Thirty Secret Years

by Robin Denniston

  A strange mixture of high governmental secrecy with close ties of family and friendship dominates the professional life of government cryptographers. Two examples may suffice. Denniston was a devoted parent and amazed the Poles at the Enigma conference outside Warsaw in July 1939 by suggesting that Mayer, the Polish chief, and he should exchange daughters during long school holidays for language study: ‘I [Mayer] would send my daughter to England to his family, and his daughter would come to Poland to stay with my family. Of course the outbreak of The War thwarted … any … such … arrangement.’ The same suggestion - the swap of daughters for educational purposes - was made many years later to William Friedman, but this also could not be followed up.

  It seems surprising that AGD should be concerned to improve his daughter’s education at a time when he was the first government employee to have his hand on an Enigma machine, and an opportunity to assist in altering the course of war history. Those working in secret intelligence carry their private lives into their work as a necessary element in their total concentration; in these cases it was an excellent way of establishing rapport. It sounds dangerous, eccentric, or self-indulgent, at least highly ad hominem. It is certainly mirrored in Denniston’s pocket-diaries of the years between 1939 and 1945. In them names and events of undoubted importance appear pari passu not only with his modest social life and the minor ailments and achievements of his children, but with his personal finances, always meticulously noted. He also notes his monthly salary which increased varied from £80 in 1939 to £161 in 1942, declining after his relegation to £100 from 1943.

  Much of the inter-war period is the subject of a 21-page memorandum which AGD wrote for the authorities at the end of 1944, shortly before his retirement. Surprising in one so loyal and discreet, he kept a copy, which is now in the archives at Churchill College, Cambridge. Christopher Andrew, professor of modern history at Cambridge, published it in full in the first number of the Journal Intelligence & National Security, since when it has been widely cited. (see Chap. 6)

  In charge of the Secret Service during the whole inter-war period was Admiral Sinclair. GC&CS was, though small, on a par with SIS and there was a considerable interchange, at formal and informal levels, of information, knowledge and friendship between the various arms of the secret services. For people working in secret, relations with colleagues and with family are a very important part of life because such work cannot be shared. Remembering the dangers of penetration by enemy interception, it is remarkable that total security was maintained concerning the work of the GC&CS despite the fact that its chiefs were all on terms of friendship of quite a strong kind with several colleagues now known to have been working for the Soviets. To AGD, Donald Maclean and Kim Philby were well-known before their defection. The Foreign Office was still a cosy enclave and the office politics continued to play their part.

  The secret Cabinet directive for the new body was ‘To study the methods of cypher communications used by foreign powers’. This was achieved by a clause in the new Official Secrets Act instructing all cable companies operating in the UK to hand over for scrutiny copies of all cable traffic passing over their systems within ten days of despatch or receipt. This secured sufficient volume of throughput without which cryptanalysis is impossible. A great deal of traffic is needed to provide sufficient material for the appropriate distribution of the resulting decodes and (as important) to establish the other side’s priorities. A flurry of activity between two nations not normally on speaking terms may herald the approach of an important new diplomatic initiative as well as provide the raw material out of which their cyphers can be broken.

  GC&CS was a poor relation of SIS, which took most of the available funds for straightforward spying and counter-espionage. GC&CS had 25 officers and a similar number of clerical staff. Pay was exiguous, but as early as 1925 recruitment from the universities began. Classicists, linguists, papyrologists were most suitable, but after AGD’s Polish visit in July 1939, the importance of mathematicians became more important. AGD himself recruited Alan Turing and Gordon Welchman. By 1939 he had realized the vital importance of GC&CS’s product to the armed forces. The most important aspect of this was the Army in the form of Brigadier Tiltman, who worked at GC&CS for a while in the 1920s and continuously from the early 1930s to the end of the war. He died in 1983 after a long career in secret intelligence, a man of great achievement and probity, a firm friend of AGD. The Air Force involvement was signalled by the appointment of another man of great brilliance and achievement, Josh Cooper, who continued in cryptanalytical work through the Spycatcher era and became head of research at GCHQ. Commander (later Sir Edward) Travis was AGD’s deputy. He gradually took over the administration of the service traffic in 1941; in 1942 the job was divided in two and he took all the service work (Denniston took the Diplomatic and Commercial traffic) and remained head of Bletchley Park till 1944.

