard to escape the notion that a few intelligence historians might well have spotted some of the more obvious mistakes that surrounded the Iraqi WMD saga in 2003. During the Cold War, we learned that estimating strategic weapons inside secure police states was hard and often resulted in estimates that were badly wrong. We also learned that handling defectors was difficult, and that their testimony could be unreliable. These sorts of lessons, learned patiently over half a century, have not been heeded since the Cold War ended in 1989. Learning from history—what Richard Neustadt and Ernest May have called “thinking in time”—is all very well, but to achieve this, we need accurate, professional contemporary history (Neustadt and May 1986). Without a sound database, there is a danger of us “learning,” not from real history, but from mythology and hagiography. This danger is exacerbated by the curious demographics now prevailing within the British intelligence and security agencies, ensuring that fewer practitioners have even a folk memory of their own organizational past. We need to invest strongly in the study of intelligence, developing substantial official history programs and encouraging unofficial historians with responsive declassification. Here, too, costs and benefits are an important part of the equation. Large-scale history projects have something in common with intelligence operations, since both are expensive forms of knowledge collection and processing. Yet both reward us with a better understanding of the world—and without them, we are likely to confront much higher risks. Notes 1. Important exceptions are Andrew 1985, 5–6; Andrew 2010, 779–780, 811–814; Jeffery 2010, 478–479, 642; the British Foreign and Commonwealth Office’s History Notes, no. 7, published under the title “My purdah lady” (1994); and Bickers 2001.



British Intelligence During the Cold War

165

2. As part of the “peace dividend,” MI6 lost 25 percent of its budget in the early 1990s, according to Davies 2005. 3. For general assessments of intelligence and Cold War generally, see Herman, McDonald, and Mastny 2006, Andrew 1998, and Aldrich 2006b. 4. Nevertheless, for general assessments of British intelligence in the Cold War, see Hennessy 2002, which is especially good on security issues, and Smith 2003, which uniquely covers the period beyond 1970. On the JIC, see Herman 2001 and Cradock 2002. 5. For assessments of American intelligence in the Cold War, see Ferris 1995, Gaddis 1989, Watt 1989, and Garthoff 2004 (covering a longer period). 6. Typically, the rising importance of JIC during the Cold War can be mapped by the increasing rank of the chairman and what he was paid. Private information. 7. COS (52) 152nd mtg. (1) Confidential Annex, 4 November 1952, DEFE 11/350. 8. Draft submission from CA (P) to S of S, “HS801R,” January 1968, DEFE 68/76. 9. In part this explains high spending by Britain on an airborne sigint capability. 10. Admiral Louis le Bailly, “Some reflections after nearly ten years,” 7., file 6, box 7, le Bailly papers, CCC. Also private information. My views are at variance with those expressed in Cradock 2002. 11. MISC 71/763, “Lessons of the Middle East War,” enclosing 1077/3/7, Annex B, “Lessons Relearned,” 5 February 1975, AIR 20/12671 (all archival reference are to the PRO). 12. William Colby attempted similar repartee with Leonid Brezhnev at the White House in 1973, but Brezhnev merely blanked him, Colby to CIA, 18 September 1988, enclosing typescript, “The More We Know,” Colby papers, folder 2, box 2, Princeton University Library. 13. Record of an interdepartmental meeting, “Nuclear Detection Schemes,” 11 February 1966, item 29, CRS A452, National Archives of Australia. 14. Omand (MoD) to SoS Defence, “Intelligence Gathering Operations—Ministerial Authority,” 6 February 1974, DEFE 13/985. 15. Private information. 16. “Operational Directive to 23 SAS and Special Reconnaissance Squadron RAC,” Annex to B 2014/10 G (Ops & Plans) 11 September 1962, WO 32/19472. 17. DOAE Project 147, “The NATO Intelligence System,” June 1969, DEFE 48/496. Also private information. 18. French approaches are related frankly in a memoir by General Paul Aussaresses, Pour la France: Services spéciaux, 1942–1954 (Monaco: Éditions du Rocher, 2001.) 19. Nor indeed is there room here to explore covert action and special operations. 20. Lawson to Stephens, “Some Observations About the British Information (& Related) Departments in London (June 29–July 9),” 23 August 1965, file UK Memos, 1 of 2, vol. VI, 7.65–, box 208, NSF Country Filed, Lyndon B. Johnson Library, Texas.

166

British Intelligence During the Cold War

Sources Aitken, Jonathan. 2000. Pride and Perjury. London. HarperCollins Aldrich, Richard J. 2001. The Hidden Hand: Britain, America and Cold War Intelligence. London: John Murray. ———. 2006a. “The US-UK Intelligence Alliance in 1975: Economies, Evaluations and Explanations.” Intelligence and National Security 21, no. 4. ———. 2006b. “Intelligence.” In Saki R. Dockrill and Geraint A. Hughes, eds., Palgrave Advances in Cold War History, 210–239. Basingstoke, UK: Palgrave Macmillan. ———. 2007. “British Intelligence, Security and Western Co-operation in Cold War Germany: The Ostpolitik Years.” In Beatrice de Graaf, Ben de Jong, and Wies Platje, eds., Battleground Western Europe: Intelligence Operations in Germany and the Netherlands in the Twentieth Century. Amsterdam: Het Spinhuis. ———. 2008. “Intelligence Within BAOR and NATO’s Northern Army Group.” Journal of Strategic Studies 30, no. 1 89–122. ———. 2010. GCHQ: The Uncensored Story of Britain’s Most Secret Intelligence Agency. London: HarperPress ———. 2011. From Bletchley Park to a Brave New World. London: Hyperion. Ambrose, Stephen, 1984. Eisenhower: The President, 1953–69. London: Allen & Unwin. Andrew, Christopher. 1985. Secret Service: The Making of the British Intelligence Community. London: Heinemann. ———. 1995. For the President’s Eyes Only: Secret Intelligence and the American Presidency from Washington to Bush. London: HarperCollins. ———.1998. “Intelligence and International Relations in the Early Cold War.” Review of International Studies 24, no. 3: 321–330. ———. 2001. “Bletchley Park in Postwar Perspective.” In Michael Smith and Ralph Erskine, eds. Action This Day: Bletchley Park . London: Bantam. ———. 2009. Defend the Realm: The Authorised History of MI5. London: Penguin Books. Andrew, Christopher, and Oleg Gordievsky. 1990. KGB: The Inside Story of Its Foreign Operations from Lenin to Gorbachev. London: Hodder & Stoughton. Aussaresses, Paul. 2001. Services spéciaux: Algérie 1955–1957. Paris: Perrin. Betts, Richard K. 1982. Surprise Attack: Lessons for Defense Planning. Washington, DC: Brookings Institution. Bickers, R. 2001. “The Business of a Secret War: Operation ‘Remorse’ and SOE Salesmanship in Wartime China.” Intelligence and National Security 16, no. 4: 11–36. Boothroyd, Betty. 2001. Betty Boothroyd: Autobiography. London: Century. Busch, Peter. 2003. All the Way with JFK? Britain, the U.S., and the Vietnam War. New York: Oxford University Press. Cradock, Percy. 2002. Knowing One’s Enemies: How the Joint Intelligence Committee Saw the World. London: John Murray. Daly, R. R., 1982. Selous Scouts: Top Secret War. Alberton, South Africa: Galago.



British Intelligence During the Cold War

167

Davies, Philip, 2005. “A Critical Look at Britain’s Spy Machinery: Collection and Analysis on Iraq.” Studies in Intelligence 49, no. 3: 41–54. Dumbrell, John. 2001, A Special Relationship: Anglo-American Relations in the Cold War and After. New York: St. Martin’s Press. Dunn, David. 1988. The Politics of Threat: Minuteman Vulnerability in American National Security Policy. London: Macmillan. Easter, David, 2004. Britain and the Confrontation with Indonesia, 1960–1966. London: I. B. Tauris. Ferris, John, 1995. “Coming in from the Cold: The Historiography of American Intelligence, 1945–1990.” Diplomatic History 10, no. 1: 87–116. Freedman, Lawrence. 2006. The Official History of the Falklands Campaign, vol. 2: War and Diplomacy. London: Routledge. Fuller, Jack. 1977. “Dateline Diego Garcia: Paved-Over Paradise.” Foreign Policy 28: 175–186. Gaddis, John L. 1989. “Intelligence, Espionage and Cold War Origins.” Diplomatic History 13, no. 2: 191–213. Garthoff, Raymond. 2004. “Foreign Intelligence and the Historiography of the Cold War.” Journal of Cold War Studies 6, no. 2: 21–56. Geraghty, Tony. 1996. Beyond the Front Line: The Untold Exploits of Britain’s Most Daring Cold War Spy Mission. London: HarperCollins. Gibson, Stevyn. 1997. The Last Mission: Behind the Iron Curtain. Stroud, UK: Sutton. Goodman, Michael S. 2007a. “The Dog That Didn’t Bark: The Joint Intelligence Committee and Warning of Aggression.” Cold War History 7, no. 4: 529–551. ———. 2007b. Spying on the Nuclear Bear: Anglo-American Intelligence and the Soviet Bomb. Stanford, CA: Stanford University Press. ———. 2008. “The Tentacles of Failure: British Intelligence, Whitehall and the Buster Crabb Affair.” International History Review 30, no. 4: 768–784. Great Britain. Foreign and Commonwealth Office. 1994, “My purdah lady”: The Foreign Office and the Secret Vote, 1782–1909. History Notes, no.7. London: Foreign and Commonwealth Office Library and Records Department Historical Branch. ———. 1998. Britain and the Soviet Union, 1968–1972, ed. Gillian Bennett and K. A. Hamilton. Vol. 1 of Documents on British Policy Overseas, Series III. London: HMSO. Hennessy, Peter. 1989. Whitehall. London: Secker & Warburg. ———. 2002. The Secret State: Whitehall and the Cold War. London: Allen Lane. Herman, Michael. 1995. Intelligence Power in Peace and War. Cambridge: Cambridge University Press. ———.2001. “The Role of the British Joint Intelligence Committee: An Historical Perspective.” In L. C. Jenssen, and O. Riste, eds., Intelligence and the Cold War: Organisation, Role and International Co-operation, 25–40. Olso: Norwegian Institute for Defence Studies. Herman, Michael, J. Kenneth McDonald, and Vojtech Mastny. 2006. Did Intelligence Matter in the Cold War? Oslo: Norwegian Institute for Institute for Defence Studies.

168

British Intelligence During the Cold War

Holbrook, J. R. 2005. Potsdam Mission: Memoir of a U.S. Army Intelligence Officer in Communist East Germany. Carmel, IN: Cork Hill Press. Hollingsworth, Mark, and Nick Fielding, 1999. Defending the Realm: MI5 and the Shayler Affair. London: André Deutsch. Hughes, Geraint, 2004. “British Policy Towards Eastern Europe and the Impact of the ‘Prague Spring,’ 1964–68.” Cold War History 4, no. 2: 115–139. Jeffery, Keith. 2010. MI6: The History of the Secret Intelligence Service, 1909–1949. London: Bloomsbury. Jeffery, Laura. 2006. “Victims and Patrons: Strategic Alliances and the Anti-Politics of Victimhood Among Displaced Chagossians and Their Supporters.” History and Anthropology 17, no. 4: 297–312. Jones, Clive, 2004. Britain and the Yemen Civil War, 1962–1965: Ministers, Mercenaries and Mandarins—Foreign Policy and the Limits of Covert Action. Brighton, UK: Sussex Academic Press. Kitson, Frank. 1986. Gangs and Counter-Gangs, London: Faber & Faber. Lashmar, Paul. 1996. Spyflights of the Cold War. Stroud, UK: Sutton. Lashmar, Paul, and James Oliver. 1993. Britain’s Secret Propaganda War. Stroud, UK: Sutton. Lewis, Julian. 2003. Changing Direction: British Military Planning for Post-War Strategic Defence, 1943–47. London: Frank Cass. Maddrell, Paul. 2005. “Operation ‘Matchbox’ and the Scientific Containment of the USSR.” In P. Jackson and J. Siegel, eds., Intelligence and Statecraft: The Use and Limits of Intelligence in International Society. Westport, CT: Praeger. ———. 2006. Spying on Science: Western Intelligence in Divided Germany, 1945–1961. Oxford: Oxford University Press. Nelson, Michael. 2011. Castro and Stockmaster: A Life in Reuters. London: Matador. Neustadt, R., and E. May. 1986. Thinking in Time: The Uses of History for Decision Makers. New York: Free Press. O’Malley, Brendan, and Ian Craig. 1999. The Cyprus Conspiracy: America, Espionage and the Turkish Invasion. New York: I. B. Tauris. Peebles, Curtis. 2000. Shadow Flights: America’s Secret Air War Against the Soviet Union. Novato, CA: Presidio Press. Pliatzky, Leo. 1982. Getting and Spending: Public Expenditure, Employment, and Inflation. Oxford: Blackwell. Reeve, John. 1999. Cocktails, Crisis and Cockroaches: A Diplomatic Tail. London: Radcliffe Press. Richelson, Jeffrey. 2006. Spying on the Bomb: American Nuclear Intelligence from Nazi Germany to Iran and North Korea. New York: Norton. Roman, Peter. 1995. Eisenhower and the Missile Gap. Ithaca, NY: Cornell University Press. Schecter, Jerold, and Peter Deriabin. 1992. The Spy Who Saved the World: How a Soviet Colonel Changed the Course of the Cold War. New York: Scribner’s.



British Intelligence During the Cold War

169

Scott, Len. 1999. “Espionage and the Cold War: Oleg Penkovsky and the Cuban Missile Crisis.” Intelligence and National Security 14, no. 4: 23–48. Sheehan, Michael. 1988. Arms Control: Theory and Practice. New York: Blackwell. Smith, Michael. 2003. The Spying Game: A Secret History of British Espionage. London: Politico’s. Vine, David. 2004. “War and Forced Migration in the Indian Ocean: The US Military Base at Diego Garcia.” International Migration 42, no. 3: 111–143. Walden, George. 1995. Lucky George: The Memoirs of an Anti-Politician. London: Penguin Books. Watt, Donald. 1989. “Intelligence and the Historian: A Comment of John Gaddis.” Diplomatic History 14, no. 2: 199–204. Wylde, Nigel, ed. 1993. The Story of Brixmis, 1946–1990. Arundel, UK: BRIXMIS Association. Young, John, and John Kent. 2004. International Relations Since 1945. Oxford: Oxford University Press.

6

The Stasi Confronts Western Strategies for Transformation, 1966–1975 Oliver Bange

If one assumes that the intensifying contacts between East and West both initiated and allowed for pressures for reform, and that this eventually led to the end of the Soviet empire and the collapse of Communist rule in Central and Eastern Europe, the GDR has to be seen as the main victim of détente.1 If, on the other hand, the codification and multilateralization of East-West contacts, behavioral standards, and individual liberties in the “Final Act” of the 1975 Helsinki Conference on Security and Co-operation in Europe (CSCE) really constituted the climax of détente, as is often claimed—then the GDR indeed collaborated in its own demise. If so, the question is when and why the GDR’s leadership entered into this seemingly irreversible process. So far, two arguments have been put forward. The starting point of the first—with a certain appeal among former GDR dissidents—is that the GDR’s ruling elite were convinced that history was on its side, that it had built a truly socialist society, which in time would lead to communism—and that therefore jeopardy could only come from the outside. The argument also assumes that the GDR’s institutions—and the Ministerium für Staatssicherheit (MfS, or Stasi), in particular—and decision-makers chronically underestimated the potential for dissent in their own society and instead focused on “counterrevolution” from the West; erroneously, as it turned out, because there was no center, not even a plan, because on the other side of the Iron Curtain détente-minded politicians had long deserted the just cause of liberating those under Communist rule.2 In this prism, the GDR’s 170



The Stasi Confronts Western Strategies for Transformation, 1966–1975

171

entanglement in the domestic consequences of the so-called CSCE-process would have been entirely self-induced, caused by ideological stubbornness, even stupidity. The second line of argument is provided by former GDR officials like Siegfried Bock or Norbert Podewin.3 It seems that this group, too, shared a common background, having been in their late thirties or early forties during the early 1970s, with distinguished and still promising careers either in the Socialist Unity Party (Sozialistische Einheitspartei Deutschlands; SED) (Podewin, for example, worked closely with Norden, Ulbricht, and Ebert) or the GDR’s ministries (Bock had risen to the head of the Foreign Ministry’s analysis and planning department)—and above all a will to tackle vitally necessary reforms. They argue that Soviet self-interest in security from and cooperation with the West essentially hindered the pursuit of socialism, both within the GDR and also across Germany. They also imply that the shift from Ulbricht to Honecker was crucial because the new party chairman did not defend GDR interests against Moscow. In effect, this amounted to treason, thereby missing what was perhaps the last opportunity for comprehensive reform in the GDR. In order to judge these two theories and to square them with research findings about the Stasi’s knowledge of Western aims, its estimate of Eastern weaknesses, and the resulting counterstrategies, we need to start with two aspects of the background: the Western strategy for change and the GDR’s position during détente. The Western Strategies for Change Versus Tactics for Maintaining the Status Quo West German Chancellor Konrad Adenauer—the epitome of a “Cold warrior”—discussed the changing geopolitical position of the Warsaw Pact, and of the Soviet Union in particular, with France’s President Charles de Gaulle in the early 1960s.4 A strong China on the USSR’s eastern borders and resurgent nationalism in its Western sphere of influence (a point particularly stressed by de Gaulle), global overstretch combined with economic and technological pressures—all seemed to put Moscow increasingly and foreseeably on the defensive. Adenauer and de Gaulle readily agreed that time was working on the side of the Western democracies. However, if the Federal Republic wished to achieve German reunification through a long-term policy of détente and entente, de Gaulle warned that it had to acknowledge GDR statehood and

172

The Stasi Confronts Western Strategies for Transformation, 1966–1975

the Oder-Neisse frontier between Germany and Poland established in 1945, denounce the 1938 Munich Pact under which Nazi Germany had annexed the so-called Sudetenland, disclaim any intention to acquire nuclear weapons, and remain economically and militarily integrated into the West. Much of this was, of course, also in France’s interest. But Adenauer did not believe he was in a position to grant any of these provisions, even though they might be necessary to initiate the kind of process envisaged. A year before his death, he finally dared to call the Soviet Union a state in search of peace in Europe5 — only to retreat immediately when faced with the outrage of the Christian Democratic Union and Christian Social Union of Bavaria (CDU/CSU) delegates present. De Gaulle lacked the partner he needed in the FRG for his détente strategy, even while Adenauer was in office and certainly thereafter— which was one reason he began taking an interest in Willy Brandt, then only the mayor of West Berlin. Meanwhile in Washington prospects of a more cooperative approach to the Soviet hemisphere emerged against the experiences of 1961–1962 in Berlin and Cuba, reminders of the dangers inherent in any East-West crisis in the nuclear age. In order to move away from purely confrontational approaches, relations with Communist regimes in Europe needed to be “civilized” (and not with the Soviet Union alone—as a prominent German historian erroneously remarked recently).6 The question was: how? And it was President Lyndon Johnson who within the first months of his assuming office in 1964 ordered a structured effort to devise a comprehensive new strategy.7 Numerous studies about the possibilities of economic détente, “cabalism” in Eastern Europe, disarmament and nonproliferation were drawn up and introduced to the public via Johnson’s “bridge-building” speech in October 1966. Although the term “competitive co-existence” was written into the speech by Zbigniew Brzezinski, rather than a “zero-sum game,” Francis Bator—the brain behind Johnson’s concept—intended simultaneous synergistic effects for both sides (which would make it even harder for Moscow and other Communist regimes to resist the offer).8 West German ideas about a new policy toward the East—meant to tap into French and American concepts aiming to end Germany’s division—were developed by Willy Brandt and Egon Bahr in Berlin, later during the grand coalition of the late 1960s by Kurt Georg Kiesinger and Günter Diehl in the Chancellor’s Office, and also throughout this period by Walter Scheel (later also Hans-Dietrich Genscher) and Wolfgang Schollwer within the FDP. All



The Stasi Confronts Western Strategies for Transformation, 1966–1975

173

of their concepts banked on the impact of “soft” power, communication, and cooperation, famously termed a policy of “small steps” by Brandt. Through a multiplication of contacts on all feasible levels, internal pressures for reform would strengthen, leading to transformation and eventually to the end of the monopoly of Communist parties. Even during the first steps toward this EastWest network, as Bahr argued, one had to provide for the “irreversibility” of the overall process.9 West German “Ostpoliticians” widely believed that they could play an important role within the West’s détente approach, because—as Brandt and Bahr believed—of the special appeal of social democracy among Communists in the East,10 and more generally because of the existence of one German nation beyond blocs and borders. Both in Paris and London, the transformation rationale behind détente strategy remained an elitist affair, throughout the Gaullist and post-Gaullist period. It seems that Britain’s Prime Minister Harold Wilson had been convinced of the virtues of a transformation approach by Brandt in the summer of 1969, but was able to throw his full weight behind it only after the Labour Party regained power in 1974. And even then the Whitehall group of experts who were fully in the know remained exceptionally small, comprising key personalities like Crispin Tickell and Roderick Braithwaite.11 Equally, in Paris, de Gaulle’s successor, Georges Pompidou, quickly had to adapt Gaullist détente to the new dynamic in European relations triggered by Brandt’s Ostpolitik after 1969. Based on the experiences of the Prague Spring of 1968, Pompidou and his adviser Jean-Bernard Raimond feared that a similar event in East Germany—which could easily be initiated by Brandt’s fast-moving opening toward Eastern Europe—might result in serious bloodshed and perhaps even war and nuclear disaster. On the other hand, Pompidou was a strong advocate of personal liberties and a driving force behind what would become Basket III of the Helsinki Final Act.12 Whereas Washington might have triggered thinking about détente across the West as early as 1961, the new Nixon administration of January 1969 took a very different course. Richard Nixon’s personal fear of Communist subversion—deeply rooted in his own political career—was matched by Henry Kissinger’s distrust of Germans, East and West. Unlike their European partners, both perceived the United States to be on the defensive against a growing ideological and military threat from Communism.13 Accordingly Nixon announced an “era of negotiations” concentrated on status-quo deals between the superpowers, while ideas about bridge-building and transformation

174

The Stasi Confronts Western Strategies for Transformation, 1966–1975

continued to influence thinking within the State Department.14 Key to the conflict between grand designs for détente was the notion of the transformability of the Communist regimes in Eastern Europe. In the summer of 1969, in a conversation with Walter Scheel of Germany’s Free Democratic Party, Nixon denied that this was a possibility, arguing that Communist rulers, if they were to accept any change at all, would only react to pressure. Scheel, then only the opposition leader of a party that commanded well below 10 percent of the national vote, immediately contradicted him: “But they could change!”15 Contrary to the established (and politically loaded) picture, there was nevertheless a kind of blueprint for a Western policy of transformation. There existed not one grand design but several; all complementary to each other, however, and intrinsically linked through a broad platform of shared assumptions, strategies, and goals. The first was a distinctive sense of representing a stronger, economically more viable, technologically advanced and certainly more liberal and democratic society. The second characteristic concerned the perception of the other side, with all these strategists holding to the mutability of Communist rule (to the point of its eventual demise). A third assumption in common was a threat perception that differed markedly from earlier in that the USSR was seen both on the defensive in Europe and cautious not to escalate into a confrontation. This was mirrored by a strong belief in the long-term effects of “soft” versus “hard” power. From this view crucial estimates emerged about likely mechanisms for transformation and the appropriate instruments to initiate, steer, and if possible control them. Perhaps the most important of these was the paradox of guaranteeing security and stability in international relations as a precondition to allow for domestic changes to eventually overcome the current state of affairs—or in Egon Bahr’s famous catchphrase: “accepting the status quo in order to change it.” Communication and contacts were the means to initiate, steer, and control. But only domestic pressures were generally believed to result in a sustainable reform process. Furthermore, the Warsaw Pact was no longer seen and treated as a single, monolithic bloc. Détente and cooperation were based on a high degree of differentiation between the Pact’s member countries and their individual problems and desires. And the relative prosperity and liberalism of Western Europe was consciously seized upon both as a magnet for Central and East Europeans and as a token of promise for an all-European future.



