The Secrets of Station X

by Michael Smith

  The simple truth is that the keys for that day were not broken until after the raid had taken place. Keith Batey, one of the Hut 6 codebreakers working on the Brown cypher, said he had

  a clear recollection of our saddest failure: although messages in early November 1940 had given notice of a special operation codenamed Mond[licht], we did not know more and to our dismay did not break any key for several days before 14 November: the result was the unhindered and disastrous bombing of Coventry. The missing keys were later broken and we could not explain the earlier failure.

  A week after the Coventry raid, Bletchley Park itself was bombed. ‘On arriving at office this morning, found it had been bombed in the night,’ Malcolm Kennedy recorded in his diary.

  Typists’ room and telephone exchange in our building blown to bits by a direct hit and the vicarage next door damaged by another bomb which landed in the garden. A third exploded in the road outside, while two more landed over at the Park, one of them bursting a bare half-dozen paces from Hut 4. By great good fortune there were no casualties. We, however, have had to give up our room to the typists and have been moved to the room used by the South American Section who, in turn, have been transferred to the Park.

  Gradually, there was more cooperation between the RAF intelligence and signals chiefs and Bletchley Park, who thanks to the reports from Cooper’s ‘Computors’ and the Brown Enigma were beginning to appreciate the value of the intelligence the codebreakers could provide. They set up a new RAF intercept site at Chicksands Priory in Bedfordshire, which took some of the burden of the Enigma traffic from the Army operators at Chatham, but it was not until early 1942, when the Bletchley Park Air Section challenged the very low Air Intelligence assessments of the number of Luftwaffe fighter aircraft – an argument borne out by Bomber Command’s lack of preparation for the German fighter defences it faced during the bombing of Germany – that it was finally allowed to issue its own daily reports to RAF customers on Luftwaffe command and control and tactics.

  CHAPTER 5

  BREAKING NAVAL ENIGMA

  While the attacks on Army and air force Enigma had produced fairly rapid results once the fighting began, the German Navy Enigma was recognised as a far more difficult nut to crack. Frank Birch recalled being told by Denniston at the start of the war that ‘all German codes were unbreakable’. There was no point in ‘putting pundits onto them’ because nothing they could do would work, Denniston said. This was particularly true of Naval Enigma. According to Hugh Alexander, who was to become the head of the Naval Enigma section, ‘when the war started probably only two people thought that the Naval Enigma could be broken – Birch, the Head of German Naval Section, and Turing, one of the leading Cambridge mathematicians who joined GC&CS. Birch thought it could be broken because it had to be broken and Turing thought it could be broken because it would be so interesting to break it. Whether or not these reasons were logically satisfactory they imbued those who held them with a determination that the problem should be solved and it is to the pertinacity and force that, in utterly different ways, both of them showed that success was ultimately due.’

  Turing himself admitted that his main reason for working on Naval Enigma rather than any other problem was because ‘no one else was doing anything about it and I could have it to myself’. He started where the Poles left off and for the first three months of the war concentrated on trying to work out the complex indicating system which was one of the factors that made Naval Enigma so much more difficult to break. He succeeded in December 1939 in solving the indicating system using Polish decrypts for the first week of May 1937, but could not make any progress against wartime Naval Enigma without a ‘pinch’, the capture of original German cypher keys or settings.

  Harry Hinsley had arrived at Bletchley in October 1939 aged just twenty, having been interviewed by Tiltman and Denniston in Charlesworth’s rooms at St John’s College, Cambridge, along with Bill Bonsall. ‘I was formally invited to join GC&CS about four days later,’ Hinsley recalled. ‘It was all done with minimum fuss and maximum dispatch.’ He was sent to Hut 4 and put in the care of Phoebe Senyard, who ensured he toured the Naval Section, learning what everyone was doing. Unlike many of the dons, Hinsley did not have a privileged background. He was a grammar school boy from Walsall, in the Black Country. His father was a wagoner who drove a horse and cart between the local ironworks and the railway station and Hinsley had won a scholarship to St John’s. A slight, bespectacled young man with wavy hair, he began working on traffic analysis of naval signals, putting together intelligence from the way in which the naval networks operated, where the individual stations or ships were located by direction-finding (DF), and the plain text messages that they passed. Following the original rehearsal, which the Naval Intelligence Division did not regard as having gone well, the Admiralty insisted that the GC&CS traffic analysts must work in the Operational Intelligence Centre (OIC) in London, which coordinated all naval intelligence. But the intelligence available from ‘log-reading’ or traffic analysis might be vital to finding cribs that would help Turing and Twinn, so Birch decided Hinsley must fill the gap together with a female intelligence analyst, 23-year-old Jocelyn Bostock, who was already working on the problem. The young and enthusiastic Hinsley was an immediate hit with Senyard.

  ‘I can remember quite well showing Harry some of the sorting and how delighted he seemed when he began to recognise the different types of signals,’ she said.

