by Nigel West
reports regarding invasion situation early on 9 June (time of report the night of 8 June): General opinion, according to statements of War Office spokesman to English and American journalists afternoon of 8 June (conference takes place thrice daily), is that conditions for Allied landing troops have improved. Impression shared in authoritative British military circles. According to statements by Harrison, an absolutely clear picture on the British side cannot yet be given as the critical period for the invading troops is only beginning. Strength so far employed as described by him as considerable, greater than was originally intended. In his opinion and according to information from other sources a second main attack across the Channel directed against the Pas de Calais is to be expected … British public very optimistic. But views in political circles more cautious. In Conservative circles the danger of too heavy losses is continually emphasized, whereas the Labour Party and other Left-wing movements are very satisfied with the beginning of the invasion.
In a futile effort to authenticate JOSEPHINE or his sources, which purported to be in London, MI5 scrutinised every word of this message but never even identified the individual supposedly named ‘Harrison’ who had been cited. One explanation considered was that ‘Harrison’ might have been a corruption for Sir Arthur Harris, then Commander-in-Chief of Bomber Command.
On this occasion JOSEPHINE appeared, not unhelpfully, to have followed the Ops (B) line, leading some of the deception planners to speculate that perhaps Krämer was drawing some of his material from assessments drafted in Berlin. However, in February 1944 he had submitted a report claiming that Allied plans for D-Day had been delayed until June because of a high-level row that had erupted between ‘the Allied Expeditionary Air Force’ and the British and American strategic bomber forces. This tale was untrue, but Krämer’s date was uncomfortably accurate.
Uncontrolled agents such as Fidrmuc, in their own ways, posed a threat to BODYGUARD, not least because of the sheer volume of their output, and various options were discussed in London, including recruitment and even elimination. Detailed analysis of the OSTRO traffic suggested that his reporting was nothing more than, as J.C. Masterman concluded, ‘rumour aided by invention and surmise’, but in the end the XX Committee recommended that he be left to continue his mischief in the expectation that his contributions would simply add to the enemy’s confusion.
The same suspicion had been raised by a Luftwaffe analyst in Paris, who noticed a curious similarity in the reporting of OSTRO and Krämer’s sources HEKTOR and JOSEPHINE, and recommended a thorough investigation, which was conducted by Leutnant Count Heynco von Posadowsky-Wehner, son of Silvius Graf von Posadowsky-Wehner, and a member of an aristocratic family that included a former interior minister in the Weimar Republic.
Krämer always exercised great care not to even hint at the identity of his sources, and little was known about HEKTOR, who seemed to be a senior Foreign Office figure, although on 27 October 1943 Guy Liddell mentioned in his diary that his messages:
deal with Ministry of Aircraft Production subjects, although the information so far received is totally inaccurate. It is of a somewhat high-level kind referring to conversations between Air Marshals Portal and Harris in Sir Stafford Cripps’ ante-room.
Almost simultaneously, MI5, SIS and the Luftwaffe set out to verify Krämer’s alleged sources, and applied rather different methodologies. Initially the Germans set HEKTOR and JOSEPHINE some test questions, which they were unable to answer satisfactorily, so Posadowsky-Wehner undertook a detailed analysis of JOSEPHINE’s reporting on FUSAG, which was completed in January 1945:
1. FUSAG’s existence was first notified on 9th January 1944, and was regularly mentioned with certainty up to 17th June. In the spring of 1944 it was increasingly reported by agents and by the press.
2 In assessing JOSEPHINE’s reports, the purport of all reports on FUSAG may be noted as follows:
‘(a) The Allied Armies under FUSAG’s command kept on changing. Finally it was stated that one English and one Allied airborne army were under the command of FUSAG, i.e. of an American Army Group.
(b) Changes in the command of FUSAG were reported quite as frequently as changes in the formations subordinate to it.
(c) When the invasion began, FUSAG was generally put down as being a second great group of forces, for a second landing operation north of the first.
