Although Jebsen had given SIS an optimistic account of his handling of Dusko Popov, he did admit that a week earlier, on 21 April, he had ducked out of a meeting he had been ordered to attend at Biarritz with Dr Aloys Schreiber, a senior Abwehr official also based in Lisbon, and a certain Major von Bohhlen. Jebsen had suspected that the location of the meeting, which had been to discuss various sums claimed by Popov, was conveniently close to the French frontier if the Gestapo wished to arrest him. Jebsen had excused himself, saying that he was anxious not to compromise his own cover as a German malcontent.
The day after Jebsen had reported to SIS he went to the German embassy in Lisbon to receive his decoration. As soon as he stepped through the door of Schreiber’s office, he was knocked to the floor and interrogated. On 1 April he was bundled, unconscious, into the trunk of a limousine carrying diplomatic plates and driven straight to France, where he received an official escort for the remainder of his journey to Berlin. After a brief spell in a Wermacht prison, he was transferred into the custody of the Gestapo, who are believed to have executed him in the Oranienburg concentration camp on an unknown date in April 1945.
The first news of this catastrophe reached Bletchley on 2 May 1944, when a signal was intercepted from Schreiber to Berlin announcing the successful completion of Operation DORA. The message confirmed that Jebsen had crossed into France at Biarritz at 1500 hours the previous day. It was subsequently established that DORA was the code name for Jebsen’s abduction.
An analysis of Lisbon’s ISOS traffic showed that DORA had been a last-minute affair, and that Schreiber had been recalled from Madrid on 24 April to prevent Jebsen’s rendezvous with the SIS, which the Abwehr believed had been arranged for the following day. It seems likely that either Jebsen’s determination to discover the real identities of the Lisbon Abstelle’s agents had led to his exposure or that his investigation of a particular spy on the German payroll had caused his downfall. All these possibilities were discussed at length by the Twenty Committee, which had no alternative but to hope that Jebsen resisted his interrogators at least until the end of D-Day. In order to reassure Tommy Harris, an MI5 officer who had not been let in on the secrets of OVERLORD was asked to review GARBO’s correspondence to date to see if he could detect a bias in favour of a particular target area on the French coast. The officer concluded that there was only the slightest of imbalance in favour of the Pas-de-Calais.
Nevertheless, undeterred by these events, GARBO and his case officer continued their work as if nothing had happened. On 1 May GARBO relayed a routine message from DONNY, who had seen officers of the 28th American Infantry Division in Tenterden and Dover. At Folkestone he had spotted the American VIII Corps; GARBO commented that these (real) units were evidently on the move. The 28th Infantry Division had been portrayed elsewhere as an assault group unit and had also been reported recently in Tenby by DORICK, shortly before he had left South Wales. As D-Day got closer, the FUSAG forces (of which the 28th Infantry Division and the US VIII Corps were constituents) were to move slowly eastward toward Kent and East Anglia, adding support to the proposition that the expected attack would be spearheaded from the Dover area.
The following day, on 2 May, GARBO described an indiscreet chat with his mistress (who was working in the war ministry), who said that there was no chance of an immediate invasion, thus contradicting the view expressed by CHAMILLUS just three days earlier.
‘I am disconcerted by what CHAMILLUS said and cannot under any circumstances advise or give information,’ radioed GARBO.
The Abwehr, clearly preferring CHAMILLUS’s version of events, replied the following evening:
I do not consider that you should attribute too much importance to the opinion of J(5) in the present situation, since it is very probable that the ministerial personnel have received very severe orders with regard to the confidential handling of all military matters at the present moment. It is also possible that subordinates of ministries are being falsely misled intentionally.
For good measure, GARBO added a report from DORICK, who had seen armoured cars and tanks of the 6th US Division at Ipswich railway station.
The next day, 4 May, CHAMILLUS vouchsafed that the 3rd Canadian Division had just left Hiltingbury with orders to embark at Southampton. GARBO transmitted the message the same night and sent a longer text the next evening:
CHAMILLUS communicates that orders have been given to clean and prepare the camp to receive troops once more. Agent supposes that these will be second-line units. This proves J(5)’s lie because she suggested, naively, today that troops in the southern area were on manoeuvres, information which has been disproved, as in this case troops would have returned. My opinion is that, assuming they have not landed on the Channel coast, the troops which embark must at this moment be moving toward their far-off objective or to join the fleet reported by BENEDICT in the Clyde.
In his next transmission, on 7 May, GARBO laid into CHAMILIUS who, he said, had
displayed the ability of a simpleton. I am very disgusted with him, though I have not let him know this. He has, today, communicated with the troops he was awaiting, which he thought would be second-line troops, who have arrived, but they are the same as previously left, which is to say the 3rd Canadian Division. The troops at the other camps have likewise returned. Though too late, he tells me that they have been in one of the many rehearsals which Churchill announced would be carried out before the second front was opened. My last comments about J(5) should be disregarded. I see that her information was true and that the fault has been partly mine through being impressed with my agent. I see that I could get more accurate information through my ministerial friends. CHAMILLUS, in excuse, says that his beliefs were confirmed by concrete military events, which I am unable to deny. In future he will make no further comments to influence me or my chiefs. I am afraid he is a little discouraged by his great stupidity.
