Secret Alliances Read online

  Relations with SOE continued to deteriorate quite sharply – to such an extent that a few weeks after he had taken over from Dalton as Minister of Economic Warfare, Lord Selborne wrote to Eden in March 1942 to express his concern at the extent of the friction, which he described as a deplorable state of affairs. He was careful not to single out either party for criticism, saying that he expected that SIS could provide a similarly formidable list of grievances to match those of SOE. He felt that the existing measures to resolve problems, consisting largely of weekly meetings between the administrative heads of the two services, were inadequate. He recommended instead a weekly meeting between the two heads of service, presided over by a conciliator, who should be someone of significant stature and authority. He attached a long minute by Jebb summarising how difficult things had become, and an even longer memo (eighteen pages) by Nelson, describing some of what SOE considered the worst examples of SIS behaviour across their common fields of operations. It is a remarkable document, both for its length – which was surely too much for a very busy minister and his senior officials – and also for the frankness with which it described the squabbles and examples of obstructive behaviour and poor faith which were damaging the effectiveness of SOE.

  In his covering minute, Jebb noted that several senior members of SOE, himself included, had either worked in SIS or had had a close association with it, so it was unjust to describe them as a bunch of amateurs. He described the extent to which SOE shared information about its activities with SIS, including their progress reports: ‘we keep nothing from them, while they keep a very great deal from us’. He criticised the ‘false beard’ mentality in SIS, acknowledged that SOE had done things which might have earned their legitimate displeasure, and emphasised that an effective liaison between them required a proper two-way exchange of information. It was for this reason that he suggested the appointment of a high-level conciliator who could preside over a weekly meeting between Menzies and Nelson, and give advisory rulings on disputes which would only be binding if accepted by both parties.

  Jebb did not go into any detail – though he might have done – about the nature and extent of the intelligence which SOE had passed to SIS in accordance with their original agreement, which would have made some of his arguments even more compelling. He could for example have mentioned the work of one of SOE’s star agents in Norway, Odd Starheim (Cheese), who, in just over four months between late February and June 1941, had provided intelligence on oil stocks throughout the whole country including petrol and aviation spirit, German troop movements, shipping, submarines, air intelligence, the Sandefjord cable, the Knaben molybdenum mines, harbour intelligence and coastal fortifications, quislings, the movements of Tirpitz and the registering of fishing boats. He also provided precise details of the effectiveness of air raids on Sola and Forus airfields near Stavanger. This followed his initial reporting which had specified the extent of the targets which were available and so led to the attacks. In a fairly unusual example of both feedback and positive comment, SIS confirmed that they were indeed interested in his reporting.9

  Nelson described the SIS view of SOE as a ‘rather ineffective and ridiculous bunch of amateurs’, who might endanger SIS and all its works if they were not kept quiet. ‘Now their attitude appears to be that we are dangerous rivals and that if we are not squashed quickly, we shall eventually squash them.’ He went on to cite a wealth of examples linked to Norway. He described how the SIS station in Stockholm had made several direct attempts to suborn contacts who had been cultivated and trained by SOE, on one occasion advising SOE that their contact was a dangerous German agent who should not be used. SOE dropped him, only to find a few weeks later that his agent network had been picked up and was being used by SIS. He claimed that another SOE agent was successfully inveigled away by SIS, who told him that his former contacts were dangerous and would land him in jail. He referred to a lack of cooperation in issuing visas to contacts whom SOE wanted to get to the UK, to avoid the risk of their arrest by the Swedes. On another occasion in late 1940, SIS had obstructed SOE attempts to interview recently arrived Norwegian refugee fishermen with a view to selecting suitable crew to participate in the first Lofoten raid – this necessitated a meeting between Menzies and Nelson to agree that half a dozen fishermen could be made available. He also quoted an incident in early 1941 when SIS had suppressed an SOE telegram to Stockholm, because they did not think it should be sent. SOE were not informed that this had happened. There were other communications difficulties too, concerning the unreliability of the radio sets provided by SIS, their unwillingness to provide sufficient frequencies and obstructiveness over an SOE request to establish a radio station in the Shetlands to facilitate their traffic with Scandinavia. This might have helped to resolve atmospheric problems which sometimes severely hampered transmissions. There was also an occasion when Linge agreed with Nagell to second an SOE agent to him for a few weeks. After his return, it was discovered that the agent had been sent by SIS to Bergen from Peterhead, at a time when SIS were using Peterhead as a separate base – without SOE being informed. There were also examples of difficulties over agreeing the availability of transport from the Shetland base, and SIS insecurity in the handling of SOE information which led to the unnecessary disclosure of some agent identities.

