Life in Secrets: Vera Atkins & the Missing Agents of WWII (55 page)

Now Vera learned what she had long suspected: that Madeleine had been sent by Buckmaster directly into Gestapo hands. But why, Vera now asked Dr. Goetz, did Kieffer not follow Madeleine after the meeting and arrest her? This, said Dr. Goetz, was another question for Kieffer and not for him. He supposed it suited Kieffer to have her run around a little longer.

In the end, said Dr. Goetz, Madeleine was arrested after a denunciation. A woman had called up Avenue Foch offering to tell them the whereabouts of a female British agent in return for money. It was a story of jealousy. The woman who denounced Nora was Renée Garry, the sister of Nora's organiser, Emile Garry. When Nora arrived in Paris in June 1943, Renée was already in love with France Antelme, but during those terrifying weeks Antelme gave not only his protection but also his affection to Nora. In the autumn of 1943 Renée Garry “sold” Nora to the Germans in revenge.

What Vera was more interested to know from Dr. Goetz was when exactly Nora's arrest had taken place. Dr. Goetz thought it must have been in September or October 1943. It was certainly early in the autumn, because he was already very busy with several “decoy transmissions,” as he called his “radio game,” and then he had to learn to play yet another radio—Madeleine's.

I hunted through the new files for the original telegrams from Dr. Goetz to see just how he had tricked London, but almost none had survived. There was one interesting message, however, in Nora's file. It came from an agent named Jacques and was sent from Berne on October 1, 1943. In it he told London that, according to a source called Sonja (“Sonja” was garbled in transmission), Nora, or Madeleine, had had “a serious accident” and was “in hospital”—code for captured or in serious trouble. The telegram also mentioned a suspect agent named Maurice Barde, who worked with another agent whose alias was Ernest.

CIPHER TEL FROM BERNE. DESP 13.57 1.10.43. REC 1820 2.10.43
IMMEDIATE
FOLLOWING FROM JACQUES.
SONJ?A RETURNED FROM PARIS 25TH REPORTS ERNEST MAURICE AND MADELEINE HAD SERIOUS ACCIDENT AND IN HOSPITAL?. MAURICE IS BARDE. MADELEINE IS W/T OPERATOR.
IF YOU GO AHEAD ON PICK UP PLAN I COULD TELL ON RECEIPT OF PHOTOGRAPH WHETHER GENUINE OR GESTAPO MAURICE.
AM TRYING TO? GET FURTHER INFORMATION VIA SONJ?A.

On the telegram were scrawled various initials showing that it was sent to “F”—Buckmaster. He replied to Berne the same day, saying: “Have had apparently genuine messages from Madeleine since 25th therefore regard Sonja's news with some doubt. Can you give us estimate Sonja's reliability?” Whether Buckmaster got such an estimate was not recorded.

The “Sonja” message was an extraordinary revelation. It showed that London had received a serious warning of Nora's capture as early as October 1, 1943, and that there was reason therefore at least to suspect that her radio was in enemy hands from that date. Yet the warning was ignored. Perhaps Buckmaster simply dismissed the information because it came from a local recruit—Sonja was hired in Paris—and local recruits were never valued as much as London-trained agents. But Jacques, who sent the message, was Jacques Weil, a valued radio operator with a small Jewish SOE circuit known as Juggler operating near Paris. Weil, who had escaped to Berne after the Prosper collapse, had provided reliable reports before. Buckmaster's refusal to follow up this clear warning of Nora's capture was yet another stunning example of the tragic incompetence that littered these files.

It was, however, the identity of Sonja herself that made this telegram particularly remarkable. Sonja was in fact Sonia Olschanesky, a Jewish Russian-born dancer who had worked as a courier for the Juggler circuit. She was also Weil's fiancée. Unknown to London, Sonia had refused to follow Weil to the safety of Berne and remained in place after the collapse of Prosper, taking immense risks by running messages between different
SOE groups. She was herself captured in February 1944 and sent to Karlsruhe prison in the same convoy as Odette Sansom and the other British SOE women. She was then sent to Natzweiler, along with Diana Rowden, Andrée Borrel, and Vera Leigh. Sonia Olschanesky was, in fact, the woman drawn by Brian Stonehouse as No. 2 on the Lagerstrasse. Vera did not recognise her from Stonehouse s drawing because she had never met her. It was Sonia's name that Vera had found in the prison records at Karlsruhe. She did not recognise the name because, as Sonia was a local recruit, she had never heard it before. Having already decided that Nora had died at Natzweiler, Vera had assumed that Sonia Olschanesky was an alias for Nora. As Vera was to discover sometime later, however, it was Sonia Olschanesky who had died at Natzweiler and not Nora Inayat Khan.

