Double Cross in Cairo

by Nigel West

  The decision to liquidate CHEESE was taken at a 30 Committee meeting held on 7 January when it was proposed that on around 10 February MISANTHROPE would relay a message from CHEESE reporting that he had failed to find the wireless buried at Chillabdariou Street in Athens, and then on 15 February announced that, ‘having got the wind up’ because of a warning from her sub-source ELIF, she was going to cease all activity for a month.

  The news that the CHEESE case was to be closed down by SIME on 17 February 1945 was an event that prompted MI5 to express the wish for the radio link to be maintained, as Guy Liddell wrote to Pope from London on 4 March.

  Alec Kellar and I have discussed this exhaustively; I, for my part, have talked it over with John Marriott (who is now head of the section formerly run by Colonel T. A. Robertson) and we are in agreement that it is in general worthwhile to continue running any double agents who are in wireless contact with the Germans as long as the latter are prepared to reply to our signals (that is to say, unless there is any very strong reason – such as, for example, the possibility of compromising other agents) for closing down other than the approaching end of hostilities.

  I am hoping to discuss with the department responsible for post-armistice security in Germany the question of possible future underground SD activities and, as soon as I have been able to acquire more information on this subject I will let you know. In the meantime the following points are perhaps worth your consideration.

  Firstly, it may be taken for granted that elements of the SD will go underground; among their objects in so doing will be the continuation of the use of agents for the purpose of rendering the task of the occupying Allies as difficult and uncomfortable as possible. In so doing they will probably endeavour to create dissention between the various Allies: it will be, of course, especially within the boundaries of the former Reich that they will attempt to achieve this aim, but it will no doubt suit their purpose if they can make trouble further afield (in passing it is worth noting that if KISS survives after the Armistice, he may appear to them to be especially suitable as a tool for any activities of this kind). Quite apart from making trouble of this kind, it also seems likely that the German underground, whatever form it may take, will be anxious to acquire information from the outside world – the more so now that its necessarily mole-like activities will cut it off from normal contact with the countries outside the former Reich. For such purpose, agents of the type of CHEESE, if they continue in being, might well be useful both (in appearance) to the Germans and (in fact) to us.

  It may, of course, be argued that any underground SD organisation will find it impossible in practice to maintain wireless communications owing to the danger of their being D/F’d by the occupying Allies. It seems to me however, that there are certain remote regions of Germany where, by nature of the terrain, it would be extremely difficult to locate a sufficiently mobile wireless station. There is also another side to this argument in that, by maintaining communication through CHEESE, or any similar agent, we can provide the Allied security authorities in Germany with a target for their D/F experts and thereby perhaps assist them to locate concealed German headquarters.

  It is proposed to enter into this whole question in greater detail when we have SIME’s opinion.

  As Pope later explained, the decision agreed by both the head of SIME and the Chairman of the 30 Committee allowed for the possibility that CHEESE might be revived, and he acknowledged that the Germans had continued to try and re-establish the link throughout March. He had arranged for the CHEESE/MISANTHROPE control station in Germany to be monitored after the traffic had been closed down on 10 February, and although no signals were reported in the period until 3 March, Pope suspected that there had been undetected transmissions, because, as he noted,

  I have had Mackenzie listening since 2 March. He has heard the Germans calling on 2, 5 and 7 March, but not since 7 March. I think it likely that the Germans will go on calling intermittently for some time.

  Pope went on to explain that

  our method of closing down the link was intended to allow for the possibility of reviving it should the necessity arise, and also, incidentally, to ensure that the Germans should go on calling us for some time. The 30 Committee have discussed your proposal to revive CHEESE at some length. The security considerations which prompted ‘A’ Force to press for the closing of the case are still valid and, in the opinion of Head of SIME and the 30 Committee, it is undesirable to revive it.

  CHAPTER NINE

  MAX AND MORITZ

  A further, unanticipated complication in the CHEESE case was the discovery, in early 1942, that an Abwehr source, codenamed MORITZ, was supplying information which appeared to originate in Cairo to the KO in Sofia. This remained a signals analysis problem for cryptanalysts in England until July 1942 when Johannes Eppler and Heinrich Sandstede were arrested and admitted to SIME, their interrogators, in July 1942 that they had been assigned those codenames for their wireless traffic, but they had never been able to achieve a single transmission. So who was MORITZ, and was there another enemy spy-ring in Egypt capable of contradicting CHEESE?

  The mystery surrounding MAX and MORITZ would last for years, but its origins are to be found in the ISOS traffic on the Sofia to Vienna circuit, with the call-signs SCHWERT and VERA, relaying information from the Middle East and, apparently, Cairo. During the period of interception, in 1942 and 1943, almost a thousand messages were intercepted and decrypted. Between December 1941 and March 1942 forty messages from MORITZ were read, and in August 1942 SIS’s Section V reviewed the texts and, having rejected the possibility that they were based on signals intelligence, judged them to have ‘a professional flavour, being up to date, terse, well-arranged and definite’, noting that the information came from ‘Syria through Iraq and Persia to Egypt and Libya’ but concluding that ‘there were no clues to the sources of the MORITZ reports.’ The content varied in quality from patently false to uncannily accurate. For example, on 23 October 1942 there was a signal with a completely authentic content, although the raid mentioned had not yet been released to the press:

  MORITZ reports – 100 British Lancaster bombers attacked Genoa port in northern Italy last night. Reports indicate accurate hits against warships and supplies at the port.

