The work of SIS in Northern and Western Europe during the first half of the war falls naturally into two parts. First, there was the period of the so-called Phoney War from September 1939 until April-May 1940 when, after the conquest of Poland (where the spoils were divided between Germany and the Soviet Union), there was a kind of armed standoff between the Allies - Britain (along with the empire) and France - and Germany. Over these six months or so SIS sought to develop and extend its coverage of German intentions and capabilities from a ring of stations in neighbouring countries, all of which were themselves anxious to remain neutral and were therefore especially sensitive about foreign intelligence agencies operating on their soil. The second period began after the German onslaught in the spring of 1940 which resulted in the occupation of Denmark, Norway, the Low Countries and France. This disrupted (and in some cases completely extinguished) SIS operations, which were further hampered by Italy’s entry into the war alongside Germany in June 1940. The Service now faced the formidable problem of re-establishing itself in German-occupied territory. At the same time there were efforts to mobilise resources in the remaining neutral countries - Sweden, Switzerland and Iberia - both to target Germany itself and to attack the Germans’ own foreign intelligence operations. Finland, which was neutral at the start, also offered apparently promising opportunities, but having fought (and lost) the Winter War against the Soviet Union in 1939-40, it moved closer and closer to Germany, becoming a full ally by the end of 1941.
Early wartime days
The first of a number of SIS withdrawals in the face of German advances during the war occurred from Poland in September 1939. On 1 September the Germans invaded. Two days later Britain and France declared war. German forces advanced so rapidly that on 5 September British diplomats, a military mission which had just arrived and the SIS station were all evacuated from Warsaw to Łuków, a small town fifty miles to the south-east. Here Sophie, the twenty-year-old wife of the SIS station chief, Major John Shelley (he was fifty-one), was killed in a German air raid. They had been married only since 21 February. Heading south towards the Romanian frontier, Shelley, using his secure SIS wireless, signalled London that ‘as matters here now look like becoming SAUVE QUI PEUT and this place will become extremely dangerous’, he had told the three female staff ‘to prepare to leave but they begged to remain’. Not everyone was as cool under fire as Shelley’s ‘three stout-hearted and courageous girls’. While the female staff could stay, Shelley asked permission to send home two male colleagues ‘who are quite useless in crisis [to] prevent panic and anxious to leave’. In fact they all escaped, with Shelley and his female secretaries reporting to Broadway on 28 September.1
In October 1939 the Polish intelligence service established its headquarters in Paris, but in the early summer of 1940 moved to London following the fall of France. In Paris a close liaison was established through the SIS station chief, Wilfred ‘Bill’ or ‘Biffy’ Dunderdale (nicknamed ‘Wilski’ by the Poles), who as head of the A.4 section at Broadway continued to be the chief SIS link with Polish intelligence after he too had to decamp from Paris to London. In Britain SIS offered financial, technical and logistical support, while the Poles (who had a very productive and extensive European-wide network of agents) agreed to pass on all the intelligence they acquired apart from any concerning internal Polish matters, an arrangement which, though not without its strains, worked thereafter to the great benefit of the Allied war effort.
For the SIS station in Helsinki across the Baltic Sea, the war effectively began on 21 August 1939 when more or less out of the blue the Soviet-German non-aggression pact was announced. This left the Finns to the mercy of their old enemies, the Russians, who invaded at the end of November. The Finns also felt badly let down by the Germans, which made it easier for Harry Carr, the long-serving SIS head of station, to work against both Germany and the Soviet Union. Carr’s formidable range of Finnish contacts included senior officers in Finnish Military Intelligence such as Colonel Reino Hallamaa, head of the Radio-Intercept and Cryptographic Branch. In January 1940 Menzies asked Carr to find out if the Finnish authorities had ‘procured any Soviet cryptographic material which could be communicated to us’. Carr immediately replied in the affirmative and it was arranged that Colonel John Tiltman of GC&CS should travel out to Finland, where he was presented by Hallamaa with a Red Army code-book taken off a dead Russian officer and which ‘bore the marks of a bullet’. GC&CS noted afterwards that it had been ‘of real assistance’ to their cryptographers.
During 1940 Finland became something of a refuge for intelligence people of various sorts. After the German invasion of Denmark on 10 April, Sidney Smith, just a fortnight after taking over as the SIS representative in Copenhagen, moved to Helsinki. Head Office had wanted him to be based in Stockholm, where he could keep in touch with his agents, but this was vetoed by both the minister, Victor Mallet, and the Stockholm SIS head of station on the grounds that it would cause ‘confusion’. With cover as a diplomatic courier, therefore, Smith travelled weekly to Sweden to meet agents. One of his contacts, a former Russian diplomat long resident in Denmark, tried to continue reporting through the United States legation in Copenhagen, but the Americans refused to pass the information on. Meanwhile, when it became apparent that the Russians were about to occupy the Baltic states, London decided to close down the Tallinn station and relocate some of the staff to Helsinki, from where they continued to run some Baltic agents. The fall of France brought an unexpected bonus for Carr in the shape of George Alexeev, head of the French secret service in Finland, who offered his whole agent network if SIS would provide the funding, an offer which London gratefully accepted. In September 1940 another Estonian contact, agent ‘Outcast’, who had been an important source on Soviet matters, turned up in Helsinki. Outcast, a Russian émigré formerly living in Berlin, had escaped from Tallinn with German help, but at the price of agreeing to work for the Abwehr Russian section. In Helsinki he told SIS that he would in future be visiting Finland regularly, and offered to work for the Service against the Germans. He appears to have been a valuable source, providing political, military and economic reports from within the Reich. In November 1940 he reported from his Abwehr control that ‘German command [was] preparing (June) campaign against U.S.S.R. which would begin Spring 1941 possibly earlier.’ In his signal to Head Office Carr added that such a startling indiscretion by an Abwehr officer to his agent seemed ‘incredible’, warning that ‘possibly [the] statement [had been] made for propaganda purposes’. London evidently agreed, since no further hard evidence remains in the archives of what would otherwise have been a unique intelligence coup.
After the German invasion of Russia in June 1941, SIS’s position in Finland became increasingly precarious. During June and July Carr managed to get all his staff out to Sweden, and made arrangements for maintaining contacts after he too had to leave. While some of the agents were able to travel to Sweden themselves and others could use secret couriers, Carr also set up a link through the United States legation (which remained in Helsinki until the summer of 1944). After Britain broke off diplomatic relations with Finland in August 1941, the USA became the ‘protecting power’ (looking after British interests) and a British representative remained in the United States mission. Carr arranged for him to receive material from Alexeev (the Frenchman, who continued to work in Finland until the autumn of 1944). There was also intelligence from a Baltic network, run by a senior Estonian officer also working for the Swedes, Germans and Japanese. He left packages for Alexeev to pick up at a ‘dead letter box’ in a Helsinki public lavatory. Contact with Outcast, however, was lost when his handler (another Russian émigré) was arrested in August 1941. But Carr had wisely set up a drill for fall-back contact in Stockholm which Outcast duly implemented, though not until May 1942.
Until shortly before the war Oslo had been a sub-station of Stockholm, but over the summer of 1939 an independent station under Lieutenant Commander J. B. Newill was established. In September Frank Foley, the longstanding SIS man in Berlin, was posted to Oslo with general responsibilities for Scandinavia as a whole, evidently on the assumption that he would be able to meet former contacts permitted to travel outside Germany, and also be well situated to recruit neutral residents who could visit the Reich. In the event not much was achieved beyond Foley putting GC&CS in touch with Norwegian cryptographers who had succeeded in breaking some German diplomatic codes and were working on German and Russian naval and military cyphers.
Following the German invasion, the Norwegian military high command withdrew from Oslo north towards Lillehammer. Foley and his staff went with them and provided secure wireless communications through which the Norwegian Commander-in-Chief, General Ruge, appealed desperately for help from the British and French governments. On 14 April Foley added his own commentary. Norway, he said, ‘has lost her arsenals and supplies in towns captured by Germans. They are fighting almost with bare fists . . . Unless help comes at once there will be a first class disaster from which allies will, I submit, find it difficult to recover.’ General Ruge ‘fears he will not be able hold out much longer. Please take his word most earnestly. You cannot conceive pitiable condition material this army but men fine types.’ Foley’s secretary, Margaret Reid, afterwards compiled a vivid account of the retreat under attack from German bombers, eventually reaching Molde on the west coast where they embarked for Scotland on 30 April. During one air raid the SIS wireless operator, H. C. Edwards, collapsed. ‘He was unconscious for over an hour,’ recalled Reid, ‘and part of the time delirious and calling out for his wife and Commander Newill.’ ‘In all probability’, she thought, it was ‘due to overstrain and too little sleep’. He died a few days later.2
On his return to London Foley was put in overall charge of operations in Scandinavia (with Newill as section head), the Low Countries and those involving de Gaulle’s Free French movement. Inevitably there was a period of confusion while the arrangements needed to underpin a successful co-ordinated clandestine effort evolved. Not only was there friction between SIS and SOE, but there were internal differences between the Norwegian armed services themselves, and establishing satisfactory liaison with the Norwegian government-in-exile took some time. Meanwhile on 10 June wireless contact was established with agents in Haugesund (on the coast south of Bergen), but the group was broken up by arrests in August 1940. With SIS clearly under pressure to re-establish a presence in Norway, two large groups of agents were infiltrated in September by a motorised fishing boat from the Shetland Islands: ‘Skylark A’ (destined for Oslo) had twelve agents, and ‘Skylark B’ (Trondheim) twenty-one. By April 1941, with the Norwegians and SOE separately sending in their own people, Newill complained that without any central control there were ‘many examples of line crossing’ which ‘increases the hazard to our agents to such an extent as to make their work almost impossible’.
