The Bletchley Park Codebreakers (29 page)

In the last twelve months of the war all the breaking of naval keys was done by four cribsters: Patrick Mahon, Joan Clarke (later Murray), Richard Pendered and myself; we worked on a shift rota around the clock. All the main keys were broken regularly during this period. While some of the cribbing became routine we enjoyed our tasks after the bombes had identified a key’s wheel order and pluggings. We still had to work out the
Ringstellung
and the
Grundstellung
.

With growing participation by OP-20-G, our opposite numbers in the US Navy, we coped successfully with the introduction of many new keys: they included Plaice for ships in the Baltic, Grampus for ships in the Black Sea, Narwhal for U-boats in the Arctic, Sunfish for blockade runners, etc. In one way this proliferation was helpful because the same announcement was often sent out in several keys. Having broken one of these we had a reliable crib for the others.

Apart from these re-encodements, weather messages remained our most frequent source of cribs. On D-Day in 1944 we broke Dolphin at daybreak thanks to an undisciplined German operator who began his report day after day with WETTERVORHERSAGEBISKAYA. At other times the disciplines must have been reimposed so that reliable weather cribs disappeared for a time. We then had to turn to other sources, including the ‘gardening’ procedure described in Chapter 11. The areas where the RAF dropped mines were carefully chosen so that the German naval grid references contained no numbers for which there were alternative spellings (NUL or NULL for ‘0’, FUNF or FUENF for ‘5’).

One day, while on duty in the morning, I was told by Intelligence that it was very important for us to read the next day’s messages as early as possible. Would gardening improve the chances of an early
break.’ I thought about it and gave the answer ‘yes’. That night was stormy and I lay in bed worrying whether my judgement had been correct or whether I had needlessly endangered the lives of the air crews. I was extremely relieved when I heard next day that there had been no flying because of the bad weather.

On rare occasions a successful crib was based on a guess that a message might contain an important topical announcement. The outstanding example occurred when the battleship
Tirpitz
was sunk by Bomber Command in north Norway in November 1944. We guessed that a certain signal might say ‘
Tirpitz
capsized’. It did and we were able to break one of the day’s keys.

Although in the last phase the number of cryptanalysts in Hut 8 had dwindled to four, we were always dependent on a big supporting cast. First of all there were the wireless operators at the various interception stations who did a fantastic job. Picking up and accurately recording a short signal, lasting only a few seconds, must have required great skill and concentration. Then there were the ladies in Hut 8 itself, eventually about 130 of them, doing the clerical work and the typing; in the days of Banburismus they also had to punch the holes in the Banbury sheets. Finally, there was the work of running the bombes, which were located in several big houses around Bletchley. This was done by Wrens working under very spartan conditions. Looking back, I think that we did not give enough credit to all these people without whom we could not have functioned.

The Germans responsible for Enigma security had tolerated all kinds of flaws but they became extremely active at the very end of the war. A complicated new system for determining the starting positions of signals was introduced on Dolphin in May 1945. We made plans to meet this challenge but it would undoubtedly have caused delays and it was perhaps fortunate that the war ended before the new procedures became effective.

At the peak there were about 9,000 people working at Bletchley Park. With, I believe, only one exception, the confidentiality of the work was never breached either then or for about thirty years after the war. We had all signed the Official Secrets Act and never doubted that we could not talk about what we did during the war. Is it conceivable that such secrecy could be maintained amid the present culture of constant leaks?

Robert Harris’s
Enigma
is a very good novel but it paints far too drab a picture of life at Bletchley. No doubt the meals served in the
canteen featured dried egg, Spam, and other wartime delicacies: they may have been stodgy but they were perfectly palatable. Most of us were billeted on householders in a wide area around Bletchley and I do not remember many complaints. I have fond memories of the elderly couple with whom I lodged in Newport Pagnell; the wife was a wonderful lady who had been maid to a duchess. No duchess could have received better care than I did.

Many of us had come straight from university and it was in many ways like having an extra four years of university life. We did work hard, taking unsocial hours for granted, but there was also a good social life. An active Dramatic Society put on plays and satirical revues of a high standard, there was a lot of music, and we played chess and bridge. Most men went about in old sports jackets and shabby corduroy trousers. Once when a visiting Admiral was taken around the site by the Director he is reported to have asked: ‘What are all these velvet arsed bastards doing here?’

