Alan Turing: The Enigma The Centenary Edition

by Andrew Hodges

  This Looking Glass ploy of taking instructions literally was one that created a similar fuss when his identity card was found unsigned, on the grounds that he had been told not to write anything on it. It came to light when he was stopped and interrogated by two policemen as he took a country walk. His awkward appearance and habit of examining wild flowers in the hedgerows had excited the imagination of a spy-conscious citizen.47

  But besides sharing in such victories over blimps and bureaucrats, there was the experience of free association with people who were among the best in British mathematics, in a sort of secret university, one in which tradition and form, together with rank, age, degrees and all such superficialities were ignored. All that mattered was the ability to think. And they had a mathematical Flash Gordon, a logical Superboy, to encourage them – someone who refused to admit defeat, or any limitations on their capacities to succeed. To Peter Hilton, Alan was

  … a very easily approachable man – though you always felt there was lots more you did not know anything about. There was always a sense of this immense power and of his ability to tackle every problem, and always from first principles. I mean, he not only … did a lot of theoretical work, but he actually designed machines to help in the solution of problems – and with all the electrical circuitry that would be involved, as well.

  He did, for instance, design a special machine to help Harry Golombek with the analysis of the particular Enigma system employed by the German motor torpedo boats. There was another designed for use on the main naval Enigma problem; there was far more to it than the Bombe. The technology was not always new; thus the Banburismus process involved the use of paper sheets on which cipher-text messages were represented as punched holes. These had to be moved against each other and coincident holes laboriously counted before the sophisticated statistical methods could be used. There was a hint of irony in the way that Alan chose to call the process ROMSing – a reference to that progressive slogan, the Resources of Modern Science. But it also represented the essential truth about the Bletchley work, and Alan Turing was at the heart of it, never too proud to get his hands dirty with the ‘dull and elementary’:

  In all these ways he always tackled the whole problem and never ran away from a calculation. If it was a question of wanting to know how something would in fact behave in practice, he would do all the numerical calculations as well.

  We were all very much inspired by him, his interest in the work but the simultaneous interest in almost everything else… And he was a delightful person to work with. He had great patience with those who were not as gifted as himself. I remember he always gave me enormous encouragement when I did anything that was at all noteworthy. And we were very very fond of him.

  Alan’s ‘great patience’ was not usually his most conspicuous characteristic, nor his approachability. But Peter Hilton was in fact the fastest thinker of the new Fish group, and drew out the most rewarding aspects of the ‘creative anarchy’ that was Alan Turing. It was pure joy to achieve something new, and show him, and have him grunt, gasp, brush back his hair, and exclaim, stabbing with his strange fingers, ‘I see! I see!’ But then it was down to earth again with the rules and regulations:

  But there again, he began to be beset by the bureaucrats who wanted him to be in at a certain time and work till five o’clock and leave. His procedure – and that of many others of us, let me say, not only he, who were really fascinated by the work – would be maybe to come in at midday and work until midnight the next day. And then, the problem being essentially solved, go off and rest up and not come back for 24 hours perhaps … they were getting much more work out of Alan Turing that way. But, as I say, the bureaucrats came along and wanted forms to be filled in and wanted us to clock in, and so on.

  Once he ordered a barrel of beer for the office, but it was ‘disallowed’. Such questions were trivial, but behind them lay more serious confrontations with the old mentality, which little by little and nearly too late had been obliged to give way to intelligence. Alan’s role in this process, however annoying to authority, was not entirely unrewarded. One day in 1942 he, Gordon Welch-man and Hugh Alexander were suddenly summoned to the Foreign Office and awarded £200 each. Alan told Joan that they could not be given decorations, so had been given money instead. He probably found it more useful.

  In September 1942 the British position was a little less hopeless, but only inasmuch as there had been no serious loss since that of Tobruk. Rommel’s eastward advance on Egypt had been checked by Auchinleck in July and by Montgomery in August, the latter being particularly helped by deciphered signals. The desert war was more like a naval war than a conventional front, and was particularly dependent upon information. It desperately required the effective integration of the three services, who had been obliged to swallow a bitter pill indeed by allowing Bletchley information and interpretation to be transmitted over the heads of the London chiefs directly to an intelligence centre in Cairo. But a more centralised system was forced upon them by the cornucopia of north Buckinghamshire. By May 1942 they were breaking every Enigma key system of the African theatre. In August this was joined by a new Hut 8 success, the breaking of the system used by Mediterranean surface ships. Rommel was now losing one quarter of his supplies through British attacks which were almost totally dependent on detailed Enigma information – sometimes enabling them to pick out the more important cargoes for destruction. News of this triumph was passed back to the analysts in Hut 8 to encourage them in their work.

  But the Mediterranean was, ultimately, an Anglo-German diversion. In the world struggle there had been a major setback for Japan at the battle of Midway, where the US Navy proved it could put its own Intelligence to devastating effect. But in Europe there was no such hint of a reverse. The Axis attack on Russia had reached Stalingrad, and the Dieppe raid had ended lingering fantasies of an easy victory in the west. More frightening than either of these developments, however, for Churchill and for everyone else, was the state of the fragile Atlantic bridge. Without it Britain was nothing.