Although Claude Dansey’s appointment in May 1941 as an adjudicator between SIS and SOE eased the inter-agency problems, progress was slow. In July 1941, other than some intermittent communications from Skylark B, all the Norwegian section could report was the early stages of four additional operations, three of which aimed to provide ship-watching reports from along the coast. During 1941 a series of operations identified by Greek code-names were initiated. Most of these emanated from suggestions by Norwegian escapees that they should return to Norway equipped with a wireless set, or that they had contacts (particularly in shipping circles) who could already operate a set and who would willingly act as observers. One operation was ‘Epsilon’, run by Newill’s successor, Eric Welsh. In particular this investigated German interest in heavy water, an important component in the development of an atomic bomb, and produced by the Norsk Hydro hydroelectric plant at Rjukan in south-central Norway. In October 1941, Leif Tronstad, a professor of chemistry in Trondheim and a member of the Skylark B network, escaped to England where he was interviewed by R. V. Jones, an Air Ministry official seconded to SIS’s scientific section (which he headed), and Professor Frederick Lindemann, Churchill’s scientific adviser.3 Based on Tronstad’s information Welsh (a scientist himself ) proposed the destruction of the plant by bombing or sabotage. ‘The removal of this source of supply’, he wrote, ‘would completely cripple any designs the Germans may have with regard to this type of weapon and on the other hand, the Allies are not in a position to use this potential weapon themselves for at least eighteen months, as they are only now considering building a suitable plant in America.’ Following the disastrous failure of a Combined Operations raid, the machinery was temporarily disabled and heavy-water stocks destroyed in a well-executed SOE sabotage attack in February 1943. This demonstrated, at least in this instance, that SIS intelligence and SOE operations could successfully and productively be integrated.
The most sustained demand for Norwegian intelligence came from the Admiralty, who wanted rapid information about German naval movements along the coast. Trondheim became particularly important after the main German battle fleet was based there in January 1942. In April the newly appointed Deputy Director/Navy, John Cordeaux, complained that ‘we are very scantily served at present in Norway, being represented only in Bodo, Trondheim, Bergen and Oslo’, but efforts to recruit and infiltrate more agents brought a marked improvement over the following year. The need to provide coverage for the Admiralty at Kirkenes on the Russian frontier in the far north-west of Norway (where there was a U-boat base) led to a disastrous co-operative venture with the Russians. Two SIS agents equipped with radios were flown by Catalina flying boat to an airbase at Lake Latkha in Russia in August 1942. The Russians had agreed to receive the agents and drop them by parachute near Kirkenes. Far from despatching them within days, as promised, the Russians held the agents incommunicado for two months and then dropped them, with only their summer equipment, in Finland instead of Norway. They were quickly apprehended and handed over to the Germans who brought them to Oslo, where they were interrogated under torture and shot. ‘I am’, minuted Cordeaux in October 1943, ‘most distressed to read this dreadful story and of course now bitterly regret that we attempted cooperation with the Russians.’
The German conquest of Norway not only brought the hitherto rather poorly resourced SIS station in Sweden into the front line, but it also naturally heightened the Swedes’ vigilance concerning any foreign clandestine activities which might threaten their neutrality.4 Besides, as with other European neutrals in the early years of the war, while there were undoubtedly individuals with pro-British leanings, there were also many pro-Germans in Sweden. The station chief in 1939, Lieutenant Commander John Martin, had no more than ten regular agents and sub-agents on his books, with four more on probation. Early in 1940 a group of Section D agents, primarily gathering shipping intelligence for the Ministry of Economic Warfare, were arrested, and all were sentenced to prison terms of hard labour ranging from eight to fifteen months.5
Worse was to come. In April 1940 an SIS Section D team was uncovered by the Swedes. They were operating under business cover and aiming to sabotage northern Swedish port installations handling iron-ore exports to Germany, an operation specifically urged by Churchill. The leader of the group, A. D. Rickman (who had attracted the attention of the Swedish Security Service by his contacts with German émigré journalists), was arrested and a mass of compromising material found in his flat, including fifty-three kilos of gelignite, detonators and electric timing devices. The minister, Victor Mallet, complained to London that ‘our sleuths seem to be thoroughly bad at their job: so far they have achieved little in Sweden beyond putting me and themselves in an awkward position’. Martin was not at all happy either. ‘Frankly I feel very bitter after all the hard work and time I have given up to help them’, he cabled London, ‘that they should, by their own negligence or inefficiency or both, have compromised me like this.’ The Rickman debacle, moreover, demoralised Martin’s most important single Swedish source, who had been reporting on the information revealed to the Swedes by Rickman and his associates. ‘Effect of this case on source deplorable,’ wrote Martin. ‘He stated frankly that previously he felt he was working with people who knew their job but now feels large portion of 22-land [UK] S.I.S. composed of inexperienced amateurs.’6
Section D’s incompetence in Sweden was one of the factors undermining Laurence Grand’s position back home and contributing to the creation of SOE as an independent agency. Martin never lost his antipathy to special operations, and the situation in Sweden illustrates SIS-SOE relations at their worst. In March 1941, for example, after the arrest of three SOE Norwegian agents, Martin complained to Menzies that he considered ‘activities of S.O. organization having serious results and impairing our relations with Swedes . . . [It] embarrasses me and increases my difficulties. They cross my lines and tap my resources with resulting serious confusion.’ Menzies, in turn, asked Frank Nelson at SOE to ‘consider issuing instructions for your people to lie low and to act with the utmost discretion for the time being’.
The fall of Denmark and Norway in 1940, and the expulsion of the British from Finland in 1941, meant that by the beginning of 1942 Sweden had become a hive of SIS activity. Additional staff were sent in, and sub-stations opened in Malmö and Gothenburg. From the capital, albeit under very close Swedish surveillance, Martin’s Stockholm station worked Swedish and Danish cases, and assisted with Norwegian operations, while Harry Carr’s Helsinki station-in-exile worked Finnish and Baltic cases. The most important of the Norwegian operations involved contact from November 1941 with scientists who had access to valuable intelligence on German progress towards an atomic bomb.
Despite this activity, London wanted more from Stockholm, and in February 1942 Menzies told Martin that he was ‘seriously alarmed at lack of [military, naval and air] information from first hand sources in North Germany, German Baltic ports and Denmark’, and that he hoped that under Carr’s supervision ‘a service to North Germany and Denmark’ might be worked up, both directly and through the Baltic states. Martin continued to have problems, not all of his own making. In April 1942 the head agent of the Czechoslovak intelligence service (run by the Czechoslovaks from the UK), along with three members of his network, was arrested. The Czechoslovaks reported, using their own code, through the SIS Stockholm station. Among the documents captured were their coding materials and, having read all their back telegrams, the Swedes took particular exception to one message referring to the possibility of sending typhus-infected lice into Germany, as they naturally felt their position would be badly compromised if the Germans got to hear of it. Martin, of course, suspected the hand of SOE behind it, and so, indeed, it turned out, SOE being in touch with at least two of those arrested. The press coverage in Sweden was considerable and Mallet was once more embarrassed.
In December 1942 Martin was replaced as Passport Control Officer by Cyril Cheshire (a fluent Russian-speaking timber merchant in civilian life), who, in the undoubtedly more favourable circumstances of 1943-5, markedly increased the productivity of the Stockholm station. Martin had certainly been under very heavy pressure in 1939-42, and he bore the brunt of the move from a peacetime to a war footing, along with the burden of coping with the fallout from the early excesses of Section D and SOE. Under the circumstances he seems to have kept his head remarkably well, and he also retained the confidence of the minister, which was no mean achievement.
In September 1939 SIS had a station in Geneva, headed by a Passport Control Officer, with an assistant and a wireless operator. Following the outbreak of war, however, Claude Dansey and the greater part of his Z Organisation relocated from London to Switzerland, which was believed to be a better base for penetrating Germany. Dansey and at least four colleagues initially set up rather insecurely in a Zurich hotel, where Dansey had to warn his staff, ‘Don’t call me colonel.’ Eventually the team was given cover as the visa section of the Zurich consulate, with Dansey as consul. After Menzies became Chief in November, Dansey was posted back to London and his deputy unsteadily (he had a drink problem) minded the station until February 1940, when he was replaced by Count Frederick ‘Fanny’ Vanden Heuvel, who remained in charge for the rest of the war. Heuvel was a cosmopolitan figure of Italian origin who had been brought up in England and had a successful business career. An acquaintance of Freddie Browning’s, he had worked for Cumming in France and Switzerland from 1916 to 1918.
An SIS agent arrested in Switzerland in October 1939, while under interrogation, had already blown Dansey’s cover as Z. So concerned was Dansey to secure this agent’s release that in late November 1939 (before Menzies’s official appointment as Chief ) he personally visited the President of the Board of Trade and got Gladwyn Jebb to see Sir Horace Wilson at the Treasury in order to sanction some sort of trade deal with the Swiss - a suggestion was an increase in the import quota of watches from £30,000 to £50,000. Dansey had good reason to be nervous, as the arrested agent alleged that he had received all his instructions from him (Dansey). ‘Why this gratuitous lie’, fumed Dansey, ‘when I had only just been put in touch with him I cannot understand.’ Dansey’s own Swiss sources now said that it was too dangerous for them to meet him in public as he was effectively ‘brûlé’ (blown). After discussions with a Swiss intermediary a deal was struck and the agent released and returned to England.
The possibilities for establishing robust networks of agents targeting Germany and Italy, which had originally been the main purpose of the Z Organisation, markedly deteriorated after the war had begun. Already highly sensitive about their neutrality, the Swiss made life difficult for all the foreign intelligence services operating in their country: French, German and Italian, as well as British. Early in 1940, they successfully penetrated the SIS Zurich station with a Swiss national who had been living in London and had applied for permission to return home, upon which the Z Organisation promptly recruited him without any apparent preparation to work in Switzerland. But he revealed himself to the Swiss authorities, who used him to keep tabs on SIS, a role he played for some time until his case-officer was alerted to his duplicity by an anonymous informant. It is scarcely possible that this man was recruited without Dansey being consulted, but the manner in which he was engaged is of a piece with the hasty quest for personnel in the difficult early wartime days.
