Codeword Overlord

by Nigel West

  Yesterday I was called to the War Office for an interview. I was asked for a lot of information about de Bona’s political views and why and how he had been brought to this country. I found out that the reason for this was a message from a British Intelligence officer in the Middle East, passing on some adverse report about de Bona which had apparently come from some Mihailovic circles in Yugoslavia. I do not know exactly what this report said but, from the questions put to me, it was clear than someone had suggested that de Bona had collaborated with the Germans. In view of de Bona’s high position now, the British are not likely to take any strong steps against him unless they have definite proof but they may be making further investigations. This is the reason why de Bona and I decided to dismantle the radio set and hide it in a safe place for the time being in case the house is searched. So far as I can judge, no suspicion has fallen on me; on the contrary, the English seem to expect me to help them find the truth about de Bona. I think this as I was asked not to mention the interview to de Bona. I think the British consider the source of the accusation a not very reliable one and very weak as practically every third or fourth Yugoslav has been accused by some Yugoslav source of being pro-German, so I do not see a very great danger in the accusation. De Bona expects that people in the highest positions will adopt a shocked attitude towards any accusation against him and will help to brush it aside. In the meantime please do not write to me in case my mail is being watched. Keep listening for the radio.

  Six weeks later, on 30 June, de Bona attempted to resume contact with an explanation for his absence:

  Had to stop for reasons given in Dusan’s letter 20th May. Have since had to travel with my master. Am just back and may be sent away on further unavoidable mission for him at short notice. Thanks to him, English doubts of me mostly overcome but some still exist.

  By this expedient, TRICYCLE and FREAK had been removed from participation in FORTITUDE at the most vital time, but the Abwehr was unimpressed and signalled on 9 July:

  Do not understand sudden and unnotified interruption of work in present critical times. Letter Dusan 20th May not received in view of postal restrictions.

  This terse exchange effectively terminated TRICYCLE’s role as a double agent, but had the merit of reassuring MI5 that Jebsen had not compromised him.

  ‡

  Another, hitherto unrecognised contribution to the BODYGUARD spies came from PRIMO, an Italian airman captured in Italy in January who was persuaded to operate as a double agent by Section V. An airline pilot and qualified radio operator before the war, PRIMO was caught in Naples with two companions, who were shipped to a prison in Algeria while he co-operated with Max Niven, the representative of the 44 Committee in which the Ops (B) equivalent operated under the theatre code name DOWAGER.12

  PRIMO was entrusted with a deception campaign, code-named VENDETTA, which was intended to direct German attention towards the south of France for thirty days from D-5, and tie down some ten divisions that might otherwise be deployed to the Normandy front. According to the VENDETTA narrative, General Alexander Patch’s Seventh Army would land at Sete and Agde on 24 June, whereas the real invasion of the French Riviera, code-named ANVIL, was scheduled for August. In reality, the US Seventh Army was composed of largely fictitious units, such as the US XXXI Corps that existed only on paper, but the genuine US 91st Infantry Division and French 9th Colonial Division undertook a very public embarkation in Oran on 9 June to ensure that the departure of some 13,000 troops and 2,000 vehicles was seen. Similarly, the aircraft carriers HMS Indomitable and Victorious, on their way to the Pacific, pretended to participate in the ‘landing exercise’, while the 91st Infantry went to Italy and the French troops occupied Elba. To complete the fiction, the Spanish authorities were asked for permission to land Allied casualties in Barcelona.

  A Section V case summary stated that PRIMO, who was managed between February and November 1944, ‘survived his early ups and downs and was a prime and most successful instrument in the implementation of all DOWAGER’s deceptive plans. MSS showed how high a value the Germans put on the case up to the very last stages.’

  When the invasion of south-west France failed to materialise, it was up to the Special Means agents to report that General Patch had cancelled the operation because the Germans had failed, as anticipated, to move their troops north to Normandy. Accordingly, VENDETTA was declared a significant achievement by the deception planners.

  Alfred Gabas. A French naval officer and radio operator in Cherbourg, code-named DESIRE, Gabas was part of a large stay-behind network intended to collect information behind the Allied lines.

  Gottfried Paul-Taboschat. The chief of the Abwehr in Barcelona, Taboschat was thought to be masterminding a massive influx of spies into England during the summer of 1944.

  Jean Senouque. A marine radio operator code-named CHARLES by the Abwehr, he was active in Granville and was turned by the Allies as a double agent code-named SKULL.

  Paul Fidrmuc. A suspected fabricator code-named OSTRO, Fidrmuc was the only German spy to correctly predict the Normandy landings, but where did he acquire his information?

  Bertie Koepke. Active across the Franco–Spanish frontier, Bertie Koepke worked under a business cover and specialised in sending spies to England disguised as refugees or evading airmen.

  Emil Kliemann. A senior Abwehr officer based in Paris, Kliemann recruited and ran the volatile Natalie Sergueiev, who was one of the four key sources in London. However, did she betray her role as the British double agent code-named TREASURE?

  Karl Kramer. The head of the Abwehr in Stockholm, Kramer ran several mysterious assets in England, but did they really exist?