  Although the first American troops had arrived in Britain early in 1942, it was the stream of war materials, tanks and aircraft in particular, that alone could make the reconquest of western Europe conceivable. That stream had to face the Atlantic U-boat fleet, which by October amounted to 196 in number. Since 1940 the numbers had trebled, and the sinkings had trebled too. Until mid-1942, American reluctance to provide coastal convoys had diverted U-boats to easy pickings off the eastern seaboard, but in August counter-measures had remedied this gap in defence. Accordingly, the U-boats had turned back to the Atlantic convoys, exploiting the area in mid-ocean where air cover was not supplied, and were now accounting for over half the merchant fleet required to supply Britain within a year. The revived American shipyards were turning out new vessels at top speed, only to have each sunk after three or so ocean voyages. But now the United States had its own pressing demands in the Pacific. The total Allied stock of shipping was actually declining, while the number of U-boats was increasing: there would be 212 at the end of 1942, with another 181 on trial.

  Fast approaching was the crisis of the western war. 1943 might see either Britain stocked up as the forward base of an impregnable American industry, or might see it slowly sink. Though more diffuse a crisis than that of the air war of September 1940, it likewise stood to see a make-or-break resolution. Ten years earlier, Alan had conceived a model of action: ‘We have a will which is able to determine the actions of the atoms probably in a small portion of the brain… The rest of the body acts so as to amplify this.’ Now he was one of the clustered nerve-cells, and around him a colossal system which had translated his ideas into concrete form: a British brain, an electric brain of relays clicking through the contradictions, perhaps the most complex logical system ever devised. Meanwhile the two years of the reprieve had rendered the rest of the body more prepared and coordinated to use its intelligence. In the Middle East it was amplifying the dim Morse signals into the
sinking of Rommel’s army. But the Atlantic was different; here Eisenhower and Marshall might be cut off on a far greater scale than Rommel unless the brain could awaken into life again.

  But those two years had seen another momentous change. The ten-fold increase in rotor positions had forced Poland to turn to the technically superior West. And now the twenty-six-fold increase had brought the United States into the electromagnetic relay race. Its Admiral King, who had been more obstinate than the British Admiralty, resisted the setting up of a tracking room until mid-1942. But the US Navy’s cryptanalysts had been quick off the mark to see what was needed. Their department had been using modern machinery since 1935, and when the black-out came in February 1942 they were not content to stand and wait until the British caught up: they could do it themselves. This did not accord at all with the British view, which held that the Americans should concentrate on the Japanese ciphers, and not duplicate the work done at Bletchley. But the US Navy was particularly insistent. Already in June, its relations with GC and CS were ‘strained’ by complaints at the delay they experienced in obtaining a promised Bombe, and then49

  in September the Navy Department announced that it had developed a more advanced machine of its own, would have built 360 copies of it by the end of the year, and intended to attack the U-boat Enigma settings forthwith.

  These were figures to send Bletchley minds reeling. The whole of GC and CS Enigma work depended upon managing with just thirty Bombes in summer 1942, although another twenty were on the way. The Americans were proposing a take-over bid for the Atlantic work, by the brute-force expedient of building twenty-six times as many Bombes as the British had available, and using them in parallel.

  But in October a second deputation from GC and CS to Washington negotiated another compromise. GC and CS ‘acceded to US desires to attack the German naval and submarine problems’ and agreed to supply the Navy Department with the intercepts and with technical assistance. In return the Navy Department … undertook to construct only 100 Bombes, accepted that GC and CS should be responsible for co-ordinating the work done by the American machines with that done by the British, and agreed to the complete and immediate exchange of the cryptanalytic results.

  There was only one person who knew everything about the methods and the machines, and who was free from day to day responsibility. The responsibility for the detailed coordination thus promised now fell to the Prof. Sorting out the tensions between American bluster and British arrogance was not at all the kind of work he enjoyed, but the Anglo-American liaison had to be made concrete. There was a war on. Accredited as an official with the British Joint Staff Mission in Washington, he was issued with a visa50 on 19 October. He told Joan, The first thing I shall do is to buy a Hershey bar.’

  This was not to be the only purpose of his visit. Now that joint operations were being planned, the Allied authorities were in need of new technology to communicate the more subtle aspects of the Grand Alliance. Telegraph communication was not enough. They required proper means for speech signals. There being no submarine Atlantic telephone cables, all speech transmissions had to go by short-wave radio. But as a June 1942 Foreign Office memorandum had stated:51

  The security device has not yet been invented which is of any protection whatever against the skilled engineers who are employed by the enemy to record every word of every conversation made.

  No one could speak except on the understanding that what they were saying was being overheard in Berlin. There was a row in September 1942, when Prince Olaf of Norway had to be refused permission to talk to his five-year-old daughter, lest it set a precedent for the transmission of uncensored messages by governments-in-exile.