Matters were not improved by the refusal of Dansey to permit his officers to do any counter-espionage work or of Valentine Vivian to deploy anyone from Section V in Switzerland. One officer, posted to Switzerland towards the end of the war, afterwards asserted that this was principally due to the personal antagonism between the two men, ‘who were at no time on speaking terms throughout the war, a deplorable state of affairs’. Another difficulty with the Swiss operation concerned communications. There was an SIS wireless set at Geneva, but it could be used only for receiving messages as the Swiss authorities did not permit foreign missions in the country to send enciphered messages except through the Swiss Post Office. Before the fall of France messages went in a diplomatic bag through Paris, and after May 1940 it was possible for British members of the mission who had diplomatic status and were over the age of forty-five to travel through Vichy France to Spain and Portugal and thence by plane to London. This ceased after the Germans occupied Vichy in November 1942, but it was thereafter found possible to bribe South American diplomats to carry the bags out. ‘Two journeys and retire for life’ was the saying. For additional security, letters for the diplomatic bags were sewn into their envelopes. Cypher telegrams could still be sent, but SIS had to use one-time pads for this, which, as Switzerland became more isolated, came to be in very short supply, as getting material into the country was as difficult as getting it out. These communications difficulties meant that often only messages of the highest importance could be sent by cable, and that much intelligence collected in Switzerland reached London only after a considerable delay. Because of the lack of continuous secure communications, moreover, London was unable to send out any signals intelligence material, which was another handicap for the Swiss station.
Among the best wartime agents in Switzerland, Halina Szymańska, remembered by one SIS officer as ‘a very attractive and formidable personality’, began working in late 1940. Szymańska’s husband had been Polish military attaché in Berlin, where the couple had become acquainted with the Abwehr chief Admiral Canaris.7 After the defeat of Poland, Canaris had arranged for Szymańska to escape to Switzerland, where he put her in touch with the Abwehr representative in Zurich, Hans-Bernd Gisevius. Although nominally working as a secretary in the Polish legation in Berne, Szymańska became a Polish intelligence agent (though paid from British funds) and began to cultivate Gisevius for information about German policy and internal conditions. As well as passing information to the Poles, with their agreement she also reported directly to SIS using the code-name ‘Z.5/1’. In January 1941, for example, she reported details of German aircraft stocks, Gisevius’s belief that there would now be ‘no invasion Great Britain but aerial bombardment on increasing scale also submarine activity’, and ‘no action Balkans before March’. Three months later Gisevius told Szymańska he was ‘convinced hostilities between Russia and Germany will start early in May’. Much of this intelligence was very sound. The German invasion of Yugoslavia and Greece was launched on 6 April, and Operation ‘Barbarossa’, the invasion of the USSR, had originally been planned for May, but was postponed until 22 June 1941.
The most intriguing aspect of the Szymańska-Gisevius relationship, however, is the role of Canaris, or ‘Theodor’ as SIS called him. There has been much speculation about the precise nature of this sensational Swiss link between the intelligence services of Britain and Germany and whether, for example, Canaris and Menzies were in personal contact. The SIS archives reveal that Gisevius (though never an SIS agent) was a regular source of intelligence. Between August 1940 and December 1942 Geneva sent London twenty-five reports with information provided by him. Almost all of this was channelled through Szymańska. Although Vanden Heuvel told London that Gisevius was ‘first and foremost acting as intermediary for Theodor’, this is not unambiguously supported by the evidence, and only nine of the reports specifically quote Canaris. Indeed, Vanden Heuvel’s assertion may have been calculated to boost his station’s intelligence product as much as anything else. One report, however, was based on a dinner Szymańska had with Canaris in Berne on 19 October 1941. This is the only recorded face-to-face meeting between Canaris and anyone reporting directly to SIS. Canaris had just returned from a tour of the Russian front and reported the difficulties the German forces were experiencing there on account of severe winter weather. Hitler, he said, had miscalculated and had ‘counted on support from dissatisfied elements in Russia itself, which had completely failed to materialise’.
Both Canaris and Gisevius were involved with opposition groups in Germany, but the extent to which their contacts with foreign intelligence agencies in Switzerland - Polish, British and American (from the spring of 1943 Gisevius also passed on information to Allen Dulles, representative in Berne of the United States Office of Strategic Services) - constituted treason is debatable. While Canaris and Gisevius may certainly have been keeping both their own and their country’s options open, cultivating Polish and British contacts was legitimate Abwehr business which could produce valuable intelligence and even provide them with a channel for spreading disinformation. On 28 March 1941, for example, Szymańska reported a definitive statement from Gisevius, ‘German troops in Libya not [emphasis added] for offence purposes’. Two days later, in fact, German and Italian forces launched their drive into Cyrenaica.
Venlo and work in the Low Countries
One consequence of the Z Organisation’s move to Switzerland was that Dansey’s man in the Netherlands, Sigismund Payne Best, was transferred to work under Richard Stevens, the SIS station chief at The Hague. Best brought with him what appeared to be extremely promising contacts with an anti-Nazi network inside the German army. Early in October 1939 Stevens told London that Best was ‘reasonably confident’ that he could arrange for ‘two highly placed 12-landers’ (‘12’ being the code for Germany), Generals von Rundstedt and Dorsheim, to visit Holland ‘in the near future’. They were, he reported, members of an organisation which sought to ‘overthrow the present regime and establish a military dictatorship’. What the Germans wanted, however, was ‘some sort of assurance’ that, if they succeeded, the British government would ‘be prepared to treat with them’. Stevens consulted Sir Nevile Bland, the British minister at The Hague, who gave his support, but because of his diplomatic status was not prepared to take part himself without instructions from London. Further enquiries by Stevens revealed that, as well as Rundstedt, a General Wiedersheim and an Oberst (Colonel) Teichmann were involved in the opposition group.
Within a fortnight direct contact had been established. On 17 October Teichmann, ‘speaking discreetly by telephone’, reported that Wiedersheim had attended an army commanders’ meeting with the Nazi leader Heinrich Himmler at which the generals had ‘refused to undertake any major action of any kind against France or England’ and had ‘insisted that everything be done to obtain peace’. Teichmann further said that conditions were ‘such that only small impetus required to set ball rolling and to get rid of Nazis’. Here was intelligence apparently coming from the very highest German military circles, and with it the seductive possibility of bringing down the Nazi regime. SIS was playing for high stakes indeed, and the potential prize was so glittering that critical faculties both in the Netherlands and in London were dangerously blunted. Years afterwards an SIS contemporary at The Hague remembered ‘the almost overbearing confidence of Stevens, who seemed to be completely in the pocket of Best’, and a diplomat at the British mission recalled that Stevens, ‘who was a man of immense ambition, saw in this a possibility literally of winning the war off his own bat, and this completely clouded his operational judgement’.
But if Stevens and Best were dazzled, so too, apparently, were the Chief in London and even his political masters. While Sinclair was in his last illness (he died on 4 November), and perhaps not fully able to engage with developments, Rex Howard, his chief of staff, and Menzies, his deputy, were kept fully informed, as was the Foreign Secretary, Lord Halifax, and the Prime Minister, Neville Chamberlain, who gave his personal approval for SIS to continue discussions with the Germans. Although some officials and politicians were not very optimistic about the possibilities of success, on 1 November the War Cabinet authorised the continuation of negotiations.8 Halifax also circumspectly briefed the French ambassador in London about the covert contacts said ‘to emanate from German military elements . . . anxious to get rid of the Nazi régime’. In the Netherlands, although it was still a neutral country and keenly anxious not to offend Germany, the head of Dutch Military Intelligence, General J. W. van Oorschot, supported SIS and provided an intelligence officer, Lieutenant Dirk Klop, to accompany the SIS men when meeting the German representatives.
There was surely an element of wish fulfilment about the whole affair. It may have seemed too good to be true, and so it was. From the start, alas, it was a brilliantly conceived and executed double-agent operation run by the Nazi Sicherheitsdienst (SD: security service). Although General Wiedersheim actually existed, he never turned up to any of the five meetings Stevens and Payne Best had with supposed emissaries. But the Germans tantalisingly increased the pressure in early November. After a meeting with a ‘Hauptmann Schaemmel’ (actually the SD officer Walter Schellenberg, who was running the operation), Stevens told London on 7 November that a ‘coup d’état will be definitely attempted’ whatever the British government’s attitude might be. The same evening he telephoned London to say that ‘to-morrow, the big man himself is going to meet us’. But the next day (phoning at midnight), although they had had another ‘satisfactory meeting’, due to a German military conference which Wiedersheim apparently could not avoid ‘big man did not come, but sent most cordial message, and probably coming tomorrow’. Following this meeting Stevens proposed flying to London to report in person and hoped to brief a senior Foreign Office official to carry on the negotiations.
The scene of the Venlo incident as depicted in the Amsterdam newspaper De Telegraaf, 30 December 1939.
So it was that on 9 November 1939 - a grey, overcast day - Stevens, Best and Klop (himself masquerading as the British ‘Captain Copper’), along with Best’s driver Jan Lemmens, headed for Venlo on the Dutch- German frontier. The rendezvous was at the Café Backus, situated beyond the Dutch border post and about 150 yards short of the German one. In a script prepared for a radio broadcast after the war, Stevens described a peaceful scene on the day: ‘No one was in sight except a German customs officer who was strolling towards the Dutch customs house and a little girl who was playing ball with a big dog in the middle of the road.’ ‘Schaemmel’, standing on the café veranda, waved a greeting. The man in charge of the German snatch squad, captured and interrogated by MI5 late in the war, told his side of the story: ‘As soon as Schellenberg recognized the approaching car of the two British agents he gave the arranged signal by taking off his hat.’ Firing machine guns in the air, the squad rushed forward and captured Stevens, Best and Lemmens. Klop opened fire, and was himself fatally wounded.