The Stasi Confronts Western Strategies for Transformation, 1966–1975

175

Finally, within this general blueprint for détente and transformation, a number of special national interests can be identified. Security from the Germans, and especially from a reunited Germany, was high on everybody’s list, though with great variations. Even the “bridge builders” in Washington laid stress on a superpower détente, for the obvious reason of averting nuclear war and achieving nuclear nonproliferation. In Bonn, the key proponents of a new Ostpolitik in all parties were acutely aware that even if a successful transformation process would lead to the demise of Communism, this did not guarantee reunification, which was why they aimed at an all-European security system within which a united Germany would pose no threat to anyone. In London, the Wilson, Heath, and Callaghan governments focused in varying degrees either on getting into the EEC or thereafter on their new role in Europe—and very much saw East-West contacts as a function of this historical watershed in Britain’s foreign policy outlook. In the smaller states both within and outside NATO and the EEC, Ostpolitik-minded politicians like Jens-Otto Krag of Denmark and Bruno Kreisky of Austria were very interested in a multilateralization of détente both as a guarantee against excessive influence from their larger partners and—particularly in the case of the neutral and nonaligned (N+N) states—as a means to overcome bloc(ked) positions.16 The GDR in the Era of Détente: Conf licting Perceptions and Strategies, 1965–1975 East Berlin’s position during détente, its reactions, and strategies underwent several shifts and one important about-turn, personalized in the changeover from Walter Ulbricht to Erich Honecker between 1969 and 1971. Ulbricht, the SED’s first secretary and the strong man in East Berlin, responded to the Neue Ostpolitik designed by Brandt and Bahr with both a (hardly surprising) antagonism and a (remarkable) parallelism. His counterstrategy was neither necessarily nor continuously the kind of policy advocated by the majority of the SED’s Politburo, particularly by the group forming around Honecker. The same was true for the perceptions and reactions in the Politburos of the CPSU and of other Warsaw Pact Communist Parties. Ulbricht’s lifelong dream was a Communist Germany. The experience of Weimar taught that the precondition for this was a united front of workers and employees—termed a Volksfront (peoples’ front) in the 1930s and an

176

The Stasi Confronts Western Strategies for Transformation, 1966–1975

Aktionsgemeinschaft (community for action) in the 1960s. In Ulbricht’s view, the German Social Democratic Party (Sozialdemokratische Partei Deutschlands; SPD), with its “bourgeois” idea of individual liberties, stood like a block between socialist revolution and its “masses.” On the other hand, only the voters, supporters, and members of West Germany’s reform-oriented social democratic party and trade union organizations could possibly bring the broad support necessary for revolutionary change. In the early 1960s—in parallel to the first détente concepts in the West—Ulbricht therefore devised a strategy of circumventing the SPD’s leadership through a direct “planned orientation toward the working classes and toward the trade unions and also toward the social democratic parties”17 in West Germany.18 Against this background it is hardly surprising that old-style revolutionaries like Ulbricht and Erich Mielke maintained a “life-long obsession”19 with German social democracy. In their self-perception, Germany’s Communists possessed unique experience and expertise in dealing with their social democratic adversaries. This they believed would enable them to identify and foil the real strategy and goals behind the new Ostpolitik and its “thrust”—as Ulbricht liked to call it—against the GDR. East German Foreign Minister Otto Winzer’s much-cited immediate characterization of Bahr’s famous speech at Tutzing (“Wandel durch Annäherung”—change though rapprochement) as a recipe for “aggression in bedroom slippers” provides an indication of the early and correct interpretation of Brandt’s Neue Ostpolitik in East Berlin. While the SPD remained in opposition, the GDR’s “Western strategy” therefore focused on presenting German Communism to SPD voters in the FRG as their natural ally in the struggle against the societal conservatism of the CDU government and its rearmament policy.20 Ulbricht remained convinced that the GDR could catch up economically with the FRG within four or five years (i.e., by the early 1970s) with the help of its allies, particularly the USSR. This would curb the FRG’s economic magnetism and the GDR would advance to become a “model” of the social progress made in the East, serving as a showroom to the West. The logic of this offensive approach demanded at least a partial opening toward the West—a truly risky affair for a state that had recently had to be stabilized by building the Berlin Wall. Consequently, Ulbricht’s repeated attempts at a political offensive against the West were vetoed by Moscow, and he was removed in 1971.21 The situation was further complicated in late 1966. After the SPD had joined the conservative parties in a grand coalition, a direct appeal to SPD



The Stasi Confronts Western Strategies for Transformation, 1966–1975

177

members to topple the FRG’s conservative rulers was tantamount to a call to fight their own party leadership—and was thus likely to backfire. With Brandt and Herbert Wehner in key ministries, West German efforts to implement a new Ostpolitik soon intensified—and in January 1967, the Romanians broke ranks and agreed to diplomatic relations with Bonn.22 For Ulbricht, 1967 became a frustrating year of defensive actions. He promised Soviet deputy foreign minister, Vladimir Semyonov, to “correct all language based on the previous situation, for example concerning reunification, national unity, a united peaceful Germany. . . . We had suggested the agreement on border passes in the interest of détente and with the intention to move the SPD further to the left. This did not succeed. This is why we now have to engage in a conflict of principles.”23 When finally, in the autumn of 1967 and in the spring of 1968, the longexpected calls for social reform and unrest gathered momentum in West Germany, Ulbricht sensed that he had arrived at the long-awaited opportunity, both for himself and for Communism.24 Even the early reforms in Czechoslovakia fitted into this overall image.25 However, when the Czechoslovak Communist Party under Alexander Dubček was seen to be increasingly losing control of the unfolding dynamic of the reform process, it was once again Ulbricht who drew the attention of the Warsaw Pact’s leaders to the corrupting influences of “social democratism.”26 He demanded the immediate intervention of the “fraternal states,” which eventually happened in August 1968, much too late for Ulbricht’s liking.27 Until the end of his life, he would accuse the Czechs of having spoiled a historical opportunity for German Communism. By October 1968, once reform movements in the Communist bloc were under control and the Warsaw Pact united as perhaps it had never been united before, Ulbricht began demanding a “counteroffensive” from Moscow against Western détente and Ostpolitik—or at least a free hand and economic support for a unilateral approach by the GDR.28 A year later, in October 1969, Ulbricht succeeded one last time in coercing the SED’s Politburo into accepting his forceful ideas on how best to handle détente and Ostpolitik. Expecting the new social-liberal government in Bonn to intensify attempts at “penetration,” Ulbricht declared: “If Brandt makes a new Ostpolitik, then will we make a new Westpolitik, and one that truly deserves its name. We will make him sweat.”29 By that time, it seems that Ulbricht was both out of luck and perhaps had also lost his sense of what could realistically be achieved. The louder

178

The Stasi Confronts Western Strategies for Transformation, 1966–1975

and more impertinent his demands became, the more concern they raised in Moscow over eventual repercussions for the USSR’s own policy of “peaceful coexistence” with the West. Without the former restrictions of an antiOstpolitik majority among conservative cabinet members, Bonn’s negotiations with other Warsaw Pact capitals advanced rapidly after 1970—adding credibility, weight, and a wealth of new tactical possibilities to Chancellor Brandt’s approach toward Central and Eastern Europe. Ulbricht’s demands for a continuation of the revolutionary fight against and within West Germany correctly followed and cited Communist ideology and dogma. But the politics pursued by Brezhnev, Ceauşescu, Kádár, Gomułka, and others had long set out on a path that Western terminology characterized as “national Communism”—a description indicating the combination of status quo policies and the prioritization of national interests. From February 1970 on, with Honecker increasingly in control, the GDR’s official position on intra-German relations became more intransigent. In fact, the course of the new SED leadership became so restrictive that Moscow already feared a public loss of face for Brandt, and accordingly warned East Berlin to start preparations in earnest for what would become the Erfurt and Kassel meetings.30 East Berlin had little say in any of Bonn’s Eastern treaties, nor in the negotiations leading up to the Quadripartite Agreement on West Berlin. This was particularly true when Bonn and Moscow compromised over a “letter on German unity” to be connected to the Treaty of Moscow—thereby acknowledging the possibility of German reunification and providing terminology for the entire détente period that worked against the GDR’s interests. Honecker and Ulbricht were merely informed by Andrei Gromyko that “the line taken by the Brandt government . . . will be supported”31 and were told in a letter by Leonid Brezhnev that the “SED’s principal course ought to be one of consistent delimitation.”32 Essentially this was a double warning: against the evil brother in the West and the big friend in the East. Within a few months, the SED Politburo’s agreed policy posture shifted from Ulbricht’s concept of intensified class struggle to Honecker’s posture of drawing a dividing line and digging in. In early 1971, the outlook was as gray as so-called actually existing socialism: “The socialist countries will have to face the task of foiling this imperialist strategy and tactic for many years ahead.”33 Throughout détente, the politics of linkage enacted by Washington, Moscow, and Bonn did not improve the GDR’s position. The ratification procedures for the Eastern treaties, the Quadripartite Agreement, SALT, and



The Stasi Confronts Western Strategies for Transformation, 1966–1975

179

eventually even the intra-German Basic Treaty were all linked, both to each other and to the CSCE and Mutual and Balanced Force Reductions (MBFR) negotiations. By then, the GDR’s leadership was indeed trapped in a process of multilateral détente, which constantly widened and increasingly unfolded its own dynamic. Significant changes to language, national images, and selfperceptions promised important long-term effects. This was nowhere more apparent then in the early CSCE process leading up to the Helsinki agreement in 1975. Human rights and free movement (enshrined in Basket III of the Final Act), better working conditions for more foreign journalists, economic and technological exchange, military détente and the “peaceful change of frontiers”—all of these issues contained special threats for the GDR. East German diplomats were under no illusion about Western objectives, perceiving that the proposals tabled by NATO and EEC member states were “aimed at eroding [our] sovereignty by means of multilateral agreements and broad ‘freedom of movement of persons and ideas.’”34 Under considerable pressure from Moscow not to stand in the way of a successful conclusion of the conference (on which Brezhnev had staked his personal reputation), Honecker finally gave in. In return, the GDR got the kind of international recognition that it had been craving for such a long time, plus its share in East-West economic contacts. It also seems that Honecker thereby also widened his personal room for maneuver, resulting in a rather special relationship between him and Helmut Schmidt, initiated during their first encounters at Helsinki.35 Three illusions shared by the Politburos in East Berlin and Moscow might have eased the CSCE decision: contempt for “soft” factors in comparison with “hard” facts (such as numbers of tanks or nuclear warheads); confidence in the abilities of the Stasi to cope; and belief that the mere existence of the “German peace border,” with a new generation growing up within it, would guarantee a socialist future for East Germany.36 But was this really the decisive moment for the GDR’s future? The Stasi Under Détente—Information, Analysis, and Strategies Information gathered is never the information processed, and information processed is never the information that serves as a basis for decision-making. This is particularly true of items and combinations of information gathered through and analyzed in secretive channels. This might serve as a warning

180

The Stasi Confronts Western Strategies for Transformation, 1966–1975

both about the fragmentation of information eventually reaching East German decision-makers and about the peculiarities of archival sources from secret services in the GDR or elsewhere. First, the quality and quantity of information obviously depends on the position and access of the sources and on the ways in which they are handled. The need to process the wealth of information gathered involves standard (or nonstandard) selection procedures and technical filters as well as ideological and personal filters. All of this constantly influences both the analysis and the interpretation of material. The result is always (and not just in the GDR) a distorted image of reality—a “perception.” And perceptions provide the essential framing for decision-making, individually or collectively. The key question for historians dealing with intelligence material must therefore be: what impact did the source, the filter, the analysis, the interpretation, and the perception have on (a) decision-makers, (b) decision-making processes, and (c) the further course of events? Estimating the Impact of Intelligence Information So where and what were the key Stasi sources? Part of the answer lies in the Stasi’s double function as an instrument of domestic control and foreign espionage. (a) The domestic framework. Whereas domestic intelligence provided information about breaches of ideological orthodoxy, public opinion surveys provided by the Stasi that fed into decision-making on international affairs were more realistic. (b) The bloc framework. The Stasi also gathered official and sometimes also less official information from the SED and Warsaw Pact meetings on the positions of other Soviet bloc countries. (c) The international framework. Since the end of the East-West conflict, the successes of Stasi espionage against the West and other neutral countries have become legend. The Stasi’s espionage department— the Hauptverwaltung Aufklärung (HVA)—placed and recruited agents at the very heart of NATO and the European Community. Before Rainer Rupp—“Topaz”—joined NATO headquarters in 1977, both organizations were already an open book to the Stasi. Information from Brussels went far beyond these institutions, also providing invaluable details about the various member countries. This source of information about the United States, Britain,



The Stasi Confronts Western Strategies for Transformation, 1966–1975

181

France, and other (non-German speaking) Western countries was complemented by relatively few spies in the respective capitals, more so by information from other “friendly” services, and—“best of all”—by information from within Bonn’s ministries.37 (d) The German framework. The FRG remained the focus of HVA espionage. The department ran informants in all parties, among them an assistant to the SPD council,38 a secretary to the CDU council, and a member—William Borm—of the FDP council.39 It also infiltrated almost every ministry in Bonn. The information gathered in the Auswärtiges Amt afforded insights into Western diplomacy around the world; and the value of material from the ministries for finance, economics, and intra-German relations was self-evident. Yet, despite numerous sources within the ministry of defense, it seems that the best information about the Bundeswehr still came from Brussels. Semi-official institutions with an assumed relevance for intra-German relations and Western détente strategies (such as the Friedrich-Ebert-Stiftung, the Konrad-Adenauer-Stiftung, the Friedrich-Naumann-Stiftung, the Stiftung Wissenschaft und Politik, the Bundesinstitut für ostwissenschaftliche und international Studien, etc.) became prime Stasi targets in the 1960s. At about the same time, the HVA infiltrated all West German institutions concerned with nuclear technology and power.40 Throughout the 1970s, industrial and technological espionage rapidly gained in importance. And contacts with Western journalists were often built on research cooperation and even trust.41 Nevertheless, despite the Stasi’s apparent omnipresence in West Germany, despite the legion of spy stories and scandals, despite its self-image and despite public belief—it never succeeded in breaking into the core group of Ostpolitik or détente decision-makers. One reason for this may have been that the secret agendas and long-term strategies behind détente/Ostpolitik/ bridge-building were generally thought up in pairs as described earlier. This, of course, allowed Western decision-makers to play their cards “close to the chest.”42 The information gathered, sometimes even in the closest circles of these core members, necessarily remained fragmented and thereby distorted. Combined with analytical tunnel vision and preselection, this led to a number of misrepresentations and even misconceptions, such as the assumption that in gaining access to the Bundesinstitut für ostwissenschaftliche und

182

The Stasi Confronts Western Strategies for Transformation, 1966–1975

international Studien in Cologne, East Germany had infiltrated the think tank of Bonn’s new Eastern policy.43 That was the general picture throughout the period under discussion, and whether and how the new electronic listening devices of the late 1970s–early 1980s changed it remains an open question.44 What, then, did the Stasi know about Ostpolitik? How did the Stasi react? What role did it play within the GDR’s own policy in the détente era? The Conceptual Phase of Ostpolitik and Détente, 1961–1966 It appears that the SED leadership, above all, Ulbricht, Norden, Hager, and Winzer, almost immediately understood the rationale emerging behind Brandt’s speeches on Ostpolitik and “small steps” in intra-German relations. This was of course much helped by the public debate that ensued between Brandt and Kennedy’s team soon after the erection of the Berlin Wall in the summer of 1961, the revision of Brandt’s strategic concept thereafter (though not of his central ideas), the leaking of key memoranda to the press, and a long series of highly publicized speeches by Brandt on East-West relations, among them those at Harvard and at Tutzing.45 All of these were surprisingly revealing, but none as much as Brandt’s speech at Tutzing in September 1963— which told even more about the concept of Ostpolitik than Bahr’s notorious speech only hours earlier.46 The elements of transformation, intellectual subversion, and counterrevolution clearly confirmed Communist preconceptions about “social democratism.” The SED leadership therefore became alarmed, and it seems that only on their explicit order did the Stasi then try to obtain as much information as possible on this new development. At the time, the Stasi possessed neither a channel to Brandt nor an informant in his direct surroundings in West Berlin.47 And while the Stasi had successfully penetrated the SPD’s Ostbüro, Brandt and his team had begun to distance themselves from its openly subversive Cold War–style operations. In early 1964, Brandt’s reflections on starting negotiations with East Berlin over temporary documents of passage—Passierscheine—for West Berliners to visit relatives and friends in the other part of the city (the first of the small steps) led to the establishment of the first unofficial contacts. Hermann von Berg, officially a journalist working for the head of the GDR government, Willi Stoph, was chosen as a go-between with Brandt via Klaus Ellrodt, a journalist from West Berlin, and Dietrich Spangenberg, the head of West Berlin’s



The Stasi Confronts Western Strategies for Transformation, 1966–1975

183

Senate Office.48 Another secret channel to Egon Bahr, Brandt’s spokesman, was established through Michael Kohl of the GDR foreign ministry. For his very first meeting with Spangenberg, von Berg had already received revealing instructions: obtain information about the balance of power within the SPD—between Brandt and Wehner and elicit SPD reactions to the “memorandum on German policy.” 49 Above all, the question remained: “What does Brandt understand by his ‘policy of small steps’ in Germany?”50 In late 1965 and early 1966, this system of secret channels intensified when Wehner tried to establish a direct link to Stoph, whom he believed to be more accessible and cooperative than the stubborn Ulbricht, 51 and when Ulbricht used the cover of Stoph to float the idea of an exchange of speakers between the SED and the SPD. Parallel to these double-edged contacts to the SPD, unofficial negotiations with the Erhard government in Bonn over Passierscheine for West Germans were initiated via the honorary president of the FRG’s employers’ organization, Constantin Paulssen. However, the history of these channels is by no means as one-sided as Stasi historians (Knabe: “all channels led directly into the hands of the state security”) have described it. The heads of the Stasi were indeed thrilled by this new development, believing that they had finally found a way to obtain essential information about Ostpolitik (much in demand with the Politburo). But they also became somewhat overexcited, in that they thought they could win over even some of Brandt’s closest associates, like Spangenberg and his speechwriter Günter Struwe (later to become undersecretary of state in Bonn and head of the ARD, the FRG’s biggest TV broadcaster). Meanwhile, the other side—and especially Spangenberg—used these channels to plant disinformation about Ostpolitik directly with the GDR leadership.52 Therefore, while Ostpolitik and détente were still in their conceptual phase (on all sides), the Stasi’s interest and knowledge (not even to speak about prognoses about its eventual impact) remained relatively marginal, particularly when compared to its effectiveness in later years. The Beginnings of Ostpolitik and Détente, 1966–1969 When the SPD joined the grand coalition in December 1966, this heralded a new focus on the FRG’s Eastern relations, necessitated a revision of Ulbricht’s strategy, and raised awareness in East Berlin about the dangers of Ostpolitik and détente—and thereby also agitated the Stasi. Four factors appear to have

184

The Stasi Confronts Western Strategies for Transformation, 1966–1975

dominated reactions in East Berlin: (a) considerations about Ostpolitik as a “new approach,” focusing on its intended instruments and embedded strategies; (b) an awareness that Bonn could not pursue an Eastern policy of its own but had to coordinate Ostpolitik with its Western allies; (c) an enhanced threat perception based on the combination between incoming information on the new strategy and prejudices about “social democratism” as the major enemy in the West; and (d) the need for Ulbricht to adapt his strategy from attempting to cooperate with the SPD to circumventing its leadership by forging coalitions at the levels of the SPD rank and file and of the trade unions. Throughout, the Stasi played a distinctive role. Stasi espionage and analysis brought essential information about the double-edged strategy behind Ostpolitik pursued by the SPD leadership under Brandt. Although the fundamental rationale of Ostpolitik, its long-term strategy and goals, had already been recognized in the early 1960s, the era of the grand coalition brought the first—and expected—opportunities to implement new openings toward the East on a wider scale. In East Berlin’s perspective, the focused efforts pursued simultaneously by SPD leaders, and particularly by Brandt and Wehner, to safeguard their own party and West German trade unions against Communist and East German influence on this process introduced a new element. Informants inside the metal workers’ union, for example, produced evidence on how and when the social democratic leaders and the head of the union, Otto Brenner,53 enacted their strategy of Linksabschirmung (protection against the Left) in this organization, deeply rooted in traditions of the workers’ movement in Germany.54 Similarly, the SED’s leadership was supplied with details of the activities of the Heinz Oskar Vetter, deputy chairman of the miners’ union and after 1969 head of the powerful German Trade Union Federation (Deutscher Gewerkschaftsbund; DGB), the central organization of all trade unions in West Germany. When pressed by the DGB’s youth organization, Vetter reacted entirely along the lines set out by Brandt: the DGB youth should sharpen its wits and defenses in disputes with Communists—but should never be allowed “to move too far” away from DGB policies and fall prey to Communist influence.55 Apparently, the DGB was also serving Brandt’s Ostpolitik outside its normal framework in that it engaged in organizing seminars on anticommunist strategy in Berlin—for example (and particularly unforgivably from an SED perspective), for Bundeswehr officers in Berlin.56 According to Stasi information, by spring 1968, the SPD’s leadership was in fact leaning so strongly on the trade unions for support in its anticommunist



The Stasi Confronts Western Strategies for Transformation, 1966–1975

185

strategy that several union leaders vehemently protested against the Mundtotmachen (“mouth killings,” or silencing) of some of their members disliked by the SPD.57 But the SPD’s mirroring of the GDR’s official “delimitation” policy was also applied to internal SPD structures—with very similar effects.58 As in the trade unions, by 1969, leftist members had become so annoyed with being controlled by their superiors that heated discussions were launched over the “right” course in dealing with the GDR, particularly concerning its recognition and that of postwar borders.59 However, what the Stasi seems not to have captured or understood was Brandt’s tactical approach to the use of leftist voices like Jochen Steffen in order to give the SPD Left a voice in the discussions over Ostpolitik—thereby integrating it into his overall strategy.60 Throughout the years of the coalition between the CDU/CSU and the SPD in Bonn, the majority of information about Ostpolitik and Western détente supplied to SED Politburo members focused on the social democrats—obviously mirroring and reconfirming the prejudices of the GDR’s leadership. Thus, for example, the Stasi submitted an analysis of Brandt’s Ostpolitik originating from an academic with contacts to Klaus Schütz, then the governing mayor of West Berlin and himself a confidant of Willy Brandt’s. The chapter on the “social democratism” embedded in the neue Ostpolitik strategy could only confirm suspicions in East Berlin. But what was perhaps even more important, through this paper, Ulbricht and his colleagues got to know that the SPD leadership had been well informed about the SED’s strategy of transforming West German trade unions and the SPD.61 Information from various sources—not just the Stasi’s spy section—and particularly the analysis of the role played by West German social democrats and trade unions during the Prague Spring of 1968 thereby served as a kind of self-fulfilling prophecy when submitted to the SED leadership. They confirmed DGB attempts—apparently in close coordination with the SPD’s leadership—to support the building of independent trade unions in the CSSR,62 large-scale financial support for Czech reformists from West Germany, and ever-intensifying contacts on all levels between the CSSR and the FRG. Mielke was alarmed when a Reuters correspondent wrote about the removal of border installations on the border with Bavaria as early as April 1968—and the HVA checked on this (false) report within forty-eight hours from both sides of the border.63 The longer the Prague Spring lasted, the more Stasi intelligence reports appear to have confirmed the fears of Politburo members. By July 1968, the SED’s leadership had become convinced that Czech “reform”

186

The Stasi Confronts Western Strategies for Transformation, 1966–1975

had already led to “counterrevolution” and could end up in the worst of all scenarios—“social-democratization.” Albert Norden—the SED’s chief Western expert—warned the Party chairman that the SPD leadership already saw itself as the “forerunner of a counterrevolutionary ‘social-democratization of Eastern Europe.’” Peter Florin telegraphed from Prague about an “increasing trend toward the social-democratization of the CP of Czechia.” And Ulbricht spoke at a decisive meeting of the “Warsaw Five” about a “systematically organized and increasing” threat.64 After the Warsaw Pact invasion of Czechoslovakia in August 1968, Stasi information once again added valuable detail to Ulbricht’s ideas for an offensive in the FRG. This comprised details about Bonn’s decision to continue with its Ostpolitik, albeit in an adapted form,65 about the SPD and the trade unions’ internal discussions about further contacts with the SED and the GDR’s trade union association, the FDGB,66 and finally about the anticommunist coordination efforts undertaken after the Prague Spring under West German guidance, particularly in the Socialist International by the SPD leadership67 and through pooling intelligence and analysis of the various institutions set up in the FRG and in NATO for academic research on the Warsaw Pact.68 The Era of Bilateral Treaties Already during the negotiations between the SPD and FDP over the eventual formation of the social-liberal coalition, Stasi intelligence obtained information that the new Ostpolitik would not only continue but would be intensified by a renunciation of the Hallstein doctrine (which prohibited diplomatic relations with any state that recognized East Germany, other than the USSR) and de facto recognition of both the Oder-Neisse line and the GDR.69 By the end of 1969—only weeks after the formation of the new cabinet—it generally acknowledged within the Stasi that this new approach, tantamount to a “silent penetration,”70 would be more threatening to the GDR’s existence than any previous Western tactics. A Stasi briefing paper for consultation with the KGB in October 1969 listed new forms of subversion and concluded a “significant intensification” of “softening up, wedge-driving, diversion” tactics through “social democratism” and the impact of the convergence theory. The description of the goals and tactics behind Ostpolitik—albeit using socialist terminology—reflected Brandt’s Fahrplan (timetable) precisely.71 The head



The Stasi Confronts Western Strategies for Transformation, 1966–1975

187

of the KGB, Yuri Andropov, shared this view in a speech in Berlin shortly afterward, declaring that “Brandt’s intention is the liquidation of the GDR . . . which is tantamount to the liquidation of socialism.”72 But in addition to societal countermeasures, he also demanded serious negotiations with Bonn, particularly regarding the recognition of borders and the renunciation of nuclear weapons by West Germany (two essential Soviet interests). In late 1969, Brezhnev asked the GDR for a policy of strict “delimitation” in its dealings with West Germany. Mielke was all too willing to follow, and as early as December 1969—only six weeks after the decision for an intensified Westpolitik in Dölln—protested in the Politburo against Ulbricht’s course. His explanations were less stringent than robust: The SPD leaders would perceive themselves as “better experts in subversion” than the Americans. Their Ostpolitik was “an intensified strategy for political-ideological softening-up”; their weapon was to be the infamous “social democratism,” and everything would be done according to a tactical blueprint. What was needed now was true “revolutionary attention.”73 Accordingly, Stasi preparations for the Stoph-Brandt meetings, first at Erfurt and then at Kassel, were geared toward demonstrating the GDR’s immovability and were aptly codenamed “Confrontation” and “Confrontation II.”74 After Brandt was nevertheless welcomed by GDR citizens in Erfurt, Stasi analysts judged that the SPD posed the “highest danger” to the GDR.75 Mielke continued to keep a watchful eye personally on the various negotiations stemming from Brandt’s now much less restricted policy of contacts with the East. His personal assistant evidently kept a special file on Bonn’s Eastern negotiations for his minister.76 Mielke and his officers attempted to monitor not only the dealings of the GDR’s Soviet “friends” with Bonn, but also the German-German negotiations between Egon Bahr and Michael Kohl. Apart from informing his minister and Honecker,77 Kohl—a senior member of the GDR’s Ministry of Foreign Affairs—also reported on these meetings directly to Mielke.78 The Stasi chief could then cross-check this information against material obtained from the Bonn ministries, against reports from an informant in Kohl’s immediate environment,79 against information provided by the SPD politicians Karl Wienand and Herbert Wehner via a secret back channel running parallel to the official FRG-GDR negotiations, unknown to Bahr,80 and occasionally also against eavesdropping.81 For obvious reasons, the negotiations in late 1971 over the so-called Traffic Treaty and the exchanges between Bonn and East Berlin over agreements on cultural exchange and

188

The Stasi Confronts Western Strategies for Transformation, 1966–1975

working conditions for journalists attracted the particular attention of Mielke and his specialists.82 Apart from sheltering the GDR’s socialist society from undue capitalist influences, Mielke also pursued the Ministerial Council’s agreed strategy of using the Traffic Treaty to maximize the GDR’s “hard” Western currency income.83 In addition, he may well have perceived the threat of traffic in persons and ideas as useful leverage in maintaining a decisive say in the GDR’s policy-making process for his ministry or himself.84 In the summer of 1971, Mielke officially vetoed further socioeconomic integration with Poland, and in late 1971, he successfully pressed for a strengthening of the border with the FRG.85 Meanwhile, the deep-seated unease of the SED leadership over the actions and motivations of their socialist “friends” was heightened by an unending flow of consecutively numbered Politburo reports by the HVA—with some of the best and most reliable information once again coming out of Bonn’s ministries. Already in late 1969, the initial reactions to West German approaches of the other Warsaw Pact member states provided important indications of what was to follow. While Polish reactions might have provided sufficient reason for concern, news about the rather restrained positions taken by Prague, Budapest, and Sofia were certainly welcome.86 The Quadripartite negotiations over West Berlin in 1971 were another occasion on which the HVA’s access to West German expertise was highly valued—and apparently not always shared within the Foreign Ministry.87 Reports like the one of January 1971 on FRG Foreign Minister Scheel’s assessment that the GDR and the United States had so far blocked an agreement over Berlin, while Bonn and Moscow favored one, certainly helped East Berlin to judge the situation beyond the diplomatic information received from Moscow.88 In early 1973, the focus of HVA information papers appears to have continuously shifted toward providing data on increasing trade between the GDR’s “friends” in eastern Europe and the FRG (while the GDR’s stagnated at this time).89 Information about Brandt and Schmidt’s coordination within the West naturally remained another priority during the détente years.90 Specialists at the Central Analysis and Information Group (ZAIG)—perhaps best described as Mielke’s think tank—were responsible for processing the enormous amount of information on détente-relevant issues. One of its analyses in August 1972—with the German-German negotiations in full swing—advised counteracting the West with demands for more GDR journalists in the West, who—if this was granted—might provide better channels