(d) After the continued withdrawal of divisions from the armies under FUSAG’s command, to reinforce the invasion forces, reports were also received as to movements of FUSAG into Central and Northern England, in connection with landing operations against Jutland and Southern Norway, and later against the Heligoland Bight.
3. It may be seen from (2) that FUSAG was certainly used by the enemy as an ‘army in being’.
Technical adviser’s personal impression is that FUSAG existed simply for this purpose.
4. During the second half of the year, reports from the agent JOSEPHINE conform to the enemy’s decoy activities. Until the middle of October 1944, they did much to uphold the ‘fiction’ of FUSAG as a strong group of forces intended for further landings.
The first reports came in very late – according to records on file, not until 6th August. Like other agents JOSEPHINE reported:
(a) FUSAG was originally to have been used for a second large-scale landing operation. As the timing of the Normandy invasion went completely wrong, it was decided not to undertake this second landing, which had been planned for the end of June; and formations from FUSAG were steadily transferred to France.
(b) As there was no longer any practical likelihood that FUSAG would be employed for a second landing, it was now mentioned in connection with a landing operation against Southern Norway and Denmark. ‘Increased troop transports from Southern and Central England to Northern England are being associated with movements of parts of FUSAG. Operations in Northern and Central Norway, starting from England, are not expected, but landings in Jutland and S. Norway.’
Replying to further enquiry he stated:
‘I was expecting your further enquiry (!), as my own suspicions had already been aroused at once. All sources however confirm that FUSAG formations have been stationed as far as the Humber. In addition there are individual reports about troop transports to Northern England/Scotland. As reported, these were not observed previously. Resumption of Swedish air traffic to England refused, although conceded at end of August. It is quite clear that either a large-scale decoy manoeuvre is planned, to cover the employment of FUSAG in Belgium–Holland–Heligoland Bight, or that an operation against Denmark is actually intended.’
(c) Later on, the idea of connecting FUSAG with the movements of troops to Northern England was withdrawn, but the increasingly doubtful FUSAG was reported to be destined for a large-scale landing in the Heligoland Bight: ‘FUSAG continues in Eastern England as far as Humber. Formations in Northern England and Scotland do not belong to FUSAG. Employment of FUSAG in Eastern Holland and Heligoland Bight after strong airborne landing has been carried out in Eastern and Northern Holland.’
5. In conclusion, it must be pointed out that JOSEPHINE was thus a participant in the enemy’s decoy plans, which were aimed at holding down strong German forces for as long as possible at various points from Norway to France.’
Thus, although Posadowsky-Wehner realised FUSAG’s true purpose as a decoy, he never understood that the entire formation was a sham. Nevertheless, he condemned JOSEPHINE as being in the hands of the enemy, a view rejected by his superior, Colonel von Dewitz. Meanwhile, MI5 assigned Anthony Blunt to track down JOSEPHINE and HEKTOR in London, and after lengthy enquiries and surveillance on the Swedish air attaché Eric Servell, he concluded that Krämer was relying on information from a conscious, or unconscious source within the Swedish embassy. Accordingly, the Swedish naval attaché, Count Johan Oxenstierna, was quietly withdrawn from his post and replaced. SIS’s approach was through the Section V officer in Stockholm, Peter Falk, who recruited Krämer’s maid to obt
ain enough evidence to suggest that there was no such individual source as either JOSEPHINE or HEKTOR, and that the code names were simply generic covers. However, Falk concluded that Krämer was not an outright fabricator, and had managed to gain access to the London embassy dispatches through a secretary in the Foreign Ministry.
An investigation into Krämer conducted by MI5 revealed that just before war was declared he had been lodging in London at 9 Sussex Gardens, and had made a hurried departure.
Various options were considered for neutralising Krämer, who was thought to have acted as a concentration camp commandant before his transfer to Stockholm, but Berlin’s growing scepticism of his reliability made plans to bribe or blackmail him redundant.