  While SIS and SOE were trying to find a way to resolve their differences, they were subject to a flank attack when the Chiefs of Staff put forward a proposal to amalgamate the two organisations under their control. This was not the first time that they had tried to take over SIS (it had happened during the First World War as well) and on this occasion they attempted some bureaucratic skulduggery by attempting to bypass the JIC. However, the two agencies were able to agree to resist this threat and the Foreign Office took the lead in seeing it off. A senior official, O. A. Scott, warned Brigadier Hollis, secretary of the Chiefs of Staff Committee, that ‘if the issue was forced, the Foreign Office would take up the challenge. We had the guns but I hoped that it would not be necessary to open fire.’10 But a resolution of the dispute between SIS and SOE was not so straightforward. Sir Findlater Stewart, a distinguished senior civil servant, was selected and agreed to take on the task of conciliator. Menzies objected on the grounds that he did not wish to submit to outside arbitration. A compromise was reached, whereby Menzies agreed to attend the fortnightly meeting which Nelson had at the Foreign Office, which could be expected to cover most of the necessary ground. There matters rested. Nelson retired because of ill health in May 1942 and was replaced by Sir Charles Hambro, a rather stronger character. Difficulties and frictions continued at a working level, but there was at least a functioning mechanism at the top by which the most serious problems could be resolved.11

  The extent of disruption to SIS operations caused by SOE

  Selborne was quite right when he wrote in his letter to Eden that he expected that SIS could provide a similarly formidable list of grievances to match those of SOE. In April 1943 Nagell wrote to General Hansteen (who had taken over as Norwegian Commander in Chief in January 1942), giving a list of eight different cases between April 1942 and April 1943 when SIS stations were disrupted and had been forced to close down as a result of SOE activities and operations in their neighbourhood. The stations were located throughout the country from Tromsø in the north down to Kristiansand in the south, and included Oslo and Bergen. There were stations which were affected too by the fallout from serious setbacks in Telavåg and Majavatn, both of which also led to extensive and harsh German reprisals and executions.12 Moreover, there were other disruptive incidents during this period which were not mentioned by Nagell in his letter, perhaps because they were considered to be misfortunes rather than the result of planned operations. Thus the SIS plan to resuscitate their station Koppa in the Ålesund area in March 1943 was frustrated when the Germans ‘bombed an SOE boat in the neighbourhood’.13 This was in fact the Bergholm, skippered by Leif Larsen, which was bombed and strafed while it was on its way
back to Shetland after completing a mission for SOE, landing passengers and stores on the edge of the Arctic Circle.14 Most of the crew were injured, four seriously, and one died. The Bergholm sank. Larsen and the survivors rowed ninety miles to Ålesund, and after four days he struck the point at which he was aiming within 100 yards, a remarkable achievement.15 They landed on an island close to the spot chosen by SIS for the delivery of their own agents.† SIS had to postpone their operation and, after several further delays caused by the short summer nights and other problems, Koppa was never revived. In fact, although neither service was aware of it, their stations had already got in each other’s way in this area. For a few months from the autumn of 1941, Koppa was located practically next door to the SOE wireless station Antrum in an isolated area close to the beach.