Beyond the wider fact that Stevens and Best had been completely taken in by the Germans, their tradecraft that day was deplorable. Stevens was carrying some coding material, and Best had a list of agents’ names and addresses with him. They travelled to Venlo in Best’s own distinctive American Lincoln Zephyr car. Perhaps because there had been two previous meetings at the same venue, no one thought to reconnoitre the area ahead of time on 9 November, and there was no contingency plan in case things went badly wrong. Stevens’s and Best’s own attempts at security, keeping the dealings with ‘Schaemmel’ and the others completely to themselves at The Hague station, meant that there was no opportunity for any of Stevens’s colleagues to assess the German emissaries or the operation as a whole. Stevens’s head agent, moreover, had himself taken on an assistant, Folkert van Koutrik (code-named ‘Wallbach’), in the summer of 1939, who was shortly thereafter turned by the German Abwehr, and proved to be a very valuable source indeed, so much so that by the autumn of 1939 the Germans had a pretty clear picture of the whole SIS operation in Holland, a fact which clearly informed their subsequent interrogation of Stevens and Best.
The Venlo disaster came at a bad time for the Service. The changeover at the top and the early wartime pressure to produce results, combined with enthusiastic initiative on the ground (not always a bad thing), meant that the operation was pressed forward with perhaps more despatch than should have been the case. The fact that Best was Dansey’s man, moreover, cannot have helped such ambitions as Dansey may have had for Sinclair’s job. Stevens and Best, however, were not acting independently and, albeit based on their own judgment of the case, proceeded only after seeking approval from London. The participation of the Dutch, who maybe had more to lose than the British if the operation went wrong, was also significant, and can only have encouraged Stevens and Best to press on with the contacts.
But the damage from the affair was very great. The Service’s reputation inevitably suffered, and the Germans made tremendous propaganda capital out of it. For the rest of the war the painful Venlo experience coloured British opinions generally (and Menzies’s specifically) about responding to apparent German opposition elements. The capture of Stevens and Best, who spent the rest of the war in a series of prisons and concentration camps, gave the Gestapo a glorious opportunity of interrogating ‘members of the British Intelligence Services in positions of authority’ and obtaining a picture of the organisation’s activity. How much actual information they got from the two SIS officers, and how much from their double agent van Koutrik, remains uncertain. By mid-December 1939, however, the Germans were able to construct detailed and largely accurate charts of both Stevens’s and Best’s agent networks and in the autumn of 1940 their Informationschaft GB provided some fairly accurate information about SIS head office and the Z Organisation, quoting both Stevens and Best. Postwar interrogation of German intelligence officers suggested that, while van Koutrik had provided names and addresses of Stevens’s Hague station agents, ‘they knew nothing of the Best organisation prior to the Venlo incident’. Whatever the truth of the matter, it is clear from the German interrogation reports on Stevens and Best that both men provided plenty of information about SIS, if only because they believed the Germans already knew a lot.9
The Dutch government were deeply embarrassed by Venlo, and Anglo-Dutch relations suffered accordingly. A complete breakdown in diplomatic relations was saved only by Sir Nevile Bland’s considerable emollient skills, but General van Oorschot had to resign and the Foreign Office instructed SIS forthwith to cease all activities in the country. Rodney Dennys, head of the counter-espionage section, which operated separately from the Stevens-Best networks and had not been compromised, protested strongly and after Menzies had weighed in on his behalf was allowed to continue working, provided he moved his headquarters to Brussels. But when the Germans invaded in May 1940, Dennys, who evidently had not completely relocated to Belgium, spent a day and a half at The Hague burning files and card indexes. During the invasion Major Monty Chidson, the fluent Dutch-speaker who had briefly headed The Hague station in 1937, in one of D Section’s few real successes managed to seize the bulk of Amsterdam’s industrial diamond stocks (essential for machine tools) and, with the help of two Dutchmen, to bring them safely to England.
SIS’s relations remained frosty with the Dutch government-in-exile after the Germans occupied the Netherlands. From the summer of 1940, nevertheless, the Service managed to slip in a few agents. A Dutch naval officer, Lodo van Hamel, for example, was dropped in by air on the night of 28-29 August. Some went in by sea, including Pieter Tazelaar, put ashore at 4.35 a.m. on 23 November at Scheveningen near the seafront casino in full evening dress and smelling of alcohol, wearing a specially designed rubber oversuit to keep him dry while landing. Rather than leaving him ‘somewhere on the dunes’, the aim was for him immediately to be able ‘to mingle with the crowd on the front’. Having landed on the beach, his colleague Erik Hazelhoff sprinkled a few drops of Hennessy XO brandy on him, to strengthen his ‘party-goer’s image’.10 But without reliable contacts inside the Netherlands these agents were all sent in blind, which made the effort especially hazardous. Van Hamel, for example, was soon captured and executed by the Germans. Of fifteen agents sent in over the eighteen months from June 1940, all but four lost their lives, a fate which intensified Dutch ill-feeling towards SIS. The Special Operations Executive, meanwhile, began working actively with the Dutch, producing a considerable amount of intelligence. When John Cordeaux became responsible for operations in the Low Countries in the spring of 1942, he found that SIS was ‘in the humiliating position of obtaining no information ourselves, but being graciously allowed to pass on . . . intelligence obtained by S.O.E.’. Cordeaux, however, managed to establish a fresh SIS liaison with the Dutch government-in-exile, and after the nightmare of Operation ‘North Pole’, when during 1942-3 SOE’s Dutch operations were disastrously penetrated by German intelligence, he was able (in conjunction with SOE) to build up productive networks over the last two years of the war.11
The SIS operation in Belgium, which had been under heavy pressure in the immediate prewar period, expanded rapidly after the outbreak of the war. Under what was known as the ‘13124 Plan’, Edward Calthrop, the head of station and Passport Control Officer in Brussels, began to recruit agents initially in the east of the country to give warning of any concentration of German troops across the frontier. Should Belgium be invaded it was hoped that the network could continue as a stay-behind organisation. The head agent was the sixty-year-old Walthère Dewé, who had run the celebrated Dame Blanche organisation during the First World War. His agents in 1940 seem to have been well selected and at least two of them survived to operate after Belgium was occupied. They were given radio transmitters and some rudimentary Morse code training, and were instructed to send their signals blind. Confirmation of receipt was to be provided through coded messages in talks and poetry readings broadcast by Radio des Beaux-Arts, an SIS Section VIII venture operating from England but ‘purporting to be a small broadcasting station financed by philanthropic old Belgian ladies desiring to bring something of beauty to an ugly war clouded world’. Sustaining this fiction proved difficult after the occupation as the technicians in Britain had to make sure that their transmissions did not overrun the German-directed electric power cuts imposed across Belgium. This perhaps over-elaborate system was not a great success, but Gambier-Parry extracted two useful lessons from the experience. ‘Ab initio training of an agent who had no knowledge whatever of signalling’ was not ‘a practical proposition’, and personnel would have to be recruited for wireless communications in the field with some existing signalling experience. The second lesson was that stay-behind schemes were inherently problematic. Gambier-Parry noted that the operation planned in Belgium ‘under pre-invasion conditions suffered the handicap that these conditions became revolutionised after occupation by the enemy’, and that the ‘subsequent infiltration’ of agents was a much more successful procedure. Inevitably this was fraught with danger. The first agent, Henri Leenaerts, was flown in on 18 August 1940, but the Lysander which carried him was unable to land and was lost on the return flight, killing the pilot and agent.
In fact, the most productive source of intelligence from inside occupied Belgium came from a partial revival of La Dame Blanche under Dewé, who ran a network successively called ‘Cleveland’ and later ‘Service Clarence’ until he was captured and shot by the Germans in 1944. Their intelligence covered a broad range of subjects, including information on aircraft hangars; troop and train movements; the state of Belgian industry; reports on the movement of munitions and munitions depots in Belgium (as well as some in neighbouring Germany); drawings of fortifications around Zeebrugge; and reports on the effects of local RAF attacks on German defences in the Low Countries.12 In May 1941 it was reported that the Cleveland organisation was ‘the only one now working in Belgium’, but by the end of the year three further networks had been established with courier lines working into Spain. These doubled as escape lines for British service personnel, which, although important and valuable, markedly increased the problems of security. As with other Allied countries, SIS operated in Belgium in co-operation with the government-in-exile in London, though here, too, there were difficulties arising from internal rivalries and the fact that the Belgian King and part of the Cabinet remained in Belgium. While the Belgian Sûreté de l’État developed a close liaison with SIS, the military Deuxième Bureau was more interested in sabotage and resistance, if not also (as some claimed) plotting to re-establish the King with a crypto-Fascist regime. During 1942 additional trained wireless operators were infiltrated and the supply of intelligence improved. In April the Air Ministry noted that reports from Belgium during February and March ‘had added very considerably to the picture of the German night fighter organisations’, while in June (echoing the achievements of La Dame Blanche a quarter of a century before) the War Office commented on the ‘accurate, comprehensive railway information’ being regularly provided by SIS sources which ‘very frequently provides important confirmation of troop movements reported by other agents’.13
The main SIS concern in France was Biffy Dunderdale’s 45000 network and his very close relationship with the French Deuxième Bureau, and subsequently with the Cinquième Bureau, formed at the outbreak of war to oversee clandestine intelligence work. Dunderdale was SIS’s chief link with the Bureau head, Colonel Louis Rivet, and with Colonel Gustave Bertrand, who ran the French cryptanalytical branch. Although he had recruited agents to gather German military intelligence, Dunderdale was unable to keep them going in the chaotic weeks following the outbreak of hostilities. There were, he wrote, ‘insurmountable difficulties in travelling across frontiers and even in France itself ’, due to the ‘very severe control and complete disorganisation of civilian traffic’. Over the winter he came to depend almost entirely on information from Rivet’s organisation. Following the rapid German advance into France after 10 May 1940 Dunderdale and his staff had to withdraw from Paris. On 10 June he found himself near Orleans, quartered next to the largest ammunition dump in France, and cheerfully reported to Menzies the observation of the French colonel in command that if the dump went up it would result in an ‘instantaneous and painless death’. But the French were beaten and Menzies called the SIS team home.