I had a happy time at Bletchley, not only because of the work but because Hut 8 was always a friendly place. There really was a spirit of camaraderie among the cryptanalysts and a sense of a common purpose. I can recall no personality clashes or big outbursts of temper. I attribute this to the fascination of the work, the satisfaction of getting results, exemplary leadership and, above all, the personalities of the individuals. I would like to conclude this account with recollections of some of them.

From addressing each other by our surnames we soon switched to first names. There were two exceptions: one of these was Alan Turing, without whom Hut 8 might not have existed. It was he who, building on pre-war discoveries by the Poles, originated the methods by which the naval Enigma was broken. We all recognized his genius; perhaps for this reason he was known as ‘Prof’. We regarded him as eccentric but I cannot remember any specific eccentricities. By the time of my arrival in Hut 8 the basic principles of the work were well established so that there was less scope for his genius. He therefore spent more and more time on assignments outside Hut 8 and left altogether in late 1942. While he was with us he was always approachable and ready to help with technical problems. It may have been my fault that I did not find it easy to communicate with him.

I could always communicate with Hugh Alexander, the former British Chess Champion, who had come in as acting head of
Hut 8 because Turing’s great abilities did not include a talent for administration. He had a wonderfully quick mind combined with tremendous energy and enthusiasm. I still think of him as the model manager, always leading from the front. He treated us as colleagues but he could be very firm when it was necessary.

Turing and Alexander have since died but Shaun Wylie remains active in his eighties. Alexander rightly described him as ‘easily the best all-rounder in the section, astonishingly quick and resourceful’. He was an outstanding leader who always set higher standards for himself than for anyone else. I have continued to admire his intellect and his integrity, more recently his superior skill in solving the weekly ‘Listener’ crossword in
The Times
to which we are both addicted.

I have already mentioned the part played by Michael Ashcroft when breaking Shark was resumed in 1942–3 with the aid of short signals. He would now be described as a caring man; he had helped to rescue and educate at least one Jewish boy from Germany. At Oxford he was active in centre-left politics. We shared various interests and met frequently after the war. I felt a great loss when he died.

Patrick Mahon was the obvious choice as head of Hut 8 when Alexander left in September 1944. He was exceptional among the cryptanalysts in being a linguist, not a mathematician. If this was a handicap, he amply compensated for it by great energy and a very methodical mind. It was he who discovered the shift system worked by some operators sending weather messages. He was given to strong views, with some youthful intolerance, but he was always popular because he had great personal charm. After the war he had a successful career in the John Lewis Partnership. We kept in touch but sadly his health failed and he died prematurely in 1972.

Richard Pendered was the brightest of the younger cryptanalysts in Hut 8. He had a memorable success when he broke Shark for 27 May 1943, using a long crib, by a purely manual method, which he himself had refined. It was a superb technical achievement: the method was used again to find the wiring of a new fourth wheel and reflector which the Germans introduced on 1 July 1943.

Joan Clarke, later Murray, was an excellent Banburist who later became an excellent cribster. It was a tribute to her ability that her equality with the men was never in question, even in those unenlightened days. She was kind and always good-tempered, but rather reserved. It was known that she had been close to Alan Turing
but that nothing had come of the relationship. She never spoke about it, but when I met her again in 1993, at the launch of
Codebreakers
(the book of reminiscences of Bletchley, edited by Harry Hinsley and Alan Stripp), she said that she had not seen Hugh Whitemore’s play
Breaking the Code
because it would have been too painful.

The other exception to the use of first names was Kendrick. I never knew what the initials F. A. stood for and it was something of a shock to find him listed in the index of
Codebreakers
as ‘Kendrick, Tony’. He was a member of Hut 8 from early 1940 until July 1942; it was said that any new suggestion had already been proposed by Kendrick at some earlier date. Severely crippled by polio, he was a very private man but he was courteous and kind, and he had a fine sense of humour. He was an even shabbier dresser than the rest of us; strips of tattered lining were seen to protrude from his threadbare suit. It was believed that this was his protest, as a career civil servant, against the abandonment of pinstriped trousers after GC&CS moved to Bletchley. Some time after the war I met him at a concert; he was working at the Ministry of Defence. ‘England is safe then,’ I said. ‘Yes,’ he replied. ‘I don’t interfere much.’