Madrid wasted no time in replying to this skilfully woven item of psychological arm twisting. GARBO received the following at 1959 hours, 8 May 1944:
I do not consider that we should reproach CHAMILLUS at all since the troops and the majority of the officers left the camps convinced that it was to be the invasion and only a few high officers knew the real objectives. If CHAMILLUS is disillusioned through his mistake, which he could not avoid, you should give him encouragement, as, if not, it might happen that when the real invasion is about to take place he will not notify this owing to over precaution.
Just half an hour later it was business as usual, with GARBO sending a report of troops from the 61st and 45th Divisions, and Nos 3 and 16 Commandos, seen by DICK between Brighton and Newhaven.
On the evening of 10 May GARBO transmitted a message in which he recounted a conversation he had had with CHAMILLUS’s American acquaintance, the sergeant known as 4(3), who had said that
the second front would open as soon as the two army groups destined for operation were ready. One of these, the 21 Army Group, is under Montgomery. The other, the First Army Group, is provisionally under the orders of Bradley. The American troops which are expected here will enter the latter army group. He assured me that Eisenhower would give a very important task to the American army group.
BENEDICT also continued to keep the shipping movements on the Clyde under surveillance and supplied GARBO with a regular flow of intelligence. On 10 and 11 May he kept watch from Greenock and later spotted a big naval assault group exercising in Loch Fyne, and mentioned having noticed troops kitted out with arctic clothing. He was convinced that this was the prelude to an important attack on Norway, and this news was reported by GARBO’s wireless on the evenings of 14 and 15 May. A similar picture was drawn by BENEDICT’s Greek agent, who noted a gradual build-up of cargo ships in Methil. Apparently, they were to supply an operation in which a large number of troops were to take part. They had already been assembled in some other ports along the Scottish east coast.
The pace was now hotting up, with a strong implication
in GARBO’s messages that perhaps a two-pronged attack across the North Sea and the Strait of Dover was being planned. The ever reliable DAGOBERT was producing some useful intelligence from the south of England. Each sounded eminently plausible, and the ISOS decrypts from Bletchley confirmed that the Germans believed them to be authentic.
On the main Romsey–Ringwood road I saw 125 military vehicles of all types, including Bren gun carriers, with the insignia of the 54th Division. I also saw vehicles with the insignia of the 47th London Division.
Meanwhile, it was realised that CHAMILLUS’s reported conversation with the American sergeant, 4(3), had made the necessary impact, for GARBO received the following questionnaire from Madrid by radio on the evening of 17 May:
With reference to the latest reports of 4(3), I send the following questionnaire, the reply to which would be of much interest. Where is the headquarters of the 21 Army Group, English? The numbering of the armies within the said army group, and their headquarters. How many and which divisions are within each of the armies of the said army group, indicating, where possible, which divisions are armoured and which are infantry.
Ten days later GARBO described a further exchange he had had with the American:
I questioned him on the 21 Army Group, but it seems that he does not know much about this formation. He was only able to say that there are a few American troops in it, but that the Americans are mainly in the First US Army Group.
Heavy emphasis was being placed on the obvious build-up of Allied forces in the south-east and eastern counties. On 25 May DAGOBERT saw ‘the sign of the panda several times’ around Ipswich, a reference to the divisional identification of the 9th British Armoured Division, which was evidently moving into East Anglia. GARBO himself made a contribution the following day by sighting ‘many troops with sign of a knight with lance’ at Lewes. This was further evidence that the 8th English Army Corps Headquarters had moved into eastern Sussex. A week later, on 1 June, DAGOBERT noticed the 9th Armoured Division outside Tilbury. The entire division had been transferred right across the country, surely more proof that the invasion was imminent and heading for the Pas-de-Calais?
While the German high command absorbed these reports, GARBO had other, more mundane, developments to report. Agent DRAKE’s behaviour, for example, had been giving cause for concern. He was supposed to have set himself up in the West Country, but on 19 April he had reported from Taunton that he had failed to get into the local prohibited zone. A month later he had sent an anxious letter from Exeter:
I entered this prohibited area where vigilance is extremely strict. I do not see any possibility of remaining here owing to the continual demands for documentation, which up to the moment I have been able to evade.
His luck had only lasted another fortnight. On 2 June 7(6) reported from Swansea that DRAKE had been arrested and sentenced to a month’s imprisonment:
Apparently, there is no indication that any suspicion exists regarding the mission with which he was entrusted by me and that it was only a routine case for not having complied with the present restrictions.
This effectively eliminated DRAKE from FORTITUDE and ensured that he did not have to experience any awkward lapses during the invasion. Having removed DRAKE, it was decided to silence 7(6) by portraying him as another of DAGOBERT’s failures. Accordingly, GARBO told Madrid:
I have received a long letter from 7(6) with reports, the majority of which were stupid. We can therefore discount the ability of this agent as an informant in spite of the repeated instructions given. His usefulness to the military information service is nil.
