The Art of Double-Cross: writers in strategic deception during World War Two

ABSTRACT

The success of British double-cross operations in World War Two is well-known. However, the techniques used by writers in MI5 to manipulate German intelligence officers and, through them, the German High Command have never been properly examined. This essay fills that gap of understanding, focussing on the most ambitious of the double-cross operations, the network of Juan Pujol, known as agent Garbo. As Michael Howard, Christopher Andrew and others acknowledge, the double-agent networks were crucial in disseminating disinformation to the enemy, including in the run-up to D-Day. As the article shows, however, they were also used more strategically, to wage a sustained campaign of manipulation against their opponents which ensured that deception plans were swallowed and acted on. By examining their tactics and strategies in detail, the essay highlights the historic contribution of writers in British intelligence during World War Two which has previously gone almost unrecognised.

KEYWORDS:

• World War Two
• strategic deception
• double-cross
• Juan Pujol Garcia
• Garbo
• operation bodyguard

Introduction

At the Tehran conference in November 1943, Winston Churchill famously called for a ‘bodyguard of lies’ to protect the long-awaited Allied counteroffensive against Nazi Germany.¹ This article examines the work of writers in strategic deception who were already hard at work making this call a reality. It focusses on the XX (double-cross) system, in which writers working for MI5 engaged in an extended written correspondence with German Intelligence. Some of the double-agents whose traffic they created were based on real people, but most were entirely imaginary and the correspondence as a whole can be considered as an extended work of fiction. Examining it in this light, this article reveals the ways XX writers drew techniques from the literary domain to manipulate their readers in the Abwehr and, through them, the German High Command.

According to Michael Howard, strategic deception ‘was possible for the British only because they enjoyed two extraordinary, perhaps unique advantages’²: their success in codebreaking and excellent domestic security within the United Kingdom. This is undoubtedly the case, but the strong hand needed to be played well: how this was done is the concern of this article. Indispensable to the success of strategic deception, I suggest, were the skills of the writers who wove the web of fiction within which German Intelligence became entangled. Guided by established authors including Dennis Wheatley in the London Controlling Section (LCS) and John Masterman through the XX Committee, dozens of agents and sub-agents were created, each with a back story and what amounted to a ‘complete life’ for their readers in Madrid and Berlin.³ Each needed to be credible in themselves, provide possibilities for strategic use, but also synchronised with the whole to avoid contradictions that risked unravelling the enterprise. For the first half of the war there was no formal deception policy and this in itself presented significant challenges, akin to mobilising a large cast in a play without knowing its plotline. At this stage the work was almost entirely speculative, as Masterman says, but nevertheless ‘it was always in the back of our minds that at some time in the distant future a great day would come when our agents would be used for a grand and final deception of the enemy’.⁴ Only when the Allies moved to the offensive could an overarching strategic deception narrative be agreed on, to which each of these individual stories could be harnessed.

For such a narrative to become established in the minds of the enemy, it was not enough to simply transmit misinformation. They also needed to establish a high degree of psychological control over key officers in German intelligence. Psychological manipulation of a reader is a necessary part of the skillset of writers, used to generate empathy, suspense, anxiety or other affects. Deception, however, was not ultimately about manipulating people to believe things but rather about manipulating them to act, and to act in a way that benefited the Allied war effort. Therefore the ultimate target of Allied deception was not merely intelligence personnel, who would pen nominally authoritative reports based on largely bogus information, but the German High Command that would read them and act. Abwehr officers therefore needed to be controlled by the XX writers almost to the point of ventriloquism, to become conduits rather than critics. The level of difficulty this posed can hardly be overstated. At the beginning of the war the established culture of the Abwehr was to communicate with their agents in a highly directive manner. Agents were employed to deliver concrete and precise information in response to detailed questionnaires on the specifics of military concentrations, movements and logistics. Early in the war Juan Pujol Garcia – the double agent codenamed ‘Garbo’ by his British handlers, whose reports formed a crucial element of the deceptions supporting Operation Overlord – attempted to break out of this straitjacket by introducing independent observations and suggestions, only to be rebuked by his German handler in Madrid:

Received yours of the 3/9 from Glasgow unnumbered, and today yours of the 10/9 from London, no. 5. The contents are of little use – information lacking. Abstain from personal observations and military propositions … . We are not interested in third hand information.⁵

When the XX Committee took over the running of Pujol and his fictional network in July 1942, his relationship with Abwehr case officers was that of subordinate to superior. When they attempted, through him, to report ‘overheard’ intelligence on Allied plans for bombing German industry, he met with firm correction: ‘We insist that you must not fail to take steps to obtain verified information’ (emphasis added).⁶ On 25 August 1942 he was further instructed to ‘reduce the amount of correspondence, send shorter letters, and omit everything not absolutely necessary’.⁷

By 1944, the team’s success in remoulding this relationship is strikingly evident in the correspondence. In the run-up to D-Day, it was now Pujol who posed questions to German intelligence officers, while theirs were ‘only to guide you as to the points which interest us so that you, being a better judge of your agents than we are can decide which of these questionnaires you can pass on to them’.⁸ It was Pujol who now directed what should be reported on and how his network should be run. As communications from the senior ranks of German intelligence in Spain show, not only had Pujol become the dominant partner in the relationship by this point, he was also seen as best placed to define their intelligence needs. The evidence overwhelmingly shows, moreover, that the XX team were able to maintain this position of psychological dominance over their opponents until well after D-Day. As senior officer Karl Kühlenthal (codename Carlos) wrote deferentially to Pujol on 31 August 1944:

It is sometimes difficult to express exactly the ideas and views which I hold with regard to your formidable organization which you have managed to set up. The results, its work, have been, and continue to be, a perfect reflection of the development of the situation in [Britain] for the use of our Headquarters. Thus it has given me the most genuine satisfaction when a few days ago I was able to transmit to you the news that you had been awarded the Iron Cross, and I wish again to repeat to you today my most cordial congratulations in which my colleagues join me. At the same time I wish you to make known to all your collaborators our profound recognition for the work they have accomplished which could not have been more perfect.⁹

On the one hand, this evolved relationship reflects the increased need, even desperation, felt in Berlin for intelligence from the UK and on the build-up of Allied forces, with the consequent imperative to husband scarce human sources. But on the other hand, it also illustrates the remarkable success of the deception planners in managing their audience. Much as the conjurers and tricksters aided Dudley Clarke’s A-force in the dessert of North Africa, careful and continuous pressure exerted by XX writers facilitated the vital suspension of disbelief prior to Overlord.¹⁰

