Six days before that drop, Allied forces had landed in Sicily. But the landings did not see A Force terminate the channel provided by Gabor Adler’s wireless set. On 20 July, after considering different projects to invade Sardinia, capture Rhodes, and land in southern Italy, Allied commanders decided to put American and British forces ashore at Salerno, south of Naples, and across the Straits of Messina. This decision led to the execution of a new deception plan, codenamed ‘Boardman’, which sought to protect those landings by weakening Axis forces in southern Italy. One strand of the plan, to quote from A Force’s records, was to heighten the threat of an Allied invasion of Sardinia. This, it was hoped, would help ‘anchor’ enemy forces there and draw reinforcements from the mainland.62 To help with this, A Force sent a series of fresh messages to Adler’s set requesting intelligence and urging readiness.63 It also introduced another theme: the spectre of Emilio Lussu.
The idea of telling the Italians that Lussu was on Sardinia had been suggested by Cecil Roseberry: as well as ‘hinting broadly at an invasion’, he felt, the ruse might persuade the Italians to go looking for Lussu and divert attention from wherever he really was in France or Italy.64 A message was sent to Sardinia stating that a ‘well known’ Sardinian anti-Fascist, with his own communications and organisation, was now on the island making preparations for immediate action in the event of large-scale Allied operations. ‘You will recognise him by small grey beard.’65
By September, with Allied landings on mainland Italy imminent, A Force was trying to cause ‘as much alarm as possible’ over Adler’s wireless set.66 ‘Is your group now ready for action?’ read one message sent to Sardinia. ‘Most important have all arrangements completed within ten days.’67 ‘Inform Serra and his friends to use password DOGFOX repeat DOGFOX to any Allied Army Field Security Police or Intelligence officers,’ read another.68 A final message included an assurance that Allied troops would arrive ‘as liberators and not conquerors’.69 Two days after that, Italy surrendered and the Sardinian end of the wireless link went dead.
For months after their capture, SOE heard nothing about the subsequent fate of Gabor Adler and Salvatore Serra. In October 1943 a report was received that the pair had been in prison in Rome when Italy surrendered. That claim was heard again in the summer of 1944 when advancing Allied forces overran the city. ‘If my memory is right, they should have been found to have been imprisoned in the Regina Coeli prison [in September 1943],’ Mario Bertacchi, the former SIM counter-espionage chief, told SOE; ‘I do not know where they are now.’70 Both men were listed officially as missing.
Once the war had moved on and the dust had settled, ‘Barclay’ was assessed to have helped fan Axis fears that an Allied attack on Sardinia had been a genuine possibility during the run-up to the Sicily invasion. ‘Boardman’, too, was judged to have worked well. But though A Force assessed that the channel provided by Adler’s set on Sardinia had been ‘used to good advantage’ and paid ‘a good dividend’, care should be taken not to exaggerate its contribution.71 Like Operation ‘Mincemeat’, it was merely one of a host of efforts aimed at manipulating the enemy’s assessment of Allied intentions in the Mediterranean in 1943.