  The villa housing SOE’s Antrum radio station near Ålesund. SIS station Koppa is out of sight, just around the corner. © NHM

  Neither knew of the presence or activities of the other, and while this must have increased the danger to both because of a greater risk of being located by German direction finding or other investigations, there is nothing to show that it happened. Koppa was later betrayed by an informer in February 1942.

  An imaginative hiding place for the SIS Koppa radio, in a grave in Borgund Cemetery near Ålesund. © NHM

  SIS progress reports provide further examples from after April 1943 of stations being forced to close so as to avoid unwanted German attention following SOE activities in their area. Some of these incidents will be considered further in due course, but for now they help to explain the lengths to which SIS went to try to ensure that SOE operations were not mounted in areas where there were ongoing operations to obtain intelligence, especially along the Norwegian coast.

  The importance attached by the Admiralty’s Naval Intelligence Division to the provision of timely reporting about German naval and merchant shipping movements along the Norwegian coast, meant that they became a powerful ally to Menzies in his attempts to ensure that SIS stations were able to operate in the most favourable circumstances for intelligence gathering and that SOE had to take second place. They were not always successful, because sometimes there were oversights leading to a failure by SOE to keep SIS informed of its intentions – as happened with Operation MARTIN in Tromsø (see below). A system of consultation was later developed which, at least until the sinking of the Tirpitz in November 1944, gave the Admiralty a chance to judge the risks to SIS operations in allowing SOE operations to take place. This is well illustrated by the discussions which took place when SOE wanted to mount a further series of attacks against shipping targets (Operations SALAMANDER and VESTIGE, using motorised submersible canoes and kayaks respectively) in the area between Ålesund and Egersund in the early autumn of 1944, following some successes the previous year. John Cordeaux, the SIS naval adviser, wrote to the deputy director of naval intelligence (DDNI), setting out what SOE wanted to do. He described in great detail (to DDNI but not to SOE) what SIS equities would be put at risk in each case, and set out which operations SIS would accept, and which it could not agree to. The DDNI accepted his advice, and only SALAMANDER II and VESTIGE V were permitted to go ahead – though even then only on condition that they would be against a large target if one appeared.16 In the event, SALAMANDER II was unsuccessful – and the coast-watching SIS station Roska, near Florø, made contact with them and was instrumental in arranging for them to be rescued without endangering its own position.17

  There was, for the most part, reasonably good cooperation over the arrangements which were made for the transport of agents to and from Norway, and the delivery of supplies. Air drops were unproblematic when weather and lack of daylight permitted them, because there were fewer difficulties in this area over the availability of aircraft or prioritising missions. Occasionally Catalina flying boats were provided at very short notice to carry out rescues of agents from under the noses of the Germans. Sea transport, mainly from the Shetland base, was occasionally more problematic because of a shortage of suitable vessels. This was sometimes resolved by the less secure procedure of delivering both SIS and SOE agents on the same vessel, though to different locations, as once happened to the SIS agent Bjørn Rørholt. Difficulties arose, however, when SIS did not inform SOE of some of the trips it was planning, particularly when it was operating from a different base at Peterhead between July 1941 and November 1943. This led to fatal errors, as happened at Telavåg in April 1942. The Borghild from Peterhead delivered an SIS agent on the island of Sotra, and moored there for several days before departing. Soon afterwards the Olaf delivered two SOE agents to almost the same spot: they remained in the area instead of moving on straight away. The combination of these activities attracted German attention and led to a raid. There was an exchange of fire between the SOE agents and the Germans, which killed both one of the SOE agents and also two Germans. One of them was a senior Gestapo officer. In retaliation, the Germans blew up the entire village, and deported all the inhabitants, sending many of the men to Sachsenhausen (where thirty-one of them died), and the rest to camps in Norway.‡ The SIS station Theta, in Bergen, reported on the incident and the fate of the two SOE agents. It received a sharp reproof from SIS: ‘there will be the devil to pay when intelligence gets mixed up with other organisations’.18 (This was not an unreasonable attitude. The previous chapter illustrated what could happen when agents became involved in more than one area of activity, leading to the break-up of the Hardware station.) Nothing daunted, Theta reported shortly afterwards that the SOE Arquebus station was unable to report because its set was damaged. On that occasion, they appear to have escaped a rebuke from SIS.19