What this message showed for the first time was that the fates of Sonia and Nora were even more inextricably entwined than that. Thanks to the bravery of Sonia Olschanesky, London was first warned of Nora's capture on October 1, 1943, but Sonia was ignored. Had Sonia's warning been heeded at that time, Dr. Goetz's “radio game” would have been exposed and probably halted there and then, saving countless SOE lives. Sonia Olschanesky herself, arrested when another Prosper subcircuit was penetrated in February 1944, would probably not have been captured.

Dr. Goetz said that very soon after Madeleine was caught, he was instructed to begin working her radio. She gave no help. “But we needed none,” said Dr. Goetz. Nora, he explained, was quite simple to impersonate because German signals staff had been intercepting her traffic for a long time. Furthermore, she had written all her letters to London en clair—that is, not in code—so they had been able to acquire all details of her signal plan and of her style through the usual method, “the mail” handed in by the agent BOE 48. Dr. Goetz had also been given a notebook belonging to Nora, found in a drawer in her apartment. The notebook contained all her back messages carefully written out. “From that we could work out her code and all her security checks.” Dr. Goetz set up
an entire fake circuit for Nora and gave her his own alias, Diana, for use within Avenue Foch.

“When was Nora sent to Germany?” asked Vera.

“I could not say. But she was one of the first to go,” said Dr. Goetz, who recalled that Nora was removed from Avenue Foch and sent on to Germany soon after being captured and certainly well before Christmas

“I told Kieffer I didn't need her to work her radio.” Now Vera was hearing something entirely new. Her investigations into Nora's fate had concluded that Nora had been sent to Germany with Odette Sansom and the other six women on the transport from Paris to Karlsruhe in May

But Dr. Goetz was telling her that Nora had been sent to Germany alone, probably in October or November 1943 and certainly many months before the other women. He did not know where Madeleine was sent, but it cannot have been too far away, he thought, because there had been a need to reinterrogate her at Christmas. At Christmas Dr. Goetz received a message for Madeleine from London, which was obviously designed to test her, so Kieffer sent a man to Germany to get the answers. The message Dr. Goetz was now referring to was composed by Vera and asked for certain information about Nora's family. Vera had received the replies from Dr. Goetz and judged them authentic.

Madeleine, however, was as usual uncooperative, said Dr. Goetz, so he had had to guess the answers himself, relying on his wits. And though Nora refused to assist at all while at Avenue Foch, she had inadvertently passed on certain personal details that proved useful, said Dr. Goetz.

Evidence from the newly opened files showed that at the end of the war as many as one-third of F Section circuits were penetrated as a result of radio deception alone. But the actual penetration of F Section was, of course, far wider, thanks to Henri Déricourt. A total of fifty-four British agents who landed in France passed through his hands. All were potentially contaminated, as were any other agents they contacted, as each one of them could have been trailed. The true extent of the penetration of F Section circuits on the eve of D-Day was therefore incalculable and
clearly concerned MI5's counterintelligence experts when they considered the disaster after the war.

MI5's own T. A. Robertson, known as TAR, who helped run the much-acclaimed British double-cross system, under which German agents captured in Britain were “turned,” commented in August 1945 on how SOE had failed to use properly “wireless finger printing,” whereby the agents' messages could be checked against an electronic trace. Had they done so, he said, “they would surely have been able to discover that their own agents were not operating the sets.”

After Christmas Dr. Goetz was worried that London had not been taken in by his reply to the Christmas questions over Nora's radio. “At this time my own impression was that although London were answering my messages they were not really deceived. At one reception for example we had asked for twelve containers but only one was dropped. This strengthened my view that they had guessed.”

Then London seemed convinced again. “At the next reception for which we asked for 500,000 francs we received precisely this amount. I therefore changed my opinion and continued to work the set and ask for large receptions. In the early months of 1944 we received not only a great deal of material but also agents.” He remembered arranging over Nora's radio for three agents to drop in February; among them came France Antelme, as well as a woman. Antelme was in a “towering rage” when he realised what had happened, but Kieffer was very excited when Antelme landed, as he knew he was an important agent, not least because Dr. Goetz received a message, meant for Antelme, congratulating him on his OBE. Kieffer even hoped he would know the date of D-Day. Antelme, however, gave nothing away. To prevent London guessing from Antelme's silence that he had been caught, thereby throwing suspicion on Madeleine's radio, Dr. Goetz then had to construct a cover story about his capture. He devised a series of complicated medical reports about an “accident” on landing.

By March 1944 Kieffer was obviously revelling in his conjuring trick
and began to think that he might match the success of his Abwehr colleague, Hermann Giskes, in Holland.