The role of writers in strategic deception

The contribution of mathematicians such as Alan Turing and Gordon Welchman to codebreaking at Bletchley Park and its significance to deception is well recognised, but that of writers in strategic deception much less so.¹¹ Part of the reason may be because relative anonymity or obscurity was more the norm than the exception for the exceptional talents of the war. While Turing’s contribution to Allied cryptography is widely celebrated and known, for example, that of Tommy Flowers, the Post Office engineer and scientist who led the team which designed and produced the ‘Colossus’ machine, widely recognised as the first electronic computer, is much less so. Among the writers who worked on the crucial deceptions of the war, moreover, another reason may be that the collaborative nature of the work makes it hard to identify standout ‘genius’ figures. This is in stark contrast to the scene from which many of these writers emerged. The literary culture of the 1930s and 1940s was strongly organised around the idea of the exceptional individual talent, despite T. S. Eliot’s insistence that Art should be an escape from the ‘personal’.¹² Masterman’s own first novel An Oxford Tragedy had been found wanting by Graham Greene because it was not individual enough,¹³ while Wheatley was lauded by critics because of his uniqueness.¹⁴ Those who adapted their skills to the art of double-cross during World War Two, however, worked in a very different way to this. Even the most celebrated and distinctive voices such as Pujol’s were, in actuality, enormous collective efforts involving multiple authors from planning to production. The names of all the writers within MI5’s B1A section who worked to manipulate Pujol’s handlers in German intelligence are not captured in Security Service records. However, it is clear from the nature and sheer quantity of correspondence – almost 5,000 pages of very carefully constructed text – that there must have been a substantial team. Case Officer Tomás Harris and Juan Pujol himself were both involved day-to-day, while Masterman and the members of the XX Committee authorised communications and exercised editorial control. Although Pujol’s imaginary network was the largest, a comparable effort was involved for each of the double-agents and networks whose traffic needed to be planned, written and dispatched from day to day.

Indeed, for individual operations the writing team could grow quickly and occasionally curiously. Strategic deception requires very careful plotting and orchestration. There is a balance to be struck between the coherence and security of the ruse, which requires a tight-knit organisation, and its application into broader operations and strategy, which requires integration with the wider defence, policy, and security machinery. But in certain instances during the war the desire to hone the perfect deception militated against the instinct for maximum secrecy, and presented a danger that too many would-be writers wanted to be involved. In the case of operation Mincemeat, the idea of passing faked documents to the enemy by planting them on the corpse of a fictional airman, Maj Arthur Martin, was borrowed by Ian Fleming from Basil Thomson’s novel The Milliner’s Hat Mystery (1937). Thomson himself likely drew much of the idea from Ferdinand Tuohy’s ‘The Lost Haversack’ (1920), an account of a successful deception he will certainly have come across in his research on intelligence in the First World War.¹⁵ In 1943 the idea was developed by Naval Intelligence Officer Ewen Montagu, who would later publish his own account in the widely read The Man Who Never Was (1953). Plans were scrutinised and refined by the LCS, where Wheatley worked under Commander Johnny Bevan, and by the XX Committee under Masterman. The documents to be planted on the corpse, among them letters purporting to be from Lord Mountbatten and the Vice-Chief of the Imperial General Staff, were edited and redrafted multiple times. As Ben Macintyre writes, ‘everyone, from the Twenty Committee to the Chiefs of Staff, had a different idea’.¹⁶ Dudley Clarke, who himself went on to be a successful writer of fiction and non-fiction, contributed a draft. An assistant to Ian Fleming, Victoire Ridsdale (thought to be a model for Miss Moneypenny in the Bond novels), contributed love letters from the dead man’s fictitious girlfriend.¹⁷ To add an additional measure of authenticity General Archibald Nye contributed a further fake letter, writing as himself.¹⁸ Described this way, the process of creating the crucial documents for the Mincemeat deception seems chaotic indeed. Nevertheless, the success of this joint effort was immediately evident when the corpse was picked up by Spanish authorities and the documents shown to the Abwehr. Ultra revealed that the Germans were taken in by the documents in the short term. And, in an echo of what would occur subsequently with the Fortitude operations supporting the Overlord invasion, also to a surprising extent in the longer term. As Christopher Andrew notes in his authorised history of the Security Service, even though ‘the Allied attack came in Sicily rather than Greece, the Germans did not doubt the authenticity of the Mincemeat documents but concluded that Allied plans had changed’.¹⁹ Upon learning of the operation’s success, an official wrote to Churchill: ‘Mincemeat swallowed rod, line and sinker’.

The large, later operations such as Fortitude necessitated far more coordination and focus, however. Where Mincemeat involved the creation of two characters and associated contexts – Major William Martin and his girlfriend Pam – Fortitude required dozens. Characterisation was often understood in this period in terms of E. M. Forster’s distinction between ‘flat’ and ‘round’ characters, and this usefully describes the way notional agents were drawn for the benefit of German Intelligence.²⁰ Most of the members of Pujol’s network, for example, were ‘flat’ characters in that they were defined almost exclusively in terms of one quality, such as hatred of the English. Interestingly, German case officers seem to have preferred to treat these sub-agents as ‘types’ and discouraged any disclosure of their personal details on the grounds of security, despite the clear incentive this procedure presented for unscrupulous agents to generate fictitious sources. As Kühlenthal told Pujol in April 1944:

In all events I shall always be glad to receive the details which you send about each of the new collaborators. By this I mean the details as to the class of the individual he is and where he is working. These particulars make it possible for me to evaluate the reports whereas other details such as names, etc., are of no interest and it is advisable that you should not communicate this either by letter or by message.²¹

‘Rounder’ characters such as Pujol himself were developed much more extensively, however. As Pujol’s case office Tomás Harris wrote, initially their approach was to style Pujol’s character around ethnic stereotypes, specifically ‘what we believed to be the German understanding of Spanish psychology as they appear to have conceived it through their association with Falangist Spain’.²² This approach tallies well with more contemporary insights about cognitive biases and deception, particularly that it is easier to reinforce existing prejudices than to challenge them. Progressively, as I will show however, Pujol’s changes of mood, his heroism tempered by vulnerability and clear-sightedness mixed with emotion and unpredictability, became key levers through which the XX writers were able to gain the initiative and manipulate the opposing side. In the case of Dusko Popov (agent Tricycle) a similar process of character development took place. While Popov’s initial stylisation was relatively crude, leaning on ideas of the gentleman spy associated with figures like Wheatley’s Gregory Sallust, he was eventually able to develop personal relationships high up within the Abwehr that allowed British Intelligence to penetrate the organisation ‘to the point of knowing all its important officers, how to influence them psychologically, and how to foresee their reactions’.²³

At different stages of the war, MI5 ran well over 50 double-agents back to German Intelligence, some based on enemy infiltrators who had surrendered voluntarily or been captured, but most entirely fictional. As in conventional authorship, a central principle for the XX writers was to ‘show not tell’. As Masterman says in The Double-Cross System, ‘the German staff officer needed facts, for with facts before him he could make his own appreciations’.²⁴ As the team discovered very early, attempts to sell false information about Allied intentions (for example by reporting a conversation between Foreign Office officials) were almost futile, while small ‘factual’ details observed in person by a lowly sub-agent were more likely to be weighted. For the German intelligence officer,

it was of far more value to him to learn, for example, that a certain division had moved to the area of one of the northern ports, or that it had had instruction in mountain warfare, or that it had been issued with arctic equipment, than it was for him to hear that ‘Lord So-and-So in the Cabinet told me in the utmost confidence that an invasion of northern Norway was being discussed by the Chiefs of Staff’’.²⁵

Creating a large-scale deception would therefore involve, on the one hand, obscuring Germany’s independent insight into what was occurring in Britain, and on the other, scattering a large number of small clues, in order that they could be picked up and pieced together into a complete picture by the enemy. Masterman’s The Case of the Four Friends neatly illustrates his own exceptional skills in this department. In this novel, a detective must try to deduce the course of a murder before it has actually taken place, much as the Abwehr needed to deduce, in advance, Allied plans to re-invade the continent. Multiple lures and snares are woven into his narrative to keep the reader in a state of engagement and calculation. For an operation like Fortitude South to be effective, similarly, the enemy needed to be kept fully engaged and invested in the XX traffic, as they sought to piece together Allied intentions for the counter-offensive.