  The involvement of Theta here illustrates the other side of the coin. While at almost all levels both the headquarters organisations of both services and their mainly British-staffed stations in Stockholm were engaged in, and distracted by, arguments and ill-feeling from time to time, the same cannot be said for their Norwegian agents. They were focused on their common fight against the German occupiers and generally chose with pragmatism not to get involved in the squabbles and differences between their employers. What is more, they were quite willing to help each other if the need arose, occasionally at some considerable risk to their own safety. For example, in October 1942 the wireless operator of the SOE station Plover took over the set of the SIS station Beta and operated it on behalf of both services at a time when SIS were unable to do so themselves.20 An even more striking example involved the SIS station Upsilon in Tromsø. In March 1943, SOE despatched Operation MARTIN, a team whose task was to organise and train resistance groups in the area, with the longer-term aim of attacking German airfields which posed a significant threat to Arctic convoys. The team were betrayed and all but one of them, Jan Baalsrud, was captured. Under interrogation, the prisoners revealed the name of Kaare Moursund, the organiser of Upsilon, which led to the arrest of other resistance members. Nonetheless, the remnants of the station were able to pass messages back to SIS in London about the location and circumstances of Baalsrud after he had been caught in a blizzard for several days and was badly affected by exposure and frostbite. They also relayed messages between different SIS agents in separate mountain valleys which enabled them to arrange for him to be rescued and transported to Sweden, and to provide updates to London which SIS forwarded to SOE.21 This earned the station a warm compliment from SIS: ‘The highest British and Norwegian authorities take this opportunity of congratulating you on the calm confidence with which you handled the difficult position in which the arrests in Tromsø placed you.’§ 22

  Despite their differences, there were many occasions when both sides behaved responsibly. There are documented examples showing the extent to which they did respect the rules. In October 1942, Welsh wrote to Nagell about a report on Haugen Sigurd Ingold from Ofoten. He thought that it was interesting, and added ‘if he has already been recruited by SOE for a special job in that area I am afraid we shall not be able to secure him, but if SOE ha
ve recruited him not intending to use him in his own district, we might bring pressure to bear to obtain him for our own coast watching system’.23

  More significantly, while they may not have received much acknowledgement from SIS, or attracted more than a small amount of information by way of reciprocation, SOE continued to honour their original agreement to pass on to SIS the intelligence which they obtained, and there is nothing to suggest that any of it was withheld. While the example of wide-ranging and high-value reporting from Cheese may have been exceptional, the Norwegian SIS files are littered with plenty of examples of the intelligence passed to it by SOE. There are also examples in Admiralty files, showing aspects of SOE’s contribution to the anti-U-boat campaign illustrated by a report from SOE in Voss, quoting a German quartermaster as stating that Bergen would become the main German U-boat base.24

  What were the practical consequences for Norway? The impact of German direction finding

  There were, fortunately, no other incidents as serious as Telavåg where insecure behaviour by both services contributed to such a tragic outcome. Nonetheless, there were quite a few cases where the ill-considered actions of one service created problems for the other, even though these were sometimes offset by the pragmatism of Norwegian (usually SIS) agents, who ignored instructions from headquarters. The greater problem was at headquarters level, where too much time needed to be spent attempting to deal with friction, mistrust and a lack of information, all of which would have been distracting and energy-sapping. What might have been achieved with closer cooperation? What were the sorts of opportunities which were foregone? For the most part, we do not know enough to be able to do more than speculate. However, there is one area where there is sufficient evidence to enable us to make a reasonable assessment: security, and the losses which were caused to SIS stations in particular by effective German technical investigations and direction finding (often known as D/Fing for short). SOE knew plenty about German direction finding activities and capabilities – but did not pass their knowledge on to SIS.