  Later historians separate off the solvers of cryptanalysis from the administrators of the work, the specific problem-solvers from those with an overall view. During the inter-war years there was no such division; indeed, very little administration was necessary. Several individuals, including AGD, received modest monetary rewards for specific cryptographic achievements. Key men included Fetterlein, Knox, Strachey, and Cooper. Hobart Hampden worked alone on the vital Japanese traffic: ‘With his knowledge of the habits of the Japanese he soon acquired an uncanny skill in never missing the important.’ That phrase alone establishes the width as well as the depth of skills needed to provide a total cryptological service - to identify and concentrate on the important and the relevant. An understanding of the psychology and thought-processes of the sender and the recipient in Whitehall was as vital as the discernment of patterns in letter frequency.

  Parallel with these developments the Americans were meeting similar problems rather differently. What might have been achieved had the two efforts been co-ordinated earlier, as they were to be from 1941 onwards, it is fruitless to speculate upon. The American effort was funded more generously and had greater access to experts, from engineers to linguists and mathematicians, of whom William Friedman was already a recognized high-profile world-class cryptoanalyst who combined remarkable intellectual concentration over long periods with a cosmopolitan’s view of world priorities. The British effort was much more modest by contrast; the British employed a few ex-Consuls who were Japanologists to provide valuable decodes of Japanese naval intercepts. Denniston’s memorandum notes:

  Yardley’s Black Chamber tells of the American success at the Washington Conference in 1919 - a revelation which of course deeply embarrassed the US government. No one will ever tell how much more accurate and reliable information (on Japanese as well as European influence) was made available to our Foreign Office and service departments during those critical years.

  The other main diplomatic achievements of GC&CS in the early inter-war period include much work on the German machine systems and Russian ciphers. Italian territorial ambitions in Abyssinia and the Eastern Mediterranean brought Italian experts to GC&CS; and a valuable basis for future German work was established when that country had re-emerged from the humiliations of Trianon to become the Number One enemy. Austria was too weak after Versailles to produce enough traffic to be worked on, Hungary was successfully attacked by Dilwyn Knox, but without much of use emerging. Summing up, Denniston writes:

  We started in 1919 at the period of bows-and-arrows methods - i.e. alphabetical books: we followed the various developments of security measures adopted in every country: we reached 1939 with a full knowledge of all the methods evolved, and with the ability to read all the diplomatic communications of all powers except those who had been forced, like Germany and Russia, to adopt OTP.*

  What was translated and submitted for circulation was up to the discretion of AGD and his team. In addition to the daily issue there was a regular summary of telegrams decoded but not circulated for the benefit of the SIS and some ministries, but this was not gre
atly drawn upon. GC&CS successfully fulfilled its allotted function, with exiguous numbers and with an absence of publicity which greatly enhanced the value of its work.

  Included in his memorandum is an interesting section on ‘clandestine activities’, in fact a nest of spies at work in a London suburb, in direct touch with a Comintern network operating from near Moscow. Italian codes were broken with considerable success thanks ‘to their habit of enciphering long political leaders from the daily press’. Via the Italians, GC&CS was introduced to the commercial Enigma and this led to the exploration of machine encipherment. This was tackled by Knox, aided by younger men such as Bodsworth and Peter Twinn. Yet another experience was acquired in traffic analysis, a process which enabled the operator to continue plotting the movement of troops or ships even when the ciphers themselves were not legible, to establish a quantitative volume indicative of activity or otherwise. This indicates the wide scope of GC&CS’ activities. Ultimately, most secret-service activities were monitored and controlled from Bletchley Park as only the seniors there had access to the total picture of clandestine activity against which each individual detail becomes significant or otherwise. In the inter-war period Malta was the focal point for all traffic between Europe and the East, and access to this traffic enabled GC&CS to watch the growth of the Axis combination at first hand - yet, paradoxically, with little ability to influence the course of events.

