by Nigel West
A post-war investigation into this appalling breach of security, based on interviews with an Amt VI interpreter, Maria Molkenteller2 and an Amt VI officer, SS-Obersturmführer Otto-Ernst Schuddekopf, revealed that the ambassador’s residence did not possess a safe, so Sir Hughe had stored his secret papers in a locked black tin box in his bedroom, but had left the key lying around so at least three members of his personal staff had access to it. Worse, the ambassador had persisted in this behaviour long after he had been warned about it in 1942. The first clue to the existence of a spy code-named CICERO, linked to Moyzisch, designated ‘6981’ appeared in the Prague–Istanbul ISOS channel dated 29 November 1943:
From 6981. Extracts from CICERO material on China can be passed on to 8027 if origin is in no way apparent. 8027.
According to Molkenteller, who had translated the CICERO product at the SD headquarters in Berlin, and was interrogated in London in 1945, it included between 130 and 150 secret Foreign Office telegrams.3 Dr Schüddekopf, a distinguished historian, confirmed that the SD was entirely satisfied by the documents’ authenticity, even though Moyzisch never identified Bazna’s woman accomplice.
A Roman Catholic, Moyzisch was partly of Jewish ancestry, although he only became aware of his father’s heritage in 1939 when he was denounced anonymously, and in peacetime had been an Austrian journalist. He had dubbed his source CICERO and passed the material, which included an important telegram from 21 December 1943, restricted to recipients with BIGOT clearances, to Berlin where Hitler’s chief of operations, General Alfred Jodl, noted in his diary on 10 February, under the heading ‘results from CICERO: OVERLORD = Major invasion from Britain’.
The telegram, No. 1751 addressed to the ambassador, was in part a copy of a Chiefs of Staff policy document sent to General Eisenhower that mentioned a British intention ‘to maintain a threat to the Germans from the eastern Mediterranean until OVERLORD is launched’.
OPERATION ‘OVERLORD’
(a) This operation will be the primary United States-British ground and air effort against the Axis in Europe. (Target date, May 1, 1944.) After securing adequate Channel ports, exploitation will be directed towards securing areas that will facilitate both ground and air operations against the enemy. Following the establishment of strong Allied forces in France, operations designed to strike at the heart of Germany and to destroy her military forces will be undertaken. (b) Balanced ground and air force to be built up for OVERLORD and there will be continuous planning for and maintenance of those forces available in the United Kingdom in readiness to take advantage of any situation permitting an opportunistic cross-Channel move into France. (c) As between Operation OVERLORD and operations in the Mediterranean, where there is a shortage of resources available, resources will be distributed and employed with the main object of ensuring the success of OVERLORD. Operations in the Mediterranean theatre will be carried out with the forces allotted at TRIDENT, except in so far as these may be varied by decision of the Combined Chiefs of Staff. We have approved the outline plan of General Morgan for Operation OVERLORD, and have authorised him to proceed with the detailed planning and with full preparations.
When the German ambassador to Turkey, Franz von Papen, read this item on 6 January 1944, he immediately interpreted it to mean that OVERLORD represented a major action to be launched from Britain. His principal mission was to either maintain Turkey’s neutrality, or to persuade the government to join the Axis, so he was keenly interested in CICERO’s material, and was easily persuaded of its authenticity.
Evidently Hitler’s chief of operations, General Alfred Jodl, had reached the same conclusion. Von Papen, a professional diplomat who had engaged in espionage from the German embassy in Washington, D.C. during the First World War, also opined that the compromised text implied a classic diversion, that a British threat to the Balkans was intended to draw attention, and doubtless the enemy’s military assets too, into the region, while some other major initiative was launched elsewhere. While this verdict may not have disclosed any exact dates or targets, it did tip off the Axis to the existence of a very specific code word.
Bazna resigned his post at the end of February 1944, having taken fright at the unexpected appearance of security investigators at the chancery and residence, and cut his ties to Moyzisch after a final rendezvous in April, but the damage had been done. At his office in Berlin’s Birknerstrasse the Amt VI chief, Walter Schellenberg,4 also grasped the significance of OVERLORD. He issued a circular to all SD staff seeking details of any other references to the code word, and imposed a special search for additional references, particularly in any decrypts of Allied communications.