In fact, it was not even the only radio game played on Sardinia with a captured Allied wireless set. Shortly after midnight on 1 July 1943, paddling ashore from three US Navy PT-boats, a five-man Secret Intelligence team of the American OSS landed, soaking wet, at the bottom of a cliff on the island’s northwest coast. OSS was a latecomer to Mediterranean operations and had embarked upon its anti-Italian planning with a degree of overconfidence and misplaced ambition well illustrated by the fate of those five men. After scaling the cliff and starting inland, this little party – all dressed in US uniforms – ran straight into an Italian observation post and were captured. Under pressure and threats from Cesare Faccio, the same SIM counter-espionage officer who was at that moment doubling Gabor Adler’s wireless set, the team’s wireless operator, Lieutenant Charles Taquey, gave up his signal plan. This allowed the Italians to double his set, too, but Taquey’s plan included a secret sign to indicate capture and OSS spotted it. Enemy signals intercepted later by Bletchley Park strongly suggest that Taquey’s set, too, was used as a channel for Allied misinformation. In August, one message sent to Berlin by the Rome office of the German Abwehr reported that, according to Italian intelligence officers, ‘the American I.S. [Intelligence Service]’ had recently instructed its agents in Sardinia to gather information about German and Italian troops and coastal and air defences, organise reception committees, explore local attitudes to the Allies and bribe officials.72 An ‘American attack on Sardinia’ was expected soon, Berlin was told.73
Despite being one ruse among many, the ‘tripling’ of Gabor Adler’s wireless set worked. Later testimonies of SIM personnel confirm Italian confidence in having controlled it and the belief that the stores dropped by parachute had been genuinely intended for him and Serra. It is evident too that the messages received over the set caused the Italians considerable alarm. Still in Sardinia when the Armistice was announced, the five OSS men captured in July were delivered safely into Allied hands on 16 September and were soon recounting their tale. ‘During our captivity,’ Lieutenant Taquey wrote in an account apparently never shared with the British, ‘the Italians proved to be afraid of a Separatist Movement that would have been helped by the Allies. From their radio contact with the British they got the impression that Lussu, the pre-fascism champion of Sardinian Autonomy, was to be smuggled into Sardinia.’ One result of this news, Taquey reported, was a raid ‘conducted by Faccio’ that saw the arrest of ‘minor people’. Another result was that SIM began to wonder whether the luckless OSS team, too, was in on the Lussu plan. ‘Faccio was persuaded that [Antonio] Camboni [the team’s Sardinian-born leader] knew everything about it and he tried to make him speak,’ Taquey recalled. ‘Camboni maintained that he did not know anything definite but that chances were that Lussu who is rather old and unable to stand physical hardships would not be found in Sardinia.’74
‘We get the final laugh out of this,’ Cecil Roseberry remarked in 1945 when learning of Sergeant-Major Silvestri’s belief that SIM had deceived the British. ‘Far from our being deceived by the doubled wireless set, we pinned our ears back right from the first message received and limited our traffic to innocuous stuff until – very soon – we were completely satisfied that the set was being worked by the Italians.’ Roseberry noted that Adler himself had provided the first clue ‘by referring to Serra by that name instead of Pisano, the name by which he was known to us. We like to think he did this deliberately.’75 Later, an A Force summary of the case recorded that Silvestri’s interrogator, ‘with a nice sense of delicacy towards one who had proved a worthy opponent … allowed the Sergeant-Major to go his way with no further inkling as to the true history’.76
It seems clear from Italian sources that Adler had deceived the Italians about his identity. Mario Bertacchi recalled that one of the agents was an ex-Carabiniere ‘of unsavoury antecedents’ called Serra, while ‘[t]he other claimed to be a British soldier named Armstrong, but he contradicted himself, appeared unwilling to talk and it was never possible to clear up the question of his position and nationality’.77 Cesare Faccio told the British that ‘Armstrong’ had claimed to be an Englishman whose father, a British-naturalised Italian, had died when Adler was young, whereupon Adler’s English mother had married a man called Armstrong who worked for the Home Office. Her son had then taken his name.78 Silvestri, too, knew of Adler only as ‘Armstrong’. It was ‘perfectly evident’ that he was not English, Silvestri added; ‘even in writing English it was noticed that he made the most obvious grammatical mistakes. Equally it was known that he could not have been an Italian, as his Italian was very far from perfect.’ Only later, Silvestri said, had SIM ‘discovered’ that Adler ‘was a Spaniard, or at least of Spanish origin, although he may genuinely have possessed British nationality’.79
To date, no records ha
ve surfaced to suggest that the Italians knew Adler by any other name than John or Giovanni Armstrong. Nor have any documents come to light that suggest anyone ever discovered that he was Jewish. As for Adler attempting to alert SOE to his capture and fate, confirmation seems to come from comparing British records against Antonio Silvestri’s testimony. At precisely the moment when the British were receiving signals from Sardinia that omitted their security checks and referred to Serra by his real name, Adler had apparently had the opportunity to manipulate their content. Silvestri recalled that when the doubling-back began and the first signals were exchanged, ‘the original messages back to Base were drafted by Armstrong personally and submitted to Cagliari for approval. After a short while things seemed to be going so well that Armstrong and Serra were sent to Rome and the Cagliari office took over the whole thing.’80