The Stasi Confronts Western Strategies for Transformation, 1966–1975

189

of influence.91 Mielke did not pursue this idea for the very same reason, reciprocity, which he apparently feared. Only five months later, the analysis group itself drew attention to the self-congratulatory mood in the SPD’s leadership in late 1972, expecting a “process of democratization” within Warsaw Pact member states and depicting the SPD’s Ostpolitik as “exemplary.”92 December 1972 did indeed see a multiplication of defensive tasks for the Stasi. The signature of the Basic Treaty between the two German states led, not only to fears of a second Erfurt (in view of the possibility of a visit to East Berlin by Brandt on this occasion),93 but also to a comprehensive “plan of action” necessitated by the GDR’s new embassies, better working conditions for more foreign journalists, and problems arising from an increase in family reunions and family visits in “border areas.”94 But there were also some endogenous reasons hindering an effective counterstrategy against the transformative intentions hidden behind Brandt’s and Bahr’s rhetoric of détente and rapprochement. One reason was the difficulties East German experts encountered in squaring the dichotomy of security, stability, intellectual exchange, and opening-up in Brandt’s concept with the FRG’s image of revanchism and militarism, until then firmly established within the GDR. This resulted in an indifferent equation of the détente concepts favored by Kiesinger, Strauß, Brandt, Bahr, Wehner, and Zbigniew Brzezinski (whom East Berlin wrongly suspected to be the mastermind behind LBJ’s “bridge-building” strategy, which was known to have been a strong influence on the conceptualization of Ostpolitik). All of these approaches were perceived as being interconnected—and therefore essentially the same. The resulting analytical framework made for a gross underestimation of the dynamic aspects of Brandt’s and Bahr’s Ostpolitik when compared to the status quo approach of the Nixon administration.95 The Conference on Security and Co-operation in Europe and the Multilateralization of Détente In October 1975, Klaus Blech—head of the planning staff in Bonn’s Foreign Ministry and for obvious reasons its chief negotiator at the Conference on Security and Co-operation in Europe96 —allowed his British colleagues a glimpse of Bonn’s long-term expectations when he sketched an optimistic outlook on the CSCE-process “if one sees [it] over a longer period—50 years—and assumes a waning of ideological confrontation over this period.”97

190

The Stasi Confronts Western Strategies for Transformation, 1966–1975

Seen from East Berlin, this was, of course, exactly what the GDR wanted to avoid. What was more, the SED leadership was informed as early as August 1969 of Brandt and Bahr’s aims for the CSCE project and the long-term impact of Ostpolitik.98 Stasi espionage provided language from Brandt’s explanations in internal SPD meetings and from Scheel’s contribution at the meeting of the British, American, French, and West German foreign ministers on the eve of the NATO summit on 3 December 1969. While Brandt clearly indicated a connection between German reunification (while carefully avoiding the term) and a “European peace settlement,” Scheel described the technicalities of this approach by linking an eventual conference on European security to constructive agreements to be encoded first in both an intra-German treaty and a Quadripartite Agreement on Berlin.99 Equally, East Berlin obtained highly relevant information through Western sources on the positions taken by other Warsaw Pact states in their respective bilateral dealings with the “class enemy.” Of particular concern must have been that Poland’s leadership was by then so keen on economic relations with the FRG that it declared the recognition of the GDR by Bonn according to international law to be “unnecessary.” As the early East-West exchanges about a European security advanced, Czech diplomats took a surprisingly “differentiating”100 position on the issue of “free movement.” Equally, Moscow’s priorities in détente—economic cooperation with the West and the CSCE to confirm its territorial gains—were clearly spelled out in Soviet talks with West German emissaries and became immediately available to the SED leadership.101 The Stasi’s international espionage department also obtained and passed on key information about Western—particularly West German and social democratic—strategies for using a future European security conference for the transformation of Communist rule in the Eastern bloc. A meeting of the Socialist International in Helsinki in May 1971, for example, discussed the promising perspectives to be opened up by a European security and cooperation conference for the rise of Eastern bloc opposition movements and how West European social democrats could support this development.102 Another scenario—much favored by Brandt and Bahr103 —on which the SED leadership was kept well informed was the prospect of major “mutually balanced” troop reductions in both blocs, resulting in a significant loss of Soviet control over Central and Eastern Europe and an enormous psychological boost for reform movements in these countries.104



The Stasi Confronts Western Strategies for Transformation, 1966–1975

191

In mid-November 1972—shortly before the signature of the Basic Treaty between Bonn and East Berlin, with numerous bilateral treaties between East and West already ratified and the Twenty-Fourth Party Congress of the CPSU and the 1972 Crimean meetings concluded105 —Mielke explained the new situation, the GDR’s goals, and the new strategies to be pursued by the Stasi to his officers.106 According to Mielke, the GDR’s prime objective in international relations now was to “secure peace in Europe,” and this could best be served through recognition of the territorial status quo within an all-European security conference. Furthermore, the GDR’s full-fledged participation in CSCE and MBFR negotiations would visibly prove East Germany’s “equality” to a global public. In order to achieve this, East Berlin’s foreign policy had to avoid “direct confrontations” with Bonn or other Western powers. At the same time, of course, the multiplication of East-West contacts resulting from the treaties—and particularly from the forthcoming Basic Treaty—had now to be both limited and controlled by a major and well-coordinated Stasi response. This resulted in the “catalogue of measures” mentioned above, agreed upon on 18 December 1972, in which the “activities of journalists” were construed as “particularly dangerous.” The irony was, of course, that while the Stasi had sufficient information on West German intentions about “journalists bringing irritations into the GDR”107 and numerous countermeasures in hand—the Helsinki Final Act would eventually codify in 1975 what the Stasi so desperately tried to prevent in 1972. A year before the CSCE’s conclusion at the Helsinki summit, the Stasi already had a clear picture of the eventual outcome and its momentous consequences for the GDR’s internal situation.108 Numerous departments were involved in an exercise listing various forms of ideological subversion expected to flood the country from the West and formulating an even more voluminous “catalogue” of ways and means to counteract this.109 The Stasi’s experts knew that by July 1974, Hans-Dietrich Genscher had informed the Americans that he was only going to sign a final document if it contained a clause on the “peaceful change of frontiers,” thereby maintaining a “reunification option.”110 It was clearly realized that an increase in the volume of East-West traffic and in economic and technical cooperation was to form an integral part of the FRG’s “policy of contacts” aiming at the “diversion” (subversion) of socialism in the GDR.111 This was truly the multilateralization of imperialist “penetration, softening-up, and wedge-driving.” What was worse, East-West discussions in the CSCE had increasingly united the Western camp,

192

The Stasi Confronts Western Strategies for Transformation, 1966–1975

and even military détente—initially thought to beef up the East’s superiority—would probably contribute to a greater degree of transparency and further contacts through “confidence-building measures” (CBMs). From NATO documents, the Stasi’s analytical unit identified three “problem fields”: (a) human contacts, (b) increase of information, (c) and expansion of cultural relations. While the West would argue that the latter constituted “apolitical” contacts, they would be political, would aim “to support or reactivate certain negative, anti-socialist elements, particularly so-called opposition forces.” In the eyes of the Stasi experts, Washington was seeking solutions to its global problems in the CSCE. The strategy behind the Western approach was plainly attributed to former Chancellor Brandt—once again reinforcing older perceptions of the dangers of “social-democratism”: “According to SPD Chairman Brandt, this long-term concept is meant to contribute to the construction of a peace order overcoming the status quo at the end of this century.” It seems that the Stasi—directly or indirectly—also contributed to the three illusions held by the political leadership of the GDR. The HVA provided a continuous flow of reports on the differences and even quarrels between Bonn and Washington over the meaning of military and economic “hardware” versus the “software” of ideas, communication, and contacts. Daydreams banking on the long-term effects of the “power of the factual” were served by expert legal reports on the durability of the “state border between the GDR and the FRG” after the final recognition of the existing postwar order in Helsinki.112 Furthermore, the Stasi demonstrated its willingness and apparent ability to deal with the resulting domestic situation in first compiling a truly comprehensive strategy and then—in the days immediately after Helsinki in August 1975—in organizing an effective “infrastructure” (including an increase in Stasi personnel until 1979) for dealing with the new realities.113 Internally, as Honecker put it with a smile to Brezhnev on their last consultative meeting before Helsinki, “there had always been the Staatssicherheit, and it is still in existence.”114 Measuring the Impact of Détente on the GDR and of the Stasi under Détente The shift from Ulbricht to Honecker was a crucial watershed in the GDR’s domestic, foreign, and German policies—and as Stasi files clearly show, it played into a very specific East German “CSCE process.” It encapsulated a



The Stasi Confronts Western Strategies for Transformation, 1966–1975

193

change of central paradigms: from an offensive to a defensive strategy, from aggressive rhetoric to détente-compatible official language, and eventually also from “delimitation” to more and more communication, contacts, and cooperation. And this was essentially what transformation strategies in the West had been aiming at. But did East Berlin’s adaptation to the détente era and especially the CSCE process contribute to the demise of the GDR? Perhaps this can only be answered by counterfactually imagining the GDR’s best-case scenarios for the outcome once multilateralization had virtually guaranteed the irreversibility of détente in Europe. The first of these scenarios might have been an effective blocking by the GDR of the reunification terminology in both the Treaty of Moscow and the Helsinki Final Act. If East Berlin—as advised by some of its diplomats115 — had refused to sign a CSCE document including the disputed phrase on the “peaceful change of frontiers,” responsibility for the success of the conference would have rested entirely with Bonn. The signatures of both German states were essential for a conference, which many, if not most, participants intended to serve as a kind of “Ersatz peace”116 over Germany, thereby codifying the territorial results of World War II. In the light of this, and contrary to its official position, Bonn might well have signed the Final Act notwithstanding in order to save the multilateralization of its Ostpolitik approach and the long-term effects expected from this.117 The second scenario would make for a more viable GDR economy. This might have been achieved if Moscow had granted the kind and quantities of raw materials Ulbricht consistently asked for; if Ulbricht’s economic reforms had not been curbed (and had been successful); and if the GDR’s industrial espionage had been utilized in a more effective and sustainable way.118 This might have resulted in a more timely linkage between the GDR’s industrial capacities and the global “technological revolution,”119 perhaps even bringing the GDR into the world economy. The third scenario comprises societal reforms, possibly in cooperation and coordination with other Warsaw Pact states, granting more individual liberties step by step. It seems that quite a few younger GDR officials—including members of the Stasi—were thinking of such a process in the early and mid1970s.120 In this scenario, the CSCE appeared more as an opportunity than a threat. A policy along these lines—necessarily also banking on an improving economic situation—would have envisaged less rather than more domestic repression and observation by the Stasi.

194

The Stasi Confronts Western Strategies for Transformation, 1966–1975

But even given these three best-case scenarios, the GDR would still have been exposed to the “Gorbachev miracle”121—which in East Berlin’s perspective was less a miracle than a nightmare as it meant increasing withdrawal of the international support necessary for the GDR’s existence. Still, the three scenarios might have provided the GDR’s leadership (either the SED or its successors after the first free elections) with a more viable negotiation platform when it came to reunification in 1990, thereby avoiding what was essentially a takeover by the FRG.122 A more palatable domestic situation within the GDR by this time—with a much less visible “Stasi”—might also have resulted in wider support for the attempts at reforming socialism by the GDR’s leftist opposition groups—which proved entirely unacceptable to the majority of East Germans in 1989–1990. This is not to say that this would have represented a preferable or— depending on one’s personal attitude—even a worse scenario. Instead, the comparison between the actual course of history and a counterfactual imagination of factors possibly working for the GDR’s leadership shows that, either way, Ostpolitik and détente, once rendered irreversible by multilateralization in Europe, would have seriously impinged on the GDR’s livelihood. Furthermore, it shows that the strategies thought up for and the roles played by the Stasi in the attempt to stem the effects of détente could not hinder, and perhaps could not even delay, the eventual outcome—and in a number of ways (by drawing on much-needed financial and intellectual resources; by nourishing fear of the authorities and distrust in public discourse) might even have contributed to it. And yet, the Stasi—as visible as it was to the GDR’s citizens when it came to matters of free movement, information, and speech—only fulfilled its envisaged function and remained the “shield and sword of the party”123 throughout. Notes Archival abbreviations: AMAE— Archives du Ministère des affaires étrangères (Paris); AN—Archives nationales (Paris), Archives de la présidence de la République—Président de Gaulle (5 AG 1) and Président Pompidou (5 AG 2); BStU—Beauftragte für Stasi-Unterlagen, archives of the GDR’s Ministry for State Security (Berlin); NARA—National Archives and Record Administration (Washington, DC); PA AA/ MfAA—Records of the GDR’s Foreign Ministry (Ministerium für Auswärtige Angelegenheiten, MfAA) in the Political Archives of the Auswärtiges Amt (Berlin); PRO— Public Record Office (London), now renamed the National Archives, Prime Minister’s



The Stasi Confronts Western Strategies for Transformation, 1966–1975

195

Files (PREM) and Foreign and Commonwealth Office files (FCO); SAPMO—Stiftung Archiv der Parteien und Massenorganisationen der DDR (Berlin). 1. This chapter was originally published as Oliver Bange, “Zwischen Bedrohungsperzeption und sozialistischem Selbstverständnis—Die DDR-Staatssicherheit und westliche Transformationsstrategien, 1966–1975,” in Torsten Diedrich and Walter Süß, eds., Militär und Staatssicherheit im Sicherheitskonzept der Warschauer-Pakt-Staaten (Berlin, 2010), 253–296. 2. See, e.g., Hubertus Knabe, “Die DDR-Opposition und ihre westdeutschen Unterstützer,” in Hans-Joachim Veen, Ulrich Mählert, and Peter März, eds., Wechselwirkungen Ost-West—Dissidenz, Opposition und Zivilgesellschaft, 1975–1989 (Cologne, 2007), 111–127, esp. 115ff. 3. Author’s conversation with Norbert Podewin, Odense, 17 November 2007. Podewin worked his firsthand experience of the SED’s decision-making elite into his book Walter Ulbricht—eine neue Biographie (Berlin, 1995). Conversations and letter exchange between the author and Siegfried Bock, 2005–2007. Bock headed both the analytical unit of the GDR’s foreign ministry and the GDR’s delegation to the CSCE negotiations until the Helsinki summit in August 1975. For his account of the constraints and missed opportunities in GDR foreign and domestic policies in the détente era, see Siegfried Bock, “The CSCE—an Era of Dissent and Consensus,” in Oliver Bange and Stephan Kieninger, eds., Negotiating One’s Own Demise? The GDR’s Foreign Ministry and the CSCE Negotiations—Plans, Preparations, Tactics and Presumptions (Mannheim, 2007), e-dossier at http://www.CSCE-1975.net; Siegfried Bock, “Die DDR im KSZE-Prozeß,” in Bock, Ingrid Muth, and Hermann Schwiesau, eds., DDR-Außenpolitik im Rückspiegel—Diplomaten im Gespräch (Münster, 2004), 102–117 (including a discussion among former GDR diplomats). 4. Various conversations and letter exchanges between Charles de Gaulle and Konrad Adenauer, 1959–1967, AN: 5 AG 1 (de Gaulle), vols. 160–168. 5. For the public outcry—particularly among CDU/CSU members—after Adenauer’s speech at his party’s election congress, see Erich Mende, Von Wende zu Wende, 1962–1982 (Munich, 1986), 216–217. 6. See Bernd Faulenbach, “Europa im Zeichen der Entspannungspolitik,” in Veen et al., eds., Wechselwirkungen, 17–29, here 22. For contextualization of the contemporary term “civilizing relations,” see the introduction by Oliver Bange and Gottfried Niedhart in Bange and Niedhart, eds., Helsinki 1975 and the Transformation of Europe (New York, 2008), 13ff. 7. For a fundamental revision of the established picture of LBJ, particularly useful on his pivotal role in the new American détente strategy, see Thomas A. Schwartz, Lyndon B. Johnson and Europe (Cambridge, MA, 2003); for Johnson’s “bridge-building” speech of 1964, see, e.g., 19ff. 8. Author’s interviews with Francis Bator (Harvard, 27 March 2004) and Henry Owen (Washington, DC, 23 April 2005). For a discussion of these concepts, see Oliver Bange, “Die USA und die oppositionellen Bewegungen in Osteuropa von 1961 bis 1990,” in Veen et al., eds., Wechselwirkungen, 79–95, esp. 81ff.

196

The Stasi Confronts Western Strategies for Transformation, 1966–1975

9. Memorandum “Considerations on the Foreign Policy of a Future Federal Government,” 18 September 1969, with a covering note from Bahr to Brandt of 21 September 1969, FRG, Auswärtiges Amt, Akten zur Auswärtigen Politik der Bundesrepublik Deutschland, 1969, doc. 296. For a detailed discussion of the central importance of “irreversibility” in Brandt’s and Bahr’s long-term strategy embedded in Ostpolitik, see Oliver Bange, “An Intricate Web: Ostpolitik, the European Security System and German Unification,” in Bange and Niedhart, eds., Helsinki 1975, 43–63. 10. Egon Bahr, Zu meiner Zeit (Munich, 1996), 547ff. See also Oliver Bange, “‘Europeanising’ and ‘Social-Democratising’ Europe—Trans-European Aspects of Willy Brandt’s Ostpolitik and the Path to a Social-Democratic Europe,” paper presented at the conference “Western Europe from the Golden Age to the Age of Uncertainty, 1965–1975,” Trento, 2006 (www.ostpolitik.net). 11. For Prime Minister Harold Wilson’s small circle of advisers on CSCE issues, see PRO: PREM 16/391 and –/392. For an historiographical analysis of the effects of this constellation, see Luca Ratti, “Britain, the German Question and the Transformation of Europe—From Ostpolitik to the Helsinki Conference (1963–1975),” in Bange and Niedhart, eds., Helsinki 1975, 127–149. 12. Jean-Bernard Raimond’s memoranda for Pompidou, 3 July, 6 November, and 8 December 1969, 24 January 1970, AN: 5 AG 2/1010 and –/1009 (Pompidou Papers). See also AMAE: Série Europe, Sous-Série Republique fédérale d’Allemagne, vol. 1726. For Pompidou’s personal commitment to human rights and to Basket III of the Helsinki Final Act, see Marie-Pierre Rey, “France and the German Question in the Context of Ostpolitik and CSCE, 1969–1974,” in Bange and Niedhart, eds., Helsinki 1975, 85–104. 13. Lawrence Eagleburger characterized this mood in a memorandum to Henry Kissinger on 26 January 1970 as one of “presiding over the partial dissolution of the American empire.” NARA: RG 59, Sonnenfeldt files, box 2. 14. Author’s interviews with James Goodby, George Vest, Henry Owen, Jonathan Dean, Robert Gerald Livingstone, and Helmut Sonnenfeldt (2005–2007). See also Stephan Kieninger, “Transformation or Status Quo: The Conflict of Stratagems in Washington over the Meaning and Purpose of CSCE and MBFR, 1969–1973,” in Bange and Niedhart, eds., Helsinki 1975, 105–127; and Kieninger, “The Survival of the Transformation Strategy During the Nixon Years,” in Poul Villaume and Odd Arne Westad, eds., Perforating the Iron Curtain: European Détente, Transatlantic Relations, and the Cold War, 1965–1985 (Copenhagen: Museum Tusculanum Press, University of Copenhagen, 2010), 101–122. 15. Memcon Nixon-Scheel, Washington, 13 June 1969, and a preparatory briefing paper for Nixon from Kissinger, 12 June 1969, NARA: Nixon Materials, NSC, CF 681 (Germany). 16. See, e.g., Thomas Fischer, “Bridging the Gap Between East and West—The N+N as Catalysts in the CSCE-Process, 1972–1983,” in Poul Villaume and Odd Arne Westad, eds., Perforating the Iron Curtain: European Détente, Transatlantic Relations, and the Cold War, 1965–1985 (Copenhagen: Museum Tusculanum Press, University of Copenhagen, 2010), 143–178; Helmut Liedermann, “Österreichs Rolle beim



The Stasi Confronts Western Strategies for Transformation, 1966–1975

197

Zustandekommen der KSZE,” in Stephan Rathkolb, Otto M. Maschke, and Stephan August Lüthgenau, eds., Mit anderen Augen gesehen—internationale Perzeptionen Österreichs, 1955–2000 (Vienna, 2002), 487–506. 17. Ulbricht to Brezhnev, Moscow, 18 September 1965, SAPMO: DY30/J IV 2/201/725. 18. On this—although exclusively based on official GDR publications—see Hans-Joachim Spanger, Die SED und der Sozialdemokratismus (Cologne, 1982). A more differentiated perspective is provided by Michael Lemke, “Eine neue Konzeption?—Die SED im Umgang mit der SPD 1956 bis 1960,” in Jürgen Kocka, ed., Historische DDR-Forschung—Aufsätze und Studien (Berlin, 1993), 361–377. Lemke shows that even this early, “social democratism” was anything but a consistent ideological concept. Rather, it constituted a container for various strategies in the fight against the West German SPD. The switch from a frontal assault on “social democratism” to partial and informal cooperation with the lower levels of West German social democracy and trade unions—a tactic preferred by Ulbricht—had thus already occurred in the late 1950s. 19. Markus Wolf, Man Without a Face—The Autobiography of Communism’s Greatest Spymaster (New York, 1999), 238. 20. For Ulbricht’s rationale, see his conversation with Brezhnev on 18 September 1965, SAPMO: DY30/J IV 2/201/725. For more detailed description and analysis of this, see Jochen Staadt, Die geheime Westpolitik der SED 1960–1970—Von der gesamtdeutschen Orientierung zur sozialistischen Nation (Berlin, 1993), esp. 107ff.; and Monika Kaiser, Machtwechsel von Ulbricht zu Honecker—Funktionsmechanismen der SEDDiktatur in Konfliktsituationen 1962 bis 1972 (Berlin, 1997), 233ff. 21. A telling example was Ulbricht’s attempt in September 1966 to get the SPD leadership to agree to a series of public debates (an ideal platform for his strategy of approaching the SPD’s electorate). When he simultaneously pressed Brezhnev for economic and political support, the latter retorted by demanding a more confrontational approach from East Berlin in its dealings with the FRG and its parties. This, as Ulbricht immediately understood, was to give the impression “that we intend to deepen Germany’s division,” thereby counteracting his entire rationale. Conversation between Ulbricht and Brezhnev in Moscow on 10 September 1966, SAPMO: (Ulbricht’s office) DY 30/3518 and –/3294. 22. For the importance of the January–February 1967 Warsaw Pact crisis for the future course of the détente era, see Oliver Bange, “Ostpolitik as a Source of IntraBloc Tensions,” in Victor Papacosma and Ann Heiss, eds., NATO and the Warsaw Pact—Intra-Bloc Conflicts (Kent, OH, 2007), 106–121. 23. Ulbricht–Semjonov conversation, 17 January 1967, SAPMO: DY 30/3520. 24. Personal and strictly confidential “Disposition für die Skizze einer langfristigen Politik gegenüber Westdeutschland,” 8 February 1968, by the AG Außen- und Deutschlandpolitik, SAPMO: (Ulbricht’s office files) DY 30/3311. 25. Author’s conversation with Norbert Podewin, 17 November 2007. In 1968, Podewin was working for Albert Norden and frequently discussed the situation in Prague with Ulbricht. See also Oliver Bange, “Die CSSR-Krise 1968—Die ‘Special Role’

198

The Stasi Confronts Western Strategies for Transformation, 1966–1975

der Neuen Ostpolitik vor dem Hintergrund westlicher Transformationsstrategien,” in Bernd Greiner and Dierk Walter, eds., Crises in the Cold War, vol. 3 (Hamburg, 2007). 26. Memorandum by Norden for Ulbricht, 12 July 1968; telegram from Florin to Winzer, 18 July 1968; draft speech by Ulbricht for the meeting in Warsaw on 14–15 July 1968, SAPMO: DY 30/3556 and –/3618. 27. Whereas Mark Kramer, in Jaromír Navrátil et al., eds., The Prague Spring 1968—A National Security Archives Documents Reader (New York, 1998), and Manfred Wilke, in Wilke, Vaclav Kural, and Lutz Priess, Die SED und der “Prager Frühling”— Politik gegen einen “Sozialismus mit menschlichem Antlitz” (Berlin, 1996), describe Ulbricht as a military hardliner, Kaiser, in Machtwechsel (cited n. 22 above), and the former East German ambassador to Prague, Karl Seidel, in Siegfried Bock et al., eds., DDR-Außenpolitik im Rückspiegel—Diplomaten im Gespräch (Münster, 2004), 60, contend that he sought a peaceful solution and was only forced to back the hardliners by Honecker and Norden. I argue for a third, different interpretation of the reaction of Ulbricht and the GDR leadership to the CSSR crisis, which occurred in several stages and was closely related to the GDR’s intra-German policies. 28. Winzer note for Ulbricht, 9 October 1968; note from Belezki, 22 October 1968; draft peace treaty, 21 October 1968, SAPMO: DY 30/3522. 29. Stenograph protocol of the Politburo meeting in Dölln on 30 October 1969 by Ulbricht’s personal assistant Wolfgang Berger, SAPMO: DY 30/3294. This appears much more reliable than Honecker’s handwritten notes, cited by Kaiser, Machtwechsel, 326–331. 30. Winzer to Ulbricht on a meeting with Abrassimov, 11 March 1970, SAPMO: DY 30/3526. 31. Ulbricht–Gromyko conversation, 29 October 1970, SAPMO: DY 30/3654. 32. Ulbricht to Brezhnev 2 June and 17 August 1970; Ulbricht–Brezhnev conversations, 19 and 21 August 1970; Brezhnev letter, 16 October 1970, SAPMO: DY 30/3530. 33. Foreign policy paper, “Die BRD in der Klassenauseinandersetzung unserer Zeit—Analyse und Prognose” (121 pp.), November 1970–January 1971, SAPMO: DY 30/3317. The study began with a “scientific” definition of “social democratism” as a pretense at transforming capitalism and an attempt at liquidating socialism step by step, the goal being “to remove the GDR from the socialist camp and to use the pretended claim of a continuing ‘unity of the nation’ finally to attach it to the FRG.” 34. Report for Winzer, “Verlauf und Ergebnisse der ersten Phase der europäischen Sicherheitskonferenz”, 23 July 1973, PA AA: MfAA C 851/75. Selected passages of this— edited and translated into English—can be found in Bange and Kieninger, eds., Negotiating One’s Own Demise? (cited n. 3 above), doc. 11. 35. See Oliver Bange, “‘Keeping Détente Alive’: Inner-German Relations Under Helmut Schmidt and Erich Honecker, 1974–1982,” in Leopoldo Nuti, ed., The Crisis of Détente in Europe: From Helsinki to Gorbachev, 1975–1985 (London, 2009), 230-243. 36. For the first two factors, see also the contribution by Süß in Militär und Staatssicherheit im Sicherheitskonzept der Warschauer-Pakt-Staaten. During his conversation with Honecker in Moscow on 18 June 1974, Brezhnev presented his own