When the war ended I could not tear myself away from decoding and spent a further year working on other ciphers. When I finally left to join the family business created by my father, I made sure that I could come back if a six months’ trial did not work out. The option was unnecessary: I am still involved in the business.

Being at Bletchley during the war was a prime example of my good fortune in life. While most of my contemporaries were risking death or injury and enduring hardship I was living comfortably and doing exciting work. At the same time I could maintain my self-respect because the work was important. I know that I was very lucky.

13

BLETCHLEY PARK AND THE BIRTH OF THE VERY SPECIAL RELATIONSHIP

STEPHEN BUDIANSK

In July 1942, a Bletchley Park liaison officer arrived in Washington and immediately began venting his considerable annoyance with Britain’s ally in a series of increasingly exasperated despatches. There was the unbearable ‘regimental gossip’ of American wives at dinner parties. There was an absurd and tedious picnic he was dragged to in Washington’s Rock Creek Park. There was ‘our fat friend Kully’ – Solomon Kullback, one of the US Army’s leading cryptanalysts – who over-organized his section, placed obstacles in the way of British requests for information, and had the habit, ‘which I am sure you must have observed in some degree, of shouting people down to contradict
something they have no intention of saying’. By September the British official was writing home:

Even with America’s entry into the war in December 1941, it was far from inevitable that the two allies would forge any real or meaningful co-operation in cryptanalysis and signals intelligence. There was certainly a lot working against it. For a start, the British were convinced that the Americans were simply incapable of safeguarding important secrets. Many Bletchley Park officials were loath to share their monopoly over the control and distribution of intelligence derived from the breaking of the German Enigma ciphers. On the American side, the US Navy might have finally forgotten the war of 1812 and the press-ganging of American sailors, but nonetheless harboured a widespread distrust and resentment of the British, and was particularly touchy about anything that might imply that America was a junior partner taking orders from the British. Moreover, many American officials (with good reason, given past practice of GC&CS) were suspicious that the British would exploit cryptologic co-operation to gain access to American codes in order to read confidential US diplomatic communications.

Finally, there was the simple transatlantic culture gap – far greater then than now – that led to repeated misunderstandings, friction and enmity on a personal level, of the sort so richly exemplified by Bletchley Park’s liaison officer, who, when he was not penning poisonous despatches, was irking the US Army code-breakers at Arlington Hall with what many took to be his pompous and overbearing manner. (One Arlington Hall veteran recalled that this officer always appeared with Sam Browne belt, polished riding boots and swagger stick.)

In the end it was a combination of practical necessity and personal diplomacy that overcame these many obstacles and helped to launch the ‘special relationship’ between the British and American intelligence
agencies, which endures to this day. The pooling of talent, effort and technology by British and American cryptanalysts proved to be of the first importance in the breaking of the German U-boat ciphers and diplomatic codes and the Japanese naval and military codes in particular. But it was a long haul, and were it not for a few far-sighted and level-headed men on both sides of the Atlantic who prevailed at critical junctures, the relationship would never have progressed beyond the ‘make polite noises’ stage at which it was largely stuck for the first couple of years of the war. The real heroes of this tale are John Tiltman on the British side and William Friedman and Telford Taylor on the American side, and perhaps even more importantly, the many lower-ranking technical experts on both sides who began working with one another at a very practical level and forged personal ties that did much to overcome the mutual suspicion and incomprehension that had threatened to poison the relationship before it had even started.

The half century that has passed since the Second World War has brought about such a thorough revolution in travel and communication that it is difficult for us today to realize just how insular Britain and America were in 1940, and the extent to which cultural differences posed a significant obstacle to establishing
co-operation.
The knowledge that even many well-educated Britons had of Americans was limited to the actors they had seen on movie screens. Within official circles in Britain, Americans were frequently portrayed in the broadest caricatures: naive in the ways of the world and incapable of keeping a secret.