Adler’s conduct deserves recognition. It should not alter the fact that he and Serra had been dispatched to Sardinia on a mission as dangerously hopeful as that on which Emilio Zappalà and Antonio Gallo had been sent to Sicily a few weeks before. It had not been long since Cecil Roseberry had described MI6’s mooted Sicily mission as ‘more or less a suicide job’ and expressed alarm at the idea of agents going ashore with no contacts or pre-arranged safe house. Roseberry had also warned SOE’s Cairo office about the difficulties of fitting young male agents ‘into the daily life of a country at war’. But British and Italian sources appear to agree that Adler and Serra had fallen foul of the very fact that they were so out of place. ‘A prisoner of war informed us that “British agents” in Sardinia were caught because they approached a shepherd whilst he was guarding his flock,’ reads one British report, tallying with Italian accounts. ‘No native would ever do this.’81
The extent to which the two agents had stepped into the dark is brutally underlined by the fact that they seemingly went ashore with no one expecting their arrival: all they appear to have had were the names of a few contacts who would have to be approached cold and might not even help. Chief among those contacts were a local lawyer, Salvatore Mannironi, and a veterinary surgeon by the name of Ennio Delogu. SOE apparently knew of these two simply because Dino Giacobbe, Emilio Lussu’s ‘red hot’ communist friend in Boston, had mentioned them to Lussu in 1941 as possible sympathisers. ‘Dr Ennio [Delogu],’ Giacobbe had told Lussu in an intercepted letter opened and read by the British, ‘is a great admirer of yours and one of the best men in Sardinia, an idealist who is ready for anything.’ As for Mannironi, Giacobbe described him as ‘a sound young fellow if ever there was one; he is head of the Catholic Action Party in Sardinia and cousin of my wife’.82 Both men lived in the same valley near the town of Nuoro in eastern Sardinia. It is no coincidence that Adler and Serra went ashore at a point on the coast not far from there. In his letter, Giacobbe had suggested that landing spot and detailed the route inland to the valley. Two early messages sent to Algiers from Adler’s set reported that he and Serra had been unable to contact ‘Mannizoni [sic]’.83 Algiers’ reply, sent before SOE decided that the set was under Italian control, advised the pair against making contact ‘unless carefully explored in advance. If made you can say Giacobbe is well, living in Boston and working for Giustizia e Liberta.’84 In the aftermath of the two agents’ capture, Mannironi and Delogu were arrested and imprisoned along with Mannironi’s brother and several other suspects. Possibly these were the ‘minor people’ whom OSS’s Lieutenant Taquey would hear had been rounded up by SIM. Indeed, Mario Bertacchi would later reflect that SIM’s exploitation of Adler and Serra had failed to produce ‘very satisfactory results’ from the point of view of counter-espionage:
Serra’s and Armstrong’s principal brief was to busy themselves in the political field in order to set up communications between Sardinian anti-Fascist elements and allied troops and thence procure arms and other means for them. But those Sardinians whom they were supposed to contact proved unprepared and unwilling to collaborate with the enemy or for some reason mistrustful.85
That explanation may be qualified by the fact that those Sardinians had known nothing about SOE or its plans and had no expectation of any sudden visitors from the sea.
A balanced picture of the Allies’ success in exploiting Gabor Adler’s capture must also acknowledge the terrible consequences of SIM’s success in exploiting Salvatore Serra. The British began to learn of this only when Mario Bertacchi was interrogated in 1944. Serra, he said, had admitted to his captors that the submarine that had landed him and Adler in Sardinia had also landed agents in Corsica. ‘It was therefore decided to transfer Serra to Corsica and escort him round the island in the hope of a chance meeting with one of those men. This did in fact lead to the capture of one of these agents and from this first step there followed a whole chain of highly successful operations.’86 Antonio Silvestri would tell his interrogators, too, of hearing in the SIM counter-espionage office in Cagliari ‘that Serra had been used on an expedition to Corsica, where he had been of great assistance in identifying and capturing other Allied agents.’87
Those Italian accounts were not quite accurate: the only agents aboard Ian McGeoch’s P.228 when it sailed from Algiers in January 1943 had been Salvatore Serra and Gabor Adler. But it had been the case that a couple of months later the Italians on Corsica had embarked on a wave of arrests and rounded up two Allied agents, both Frenchmen, who, weeks earlier, had been in Algiers and under SOE’s wing at the same time as Serra and Adler. Their names were Godefroy (‘Fred’) Scamaroni and Jean-Baptiste Hellier. Both belonged to SOE’s RF Section, which worked to support and supply Charles de Gaulle’s Free French Forces. The 28-year-old Scamaroni was a Corsican civil servant who had joined de Gaulle in 1940 and attempted once already to encourage resistance on the island. Hellier was Scamaroni’s wireless operator. He was thirty-six, came from Limoges, had worked before the war as a heating engineer and stove-fitter, and had fought in the French infantry against the Germans in 1940. Together with a third agent, a man named Jickell who was to work as Scamaroni’s assistant, the pair had left Algiers on 30 December 1942 aboard HM Submarine Tribune and gone ashore on the Corsican coast, twenty miles south of Ajaccio, Corsica’s capital, during the early hours of 7 January.