Born in Saarbrücken in 1910, Schellenberg had qualified as a lawyer and joined the Nazi Party in 1933. Two years later he was recruited into the SD and in November 1939 was entrusted by his chief, Reinhard Heydrich, with a delicate assignment, the abduction of two British intelligence officers at the Venlo border crossing into Holland, a highly successful mission for which he received the Iron Cross (first class) and promotion to Amt IV and counter-intelligence operations conducted with Abwehr III. He would also liaise closely, and develop personal friendships, with his Swiss counterpart, Roger Masson; the chief of the Turkish intelligence service, Mehmet Naci Perkel; and the head of the Swedish Security Police, Martin Lundquist. Suave, sophisticated and cosmopolitan, Schellenberg was the consummate counter-intelligence professional and he would play one of the key roles in the OVERLORD saga.
On 12 February 1944 Hitler signed an order to create a unified intelligence organisation, the Reich Security Agency (RHSA) headed by Ernst Kaltenbrunner, under the direction of Heinrich Himmler, which would absorb the Abwehr. The catalyst for this radical reform was the recent defection to the British of Erich and Elizabeth Vermehren in Istanbul, and the recruitment by the American Office of Strategic Services (OSS) in Ankara of Moyzisch’s young secretary, Cornelia Kapp. The daughter of Karl Kapp, the German consul-general in Sofia, Nellie Kapp had been educated in Bombay and Cleveland, Ohio, and was an anti-Nazi. According to an SIS report addressed to MI5’s Alex Kellar dated 29 May 1944, Moyzisch had survived the defection (although his superior Leverkühn had not) because of support from von Papen, for whom he had worked in Vienna, and because Nellie had been certified ‘mentally deficient’.5
Both defectors had possessed valuable information and the damage they inflicted would be exacerbated by three further Abwehr defections, those of Willi Hamburger and Karl and Stella Kleczkowski. One of the first casualties of the Turkish debacle was the Abwehr’s chief, Admiral Wilhelm Canaris, who was dismissed at the end of January, and then in early August arrested at his home at Zehlendorf in the Betazielestrasse by Schellenberg, who escorted him to an extended period of house arrest at the SIPO training school at Furstenberg in Mecklenburg.
These events were discussed at a conference called by General August Winter at ZEPPELIN, the staff headquarters at Zossen, which established a new section, designated Mil Amt, which would be responsible for the collection of military intelligence, with a particular requirement to provide advance warning of future enemy landings. This decision was endorsed by the Abwehr attendees, consisting of Admiral Leopold Burkner, Georg Hansen and Baron Wessel Freytag-Loringhoven of Abt. II; Franz von Bentivegni of Abt. III; and Colonel Heinrich, and effectively left responsibility for the collection of all military intelligence in the hands of the Abwehr’s Abteilung I. It was also agreed that Mil Amt, with a headquarters at Waldberg, near Furstenwalde, would be headed jointly by Hansen and Schellenberg, who would deputise for each other. When the ambitious Kaltenbrunner objected to this arrangement he was overruled by Himmler who had been persuaded by his friend and protégé Schellenberg that the SD had no experience of military matters, whereas the Abwehr had developed an expertise in the field. In practical terms the impact of this new structure was minimal, and the Kriegsorganisationen (KO) in Lisbon and Madrid were largely unaffected. Thus, although ostensibly the Abwehr had been absorbed into the Re
ich Security Agency (RHSA), much the same personnel continued their work in the collection of military intelligence. There would be more radical changes within the German intelligence monolith after D-Day, but during the vital period prior to the Allied landings, the individuals responsible for the recruitment and management of agents, and the assessment of information, remained mainly unchanged.
The extent to which the German analysts grasped the significance of the code word OVERLORD is not open to doubt, because on 8 February 1944 FHW issued an assessment on the subject:
For 1944 an operation is planned outside the Mediterranean that will seek to force a decision and therefore will be carried out with all available forces. This operation is probably being prepared under the codename of OVERLORD.