  The essential difficulty of providing speech secrecy lay in the overwhelming redundancy of speech, as compared with writing. While the modular sum of two written messages would require painstaking-work to unravel, the ear and brain could, almost without thinking, analyse and separate a sound signal into conversation, music and background noise. This was only possible because speech signals carried so much more information than was necessary for comprehension. Cryptanalysis thrived on redundancy, whether on routine ‘Probable Words’, repeated triplets in indicators, or re-enciphered messages.’ Any secure form of speech encipherment would have to remove it. The systems in use in 1942 did not attempt to satisfy that requirement. Systems existed which split up speech into pitch levels and then permuted them, preventing casual overhearing. But such ‘scramblers’ were easily penetrated by doing jigsaw puzzle work on a sound spectrograph of the resulting signal. They did not tackle the essential problem. Some efforts were being made at Dollis Hill to create a more sophisticated system, but American developments were much further ahead, and part of Alan’s assignment was to investigate them. It marked a shift from the cryptanalytic to the cryptographic side, reflecting the demands of the more offensive Allied war now being planned.

  Speech secrecy was in another sense also a problem to the British authorities. In 1940 it had not been too difficult: a few jolly clever chaps in a country house were having a go at breaking German codes. In 1941 it changed: Churchill was getting the most important information from a source which only a select few knew about. The problem was to throw a ring of secrecy around this mushrooming department which lay outside the normal structure of the state. But by 1942 the problem had changed again. Bletchley Park was no longer outside the ordinary channels: it dominated them. Its productions were not the spice added to some other body of knowledge. It was nearly all that they had – photo-reconnaissance and POW interrogation adding points of important detail but never matching in scale what they had fresh from the horse’s mouth. There were sixty keysystems broken, producing fifty thousand decrypted messages a month – one every minute. The old days of ‘Red’ and ‘Yellow’ were long over and the soaring imagination of the analysts, exhausting the colours of the rainbow, had plundered the vegetable and animal kingdoms: Quince for the SS key, Chaffinch for Rommel’s reports to Berlin, Vulture for the Wehrmacht on the Russian front. Certain key-systems were used with the proper safeguards, and with these Bletchley was powerless. The Shark system, as they called the U-boat key, still remained intact but for those three days in February and March 1942. But except for these gaps, the German radio communication system had become an open book – to an élite.

  It meant that a cloud of mystery and obscurity was settling over the whole British war. All of its documentation had to be falsified, acting out a charade in which ‘the old procedures’, as Muggeridge saw it,52

  like the setting of agents, the suborning of informants, the sending of messages written in invisible ink, the masquerading, the dressing-up, the secret transmitters, and the examining of the contents of waste paper-baskets, all turned out to be largely cover for this other source; as one might keep some old-fashioned business in rare books going in order to be able, under cover of it, to do a thriving trade in pornography and erotica.

  It was the ability of the British system to absorb the necessary innovations in an ad hoc fashion, when forced to do so, that constituted its real secret weapon. Without that flexibility, all the mathematical and linguistic skill would have been of no avail. Here, perhaps, the habit of paternal Empire enjoyed its triumph. While A.V. Alexander, the trade union man who followed Churchill as First Lord of the Admiralty, was never allowed to know about naval intelligence, let alone cryptanalysis, the prefect layers of the stratified British system were characterised by a trust that allowed them to keep in control and in communication with each other. There were bitter conflicts at every level over what was the most important and exciting development ever to be thrust into the unready hands of British government. But the conflict took place within a club where the rules were agreed tacitly, not enforced by legalities or by handcuffs. Alan Turing could hardly have survived in any other system – certainly not in the German, obsessed by spying and treachery, and perhaps not in the American. He was no team man himself, preferring always to have somethi
ng tidily his own, but they were able to use him as the sixth form Maths Brain, distinguished, as the headmaster had said, within his sphere.

  For those in charge, the fruits of his labours presented logical problems no less difficult than those of Bertrand Russell. Who was to know what, and who to know they knew? Liaison with the very differently organised American system was just one problem; there was the deception of Dominions, free forces and Russians. The capture of cipher material had to be avoided when it was not needed; they had to prevent the ‘indoctrinated’ from ever falling into enemy hands, and above all, they had to arrange convincing explanations for the foreknowledge that successful operations might betray. But how could this be done without vast numbers of people knowing that something strange was happening, and how could the information be used without giving itself away?

  It could not. Bletchley’s continued successes depended upon the willingness of German authorities to believe that ciphers were proved secure, instead of asking whether they actually were. It was a military Gödel theorem, in which systematic inertia rendered German leadership incapable of looking at their system from the outside. Nor was the principle of ‘need to know’ one that worked like a complete and consistent logical system. At Cambridge – and at Sherborne School – people guessed the nature of work being done at Bletchley. In 1941 the Daily Mirror had carried an article53 headed SPIES TAP NAZI CODE, which dwelt proudly upon the work of amateur radio operators who were ‘taking down the Morse messages which fill the air’. In the hands of ‘code experts’, the article explained, ‘they might produce a message of vital importance to our Intelligence Service’. ‘A letter of thanks from headquarters telling us that we have been able to supply some useful information is all the reward we ask,’ said the radio spies themselves. And more significantly, on another part of the board, the initiative of the Red Queen was allowing Soviet authorities access to Enigma decrypts. Yet the system still held together.