Partly due to French sensitivities, stay-behind preparations in France were continually frustrated until May 1940, by which time it was too late to achieve anything meaningful. At the last minute the head of Dunderdale’s Special Communication Unit managed to organise a wireless link with Bertrand, which in the immediate aftermath of the capitulation was the only direct SIS contact with France. A further difficulty lay in the fact that the majority of SIS’s French contacts were with men who continued to serve under the Vichy regime which, after the surrender to the Germans in June 1940, survived as a quasi-independent government responsible for a large part of southern and south-eastern France. The Vichyists were vehemently opposed to the Free French General de Gaulle, while in their turn the Free Frenchmen (who inevitably constituted a highly promising source of potential agents) heartily despised Vichy and all its works. Faced with this conundrum, and under immense pressure from the government and service ministries to get intelligence out of France, Menzies deputed Dunderdale to head Section A.4 and continue to foster contacts with Bertrand and his Vichy colleagues, while Commander Kenneth Cohen, who had been in Paris as part of Dansey’s Z Organisation, was made head of Section A.5 and deployed to work with the Free French. Although in the anxious summer days of 1940 this division of responsibility may have seemed to Menzies to be the only option, the potential for competition, disagreement and friction was very great, and was heightened by the fact that, while Cohen was responsible to Dansey (the Assistant Chief), Dunderdale reported directly to Menzies himself.
By late June 1940 the War Office was urging SIS to get agents into France as soon as possible. Clandestine pick-up operations and insertions started as early as 18 June, when Major Norman Hope of Section D led an abortive mission to rescue General de Gaulle’s wife. On 23 June, three men were landed on the French coast to ascertain the morale of the people. But, as Major Hubert Hatton-Hall, head of SIS’s Army Section, argued, wherever the target area was, the ‘important thing is to enlist agents who belong to those areas & who know their way about & to provide them with WT communications’. None of this could be magicked out of thin air, but in late July Commander Frank Slocum’s new Operations Section landed Dunderdale’s first A.4 agent, a young cavalry cadet called Hubert Moreau, the son of a French admiral, near Douarnenez in western Brittany. Moreau stayed for four days, and, operating out of Mylor Creek in Falmouth Harbour, Cornwall, he went on to make two more trips in August and September.14Above and beyond whatever intelligence was acquired, the expeditions demonstrated that, although difficult, it was possible to infiltrate agents by sea, and contacts made in these early days contributed to the marine reporting system which developed along the French coast.
Evidently hoping to raise spirits at Christmas 1940, this letter from Menzies to Archie Boyle of Air Intelligence, asserts that the festival had been abolished ‘by German authority’.
Clandestine landing by sea was not the only option and on 20 August 1940 the Royal Air Force formed No. 419 Flight at the North Weald fighter station to assist in the delivery and extraction of SIS (and SOE) agents in enemy territory. New aviation techniques had to be learned by the unit employing the limited available aircraft (obsolete Whitley bombers and army-support Lysanders). The first attempted operation pre-dated the formation of the flight by several days. In a report prepared in April 1942 Claude Dansey summarised it: ‘In August, 1940, our first effort was made of landing an agent by Lysander. We were inexperienced, we were groping in the dark, and the first essay was a failure. What happened has never been known with exactitude as the plane did not return and we never heard of the agent or his wireless set.’ It was a bad start but things soon improved; for the period October 1940 to August 1941, ninety operations were planned, of which forty-eight were cancelled before take-off. Of the remaining forty-two, four were unsuccessfully attempted, with the remaining thirty-eight involving the delivery of fifty-seven agents and the return of eighteen with the loss of three aircraft. It was not just a case of transporting people and delivering supplies; the operations were able to bring important documents back to Britain from the field.
Through the summer and autumn of 1940, the intelligence return was nevertheless disappointing. In November all three service liaison officers in SIS complained about the meagre product. The head of the Air Section, Group Captain F. W. Winterbotham, declared that ‘with the exception of some reports from deserters and Free Frenchmen who have managed to get across to England, reports from agents on the German Air Force in France and Belgium have been practically nil’. Dansey responded by saying that to give a list of agents put back into occupied France would convey ‘little or nothing’ in itself, though it ‘would show a record of disappointments’. The main problem was one of communications; satisfactory radio links had not yet been established and other means of communication, for example through Switzerland, meant that information when received was ‘completely stale’.
Matters had improved by early 1941, when Dunderdale’s organisation had agents and postboxes in occupied and unoccupied France. One important network (code-named ‘Fitzroy’ and later ‘Jade’) was set up by Claude Lamirault, who had been active in prewar extreme right-wing and anti-Communist circles. John Cordeaux wrote that he was ‘as tough as any Chicago gangster’ and ‘rather an “ugly customer”’ (so much so that SIS ‘had certain misgivings about him’), but his ‘burning hatred of the Germans’ evidently outweighed his pro-Fascist political leanings. For the first two years of the war, moreover, the Russo-German non-aggression pact could provide anti-Communists with an additional justification for resisting the invader. Already versed in clandestine activity, Lamirault recruited agents ‘from the most varied walks of life’, including government officials, leading businessmen, factory workers, railway officials and prostitutes. Initially these agents provided information on economic and railway subjects.
Another network, code-named ‘Johnny’, provided good intelligence about the naval base at Brest, from March to June 1941 reporting on the positions, condition after bombing and state of seaworthiness of the German battle cruisers Scharnhorst and Gneisenau, while ‘Felix’ was the first to report the arrival of the heavy cruiser Prinz Eugen in June 1941. Human intelligence of this type was especially valuable, as it was only from the early summer of 1941 that the code-breakers at Bletchley Park were able to crack the German naval and dockyard Enigma cyphers. But the A.4 networks suffered a series of setbacks in 1942. In February Felix was disbanded after the capture of its wireless operator, and in April the Germans raided the Fitzroy headquarters at Sartrouville, capturing another radio operator. Lamirault was back in England at the time and insisted on being dropped blind into France, where he salvaged the remains of his group and managed to evacuate six important agents out through Spain and Gibraltar.
After the fall of France Dunderdale re-established contact with Rivet and Bertrand, who, although continuing to work for the Vichy regime, offered to supply SIS with information. Dunderdale, travelling as ‘John Green’, flew to Lisbon where on 5 September 1940 he met Rivet’s representative, ‘Victor’, in the church of St Geronimo. Victor expressed his willingness to work on a reciprocal basis and also help the British with potential sabotage work in Spain and Libya. Dunderdale gave him a wireless transmitter and codes and reported to Head Office that for him reciprocity meant sending the French ‘harmless stuff’ while ‘exploiting every opportunity of obtaining information’. Dansey, responsible for developing work with the Free French, meanwhile warned Menzies that liaison with anti-Gaullists could cause friction and political complications (as in fact it did), but SIS’s contacts with the established French intelligence community had been so close and productive that, as Menzies put it in the spring of 1941, ‘although my policy was queried, I insisted on renewing W/T contact with the French Deuxieme Bureau’, after the fall of France.15 It was a good decision, since through his unique position Bertrand (or ‘Bertie’ as he was referred to in telegrams) was able to provide advance warning of Abwehr and Gestapo intentions against SIS agents, as well as information on Italian and German troop and ship movements to North Africa, German dispositions in the Balkans and losses on the Russian front.
The false passport used by ‘Biffy’ Dunderdale (travelling as ‘John Green’) when meeting a representative of the Vichy French intelligence service in Lisbon in September 1940.
Back in England Kenneth Cohen at the head of A.5 had the difficult task of liaising with Colonel André Dewavrin - whose nom de guerre was ‘Colonel Passy’ - the head of the Free French Forces’ Intelligence. Agents infiltrated into France through Spain included Gilbert Renault, an ardent Gaullist, who by mid-1941 had started networks along the French Atlantic coast and established a courier line to Spain. One agent, employed by the Germans in Brest, sent him complete plans of the harbour defences and reported movements of German naval craft. Renault’s organisation, which became known as the Confrérie de Notre-Dame, eventually covered most of France. Another extensive network was run by ‘Navarre’, a right-wing hero of the Great War and leading member of the veterans’ organisation Les Anciens Combattants. Using war veterans’ clubs and various charity enterprises as cover, by mid-1941 his organisation (‘Kul’) had spread into much of the unoccupied zone and, to a lesser extent, Brittany, parts of northern France and Paris, as well as having agents in French North Africa. ‘Most of these agents’, noted Cohen, were ‘what might be termed “enthusiastic volunteers” of the officer type, rather than trained spies’, and they were consequently sometimes ill prepared for the efficiency of the German security forces in France. Navarre himself was arrested in July 1941, but, thanks to the efforts of Marie-Madeleine Meric (later Fourcade), Kul continued to function. With Fourcade taking the code-name ‘Hérisson’ (Hedgehog), and later under the new name ‘Alliance’, it established its headquarters in Marseilles.16
By the end of 1941 A.5 had established particularly good coverage of the French Atlantic ports, where German U-boats and commerce raiders were based. But the growing size and consequent vulnerability of the networks resulted in a number of them, including the Confrérie de Notre-Dame, being penetrated by the Germans. Into 1942 there was a growing list of arrests, especially of wireless operators, but the Confrérie and Alliance nevertheless managed to provide a large volume of information from most parts of France. This was passed to the United Kingdom by wireless, by couriers through Spain, by sea operations to the Brittany coast and by aircraft pick-ups (of which by mid-1941 there was about one every month). Probably the most important intelligence was the very detailed coverage by a Confrérie cell of the German radar station and its defences at Saint-Bruneval, north of Le Havre on the French Channel coast. Along with air reconnaissance, this helped enable the successful British raid in February 1942 in which precious examples of the German equipment were captured. 17
During 1942 Fourcade’s Alliance organisation was threatened by the activities of ‘Bla’, a thirty-nine-year-old former farm manager in Normandy who had lived in France for about twenty years before the war. Escaping from the German advance, he had made his way to England in June 1940 and been recruited as an agent by SIS. He parachuted into southern France near Pau in August 1941, intending to operate a wireless set for Alliance and establish a network in Normandy. He began to report directly to London in February 1942, but inconsistencies in his identity checks, among other things, had begun to lead London to suspect that he might have been compromised. In May SIS received reliable information that Bla had been arrested by the Germans, who were using him for a deception operation. Hoping that he might still be loyal, Menzies decided in turn to play him as a double agent through the XX (or ‘Twenty’) Committee, an interdepartmental body set up in January 1941 which ran the tremendously successful British wartime Double-Cross System. He also wanted to give Bla the chance ‘to escape from the enemy’s clutches’ and return to Great Britain. A rendezvous was arranged in July, when Bla did not appear, but one of the leading Alliance agents narrowly evaded arrest by Abwehr officers lying in wait for him. The following month, with mounting concern, London instructed Bla to hide his set and get to Spain or Switzerland where he should report to the British authorities. By this stage the alarm bells were also ringing at SOE, who knew Bla as ‘Blanchet’. One of their agents had spotted him in Lyons, and Major Maurice Buckmaster, the head of SOE’s F (Independent French) Section, told SIS that he was signalling to France to warn them that Blanchet was a ‘most dangerous individual’ and they ‘should have no dealings with him whatsoever . . . If possible suggest to him that he makes for Spain,’ he continued. ‘If he persists in worrying you you are fully authorised to dispose of him as neatly as possible.’ In September 1942 there were various sightings of Bla in Pau and Toulouse, and a chance encounter in Marseilles with another British agent, ‘Heron’, led to his capture in early October by Alliance agents masquerading as policemen.