2
GERMAN SIGINT
‘The war will be won or lost on the beaches.’
Field Marshal Erwin Rommel
31 December 1943
In the last months of the war a group of specially indoctrinated American and British intelligence officers was deployed in Europe, usually behind enemy lines in TICOM teams, to capture cryptographic equipment and archives, and to interrogate prisoners of war who were suspected of possessing a knowledge of German code-breaking.
The very existence of the Target Intelligence Committee was a closely guarded secret until details were declassified in 2010. TICOM personnel snatched large quantities of documents and materiél to prevent them from falling into the hands of the Red Army and played a key role in acquiring information that would assist the western Allies in any future conflict with the Soviets.
The TICOM project had been conceived by Colonel George A. Bicher, Director of the US Signal Intelligence Division, in the summer of 1944, and fully established by October when six TICOM teams were deployed in Europe to gather information about all aspects of the enemy’s activities in the signals intelligence (SIGINT) arena. Through the examination of documents and the interrogation of prisoners, TICOM uncovered the scale of the Reich’s cryptographic operations, which proved to be far more extensive than had ever been suspected.
The first exploitation team was dispatched in April 1945 to the Neurouenster–Flensburg area, and others were quickly assigned to different combat zones as soon as they were overrun. Approximately 4,000 separate German document files were captured, weighing 5 tons; large quantities of cryptographic equipment devices were secured and 196 PoW reports completed. Of particular interest were five SIGINT specialists who were interviewed at Nuremburg in September 1945. All the files and equipment were then flown directly to England for examination by Anglo-American analysts, who came to admire the professionalism of their adversaries.
TICOM’s research uncovered the complexity of overlapping organisations that had collected Allied signals traffic and then subjected it to prolonged analysis. German SIGINT specialists excelled at traffic analysis (T/A) and were particularly skilled at the associated disciplines of direction-finding (D/F), call sign analysis, frequency allocation, plain-text analysis and operators’ chat. They had also pioneered airborne radar route tracking, and the monitoring of transmitter zero beat tuning. Thus, before a single cipher group had been solved by a cryptographer, the analysts had accumulated a great deal of information about a particular signal and its carrier channel. Even if the text itself resisted attack, T/A could often identify the sender, the type of message, the operator’s location, the unit to which he was attached, and characteristics of the net on which he was transmitting.
The extent of the Axis investment in signals intelligence proved enormous and consisted of six principal organisations employing a staff of around 30,000. Italy possessed two main SIGINT agencies, with Finland, Austria, and Hungary one each, making a total combined strength of 36,000. In comparison, the Allies fielded an estimated 60,000, of which 28,000 were US Army personnel.
Four of the German organisations were military, being the Oberkommando des Heeres, General der Nachrichtenaufklärung (OKH/GdNA), which dealt with enemy army traffic; the Kriegsmarine’s Seekriegsleitung III (OKM/SKL III), which handled enemy naval traffic, the Luftwaffe’s Luftnachrichten Abteilung 350 (OKL/LN Abt 350); and the Wehrmacht’s Oberkommando der Wehrmacht, Chiffrierabteilung (OKH/Chi). The two civilian organisations were the Foreign Ministry’s Cryptanalytic Section (Pers Z S), and Hermann Göring’s Research Bureau (FA), a Nazi Party agency that also dealt with diplomatic traffic, news releases, broadcast monitoring, telephone tapping, and other types of communications intelligence, irrespective of whether it was enemy, neutral, or friendly.