  On the following day there was another message in a similar vein;

  MORITZ reports – General Montgomery is leading a new British operation, Operation LIGHTFOOT near el Alamein, west of Alexandria. The Germans have begun to retreat westwards.

  At a particularly critical moment in the North African campaign, on 9 November 1942,

  MORITZ reports – Last night American forces invaded North Africa in Operation TORCH. The British launched Operation SUPERCHARGE from the east and advanced towards Tunis. The double offensives against the Germans from east and west will trap them and destroy Rommel’s entire Korps.

  The very next day, 10 November 1942,

  MORITZ reports – Warships crossed the Suez Canal from the south. They will join the marine forces in the Mediterranean Sea.

  Later the same month an inaccurate message dated 30 November 1942 stated:

  MORITZ reports – Yesterday and today American bombers attacked Bizerte, Tunis, from the air. Air raids against German targets in Bizerte will continue.

  A few days later, on 8 December 1942, there was an accurate text:

  MORITZ reports – Large British forces are making preparations for an offensive near Sirte Bay in Libya.

  These messages, which would include five purporting to give a daily commentary on the British attack on Tobruk, and suggesting that Rommel’s DAK was on the verge of collapse, prompted a review of the material to identify clues to the identity of the network or spy responsible for gathering the information.

  The most striking feature of the MORITZ reports was that they were at the same time detailed and for the most part so inaccurate that they seemed to be reasonably explained either as a deliberate attempt to deceive the Germans o
r as a concoction put out for mercenary reasons.

  The resulting analysis concluded that a transmitter in the Spanish embassy in Ankara codenamed ANKER, was in daily contact with SCHWERT, the Abstelle in Sofia, using a hand cipher. The exact same messages were then relayed from Sofia to VERA in Vienna on an Enigma channel at the same time each afternoon, except Sundays. Each of the transmissions consisted of four or five messages from MAX, and just one or two from MORITZ. VERA, the Vienna Abstelle headed by the Graf Rudolf Marogna-Redwitz, then passed the messages on to BURG, the Abwehr’s headquarters in Berlin. Within half an hour, BURG was circulating the reports to Fremde Heere Ost at the Boyen Fortress in Lötzen, eastern Prussia, and to the OKW headquarters in Rome. The circuit had opened in October 1940 and had broadcast a final signal on 13 February 1945.

  A lengthy study of the traffic conducted by the Radio Security Service (RSS) revealed that MAX and MORITZ were part of an Abwehr dienstelle headed by a certain Fritz Klatt, a name that turnd out to be an alias adopted by a Czech Jew, Richard Kauder, whose network of more than seventy agents, of whom around a dozen were Jews, accounted for much of the Sofia Abstelle’s activities. Kauder operated from a company that he owned, the Mittermeyer Import-Export Company, based at 55 Skoolev Street, not far from Otto Wagner’s KO at 57 Patriarch Aphtimey Street.

  The RSS study, conducted by Gilbert Ryle, posed a series of questions. How could Klatt’s network submit reports from Cairo to Sofia on the same day? Some of the messages referred to events that had taken place across the Middle East the previous day. One significant characteristic of the traffic was the frequent misspellings of names, and the use of Russian names, for example Galiopolia for Heliopolis. Similarly, fighter aircraft were described as isterbaitle. Broadly, the information from the Middle East did not pose a serious threat to the Allies, and much of it could have been gleaned by the Germans from other sources. The reporting was long on generalities but short on specifics, so the decision was taken to inform the Soviets of the existence of MAX and MORITZ, and the news was delivered at a meeting convened with Stalin in Moscow by the SIS representative, Cecil Barclay, on 19 April 1943, in the presence of the ambassador, Archie Clark Kerr. Curiously, the Soviets seemed disinterested, despite a further conference with the DMI, General Fyodor Kuznetsov, on 29 July, and the traffic continued, apparently unaffected, so the British concluded that the entire organisation was operating under the NKVD’s supervision and therefore was part of some elaborate deception campaign.

  That verdict had been reached after the most intensive scrutiny conducted by both MI5 and Section V which included a veracity check on forty-nine messages decrypted in June and July 1943, of which only five were thought to have any value. Thirty-three were shown to be useless, and eleven could not be subjected to any comparison.

  Research into MAX and MORITZ escalated after the war when captured Abwehr officers underwent interrogation by the Allies in an effort to identify the organisation’s sources. MAX turned out to be General Anton Turkul, a Ukrainian who had fought against the Bolsheviks in the Russian Civil War with the whites, and had later settled in comfortable exile in Paris where he had become an informant for Dick Ellis at the SIS station. When interviewed, Turkul revealed that the codenames MAX and MORITZ were not individual agents, but codenames for groups of geographically based networks, MAX being the Ukraine and Russia, and MORITZ being Turkey and the Middle East. He alleged that the names had been inspired by Wilhelm Busch’s illustrated rhyming tales, Max and Moritz: A Story of Seven Boyish Pranks, published in 1865, and admitted that he had worked for the NKVD for years, and that Kauder had quickly realised that his own organisation was actually sponsored by the Soviets.