The Stasi Confronts Western Strategies for Transformation, 1966–1975

199

interpretation of the power of the factual: step by step, the old generation would be replaced by Young Pioneers with a socialist education, leading to a “natural form of delimitation between the GDR and the FRG.” SAPMO: DY 30/ J IV 2/201/1303. 37. Author’s conversations with Rainer Rupp, Gabriele Gast, and Bernd Fischer (who made this statement about the Middle East, his area of responsibility in the HVA), Odense, 16–18 October 2007. See also Bernd Fischer, Als Diplomat mit zwei Berufen—Die DDR Aufklärung in der Dritten Welt (Berlin, 2009). 38. Rudolf Maerker was also head of the regional SPD organization in Bonn. See “Stasi-Spionage: Spitzel ‘Max’ enttarnt,” Tagesspiegel, 9 August 2000. Maerker had worked for the Stasi since October 1968. 39. Wehner’s close collaborator, Karl Wienand (born 1926, SPD member of parliament, 1963–1974), was convicted in 1996 for spying against the FRG. However, with regard to occasional disinformation planted by Wienand on behalf of Wehner during his back-channel conversations with East Berlin, the author tends to believe that Wienand, at least until 1975, did not provide information without party or government backing in Bonn. See Gerd Lotze, Karl Wienand—Der Drahtzieher(Cologne, 1995). William Borm (1895–1987) was an FDP member of parliament from 1965 until 1972, and from 1960 until 1982 was a member of the FDP’s national council. He had worked for the Stasi since the late 1950s. See Klaus Marxen and Gerhard Werle, eds., Strafjustiz und DDR-Unrecht—Dokumente, vol. 4 (legal cases against members of the GDR’s spy service, the HVA of the Stasi) (Berlin, 2004), 66ff. Particularly with regard to Borm’s nine years in East German prisons, contemporaries within the FDP still regard his case as “tragic.” Author’s interview and conversations with Barthold C. Witte, 2003– 2005; see also Witte, Für die Freiheit eine Gasse- aus dem Leben eines liberalen Bürgers (Stuttgart, 2003). For an analysis of East German recruitment methods in the FRG and West Germans’ motives to work for the GDR’s spy service, see Georg Herbstritt, Bundesbürger im Dienst der DDR-Spionage—Eine analytische Studie (Göttingen, 2007). 40. Author’s conversation with Manfred Süß and Horst Vogel in Odense, 18 November 2007 (following a presentation by Süß in which he made the above-mentioned claim). Süß and Vogel were head and deputy head of the HVA’s department of espionage in science and technology. 41. For examples, see Hubertus Knabe, Der diskrete Charme der DDR—Stasi und Westmedien (Munich, 2001). 42. Bahr, Zu meiner Zeit, 547ff. 43. This was, however, an illusion. The institute had been founded by Konrad Adenauer, and its head, Boris Meissner, enjoyed direct access to the Kanzleramt (to which the institute was also formally attached—instead of to the Auswärtiges Amt or the Ministry for All-German Affairs, to which it should nominally have been attached). This triggered a residual mistrust within both the SPD and FDP leaderships, who favored (and even founded) new research institutions after 1969. Author’s conversations with Walter Süß and Gerhard Wettig. For a general outline of internal West German disputes and their repercussions on attempts at pooling Western knowledge about the Eastern bloc, see Giles Scott-Smith, “Confronting Peaceful

200

The Stasi Confronts Western Strategies for Transformation, 1966–1975

Co-existence—Psychological Warfare and the Role of Interdoc, 1963–1972,” Cold War History 7, no. 1 (February 2007): 19–43. 44. Author’s conversation with Karl Rehbaum, head of the HVA’s department of strategic military espionage, Odense 17 November 2007. 45. On the public aspects of Brandt’s early Ostpolitik, see Gottfried Niedhart, “The East-West Problem as Seen from Berlin—Willy Brandt’s Early Ostpolitik,” in Wilfried Loth, ed., Europe, Cold War and Coexistence, 1953–1965 (London, 2004), 285–296; Arne Hofmann, The Emergence of Détente in Europe—Brandt, Kennedy and the Formation of Ostpolitik (London, 2007); Oliver Bange, “‘Scenes from a Marriage’—East-West Détente and Its Impact on the Atlantic Community, 1961–77,” in Giles Scott-Smith and Valerie Aubourg, eds., Atlantic, Euratlantic, or Europe-America? (Paris, 2011), 262–283. 46. The speeches by Brandt and Bahr at Tutzing on 15 July 1963 are edited in Dokumente zur Deutschlandpolitik, ser. 4, vol. 9 (Frankfurt a/M, 1978), 565–575. Originally, it was intended that Brandt should speak first, but he was delayed, and Bahr was asked to do so, “stole the attention,” and received most of the criticism (Willy Brandt, Erinnerungen, [Frankfurt/M, 1989], 79). 47. Wolf, Man Without a Face, 117, prides himself in the fact that his informant “Freddy” was “a close associate of the young mayor.” However, while “Freddy”—Josef Braun, the deputy head of West Berlin’s SPD in the 1950s until the late 1960s—was certainly well placed within the party, he did not belong to Brandt’s small team of Ostpolitik advisers. 48. Von Berg denied ever having been a Stasi agent when confronted with his MfS file in August 1993. He claimed that there had only been a few informal meetings (Abschöpfung), but his formal designation as an “informal employee” and the fact that a case officer organized his trips to the West, advised and debriefed him, and telling him what the Stasi wanted him to convey in West Berlin suggest otherwise. BStU: GH 25/87. After being charged with “social-democratism” in East Berlin and personal intervention by Egon Bahr, von Berg was able to leave the GDR—and in turn accused the SPD leadership of having given up their fight against the GDR. See Hermann von Berg, “Positionen der Linken zur nationalen Problematik in Deutschland,” in Konrad Löw, ed., Verratene Treue—Die SPD und die Opfer des Kommunismus (Cologne, 1994), 207–234. 49. It is not entirely clear to what the mention of a “memorandum on German policy” referred. The timing perhaps hints at Ulbricht’s proposal on 25 April 1964 that there be an exchange of West and East German newspapers (eventually blocked by the governing parties in Bonn on legal grounds); see Mende, Von Wende zu Wende (cited n. 5 above), 138ff. For an official account of Ulbricht’s speech at Bitterfeld, see Lieselotte Thoms et al., Walter Ulbricht—Arbeiter, Revolutionär, Staatsmann (Berlin, 1968), 260ff. 50. This list of questions to be pursued by von Berg is attached to a report on the conversation between von Berg and Spangenberg on 21 May 1964. BStU: GH 25/87. 51. BStU: GH 65/88. Wehner let it be known via Robert Stengel’s messages to von Berg on 10 March 1966 and 25 October 1966 that he would have liked a direct contact to



The Stasi Confronts Western Strategies for Transformation, 1966–1975

201

Stoph or Honecker—depending on “who is the next man?” On Wehner’s belief that Stoph might be more accessible and flexible than Ulbricht, see also Christoph Meyer, Herbert Wehner—eine Biographie (Munich, 2006). 52. Spangenberg—but also Wienand, Wehner, and Bahr—repeatedly provided East Berlin with information to the effect that ideas about a realistic reunification of the FRG and the GDR had virtually been given up by the SPD and later by the coalition government of SPD and FDP in Bonn. For example, on 28 August 1966, Spangenberg shared his personal opinion (from the office of the governing mayor of West Berlin!) that “only a pathway to a confederation is—if anything at all is—still possible.” BStU: GH 25/87. 53. See, e.g., HVA Info no. 1290/69, 31 December 1969, “Attitude of Trade Unions Toward Ostpolitik and FDGB Contacts” and Brenner’s fears therein of contacts “on a broad platform”! BStU: HVA 156. Another significant item of evidence to this effect is provided in Info no. 1103/69, 10 December 1969, “Reactions of DGB on (Wild) Strikes of September 1969.” IGM and DGB had acted against strikes not authorized by them. In internal trade union circles, Brenner defended this action, arguing that the DKP (the West German Communist party) had tried to “inflame” the situation. BStU: HVA 154. Otto Brenner (1907–1972), like Willy Brandt in his youth a member of the leftist SAP, head of the metal workers’ union in 1956, head of the International Metal Workers’ Union in 1961, head of the European Trade Union Confederation in 1971. See Klaus Ullrich, “Otto Brenner—8. November 1907–15. April 1972,” in Claus Hinrich Casdorff, ed., Demokraten—Profile unserer Republik (Königstein, 1983), 79–87; Rainer Kalbitz, Die Ära Otto Brenner in der IG Metall (Frankfurt a/M, 2001). Heinz Oskar Vetter (1917–1990), head of the DGB, 1969–1982, deputy chairman of the trade union for mining and energy in 1960, head of the European Trade Union Confederation,1974–1979. For his vision of socialist reform and anticommunism, see Vetter, Notizen—Anmerkungen zur internationalen Politik (Cologne, 1983); Vetter, “Gewerkschaftspolitik—Beitrag zur Gesellschaftsreform” (presentation to the 9th DGB national conference in Berlin on 28 June 1972), Düsseldorf, 1972. 54. Info no. 1346/68, 23 December 1968, on “the inner situation of the IGM.” BStU: HVA 139. 55. Info no. 1051/68, October 1968. BStU: HVA 153. 56. Info no. 1370/68, 19 December 1968. BStU: HVA 139. 57. Info no. 369/68, 4 April 1968, on “relationship between SPD and trade unions.” BStU: HVA 128. 58. Info no. 1377/69, 20 December 1969, on SPD election campaign strategy. According to this, Wehner had thought up an anti-SED strategy as an electioneering platform for the SPD. BStU: HVA 139. 59. Info no. 1046/69 (n.d., but from October 1969) drew attention to the fact that Jochen Steffen, head of the SPD in Schleswig-Holstein, was “clearly against the Herbert Wehner-Helmut Schmidt course.” BStU: HVA 153. See also memorandum in Mielke’s office papers, 7 August 1968, on Steffen’s ideas for a possible SPD strategy

202

The Stasi Confronts Western Strategies for Transformation, 1966–1975

for the federal elections in 1969, combining leftist political positioning with a strict anticommunist front. BStU: SdM 1471. Further evidence came in the form of a confidential explanation by Franz Woscheck, member of the DGB council, during a meeting with the Kuratorium Unteilbares Deutschland (KUD) on 29–30 October 1969, on contacts between the DGB Youth and its East German equivalent, the FDJ. According to this statement, a decision by the DGB was still pending, and a meeting would only become possible after the DGB had eventually reached its verdict. Info no. 1165/69, 13 November 1969. BStU: HVA 154. 60. For Steffen’s role in giving the Left an audible voice in internal party discussions over Ostpolitik—thereby integrating it with the strategy pursued by Brandt— see Jens Schultz, “Sozialdemokratie und Kommunismus—Die Auseinandersetzung der SPD mit dem Kommunismus im Zeichen der Neuen Ostpolitik 1969–1974” (diss., Mannheim, 2008). 61. Info no. 1081/69, 10 December 1969. BStU: HVA 153. The 70-page study was undertaken by Wolfgang Göbel, a political scientist and SPD member then working for West Berlin’s press office under Peter Herz. Göbel undertook the study on East German reactions to Bonn’s Ostpolitik in June 1969 on behalf of the FRG Ministry for Pan-German Questions, with the support of the Otto-Suhr Institute of West Berlin’s Free University. 62. See info no. 1347/68, 16 December 1968, with details on Brenner’s role. BStU: HVA 139. 63. Reports from 16–18 April 1968. BStU: SdM 1475. 64. Norden, memorandum for Ulbricht, 12 July 1968; telegram from Florin to Winzer, 18 July 1968; draft speech by Ulbricht for the meeting in Warsaw, 14–15 July 1968. SAPMO: DY 30/3556 and –/3618. 65. Report about the discussions at the annual meeting of the KUD, info no. 1381/68, 23 December 1968. BStU: HVA 139. 66. This contained material highly relevant for Ulbricht’s strategy of directly approaching the SPD and DGB members by showing up the different attitudes and positions. Thus IG-Metall rejected spearheading any talks with the FDGB, a role eventually delegated to the deputy head of IG Chemie, Küpper. Info no. 1073/69, 2 September 1969. BStU: HVA 153. Karl Küpper was widely perceived as left-wing trade union leader. Since the foundation of the FRG, he had publicly denounced nuclear weapons, had participated in a famous SDS conference against the Vietnam War on 22 May 1966 in Frankfurt a/M, and in 1968 had joined student protests—like Otto Brenner—against the draft legislation for an eventual state of emergency, then prepared by the parties of the grand coalition. On the international activities of Küpper and his trade union in this era, see Jörg Rumpf, “Deutsche Industriegewerkschaft und Europäische Gemeinschaft—Die Europapolitik der Industriegewerkschaft Metall und der Industriegewerkschaft Chemie, Papier, Keramik im Zeitraum zwischen dem Inkrafttreten der Römischen Verträge und den ersten Direktwahlen zum Europäischen Parlament” (diss., Bochum, 2001). 67. Report on the SI meeting in Berlin, specifically called to discuss relations with Communism, at which Herbert Wehner and other West German social democrats



The Stasi Confronts Western Strategies for Transformation, 1966–1975

203

apparently played a crucial role in constructing a common front. Info no. 1340/68, 11 December 1968. BStU: HVA 139. 68. Unnumbered Stasi report, 22 October 1969, BStU: HVA 153, and info no. 1394/69, 8 January 1969, BStU: HVA 139, on a special meeting in the Chancellor’s Office in Bonn in April 1969, on the Federal Institute for Eastern Studies in Cologne, and on the founding of Interdoc at the Hague. 69. Info nos. 1074/69 and 1076/69, 13 October 1969. BStU: HVA 153. 70. Info no. 1296/69, 16 December 1969. BStU: HVA 156. 71. MfS, “Disposition für Beratung” (n.d., but October 1969) and Markus Wolf, head of the HVA, note to Minister Erich Mielke on consultations with the KfS (KGB), 16 October 1969. BStU: SdM 1471. 72. “J. W. Andropow zur Einschätzung der Lage,” Berlin 17 November 1969. BStU: SdM 1473. For the printed version distributed in the ministry, the protocol was changed and the following sentence inserted: “[We] entirely agree with the considerations of Com. Mielke.” English excerpts from this speech can be found in Oliver Bange, ed., “And there will always be the State Security”—The GDR’s Ministry for State Security and the CSCE negotiations, 1969–1975 (Mannheim, 2008), e-dossier (www. CSCE-1975.net), doc. 2. The English translation was made possible by funds from the Hermann-Weber-Foundation, to which the author expresses his gratitude. 73. Twelfth meeting of the SED’s Central Committee, 12–13 December 1969. SAPMO: DY 30/IV 2/1/404. The Stasi’s analysis of Brandt’s and Bahr’s self-perception is mirrored in Bahr’s memoir Zu meiner Zeit (cited n. 10 above), 552-574. For details of the Stasi’s plans for “Confrontation” and “Confrontation II,” see BStU: SdM 1471. See also Detlef Nakath, Erfurt und Kassel—Zu den Gesprächen zwischen dem BRD-Bundeskanzler Willy Brandt und dem DDR-Ministerratsvorsitzenden Willi Stoph im Frühjahr 1970, Hefte zur DDR-Geschichte, vol. 24 (Berlin, 1995); Nakath and Gerd-Rüdiger Stephan, Erfurt and Kassel—Die ersten deutschen-deutschen Gipfeltreffen im Frühjahr 1970 (Berlin, 2000); Mary E. Sarotte, “A Small Town in (East) Germany—The Erfurt Meeting of 1970 and the Dynamics of Cold War Détente,” Diplomatic History 25, no. 1 (2001): 85–104; Sarotte, Dealing with the Devil—East Germany, Détente, and Ostpolitik 1969–1973 (Chapel Hill, NC, 2001); for a GDR insider’s account of the proceedings, see Karl Seidel, Berlin-Bonner Balance—20 Jahre deutsch-deutsche Beziehungen (Berlin, 2002). 75. Memorandum, 29 May 1970. BStU: ZAIG 4739. 76. BStU: SdM 1837. 77. The East and West German records of the negotiations between Egon Bahr and Michael Kohl from 1970 until 1973 are edited in Dokumente zur Deutschlandpolitik, ser. 4, vol. 2 (Munich, 2004). Because of the vast number of lengthy meetings, only selected records could be included in the volume, the rest being supplied on a CD-Rom. This edition is based on documents found in Bundesarchiv Koblenz (B 137), Bundesarchiv Berlin-Lichterfelde (DC 20 and DY 30), BStU (SdM), and the Archiv der sozialen Demokratie (Depositum Bahr). Additional material can be found in the relevant file series of the GDR’s Ministry for External Affairs (MfAA) in the Politisches Archiv des Auswärtigen Amts, esp. MfAA: G-A 82–96, 125–137, 150–169, 199–200, 243–244, 256–272.

204

The Stasi Confronts Western Strategies for Transformation, 1966–1975

78. For Michael Kohl’s reports to Erich Mielke, see BStU: SdM 233. 79. This at least is borne out by highly detailed reports on negotiating trips to Bonn, indicating that the informant was in the car, but not its driver—who, as Kohl knew, was also a member of the Stasi. See, e.g., the report on 13 November 1972 by the driver, Stasi Captain Striegler, on Kohl’s trip to Bonn on 9 November 1972. BStU: SdM 1837. 80. BStU: SdM 305. At least on one occasion, 13 July 1971, the KGB felt itself to be uninformed about a communication transmitted via this channel and demanded access—and thereby a major review of all contacts with Wehner. BStU: SdM 1423. 81. Several examples of this can be found in BStU: SdM 233. 82. For the negotiations between Günther Gaus and Kurt Nier in 1973–1974, see BStU: SdM 1847–1848; for the traffic treaty, see BStU: SdM 2392, and for the official Bahr-Kohl meetings thereon, BStU: HA XIX—272. 83. Decision of the GDR’s Ministerial Council of 25 May 1971 and Ministry of State Security follow-up of 7 June 1971. BStU: SdM 2392. 84. Legal arguments would certainly have been welcome on this point, and Mielke continually demanded and received memoranda from the Stasi’s legal department on these negotiations. BStU: RS 120–121 (June–November 1972). 85. See Ministerial Council decision, 27 July 1971, on the socioeconomic integration of the GDR and Poland; Mielke to the chairman of the Ministerial Council (the GDR’s nominal head of government), 8 September 1971, on “improvement in security and order on the state borders”; Mielke’s intervention in the decision of the Ministerial Council on 3 December 1971 on a West Berlin transit agreement, after which the Ministerial Council decided on 15 December 1971 to delete more lenient arrangements for visitors from West Berlin. BStU: SdM 2392. 86. Info no. 1195/69, 10 December 1969, on Warsaw’s policies toward the FRG. BStU: HVA 155. Info nos. 1349/69, and 1348/69 from December 1969 on reactions in Prague and Budapest, BStU: HVA 157. 87. See Mielke’s office file on the Quadripartite negotiations over Berlin in BStU: SdM 1836. Memorandum from Scheel’s office in Bonn’s Auswärtiges Amt of December 1970, cited in info no. 1332/70, 7 January 1970. BStU: HVA 176. 88. Judging by the office files of Walter Ulbricht and Erich Honecker and of Foreign Minister Otto Winzer, Soviet information about the status of the negotiations for a Quadripartite Agreement on (West) Berlin remained formal and relatively general and avoided referring to the secret back-channel negotiations between Falin, Rush, and Bahr over the issues involved. See MfAA: G-A 444; and SAPMO: DY 30/3537 and –/2372. 89. Compare BStU: HVA 88, 104, 386, 390. 90. This particularly concerned information about the meetings of the Socialist International and the role played by West German social democrats in these meetings. See, e.g., info nos. 329/72, 24 April 1972, on the SI meeting in Igls on 5–6 February 1972, and 862/71, 8 September 1971, on the SI council meeting of May 1971 in Helsinki. BStU: HVA 382 and 393.



The Stasi Confronts Western Strategies for Transformation, 1966–1975

205

91. Memorandum, “Survey of Possible Agreement/Cooperation with the FRG,” 14 August 1972. BStU: ZAIG 4632. 92. Memorandum, “Problems Associated with the Signing of the Basic Treaty in the GDR’s Capital,” 27 November 1972. BStU: ZAIG 4636. 93. For details on these fears and possible actions to prevent a renewal of the Erfurt experience, see the memoranda by the ZAIG from October and December 1972 in BStU: ZAIG 4641 and 4638. 94. See Mielke’s briefing papers for the meeting of the Stasi steering group (“Kollegiumssitzung”) on 18 December 1972 and the ZAIG memorandum “Political-Operative Problems and Tasks Resulting from the ‘Treaty on the Principles of the Relations Between the GDR and the FRG’ and Relevant Additional Documents,” (5 May) 1972. BStU: ZAIG 4640 and 4641. 95. For the long-term impact of these dynamic effects versus the status-quo oriented approach of the Nixon White House, see Kieninger, “Transformation or Status Quo” (cited n. 14 above); Bange, “Die USA und die oppositionellen Bewegungen” (cited n. 8 above); Gottfried Niedhart, “Frankreich und die USA im Dialog über Détente und Ostpolitik, 1969–1970,” Francia 31, no. 3 (2004): 65–85. 96. The VW-project on “CSCE and the transformation of Europe” at the University of Mannheim is providing an essential Internet-based introduction to the “GDR and the CSCE negotiations, 1969–1975.” The undertaking consists of three parts, highlighting the roles of the SED leadership, the GDR’s Foreign Ministry, and the Stasi. Each part contains a historiographical introduction, if possible commentaries from contemporaries, and a selection of key documents in English (see www.CSCE-1975. net). 97. Summary of planning staff talks, 31 October 1975. PRO: FCO 49/589. While Reginald Hibbert for the British side wondered in a commentary of 10 November 1975 if they should not admit that Basket III was “a form of ideological confrontation on our side,” Klaus Blech advised against further confrontation over the implementation of results during the follow-up conference at Belgrade, obviously banking on the longterm effects of the CSCE process, while not willing to put this in jeopardy. 98. The HVA obtained an Auswärtiges Amt memorandum dated August 1969 that clearly set out why bilateral Ostpolitik could neither induce long-term change nor render any transformative developments irreversible, and how multilateralization through the CSCE might do just this. Info no. 1066/69, 10 December 1969. BStU: HVA 153. 99. Brandt at a meeting of the SPD leadership of West Berlin on 13 December 1969, cited in info. no. 1306/69, 19 December 1969. Scheel’s statement at the “3+1” meeting between the three Western powers and the West German government on 3 December 1969 was cited in a memorandum of the Auswärtiges Amt on the reactions of the Western powers—and this memorandum had then been obtained by the GDR’s espionage service. See info no. 1320/69, 20 December 1969. BStU: HVA 157. 100. Apparently, Czech diplomats showed this “differentiating” attitude in talks with Dutch and Yugoslav colleagues. This information found its way into West

206

The Stasi Confronts Western Strategies for Transformation, 1966–1975

German documents obtained by the HVA and submitted to the SED’s Politburo in info no. 1216/71, 7 January 1972. BStU: HVA 398. 101. West German analysis of a visit to Moscow by Karl Wienand and Alois Mertes on 16–20 February 1971, along with the conversations of the emissaries of the SPD and FDP and the conclusions drawn in Bonn, was, e.g., quickly brought to the attention of the East German leadership by info no. 178/71, 4 March 1971. BStU: HVA 179. 102. BStU: HVA 393 of 8.9.1971. The Socialist International council agreed on a resolution in favor of a European Security Conference at its meeting in Helsinki on 25–27 May 1971. Info no. 497/71, 7 June 1971. BStU: HVA 182. 103. On the importance of an all-European security system in Brandt’s and Bahr’s long-term plans for reunification, see Bange, “Intricate Web” (cited n. 9 above). 104. See, e.g., Stasi info no. 757/71, 30 August 1971, about the forthcoming NATO special conference on MBFR; info no. 742/71 (n.d., but August 1971) from a source close to Wehner—apparently Karl Wienand—on the West German link between ESC and MBFR, and opposition in the Auswärtiges Amt to troop cuts of between 5 and 10 percent. BStU: HVA 392. Info no. 908/71, 27 September 1971, contains details about Bundeswehr planning for the MBFR. BStU: HVA 393. 105. For the East German stenographic records of the Crimean meeting on 31 July 1972, see Jordan Baev and Anna Locher, eds., Brezhnev’s Crimea Meetings in the 1970s, e-dossier (www.isn.ethz.ch/isn/Digital-Library/Publications/ Detail/?ots591=0c54e3b3-1e9c-be1e-2c24-a6a8c7060233&lng=en&id=108635). 106. Mielke made an almost 100-page long “presentation at the central service conference” on 16 November 1972. BStU: BdL doc. 005700. This position was also reflected in Erich Honecker’s explanations before the Sixth Central Committee plenum. Handwritten notes by Mielke. BStU: SdM 1423. 107. The quotation is from Peter Christian Ludz (1931–1979), a well-known adviser to the proponents of a new West German Ostpolitik, who wrote his Habilitation at the Free University of Berlin in 1967 on elite changes in the GDR. Both citations come from a memorandum titled “Indications in Connection with the Basic Treaty on the GDR’s Relations with the FRG,” 23 November 1972. BStU: ZAIG 4635. For the preparatory papers for the decisive meeting of the working group on the catalogue of measures, 18 December 1972, see BStU: ZAIG 4640 and 4771. 108. Memorandum from ZAIG (the analytical unit of the Stasi), 15 July 1974, on Western aims in connection with CSCE, filed in HA IX (the prosecution department). BStU: HA IX 1943; ZAIG 4645. 109. Stasi planning on CSCE came to a head in the summer of 1975. Many memoranda from the numerous departments involved in this exercise were sent to the legal department for comment and—once filed—formed a comprehensive overview of the various considerations and methods put forward. BStU: RS 289, 290, 347. 110. Exclusive “aktuelle Information” no. 2/74 by the HVA for Honecker, Stoph, and Sindermann, 2 September 1974. BStU: HVA 121. 111. This and the following citations are from the ZAIG memorandum of 15 July 1974, cited in n. 108 above.