On 15 November 1940, in a memorandum discussing the first feelers for co-operation with the US signals intelligence agencies, Alastair Denniston – then the operational head of GC&CS – warned that ‘we are entitled to recall that the Americans sent over at the end of the last war the now notorious Colonel [Herbert O.] Yardley for purposes of co-operation. He went so far as to publish the story of his co-operation in book form.’ Even after the US Navy had begun convoying merchant ships to Iceland and had become engaged in what was for all intents and purposes an undeclared war in the Atlantic in 1941, British officialdom was set against sharing the fruits of signals intelligence with the United States. Brigadier Stewart Menzies, who as ‘C’ oversaw both GC&CS and the Secret Intelligence Service (MI6), wrote to Prime Minister Winston Churchill on 24 June 1941 that he had considered the matter ‘from all angles’; true, US Navy units
were
being chased by U-boats – just four days earlier U-203 had pursued the American battleship
Texas
for 140 miles between Greenland and Iceland and had repeatedly tried to manoeuvre into position to fire a torpedo – and, true, GC&CS was reading Enigma traffic that contained orders from the German Admiral Commanding U-boats to his captains at sea, which might give vital warning to the Americans. But, ‘C’ concluded:

The attached article that had caught ‘C”s attention reported that Colonel William ‘Wild Bill’ Donovan ‘has a new hush-hush mission – to supervise the United States Secret Service and ally it with the British Secret Service … The American “Mr. X”, as he is known privately, will report direct to the President.’ It wasn’t exactly ‘hush-hush’ by the time the story had appeared in the
Daily Express
, and ‘C’ had a point.

Although the US Army’s Signal Intelligence Service, under the direction of the renowned cryptanalyst William Friedman, was eager to initiate a sharing of technical data with GC&CS in the autumn of 1940, the US Navy was almost completely opposed to all of Friedman’s recommendations. Commander Laurance Safford, Friedman’s counterpart as head of the Navy’s OP-20-G, rejected any exchange of cryptanalytic data and techniques or training materials. Neither Friedman nor Safford, however, was interested in letting the British in on American
cryptographic
secrets – that is, the business of codemaking as opposed to code-breaking. In particular, Friedman ruled out giving the British ‘information of any kind’ relating to America’s SIGABA cipher machine, even the very fact of its existence. There was considerable suspicion – again not unjustified, as it
happened – that the British were attempting to read American codes. Even while proposing an intimate exchange of highly secret material on codebreaking Friedman was not going to give the British anything that might help them break American systems. (After America’s entry into the war following the Pearl Harbor attack, Churchill felt he needed to warn his new ally about the insecurity of her diplomatic codes; in a letter to President Roosevelt on 25 February 1942, the Prime Minister wrote:

On the one hand it was a considerate gesture, but it also confirmed the US experts’ worst fears about the British. The following year, Colonel Alfred McCormack, a US military intelligence officer, visited the British radio intercept station at Beaumanor, and while the station’s commander was showing him around he casually remarked that he used to read the US State Department ciphers and that it had been ‘lots of fun’.)

Of course, a far greater obstacle was that the British and American signals intelligence establishments had very different aims; each wanted to extract something quite different from the other, and some of those goals were mutually exclusive. When the first technical exchanges between the two sides began in February 1941 with the arrival of a combined US Army–Navy liaison mission, Britain’s chief concern was gaining assistance in dealing with Japanese codes. Though the British had begun in the 1930s to intercept Japanese naval traffic from a station in Hong Kong (the station moved to Singapore in September 1939 and Kilindini, Kenya, in 1942 as the Japanese Army advanced through South Asia) – and though they had made some progress against the main Japanese naval operations code (the system that would later be known as JN-25) – Britain had a critical shortage of Japanese linguists. The codebook change that the Japanese Navy introduced in December 1940, replacing the JN-25A code with the much more complex JN-25B code, was a considerable setback, and the British frankly conceded that they very much needed American
assistance. In fact, the British offered to turn over their entire Far East codebreaking operation to the Americans if the United States would supply the needed translators.

In its efforts against the high-level Japanese diplomatic cipher, Purple, which was generated by a machine roughly equivalent in complexity to the German Enigma, GC&CS was also stymied and very eager for US help. GC&CS had broken the relatively simple Japanese Red machine, which had been used for diplomatic traffic for much of the 1930s, but had actually given up trying to solve the more complex Purple machine, which came into use on 20 February 1938. From the start, the British consistently sought to keep Anglo-American cryptanalytical co-operation principally limited to the one area where they knew they needed help – Japan.