  Ludwig Moyzisch. The Sicherheitsdienst representative in Ankara, Moyzisch managed a spy in the British embassy who photographed documents in 1943 that compromised the true meaning of the code name OVERLORD.

  A captured OKH map issued on 19 June 1944 identifying FUSAG units assessed as being located in south-east England, ready for the imminent ‘second wave’ attack across the Channel.

  Das Britische Kriegsheer. The German High Command’s assessment of the British military order of battle. Captured by commandos at Dieppe in August 1942, this 460-page intelligence handbook became the basis of all Axis judgements about Allied strengths.

  First page of Das Britische Kriegsheer.

  The German assessment of FUSAG’s entirely bogus order of battle, dated 15 June 1944.

  Erwin Rommel on an inspection of the Atlantic Wall. He persuaded himself that his deployment of armoured reserves in the Pas-de-Calais had deterred a planned second assault on the French coast.

  MI5’s assessment of the Madrid Kriegsorganisation’s internal structure, based on the interrogation of defectors.

  Harry Williamson, code-named TATE, returns to the field in Cambridgeshire where he landed by parachute in September 1940, accompanied by the author.

  The OKW telegram reporting ALARIC’s message of 9 June predicting another invasion in the Pas-de-Calais. The annotations indicate that it was handled by General Alfred Jodl and Adolf Hitler.

  The GARBO network of Juan Pujol’s notional spy ring.

  Juan Pujol at Buckingham Place for his investiture in June 1984, having emerged from hiding in Venezuela, to receive his previously ungazetted wartime medal.

  Baron Oshima Hiroshi, the Japanese ambassador in Berlin, whose detailed reports on the construction of the Atlantic Wall were transmitted in a diplomatic machine cipher that had been solved by Bletchley Park cryptographers. (Public domain)

  Dusan Popov, code-named TRICYCLE, whose friendship with an Abwehr officer, Johann Jebsen, nearly endangered the FORTITUDE deception campaign. (Public domain)

  7

  THE INTELLIGENCE ASSESSMENT

  ‘The British could be expected to do the unexpected.’

  Erwin Rommel, quoted by Hans von Luck

  21st Panzer Division

  During the early years of the war a major criticism of the German intellig
ence structure was its rigid compartmentalisation within the three major armed services without regard to the requirements of the military planners. They did not have the benefit of daily briefings until November 1942, when the OKW’s failure to anticipate the Allied landings in North Africa forced the issue and prompted the creation of an assessment group within Fremde Heere West (FHW) under Colonel Friedrich-Adolf Krummacher, consisting of thirty officers and 110 other ranks. An artillery officer who had served before the war as a military adviser in China to Chiang Kai-shek, Krummacher’s role was largely one of liaison, with the responsibility to draft and circulate a daily situation report, the Lagebericht West, compiled from all-source material supplied by the OKM, OKW and OKL, with his own analysis added. These circulars, distributed under the name of the FHW chief Ulrich Liss, would acquire great significance in the months leading up to D-Day and another experienced intelligence officer, Baron Alexis von Roenne, was assigned the task of predicting the Allied invasion.

  From Latvian–German nobility, von Roenne was fluent in Russian and had supervised the interrogation of PoWs on the eastern front before his transfer to FHW at Zossen. However, just as Liss was blamed for not anticipating the TORCH landings, von Roenne’s initial assessments of Allied strategy neglected to warn of the landings in Sicily, nor of the later assaults at Anzio and Salerno. An Anglophile artillery officer and skilled linguist, Liss was transferred to command a regiment on the eastern front and later promoted to general.

  Liss was perceived to have missed the telltale signs of an impending Allied assault on Algiers, Oran and Casablanca in November 1942, in the absence of any relevant agent reporting from the region, apart from suggestions of an expedition destined for Dakar. The B-Dienst had described a significant build-up of shipping on the Clyde on 21 April, estimating an increase over six days of eight vessels to forty-three. However, the B-Dienst could not identify the assembling convoy’s destination, nor detect the American invasion fleet exercising radio silence as it sailed across the Atlantic.

  A devout Christian, von Roenne would be implicated in the 20 July plot to assassinate Hitler through his friendships with those directly involved, and he was arrested on 9 August and executed at Plötzensee prison in October 1944. His replacement at FHW would be another experienced intelligence officer, Will Burklein.

  The foundation of FHW’s study was the Führer Directive No. 51, dated 3 November 1943 (see Appendix I) which was followed two months later by an OKW document entitled ‘Guidelines for Reconnaissance of Enemy Preparations for the Creation of a Second Front in the West’ and signed by General Jodl.

  The FHW’s Lagebericht West was delivered daily by encrypted radio teletype to the former royal palace, the Château Saint-Germain-en-Laye, just west of Paris, and the headquarters of the commander in chief, Field Marshal Gerd von Rundstedt, where von Roenne’s counterpart was Wilhelm Mayer-Detring. Subordinate to von Rundstedt was Erwin Rommel’s Army Group B in northern France, composed of the 15th Army and the 7th Army, based 40 miles away at the Château La Roche Guyon in the Val d’Oise; and Army Group G in southern France, consisting of the 1st Army and the 19th Army, commanded by General Johannes Blanowitz from his headquarters at Rouffiac, near Toulouse.