The Stasi Confronts Western Strategies for Transformation, 1966–1975

207

112. For this rationale, see, e.g., the February 1975 report by [Major Klaus?] Emmerich of the Stasi’s legal department on the “Staatsgrenze zwischen der DDR und der BRD.” BStU: HA II/13–2060. Erich Mielke to his officers, 6 August 1975, depicting the CSCE as recognition of the existing postwar order and demanding close study of the speeches by Brezhnev and Honecker at Helsinki. BStU: HA VI 6298. 113. For the Stasi planning exercises between July and September 1975, see BStU: RS 289, 290, 347. Demands (carefully phrased as implications) for more personnel were a common feature in the reports submitted by all departments involved. 114. Conversation between Honecker and Brezhnev, Berlin, 18 June 1975. SAPMO: DY 30/ J IV 2/2/1567. 115. Conversations and letter exchange (2005–2007) between the author and Siegfried Bock, head of the GDR’s delegation to the CSCE negotiations. See also Siegfried Bock, “The CSCE—an Era of Dissent and Consensus,” in Bange and Kieninger, eds., Negotiating One’s Own Demise? (cited n. 3 above). 116. John J. Maresca, To Helsinki—The Conference on Security and Cooperation in Europe, 1973–1975 (Durham, 1987), 80. Maresca was a junior member of the U.S. delegation to the CSCE negotiations. 117. Perhaps aiming at adding another letter on German unification to the CSCE Final Act as had been done in the Treaty of Moscow in 1970. In retrospect—fifty years after Helsinki in 2005—former Foreign Minister Hans-Dietrich Genscher still claimed that this would have prevented the Federal Government from signing the Final Act, while Egon Bahr thought that the conference had been so important to the goals of Ostpolitik that they would not have allowed this passage to ruin its outcome. Cited in Bock, “The CSCE—an Era of Dissent and Consensus,” in Bange and Kieninger, eds., Negotiating One’s Own Demise? (cited n. 3 above), 3. 118. On the gap between the HVA’s highly successful industrial espionage department (Sektor Wissenschaft und Technik) and the much less convincing utilization of Western know-how (or know-how about the West) for the GDR’s economy, see Jörg Roesler, “Operatives Wissen und praktische Verwertung,” paper presented by Bernd-Rainer Barth at the conference “Hauptverwaltung A—Geschichte, Aufgaben, Einsichten,” Odense, 17–18 November 2007. For the general framing of this structural deficit, see Barth, Ostdeutsche Wirtschaft im Umbruch 1970–2000 (Cologne, 2003); and André Steiner, Von Plan zu Plan—Eine Wirtschaftsgeschichte der DDR (Munich, 2004). 119. The issue of the apparently ever-widening “technology gap” between East and West played a central role in the meetings of Communist leaders in the so-called Permanent Consultative Assembly of the Warsaw Pact, particularly in the mid- and late 1970s. The term itself was introduced into public use by American publicists in 1970. 120. Author’s interviews with various members of the MfAA, the Stasi, and the SED CC (2004–2007). 121. Olav Njølstad introduced the term “Gorbachev miracle” into the discussion at the conference “European and Transatlantic Strategies in the Late Cold War Period (c. 1965–1985) to Overcome the East-West Division of Europe,” University of Copenhagen, 30 November–1 December 2007. The conference proceedings were edited by Poul

208

The Stasi Confronts Western Strategies for Transformation, 1966–1975

Villaume and Odd Arne Westad under the title Perforating the Iron Curtain: European Détente, Transatlantic Relations, and the Cold War, 1965–1985 (Copenhagen: Museum Tusculanum Press, University of Copenhagen, 2010). See therein also Oliver Bange, “The GDR in the Era of Détente—Conflicting Perceptions and Strategies, 1965–1975” (57–78). 122. The GDR’s “bankruptcy” and the decision for a takeover were already implied in Kohl’s ten-point plan in 1989. See, e.g., Helmut Kohl, Ich wollte Deutschlands Einheit (Berlin, 1996), 53ff, 157ff; Wolfgang Schäuble, Der Vertrag—Wie ich über die deutsche Einheit verhandelte (Munich, 1993), 13ff.; Hans-Dietrich Genscher, Erinnerungen (Munich, 1997), 669ff.; various contributions in Theo Waigel and Manfred Schell, eds., Tage, die Deutschland und die Welt veränderten—Vom Mauerfall zum Kaukasus—Die deutsche Währungsunion (Munich, 1994); Frank Elbe and Richard Kiessler, A Round Table With Sharp Corners—The Diplomatic Path to German Unity (Baden-Baden, 1996), 48ff. See also Dokumente zur Deutschlandpolitik, special edition on German Unity, 1989–1990, containing documents from the Chancellor’s Office, ed. Hanns Jürgen Küsters and Daniel Hofmann (Munich, 1998). 123. Jens Gieseke with Doris Hubert, Die DDR-Staatssicherheit—Schild und Schwert der Partei (Bonn, 2000). The Stasi’s coat of arms (“The shield and the sword”), which symbolized its double task of shielding and spreading socialism both inside and outside the GDR, still serves as a popular icon for it. The Stasi appears to have lived up to it inasmuch as it remained—like a shield or a sword—an instrument of the GDR’s political leadership rather than an actor in its own right.

7

The West German Secret Services During the Cold War Holger Afflerbach

Between the “Steamroller” and the “House of Cards”: German Perceptions of Russia in the Twentieth Century The West German secret services after 1945 are a remarkable historical topic.1 They are an important part of the history of the Cold War, but they are also a part of Western Germany’s political culture after 1945. Sometimes they were blamed for being totally dominated by the Americans. Sometimes they were accused of being undermined by old Nazis, by former soldiers of the Wehrmacht, or, even worse, by former members of the SS and the SD. These complaints were expressed not only in Western Europe,2 but even more ferociously in the Soviet bloc, especially in the GDR. Finally, several spectacular failures by the German secret services, which seemed to prove their total incompetence, were the targets of merciless public critique and sarcasm. I focus in this chapter on the main field of activities of the West German secret services during the Cold War—Eastern Europe, especially the USSR and the GDR.3 I analyze the views the services had of the Soviet Union and the possible influence that these views had on political decisions. West German intelligence was not only dealing with the Soviet Union and the Eastern bloc, of course, but also with many other countries. However, it is obvious that in West Germany, a frontline state during the Cold War, the main interest of any intelligence service was the potential threat from the Soviet Union and its allies.



209

210

The West German Secret Services During the Cold War

There is also another reason for this focus: the question of continuities in German perceptions of neighboring countries and enemies. It has frequently been observed that the German perception of Russia has oscillated remarkably between two extremes over the past two hundred years. The first extreme is the notion of the Russian “steamroller”; the second is the idea of Russia as a “colossus with feet of clay” or a “house of cards.”4 These two perceptions of Russia’s overwhelming strength and weakness reflect some realities of its internal structure and are partly a consequence of internal conditions in this vast empire. It is striking, however, that there does not seem to have been any middle position, seeing strengths and weaknesses at the same time, in German perceptions of Russia. One of the two viewpoints always prevailed over the other—and the changes between them were surprisingly sudden. The phenomenon of dramatic and sudden changes in the German perception of Russia is a first-rank problem of German political and military history. Knowing one’s enemy was one of the central problems of twentieth-century German history. Poor intelligence—or poor use of good intelligence—was one of the main reasons for the German defeat in both world wars. Because of the political importance of intelligence, its history should not be only the research field of a small number of specialists, but an integral part of political history. Analyzing intelligence after 1945 raises the questions of whether West German society was the victim of a misperception of Soviet power, and the role of intelligence services in creating these ideas. The German Secret Services After 1945 How many secret services were there in West Germany after 1945? The International Spy Museum in Washington, DC, tells us that the average number of secret services per state is seven, and that one state—it is not named—has as many as one hundred separate secret services. The Federal Republic of Germany is represented by the Bundesnachrichtendienst (BND; Federal Intelligence Service), the Militärischer Abschirmdienst (MAD; Military Protection Service), the Bundesamt für Verfassungsschutz and its sixteen regional Landesämter, the Amt für Nachrichtenwesen der Bundeswehr (ANBW, later ZnBW), the Zentrum für Verifikation in Geilenkirchen, and the A2/G2 functions in the staffs of the armed forces. Moreover, a number of intelligence services no longer exist, like the Friedrich Wilhelm Heinz Dienst and the Social Democrats’ Ostbüro.



The West German Secret Services During the Cold War

211

The Bundesamt für Verfassungsschutz deals with security issues, but not with information about foreign countries. It was founded in 1950 and had a federal branch with around 2,500 employees and sixteen regional branches in the different German federal states. Its task was to protect the German constitution and to observe people and movements that might endanger German democracy. It was influential and was publicly criticized in the 1970s for hunting Communists. Its history is not free of scandals and failures. The first president, Dr. Otto John, fled to East Germany in 1954. After a few months he came back, complaining—most likely falsely—that he had been kidnapped by Eastern bloc services. He went to prison.5 But the Verfassungsschutz’s most spectacular disaster was Chancellor Willy Brandt’s secretary Günther Guillaume. Finding out much too late that Guillaume was an East German spy—and then delaying his arrest to clarify how this had come about—led to Brandt’s early resignation as chancellor in 1974. In 1985, the high-ranking Verfassungsschutz executive Hansjoachim Tiedge fled to East Germany. The second secret service, MAD, is tasked with protecting the armed forces against foreign intelligence, spies, and enemies of the constitution. MAD employs around 1,300 people. Like the Verfassungsschutz, it is an internal intelligence agency and has no mandate or means for observations outside the country. MAD’s most famous failure was the false accusation that the four-star General Günter Kiessling was homosexual. This led to his dismissal, but shortly afterward, it became clear that the charge was unfounded. The responsible minister, Manfred Wörner, was then weggelobt (kicked upstairs)—he was persuaded to apply to become NATO’s secretary-general in Brussels. The West German armed forces also had an agency called the Amt für Nachrichtenwesen der Bundeswehr (ANBw), later the Zentrum für Nachrichtenwesen der Bundeswehr (ZNBw), which analyzed security issues. Founded in 1976 by Federal Minister of Defense Hans Apel, this had around 650 members. It was more a client of intelligence than a producer of it. Its brief was similar to that of the BND, into which it was incorporated on 31 December 2007. The BND is responsible for providing intelligence about foreign powers, and its activities during the Cold War are therefore the main focus of this chapter. Reinhard Gehlen and the BND The BND had—and has—the double task of collecting intelligence abroad and also evaluating it. It also conducts economic and scientific espionage, but

212

The West German Secret Services During the Cold War

during the Cold War its focus was mainly on military and strategic questions. The history of this organization is remarkable.6 It is, to a very great extent, linked with the story of one man—General Reinhard Gehlen.7 Born in 1902, Gehlen was a general staff officer who was famous for his organizational talents and his ability to work hard. For that reason Chief of Staff General Franz Halder appointed him head of the Fremde Heere Ost (FHO; Foreign Armies East) department of the Wehrmacht’s Supreme High Command in 1942. The task of this department of the general staff was to analyze Soviet military strength. After the shock of discovering Russia’s unexpected strength, Halder considered rightly that this department was highly ineffective. Gehlen took over and changed the FHO dramatically. Whether his work was effective, however, is highly debated. He claims to have foreseen every Russian offensive, which is definitively not true for 1942 and the battle of Stalingrad. Gehlen considered the Nazis’ harsh treatment of the Russian population to be a huge mistake. He was a convinced and lifelong anticommunist and favored close cooperation with Russian anticommunists like General A. A. Vlasov. But this was permitted by Hitler only in the final stages of the war. Gehlen gave correct estimates of the Russian Army’s strength in 1945 and was finally dismissed by Hitler, who considered him “a fool” and “defeatist” who belonged in a madhouse.8 Gehlen and his FHO staff were already planning far ahead. Seeing German defeat as inevitable, they reflected as early as 1943 on the question of how to safeguard their knowledge of the Russian Army.9 Gehlen ordered copies to be made of all of the FHO’s intelligence material, to evacuate it to Bavaria and hide it there in fifty waterproof metal boxes. He also ordered his staff to hide somewhere nearby. After the German surrender, Gehlen contacted the American authorities and offered them his services. His attempts to keep his information and also his manpower together were surprisingly successful. After a short time in American POW camps, Gehlen’s former FHO staff went to work for the United States, paid with American money and making additional money from black market operations such as the selling of coffee and penicillin.10 They moved into a camp in Pullach, near Munich, where they lived nearly hermetically secluded from the surrounding world. In the first years, even the families of the service members lived “on campus.” Gehlen offered his employees American living standards and helped many former officers who otherwise would have been unemployed.



The West German Secret Services During the Cold War

213

The main task of the “Organization Gehlen,” as the service was called until 1955, was, according to the head of the U.S. G-2 service in Europe, “to deliver to American authorities information on Russian military, political and economical strength in that order of priority.”11 Gehlen had rightly understood that the Americans, who had dissolved their wartime intelligence services, lacked intelligence on their wartime ally, the Soviet Union; and he could offer them what they needed. By 1946, Gehlen’s organization —the general himself was at that time in the United States—had already delivered sixty-nine major reports to its American masters.12 The political atmosphere started to change; the Cold War began. An insider argued: “The CIA loved Gehlen because he told us what we wanted to hear.”13 Some American officers complained that the reports coming from the “Organisation Gehlen” provided details that were purely tactical and not of major strategic importance; but they had no better information. General Curtis LeMay claimed later that the first nuclear planning of the Strategic Air Command had been based 70 to 80 percent on intelligence delivered by the “Organisation Gehlen,” 14 and, according to estimates, 70 percent of the American figures on the Soviet Army were also based on intelligence provided by Gehlen and his agents.15 Therefore, the “Organisation Gehlen” played an important role at the very beginning of the Cold War. Gehlen, a German nationalist as well as a convinced anticommunist, tried to implant his worldview on the Americans as well as on his fellow Germans. Gehlen was convinced that the Soviet “steamroller” would not suddenly stop its successful drive westward, but would in the near future try to overrun Western Europe.16 It is astonishing that Gehlen’s staff survived defeat and surrender and continued to work. The “Organisation Gehlen” was taken over in 1956 by the Federal Republic of Germany and renamed the BND. Gehlen was an unusually gifted networker. He had contacts with Chancellor Konrad Adenauer of the Christian Democratic Union (Christlich Demokratische Union; CDU) as well with the head of the German Social Democratic Party (Sozialdemokratische Partei; SPD), Dr. Kurt Schumacher. He employed men such as General Adolf Heusinger, one of the founding fathers of the Bundeswehr, helping him and others to bridge the gap between German disarmament in 1945 and rearmament in 1955.17 Gehlen also networked with Generals Hans Speidel and Hermann Foertsch, two other founding fathers of the Bundeswehr. He became extremely influential through arranging the reemployment of former Wehrmacht specialists in the new West German Army.18 Gehlen and his

214

The West German Secret Services During the Cold War

organization thus played a key role in the making of West German security policy after 1945, although some high-ranking officers and politicians complained that it sabotaged the careers and appointments of those it did not favor by defaming them in reports.19 Gehlen was not only an able and inventive organizer, but also an eccentric. He did not behave like the head of a large intelligence agency, but rather like a spy. He declared openly that it was his intention to make the structure of the service as unclear as possible so as to make it more difficult for foreign services to penetrate it.20 It was impossible to apply to work for the BND—for security reasons, it recruited new members by networking. It thus had the structure of an “old boys’ network,” and a large number of former members of the Nazi Reichssicherheitshauptamt (Reich Security Head Office) got in— brothers-in-arms brought in their former comrades. After a while, nobody, not even Gehlen himself, was able to understand the chaotic structure of the BND. The regional branches were given names as if the BND were an insurance agency,21 and incoming intelligence went through slow and complicated procedures to protect sources in the Eastern bloc. Gehlen thought it inevitable; a secret service had to be like this. He traveled under the pseudonym “Dr. Schneider” and took different routes to work every day. It was said that it was his ambition to compete with Greta Garbo in being invisible to the public.22 Gehlen was called a mixture of an intelligence specialist, an actor, and a gambler.23 He was all of these things, but he was even more: perhaps by nature, perhaps as a result of his profession, he was paranoid. During the Adenauer era, he became a legendary figure, and it was widely believed that he headed the most effective intelligence service in Europe.24 But later some spectacular failures eroded this prestige. The most spectacular case was the discovery that one of the BND’s top agents, Hans Felfe, who had enjoyed a phenomenal rise in the service, was a Soviet agent.25 The damage was enormous. A former British intelligence officer said that even in the 1980s, MI6 was reluctant to share its information with the BND, because it thought that the German service was undermined by Russian spies. Other factors added to diminish the prestige of the BND. In the more liberal climate of the late 1960s, Gehlen’s stubborn anticommunism, his strange behavior, his nepotism, 26 and especially the increasing organizational chaos in the BND made him an increasingly odd figure. His time was over, and he was succeeded by General Gerhard Wessel, who had more managerial ideas about how to structure and lead a big intelligence service.



The West German Secret Services During the Cold War

215

Where was the focus of the BND’s work? Its predecessor, Fremde Heere Ost, had been a military intelligence service. Originally, its task had been purely analytical, studying intelligence about the Soviet Army, its strength and intentions. But the FHO did not generate the intelligence itself. This was provided by other organizations, by army units, by interrogating Soviet prisoners of war, from papers captured by German troops, by spies and agents, and so on.27 Only in 1944 did the FHO start to collect some of its intelligence itself. Inasmuch as Gehlen’s services were attractive to the Americans because he offered a spy network in Eastern Europe, the “Organisation Gehlen” and later the BND both collected intelligence and analyzed it. It is not surprising that the BND interpreted its raison d’être as being similar to that of the old FHO. BND personnel regarded it as a military intelligence service, and their main efforts were directed toward the Soviet Union. They collected and analyzed intelligence on Russian intentions and the order of battle of the Red Army and its allies. Here ideology also came into play. Gehlen’s first successor at the head of the BND, Wessel, thought that the analysis provided by the BND about the stability and structure of states, societies, and ideologies of foreign powers was at least as important for policy-makers as classical intelligence.28 The BND never tired of analyzing Soviet Communism and its expansionist idea of world revolution; Gehlen’s memoirs are the best example of this. The BND had an ideological mission and warned against the permanent threats of Communism for West German society. It feared a military attack from the Soviet Union as well as the subversive activities of Communist agents undermining German society.29 The BND tried several times to protect Germany against real or imaginary internal enemies. Politicians with leftist sympathies or suspicious contacts in Eastern Europe, like Gustav Heinemann and Egon Bahr, were under observation.30 The BND took its self-imposed task of guarding against hostile ideologies very seriously and perhaps even prioritized it. The BND delivered its intelligence to the Bundeskanzleramt (Federal Chancellor’s Office), which was therefore its direct superior. Hans Globke, the head of this office under Adenauer, was an able administrator and one of the chancellor’s most important advisers, but he came under heavy fire for having helped draft the 1935 Nuremberg laws. Gehlen and Globke worked very well together. The BND also delivered its intelligence to the Foreign Office, to the Ministry of Defense, to the Bundesamt für Verfassungsschutz, and to other agencies. The intelligence was presented in

216

The West German Secret Services During the Cold War

the form of daily and weekly reports, studies of single countries and problems, and, during the Cold War, the Lageberichte Ost (situation reports on Eastern Europe).31 Globke also used the BND for his own purposes. He financed the service with special funds and ordered it to report on individuals within Western Germany.32 In that way, the BND became a part of the “Adenauer state,”33 one of the pillars of so-called Kanzlerdemokratie (chancellor democracy). The BND, Adenauer, and West German Rearmament Adenauer liked Gehlen, whose anticommunist convictions he shared, and addressed him as “my dear General.”34 He was a grateful recipient of Gehlen’s weekly reports and used them to correct information from other services, for example from the Foreign Office.35 The historian Arnulf Baring even claims that the reports of the “Organisation Gehlen” were the main sources for Bundeskanzleramt information about Eastern Europe.36 These reports took a “mainly military and strategic point of view. They described the strategic situation as highly critical.”37 They had an influential role in the beginning of the Cold War by alarming the politicians in Bonn and Washington. Especially after the outbreak of the Korean War, the fear was growing that a Soviet invasion of Western Europe was impending. The “Organisation Gehlen” had contributed to raising this fear—and shared it. Gehlen and his men were convinced that Western Europe would be lost without a defense contribution from West Germany. They had warned against the Soviet aggression and reported on the clandestine rearmament of East Germany in the shape of the “Kasernierte Volkspolizei” (KVP; paramilitary people’s police housed in barracks).38 It seemed possible that something like what was happening in Korea could happen to West Germany—East German forces could penetrate West German territory while Russia stood aside. The Western powers refused to give the German government a guarantee for its security. Would they start a world war to stop limited Soviet bloc aggression? West German rearmament seemed necessary to counter a surprise attack. The Western allies wanted German help for the defense of Western Europe. But shortly after the war, dislike of rearming West Germany was still very strong, and there was evident reluctance to build up an independent German army. Gehlen and his people wanted one allied with the Western powers and guaranteeing German independence. The Western powers instead wanted small German contingents under their own command. This was unacceptable to the German government



The West German Secret Services During the Cold War

217

and also to Gehlen’s service, whose role as anticommunists and Westerners now conflicted with their self-perception as German patriots. Adenauer wanted rearmament and the buildup of a German army only if the Western allies paid a price for it: he wanted full sovereignty for Germany. But rearming Germany hit another obstacle: the overwhelming majority of all Germans were against it.39 Gehlen and his agency had an important role in this complicated game. They had provided the government and the allies with the alarming news from Eastern Europe, and they did even more. Gehlen provided General Speidel with figures about Soviet military strength so that Speidel could write a memorandum on the size and structure of a future West German army.40 Adenauer had appointed a security adviser who had to research all these questions of German rearmament. Gehlen was one of his top candidates for the post, but could not do it because of his close connections with the Americans. The British and the French would have objected.41 General Graf Gerhard von Schwerin therefore got the post. He built up his own secret service, the Friedrich Wilhelm Heinz Dienst, led by former Colonel Heinz. Heinz challenged Gehlen’s monopoly of intelligence, but he had no chance against him. He became victim of certain exaggerations in his reports about the Soviet ability to wage war.42 His service was short-lived. The Gehlen Organization played an important role in the entire question of rearmament. It helped to sustain the atmosphere of fear that was necessary to influence the government to make the highly unpopular rearmament a reality. It is, of course, a question whether its contribution was indispensable, or if everything would have happened quite similarly without it. Gehlen’s intelligence was only part of the overall picture, but an important part. Much of the information offered by his agency was of some use to Adenauer. Gehlen delivered welcome facts; he helped the chancellor to strengthen his own arguments, but he did not form Adenauer’s opinion. This is certainly the final verdict of historians such as Adenauer’s biographer Hans Peter Schwarz.43 Adenauer’s fear of the Soviet Union had a much larger basis than purely military considerations; it was primarily politically and psychologically motivated. Military Espionage in Eastern Europe The BND’s main target remained the armies of the Warsaw Pact, especially the National People’s Army (Nationale Volksarmee; NVA) of the GDR, but

218

The West German Secret Services During the Cold War

also the Group of Soviet Forces in Germany (GSFG). This amounted to a very considerable force of over 400,000–600,000 men,44 and Western intelligence services consistently reported that these troops were ready to strike at very short notice.45 Soviet leaders like Generals Zhukov or Malinovsky thought that in case of war, the Soviet Army would be able to break through very quickly and reach the Channel after a few days.46 This was widely believed in the late 1940s and early 1950s. The BND reported in 1967 that the Russians still believed that the Red Army would be able to reach the French, Spanish, and Portuguese coasts in only twenty days. It seems that the BND considered Soviet planning unrealistic—for very good military reasons.47 But the service warned in its reports that the Warsaw Pact armies were formidable and a dangerous opponent, able to attack without warning. They wrote too that the Eastern bloc was still improving its military power, despite its official détente feelers.48 The BND considered it of primary importance to check every move of this powerful opponent. The first phase of observation began immediately after the war. Former German POWs, coming back from prisoner camps somewhere in the Soviet Union, had to fill out questionnaires. Their information gave an accurate idea about the internal situation of the Soviet Union. Other information was gained by risky humint operations inside the USSR. But spying inside the Soviet Union was dangerous. The Soviet authorities executed more than 2,000 Germans in the early 1950s; more than half for espionage.49 The second phase lasted from the beginnings of the GDR until the erection of the Berlin Wall (1949–1961). The West German secret service spied successfully in East Germany, using a large number of volunteers among the East German population. American sources estimated that more than 4,000 spies worked for the Western powers behind the Iron Curtain.50 Detailed spying targeted both the GSFG and the East German NVA and Kasernierte Volkspolizei (paramilitary police). Airports, aircraft, military units, tanks, artillery, and equipment were all carefully observed. West Berlin, easily approachable from east and west, was the center of exchange of information. The BND compiled the intelligence on file cards (Karteikarten) in a series begun by Fremde Heere Ost.51 The BND observed GSFG 162. The number of file cards shows the density of the operation: only twenty garrisons got fewer than six file cards full of information, sixty-two got more than twenty cards, and eighteen more than forty.52 The most highly observed garrison was Krampnitz, where the 219th Soviet Anti-Aircraft regiment and the 29th Motorized Infantry