  He joined up with Miss Bostock, who was working on frequencies and callsigns. I then had to pass to Harry any strange, new or unknown signals, break of routine, or change of procedure which I noticed, for by that time one of my duties was the recognition of all German naval signals. No one was allowed to pass any cypher message out of the section except via me. This order of Mr Birch’s was very definite as we were on the lookout for any changes or new types of code, or for any slip up on the part of the Germans, and I was able, by long use, to identify German naval procedure and to tie new methods with old, passing them immediately to Harry who was on the lookout for these things too. Those were very enjoyable days indeed. We were all very happy and cheerful, working in close cooperation with each other. If I was in difficulty, I knew I could go to Harry. It was a pleasure because he was always interested in everything and took great pains to find out what it was and why.

  Relations between Hut 4 and the Admiralty were however far from cordial. While Hut 3 worked to the War Office and the Air Ministry via MI6, the Naval Section was officially NID12, part of the Naval Intelligence Division, and in direct contact with the OIC. But whenever any of the codebreakers attempted to talk to the intelligence officers in the Admiralty, their advice was virtually ignored.

  ‘Communication with the Admiralty was distinctly primitive,’ Hinsley recalled.

  I used a direct telephone line which I had to activate by turning a handle energetically before speaking. On this I spoke, a disembodied voice, to people who had never met me. They rarely took the initiative in turning the handle to speak to me and they showed little interest in what I said to them.

  In the second half of May 1940, Hinsley began to report evidence that a number of German warships were about to break out of the Baltic. The OIC ignored his warnings. This was to lead to one of the Royal Navy’s worst disasters of the Second World War, the sinking of the aircraft carrier HMS Glorious.

  ‘For about a fortnight beforehand, I pretty well rang the OIC once or twice a day and said: “Look you ought surely to pass a signal out on this. Can you possibly pass a signal out?”,’ Hinsley recalled. ‘They showed some interest. But were not sufficiently convinced to send a warning to the Home Fleet.’

  On 7 June 1940, the Glorious and her two escort destroyers were spotted by a German flotilla which included the pocket battleships the Gneisenau and the Scharnhorst, Hinsley said.

  On that day more than ever, I was saying to the duty officer: ‘For goodness sakes, can’t you just persuade them to send an alert or even “It
may be the case”.’ He said: ‘I can’t because first of all my traffic analysis group doesn’t agree with your interpretation. It doesn’t see the point, doesn’t see there’s any evidence and secondly my boss, the chief of the OIC, will not go to the operational chaps and say send this kind of signal out on your kind of information’.

  The next day HMS Glorious and her two escort destroyers, HMS Acasta and HMS Ardent, were sunk with the loss of 1,500 men.

  ‘The Glorious was capable of making a limited torpedo strike and could have flown defensive patrols if she had received even a qualified warning,’ Hinsley said.

  But the OIC, for all that it included these indications in its daily bulletin, resisted Bletchley’s suggestion that such a warning should be sent to ships at sea. It was not prepared to accept inferences drawn from an untried technique by civilians as yet unknown to its staff.

  Bletchley Park’s warnings to the Admiralty were the subject of a sustained cover-up. Even today, and despite the fact that Hinsley detailed the advice given to the Admiralty in the official history of British Intelligence in the Second World War, the Ministry of Defence still claims that ‘British intelligence sources failed to discover that the German force had sailed.’

  After the sinking of the Glorious, the OIC began to take more interest in what Bletchley Park said, and the GC&CS traffic analysts who had been detached to the Admiralty were sent back to the Naval Section. ‘Steps were taken not only to improve the working relations between the operational and the intelligence staffs in the Admiralty, but also to bring the OIC and the Naval Section at GC&CS closer together,’ Hinsley said. ‘It was as a direct result of the loss of the Glorious that regular visits between the two groups were instituted and that the OIC recognised that it had to rely on Naval Section’s greater familiarity with the German naval wireless system and to co-operate with the Naval Section in relating this knowledge to other operational information. It did so with a will.’ Nevertheless, it was clear that some naval intelligence officers clearly resented its very existence. ‘There was more than a suspicion of professional jealousy,’ noted Charles Morgan, who served in Naval Intelligence. ‘It was almost a point of honour to find the answer from our own records, even if a trifle incomplete, rather than have recourse to a BP telephone extension.’

  Reviewing what happened after the event, Birch had a more dispassionate view of the situation, identifying the problem as being more a case of the OIC not being persuaded of the reliability of the traffic analysis process. ‘Admiralty was by no means convinced of its reliability,’ Birch said,

  and there were good grounds for scepticism, for this inferential technique, which called for the severest mental discipline, seemed to have a fatal attraction for the romantically minded with more flair than judgement. Enthusiasts made extravagant claims; critics dismissed the whole subject as crystal-gazing.

  In an apparent effort to heal the rift, Hinsley was called down to the Admiralty and even sent to visit the Home Fleet at Scapa Flow. This caused some minor problems since, like many of the civilians at Bletchley, Hinsley wore only casual clothes, corduroy trousers and jumpers, and had no presentable clothes. ‘Scrounger’ Green said he was called in by Birch and told ‘to produce a suit of clothes – a hat he would not wear – for a shining light who was summoned to the Admiralty and who had nothing else to wear but a pair of corduroys and a Fair Isle pullover’.