The ‘complete picture’ to be reconstructed in this case was the false order of battle that kept a large concentration of German forces at Calais rather than reinforcing Normandy. From a military and strategic perspective, as Dennis Wheatley recognised, a successful deception operation needed to be at least as credible and plausible as the real operation it was designed to cover. Prior to his work as a deception planner Wheatley had also demonstrated exceptional imaginative skills in this area with his ‘wartime papers’. These took a series of hypothetical, but potentially significant adverse scenarios and drew out likely consequences and sequences of events. Topics ranged from civil unrest during the Blitz to a German invasion of the UK mainland. As Wheatley says in his memoirs, by the time he was appointed to the LCS in autumn 1941 he had ‘written over half a million words for the Joint Planning Staff’.²⁶ Under Johnny Bevan, he took a major role in planning the overall narrative which was to be ‘reconstructed’ by German Intelligence for many of the major deception operations including Bodyguard.

The XX Committee controlled all leakages of true information, digested decrypts of German communications and exercised editorial control over the stories of all the double-agents. Teams led by case officers then worked to develop and sustain what they described as a ‘complete life’²⁷ for each agent, ensuring continuity and consistency. To be most effective, Masterman writes, it was necessary for the writers of each agent’s traffic to live this fictitious life vicariously, as far as possible. Such were the demands on the team as a whole that, unfortunately, not all promising cases could be developed or exploited. As Masterman says, ‘studying the luxuriant prose of GARBO or the terse and virile telegraphese of TATE, one may speculate a little sadly on the “mute inglorious Miltons” who … could not find a publisher’.²⁸

Susceptible readers: the role of the Abwehr

They did, however, have readers. Double-agents’ reports were transmitted to handlers in Germany’s Abwehr, who then digested and circulated their intelligence to the wider military and political community. Ruses require audiences and a successful deception requires someone to be deceived at the least, and at best that the target becomes complicit. Strong Allied security helped enormously in this respect, certainly, but German analytical culture also offered deception planners an opportunity. In exploring the success of the XX system, therefore, there are aspects of the Abwehr’s operations, and that of the Nazis’ intelligence machinery in general, that merit consideration. Crucially, the Allies’ capacity to eavesdrop on German communications offered them a window through which they could observe and study this machine, enabling them to exploit its weaknesses.

The Abwehr was Germany’s principal intelligence service until 1944, when it was abruptly dissolved and espionage transferred to the Reich Security Administration.²⁹ Its organisational roots can be traced back to the Treaty of Versailles, when the Intelligence Office of the General Staff became an Intelligence Group attached to the Foreign Armies Branch. Nominally a counterespionage shop – literally ‘defence’ – its work included espionage from the outset.³⁰ Following various organisational reshuffles in the 1920s, it absorbed several other agencies to become the military espionage agency of the German High Command of the Armed Forces. It was headquartered with the OKW in Berlin and from January 1935 was headed by Wilhelm Canaris, a Navy officer. Under his leadership the organisation, for a time, settled many inter-service rivalries, notably with Reinhardt Heydrich and his Security Organisation, and secured responsibility for foreign espionage.³¹

In this mission the Abwehr achieved notable successes, including running a sustained and successful double-cross operation against the British, with operation Nordpol.³² But despite being an adept bureaucratic pugilist, Canaris’ intelligence service, and indeed, the wider intelligence machinery of which it was a component, bore several fundamental flaws which made it susceptible to accepting and disseminating deceptive information. The first (though debated) is the question of its loyalty to the Third Reich. There can be no doubt that Canaris was not enamoured by Hitler’s regime, though not necessarily against the war itself, and on several occasions either misled Hitler deliberately, as he did in discouraging Hitler from an operation to invade Switzerland, or was complicit to a greater or lesser extent in plots to assassinate the Fürher. Moreover he, along with many of his people, maintained contacts with Allied intelligence agencies from time to time. His SIS cryptonym was ‘Theodor’. In Keith Jeffery’s judgement, ‘Both Canaris and Gisevius [the Abwehr’s representative in Zurich] were involved with opposition groups in Germany, but the extent to which their contacts with foreign intelligence agencies in Switzerland … constituted treason is debatable’.³³ Debatable, perhaps, but an intelligence agency working at odds with its political and military leadership was both problematic for the Nazi regime and a gift for the deception planners in London. For even if it was not the case that Abwehr officers were deliberately complicit in the deception operations, passing on intelligence that they knew to be false (as has been claimed) such a serious disjoint between producer and consumer was eminently exploitable by Allied intelligence.³⁴

Second, structural and cultural issues within the Abwehr and the wider intelligence machinery rendered it less effective in weeding out good intelligence from bad. Frequently its officers substituted volume for quality, passing on dubious information or here-say to handlers, even in the absence of deception operations.³⁵ An early post-war account, based on captured documents and on interrogations of Abwehr officers, paints a bleak picture of the quality control in the organisation. This applied both to officers and to the product they developed, especially once Germany ceased to be on the offensive. These negative qualities, US analysts judged, were very much a reflection of Canaris as leader – ‘more of a professional intriguant rather than an organizer’ – who both appointed weak officers and ‘gave them practical independence’.³⁶ They had free enough rein to enjoy the benefits of their postings as long as they submitted the required number of reports. One officer noted to his interrogators that ‘it was better to have a controlled agent than none at all’.³⁷ This was accompanied by a structural and cultural disinclination to engage in robust analysis and assessment.

This, from the top, was driven by Hitler, who as the war progressed became increasingly dictatorial in directing what was and was not an intelligence priority.³⁸ As countless studies of intelligence failure and politicisation have illustrated, leaders who make it expressly clear what they do and do not want to hear tend to receive intelligence to please, not necessarily the intelligence they need. Not only does this create a disincentive to careful analysis and assessment, it creates a disinclination to deliver bad news, and, frequently, it encourages unhealthy competition between or within agencies, as senior and junior officers vie for the favour of their masters.³⁹ Hitler’s model of management, with his tolerance for competing agencies, was not conducive to a critical culture nor to deep cooperation – the Abwehr, for instance, struggled to gain decent access to SIGINT. Inevitably, the Abwehr below Hitler adopted many of these negative traits. Crucially, it did not develop processes for careful all-source analysis. The German system was fragmented. Unlike the British, who created a Joint Intelligence Committee supported by a Joint Intelligence Staff to evaluate intelligence on significant matters on an inter-service basis and with access to all sources, the Abwehr passed its reports on to the service departments, which then assessed them but lacked a mechanism for properly evaluating the reliability of the material.⁴⁰ As the war progressed and the pressure mounted on Germany, the process became increasingly chaotic.