The West German Secret Services During the Cold War

219

Regiment were located. The BND collected 102 file cards full of information about these units, giving details of their strength, location, equipment, and general condition.53 In the opinion of historians such as Paul Maddrell, Armin Wagner, and Matthias Uhl, who have made extensive use of the archives of the East German Ministerium für Staatssicherheit (MfS; Stasi), the GDR was undermined by Western espionage. They have challenged the traditional view of a powerful Stasi undermining West Germany. The opposite seems to be more true, despite the quip that East Germany was not a state with a secret service, but a secret service with a state. Nevertheless, the mighty Stasi was unable to protect East Germany from Western espionage—and they knew it. They tried energetically to change this situation. Ernst Wollweber, who headed the Stasi from 1953 to 1958, worked hard to destroy the BND network, but with only partial success.54 The construction of the Berlin Wall in 1961 stopped, not only the unsupportable drain of refugees, however, but also the ease with which the BND and the other Western intelligence services could maintain spy networks in the GDR.55 The Wall strangled the BND’s humint network. Slowly this network of spies in the GDR dried up. The BND substituted so-called Reise und Transitspione—West Berlin or West German travelers and truck drivers and elderly East Germans, who were allowed to travel after retirement. Retired East Germans, some of them nearly eighty years old, spied for the BND.56 The Stasi arrested some 5,000 Western spies. 80 percent of them caught while spying on military installations.57 An important part in this game was also played by military attachés in Eastern Europe, who used diplomatic immunity to spy on tanks, artillery, airplanes, airports, barracks, radio communications, and radar stations, sometimes photographing these from their cars while driving by.58 Another important means of obtaining information was sigint: electronic spying, or observation of radio communications (Fernmelde- und elektronische Aufklärung). It had its origins in World War II, but was growing in importance. A series of spy stations were built along the border of East Germany to monitor military radio communications and draw conclusions about the movements and training of Warsaw Pact troops. Sigint started to replace the risky observation of military units by agents. According to statistics, the BND got 75 percent of its information from open sources and only 25 percent from intelligence. Three-quarters of the latter, or 18.5 percent of all the BND’s information, derived from sigint.59 The BND collected around one hundred

220

The West German Secret Services During the Cold War

thousand reports in 1989, but only 450 contained useful information.60 Not only sigint, but also imagery intelligence, or imint, played a significant role, especially because one-sixth of the territory of East Germany could be spied on by planes flying to and from West Berlin. The British and American intelligence services had greater imint potential because of their large-frame computers and excellent electronic spying technology, especially spy satellites. In the early 1960s, Franz Joseph Strauß and some other West German politicians came to think that the other Western intelligence services were far ahead of the BND. They complained that, as a consequence of this backwardness of its secret service, the Federal Republic had no independent information about the military situation in East Germany and was dependent on the Americans. In their eyes, Gehlen was blowing up the importance of the limited information he got.61 This seems to be too harsh. The BND was able to continue its successful work in East Germany by means of sigint.62 In the 1980s, the BND even used electronic devices that allowed it to check on the deployment of atomic weapons on railroad trains by measuring their radioactivity.63 How can we judge military espionage of the BND in Eastern Europe and its political importance? In total, the BND seems to have been successful. This was at any rate the opinion of the Stasi, which noted on 2 February 1984: “The enemy was successful in ascertaining the deployment, structure, and main equipment of the National People’s Army. They gave roughly correct estimates of our manning levels, of our operational readiness.”64 The BND’s military espionage was thus successful. But was this intelligence important? Did the BND produce something politically relevant? The BND produced annual reports of the Eastern bloc’s military strength.65 It also exchanged intelligence with other Western intelligence services, and it seems that it was accurate. The BND maintained a kind of alarm system to guard against a Soviet surprise attack. The real test would have been an aggressive war started by the Eastern bloc. This did not take place and was, as we know today, also never planned by the Soviet leaders. But the BND was able to locate movements of Warsaw Pact troops, to recognize relocations of military units, and to draw the right conclusions. This was demonstrated when the Warsaw pact invaded Czechoslovakia in 1968. The BND foresaw this intervention by recognizing the deployment of troops,66 and it also interpreted the sudden silence of all Eastern bloc military radio communication as an unmistakable sign for imminent action. It is undeniable that this ability to



The West German Secret Services During the Cold War

221

foresee a Soviet attack would have been a great asset in case of war. And the BND claims to have been the best military intelligence service in Europe. Its agents estimated that in the early 1970s, they provided 70 percent of all military information about the Warsaw pact armies.67 On the other hand, BND agents were highly ideological. They were very good at evaluating military information, but their anticommunism made it difficult for them to be clear in their judgment. This is true especially of the early years. After a conference in June 1951, for example, Otto Lenz, the secretary in the Bundeskanzleramt (Federal German Chancellery), noted: “The gentlemen [of the BND] were afraid that a war was imminent in a short time; they had their observations. They recognized that the Russians were preparing their troops in East Germany step by step for constant combat readiness. On the other hand, Russia and the East were industrially so backward that it was not expected that they would risk a war in their actual economic condition.”68 The BND wavered, torn between observation and ideology. It saw the military strength of the Soviet Union and believed that the Eastern bloc was willing to attack despite the fact that it was not ready for war. And its belief in Eastern aggressiveness led it to interpret its military intelligence in ways that were not always justified. The BND also undertook military espionage in the West. In 1956, the British government tried to convince the American government to base the defense of Europe on atomic weapons so as to reduce expenses for conventional armies. The British and American governments wanted to keep these plans secret from their French and German allies, because it meant a “thermonuclear bomb or nothing” and “liberation by devastation”—a disastrous concept that could not find favor with continental Europeans. The BND discovered this, and Gehlen informed Adenauer, much to the annoyance of the U.S. State Department, which complained that Gehlen was basing his analysis on “miscellaneous data gathered from sundry sources.”69 The BND’s Political Prognosis What of the political prognoses of the BND? This was a field where humint as well as the ideology of the analysts played a crucial role. Some in Bonn complained about the quality of BND reports, which were criticized in the 1950s as blown-up newspaper articles and reports on trips and conversations,

222

The West German Secret Services During the Cold War

but vaguer than newspapers, because the BND concealed its sources.70 Interior Minister Gustav Heinemann suggested that the BND could halve its research staff if it subscribed to the Neue Zürcher Zeitung (a leading Swiss newspaper).71 Chancellor Helmut Schmidt is reported to have complained that BND reports told him nothing he had not previously read in the newspapers. Indeed, the work of analysts and that of journalists are quite similar: they both have to collect facts and give a sensible interpretation to them. It depends on the personal abilities of a BND resident at a German embassy (normally known as Pol 2) if he knows more and is better informed than the foreign correspondent of an important international newspaper or of a news agency. Complaints that the BND did not know more than the newspapers were thus justified. A high-ranking BND official complained in the early 1970s that the politicians sometimes expected the impossible: “We do not have a man sitting under Ulbricht’s desk.”72 The BND could not read Ulbricht’s mind either. It had high-ranking spies, but even they could not predict spontaneous events or decisions made by a very small group of politicians. Nevertheless, the BND was able to deliver important intelligence. During the Berlin blockade, the Americans used the intelligence delivered by the Gehlen Organization to calculate the military risks in this tense situation.73 They wanted to know what the Russians were going to do; if they were willing to escalate the blockade into a war. Information about Soviet Army movements provided by the “Organisation Gehlen” was thus important. It was discovered, for example, that the Russians had brought reinforcements to the Berlin area from Austria.74 Later, the “Organisation Gehlen” informed both the German government and the Americans about the first secret steps toward East German rearmament, the creation of the Kasernierte Volkspolizei. A mark of the ability of the “Organisation Gehlen” to foresee political events was the East German uprising of 17 June 1953. Desperate East German workers revolted against the Communist government, which had imposed unrealistically high work loads on them in order to fulfill an overambitious economic program. The “Organisation Gehlen” had a spy in a high-ranking position on the GFR’s economic planning committee. Gehlen’s men knew that the new economic plan was unrealistic,75 and they knew about the discontent. But they did not foresee the rebellion. The Americans learned about the uprising from newspapers and complained about not receiving any advance warning of it from Gehlen’s agency, which had cost them so much money.76 The



The West German Secret Services During the Cold War

223

“Organisation Gehlen” subsequently delivered a good analysis of the events, however, using hundreds of reports. The agency analyzed its work in detail and saw no reason for regret. It admitted that there had been some mistakes, but concluded that it was impossible to foresee spontaneous events. Moreover, it was essential for it to protect its sources in East Germany. Since a minimum of two days was needed to transfer secret information to the West, it had been impossible to give advance notice of the uprising on 17 June.77 The BND was also unable to foresee another prominent, GDR-related event: the erection of the Berlin Wall in August 1961. It had foreseen that something was going to happen; the drain of refugees was growing, and the East German authorities had to act. But the BND thought that something like the Berlin blockade would happen again. They had observed that the East German troops were equipped with materials for blockading streets; but they did not foresee the construction of a solid wall or the exact date. This was a lasting disappointment for Willy Brandt, who was at that time mayor of West Berlin. He continued to be very skeptical about the quality of the BND’s political reports. Sometimes the BND delivered good news. For example, in November 1962, it observed Soviet soldiers in East Germany going out and having fun. The alarming situation around the Cuban Missile Crisis was obviously over, and the Soviet Army was no longer under alert.78 The BND had excellent connections in the Middle East—with some Arab countries as well as with Israel. It was able to foresee the Israeli attack in 1967 up to the day and informed other Western intelligence services, notably the CIA.79 The BND also foresaw the blocking of transit routes to West Berlin by the East German authorities; but it did not foresee the revolts in Poland in 1970, giving Brandt reason to complain.80 Did the BND foresee the downfall of East Germany? According to HansGeorg Wieck, who headed the BND from 1985 to 1990, the service had a more precise idea of the economic difficulties in East Germany than other West German government agencies like the Bundeskanzleramt or the Ständige Vertretung (Permanent Representation) in East Berlin, which believed falsified official GDR statistics.81 In 1986, the BND also started to interview batches of six hundred refugees from East Germany every six months. It discovered that the vast majority of them—more than 72 percent—favored German reunification.82 This information and the reports about the dangerous economic weakness of the GDR were not used by the West German

224

The West German Secret Services During the Cold War

government, because according to Wieck it was trying to avoid an untimely reopening of the “German question.”83 He explained this in terms of the Bonn government’s notoriously poor relations with the BND, which it mistrusted. Despite its knowledge of the GDR’s economic difficulties, the BND did not foresee the sudden breakdown of East Germany and the opening of the Wall. It thought that the GDR would continue to exist as long as Soviet forces remained—and thought that the Russians would never go.84 According to Wieck, it “believed for a very long time that the East German government would be able to keep its state running, whatever the economic conditions.”85 The BND was not alone in this opinion—everybody in West and East Germany thought so until the summer of 1989.

Conclusion The BND did good work in one area—military espionage. Politicians like Brandt or Schmidt complained about its reports in the political field. But Brandt admitted that its military reports were accurate; and Schmidt’s assistant reports that he also studied them.86 Copious use of both humint and sigint, combined with very good knowledge of the Soviet Army, helped make the BND up to its task as a military intelligence service. On the political side—not in the field of getting intelligence, but in the field of analyzing it and constructing a bigger picture—the balance is less brilliant. The BND was a veteran of Adenauer’s Germany. It was conservative, highly ideological, opposed to détente, anti–social democratic,87 and very anticommunist. The BND was an integral part of Germany during the Cold War, causing political developments of that time as well as reacting on them. Heinz Höhne and Hermann Zolling wrote in their classic study of the BND in 1971 that Gehlen’s men “had no understanding for political nuances and strategic motives. They had a tendency to demonize the enemy; this made it impossible for them to understand the inner mentality of the Soviet Union. But they were masters of military intelligence.”88 Having consulted a selection of the BND files now open for examination at the Bundesarchiv in Koblenz, one is inclined to agree. This mixed verdict quite accurately describes the achievements and the limits of the BND and its importance for political decision-making in the Federal Republic of Germany. Aside from this, the BND and its bizarre founder offer more than



The West German Secret Services During the Cold War

225

enough material for a cultural history, which would also shed interesting light on the general political history of postwar West Germany. Notes 1. This chapter draws on material in the Bundesarchiv Koblenz, where a number of files are available for public use. See Elke Ursula Hammer, “Archivwesen im BND und Bestand B 206 im Bundesarchiv: Von Quellen und Methodenschutz und dem historischen Interesse,” www.bundesarchiv.de/aktuelles/aus_dem_archiv/nitteilungen/00087/index. The Bundesarchiv has good finding aids for “B 206” BND files, and Bundeskanzleramt records (e.g., BA, B 136/4869, 4870, and 4887) offer additional information. The BA Koblenz has also the monthly reports of the Verfassungsschutz from 1950 to 1976 (BA, B 443/527–570: numbers are not chronological); they contain interesting overviews on political extremism in Western Germany and also some material on political oppression in East Germany. In evaluating the political importance of the secret services, I was able to make use of some excellent books on the BND. Extremely informative: Heinz Höhne and Hermann Zolling: Pullach intern: General Gehlen und die Geschichte des Bundesnachrichtendienstes (Hamburg, 1971), trans. as The General Was a Spy: The Truth About General Gehlen and His Spy Ring (New York, 1972); this extremely readable analysis has stood the test of time well. Höhne and Zolling did not have archival material at their disposal, but a lot of insider information. Their account—and also their judgment—is generally accurate. For new details and some additional precision, see Peter F. Müller and Michael Mueller, Gegen Freund und Feind: Der BND, geheime Politik und schmutzige Geschäfte (Reinbek bei Hamburg, 2002). Useful: Wolfgang Krieger and Jürgen Weber: Spionage für den Frieden: Nachrichtendienste in Deutschland während des Kalten Krieges (Munich, 2001). A highly interesting and readable collection of biographies of the chiefs of the West and East German secret services is provided by Dieter Krüger and Armin Wagner, eds., Konspiration als Beruf. Deutsche Geheimdienstchefs im Kalten Krieg (Berlin, 2003); see in particular Dieter Krüger, “Reinhard Gehlen. Der BND-Chef als Schattenmann der Ära Adenauer,” ibid., 207–236. Armin Wagner and Matthias Uhl, BND contra Sovietarmee. Westdeutsche Militärspionage in der DDR (Berlin, 2008), a sober account and an extremely valuable source for all questions of military espionage, uses German as well as Soviet archival material and also contains much information about the East German secret service. Another important book on the topic is Mary Ellen Reese, Organisation Gehlen. Der Kalte Krieg und der Aufbau des deutschen Geheimdienstes (Berlin, 1992). Reese is less critical of Gehlen than many others, including Höhne, who provided an introduction. James H. Critchfield, Auftrag Pullach: Die Organisation Gehlen, 1948–1956 (Hamburg, 2005), deals with the early phases of the “Organisation Gehlen.” The numerous books and articles of Erich Schmidt-Eenboom, a prominent German secret-service specialist, benefit from the expertise of their author, but mainly

226

The West German Secret Services During the Cold War

cover the past twenty years and less so the earlier period: see Schmidt-Eenboom: Der BND—die un heimliche Macht im Staate: Schnüffler ohne Nase (Düsseldorf, 1993); id., Geheimdienst, Politik und Medien:Meinungsmache Undercover, Edition Zeitgeschichte 16 (Berlin, 2004); id., Der Schattenkrieger: Klaus Kinkel und der BND (Düsseldorf, 1995); id., Undercover: Wie der BND die deutschen Medien steuert (Munich, 1999). Schmidt-Eenboom’s books suffer in general from a bias: on the one hand, he blames the services for being incompetent; on the other, he warns that they are an extremely dangerous and powerful force, which seeks to undermine German society. In his books Der Dienst: Erinnerungen, 1942–1971 (Mainz, 1971), trans. as The Service: The Memoirs of General Reinhard Gehlen (New York: World, 1972), Zeichen der Zeit: Gedanken und Analysen zur weltpolitischen Entwicklung (Mainz, 1973), and Verschlußsache (Mainz, 1980), Reinhard Gehlen offers more a political and ideological analysis from his own anticommunist point of view than an analysis of his secret service work, and he tends to leave out unpleasant facts. Heinz Felfe, Im Dienst des Gegners. 10 Jahre Moskaus Mann im BND (Hamburg, 1986), is a polemical and satirical—and therefore highly readable—account written by the most famous double agent inside the BND. Felfe was famous for his arrogance, and the style of his memoirs reflects it. 2. Krüger, Gehlen, 224, quotes an article by Sefton Delmer in the Daily Express, 17 March 1952, in which Delmer described Gehlen as a loyal Nazi who also employed a significant number of former Nazis in his service. 3. Felfe, Im Dienst des Gegners, 271, ranks BND priorities: East Germany was no. 1, the Soviet Union no. 2, Poland no. 3, the CSSR no. 4; then the other East European Communist countries; China was no. 10, North Korea, no. 11. 4. Andreas Hillgruber: “Deutsche Rußland-Politik, 1871–1918: Grundlagen– Grundmuster–Grundprobleme,” Saeculum 27 (1976): 94–108; Walter Laqueur, Deutschland und Rußland (Berlin, 1965). 5. Felfe, Im Dienst des Gegners, 175–188. 6. Höhne and Zolling, Pullach intern; Müller and Mueller, Gegen Freund und Feind; Reese, Organisation Gehlen; Critchfield, Auftrag Pullach. 7. On Gehlen, see first Krüger, Gehlen, and the sources cited in nn. 1 and 6 above. 8. Reese, Organisation Gehlen, 31–32. 9. Höhne and Zolling, Pullach intern, 96. 10. Felfe, Im Dienst des Gegners, 154 –156, with a description of these illegal activities; Krüger, Gehlen, 221. 11. Samuel Bossard to Director CIG, 5 May 1947, cited in Wagner and Uhl, BND, 48. 12. Ibid. 13. Krüger, Gehlen, 222. 14. Wagner and Uhl, BND, 49. Krüger, Gehlen, 221–222. 15. Höhne and Zolling, Pullach intern, 135. 16. Ibid., 134. 17. One of Gehlen’s collaborators, Heinz Herre, wrote later: “Es kam damals darauf an, möglichst viele Offiziere von der Straße zu holen [It was necessary to employ as many officers as possible].” Quoted in Felfe, Im Dienst des Gegners, 157.



The West German Secret Services During the Cold War

227

18. Werner Abelshauser and Walter Schwengler, Wirtschaft und Rüstung, Souveränität und Sicherheit, vol. 4 of Militärgeschichtliches Forschungsamt, ed., Anfänge westdeutscher Sicherheitspolitik: 1945–1956 (Munich, 1997), 137. 19. Lutz Köllner, Die EVG-Phase, vol. 2, of Militärgeschichtliches Forschungsamt, ed., Anfänge westdeutscher Sicherheitspolitik: 1945–1956 (Munich, 1989), 600. 20. Gehlen, Dienst, 238. 21. Felfe, Im Dienst des Gegners, 160, with a colorful description of a “General Agency” in Karlsruhe. 22. Höhne and Zolling, Pullach intern, 302. 23. Krüger, Gehlen, 236, quoting General Gerd Schmückle. 24. Wagner and Uhl, BND, 65. 25. Höhne and Zolling, Pullach intern, 292. 26. Many examples of Gehlen’s nepotism in Felfe, Im Dienst des Gegners, 188–213. 27. Bundesarchiv, B 206/99–100; also B 206/101–106. 28. Wagner and Uhl, BND, 193, quoting Gerhard Wessel “BND—der geheime Auslandsnachrichtendienst der Bundesrepublik Deutschland,” Beiträge zur Konfliktforschung 15, no. 2 (1985): 5–23, from 12. 29. Gehlen, Dienst, passim; esp. 418. 30. Höhne and Zolling, Pullach intern, 321. 31. Bundesarchiv, B 206/117–177, Militärische Lageberichte Ost, Jahresberichte 1960–1991); B 206/178–204, Militärische Lageberichte Ost, Monatsberichte 1/1960– 12/1974; B 206/205–414, Militärische Lageberichte Ost, Wochenberichte 1960–1992. There were also Lageberichte West, but they were much less frequent (38 vol.). See also Höhne and Zolling, Pullach intern, 16. 32. Krüger, Gehlen, 223. 33. Ibid., 225–226. 34. Höhne and Zolling, Pullach intern, 237. 35. Gehlen, Dienst, 211. Höhne and Zolling, Pullach intern, 268, write that Adenauer checked Foreign Office and Verfassungsschutz (Federal Office for the Protection of the Constitution) against BND reports, and Foreign Minister Heinrich von Brentano was told in Adenauer’s marginalia in the latter that the BND was better informed than the diplomats. Adenauer’s enthusiasm was not shared by the Foreign Office. Diplomats complained that BND reports read like “reports of travels and conversations in newspapers, but less detailed, with unclear sources, full of ‘Secret’ stamps and carefully camouflaged informants.” 36. According to Arnulf Baring, Außenpolitik in Adenauers Kanzlerdemokratie, 2 vols. (Munich, 1971), 1: 36, the “Organisation Gehlen” was the Federal Chancellor’s Office’s main source of information on Eastern Europe. See also Gehlen, Dienst, 182, 209. 37. Gero von Gersdorff, Adenauers Außenpolitik gegenüber den Siegermächten 1954: Westdeutsche Bewaffnung und internationale Politik (Munich, 1994), 58. 38. Bruno Thoss, ed., Volksarmee schaffen—ohne Geschrei! Studien zu den Anfängen einer “verdeckten Aufrüstung” in der SBZ/DDR, 1947–1952 (Munich, 1994).

228

The West German Secret Services During the Cold War

39. Holger Afflerbach, “Das Militär in der deutschen Gesellschaft nach 1945,” in Afflerbach and Christoph Cornelißen, eds., Sieger und Verlierer. Materielle und ideelle Neuorientierungen nach 1945, Kultur und Erkenntnis 16 (Tübingen, 1997), 249–272. 40. Roland G. Foerster et al., Von der Kapitulation bis zum Pleven-Plan, vol. 1 of Militärgeschichtliches Forschungsamt, ed., Anfänge westdeutscher Sicherheitspolitik: 1945–1956 (Munich, 1982), 414. 41. Ibid., 459; Höhne and Zolling, Pullach intern, 237. 42. Foerster et al., Von der Kapitulation bis zum Pleven-Plan, 468. 43. Hans Peter Schwarz, Adenauer: Der Aufstieg, 1876–1952 (Stuttgart, 1986); Der Staatsmann, 1952–1967 (Stuttgart, 1991); Norbert Wiggershaus, “Bedrohungsvorstellungen Bundeskanzler Adenauers nach Ausbruch des Korea-Krieges,” Militärgeschichtliche Mitteilungen 15 (1979): 79–122. 44. Wagner and Uhl, BND, 81. 45. Ibid., 201, quoting a 1975 source. Similar warnings can be found in any BND report from the 1960s and 1970s. See, e.g., the Bundesarchiv sources cited in n. 31 above. 46. Wagner and Uhl, BND, 86; Vojtech Mastny: “Planning for the Unplannable,” www.php.isn.ethz.ch/collections/coll_warplan/introduction_mastny.cfm?navinfo=1 5365&print=1&show=0. 47. That the BND did not believe Soviet planning was realistic is shown by Bundesarchiv, B 206/130, Militärischer Lagebericht Ost, Jahresabschlußbericht 1968, B V, “Die Grundzüge der operativen Planung des WP,” 36–37. 48. Wagner and Uhl, BND, 129, misread Bundesarchiv, B206/128, Militärischer Lagebericht Ost, Jahresabschlussbericht 1967 vom 30.12.1967, which analyzes Soviet capabilities for an offensive in Western Europe and the Soviet ability to reach the Atlantic coast of France, Spain, and Portugal in twenty days. The BND did not assert (as Uhl and Wagner claim) that the Soviet Army could do this, but merely reported on Soviet planning without comment. 49. Wagner and Uhl, BND, 68–69. 50. Ibid., 93 51. Bundesarchiv, B 206/107–116. 52. Wagner and Uhl, BND, 94–97. 53. Ibid., 98. 54. See Roger Engelmann, “Ernst Wollweber (1898–1967). Chefsaboteur der Sowjets und Zuchtmeister der Stasi,” in Dieter Krüger and Armin Wagner, Konspiration als Beruf: Deutsche Geheimdienstchefs im Kalten Krieg (Berlin, 2003), 179–206. 55. Paul Maddrell, Spying on Science: Western Intelligence in Divided Germany, 1945–1961, (Oxford, 2006); Maddrell, “The Western Secret Services, the East German Ministry of State Security and the Building of the Berlin Wall,” Intelligence and National Security 21, no. 5 (2006): 829–847. 56. Wagner and Uhl, BND, 147–148. 57. Ibid., 188. 58. I checked 1980s FRG Prague military attaché files, but these remain classified, so further details cannot be given here.



The West German Secret Services During the Cold War

229

59. See, e.g., Erich Schmidt-Eenboom, “The Bundesnachrichtendienst, the Bundeswehr and SIGINT in the Cold War and After,” Intelligence and National Security 16 (2001): 129–176, at 129. Höhne and Zolling, Pullach intern, 313–314, cite similar numbers: in 1970, the CIA got 20 percent of its information from secret sources, 25 percent from open sources, 25 percent from routine reports from foreign and defense ministries, and 30 percent from military attachés’ reports. 60. Schmidt-Eenboom, “Bundesnachrichtendienst,” 129. 61. See Krüger, Gehlen, 229. 62. Wagner and Uhl, BND, 141. 63. Ibid., 188–189. 64. Ibid., 140, quoting Stasi Hauptabteilung I, 2 February 1984. 65. Bundesarchiv Koblenz, B 206, BA, B 206/117–177, Militärische Lageberichte Ost, Jahresberichte 1960–1991. These reports are extremely detailed and seem to be quite well informed and accurate, but they do not identify sources. 66. Höhne and Zolling, Pullach intern, 323–324. 67. Ibid., 329. 68. Otto Lenz, Im Zentrum der Macht: Das Tagebuch von Staatssekretär Lenz, 1951– 1953 (Düsseldorf, 1989), 91–92, quoted by Wagner and Uhl, BND, 52. 69. Hans Ehlert, Die NATO-Option, vol. 3 of Militärgeschichtliches Forschungsamt, ed., Anfänge westdeutscher Sicherheitspolitik: 1945–1956 (Munich, 1993), 38–39. 70. Höhne and Zolling, Pullach intern, 268. 71. Ibid., 324. 72. Ibid., 325. 73. Wagner and Uhl, BND, 50. 74. Ibid., 79–80. 75. Ibid., 89. 76. Ibid., 90–91. 77. Bundesarchiv, B 206/ 933, “Juniaufruhr in Ostberlin und der sowjetisch besetzten Zone Deutschlands,” and B 206/934, “Die Unruhen in der Ostzone: Überblick und Betrachtungen im Spiegel der 50–D-Tätigkeit”—Analyse über die operative Beschaffung während des Volksaufstandes am 17. Juni 1953 in der DDR.” The latter report, from the end of July 1953, discusses the ways informers could send messages from East to West Germany (52ff.), saying that it was not possible to get information about the uprising out earlier than 19 June, and more from 20 June 1953 on. Permission to use radio communication was given only on 19 June (56). 78. Wagner and Uhl, BND, 99. 79. Höhne and Zolling, Pullach intern, 299–300. 80. Ibid., 333–334. 81. Wagner and Uhl, BND, 192–193. 82. Hermann Wentker “Die DDR in den Augen des BND (1985–1990). Ein Interview mit Dr. Hans-Georg Wieck, Vierteljahreshefte für Zeitgeschichte 56, no. 2 (2008): 323–358, at 339. 83. Ibid., 347–349.