Another suspected hoaxer, who exercised considerable influence in his role on behalf of Ast Berlin in Lisbon, was Juan Brandes, a 24-year-old adventurer entrusted with his own wireless circuit to Berlin that, according to TRICYCLE, who partied with him at Estoril, was of partly Jewish heritage and a committed Anglophile.11 However, an analysis of his ISOS traffic revealed that he claimed to be supervising a large spy ring of Swiss diplomats that stretched from the United States and South America to the Middle East:
A number of them appear to be Jewish, or at least to be based in Switzerland. The particulars available about them on secret sources suggest that they are, at least ostensibly, of a higher grade than most Abwehr agents. The principal ones of whom we have knowledge are:
INSTERBURG, tentatively identified by SIS as Paul Rene Keller of a Swiss transport firm in Lisbon.
AESCULAP, who was said to have been in Cairo in July 1942, and subsequently to have left for de Gaulle’s headquarters at Fort Lamy. He was later reported in Portuguese Guinea and last heard of in Freetown in January 1943.
BUNSEN, probably a Swiss, who was in New York in October when he returned to Switzerland via London. In December of the same year he was said to be about to return to England. More recent evidence shows that he has now returned to Switzerland after having apparently been in London.
The IRA Man. Nothing is known of this man except that he is an Irishman member of the IRA, who was intended in November 1942 for Estonia, which may, in that context, have been a cover name for England.
PETTERMANN. Another agent, possibly connected in some way with AESCULAP, who was intended for this country in January 1941. We know nothing of him except that according to Brandes he was full of good prospects.
BARINKI d’ARNOUX @ DANNEMANN. A Frenchman who was in Vichy in March 1943, awaiting a visa for the USA where he was to take up the position of secretary to the representative of the Swiss Mercantile Marine in New York.
No actual reports from any of Brandes’ agents have so far appeared on secret sources. There is not the same positive evidence, therefore, as there is with Fidrmuc, that the organisation which he claims to run is fictitious. There are, nevertheless, reasonable grounds for supposing that this may in fact be the case. The following facts may be cited:
Endeavours to identity BUNSEN and AESCALAP were attended by wholly negative results, though in each case the facts available should have been sufficient to produce, if not an identification, at least a short list of possible candidates.
Similar attempts to identify DANNEMANN failed completely, even though we were informed of his real name. According to the FBI, no application was made in Vichy at the material time for a US visa by anyone named d’Arnoux, nor did any alternative candidates suggest themselves.
According to both TRICYCLE and ARTIST, Brandes is not only anti-Nazi but positively pro-British and has, according to his own statement, been a party to at least one deliberate fraud on the Abwehr.
The existence of a spy named Bunsen caused alarm at MI5, where Herbert Hart kept a watch for references to him in ISOS, and on 24 January Guy Liddell noted:
Herbert Hart disclosed that ISOS indicated that the unidentified German agent Bunsen, thought to be a Swiss, had left this country for Lisbon and Switzerland somewhere round the middle of January, where contact was to be made with an Abwehr representative from Berlin. This is rather disconcerting since it means that we have had a German agent here since December 1942. We should, however, be able to pinpoint him on the present evidence.
Five days later, Hart had more to report to Liddell:
Herbert Hart is still trying to follow up the Bunsen case. There is now some suggestion that it may be purely notional, since Hans Brandes, the controlling officer is putting up a notional agent to headquarters in order that he can stay in Lisbon, and has also told TRICYCLE not to bring back any good information from this country as it will only prolong the war.
By 2 February, Hart had come to suspect that Bunsen did not exist, and reported his opinion to Liddell:
At the weekly meeting Herbert Hart mentioned that he had not been able to make any further progress with the case of Bunsen. He had examined the cases of all Swedes but none of them fitted the facts. He was beginning to think that Bunsen might well be notional.