Bla was taken to an Alliance safe house where he was interrogated by Marie-Madeleine Fourcade (among others) and confessed to ‘having given to the Boches all details known to him about us’. They first tried to kill him ‘without him knowing it’, by putting lethal drugs in his food, but this failed and merely alerted the unfortunate man to ‘the attempt we were making’. When he was killed he faced his fate with what Fourcade reported as ‘extraordinary moral courage’ and ‘astounded us by his calm attitude in facing punishment’. ‘The way he died’, she wrote, ‘did something to mitigate his past record.’ Although in her memoirs Fourcade relates that an ‘execution order’ was received from London, nothing so explicit survives in the relevant files, and her contemporaneous report asserted that over the weekend when Bla was in Alliance hands no contact was established with London. A subsequent minute, nevertheless, by the Free French Section in Broadway recorded that as the leading members of Alliance were ‘known personally to [Bla] the danger was such that eventually we instructed them to do away with him should the opportunity occur’. Bla’s fate was naturally suppressed. When his widow made enquiries about him towards the end of 1944 she was simply told that the authorities had had no information of his whereabouts ‘since 1942’. As one SIS officer minuted, ‘if any sleeping dogs should be let lie I think this is one’.
A.4 and A.5 represented passionately opposed French political opinions. The files contain accusations and counter-accusations, with the Free French accusing Dunderdale of purloining Frenchmen before they had a chance to declare themselves for de Gaulle, and Dunderdale complaining about the unwillingness of the Free French to co-operate. After Dewavrin had complained to Dansey about Dunderdale’s hostility to de Gaulle, in February 1941 Menzies tried to calm things down by giving Dunderdale the primary task of collecting intelligence from the ports in German-occupied France, authorising him to contact de Gaulle’s staff direct for any personnel required ‘so long as there is no possible overlapping’, while adding that A.4 agents ‘need not of necessity join De Gaulle, but if they can be persuaded to do so, the less bother will occur’. The maintenance of two parallel SIS French sections, while administratively inefficient, was a pragmatic solution to the intelligence opportunities which emerged after the fall of France. But it certainly ran risks. As the Beirut station warned in May 1942, ‘any suspicion that [the] British were in contact with Bertrand would sow deepest seed of distrust in heart of even our best friends among Free French’. Dunderdale also provided SIS liaison with Polish intelligence organisations in France, drawing on contacts he had made before the war through Bertrand. This also irritated the Free French, who complained that the Poles were not only allowed to operate much more independently than they were, but were also permitted directly to recruit French agents of their own. Towards the end of 1941 the Polish F2 network in France had 210 French and only forty Poles. But it was very productive. The Marine Section, based in Bordeaux under ‘Doctor’, noted German submarine movements out of Bordeaux, Brest and Le Havre. Doctor also ran a cell called ‘Italie’, which reported German troop movements on Italian trains. It was the collation of information about troop transports and a tank division going south with a report about desert-war training in a Prussian camp that first suggested to London that the Germans might invade North Africa.18
The threat of invasion and conditions in Germany
For about a year from the fall of France SIS was under sustained pressure to provide intelligence on the likelihood of a German invasion of Britain. What the Service was able to glean can be followed in the reports for the Foreign Office and the service ministries, which from May 1940 were also supplied to Desmond Morton for the Prime Minister. The available human intelligence fell into two main categories: detailed eyewitness observations of German operational preparations, and higher-level information, mostly from Berlin, about German strategic intentions. In July 1940 agents reported the arrival of paratroop battalions in Norway and Belgium, including a Landungskorps (special air landing force) which had been transferred from Austria. In September Menzies sent Morton (for Churchill) a very detailed series of reports from an A.4 agent who had just made his second visit to northern France. The agent had cycled along the north coast of Brittany and returned with maps showing German garrisons, gun-emplacements, airfields and troop concentrations, photographs of German activities at Douarnenez and Lorient, and reports of German soldiers practising amphibious landings. On into the autumn, agents’ reports confirmed the probability of an invasion. On 30 September fully equipped German troops were observed to have embarked on seven cargo ships at Bayonne, only to disembark two days later. On 20 October the German War Ministry ordered that three Einsatztruppe (special contingents) were to be despatched, one to Oslo and two to The Hague, each contingent being accompanied by two English-speaking interpreters. The Navy Section weekly summary for 21-28 October noted ships of four and five thousand tons lying in Hamburg harbour waiting to take on troops, and on 4 November a French agent watched a German invasion exercise at Étaples, using rubber boats with ten to twenty soldiers in each.
The best of the higher-level reports about possible invasion came from Menzies’s ‘very well placed and reliable German source’. This was ‘A.54’, Paul Thümmel, who had offered his services to the Czechoslovaks in 1936. They continued to run him, first from the Netherlands and later from London after Colonel František Moravec and the core of the Czechoslovak Intelligence Service had been spirited out of Prague by SIS in March 1939. Thümmel was a well-placed Abwehr officer who was able to provide first-class intelligence (which came via Switzerland) about German capabilities and intentions.19 Between April and December 1940, designated ‘12022/A’ by SIS, he supplied at least fourteen reports specifically concerning the invasion question, and the way in which they were processed demonstrates how SIS broadly handled intelligence at this time. The subject matter was classified as political, which meant that the primary circulation of the material was to the relevant section of the Foreign Office, in this case the Central Department. Other copies (of which, apparently, no more than thirteen were made) were sent to the Permanent Under-Secretary (who was only personally sent reports judged to be of special importance, and internal evidence suggests were for him to show to the Foreign Secretary, Lord Halifax), the PUS’s private secretary (who was sent copies of all reports), Major Morton (for the Prime Minister) and the service departments. The only copies of the reports which have survived are those sent to Morton and which he (as required) returned to SIS.
Thümmel’s first mention of an assault on Britain was in ‘an unavoidably delayed report’ of 2 May, circulated on 19 May 1940. ‘Intense preparations for air attacks on England’, he warned, ‘are proceeding,’ but the invasion itself had ‘been postponed, because all available troops are being used in Norway’. In July he reported that the invasion had been further postponed pending clarification of Soviet intentions in the Balkans. On 13 August SIS circulated a report which had been received ‘within the last few days’ from their ‘very well placed German source’. It asserted that the attack on Britain was ‘not to be expected within the next fourteen days’. An expeditionary force was being assembled in Paris, Brussels and The Hague, but the troops would not be ready for at least three weeks. This intelligence was clearly of interest at the highest level. ‘P.M. saw Lord Halifax’s copy,’ noted Morton. On 19 September Thümmel reported that the ‘fact that the attack on Britain had not yet taken place’ was ‘due chiefly to weather conditions and secondly to the British air raids’. This report was circulated on 21 September, just two days after being made, a remarkably quick process repeated in the next report, five days later, which also put the delay down to bad weather. By October Thümmel was reporting that the attack would not now be until ‘early 1941’, and on 28 December, having been present at a speech given by Field Marshal Keitel (Chief of the Oberkommando der Wehrmacht - the German Armed Forces High Command), he said that Hitler had decided on a spring invasion of Britain, perhaps at the end of March 1941. Keitel also forecast an attack on Greece in support of the Italians, but ‘said he would let the Italians stew till the spring as the main German purpose was to syphon off British troops from North Africa’.