230

The West German Secret Services During the Cold War

84. Ibid., 353. 85. www.hans-georg-wieck.com/homepage.htm. 86. Wagner and Uhl, BND, 193. 87. Müller and Mueller, Gegen Freund und Feind, give many striking examples of very strong feelings against the SPD among the CSU old boy network (Seilschaften) in the BND; see passim, esp., e.g., 390. 88. Höhne and Zolling, Pullach intern, 135; Wagner and Uhl, BND, 51.

Index

Abetz, Otto, 96 Achtung–Panzer! (Guderian), 109 Adenauer, Konrad, 171–72, 199n43, 213–17, 221, 224, 227n35 Afflerbach, Holger, 7, 10, 209–30 Agabekov, Georgii, 23 Agayants, Alexander, 30 Aldrich, Richard J., 9, 10, 149–69 Alexander, Martin, 84 Algeria: French torture, 160; wars (1954–62), 132 Allende, Salvador, 7 Allied intelligence, 158–59; French intelligence cooperating with, 144– 45; May’s critique of, 85–87; TOTEM, 142, 144–45; “Ultra” and “Magic,” 101; West Berlin, 9. See also Britain; France; United States Ampersand Ltd, 163 “Analysis, War and Decision: Why Intelligence Failures Are Inevitable” (Betts), 6–7 Andrew, Christopher, 2, 22, 33n29 Andropov, Yuri, 186–87 Anfilov, Viktor, 46 Antonov, A. I., 65

Apel, Hans, 211 Arab-Israeli wars, 98, 153–54, 223 Arcos trade delegation, Soviet, 19, 21 Argentina, Falklands invasion, 154 arms control, 155, 178–79 Artuzov (né Fraucci), Artur, 25, 29 Atlantic Alliance, 129, 132 atomic bomb: Soviet atomic intelligence, 36, 37, 55–68, 70, 73n24, 77nn97,101, 162. See also nuclear research/ weapons Auriol, Vincent, 133 Austria: French intelligence on, 107; Kreisky, 175; Soviet espionage, 28–29 Bahr, Egon, 201n52, 215; Kohl negotiations, 183, 187, 203n77; Ostpolitik, 172–76, 182, 183, 187, 189– 90, 200n46, 207n117; Tutzing speech (15 July 1963), 176, 200n46 Baldwin, Stanley, 19 Bange, Oliver, 7, 10, 170–208 Barbarossa. See Operation Barbarossa Baril, Paul, 86, 87–88, 116n6 Baring, Arnulf, 216 Barkovskii, V. B., 77n96 231

232

Index

Barrass, Gordon, 6, 8 Bator, Francis, 172 Baudouin, Paul, 114 Beevor, Antony, 63, 79n126 Belgium: British forces, 85, 86; French forces, 82–86, 105, 114, 121n66; Gembloux Gap, 85, 110; German forces, 91; intelligence cooperation limits, 83; neutrality, 105; NORTHAG, 152–53; Soviet espionage, 20; van Overstraeten, 83, 117n12 Ben Barka scandal, 137–38 Beneš, Edvard, 107 Beria, L. P., 13, 22, 42; atomic intelligence/nuclear research, 58–59, 60, 63–64, 70, 79n126; “information filter,” 42, 73n35; London office closure, 29, 30 Berlin: blockade (1948), 222; crisis (1961), 130, 131–32, 145; Wall, 138, 176, 182, 218, 219, 223, 224. See also West Berlin Berlings, Orestes (“Litseist”), 45 Bertrand, Gustave, 81, 88, 98 Berzin, Jan, 22, 28 Betts, Richard, 6–7, 101–2 Bidault, Georges, 100 Blech, Klaus, 189, 205n97 blitzkrieg, Wehrmacht, 83, 98 Bloch, Marc, 98 Blum, Léon, 106 Blunt, Antony, 26, 30 BND, West German, 7, 157, 210–30 Bock, Siegfried, 171, 195n3 Bohr, Niels, 77n97 Bokii, Gleb, 16–17, 21–22 Bolsheviks, 12–25, 29, 39, 50 bomb. See nuclear research/weapons Boothroyd, Betty, 163 Borm, William, 181, 199n39 Braithwaite, Roderick, 173 Brandt, Willy: complaints about BND, 223, 224; mayor of West Berlin, 172, 223; Ostpolitik, 172–78, 182–92,

200nn46,47; secretary Guillaume, 211; Stoph meetings/”Confrontation” and “Confrontation II” (Erfurt and Kassel), 178, 187, 189; Tutzing speeches (1963), 176, 182, 200n46; youth, 201n53 Brenner, Otto, 184, 201n53, 202n66 Brezhnev, Leonid, 72n19, 165n12; GDR and, 178, 187, 192, 197n21, 198–99n36 Briand, Aristide, 93, 104–5 Britain: Baldwin, 19; British Army of the Rhine (BAOR), 153; Callaghan government, 175; Churchill, 39, 49, 55–56, 64, 71n1, 156; Communist Party, 18, 26; and CSCE, 189; EastWest détente, 173, 175; and France, 82–86, 104, 106–7, 114, 121n57, 129–32; and Germany (1940s), 38–39, 43–48, 50, 85, 116n7, 118–19n29; Heath government, 175; Labour Party, 130, 162, 173; NORTHAG, 152–53, 158; nuclear research/weapons, 36, 55–56, 58, 62–65, 67, 69, 80n142; population after World War I, 103; RAF, 156; red scare, 18, 163; Soviet espionage, 20, 26–31, 47–48, 162, 163; West German military espionage, 221; Wilson government, 173, 175; World War I, 23, 100, 103. See also British intelligence British intelligence, 149–69; anthropology, 157–58; archivebased history of, 160; British Commander-in-Chief ’s Mission to the Soviet Forces in Germany (BRIXMIS), 157–58; Cambridge Five, 25–30, 56, 59, 77n96; Cold War, 149–69; counterintelligence, 151, 161–63; cryptography, 5, 15, 16, 18–24, 95, 121n57; Defence Intelligence Staff (DIS), 159; global reach, 159–61; Government Code and Cypher School (GC&CS), 5, 19–21, 23; Government Communications Headquarters (GCHQ), 15, 151–52, 154, 159;



imint, 220; Information Research Department (IRD), 163; Joint Intelligence Committee (JIC), 153, 154, 165n6; Maud Committee/ Maud Report, 55–56, 58, 62, 66; MI3, 116n7; MI5, 1–2, 30, 161, 163; military, 159; Room 40, 23, 90; Scientific and Technical Intelligence Bureau, 157; security, 161–63; seismology, 155; sigint, 151, 155–61, 220; Special Air Service, 158; Special Branch, 161–62; spending, 10, 149, 150–52, 159, 161, 164; spyf light activity, 156; stability, 154–56; vetting, 163; warning, 152–54; World War II, 5. See also Allied intelligence; MI6 Brook, Norman, 152 Brüning, Heinrich, 103 Brussels Pact, 129, 132 Brzezinski, Zbigniew, 172, 189 Bullitt, William, 109, 113 Bundy, McGeorge, 56, 62 Burgess, Guy, 26, 94, 120n51 Bush, Vannevar, 56 Cairncross, John, 26, 56, 59, 66–67, 76n96 Cambridge Five, 25–30, 56, 59, 77n96 Carr, E. H., 33n29 CDU (Christian Democratic Union), Germany, 172, 176, 181, 185, 213 Ceauşescu, Nicolae, 178 Central and Eastern Europe: fall of communism (1989), 3, 170; French intelligence, 130–44, 177; Hungary crisis (1956), 134; West German secret services, 217–21. See also Warsaw Pact; individual countries Cheka (Soviet secret police): foreign department–(INO), 15–16, 17; military intelligence, 27–28; Special Section (Osobii Otdel), 14; Spetsotdel/SPEKO, 16–18, 22, 24, 29 Chicherin, Georgy, 16, 18, 20

Index

233

Chile, Allende overthrow, 7 China: Britain, 19; People’s Republic, 69; Soviets and, 171 Church, Frank, 7 Churchill, Winston, 39, 49, 55–56, 64, 71n1, 156 CIA (Central Intelligence Agency), U.S., 1, 154–55, 163, 213, 223, 229n59 Clausewitz, Claus von, 101 Clemenceau, Georges, 104, 113–14 Codeword Barbarossa (Whaley), 39–40 Colby, William, 165n12 Cold War, 1, 9, 129; British intelligence, 149–69; end of, 154; West German secret services, 209–30 colonies: British and, 159–60, 161; French and, 96, 145, 160; Weimar regime, 103 Colson, Louis, 86 COMECON, 133, 134, 138, 140 Cominform (Communist Information Bureau), 131 comint (communication intelligence), 8. See also sigint Comintern (Communist International): created, 12; cryptography, 18–19, 21; Dimitrov, 49; dissolved (1947), 131; executive (IKKI), 27; France, 93; internationalism, 12–13, 25; sigint and humint, 15; spying directive (1924), 27; Zinoviev, 18–19 communications, French military, 110–11 communism: Bolsheviks, 12–14; British red scare, 18, 163; fall of (1989), 3, 170; German failures (1920s), 13; “national Communism,” 178; Nixon and, 173– 74; West Germany and, 172–78, 211, 212, 215, 216, 224; world revolution, 12, 107, 215. See also Communist parties Communist parties: British, 18, 26; Czechoslovakia, 177, 186; East-West détente and, 172–208; France (PCF), 27, 89, 93, 108, 111, 120n48, 135; Germany, 170, 172–78, 182–87, 201n53, 202n59; Italy, 142–43; Poland, 130; Russian

234

Index

Communist Party, 14; Soviet Union (CPSU), 39, 175, 191; Spanish, 21; worker correspondents, 27. See also Cominform; Comintern; communism Cooper, J. E. S. (“Josh”), 20–21 Corap, André, 111–12 Cot, Pierre, 108 counterinsurgency, 101, 160 counterintelligence: British, 151, 161–63; France, 81, 89–95, 115; German, 45; informants from West to East Germany, 181–84, 187, 199n39, 200–201nn47,48,51,52, 204n79, 206n101, 211; Soviet, 24–27. See also disinformation counterrevolution: GDR and, 170, 182, 185–86; Soviets and, 8, 13, 16–17, 20, 25 Crabb, Buster, 156 Crampton, John, 156 Crimean meetings (1972), 191 Crowe, Eyre, 6 cryptography, 8; British, 5, 15, 16, 18–24, 95, 121n57; France, 88, 90–92, 95, 131; Polish, 22–23, 98; Soviet, 15–24, 29, 31. See also Enigma machine; sigint CSCE (Conference on Security and Co-operation in Europe), Helsinki (1975), 179, 189–93; Stasi and, 170–71, 190–93, 195n3, 206n109. See also Final Act CSU (Christian Social Union of Bavaria), 172, 185, 230n87 Cuban Missile Crisis (1962), 141, 145, 223 Curzon, Lord, 18, 95 Cyprus, UKUSA alliance, 160–61 Czechoslovakia, 177; Communist Party, 177, 186; Dubček, 177; East-West détente, 190; French alliance sought by, 107; Prague Spring (1968), 144, 145, 173, 185–86; Soviet invasion possibility (1968), 153; Warsaw Pact invasion (1968), 186, 220 Daladier, Édouard, 86, 96, 105–14,

118n24; defensive strategy, 105, 106, 117n10; military deficiencies, 99, 106, 108–9, 111; newspaper corruption, 100; Pfeiffer as chef de cabinet, 94; political dysfunction, 112–13, 114, 118n24; Reynaud and, 112–14, 118n24, 121–22n68 Darlan, François, 86–87 Dautry, Raoul, 111 DDR. See East Germany decryption. See cryptography de Gaulle, Charles, 10, 89, 109, 115, 129, 171–73 Dekanozov, Vladimir, 30 delimitation policy, GDR, 178, 185, 187, 193, 199n36 de Margerie, Pierre, 97 de Margerie, Roland, 112 Denikin, Anton, 14 Denmark, Krag, 175 Denniston, Alastair, 23 de Rochegonde, Major, 143 de Sainte-Suzanne, Raymond, 112 détente: East-West, 170–208; economic, 172, 190; military, 192 Deutsch, Arnold/Stefan Lang, 25–27 de Villelume, Paul, 112, 114 Dewavrin, André, 115 Didelet, Henri, 97 Diego Garcia island, Indian Ocean, 161 Diehl, Günter, 172 Dien Bien Phu disaster, 136 Dilks, David, 2 Dimitrov, Georgii, 49, 124n99 diplomatic history, 3, 5. See also international relations disinformation: atomic intelligence, 59, 62–63; British, 48; Nazi German, 40, 41, 44–48; West German, 183, 199n39. See also counterintelligence Doumenc, Aimé, 86, 107 Dreyfus affair, 89, 90 Dubček, Alexander, 177 Duclos, Jacques, 93



Dufieux, Julien, 109 Dulles, Allen, 154–55 Dzerzhinsky, Felix, 17, 18, 20 Easter, David, 160 Eastern Europe. See Central and Eastern Europe East Germany (GDR), 170–208, 219; British Commander-in-Chief’s Mission to the Soviet Forces in Germany (BRIXMIS), 157–58; British intelligence in, 157; delimitation policy, 178, 185, 187, 193, 199n36; downfall, 170, 223–24; East-West détente, 170–208; economics, 193, 208n122, 222, 223–24; French intelligence and, 132, 135–36; NVA/National People’s Army/ Nationale Volksarmee, 217–18, 220; rearmament/Kasernierte Volkspolizei/KVP, 216, 218, 222; SED, 171, 175–90, 194, 201n58; Soviet Union and, 132, 157–58, 170–71, 177–78, 186–88, 197n21, 198–99n36, 218–19, 222, 224; Stoph-Brandt meetings/”Confrontation” and “Confrontation II” (Erfurt and Kassel), 178, 187, 189; trade unions, 186, 202n66; uprising (17 June 1953), 222–23, 229n77; West German secret services and, 209, 216, 218–20, 222–24; Westpolitik/”Western strategy,” 176, 177, 187. See also Honecker, Erich; Stasi; Ulbricht, Walter East-West détente, Germany, 170–208; Ostpolitik, 173–94, 200n47, 205n98, 206n107, 207n117. See also reunification Ebert, Friedrich, 171, 181 economics: BND espionage, 211–12; economic détente, 172, 190; GDR, 193, 208n122, 222, 223–24; Soviet New Economic Policy, 16. See also spending

Index

235

EEC, 175 Egypt, Israel attacked by (October 1973), 98, 153–54 Eisenhower, Dwight D., 152 electronic spying. See sigint Ellrodt, Klaus, 182 Enigma machine: French, 131; Nazi Germany, 22–23, 84, 86, 88, 97–98, 131 L’étrange défaite (Strange Defeat), Bloch, 98 Europe. See Central and Eastern Europe; individual countries and international organizations Falklands, Argentine invasion, 154 FDP, Germany, 172–73, 181, 186, 199nn39,43, 201n52, 206n101 Felfe, Hans, 214 Fermi, Enrico, 57, 58, 77n97 Fetterlein, Ernst (“Fetty”), 20 FHO (Fremde Heere Ost), 82, 212, 215, 218 Final Act, CSCE (Helsinki, 1975), 170, 191, 193, 207n117; Basket III, 173, 179, 205n97 finances. See economics; spending Finland: Soviet Winter War and armistice (1940), 38–39, 50, 113. See also Helsinki Fitin, Pavel M., 30, 42 Florin, Peter, 186 Foch, Ferdinand, 111 Foertsch, Hermann, 213 Forcade, Olivier, 84, 87–88 foreign intelligence: BND, 7, 157, 210–30; France (SR), 81, 84, 87, 88, 91–98, 102, 104, 114, 115, 116n6, 117n13; Soviet, 1, 15–30, 37, 42, 44, 45, 56–57, 71n4; spending on, 161 foreign language competence, 3, 6, 17 France: British alliance, 82–86, 104, 106– 7; Communist Party (PCF), 27, 89, 93, 108, 111, 120n48, 135; de Gaulle, 10, 89, 109, 115, 129, 171–73; dysfunctional institutions, 86–87, 99–100, 112–14; Fifth Republic, 89, 115, 140–41, 142;

236

Index

Fourth Republic, 139, 146n17; FrancoSoviet non-aggression pact (1935), 107; Germany defeating (1940), 38, 39, 50, 52, 85; Italian alliance, 105, 106–7; Khrushchev summit, 155; Operations (3ème Bureau), 84, 112; Paris peace conference (1919), 104; Pompidou, 173; Popular Front, 13, 86–87, 99–100, 106; population after World War I, 103; Resistance, 100, 108, 120n48; Soviet espionage, 20, 93; Soviet kidnappings, 13, 20; Third Republic, 81, 89, 99–100, 110–11, 115; Vichy, 89, 96, 100, 115; World War I, 88, 90–92, 94, 96, 101, 103, 117n10. See also French intelligence; French military François-Poncet, André, 92, 97 Frankenstein, Robert, 106 Fremde Heere Ost (FHO), 82, 212, 215, 218 French intelligence, 81–148; agencies, 86–87, 90, 95–96, 142–43; Bulletin mensuel: Pays du Bloc soviétique et Yougoslavie, 141; Bulletins particuliers de renseignement, 142; Bureau central de renseignement et d’action (BCRA), 115; CER, 134, 139–41, 143–44; and colonies, 96, 145, 160; cooperation between agencies, 142–45; DGER (Direction générale des études et recherches), 129, 130; about East (1945–68), 128–48; EMGDN, 129, 131, 135, 139; foreign (SR), 81, 84, 87, 88, 91–98, 102, 104, 114, 115, 116n6, 117n13; military attachés, 90, 95–97, 128, 135–44; National Defense Intelligence Plan, 129–30; Nazi Germany/World War II, 10, 81–132; Northeast Front GHQ, 84, 97, 116n6, 118n24; notes de renseignement, 139–40, 142; overambitious, 128–32; postwar, 9, 10–11, 128–48; SDECE, 115, 128–34, 139–45; SGDN, 128, 129, 131, 133, 134, 137–40, 142–43, 145; sigint, 81, 84, 90–91, 92, 95, 131–32; on Soviets,

129–45; and Warsaw Pact/Central and Eastern Europe, 130–44, 171. See also Allied intelligence; French military intelligence; French police French military, 82–83; Armée de l’Air, 83, 87–88, 97, 108–9, 124n101, 125n105; civilian distrust of, 89, 101; communications, 110–11; “continuous front,” 98–99; DCRs (reserve armored divisions), 110; defensive strategy, 105, 106, 117n10, 121n66; dysfunctional, 86–113, 125n105; Maginot Line, 105, 112, 121n66; offensive strategy, 91, 121n66; spending, 106, 121–22n68 French military attachés, 90, 95–97, 128, 135–44 French military intelligence: 2ème Bureau/5ème Bureau, 81–96, 102, 106, 107, 115, 116n6, 129, 131, 134, 135; EMFA, 135–37 French police: DST, 115, 135, 138, 143; Sûreté générale, 90, 93–94 FRG. See West Germany (Federal Republic/FRG) Frieser, Karl-Heinz, 83 Frisch, Otto, 55 Fuchs, Klaus, 56–57, 66–67, 76n95 Gamelin, Maurice, 82–88, 96–101, 106–7; Breda plan, 99, 112, 125–26n113; defensive strategy, 117n10; and Georges, 86, 112, 118n24; military deficiencies, 111, 112; and Reynaud, 114, 118n24 Garton Ash, Timothy, 154 Gauché, Maurice, 81, 82, 84, 87–88, 96, 116n7 GC&CS, Britain, 5, 19–21, 23 GDR. See East Germany Gehlen, Reinhard/Gehlen Organization, 7, 211–17, 220–25, 226n17 Geneva Disarmament Conference (1932–33), 106



Genscher, Hans-Dietrich, 172, 191, 207n117 Georges, Joseph, 84, 86, 112, 116n6, 118n24 Géraud, André (“Pertinax”), 108 Germany: Bolsheviks and, 12–14; communist failures (1920s), 13; disarmament (after World War I), 104, 106; disarmament (after World War II), 104, 213; East-West détente, 170–208; FDP, 172–73, 181, 186, 199nn39, 43, 201n52, 206n101; FHO, 82, 212, 215, 218; intra-German Basic Treaty, 179, 189, 190, 191; NORTHAG, 152–53, 158; population after World War I, 103; reunification, 171, 175, 178, 190–94, 201n52, 223–24; Weimar, 14, 103–4, 175; World War I, 92; world war intelligence defeats, 210. See also Berlin; East Germany; Nazi Germany; SPD; West Germany Giraud, Henri, 89, 115 Globke, Hans, 215–16 gold standard, 105 Golikov, F. I., 42, 45, 46, 48, 49, 50 Gomułka, W., 178 Gorbachev, Mikhail, 162, 194, 207n121 Gordievsky, Oleg, 22, 162 Gordon, Boris, 30 Göring, Hermann, 49, 85, 95, 97 Gorodetsky, Gabriel, 41, 49, 75n69 Gor’skii, A. V., 56 Gorsky, Anatoly, 30–31 Gort, Lord, 118–19n29 Gouzenko, Igor, 37 Gowing, Margaret, 62 Grand Delusion (Gorodetsky), 49, 75n69 Gromyko, Andrei, 64, 178 Guderian, Heinz, 86, 109 GUGB, 13, 22, 71n4. See also NKVD Guillaume, Augustin, 138 Guillaume, Günther, 211 Hager, Kurt, 182 Halder, Franz, 83, 212

Index

237

Hall, Theodore, 57 Hallstein doctrine, 186 Hampshire, Stuart, 151–52 Handel, Michael, 102 Hankey, Maurice, 56, 100 Harnack, Arvid (“Korsikanets”), 44 Harriman, Averell, 65–66 Hasegawa, Tsuyoshi, 63 Haslam, Jonathan, 1–35 Heinemann, Gustav, 215, 222 Heisenberg, Werner, 56 Helfand, Leon, 23 Helsinki: Socialist International (1971), 190. See also CSCE Hennessy, Peter, 162–63 Herman, Michael, 24–25, 31, 154, 160 Heusinger, Adolf, 213 Hinsley, Harry, 2, 5 Hiroshima, atomic bomb, 38, 56–70 history. See diplomatic history; intelligence history; military history Hitler, Adolf: French intelligence on, 10, 83, 85, 106; German intelligence and, 83–84, 212; Mein Kampf, 48, 50; Soviets and, 37, 39–55, 66, 67, 73n24. See also Nazi Germany Hoffmann, Stanley, 99 Höhne, Heinz, 224 Holloway, David, 7, 9–10, 36–80 Honecker, Erich, 178, 187, 198n27, 200–201n51; CSCE, 179, 192–93; Quadripartite Agreement on (West) Berlin, 178, 204n88; shift from Ulbricht to, 171, 175, 192–93; Sixth Central Committee plenum, 206n106; Soviet Union and, 171, 178, 179, 192, 198–99n36 humint (human intelligence), 8; BND, 218, 219, 221–22, 224; French, 90, 93, 131; Soviet, 12–35 Hungary crisis (1956), 134 Huntziger, Charles, 112 imint (imagery intelligence), 220

238

Index

Indian Ocean, Diego Garcia island, 161 individual liberties, East-West détente and, 170, 176, 193 Indochina: French torture, 160; wars (1946–54), 133 Indonesia, “Confrontasi,” 160 industrial espionage: Soviets in France, 93; Stasi, 181, 193, 207n118 intelligence, 8; atomic, 36, 37, 55–68, 70, 73n24, 77nn97,101, 162; failures, 101–2; field, 8; imint, 220; importance, 68–71; national “culture” of, 89; political, 8, 92, 132, 163, 221–24; security, 162. See also comint; counterintelligence; foreign intelligence; humint; military intelligence; sigint; strategic intelligence; individual countries intelligence history, 1–2, 4; access to reliable sources, 8–9; archival sources, 3, 160; collection/analysis/ policy, 5–8, 42–48, 67–70, 89–90; consequences of intelligence, 9, 68–71; and diplomatic and military history, 3; foreign language competence and, 3, 6; political importance, 210; problems of, 2–4, 150; spending on, 164 internationalism, Bolshevik, 12–13, 25 international relations: consequences of intelligence for, 68–71; history, 3; psychology applied to, 7; value of secret intelligence to, 5–8 International Spy Museum, Washington, DC, 210 Ioffe, Abram, 58 Iraq, Kuwait invasion by (1991), 154 Ireland, Soviet target, 20 Israel: Arab-Israeli wars, 98, 153–54, 223; BND connections, 223 Isserson, G. S., 54, 76n88 Italy: Communist Party, 142–43; French alliance, 106–7; Mussolini, 107; Vatican, 130

Jackson, Peter, 85 Japan: cryptography, 24; Pearl Harbor attack, 39–40, 86, 98; Soviet attack on, 63–66; U.S. atomic bomb vs., 38, 56–70 Jervis, Robert, 6, 7, 102 “Jewish Bolshevism,” 39, 50 Joffre, Joseph, 88, 91–92 John, Otto, 211 Johnson, Lyndon (LBJ), 172, 189 Jones, Clive, 160 journalists: East-West working conditions, 187–88; intelligence analysts and, 222; Stasi contacts, 181 Kádár, János, 178 Kaftanov, S. V., 58 Kahn, David, 2 Kanzlerdemokratie, 216 Kegel, Gerhard, 73n38 Kennan, George, 6 Kennedy, John F., 182 KGB, 1, 93; Cambridge Five, 77n96; Hitler’s surprise attack, 73n24; London, 162; Paris, 93, 108; West Germany, 186–87, 204n80 Khariton, Iu. B., 60 Khrushchev, Nikita, 74n53, 139, 155; French intelligence on, 132–33; “Secret Speech,” 39, 40, 72n19, 140; U.S. visit, 132–33, 154–55 kidnappings, Soviet, 13, 20, 93 Kiesinger, Kurt Georg, 172, 189 Kiessling, Günter, 211 Kikoin, I. K., 60 Kim Il Sung, 69 King, John, 24 Kirov, Sergei, 29 Kissinger, Henry, 173 Knowing One’s Enemies: Intelligence Assessment Before the Two World Wars (May), 1, 2, 8 Kobulov, Anayak Z., 30, 45 Koeltz, Louis, 96