  Kauder was freed from prison in Vienna by American troops in May 1945 and under interrogation he admitted that his entire organisation had been run by the NKVD, but he had decided, for reasons of self-preservation (because he was already in fear of the Gestapo), not to tell the Abwehr. He was later transferred to Camp King in Oberrursel where he was questioned by Klop Ustinov, and then in 1946 by Gilbert Ryle.

  Kauder’s candour helped solve the mystery of MAX and MORITZ and provided eloquent proof that if the NKVD was ruthless enough to accept hideous sacrifices to enhance the status of a valued double agent, the Abwehr across the Mediterranean region was inherently corrupt.

  CHAPTER TEN

  FINALE

  After the war Count Scirombo was interrogated, and he shed new light on the CHEESE case. He confirmed that Levi’s original recruitment had been conducted jointly by SIM and the Abwehr. Without realising the double agent’s true role, Scirombo confirmed that Levi had been sent to Egypt posing as a Jewish refugee seeking to escape religious persecution. He also said that Levi’s sub-agent had communicated first with a SIM wireless station at Forte Braschi, but control had later been switched to an exclusively German facility in Athens.

  Scirombo claimed that Levi had raised suspicions when it was learned that he had visited the British delegation in Belgrade while on his way to Turkey. The decision then had been taken to question Levi through an informant who happened to be one of his friends. This had been the encounter with the Italian air force officer, Captain Alessi, and Scirombo said that Levi had

  talked very freely to the informer and told him all the details of the mission he had undertaken on behalf of SIM and the German Intelligence Service. He added that he was expecting shortly to receive instructions to return to Egypt, but had no intention of doing so.

  The Germans endeavoured to bring pressure to bear on SIM to obtain Levi’s release. They pointed out that Levi had accomplished a very successful mission and, in an effort to convince [the] subject, showed him copies of the messages which were being received from the agent in Cairo. Subject states that the quality of the traffic was excellent and the information given on Allied troop movements were checked and found to be accurate. The agent quoted as one of the principal sources of information was an Allied NCO working in a military headquarters situated near the Italian consulate in Cairo. Subject also remembers seeing messages requesting funds. In spite of all this SIM refused to adhere to the repeated requests from the German Intelligence Service for Levi’s release and the latter remained in a concentration camp until the Armistice when subject supposes he was set free by the Germans. (Note: subject has no idea of his present whereabouts.) In December 1943, when subject was evacuated from Athens by the Germans, the wireless contact established by Levi in Cairo was still functioning.

  Another loose end was represented by Johannes Eppler and Heinrich Sandstede who, in February 1943 had come to the end of their usefulness. Accordingly, SIME asked Herbert Hart, of MI5’s B1(b) section, whether there was any point in sending them to London.

  In our view all the intelligence these characters have to give has probably been extracted from them already, and even if it has not we have no better facilities for extracting it than Mid East, with their knowledge of local conditions, background, etc. We therefore have no objection to their being dealt with locally.

  Consequently, Eppler and his partner were detained as prisoners of war in Egypt until March 1946 when Eppler and Sandstede were repatriated to Germany. The first details of their story emerged in 1958 when the war correspondent Leonard Mosley published The Cat and the Mouse, a highly inaccurate account in which he claimed that Eppler had achieved radio contact with Rommel’s headquarters, and had employed a book code based on Daphne du Maurier’s Rebecca. In fact, of course, the code was based on The Unwarranted Death.

  Another version of Eppler’s arrest emerged in 1965 when Colonel A. W. Sansom, who had served with the Field Security Wing in Cairo in 1942, published I Spied Spies and claimed that he had masterminded the raid on the German agents. According to him, the first clue that a spy had reached Egypt was when New Zealand troops had overrun the Bir Hachim intercept site and found two PoWs, who did not speak English, in possession of a copy of Rebecca that had been sold in a Lisbon shop.

  I sent a cable to Lon
don and asked them to investigate. The answer gave me the first, as yet insignificant, lead on Eppler. At least now I knew for certain that this must have been the code book of an agent and that the man concerned was certainly in this country. It wasn’t much – but it was a start.

  In Eppler’s own account, Rommel ruft Kairo which appeared first in 1960, with an English language edition, Operation Condor, released in 1977, he provided a very different tale, diametrically contradicting his CSDIC statements. He claimed that he had been recruited by the Abwehr in Beirut in May 1937, and since that date had undertaken numerous missions across eastern Europe and the Middle East, acted as an interpreter for Adolf Hitler, and operated as a spy in Alexandria long before his final participation in Operation CONDOR. Furthermore, he and his companion had established good radio contact with Rommel’s headquarters, and his belly-dancing girlfriend Hekmat Fahmy had been a very active co-conspirator.