The art of deception: writing the double agent ‘Garbo’

As Michael Howard says in Strategic Deception in the Second World War, the object of deception is not merely to influence the way an enemy thinks, but to influence their actions. ‘When seeking to manipulate junior commanders this may not be too challenging’, he argues, but to target the enemy high command is ‘normally so difficult that it has seldom been attempted’.⁴¹ In Howard’s assessment, the XX system was pivotal to Allied deception in World War Two because of its success in feeding ‘carefully orchestrated misinformation into [the German] intelligence system’.⁴² Similarly, Christopher Andrew characterises the system as a means of passing ‘a mixture of information and disinformation with which they could both impress and deceive German case officers’.⁴³ In fact, as this article will show, sharing tit-bits of real intelligence was not the only – or even the primary – means by which the XX team were able to gain the initiative over their opponents. Abwehr case officers needed to be made into reliable mouthpieces for British Intelligence and this implied a high degree of psychological control. As we have already seen, such control went strongly against the ingrained culture of the Abwehr, in which officers were resistant to receiving any strategic or tactical analysis and wanted only unvarnished ‘facts’ to form their own appreciations and to meet the expectations of their commanders and customers. Of course, their successful deception during World War Two was achieved by a number of means, including wireless and visual deception. Surveillance flights were flown over the wrong targets and misleading naval manoeuvres took place. However, between 1941 and 1944 a very heavy burden fell on the use of double-agents: that is, on an enormous volume of written correspondence between XX writers and German Intelligence. Hitherto, the mechanics of this correspondence and the writerly techniques that drove it have never been closely examined. This essay seeks to do so by focusing on the most ambitious double-cross project, Juan Pujol, known to the British as ‘Garbo’.

To understand how this vast epistolary novel developed, it is useful to start with the relationship the team inherited when Pujol came under MI5 control. Established practice within the Abwehr was to address agents in a highly directive style, with questionnaires whose bluntness and specificity were very difficult to evade. When Pujol attempted to diverge from this in early communications, he was reminded of expectations:

We are interested in concrete information with dates. Work with competent people – don’t communicate third-hand information. We are interested in … troop movements, types of uniform, units, different armaments, the position of aerial and naval gun emplacements, the position of aerodromes.⁴⁴

A first step in shifting the character of this relationship was to put the Abwehr officers themselves to work. Having no deception policy to work to, the writers ‘endeavoured to report as much confusing bulk as possible’, to ‘swamp the Germans with information, misinformation and problems’, Harris writes.⁴⁵ Considerable labour was required on the part of the Felipe Stelle – the Abwehr branch in Madrid to which Pujol reported – to develop secret ink, decipher script written tightly over cover letters and decrypt encoded messages. The desire to make the enemy work, then, provided an incentive to continually expand the number of invented sub-agents working under Pujol. The larger his network grew, however, the greater the pretext it also provided for developing long and detailed narrative threads within which it might be possible to ensnare enemy intelligence. Indeed, examination of the correspondence reveals that far more happened in terms of manipulation than simply causing the Abwehr needless work.

To understand the techniques used to achieve this, I will be drawing here on a few terms from Roland Barthes’ narratological classic S/Z. First among these is the ‘feint’.⁴⁶ Familiar in many forms of combat, the feint is a mock attack, usually designed to open the opponent’s guard. Importantly for our purposes here, it works by creating a risk for the adversary, forcing their hand in so doing. From the beginning of their correspondence with the Abwehr, it is notable how adept the XX team were in the use of this technique, as their outgoing letter dated 28 August 1942 beautifully illustrates. Here, Pujol begins for the first time to paint a picture for his handlers of his difficult position as a Nazi spy in Britain, almost penniless and at constant risk of capture and execution. In a first attack, he complains that the secret ink supplied to him is not as they had claimed, ‘impossible to discover’, but is in fact easily developed. ‘Should the English censor test the correspondence from the continent they would easily find the ink’, he writes, ‘the consequence of which to me would be that I would have to pay for it with my life, a very remarkable thing’. Here Abwehr officers are accused of having insufficient regard for his safety but also implicitly of incompetence. Next, he amplifies an already-established refrain, the inadequate funding of his network:

With regard to the question of money … it is extremely easy to calculate what my requirements are for each fortnight. Therefore I feel that the responsibility for my not having received funds for at least a month ahead is only on your shoulders … . All this has worried me a great deal and I want you to know that were it not for the esteem which I feel personally for you, which I feel you reciprocate, as well as the interest in helping our cause … that in all sincerity, and as a friend, that I would have returned to Spain some time ago … . I consider that now less than ever should I be exposing myself to useless risks which are in our power to avoid … . I have been passing through a long period of nervous strain which affects one’s morale due to the responsibilities and delicacy of the work.⁴⁷

Risk is created for the survival of Pujol’s network on three levels here. Firstly, the Abwehr have shown themselves unable to protect his personal safety. Secondly, they have been unwilling or unable to fund their own network adequately. This effectively confronts the officers with a choice, of either allowing the network to disintegrate (and answering to Berlin for that failure) or securing better finance for it. Thirdly, Pujol’s fragile mental health necessitates a change in the way he is handled, if he is to continue as an agent, thus forcing a more personal approach. Finally, in this first long letter, the team take the opportunity to introduce a fourth, larger horizon of risk, opening a theme that would come to dominate the entire correspondence: the expected counter-offensive or Second Front of the ‘English Mangantes [brutes, but also fraudsters]’.⁴⁸ The feint, in creating the illusion of peril for the network, conjures a more visceral fear for his handler: his position and privilege.

Conjured out of an entirely fictional scenario, each of these four rather different feints were designed to elicit a reaction from their readers. It is interesting to see, therefore, how quickly and effectively the second and fourth provoked responses. In his reply case officer Friedrich Knappe (codename Frederico) details the arrangements he has put in place to provide ‘larger remittances’ with an immediate £500 and a promise of £1000–2000 to follow. This, in itself, represented a watershed in terms of the Abwehr’s financial commitment to the network. Swallowing the fourth feint, Knappe continues: ‘Second Front! Very important! It is of the greatest importance that you should intensify all your efforts to try to get extensive information and transmit it to us direct here’.⁴⁹ There is tacit agreement here that, rather than simply answering questionnaires, agents should now be taking the initiative in information gathering. From the point of view of deception, clearly, this significantly enhanced their potential. Feints one and three, the team’s attempts to elicit a more personal and supportive approach to Pujol, took longer to bear fruit as we will see, but in the long term proved to be very important in changing the tenor of the relationship.

Related to the feint in Barthes’ narratology is the ‘snare’. In everyday language, a snare is a baited trap which entangles the prey and prevents its escape: in a text, it works similarly. In a murder mystery, for example, the reader might be tempted with a clue to the identity of the next victim, while actually being entangled in falsehood and psychologically prepared for the author’s next trick. Perhaps the most important snare in the Pujol correspondence was the unfolding story of a vast military storage complex being built in the Chislehurst caves in Kent, fed out in dozens of communications from July 1942. According to case officer Tomás Harris, this was ‘intended to provide the opposition with a great bulk of apparently interesting information which would lead them nowhere … . All the information passed over in this connection was untrue, it being entirely imaginary’.⁵⁰ Once again, however, when the exchanges are closely examined it can be seen that much more manipulation took place around Chislehurst than this account implies.