  The new liaison had a limited effect, Hinsley recalled. ‘They began to realise that in spite of the fact that we were scruffy and young, and civilian, we had something to contribute. They took great pains thereafter always to be in close touch and always to argue and listen to us, taking the trouble to appoint a liaison officer to Bletchley to whom we could always show the facts.’ But despite the closer liaison, the difficulties continued. A few months later, Alec Dakin, an Egyptologist from Brasenose College who had been newly recruited into Hut 4, visited the Admiralty and was aghast to discover that the OIC appeared to be deliberately ‘obstructive and dismissive’ of Bletchley Park. Asked by Birch and Travis what was behind the problem, Hinsley blamed it on ‘a competitive spirit which instead of being of a healthy type is obviously personal. It couches itself in a show of independence and an air of obstruction. It appears to be based on personal opposition to Bletchley Park. I suspect that another reason for their inadequacy is incapacity pure and simple.’

  Turing and Twinn were working on the German Navy’s ‘Home Waters’ cypher, which they dubbed Dolphin. ‘There were at the time only two (daily-changing) Naval Enigma keys – for home and foreign waters respectively,’ Birch said. ‘As the latter never carried enough traffic for cryptanalysts to work on, the Home Waters key was, in effect, the only objective.’

  The Naval Enigma Section would eventually be housed in Hut 8 but at the beginning it was set up in a small corner of Hut 4, the Naval Section’s base to the left of the mansion, Senyard said

  Around about the beginning of April 1940, Mr Birch sent around a circular to the effect that there would be a new cryptographic section formed and that we should probably be very uncomfortable for about a fortnight or more in the endeavours to house them.

  We put our backs into it in order to welcome the newcomers, by tidying up our files and papers, binding and storing into cupboards all signals and books such as not in current use. Everyone who could be spared temporarily from their jobs was pressed into service and room was made for it, but it was a tight squeeze. We almost felt as if we ought to all breathe in together.

  The move to Hut 4 was followed by a key pinch. Two of the new wartime wheels for the Naval Enigma had been captured in February when the U33 was sunk while trying to lay mines in the Firth of Clyde, then on 26 April 1940, HMS Griffin captured a German Navy patrol boat, a converted trawler, off the Norwegian coast and towed her into the Orkneys. Unfortunately, after the German ship arrived at the naval base of Lyness, it was looted by British sailors looking for trophies. But some cypher tables and some sheets of signal pad on which encyphering had been carried out were recovered and these allowed Turing to confirm how the indicating system worked as well as providing him with letter for letter cribs for messages encyphered on the 25 and 26 April. The information on the relatively few pieces of paper that could be retrieved led to the breaking of six days’ worth of traffic, Birch said. ‘In fact, special intelligence from Naval Enigma might have been continuously forthcoming from then on – a whole year earlier than in the event – if the German trawler had not been looted before examination.’ Even with the captured material, the keys for 26 April were difficult to get out. Since the first Bombe, nicknamed Victory, had been introduced in March 1940, it was used to find the solution. It was a measure of the difficulties involved both in breaking Naval Enigma and with the first Bombe that it took a fortnight before it finally came up with the answer, scarcely a practical solution to solving Naval Enigma. Welchman’s clever modification to the Bombe, known as the diagonal board, would make it much more efficient, but the Bombes fitted with the diagonal board would not be introduced until August.

  The frustration felt by Hut 8 and in the Naval Section was exacerbated by the fact that the entire second half of 1940 ‘produced depressingly few results’, while the German submarines, the U-Boats, began to cut into the vital Atlantic convoys bringing food, machinery and oil to the UK. The Fall of France had given the U-Boats new bases on the French Atlantic coast that allowed them much easier access to the Allied convoys in what was effectively the beginning of the Battle of the Atlantic. The U-Boats used the Rudeltaktik, or ‘wolf pack’ tactic, devised by their commander Admiral Karl Dönitz while he was a prisoner of the British during the First World War. The German Navy had broken the British Merchant Navy Code and was reading many Royal Navy operational messages so the wolf packs knew the routes to be taken by the convoys and could lie in wait.

  They lined up north to south across the shipping routes. Once contact was made with a convoy, the closest U-Boats shadowed it, sending out homing si
gnals to draw in the other members of the pack. When all the U-Boats were assembled, they pounced en masse. This would peel the Royal Navy escorts away as they chased off individual U-Boats, allowing the rest of the pack to attack with impunity. In the immediate aftermath of the Fall of France, the U-Boats made hay against the convoys and although they did not start to use the wolf pack tactic routinely until September 1940, from June onwards, they were sinking increasing numbers of convoys, eating into Britain’s vital supplies, making it imperative that the U-Boat Enigma was cracked as soon as possible.

  It was the job of the Admiralty’s Operational Intelligence Centre (OIC) to re-route the convoys around the U-Boats and to do this it needed Bletchley Park to report where the U-Boats were. Turing and Twinn had moved into Hut 8 in June 1940 and had been bolstered by the addition of Tony Kendrick and Joan Clarke, who would at one point become briefly engaged to Turing, despite his homosexuality.