TICOM would produce a series of reports that were then, and remain today, the most comprehensive assessment of the Axis commitment to SIGINT, confirming that the source provided the foundation upon which other material was assembled, such as agent reports. Indeed, General Albert Praun, the OKW’s chief signals officer, asserted that most of the OKW’s actionable intelligence originated from SIGINT, not human sources, observing that, ‘In 1944–45 the results obtained by communications intelligence probably amounted to as much as 75 per cent of all the tactical information available to division commander.’1 However, Praun also acknowledged that Hitler and Jodl were often unimpressed by signals intelligence:
Throughout the war General Jodl, as well as Hitler himself, frequently showed a lack of confidence in communications intelligence, especially if the reports were unfavourable. However, orders were issued as early as the Salerno landing that all favourable reports should be given top priority and dispatched immediately, regardless of the time of day. Moreover, Communication Intelligence West was required to furnish a compilation of all reports unfavourable to the enemy derived from calls for help, casualty lists, and the like. Even during the first days of the invasion, American units in particular sent out messages containing high casualty figures, [and] the OKW was duly impressed. In contrast, the estimate of the situation prepared by the Western Intelligence Branch was absolutely realistic and in no way coloured by optimistic hopes.2
The Allied analyses revealed that the Germans had adopted a fragmented approach to the collection and processing of SIGINT, and that there had never been one single, centralised organisation to manage interception, traffic analysis, decryption and distribution, in sharp contrast to the Allied model. After a lengthy examination of captured documents and, having collated numerous interrogation reports, TICOM provided an overview of the various components of Axis SIGINT.
The OKH/GdNA, employing 12,000 staff and based at Juterbog, about 60 miles south-west of Berlin, was responsible for the cryptanalysis and evaluation of Allied army traffic, at any level whether strategic or operational, and undertook some radio broadcast monitoring. It was supported by two intercept stations targeted against high-level Allied traffic and nine field Signal Intelligence Regiments (KONA) assigned to various Army Groups to undertake interception, traffic analysis, cryptanalysis, and evaluation of Allied army low-level tactical traffic in the relevant Army Group areas. This was in addition to a small Signal Intelligence Section, assigned to the Army Commander-in-Chief West, which acted as a co-ordinating section for the two KONA regiments on the Western front. The unit issued three reports daily, circulated to the three High Commands (OKW, OKM and OKL) and to the Supreme Command, Armed Forces, to Heinrich Himmler, and probably to the Reich Security Agency (RHSA). These were also circulated directly to commanders at army group, army, and corps levels, and the nine KONA field units co-operated closely with their local counterpart Luftwaffe signals regiments. Until 1944, when the responsibility was passed to the OKW/Chi, the OKH/GdNA branch designated Inspektion 7/lV (abbreviated to In 7/IV) issued all the Wehrmacht’s codes and ciphers.
All branches of the German armed forces were heavily reliant on signals intelligence and much of the preparation for the expected Allied invasion was based on a very complex and comprehensive organisation. For most of the war SIGINT on the western front was the responsibility of KONA 5, until the establishment of KONA 7 in February 1943. Prior to February 1944, KO
NA 5 consisted of a SIGINT evaluation centre, Nachrichtenaufklärungs Auswertestelle 6 (NAAS 6), and four stationary intercept companies, Feste Nachrichten Aufklärungsstelle 2, 3, 9 and 12, as well as two long-range signal intelligence companies, Nachrichten-Fernaufklärungs-Kompanie, FAK 613 and FAK 624.3
The NAAS was based at St Germain-en-Laye and consisted of about 150 personnel, made up of interpreters, cryptanalysts, evaluators, draughtsmen, telephonists, drivers and clerks. It also employed some women auxiliaries who were usually assigned to the telephone switchboard.
Feste 2 was composed of a radio intercept platoon, a D/F platoon, and an evaluation platoon consisting of two sections: one for the assessment of content of messages, content evaluation, known as Inhaltsauswertung, and one for the evaluation of traffic, traffic analysis, Verkehrsauswertung. The unit had been formed originally to man the Wehrmacht intercept site at Münster, and in November 1944 Feste 2 amalgamated with Feste 9 and FAK 613 to form NAA 13.
Originally Feste 3 was the unit manning the Wehrmacht intercept site at Euskirchen but early in the war it had been subordinated to KONA 5. According to a member of its staff, Leutnant Hans Lehwald, it consisted of a radio reception platoon of approximately seventy receivers, and an evaluation platoon of twenty-five to thirty men. Evaluation was divided into a section for traffic analysis, cryptanalysis, evaluation, D/F and a filing section for diagrams of the radio nets, call signs, personalities, code names and D/F results. Later in 1944 Feste 3 would amalgamate with FAK 626 to form NAA 14.