Thümmel was not the only person reporting about Germany in 1940. Regular reports on morale and other matters were assembled from a number of sources. A report in September contained information from two ‘North Europeans’ (most probably Swedes), who had visited Berlin, and an ‘eminent Swiss physician’, who had visited Munich and Vienna. A ‘German manufacturer who arrived in Budapest from Hamburg’ had said that the shipbuilding yards of Blohm and Voss had been completely destroyed by air raids. The following month a report entitled ‘Germany: miscellaneous indications’ drew on information from ‘a Swede of standing, who has many friends in Berlin’, ‘a traveller of Austrian nationality, of a Swiss firm’, ‘the correspondent in Finland of an important German newspaper’ who had just spent a month in Germany and ‘a well placed neutral in Berlin’. These people were not ‘agents’, as the report explained; all their information had been acquired ‘by trustworthy indirect means, i.e. none of these individuals is a conscious source of ours’. The SIS team in Switzerland were always on the lookout for information. Late in 1940 the Geneva station was advised that an employee of the United States embassy in Berlin would shortly be visiting Switzerland. ‘She should only be asked to give information about conditions in Germany,’ came the instruction, ‘and not asked to work for us.’ Hard military information was the most difficult intelligence to acquire. On 27 December 1940 Malcolm Woollcombe noted: ‘It is piteous to find ourselves in this state of ignorance’ about events and production inside Germany. Nevertheless, one agent did get through at this dark hour, a Yugoslav ex-naval officer, ‘Rauf ’, who had been recruited in Trieste. Rauf visited Berlin in late 1940 and (according to his postwar medal citation, now the only record of his work) reported on the Junkers aircraft factory, and also ‘gave important information on a newly-established aerodrome in the vicinity’.
Menzies’s response to an enquiry about the identity of the German propaganda broadcaster, ‘Lord Haw Haw’.
Once the Germans had occupied western France down to the Pyrenees and the Spanish frontier, it was widely feared that Spain would be their next victim. The right-wing government of General Franco, moreover, provided an accommodating environment for pro-German groups and SIS found itself having to devote considerable efforts to counter-espionage work. There was also the problem of dealing with Sir Samuel Hoare, ambassador in Madrid from June 1940 to December 1944, whose prickly combination of knowledge and authority (as both a former Foreign Secretary and a former intelligence officer) with a tendency to windy overreaction made him a difficult colleague. This was demonstrated in September 1940 after Hoare was told by his own ‘secret sources’ of an alleged ‘confession’ to the Spanish Security Police by three men posing as Belgians that they were British agents working for the Passport Control Officer (and SIS representative). Although the men were ‘plants’, apparently engineered by Falangist (Spanish Fascist) and German elements, Hoare, fearing that the incident could escalate into a major anti-British demonstration, panicked and ordered the Passport Control Officer to leave the country. He also wanted the newly arrived SIS head of station, Leonard Hamilton Stokes, to go; SIS to cease all work (especially along the frontier with France where Hamilton Stokes had been setting up an early-warning network against a possible German invasion) and to burn its archives; and the Passport Control Office to be run on a skeleton basis, dealing only with legitimate passport matters. Hamilton Stokes complained to Menzies that his ‘position vis-à-vis YP [the ambassador] has become extremely difficult and embarrassing and reputation of S.I.S. has suffered’. Menzies responded with cool good sense. ‘A mere statement unsupported by reliable evidence that certain persons are working for the Passport Control Officer’, he signalled, ‘should neither perturb the Embassy nor the Spanish authorities who happen to be friendly or neutral. This is just a bluff by the Germans to get our organisation shut down and YP has fallen for it.’ Hamilton Stokes, he said, should ‘tactfully’ point out to Hoare that the men had ‘never been employed by us’ and that as the Passport Control Officer’s name was well known, he was ‘naturally a scapegoat’.
In the end the Passport Control Officer went home and Hamilton Stokes grumpily stayed on in Spain. Not only were relations with Hoare difficult, but he also had to work with the naval attaché, Captain Alan Hillgarth, an adventure novelist and personal friend of Churchill, who played a central role in collecting intelligence about enemy submarines in Spanish waters (Hoare described him as ‘a veritable sleuth’),20 and was permitted to communicate directly with Menzies in London. Over the winter of 1940-1 Hamilton Stokes endeavoured to build up his networks, exploring possibilities with, for example, opposition elements in the Confederación Nacional del Trabajo (National Labour Confederation). The potential for the public embarrassment (or worse) of the British in Spain if contacts such as these became known was another factor in Hoare’s nervousness about SIS activities. This was exacerbated by Hamilton Stokes’s liaising with the British military authorities to set up escape lines for Allied servicemen out of France. Already in August 1940 Frank Foley had raised the possibility of working with the Basques who had ‘a complete S.I.S. organisation started during the Spanish Civil War’ with excellent secret contacts and routes across the French frontier. But, as Foley delicately put it, arrangements with any such groups raised ‘certain difficulties caused by considerations of higher policy’, especially as the Basques, it seemed, in return for co-operation would demand British support for their political claims.
In the spring of 1941 SIS representation in Spain was boosted by the addition of a Section V officer, Kenneth Benton, to work under Hamilton Stokes. But, reflecting the animosity between Dansey and Valentine Vivian in London, Benton was separately instructed by the former to concentrate on developing communications across the Franco-Spanish frontier and stay-behind networks in Spain, and by the latter to devote himself to counter-espionage duties. After some irritable exchanges, Hamilton Stokes was confirmed as in overall charge of SIS activities in Spain, while Benton focused on the developing work in the territory south of Madrid, as well as counter-espionage across the whole country. Meanwhile Hillgarth (who doubled up as the SOE representative in Spain) had set up his own counter-espionage system, code-named ‘Secolo’, targeting German attempts to sabotage British ships in Spanish ports. Some positive work was at last being done. By the early summer of 1941 Hamilton Stokes had become cautiously optimistic and reported that over many months of patient and difficult work he had gradually re-established Samuel Hoare’s confidence in SIS.
But it was not to last. Two high-profile incidents disturbed the harmony, the first of which was the Claire case. Early in 1941, ‘Paul Lewis Claire’, a French naval officer who had transferred to the Royal Navy after the fall of France, was taken on by Commander Slocum’s O Section in SIS to help run clandestine sea operations in French waters. By July Claire was in Spain where he was arrested at the frontier apparently trying to get to Vichy France. Subsequently, as Hoare cabled to London on 23 July, Claire had gone to the Vichy French naval attaché in Madrid, to whom he had ‘divulged many secrets of British S.S. in Spain and elsewhere’. Since ‘French Embassy are now trying by every possible means to get him to Vichy’, Hoare was ‘accordingly faced with the dilemma (1) capturing or killing him and running the risk of irrevocably compromising the Mission. (2) Letting him get to France with the information that will destroy our present intelligence organisation in Spain and do even greater harm elsewhere.’ Hoare wanted immediate instructions as Claire was expected to visit the embassy the following day to get his British passport. At SIS headquarters Frank Slocum advised that Hillgarth had been ‘asked to take what steps he can to intercept Claire’. He also suggested that SOE might be approached ‘to see if there is any possible step which could be taken in Spain or France to liquidate Claire’ and wondered ‘whether we could kidnap Claire’s wife or any member of his family in an effort to bring Claire to heel’. Convinced he was a traitor (and Hillgarth afterwards said Claire had admitted his treachery), and aware that he possessed damaging information, including the identities of at least one agent network in northern France, Menzies said that if Claire could be intercepted, ‘belief that he may learn something more to divulge may make it possible to get him to Gibraltar by road with assurance car will wait and bring him back to Madrid. Action Gibraltar’, he added, ‘will be arranged.’
What then happened can be followed in a series of telegrams mostly from Hamilton Stokes ‘for C.S.S. only’. At 1.00 a.m. on 25 July he reported that ‘in conjunction with Y.N. [Hillgarth] I have lured Claire to Embassy and rendered him unconscious. He leaves tonight under Morphia for Gibraltar by car.’ There was good news at 6.13 p.m.: the kidnap had been ‘satisfactorily accomplished’. Menzies responded by drafting a signal for Gibraltar instructing that Claire was ‘to be arrested on arrival, charged with treason and kept under close arrest until opportunity of sending him home. On no repeat no account whatsoever is he to be allowed to escape.’ But this signal was cancelled when, early the following morning, ‘51000’ (the SIS representative in Morocco, currently in Gibraltar) informed Hamilton Stokes that ‘consignment arrived in his ?town completely destroyed ?owing to over attention in transit. 51,000 states salvage being quietly disposed of tonight . . . Damage regretted but I submit it is for best.’ In a despatch a few days later 51000 explained that during the drive south the unfortunate Claire had recovered from the drug and had begun ‘shrieking for help and calling [sic] so much attention’. His captors had ‘hit him on head with a revolver’ with the result that he had died. ‘It could not have been a worse affair,’ wrote Hoare to Anthony Eden. ‘As to ignorance of conditions the airy suggestion that we should intercept him in his journey to France shows how totally ignorant S.I.S. are of work in a semi hostile country stiff with German Gestapo. As it is with great personal risk to ourselves and still greater to the existence of the mission we have got rid of him - he is dead.’ Though SIS was certainly at fault - Menzies frankly admitted that Claire should not have been sent to Spain - this criticism was less than fair. Hoare had himself raised the admittedly risky possibility of kidnapping Claire, and he later told Cadogan that ‘had it not been for Hamilton Stokes’ very efficient service, the man would now be in Vichy giving away every kind of secret to the Germans’. Rather ignoring the fact that Hamilton Stokes was himself an SIS officer, Hoare asserted that ‘once again we have had here to save S.I.S. from a catastrophe’.21
There was some embarrassing political fallout. German and Vichy French diplomats in Madrid complained to the Spanish Foreign Ministry, and on 12 August Radio France broadcast a report of the affair, asserting that as the car passed through ‘a small village in Andalusia’ the prisoner ‘tried to attract people’s attention by shouting . . . [but] the Englishmen declared “Don’t get upset, it’s only a member of the Embassy gone mad and we are taking him to a Sanatorium”’. Two days later the London Daily Telegraph, under the headline ‘Nazis invent a kidnapping’, reported ‘one of the highest flights of fancy from the Nazis’ propaganda department’, which alleged ‘a gangster-like kidnapping of a Frenchman by British secret service agents in Spain’. In fact the German story was pretty accurate, but one Foreign Office official thought that ‘the D.T. has done quite well and no doubt other papers would take the same line’, a sentiment with which Menzies agreed.22 The official story (as communicated by Commander Ian Fleming of Naval Intelligence to the British Red Cross in July 1942) represented Claire as ‘missing believed drowned’, en route to Britain on the SS Empire Hurst, sunk by enemy aircraft on 11 August 1941. After the war, in order to protect SIS, it was decided to pay Claire’s widow a pension, ‘however repugnant it may be to reward the dependants of a traitor’. As Valentine Vivian minuted about the matter, ‘It may be that “murder will out” in any case, but we stand to lose so much in this case if it does, that we ought not voluntarily to take the slightest scintilla of risk.’