This ploy was designed to add some authenticity to GARBO’s work. After all, however brilliant an organiser he might be, not all his agents could plausibly be presented as excellent performers. DRAKE and 7(6) proved that occasionally GARBO experienced a setback. Nevertheless, GARBO’s network was maturing well, and the letter and wireless traffic to Madrid was gradually being built up. In fact, the volume got so great that Harris became concerned that it was probably too much for a single person, so GARBO sought permission to bring BENEDICT the Venezuelan down from Glasgow to act as his deputy. He also asked for Mrs Gerbers, Agent two’s widow, to be allowed to help with the mass of ciphering work. Madrid responded with enthusiasm. Incredibly, they also accepted GARBO’s suggestion that all future communications should be written in English, to assist Mrs Gerbers.
One agent who had excelled was J(3), the unconscious source who worked at the Ministry of Information. As Tommy Harris had already pointed out, this individual could have been identified as the head of the Spanish section and, purely by chance, the real person actually went to Madrid early in May 1944. GARBO reported his departure, and MI5 requested that the British embassy send him home at the earliest opportunity, before the Germans took any initiative. Nothing untoward happened to the entirely innocent official, and on 22 May GARBO described a conversation he had had with him on his return:
Result of interview today with J(3). He returned from Madrid for reasons connected with propaganda preparations of the second front. He proposed that I should help him, offering good remuneration for work which I could do at home. I accepted provisionally, telling him that I wanted to consult with my family, in order to allow time to receive approval from you, since it concerns work for the enemy which I could only accept as a sacrifice to be compensated by getting important information.
Not surprisingly, when tempted with such an attractive prize, the Germans wasted no time in telling their star agent to take up the MOI official’s tantalising offer. They radioed the next day:
Have studied carefully the question of J(3). Am of the opinion that you should accept the offer as apart from his being able to facilitate good information, this work assists your cover in every respect.
On 28 May, GARBO announced that he had taken the job … and had been made to sign a special document in which he had undertaken not to breach the Official Secrets Act.
I attach much importance to this because I learned that only people who may get to learn details which may compromise secret plans are obliged to sign this.
Even forty years later it is not difficult to imagine the fun that GARBO and his case officer had while compiling this particular message. But there was also a serious side to the Abstelle’s signal of 22 May. Madrid Abstelle had also inquired about FORTITUDE’s fictitious units in Scotland:
I am particularly interested to know urgently whether the 52nd Division is still in the camps in the Glasgow area after finishing the manoeuvres on the 11th, in accordance with the message of BENEDICT. I should be grateful to have your reply as soon as possible. Please take every measure to ensure that you are notified of every movement of that division by the quickest possible means. I should be grateful if you would tell me how much time will be lost from the moment the division starts its embarkation operation until the news reaches you for the transmission to us by message.
One interesting aspect to this text was the final sentence concerning the delay involved in passing on news of any embarkation. MI5 knew, from the ISOS intercepts, that this exact question had been put to the Madrid Abstelle by the Berlin Abwehr; Madrid had promised Berlin that the information about embarkation would reach the Spanish capital within forty-eight hours. On 23 May GARBO offered to send it on within twelve hours, which must have pleased Madrid.
This request was also convenient because it offered an excuse for GARBO to demand that the German receiving station in Madrid stay on the air later than its current closedown of eleven o’clock at night. MI5’s motive for obtaining longer hours from the enemy’s wireless service lay in a controversial plan to warn the Abwehr that the Normandy invasion had begun shortly before the landing craft really did hit the beaches. Both GARBO and Harris were adamant that the network’s credibility would be enhanced greatly if it could send a message to Madrid in time. The Abwehr would then be more likely to believe that this first attack was a feint and would then listen to GARBO’s informatio
n about preparations for the main landing in the Pas-de-Calais area. With GARBO’s increased reputation, Harris had argued, he would be an even more effective weapon of deception. He also pointed out that if the warning was received a few hours before the invasion started, it would still be too late for the Germans to do much about it. MI5 knew from ISOS that it would take the Abwehr about three hours to relay the warning to the relevant sectors via Berlin, so there was no danger of actually betraying or endangering the operation itself. Naturally, this extraordinary proposal got a mixed reception from the services responsible for conveying the troops across the Channel safely. Some thought that what Harris was suggesting was uncomfortably close to treachery. After lengthy negotiations with General Eisenhower’s staff, the supreme commander eventually gave his consent for a transmission to be made on condition that it should go out no more than three and half hours ahead of the attack. The idea of signalling just before eleven the previous evening was vetoed. The master plan for D-Day had the first troops landing on enemy territory at exactly 6.30 on the morning of 5 June 1944, so GARBO was authorised to make his signal at about three o’clock in the morning. Unfortunately, the Abwehr operators went off the air between eleven and seven, so GARBO was obliged to find an excuse to keep them up all night. His scheme was particularly ingenious and involved the creation of a spurious crisis concerning his Greek agent.
Operation Garbo Page 15