  * * *

  III

  AGD’s career at this time involved attention to the ‘how’ of interception, the establishment of the Y committee, and the university recruiting drives of 1937 onwards. The memorandum ends:

  From August 15th 1939 onwards when the diplomatic and commercial sections moved to BP the university recruits began to join, so that by September 1st GC&CS was in action at its war station, already in process of growth towards that vast and successful body whose full story will perhaps never be told.

  What is certain is that a very minor operation, underfunded to such an extent that a major bureaucratic operation had to be mounted in the mid-1930s to recruit half a dozen war veterans who had an entitlement to only half of their proper disability pension (all this is attested in the National Archives files), became in 1938-39 an important but unrecognized unit which would serve the Allies through the war with invaluable material, acknowledged by Eisenhower and used by all top military commanders, including Montgomery. It was understood by none better than Churchill himself, who knew the value of the naval intercepts from his experiences as First Lord of the Admiralty in World War I. He made visits to Bletchley Park, charted the daily results of the intercepts on his war maps, and eventually brought success to the British war effort and created, inexpertly, the ensuing peace.

  Two events of historic importance appear in the pocket-diary for 1939. They both concern the developing interest of the French and the British in the intermittently successful attempts by a group of Polish government cryptoanalysts working near Warsaw to solve the secrets of the Enigma machine. Much has been written about this, of course, including a whole appendix in Vol. 1 of Hinsley et al, British Intelligence in the Second World War (HMSO 1979) pp. 487-495. At first the French had taken a more positive view of the Polish achievement than the British, and a conference of the three cryptographic establishments (Denniston was assigned the code-name Crypto) in January 1939 was generally regarded as only a modest success. The obsessive secrecy of GC&CS following the political disclosures of the early 1920s had not abated, and it was a not unnatural reluctance to share Dilly Knox’s theoretical attacks on the Enigma problems coupled with doubts about the genuineness of the Polish achievements and (I would guess) an ancestral suspicion of the French that caused both Poles and French to register something like contempt for the British contribution to that conference, whose dates, whereabouts, and main participants duly appear along with some exiguous expenses in the Denniston diary for the week. In fact, all three nations kept their secrets to themselves.

  However little that conference achieved, events later in the year brought the Poles to the heroic point of handing over their complete work to both French and British. This took place in July 1939, and there are various accounts of what happened. The Britons were AGD as head of GC&CS, Knox as the Enigma expert, and a naval communications officer, Commander Sandwith. AGD’s account of the visit, with subsequent additions, is now to be found in the National Archives and in chapter 7.

  Much has now been written about the trip to Poland, to discuss with the French and the Poles the solving of the Enigma, the making of the Bombe, and the receipt by the British and French not only of the Poles’ vital findings but of specimens of the machine itself. The party went from Dover to Ostend or Calais and by train across Europe. The passport, issued the previous year (possibly because they had expected to make this contact earlier), shows arrival at Calais on 24 July and Lille on the same day, passing back through Germany from Poland on the 28 July. On 14 August Bertrand brought the Enigma machine to Menzies, who handed it over to Denniston.

  In December of that year, the phony war now in progress, there was a conference chaired by AGD in London. The French were represented by Bertrand and the Poles by Langer. ‘During this conference’, writes T. Lisciki, a graduate of the Warsaw Polytechnic who commanded a Polish signals unit and has recently set the record straight regarding the Polish contribution to cryptanalytical matters in this period, ‘it was decided to establish this direct link and the question of which cypher to use arose. As the British Typex machine was super secret and the French did not have any, your father had a brilliant idea to use the German machine. This machine was used until the collapse of France and I doubt if the Germans ever had any suspicion of such “affront”’.