The second affair which threatened to disturb SIS’s position in Spain was that of Lieutenant Colonel Dudley Wrangel Clarke, who had been running a strategic deception section at GHQ Middle East in Cairo since the late autumn of 1940. Using cover as a war correspondent for The Times, and aiming with the help of SIS to make contacts to assist his deception work, Clarke travelled to Lisbon and Madrid where, some time after calling on Hamilton Stokes, on 17 October 1941 (as the British embassy reported to London the following day) he ‘was arrested in a main street dressed, down to a brassière, as a woman’. He initially told the Spanish police that he ‘was a novelist and wanted to study the reactions of men to women in the streets’. Later he asserted that in fact ‘he was taking the feminine garments to a lady in Gibraltar and thought that he would try them on for a prank. This’, added the embassy, ‘hardly squares with the fact that the garments and shoes fitted him.’23 A telephoned press report from Madrid to Berlin which SIS intercepted said that ‘Wrangal Craker, the Madrid correspondent of the London “Times”’, had been arrested. Reflecting the ‘unusual circumstances of these times’ and the ‘working methods of British agents’, he had been discovered ‘dressed as a woman’. ‘Craker’ had ‘unusually big feet with a remarkable . . .’ and here the intercept tantalisingly broke off with an ‘indecipherable passage’. In response to anxious cables from Menzies, concerned about a possible leakage of information, Hamilton Stokes reported that while the police were inclined to the theory that Clarke was a ‘homosexualist’, the German Gestapo had intervened and alleged he was a spy. On 21 October, however, Clarke was released from prison and the embassy got him away to Gibraltar the same evening.
On arrival in Gibraltar Clarke insouciantly claimed that the ‘incident in Madrid was carefully calculated’ and ‘nothing (repeat nothing) whatever compromised’. He asserted, indeed, that the affair had usefully confirmed that his cover still held with both the Spaniards and the Germans. Clarke’s eccentricity evidently reinforced his aptitude for secret work and the episode does not seem to have affected his own position, since he returned to Cairo and went on to have a brilliant career in deception. Hamilton Stokes, meanwhile, worried that his situation might have been jeopardised, while Hoare inevitably blamed the unwelcome publicity on SIS, though the Service could scarcely have predicted Clarke’s foray into undercover work, if that is what it was.
Hoare’s ambivalence about SIS, on the one hand anxious that covert activities be limited as much as possible so as not to upset relations with the host country, and on the other supportive in some areas (especially counter-espionage) and quick to exploit the capabilities of officers on the ground, reflected a common ambassadorial attitude towards intelligence matters. Late in 1941 there was a row involving Menzies, Hoare, the Foreign Office and Hamilton Stokes over Menzies’s (and apparently Hoare’s) desire to boost the Passport Control staff in Spain to monitor suspect individuals travelling from Europe across the Atlantic. But when Hamilton Stokes injudiciously revealed London’s exasperation with Hoare over his refusal to contemplate embassy cover for extra SIS staff, the ambassador bluntly declared that ‘after the episodes of [the PCO in September 1940], Claire and Clarke I have ceased to have any faith in SIS London’. In January 1942 Hoare, asserting that ‘various S.I.S. agents’ had ‘apparently become dupes of the Germans in the war of nerves’ and were merely peddling ‘sensations’, suggested to Alexander Cadogan that he should look through recent SIS reports ‘and thus check their usefulness’. The Foreign Office view, however, was that intelligence reports from Spain had ‘been particularly good lately, and not in the least alarmist’. Peter Loxley (Cadogan’s private secretary) thought that Hoare was simply ‘running a hunt against the S.I.S. these days’ and that his ‘present strictures’ were ‘quite unjustified’. Cadogan firmly told the ambassador that over the past month SIS reports had in general been ‘balanced and accurate’.24
Later that year SIS had a striking success when (through agent networks and signals intelligence) it helped finally to neutralise Germany’s infrared surveillance system (code-named ‘Bodden’) aimed at detecting Allied shipping passing through the Straits of Gibraltar. Intelligence about Bodden obtained by SIS and the Admiralty was used by Hoare from May 1942 onwards to embarrass Franco into ordering the abandonment of the whole undertaking, which was being carried on by the Axis in Spain and Spanish Morocco. By December 1942 Menzies could happily tell Peter Loxley at the Foreign Office that he had learned from ‘most secret sources’ (signals intelligence) that Hoare’s protest of 20 October had ‘had a very healthy effect in Spain’, leading to the dismantling of Bodden.25
The intelligence challenge in Portugal was similar to that in Spain, though at the beginning the work was bedevilled by staffing problems. As in Spain, SIS objectives included monitoring German fifth-column activities, establishing a stay-behind network in case of a German invasion and penetrating occupied France, though an additional priority was Italy. Yet when, an ex-banker and MI5 officer, was sent out in June 1940 to assist the existing one-man SIS operation, he spent his first three months complaining (albeit with some justification) about the poor level of security in Lisbon, so much so that Menzies eventually lost patience with him. ‘The war is at [a] stage’, he wrote in September, ‘at which risks must be taken and the question of being compromised [must] take a back seat.’ This seems to have had some effect and in November he reported that he had established a line of agents in northern Portugal whereby people, letters or parcels could be smuggled over the Spanish frontier in either direction. In February 1941 a Section V officer, Ralph Jarvis, was sent out to take over as Passport Control Officer and build up the counter-espionage side. Jarvis and his newly appointed assistant, whose health was poor, did not get on, and the assistant was replaced by Philip Johns in June 1941.
Between 1940 and 1942, while the SIS team in Portugal failed to obtain any significant intelligence from Italy, it did manage to build up a reasonable coverage of shipping movements in and out of the country. Lisbon was also a main base for the transatlantic Ship Observers Scheme, run by William Stephenson in New York during 1941, whereby ‘observers’ were planted among the crews of neutral ships with the job of reporting on any suspicious matters. Through a combination of signals intelligence and Section V information, by 1942 Jarvis was well on his way to achieving almost complete coverage of enemy intelligence operations in Portugal, including the German Waterfront Organisation set up to obtain shipping information from Allied seamen in Lisbon. A dossier of its illegal activities was prepared which the ambassador submitted to President Salazar, with the result that the organisation was suppressed. Despite these local successes, London was not satisfied with the overall quality of the Lisbon station’s work. ‘Cut out ruthlessly redundant Economic and C.E. [counter-espionage] stuff, and concentrate on essential, viz. Armed Forces and Italy,’ Menzies instructed Johns in April 1942. The following month Claude Dansey told him that the most important requirement was the development of an organisation ‘for watching over North-West ports in 23-land [Spain]’. The British ambassador, moreover, thought that Johns was too closely involved with anti-Salazar groups, and (according to Dansey) had described some of his political reports as ‘rubbish’, so it was decided to move him, and he was replaced in December 1942.26
Some of SIS’s Iberian work was linked to that of MI9, a Directorate of Military Intelligence section formed in December 1939 to provide an organisation to facilitate the escape of British personnel from prisoner-of-war camps and to develop techniques to assist servicemen to evade capture if stranded behind enemy lines. The military disasters of the spring and summer of 1940 resulted in the capture of thousands of British troops, while others sought sanctuary by crossing the frontier into neutral Switzerland or Spain. Members of Allied aircrew shot down over enemy territory added to their number. Embryonic escape lines soon began to establish themselves in the enemy-occupied Netherlands, Belgium and northern France and similar, if less dangerous, activities began in the Vichy-controlled southern zone. The work was among the earliest forms of resistance to occupation and relied heavily on the zeal of the population and a few intrepid British servicemen, such as Seaforth Highlander subaltern Ian Garrow, who declined to escape across the Pyrenees in order to set up a sound and effective network in France.
Inevitably word of these activities reached SIS. An authoritative work on the subject states that Menzies met with the head of MI9 on 6 August 1940 and ‘offered to set up an escape line to run from Marseilles into Spain’.27 Unfortunately no record of the meeting has survived in Menzies’s (somewhat sparsely maintained) appointments diary, but there is no question that SIS took a keen and early interest in the lines that had begun to stretch from the Low Countries through the occupied and unoccupied zones of France and across the frontier into neutral Spain, and along which intelligence as well as people could travel. Claude Dansey managed the SIS side. He maintained his customary caustic view of operations and frequently gave the impression that his engagement was as much to deny any other government department the opportunity to meddle on the Continent as it was to rescue British personnel. He employed a small group of army officers on MI9’s books, such as J. M. (Jimmy) Langley and Airey Neave (later to become a Conservative MP and be murdered by Irish terrorists), who possessed first-hand experience of escape and handled the organisation’s affairs in London with skill and sensitivity. Other SIS/MI9 officers, such as Donald Darling and Michael Creswell, were employed in Spain, Portugal and Gibraltar to oversee the end of the lines.
MI9 attracted a remarkable collection of characters. The range extended from the outstanding bravery of men and women such as the Belgians Albert-Marie Guérisse (‘Pat O’Leary’) and Andrée de Jongh, to the despicable behaviour of the British traitor Harold Cole, whose treachery led to the deaths of some fifty escape-line helpers. The bulk of the work was done by men, women and sometimes children, who offered safe houses and acted as couriers for the extremely dangerous business of moving Allied servicemen through enemy territory and across closely guarded borders. The German security forces tended to treat the organisation as seriously as any of SIS’s intelligence-gathering networks, and the penalties for anyone arrested were as grave as those for being caught carrying out espionage.