As a narrative, Chislehurst can be said to create both risk and temptation for German intelligence. On one hand the mass storage of weaponry implies that the ‘Second Front’ may be imminent, but on the other it provides the opportunity for an intelligence coup. Detailed developments were fed out over the course of a year with the help of Pujol’s (notional) agent 4, who was able to find employment in the Chislehurst complex itself. On the 24 April 1943 Pujol reported huge stores of small arms including rifles, grenade launchers and rapid-deployment weapons for use against armoured vehicles. Great quantities of munitions together with stockpiles of French and Belgian army uniforms were also being stored ‘to arm the Fifth Column of Europe at the time of opening a Second Front’.⁵¹ This was further reinforced on 3 June with a mass of detail on positions of small tunnels, stations and railway cuttings around Chislehurst’s exterior – a make-work encouraging German case officers to waste time and effort deciphering their collective significance. Agent 4 was further able to confirm the presence of ‘special asphalt compounds’ underground, intended to be used as laboratories to prepare a new weapon. Pujol reports that ‘it is considered safer to fill it immediately before using it, in the same depot, to avoid deterioration of the chemicals which constitute the explosive’. In the by-now-established pattern, pieces of the jigsaw but not the full picture are provided, as he says that ‘he has not been able to discover what kind of weapon it is’.⁵²

If we consider the discovery of this important complex as ‘bait’, the snare prepared by the team is laid on 3 June 1943, when Pujol proposes a plan to sabotage and disable the entire complex. ‘If this were possible, I am sure that we could immobilize an enormous quantity of supplies for a long time … . That is to say, we would disorganize the continuation of an offensive’.⁵³ Pujol’s principal proposal is to blow up a train as it passed through the central tunnel. Developing multiple possibilities for carrying this out at some length, he concludes however that:

I have no intention of deciding anything unless I have an opportunity of discussing it personally with one of you. I remember how time and again Frederico [case officer Friedrich Knappe] was envious of me when I left on my mission to this country and thinking how he would like the opportunity to come to this country without danger I have considered a plan for him to do so. I am convinced that if he, or one of you, were here with us, we could discuss all the details and future plans much better than if we wrote hundreds of letters. For the plan to be put into operation it would be necessary that I should have your cooperation.⁵⁴

Decrypts of communications between the Abwehr in Madrid and Berlin reveal that, while Pujol’s report on the development of military facilities in the Chislehurst caves were forwarded to Berlin in the form of a detailed report, both the sabotage plan and the invitation to Frederico (Friedrich Knappe) to supervise it were not. The Felipe Stelle in Madrid will have known very well that, if identified as an enemy spy on British soil, Knappe could expect to be interrogated and executed. The Stelle were content for Pujol and his agents to take these risks on a continuous basis, but were clearly unwilling to do so themselves. After much prevarication Kühlenthal eventually writes to Pujol by way of justification:

I have already let you know in a previous letter that your work on the undergrounds has been recognised from all points of view. Believe me, I understand very well the reasons which have induced you to make us the proposition [that Knappe should come to Britain to supervise the sabotage plan]. The ideas which you put forward fall completely outside the territory of our work and to realize it we would have to use other people foreign to us to take charge which would mean that we would have to abandon all the immense work and the network which you have organized.⁵⁵

The Chislehurst narrative was, as we have seen, nothing more than a work of fiction. Through it, however, it is possible to see how the XX team effect an important shift in the relationship between Pujol and his handlers. Pujol and his agents are willing to risk their own lives to protect German soldiers and to advance the German cause, while their case officers are not. Pujol is not only capable of gathering critical military intelligence but digesting it, forming appreciations and offering real-world proposals for turning it to Germany’s advantage. It is clear from decrypts that Madrid accepts the strategic importance of the Chislehurst complex. As their response makes plain, however, their primary motivation is to retain their intelligence asset. They prefer this over the chance to strike a blow against Allied preparations for the Second Front – prioritising their own position by delivering only what was expected of them, a quantity of military information. An implicit contrast is therefore established here between the bold, effective and strategic Pujol network and their self-interested, uncommitted and somewhat deceitful handlers. By failing to refer the sabotage plan to their superiors, the Felipe Stelle may have materially damaged Germany’s position.

The XX team, unlike their opponents, had the benefit of decrypts which revealed exactly how their reports and proposals were being received and what the Abwehr in Madrid chose to pass on or withhold from their superiors. In any game of wits, the ability to see the opponent’s hand is clearly an enormous advantage. The creativity with which the XX team were able to exploit their privileged knowledge has tended to go unrecognised, though. This is partly because the post-war reports written by case officer Tomás Harris and J. C. Masterman as chair of the XX Committee are almost completely silent on the subject of psychological manipulation, an aspect of the work which was essential if Abwehr officers were to be used effectively to influence the German High Command. Therefore, it is necessary to go back to the exchanges themselves to see first-hand the techniques that were used. In this case it is impressive and somewhat amusing to see how, on 2 August 1943, the team worked to intensify the discomfiture of German case officers. Pujol writes:

I do not understand and I condemn the interpretation you give to the UNDERGROUND matter. I wish to know urgently the opinion of the experts [to whom Madrid claimed to have referred the sabotage plan] so as to know whether or not to bother further with it, in which case the responsibility will not be mine.⁵⁶

Much was at stake here for the Felipe Stelle in relation to the expected Second Front, and Pujol now had a significant hold over them. Still a long way from D-Day, a dynamic had become established in which Pujol and his agents were pro-active and patriotic, while the Abwehr were ambivalent and compromised. It was now much less easy to treat Pujol as a subordinate, obeying orders. As an analyst of critical intelligence he was on his way to establishing a position of full parity with officers in Madrid and was already their moral superior. As has been seen time and again since, when the balance of power between an agent and an officer is effectively reversed and critical faculties are suspended, the potential is enhanced for deception, self-deception and failure.⁵⁷

Deciphering the enigma: the allied second front

In Roland Barthes’ scheme of analysis, a text’s use of feints and snares are features of what he calls its ‘hermeneutic code’.⁵⁸ This is one of five codes identified in his S/Z which, woven together in many different ways, endow a text with meaning. Importantly for us here, the hermeneutic code encompasses the ways in which a text plays with what Barthes calls an ‘enigma’,⁵⁹ or key unanswered question. In a murder mystery, for example, this might be ‘who is the murderer? – clearly, if the reader is to remain invested, this enigma must not give itself up easily. The hermeneutic code therefore comprises all the ways in which, instead, possible answers to this question are glimpsed, deferred, delayed and ‘held in suspense’ before finally finding a solution.⁶⁰ As Barthes says, in a work of fiction the story is continually unfolding from sentence to sentence, but at the same time ‘the hermeneutic code performs an opposite action: it must set up delays (obstacles, stoppages, deviations)’ thickening the enigma and, in so doing, progressively ensnaring the reader. ‘The more signs there are, the more truth will be obscured, the harder one will try to figure it out’, he argues. ‘The truth is thereby long desired and avoided, kept in a kind of pregnancy for its full term, a pregnancy whose end [will be] both liberating and catastrophic’.⁶¹

Viewing the Garbo traffic in the terms Barthes set out here sheds further light on the writerly techniques used to ensnare its readers. From almost the outset, the German conception of Pujol’s usefulness is clearly that he and his network can provide information on troop concentrations and other military matters which, when pieced together en masse, might be used to decipher Allied intentions for the Second Front. The central enigma here, the Allies’ order of battle, is something Pujol begins to allude to as early as 1942, as we have seen. On 29 May he reports that he has ‘told Agents of the necessity of investigating everything that might be connected with a possible attack’, subtly attempting to reposition them as investigators rather than observers and, hence, opening the possibility for reporting on a wide range of matters outside those specified in questionnaires. Through these agents (all of them fictional) the team pass hundreds of communications related to Allied preparations over the course of the next two years, judiciously mixing information with misinformation, as Howard and others note. On 19 October 1942, for example, Pujol reports via agent 3:

Kilmarnock and Fenwick. Ten heavy tanks, four light Matilda tanks, stationed in a field to the west of Fenwick camouflaged with green and brown nets. Two of the heavy tanks were unloaded from transport lorries and these later took the road to Ayr. Personnel of Canadian and British tank Corps stationed in both towns wear a shoulder badge 17/21. L.⁶²

Such observations were usually highly specific and concrete, and therefore do not read as evasions. However, they were also fragmentary and usually slyly unverifiable. Notably, the chickenfeed fed by the agents, necessary for their credibility and frequently plucked from open sources like newspapers, constituted some of the best intelligence the Abwehr had on what was occurring in Britain: such was the extent to which the British had tightened the screw.⁶³ It should be remembered that operation Bodyguard had not been authorised by this stage, so as temptingly as these jigsaw pieces were scattered for assessment by Abwehr officers, there was no ‘complete picture’ into which they could be assembled.

Meanwhile Pujol’s longer letters worked to ‘thicken the enigma’ of the Second Front in various other ways, reporting such things as ‘the confidence which exists in official circles’ about ‘an operation of great importance [which is] in embryo’.⁶⁴ Small-scale games of suspense are played throughout the correspondence, for example by promising significant revelations in the next letter, which might not arrive for some time or be lost. The ways in which these techniques of delay and deferral are used by the team are neatly exemplified by the case of Pujol’s (notional) Agent 5, a Venezuelan and brother to one of the earliest members of his network. Initially he is stationed in Scotland, reinforcing the theme of military preparations in the north of the UK. Since it was believed that Germany had no good sources of information on preparations for the Second Front in North America, however, it was decided that this agent should smuggle himself into Canada in order to form a sub-network there. Pujol initially proposed that he should set up a refuelling station for German submarines – a ruse with clear potential for exploitation by the Allies. Like all of Pujol’s proposals for concrete action in support of Germany’s war effort, however, this was rejected by the Felipe Stelle as impractical. Since it was inevitable that much of the capacity for the Allied counter-offensive would come from North America, nevertheless, Agent 5 was able to successfully position himself to monitor troop movements, shipping convoys and other preparations on the east coast. Strategically important intelligence was in prospect for the Abwehr here, but there were difficulties and delays in communications. These were most unfortunate given the urgency of the situation, but eventually Pujol was able to arrange for him to communicate directly using secret ink which was requested from Madrid. Unfortunately again, however, Agent 5 had a tendency to depression and low morale and was unable to use this consistently.

It is illuminating to observe how the head of the Felipe Stelle Karl Kühlenthal allowed himself to be strung along for months by this narrative, seemingly unaware of its farcical quality. His enthusiasm for the idea of having eyes in North America continued undimmed, even as he wrote,

I am still unable to give you a definite opinion about FIVE’s work since a considerable number of the letters he has sent us through you have not come out well due to the fact that he has not used the ink to perfection [although] from what we have been able to decipher it can be seen that he knows very well what is of interest to us.⁶⁵

It is easy to see here how Kühlenthal’s investment in the preservation of the network makes him a willing participant in his own deception – aided, of course, by a weak organisation that paid little heed to the matter of validation and assessment. Strikingly, he specifically instructs Pujol not to tell Agent 5 ‘anything about part of his letters not having come out, but tell him that we are very pleased and have confidence in counting on his collaboration’.⁶⁶ Perversely, the morale of this agent and his willingness to continue seems to supersede the value of gaining reliable intelligence from him. As with the Chislehurst caves, the maintenance of the Stelle’s key intelligence asset – Pujol’s network of agents – has become established as the one indispensable priority. The ambivalent position this placed Abwehr officers in was something the XX team were able to use to great advantage.

As D-Day approached, the team was focussed on the different elements of operation Bodyguard, the deception plans formulated by the LCS to cover the Normandy landings. The vast majority of their communications were in short form and transmitted by wireless, providing a multitude of ‘factual’ observations related to Allied preparations for the re-invasion of Europe. A large number of these concerned the formation of the First United States Army Group (FUSAG) in South East England. As Thaddeus Holt suggests, the existence of FUSAG and its threat to the Pas de Calais were primarily built up by simulated wireless traffic ‘representing the notional divisions training with their associated naval forces’⁶⁷ and double-agents’ reports, especially from Pujol’s network. It constituted, in Michael Howard’s words, a ‘huge and spectral force’, ‘which, had it existed, would have totalled some 150,000 men’.⁶⁸ As with the cores of all great deceptions it tapped into what the target most wanted, or in this case feared. Learning of this force was how the Abwehr justified its existence. It was therefore necessary for the XX Committee and the teams within B1A to coordinate this broad stream of information very carefully indeed, as well as to monitor its reception through decrypts.

At the same time, however, psychological games continued to be played with German officers through the medium of Pujol’s letters. In advance of the Tehran Conference in November 1943, for example, Pujol reported intelligence of ‘exceptionally high value’ concerning secret negotiations between the US, UK and Soviet Union. Particulars came courtesy of his (notional) girlfriend, a secretary at the Ministry of War, and the ‘wanton tricks I employed to get them out of her’.⁶⁹ Because of the ‘nervous strain’ created by his work, however, the same letter reported that he had been ordered by a doctor to take an immediate rest and therefore intended to cease communication for eight-to-ten days.The promise of illuminating the ‘enigma’ of the Second Front is, as before, immediately followed by deferral.

Since the preservation of the network had emerged so strongly as the priority for the Felipe Stelle, risk continued to be created for its survival. In an undated letter from early 1944, Pujol was forced to report a new ‘matter of life and death for the service’.⁷⁰ In a pair of linked feints, the team introduced news of two serious betrayals. One of Pujol’s most established unconscious collaborators, another fictional creation known as ‘the Courier’, had been used to swap letters and packages with the Abwehr since July 1941, using his position with an international airline to evade the censors. ‘By means which were never disclosed’,⁷¹ he had now discovered that a cover address used by Pujol was controlled by German Intelligence and had therefore concluded that Pujol must be a Nazi spy, rather than a Spanish refugee as he had claimed. Pujol’s account of the ensuing blackmail scene was written beautifully and in a highly engaging fashion by the XX team. Having demanded £2,000, the Courier himself was then used to deliver the second feint: the Lisbon cover address had been blown to the British Embassy by an (unknown) German intelligence officer in exchange for money.