  Rommel’s chief intelligence officer was a FHW veteran, Colonel Anton Staubwasser, who had been supplied with some of the pieces of the jigsaw: in January 1944 British newspapers had been banned from speculating about the Second Front; on 1 April a movements ban had been imposed along the coast from The Wash to Land’s End, and restrictions placed on travel to Eire; on 8 April all leave was cancelled; on 18 April neutral diplomats were prohibited from returning home, and were required to transmit all radio messages en clair. In the intelligence business, then as now, such indicators are known as ‘clues’.1

  To a limited extent the Allies could monitor the enemy’s deliberations, but usually only on the rare occasions that such items were distributed by radio, and on 6 January MI-14 circulated an OKM report from the previous month, dated 16 December 1943. Without disclosing its sources, the OKM had declared that a noticeable increase in shipping in the Bristol Channel and Scottish anchorages was linked to ‘the great landing planned against Western Europe’ suggesting ‘a high degree of readiness’ on the part of the Allies.

  There were other early signs that the enemy, by being demonstrably predisposed to believe in a cross-Channel assault, was susceptible to a co-ordinated deception campaign. On 14 December 1943 von Rundstedt had circulated an assessment by radio, which was read at Bletchley Park within two days, mentioning German interception of traffic generated by landing exercises conducted along the coast between Portsmouth and Plymouth.

  In May Staubwasser identified some of the Allied units assembling on the other side of the Channel, such as the 51st Highland Division and the 7th Armoured Division, as combat-hardened veterans of the Libyan desert and Sicily campaigns who evidently had been withdrawn from the Mediterranean to participate in the imminent landings. He also recognised the 1st US Infantry, transferred from Italy in November 1943; the 9th US Infantry, participants in TORCH and HUSKY; and the British 1st Airborne, also back from Sicily. All were genuine formations, well-known to the OKW, and to Rommel. By Staubwasser’s calculations, there were a total of sixty-five Allied divisions in the UK by early June, and he resented the lack of Luftwaffe reconnaissance that should have provided him with confirmation. When in January 1944 Staubwasser was transferred to Rommel’s staff he was replaced by Major Roger Michael, who would later explain his methodology in a CSDIC interview:

  As sources to predict the Anglo-American attack from Great Britain I had the usual intercepted radio messages and statements by prisoners in Italy. Particularly valuable were messages sent through the Royal Air Force network, especially when they provided information on the four Allied Armies already in action. Radio messages from the convoys en route from North America were also useful. Our infiltration of the French resistance movement provided us with some very valuable indications and clues. The invasion itself confirmed our picture of the enemy.2

  On 5 May 1944, with the recent transfer of the 9th and 10th SS Panzer Divisions to Russia, the Axis forces across France and the Low Countries amounted to fifty-eight divisions. These consisted of ten panzer-type divisions, seventeen field-type, twenty-four lower establishments and seven training divisions. An American assessment of the enemy situation noted that ‘six infantry divisions and three panzer divisions fronted the NEPTUNE beaches. Six field infantry divisions and three panzer divisions were located north of the Seine with ten lower category divisions; these large forces waited not for the real but the notional FORTITUDE force to assault across the narrow channel at the Pas-de-Calais.’

  On 19 January 1944, on the day Rommel wrote to his wife that, ‘In the West: I believe that we’ll be able to beat off the assault,’ FHW predicted an invasion in about a month, asserting that ‘experience shows that a sufficiently long fine weather period suitable for landing operations may be expected at this time’. However, in the absence of the predicted assault, on 12 February FHW stated:

  To sum up, the continuation of invasion preparations of every kind is to be seen which, according to present opinion, points for the first time to the second half of February as a critical period.

  After so many false alarms, FHW conceded on 3 April:

  There is a complete absence of information, apart from worthless gossip, regarding the date of the assault. The only protection against surprise is therefore to be found in intensively organized sea reconnaissance by air and naval forces (for example U-boats).

  However, the same circular assessed the deployment of British and American troops across the country, in anticipation of embarkation:

  The troop movements in England have mainly been completed since the assembly areas seem to have been reached. The movements have above all led to the concentration of the central English group of forces, which has hitherto been spread over a wide area compared with the main group in the So
uth of England. It has reached a total strength of some twenty divisions by the addition of one or two American formations. It also includes strong parachute formations.

  Three days later, on 6 April, Rommel wrote to his wife, ‘the tension is growing day to day … It will probably be only weeks that separate us from the decisive event.’

  As D-Day approached, there was a distinct German effort to move troops into the Pas-de-Calais, defended by nineteen divisions of the 15th Army, with ten in Normandy, backed by eight in Brittany. These reinforcements were monitored in England, where the 21st Army Group circulated a Weekly Review appreciation on 9 April commenting that it appeared that:

  from the shaping dispositions of the remaining Panzer and the layback Infantry divisions in France that the enemy expects an assault anywhere from the Pas de Calais to the Bay of Biscay with an added emphasis on Brittany and a lessened risk to the Low Countries and the Mediterranean littoral.