Scamaroni and his team had then spent two months seeking to establish themselves in Corsica and organise a resistance movement. Possible dropping zones and landing beaches were located. Sympathisers were recruited. Then, while drinking in an Ajaccio bar, Hellier was arrested. It was said afterwards that three Italians had been seen to approach him, hustle him into a car, and drive off; and that night, apparently dragging a beaten and bleeding Hellier, Italian soldiers burst into the house in the capital where Scamaroni was sleeping and arrested him, too. In time, dozens of their local helpers were rounded up. Imprisoned in the citadel at Ajaccio, Scamaroni refused to give up his secrets and suffered appalling tortures. His body was burned with hot irons. His fingernails were ripped out. Then, according to one account, he managed, while unobserved in his cell, to work a piece of wire into his neck, pass it behind his windpipe, and tear out his throat. When his body was discovered, Scamaroni had apparently written on the wall in his own blood: ‘Je n’ai pas parlé. Vive de Gaulle! Vive la France!’88 Hellier was shot in Bastia in July.
Some suspected Hellier of betraying his friends. Even a recent study of the French Resistance describes him as ‘unreliable,’ drinking too much, and ‘proclaiming he was going to give himself up’.89 This is not the character that emerges from SOE records. Before he went into Corsica, Hellier had been considered ‘a very capable and dependable man, of whose genuine loyalty and sincerity there can be little doubt’.90 After his arrest, SOE thought it ‘inconceivable that he would go over to the enemy’ unless he thought he could outwit them.91 Today, the memoirs of a SIM officer called Virginio Sias confirm that Hellier did not go over willingly. They also suggest that there be truth in a local tale that Hellier had said to a prison-mate in Ajaccio that a team of agents who had been with him in Algiers h
ad ‘sold’ him to the enemy.92 At the beginning of 1943, Sias was working for SIM in Trieste; dispatched to Ajaccio in March to strengthen the SIM counter-intelligence office, he did not arrive alone: with him, under escort, was Salvatore Serra. After being moved from Sardinia to Rome’s Regina Coeli prison, Serra, fearing execution, had accepted a proposal to collaborate actively. Sias recounts how Serra identified Hellier by the eczema on his earlobe.93
Notes
1 Report by Leading Seaman F. R. Taylor, TNA HS 6/877.
2 McGeoch’s report also explains the submarine’s temporary disappearance: ‘0540 [hours]. Boat had not returned, but it was necessary to dive for the dawn.’ Patrol Report, 25 January 1943, TNA ADM 199/1838.
3 Interrogation report on Salvatore Serra, 27 May 1945, TNA HS 9/1343/1.
4 History Sheet, TNA HS 9/1343/1.
5 C. Mackenzie to SOE London, 16 February 1942, TNA HS 9/1343/1.
6 Report by F. Jackson, 10 March 1942, TNA HS 9/1343/1.
7 Report by Corporal Morris, 25 March 1942, TNA HS 9/1343/1.
8 Report by Corporal Saunders, 2 April 1942, TNA HS 9/1343/1.
9 Report by Corporal Saunders, 10 April 1942, TNA HS 9/1343/1.
10 Report by Corporal Saunders, 30 April 1942, TNA HS 9/1343/1.
11 Report by Corporal Saunders, 22 May 1942, TNA HS 9/1343/1.
12 Report, 22 June 1942, TNA HS 9/1343/1.
Target Page 31