As with previous feints, risk is created for the Abwehr on several levels here. Firstly, the survival of the network now teeters on a knife-edge. Secondly, their own organisation is compromised since an unidentified officer in a neighbouring country has been selling information to the enemy. Thirdly, Pujol’s own irascible temperament renders his own reactions a matter of concern:

I had already warned you that this man was a dangerous individual since he was mixed up in illegal business … but my surprise was greater when I realized that a man employed in an official job was capable of selling himself to another country purely for money without having any idealistic motive for his work. I must say that for a moment I felt so disgusted that … I would have cut the conversation short telling him that not only would I denounce him for libel but also for blackmail.⁷²

Once again, blame is directed at the Abwehr, who had wanted to continue with the Courier despite Pujol’s prior warnings, and who now bear responsibility for compromising the network because of their own poor internal security. In the circumstances Pujol concludes that ‘it is probably better that I should escape from England and not take any more risks’. In an amusing sideswipe at case officers as men of the desk rather than men of action, he adds, ‘put yourself only for a moment in my place and I believe that just thinking about it you will sweat ink’.⁷³

Having established a relationship of dependency, putting the network at risk in different ways was therefore now a tried-and-tested method for exerting psychological control over the Abwehr. At the same time, moreover, it also provided opportunities for the XX team to monitor their own success through the responses their letters provoked. Kühlenthal’s reply dated 14 February 1944 illustrates the incredible transformation they had managed to bring about in Pujol’s relationship with the Felipe Stelle over the previous two years. Beginning with an apology for his previous ‘cut down’ communications, Kühlenthal expresses a desire to communicate ‘more fully and more personally’ in future. Reprimands for poor performance are nowhere to be seen and there are no tersely delivered orders. Indeed, his mode of address to Pujol is now almost obsequious:

Your outlook shows such a comprehension of all our ideas that I am proud to be able to rely on a companion and collaborator such as yourself … . I want you to know that we are constantly occupied with all the questions relating to your organization [but] I find everything so perfect that there is practically nothing I can say with regard to the measures you have taken [in finally acceding to the Courier’s blackmail]. On all questions of organization you are the one best fitted to decide what should be done.⁷⁴

At this stage, it seems, nothing Pujol suggests can be taken amiss. He hatches a ludicrous scheme for hiding critical documents in a hen house: Kühlenthal find this ‘magnificent’. ‘We are now entering into the most decisive phase of the war’, he tells Pujol, but ‘I am certain that your present organization will make it possible to obtain the most essential details for the guidance of our Command’.⁷⁵ In a further radical change, the Abwehr are now very interested indeed in overheard conversations. Pujol’s girlfriend at the Ministry of War and an indiscreet friend at the Ministry of Information are now ‘your magnificent contacts’ who may reveal important details about Allied intentions.

Through a combination of short-form wireless messaging and the more expansive style used in these letters, then, it is clear that Pujol’s network is now fulfilling two functions for German intelligence. In the first case, it is delivering large quantities of highly detailed observations on military logistics and preparations. In the second case, Pujol has now achieved the status of a ‘comrade’ whose position in the enemy camp enables him to provide better appreciations and strategic analysis than his colleagues in Madrid can. Wireless communications allow the XX team to continue scattering multiple fragments of information for the Germans to fit together. At the same time, however, Pujol’s letters give them the means to make sure that enemy intelligence assemble the ‘complete picture’ as intended. It is the combination of these methods, I argue, that allowed the XX system to make such a decisive contribution to the Bodyguard deceptions. There was indeed a carefully orchestrated campaign of misinformation as Howard and Andrew suggest but, crucially, this was coupled with a sustained programme of psychological manipulation targeted at German intelligence officers by the writers of the XX system.

Considering the correspondence in its entirety, their success in controlling and exploiting their readers’ responses is strikingly evident. By creating various crises for the network, the team were able to escape the Abwehr’s preferred question-and-answer format as early as 1942 and to introduce a much more expansive and elastic mode of communication. At first this met with resistance, as we have seen. In due course as Harris says, however, officers were ‘persuaded to accept GARBO’s verbose style until they, themselves became infected by it’. Over a period of two years from the Abwehr’s ‘first telegraphic message in secret writing, consisting of a few lines, they were worked up to the climax of sending us no less than 25 foolscap pages of secret text in one letter’.⁷⁶ Meanwhile, such was their belief in Pujol’s effectiveness that, as decrypts show, observation reports submitted by his network were transmitted faithfully and without additional appreciations to Berlin from August 1943 until D-Day and beyond.⁷⁷

Conclusion

The great deceptions of the Second World War have received significant historical scrutiny over the decades since they emerged into the public domain. Of obvious intrinsic interest to observers of the war, their impact on the course of operations and strategy have been widely considered by both official academic studies and more recent popular writers, notably Howard and Macintyre. They have also been extensively written about by some of the practitioners who imagined, designed, and implemented the secret operations that have now become household names. Some, such as Dudley Clarke, offered their reflections on matters of craft and their practice. Interwoven into the raft of work in this area is an appreciation, both explicit and implicit, of the extraordinary creativity that strategic deception requires. The mathematicians like Alan Turing; the conjurers like Jasper Maskelyne; many of the talents of Churchill’s wizards have since been recognised. But a crucial element of the entire enterprise, the way in which writers worked to manipulate and ventriloquise their readers in the German intelligence machine, has hitherto received very little attention.

As the Allies moved to the offensive in the latter part of the war, however, a very heavy burden fell on the traffic of double-agents: that is, on writers working within the XX system and the correspondence they sustained with German intelligence. Unprecedented successes in codebreaking enabled them to be carefully attuned to the weaknesses of the German intelligence machinery and helped create the space for deceptions great and minor by careful and deliberate management of their readers. The analysis of their techniques provided here supplements other work which has considered elements of Allied deception from a psychological perspective, including Timothy Smith’s Overlord/Bodyguard. I do not dispute Howard’s judgement that the core enablers of strategic deception were derived through intelligence, which provided insight into the target’s communications and ensured security in Britain, thus rendering core elements of Germany’s machine both transparent, to a degree, and largely blind. Had Germany had better intelligence and counterintelligence it is unlikely that they would have fallen victim to deceptions so frequently. That they did not reflected both the nature of Hitler’s regime and his management of it, which was chaotic, encouraging conflict rather than cooperation, and of intrinsic weaknesses in the Abwehr itself. The weakness was partly structural and organisational, particularly in assessment, unwillingness to deliver bad news, and poor management. But it was also clearly linked to the ambivalence of many Abwehr officers about the Third Reich and the direction in which it was taking the war and Germany. Britain’s insight into these weaknesses offered writers working in deception a thread on which to pull.

Understanding the working practices of Abwehr officers who ran controlled sources, the XX committee found themselves in favourable conditions to exploit this knowledge by establishing a relationship of stress and dependency which allowed British intelligence to effectively penetrate the Abwehr. The survey of the extremely voluminous body of messages between Juan Pujol and his handlers offered here illustrates the careful and deliberate way they worked to alter the balance of power between officer and agent, turning the directed into the director. Using terminology from Roland Barthes, I have sought to reveal some of the key mechanics of this exchange, which have previously gone unexamined. Key among these are the ways in which Abwehr officers were set to work by the XX team, how stress and risk were created for them, and how their relationship with the German High Command was compromised. Ultra decrypts gave the team an enormous strategic advantage over their opponents, certainly. Nevertheless, the degree of control established over German intelligence officers by writers in British intelligence remains extraordinary. Even as Allied troops were establishing themselves on the Normandy coast on 7 June 1944, senior officer Karl Kühlenthal wrote to Pujol, ‘I wish to stress in the clearest terms that your work over the last few weeks has made it possible for our command to be completely forewarned and prepared’.⁷⁸ Two and a half months later, four days before the liberation of Paris, belief in the elaborate fiction woven by the XX writers remained unshakeable, both for him and his superiors:

Message for J. Headquarters entrust me with the mission which I fulfil with the utmost satisfaction of again expressing to you our special recognition for the results which have been achieved by you and your organization. They also make reference in this connection to the information which you have supplied to us since the invasion of France, news which has been of utmost value to them. Please make this recognition known fully to your collaborators.⁷⁹

The irony implicit in this message of praise and thanks from the highest level will not have been lost on Pujol and his fellow writers within MI5. The same cannot be said for his network of collaborators who had worked so long and hard to support Germany’s war effort, however, because they were entirely fictional.

Acknowledgements

With thanks to Huw Dylan for his invaluable contributions on the culture of the Abwehr and the history of German intelligence.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

Research for this article was supported by the Arts & Humanities Research Council (Grant AH/V001000/1)

Notes on contributors

Jago Morrison

Jago Morrison is Professor of Modern Literature at Brunel University and Principal Investigator for the project Writers in British Intelligence: The Secret State and the Public Sphere. Jago Morrison is Professor of English at Brunel University, with a special interest in literature and security. He is Principal Investigator on the AHRC-funded project Writers in British Intelligence: The Secret State and the Public Sphere [AH/V001000/1]. Recent publications have explored themes of radicalisation, terrorism and the changing security paradigm from the end of the Cold War to 9/11. Morrison has also worked extensively on Nigerian Writing and its mediation of a changing security environment post-independence. He edits the blog WritersinIntelligence.org

Notes

1. Brown, Bodyguard of Lies, 32.

2. Howard, Strategic Deception in the Second World War, x.

3. Masterman, The Double-Cross System, 14.

4. Ibid., 71.

5. TNA, KV 2/63, incoming no. 3, September 25, 1941.

6. TNA, KV 2/64, incoming no. 23, September 25, 1942, in response to outgoing no. 63, July 3, 1942.

7. TNA, KV 2/64, incoming letter no 28, August 25, 1942.

8. TNA, KV 2/86, incoming message May 19, 1944 (unnumbered).

9. TNA, KV 2/69, incoming letter August 31, 1944 (unnumbered).

10. A-force is discussed in several sources, including Bendeck, ‘A’ Force.

11. Rankin and Seamark, Churchill’s Wizards.

12. Eliot, “Tradition and the Individual Talent”, 53.

13. Greene, Grand Canary. By A. J. Cronin. Durbar. By Denis Kincaid. Mrs. Van Kleek. By Elinor Mordaunt. The Enemy at the Gate. By A. t’Serstevens. An Oxford Tragedy. By J. C. Masterman (Book Review).

14. New Novels.

15. Tuohy, The Secret Corps, 283–9. Thanks to Alan Burton for this reference.

16. Macintyre, Operation Mincemeat, 116.

17. Smyth, Deathly Deception, 36.

18. Macintyre, Operation Mincemeat, 119.

19. Andrew, The Defence of the Realm, 287.

20. Forster, Aspects of the Novel.

21. TNA, KV 2/68, incoming letter (unnumbered) April 3, 1944.

22. Harris, Garbo, 70.

23. Popov, Spy/Counter-Spy, 233.

24. Masterman, The Double-Cross System, 32.

25. Ibid.

26. Wheatley, The Time Has Come, 696.

27. Masterman, The Double-Cross System, 14.

28. Ibid., 16–17. Quotation from Thomas Gray ‘Elegy Written in a Country Churchyard’ (1751).

29. CIA ERR, W. R. E., “The German Intelligence Service and the War”, (1945), 1 < https://www.cia.gov/readingroom/docs/CIA-RDP78-03362A002500070002-3.pdf >

30. Mowry, Cryptologic Aspects,1.

31. Ibid, 1–2.

32. See Murphy, Security and Special Operations,148–169. The parallels with later British operations are interesting, particularly Britain’s reluctance to conclude that their operations was penetrated and their radio operator working under duress.

33. Jeffery, MI6, 380–382.

34. Claims in Mure, Masters of Deception, for example, 165–166.

35. Guensberg, “The Abwehr Myth”, 39.

36. CIA ERR, SCF. 52/50 ‘The German Intelligence Service and the War’, 1 December 1945, 2. < https://www.cia.gov/readingroom/docs/CIA-RDP78-03362A002500070002-3.pdf >

37. Ibid, 3.

38. See https://www.jewishvirtuallibrary.org/the-abwehr

39. In a more contemporary context, see Joseph Hatfield, ‘Intelligence under Democracy and Authoritarianism: a Philosophical Analysis’, Intelligence and National Security 37 (2022): 903–919 and Huw Dylan and Thomas Maguire, ‘The Ukraine War: A Public Crucible for Secret Intelligence’, Survival 64, no. 5 (2022): 33–74.

40. “The German Intelligence Service and the War”, 7.

41. Howard, Strategic Deception in the Second World War, x.

42. Ibid.

43. Andrew, The Defence of the Realm, 255.

44. TNA, KV 2/63, incoming letter no. 3, September 25, 1941.

45. Harris, Garbo, 69.

46. Barthes, S/Z, 32.

47. TNA, KV 2/64, outgoing letter August 28, 1942 (unnumbered).

48. Ibid.

49. TNA, KV 2/64, incoming letter October 14, 1942 (unnumbered).

50. Harris, Garbo, 115–16.

51. TNA, KV 2/65, outgoing letter no. 189, April 25, 1943.

52. TNA, KV 2/66, outgoing letter no. 200, June 3, 1943.

53. Ibid.

54. Ibid.

55. TNA, KV 2/68, incoming letter February 28, 1944 (unnumbered).

56. TNA, KV 2/66, outgoing letter no. 219, August 2, 1943.

57. A more contemporary example is provided by the “Curveball” case in 1999, in which an Iraqi defector claimed to provide information on weapons of mass destruction in Iraq. Enormous political capital was invested in his claims both in the US and UK, before they were finally debunked in 2007.

58. Barthes, S/Z, 17.

59. Ibid.

60. Ibid., 19.

61. Ibid., 62.

62. TNA KV 2/64, outgoing letter no. 101, October 19, 1942.

63. “The German Intelligence Service and the War”, 3.

64. Ibid.; outgoing letter no. 111, November 10, 1942.

65. TNA, KV 2/68, incoming letter April 3, 1944 (unnumbered).

66. Ibid.

67. Holt, The Deceivers, 588.

68. Howard, Strategic Deception in the Second World War, 121.

69. TNA, KV 2/67, outgoing letter no. 267, November 16, 1943.

70. TNA, KV 2/67, undated and unnumbered “Courier” letter. The letter is incorrectly marked as having been dispatched on 26 February 1944 – it had already received a reply twelve days earlier, so can roughly be dated to the beginning of February 1944.

71. Harris, Garbo, 294.

72. Ibid.

73. Ibid.

74. TNA, KV 2/68 incoming letter February 14, 1944 (unnumbered).

75. Ibid.

76. Harris, Garbo, 71.

77. TNA, KV 2/42 provides a detailed tabular comparison of GARBO messages and ISOS decrypts of associated communications between Madrid and Berlin. There is a striking correspondence between the two throughout.

78. TNA, KV 2/69, incoming message June 7, 1944.

79. Ibid, incoming message August 21, 1944.