Kohl, Michael, 183, 187, 203n77, 204n79, 208n122 Korean War, 69, 216 Krag, Jens-Otto, 175 Kreisky, Bruno, 175 Krivitsky, Walter, 24, 27–28, 29, 30 Kurchatov, Igor, 55, 57–62, 68, 70 Kutepov, Alexander, 20 Kuwait, Iraqi invasion (1991), 154 La Chambre, Guy, 108–9 Lacoste, Pierre, 89 language competence, foreign, 3, 6, 17 Latvians, 22, 29, 45 Laval, Pierre, 105, 106–7 le Bailly, Louis, 153 le Carré, John, 4 Lee, Bradford, 99 Léger, Alexis, 95, 113, 114 LeMay, Curtis, 213 Lenin, 12–16 Leopold III, 117n12 Liss, Ulrich, 81, 83–84 Litvinov, Maxim, 12, 18 Los Alamos laboratory, U.S., 56–57 Lota, Vladimir, 41 Louis XIII, 89 Louis XV, 89 Machiavelli, 11 Maclean, Donald, 26, 77n96 Maddrell, Paul, 219 Maginot Line, 105, 112, 121n66 Malaya, British intelligence and, 160 Malvy, Louis, 92, 120n43 Mandelson, Peter, 163 Manhattan Project, U.S., 36, 56–62, 67 Marshall, Andrew, 8 Marxism: Bolsheviks and, 14. See also communism Mason, Roy, 159 Massigli, René, 131 Maud Committee/Maud Report, Britain, 55–56, 58, 62, 66

Index

239

Mauritius, and Diego Garcia island, 161 May, Ernest R., 4, 85–87, 89, 164; Knowing One’s Enemies: Intelligence Assessment Before the Two World Wars, 1, 2, 8 MBFR (Mutual and Balanced Force Reductions), 179, 191 McCarthyism, 163 Mechelen incident, 83 Mein Kampf (Hitler), 48, 50 Menzhinsky, Vyacheslav, 18, 20 Meretskov, K. A., 48, 52 Merkulov, V. N., 42, 44–45, 63–64 Messimy, Adolphe, 91 MI3, 116n7 MI5, 1–2, 30, 161, 163 MI6, 1–2; Barrass, 6; and BND, 214; budget, 161; French SDECE and, 132; Reilly, 25; Sinclair, 14; on Soviets, 6, 14, 19, 24, 25, 131; Vivian, 24; Zinoviev Letter, 6, 19 Middle East: BND connections, 223; Soviet penetration, 133; UKUSA alliance, 160–61. See also individual countries Mielke, Erich, 176, 185–91, 203n72, 204n84, 206n106 military: British Army of the Rhine (BAOR), 153; détente, 192; GDR, 217–18, 220; Group of Soviet Forces in Germany (GSFG), 218–19; MBFR, 179, 191; NATO, 152–53, 158. See also French military; military...; Red Army; Wehrmacht military attachés: French, 90, 95–97, 128, 135–44; Western spies in Eastern Europe, 137–39, 219, 229n59 military doctrine: French, 85, 98; Soviet, 53–54 military history, 3 military intelligence, 8; BND, 211–12, 215, 220–21, 224; British, 159; FHO, 82, 212, 215, 218; NATO, 145; Soviet, 18–19, 22, 27–29, 36–55, 67–69, 71n4, 73n38, 80n142. See also French military intelligence

240

Index

Miller, Evgenii, 13 The Missing Dimension: Governments and Intelligence Communities in the Twentieth Century (Andrew and Dilks), 2 Mitrokhin, Vassili, 162 Molotov, Vyacheslav, 17, 46–47, 59, 64, 74n54, 124n99 Moravec, František, 107 Munich conference (1938), 85, 97, 107, 111–12, 172 Mussolini, B., 107 Nagasaki, atomic bomb, 38, 56, 62 “national technical means,” arms control talks (1960s), 155 NATO, 144–45, 175; NORTHAG, 152–53, 158; Stasi espionage, 180–81, 186, 190, 192; summit (3 December 1969), 190; warning, 152–54; Wörner, 211 Navarre, Henri, 81, 116n6 Nazi Germany, 10, 30, 37–39; Enigma machine, 22–23, 84, 86, 88, 97–98, 131; French intelligence failure with, 10, 81–127; French newspaper propaganda, 100; harsh treatment of Russians, 212; ideological war, 50, 67; Munich conference (1938), 85, 97, 107, 111–12, 172; Munich Pact, 172; nuclear research/uranium/nuclear weapons, 56, 57, 59, 63, 77n101, 172; rearmament, 105–6; Soviet espionage, 30, 37, 42–50, 73n38; Stalin and, 5, 9–10, 13, 31, 36, 38–55, 67–70, 72n19, 107; West German secret services and, 209. See also Hitler, Adolf; Operation Barbarossa; Wehrmacht Nazi-Soviet Pact, 38–39 Nekrich, A. M., 72n19 Netherlands: army, 125–26n113; intelligence from Germany, 83; NORTHAG, 152–53 Neustadt, Richard, 164 Nicoll, Douglas, 154

1941 god (The Year 1941), 40–41 Nixon, Richard, 173–74, 189 NKVD, 25; analyses of incoming intelligence, 31, 42; Beria, 13, 22, 29, 30, 42; cryptography, 21, 22, 24; Fitin, 30, 42; GUGB, 13, 22, 71n4; Krivitsky, 24, 27–28, 29, 30; NKGB, 37, 42, 44, 45, 56–57, 71n4, 77n96; OGPU absorbed into, 20, 22, 93; Sudoplatov, 13 Norden, Albert, 171, 182, 186, 198n27 NORTHAG (Northern Army Group), NATO, 152–53, 158 NSA (National Security Agency), U.S., 4, 151–52, 161, 163 nuclear research/weapons: arms control, 155, 178–79; Britain, 36, 55–56, 58, 62–65, 67, 69, 80n142, 221; Germany and, 56, 57, 59, 63, 77n101, 172, 187; vs. Japan, 38, 56–70; plutonium bomb, 57–58, 61–64, 66; Soviet atomic intelligence and, 36, 37, 55–70, 73n24, 77nn97,101, 79n126, 162; U.S., 10, 36, 38, 56–70, 80n142, 152, 221 Nuremberg laws (1935), 215 Oder-Neisse line, 186 Odom, Bill, 4, 6 Operation Barbarossa (German invasion of Soviet Union in 1941): Soviet intelligence and, 5, 9–10, 24, 31, 36–55, 66–74; surprise, 40, 53–55, 71n1, 73n24, 86 Oppenheimer, Robert, 77n97 Orlov, Alexander (Leiba Fel’dbin), 25, 26, 29, 30 Oster, Hans, 83 Ostpolitik, West Germany, 173–94, 200nn46,47, 205n98, 206n107, 207n117 Paillole, Paul, 81, 87, 116n2 Paulssen, Constantin, 183 Pearl Harbor attack, 39–40, 86, 98 Peierls, Rudolf, 55, 56 Pétain, Philippe, 87



Pfeiffer, Édouard, 94, 120n51 Philby, Kim, 26 Piłsudski, Józef, 14 Pliatzky, Leo, 150 plutonium bomb, 57–58, 61–64, 66 Podewin, Norbert, 171, 195n3 Poincaré, Raymond, 94 Poland: Communist Party, 130; cryptography, 22–23, 98; East-West détente, 188, 190; French intelligence and, 130, 132, 137, 138, 140, 142; Germany’s victory over, 52, 54, 113; Red Army, 13–15, 53; revolts (1970), 223 police. See Cheka; French police political intelligence, 8, 92, 132, 163, 221–24 politics: damaging impact of, 9, 89–91, 102–3, 112–14, 118n24, 163; French intelligence, 89–91, 93, 115, 118n24, 139–42; intelligence history and, 210; notes de renseignement, 139–40. See also communism Pompidou, Georges, 173 Ponomarev, Boris, 142–43 population, Europe after World War I, 103 Porch, Douglas, 84 Portuguese, torture in Africa, 160 Potsdam Conference (1945), 63–66 Pourquoi l’Armée rouge a vaincu (A. Guillaume), 138 Powers, Gary, 156 Prague Spring (1968), 144, 145, 173, 185–86 Prételat, André-Gaston, 111–12 Pudin, Vasilii, 24 Pyatnitsky (né Tarshis), Osip, 19 Quadripartite Agreement on West Berlin (1971), 178–79, 188, 190, 204n88 Radek, Karl, 14 radio communications: French military, 110. See also comint; sigint

Index

241

Raimond, Jean-Bernard, 173 Ramsay MacDonald, James, 10 Razvedupr, Soviet, 18–19, 22, 27–28, 29, 71n4 Reagan, Ronald, 4 Red Army, 22, 37; BND knowledge of, 212, 215, 218–19, 224; Cuban Missile Crisis (1962), 223; Finland Winter War (1940), 38–39, 50; vs. German attack, 36, 38–39, 50–55, 68, 69; offensive, 46, 53–54, 64, 65, 136, 137; organizational restructuring and rearmament (1941), 50–51; Poland, 13–15, 53; “Stalin line,” 53; Trotsky as creator of, 28 Reeve, James, 155 Reilly, Sidney, 25 Renondeau, Georges, 97 reunification, Germany, 171, 175, 178, 190–94, 201n52, 223–24 Reynaud, Paul, 112–14, 118n24, 121–22n68 Ribbentrop, Joachim von, 49, 124n99 Richelieu, Cardinal, 89 Riom trial (1942–43), 100 Rivet, Louis, 81, 88, 95–96 Roberts, Geoffrey, 41 Robertson-Malinin Agreement, 157 Romania: diplomacy with Bonn (1967), 177; Soviet invasion possibility (1969), 153 Rommel, Erwin, 86 Roosevelt, F. D., 56, 64 Rumsfeld, Donald, 9 Rupp, Rainer (“Topaz”), 180 Russia: Bolsheviks, 12–23; Nazi’s harsh treatment of people, 212; provincialism, 13; West German perceptions of, 209–10. See also Soviet Union SALT (Strategic Arms Limitation Talks), 155, 178–79 satellites, sigint, 155, 159, 220 Savinkov, Boris, 25

242

Index

Scheel, Walter, 172, 174, 188, 190 Schlesser, Guy, 93, 94, 109 Schlieffen Plan, 91, 119n39 Schmidt, Hans-Thilo, 83, 97–98 Schmidt, Helmut, 179, 188, 201n59, 222, 224 Schollwer, Wolfgang, 172 Schuker, Stephen A., 7, 10, 81–127 Schumacher, Kurt, 213 Schütz, Klaus, 185 Schwarz, Hans Peter, 217 Schwerin, Graf Gerhard von, 217 scientific and technical intelligence: arms control, 155, 178–79; BND, 211–12; British, 157; France, 137; global “technological revolution,” 193; Stasi, 199n40; “technology gap” between East and West, 207n119. See also nuclear research/weapons secret police: French DST, 115, 135, 138, 143. See also Cheka Sedan, French military, 86, 87, 101, 112 SED (Socialist Unity Party), GDR, 171, 175–90, 194, 201n58 seismology, British, 155 Semyonov, Vladimir, 177 Serebryanskii, Yakov, 20 Shaposhnikov, B. M., 52, 53 Shtemenko, S. M., 64–65 Shulze-Boyson, Harro (“Starshina”), 44, 49 sickle-cut strategy, Wehrmacht, 83 Siegfried Line, 107, 117n13 sigint (signals intelligence), 8; BND, 219–20, 224; British, 151, 155–61, 220; French, 81, 84, 90–91, 92, 95, 131–32; satellites, 155, 159, 220; Soviet, 15, 37, 71n7; UKUSA alliance, 158–61; U.S., 155, 158–61, 220. See also comint; cryptography Sinclair, John, 14 Smyth Report, 62 social democracy: GDR vs., 173, 176–77, 182–87, 190, 192, 197n18, 198n33, 200n48. See also SPD

Socialist International, 186; Helsinki (1971), 190 Sorge, Richard (“Ramsay”), 27–28, 45 Soustelle, Jacques, 115, 129 Soutou, Georges-Henri, 9, 10–11, 128–48 Soviet intelligence, 5–74; atomic, 36, 37, 55–68, 70, 73n24, 77nn97,101, 162; Austria, 28–29; Belgium, 20; Britain, 20, 26–31, 47–48, 162, 163; cryptography, 15–24, 29, 31; foreign, 1, 15–30, 37, 42, 44, 45, 56–57, 71n4; France, 20; and French Communists, 27, 93, 108, 120n48; GPU/OGPU/ INOGPU, 17–30, 93; military, 18–19, 22, 27–29, 36–55, 67–69, 71n4, 73n38, 80n142; Narkomindel (People’s Commissariat of Foreign Affairs), 16–19, 30; Nazi Germany, 30, 37, 42–50, 73n38; OMS, 19, 27; Operation Barbarossa, 5, 9–10, 24, 31, 36–55, 66–74; OTP (one-time pads), 21; Rote Kapelle, 28, 44; shifrogrammy/secret telegrams, 20–21; sigint, 37, 71n7; SVR, 1; “Venona” interceptions, 162; World War II, 36–80; “Yasha’s Group,” 20. See also Cheka; KGB; NKVD Soviet Union: Arcos trade delegation, 19, 21; Bolsheviks, 12–23; British Commander-in-Chief’s Mission to the Soviet Forces in Germany (BRIXMIS), 157–58; British diplomatic break (1927), 19; British intelligence on, 152, 156, 159, 162–63; Communist Party (CPSU), 39, 175, 191; East-West détente, 171, 177–78, 186–88, 190; Finland Winter War and armistice (1940), 38–39, 50, 113; Franco-Soviet nonaggression pact (1935), 107; French intelligence on, 129–45; GDR and, 132, 157–58, 171, 177–78, 186–88, 197n21, 198–99n36, 218–19, 222, 224; German POWs, 157; Gorbachev,



162, 194, 207n121; Japan attacked by, 63–66; kidnappings by, 13, 20, 93; Ministry of Defense’s Nachal’nyi period voiny (The Initial Period of War), 40; MOPR (international workers’ aid), 19; Nazi-Soviet Pact, 38–39; New Economic Policy, 16; nuclear research/weapons, 36, 37, 55–70, 73n24, 79n126, 152; “peaceful coexistence” policy, 178; Profintern (trades union international), 19; Sovnarkom (Council of People’s Commissars), 17; Treaty of Moscow, 178, 193, 207n117; U.S. intelligence on, 162, 213; Western spies, 218–19; West German secret services and, 209, 212–21, 228n48. See also Brezhnev, Leonid; Khrushchev, Nikita; Operation Barbarossa; Red Army; Russia; Soviet intelligence; Stalin, Joseph Spain, civil war, 21 Spangenberg, Dietrich, 182–83, 201n52 SPD (German Social Democratic Party): CSU and, 230n87; GDR vs., 176–77, 181–92, 197nn18,21, 199n38; Maerker, 199n38; Ostbüro, 182, 210; Schumacher, 213; Stasi informants in, 181–84, 187, 199n39, 200– 201nn47,48,51,52, 206n101; Steffen, 185, 201–2n59; think tank, 199n43; Wehner, 183, 184, 187, 199n39, 201n58 Spears, Edward, 114 Speidel, Hans, 213, 217 spending: BND, 216; British intelligence, 149, 150–52, 159, 161, 164; French military, 106, 121–22n68; on intelligence history, 164; nuclear weapons, 152; U.S. intelligence, 149, 151, 161; Wehrmacht, 106, 121n68 spy-flight activity/U-2 spy plane, 155, 156 SR (French foreign intelligence), 81, 84, 87, 88, 91–98, 102, 104, 114, 115, 116n6, 117n13

Index

243

Stalin, Joseph, 13, 74n53; Bokii and, 17; Helfand and, 23; “his own intelligence boss,” 15, 46–50, 68; humint, 12–35; Kennan and, 6; Khrushchev’s “Secret Speech” and, 39, 40, 72n19, 140; Molotov-Ribbentrop pact, 124n99; Nazi-Soviet agreements, 38–39, 107; nuclear research/atomic bomb, 58–66, 67, 79n126; personality, 38; terror/purges, 13, 26, 29–30, 37–39, 46, 66; and Trotsky, 13, 19, 25; unpreparedness (1940–41), 5, 9–10, 31, 36, 38–55, 67–70, 72n19; Volkogonov biography of, 40 Stalin’s Wars (Roberts), 41 Stasi (GDR secret services), 7, 10, 157, 170–208, 219, 220; bloc framework, 180; BND successes, 220; coat of arms (“the shield and the sword”), 208n123; and CSCE, 170–71, 190–93, 195n3, 206n109; domestic framework, 180; German framework, 181; HVA (Hauptverwaltung Aufklärung), 180–81, 185, 188, 192, 199n40, 205– 6nn98,100; industrial espionage, 181, 193, 207n118; international framework, 180–81; Western spies, 219; West German informants to, 181–84, 187, 199n39, 200– 201nn47,48,51,52, 204n79, 206n101, 211; ZAIG (Central Analysis and Information Group), 188–89 Steffen, Jochen, 185, 201–2n59 Stehlin, Paul, 97 Stöbe, Ilse (“Alta”), 43, 44 Stoph, Willi, 182, 183, 187, 200–201n51; Brandt meetings/”Confrontation” and “Confrontation II” (Erfurt and Kassel), 178, 187, 189 Strang, William, 131 strategic intelligence, 8, 67; BND, 211–12, 216; warning, 152–54. See also military doctrine; military intelligence

244

Index

Strauß, Franz Joseph, 189, 220 Straw, Jack, 163 Stresemann, Gustav, 103 Struwe, Günter, 183 Sudetenland, Germany annexing (1938), 103, 107, 172 Sudoplatov, Pavel, 13, 25, 31, 77n97 Suez crisis (1956), 134, 138, 145 Sûreté générale, 90, 93–94 surprise attack: Arab-Israeli war (October 1973), 98, 153–54; Bolshevik, 13; German invasion of France, 10, 86, 98; German invasion of Soviet Union/Operation Barbarossa, 40, 53–55, 71n1, 73n24, 86; Hiroshima bombing, 63; NATO and, 153–54; nuclear attack (1980s), 73n24; Pearl Harbor, 86, 98; warning systems, 152–54, 220–21; West German rearmament and, 216 Suvorov (Rezun), Viktor, 75n69 Szilard, Leo, 77n97 Taittinger, Pierre, 112 technical intelligence. See scientific and technical intelligence “technological revolution,” global, 193 “technology gap,” between East and West, 207n119 telecommunications, French military, 110–11 telegraphy, international, 90 terror, Stalin’s terror/purges, 13, 26, 29–30, 37–39, 46, 66 Thatcher, Margaret, 162 Tickell, Crispin, 173 Tiedge, Hansjoachim, 211 Tiltman, John, 20–21 Timoshenko, S. K., 48, 51, 52, 53–54, 68, 69 Tippelskirch, Kurt von, 86, 118n23 Tirard, Paul, 94–95 Tisserant, Éugène, 130 Tonkin, Derek, 159

torture, colonial, 160 trade unions: FRG, 176, 184–85, 186, 197n18, 201n53; GDR, 186, 202n66; Profintern, 19 Traffic Treaty, 187–88 Transiberian Railway, 138 Treaty of Moscow, 178, 193, 207n117 Trend, Burke, 150, 152 Trepper, Leopold, 28, 30 trinitarian analysis, 101 Trossin, Julius, 29 Trotsky, Leon, 12, 13, 19, 25, 28 Truman, Harry, 6, 64, 66 the Trust, 25 trust: Bolsheviks in cryptography, 15; France and Stalin, 13; humint and, 12–35; Stalin and, 13, 31, 38, 47 Tukhachevskii, Mikhail, 14 Turkey, U.S. bases, 161 Uhl, Matthias, 219, 228n48 Ukraine: Bokii from, 16; Bolsheviks and, 14; British spy-flight activity, 156; Hitler attack plans, 39, 43, 48 Ulbricht, Walter, 175–78, 182–87, 198n27, 200n49, 201n51; BND intelligence of, 222; Podewin and, 171, 195n3; Quadripartite Agreement on (West) Berlin, 178, 204n88; shift to Honecker from, 171, 175, 192–93; and social democracy, 176, 177, 185–87, 197n18; Soviet Union and, 176, 177–78, 193, 197n21; and SPD, 176, 177, 183–87, 197n21, 202n66 unions. See trade unions United States: Air Force, 156; “bridgebuilding” strategy, 172, 173–74, 181, 189; British alliance, 151–52, 158–61; British-based missions, 156; CIA, 1, 154–55, 163, 213, 223, 229n59; Cold War intelligence, 150; East-West détente, 172–75, 178, 181, 182, 188, 189, 192; and France, 104, 106, 107, 108–9, 113, 129–32; imint, 220; intelligence



spending, 149, 151, 161; Japan bombed by, 38, 56–70; Khrushchev visit, 132– 33, 154–55; LBJ, 172, 189; Los Alamos laboratory, 56–57; Manhattan Project, 36, 56–62, 67; McCarthyism, 163; Nixon, 173–74, 189; NSA, 4, 151–52, 161, 163; nuclear research/ weapons, 10, 36, 38, 56–70, 80n142, 152; Reagan, 4; Roosevelt, 56, 64; Senate Select Committee to Study Governmental Operations with Respect to Intelligence Activities, 7; sigint, 155, 158–61; Soviet atomic intelligence and, 68; spy-flight activity/U-2 spy plane, 155, 156; Vietnam War, 150; West German secret services and, 209, 212–13, 215, 221; World War I, 103. See also Allied intelligence Unshlikht, Iosif, 18 uranium: German, 56, 63, 77n101; Soviet, 57, 58, 61, 63 Urbach, Karina, 1–11 USSR. See Soviet Union van Creveld, Martin, 101 Vannikov, Boris, 60, 61 van Overstraeten, Raoul, 83, 117n12 Vasilevskii, A. M., 50, 64–65 Vatican, French intelligence on, 130 Versailles treaty, 104, 106 Verzat, Lieutenant-Colonel, 143–44 Vetter, Heinz Oskar, 184, 201n53 Vinarov, Ivan (“Mart”), 28–29 Vivian, Valentine, 24 Vlasov, A. A., 212 Volkogonov, Dmitrii, 40 von Berg, Hermann, 182–83, 200– 201nn48,51 von Manstein, Erich, 83 Von Scheliha, Rudolf (“Ariets”), 43, 44 von Schulenberg, Werner, 49 Voroshilov, Kliment, 19–20 Vuillemin, Joseph, 97, 107

Index

245

Wagner, Armin, 219, 228n48 Warsaw Pact: British intelligence on, 154, 158; Czechoslovakia invasion, 186, 220; East-West détente, 170–208; France and, 130–44, 171; Permanent Consultative Assembly, 207n119; Stasi espionage, 180, 186, 190; West Germany and, 171, 217–21. See also Central and Eastern Europe Wehner, Herbert, 177, 189, 201n59, 204n80; SPD, 183, 184, 187, 199n39, 201n58; Stasi informant, 183, 184, 187, 199n39, 200–201n51 Wehrmacht: blitzkrieg, 83, 98; British and, 85, 116n7, 118–19n29; FHO, 82, 212, 215, 218; France and, 82–89, 97, 106, 108–12; Luftwaffe, 49, 84, 87, 88, 97, 109, 114; multi-pronged attack, 83; offensive strategy, 85–86; Panzer forces, 84, 86, 87, 110, 116n6; sickle-cut strategy, 83; Soviets and, 38–39, 40, 51, 53–55, 67; spending, 106, 121n68; West German secret services and, 209 Wessel, Gerhard, 214, 215 West: East-West détente, 170–208. See also Allied intelligence; Britain; France; NATO; United States; West Germany West Berlin: Allied intelligence, 9; Brandt as mayor of, 172, 223; Quadripartite Agreement (1971), 178–79, 188, 190, 204n88; spy center, 218 West German secret services, 210–30; A2/G2 functions in staffs of armed forces, 210; ANBw, 210, 211; BND, 7, 157, 210–30; Bundesamt für Verfassungsschutz, 210, 211, 215; Cold War, 209–30; disinformation, 183, 199n39; Friedrich Wilhelm Heinz Dienst, 210, 217; Lageberichte Ost, 216; MAD, 210, 211; Ostbüro, 182, 210; Russian spies in, 214; ZNBw, 210, 211 West Germany (Federal Republic/

246

Index

FRG), 170–208; Adenauer, 171–72, 199n43, 213–17, 221, 224, 227n35; Brandt, 172–78, 182–92, 200nn46,47, 201n53; Bundeskanzleramt (Federal Chancellor’s Office), 215, 216, 221, 223; CDU, 172, 176, 181, 185, 213; CSU, 172, 185, 230n87; DGB (German Trade Union Federation), 184–85, 201n53, 202n59; East-West détente, 171–208; Erhard government, 183; and nuclear weapons, 187; Ostpolitik, 173–94, 200nn46,47, 205n98, 206n107, 207n117; rearmament, 176, 213, 216–17; Stasi espionage, 181–208; Stasi informants in, 181–84, 187, 199n39, 200–201nn47,48,51,52, 204n79, 206n101, 211; Stoph-Brandt meetings/”Confrontation” and “Confrontation II” (Erfurt and Kassel), 178, 187, 189; trade unions, 176, 184–85, 186, 197n18, 201n53; Treaty of Moscow, 178, 193, 207n117. See also SPD; West Berlin; West German secret services Weygand, Maxime, 97, 105, 114 Whaley, Barton, 39–40 White, Dick, 150, 153 Wieck, Hans-Georg, 223–24 Wienand, Karl, 187, 199n39m187, 199n39, 201n52 Wilson, Harold, 173, 175 Wilson, Woodrow, 104 Winzer, Otto, 176, 182

Wirth, Joseph, 103 Wohlstetter, Roberta, 39–40 Wollweber, Ernst, 219 workers: worker correspondents, 27. See also trade unions world revolution, communist, 12, 107, 215 World War I: Britain, 23, 100, 103; “continuous front” doctrine, 98–99; European population after, 103; France, 88, 90–92, 94, 96, 101, 103, 117n10; Netherlands, 83; U.S., 103 World War II: atomic bombing of Japan, 38, 56–70; British intelligence, 5; French intelligence, 81–127; Pearl Harbor attack, 39–40, 86, 98; Soviet intelligence, 36–80; “Ultra” and “Magic,” 101. See also Nazi Germany Wörner, Manfred, 211 Yagoda, Genrikh, 21 Yemen, British intelligence, 160 Yugoslavia: French intelligence on, 134, 142, 143, 144; French military attachés, 139 “zero-sum game,” 172 Zhukov, G. K., 42, 51–54, 64, 68, 69, 218 Zhunkovskii, Vladimir, 25 Zinoviev, Grigorii/Zinoviev Letter, 6, 18–19 Zolling, Hermann, 224