  The admiral, that is to say Admiral Hugh Sinclair, Menzies’ predecessor as head of the British SIS, bought Bletchley Park. After a successful 1938 dummy run those in control were more than ready, with Enigma in their hands and war declared on 3 September 1939, to crack the German codes which had eluded them for so long. And the university recruits came, ‘the Professor types’, mostly from Cambridge. They were paid £600 a year. In August Professors Waterhouse, Gore Brown, Vincent, and Boase joined and others, including the WWI veteran Nigel de Grey, rejoined. In September Gordon Welchman and Alan Turing joined and Frank Birch, Frank Adcock, and William Hope rejoined. Lord Dacre, in charge of the radio security service, told me how he and AGD had got across each other during the war and had made it up at dinner at Christ Church afterwards; and he told a mutual acquaintance that in his view the state of friendly informality verging on apparent anarchy, which was such a feature of the early years at Bletchley, was AGD’s particular contribution, that diminished with his departure to Berkeley Street with the diplomatic and commercial sections from BP. Travis was a more dictatorial supremo, probably what BP needed, but some Huts 3 and 6 veterans missed the former collegial atmosphere.

  Many others came, mostly women. Some early reminiscences of their reception at BP are circulated. One published one, by Aileen Clayton, The Enemy Is Listening (Hutchinson, 1980), recalls:

  ‘Station X … then under the control of Alastair Denniston, a quiet middle-aged man who seemed more like a professor than a naval officer. It was to him that I had to report, and I was immediately impressed by his kindness and by the courtesy with which he greeted me.’

  The author goes on to give a vivid picture of Y intelligence at work in her first interview with Josh Cooper.

  George Steiner wrote in the Sunday Times (23 October 1983) that increasingly ‘it looks as if Bletchley Park is the single greatest achievement of Britain during 1939-45, perhaps during this century as a whole’. I am also indebted to Neil Webster for the following passage from his unpublished paper on intelligence:

  The cryptographic organisation at Bletchley was highly efficient. Indeed it was the most efficient working organisation I have met, perhaps because there were no trade unions and little or no financial control and because it was run mainly neit
her by business men nor career civil servants but by mathematicians and chess players, who brought detached and decisive minds to the solution of cryptographic, organisational and human problems. Contributory factors were the devotion, high morale, and esprit de corps of the picked band of workers. Gifted people were willing to work on boring and repetitive tasks if it was important that these should be done by people capable of spotting the occasional small nugget, which might turn up in the sieve.

  Some photographs at the time bear out the informality and also the tension - AGD watching a game of rounders with his hand characteristically over his mouth, his double-breasted suit. Because of his small stature he and my mother always took particular care of his appearance. In addition to his interest in technical matters, he devoted much time and thought to the building programme round BP, necessitated by the exponentially growing numbers.

  His diary records on 18 April 1940: ‘Birch to have copies of all naval traffic so that Hope can consult.’ On the same day: ‘esquimaux traffic: Stern not wanted: Falconer not wanted majority of RR material is by post; however TP’s necessary’. This was the setting up of Hut 3 under Frank Birch, handling all the naval traffic. Meanwhile Enigma work, assigned to Travis, was concentrated in Hut 6 with Welchman, Turing, Babbage, Milner Barry and de Grey. And this is where office politics become inextricably tied up with winning, or trying not to lose, a world war.

  IV

  When we holidayed at Loch Striven in south Argyllshire in 1937 – one of several summer holidays cut short for my father by the exigencies of GC&CS as war approached - we included Ian Schiller in the party. Uncle Ian was a German who had come to adopt our family and made Y his honorary god-daughter. Despite infantile paralysis which left him partially lame he was adept at messing about in boats and since all our chattels had to be rowed the mile along the coast north from where the shepherd lived to our holiday cottage his skills were vital. My father loved boats and the sea, having been brought up near them in Dunoon, and I got the urge - He taught me the rudimentary skills of managing a dinghy. When war came my father’s gift for friendship extended to the Americans who came to liaise with BP and Berkeley Street, chief among them Telford Taylor and his successor Major Bancroft Littlefield. Both men used to weekend at Ashtead in 1944 during the V1 and 2 bombing. Bancroft picked and ate our soft fruit while watching these lethal machines whizz overhead to do serious damage to more western parts of Surrey, while Telford played the piano (including duets with me), and tennis at which he was far better than us. In the evenings he and my father used to drink the whisky he always brought with him, and then no doubt my father opened up and told him more of the truth of his 30 years in signals intelligence.