As an accredited Abwehr officer under the protection of Admiral Canaris, and exempted from military service through alleged ill health, Brandes posed a rather different problem to the other uncontrolled sources. When Jebsen began to contemplate switching sides he started to cultivate Brandes, presumably in an effort to learn more about his network so he had more to trade in his defection negotiations with the Allies. Naturally, Brandes took fright at Jebsen’s unwelcome inquisitiveness and denounced him to Berlin, where the seeds of doubt, and probably jealousy, had already been sown. Ostensibly, Jebsen’s relationship with Popov was one of handler and agent, but their cover was the development of an escape route along which evading flyers and others could be smuggled over several frontiers to reach Lisbon and ultimately England. This was the method that had allowed several of Popov’s nominees, including his brother Ivo, and his friends Eugn Sostaric and Frano de Bona, to join him in London and work as German spies. In reality, of course, all three had acted as double agents (code-named DREADNOUGHT, METEOR and FREAK, respectively), but the arrangement gave Popov a pretext for his travel to Lisbon, and for his meetings with Jebsen.
Brandes and Jebsen became close friends, but did not entirely trust each other. At a meeting with SIS at the end of January 1944, Jebsen claimed that he had seen Brandes and his assistant, Werner Duessen, poring over newly arrived English newspapers to assemble what they termed ‘constructive fantasy’.
Brandes has claimed to have four agents, one in England, one in the United States, and two in South America, and is supposed to run these agents with assistance from either the Swiss Secret Service or from Swiss diplomats.
Furthermore, Jebsen expressed the view that while Brandes pretended he received his information in letters from Switzerland containing secret writing, the correspondence was simply a cover and did not really conceal anything. This was interpreted in London as tantamount to an admission that Brandes was a fabricator, and demonstrated Jebsen’s value. Soon afterwards, on 3 February, he reported on a conversation with von Auenrode:
Karstoff told me that according to the latest reports which he has received there is a possibility that the landing in West Europe will not take place till next spring because failure cannot be risked on account of the American presidential elections. Karstoff believes these reports and has repeated them to Berlin. When Brandes returns from Berlin I shall know whether the Admiral believes them and has passed them on to the General Staff, because the Admiral will certainly discuss with him reports coming from Lisbon.
On 7 March, following Brandes’ return to Portugal, Jebsen had more to tell Tommy Robertson:
Brandes had told ARTIST that when he saw Canaris in Berlin Canaris had, in conversation, expressed his opinion that the Allies would not try and land in France where the defences were too strong but would seek to invade Germany through Norway and the Balkans, putting increased pressure on Turkey and Sweden. Canaris expected a German breakdown to follow very quickly if a successful landing was made at any f
resh point and if bombing continued.
Following a further rendezvous on 9 March, SIS reported that Jebsen ‘is certain that Brandes has no agents and makes up all his reports from newspapers, etc’.
At some point Jebsen learned, or guessed, that Popov was truly a double agent loyal to the Allies, and this realisation led him to his contact with SIS, which he declared, presumably in somewhat inexact terms, to Berlin. As these events coincided with the very public defection in Istanbul of his acquaintance Erich Vermehren, through his mother Petra, who was a journalist living in Lisbon, Jebsen’s links with the enemy, combined with Brandes’ denunciation, were enough to seal his fate and he was arrested by the SD at the end of April 1944. This drama was monitored remotely by MI5 through ISOS, raising the fear that under interrogation Jebsen would compromise Popov, but apparently this never happened, and the focus of the investigation was financial misconduct rather than outright treason.
Nevertheless, while Jebsen’s fate remained in the balance, pressure built in London to terminate TRICYCLE, or at least remove him from the FORTITUDE campaign in which he was becoming a central player, as indicated by a German demand on 18 May: ‘Where do you locate 21 Army Group? Which armies and divisions belong to it?’
As luck would have it, this was the moment that MI5 learned that FREAK had been jeopardised. A leak from within Draža Mihailović’s Chetnik movement had disclosed that de Bona was working for the Germans, and news of this tip had reached the Abwehr. Naturally, Berlin anticipated that British action would follow, even if de Bona had been appointed aide-de-camp to King Peter, and Popov confirmed this on 20 May in a letter written in secret ink: