The Spies of Winter

by Sinclair McKay

  In an age when the term ‘Northern Powerhouse’ is witlessly deployed by public relations officers, it is instructive to think of the pioneering work that was taking place 70-odd years ago within the soot-smirched gothic buildings of Manchester University.

  Naturally, while figures such as Turing had a passion for pursuing a vision of thinking machines as an end in itself, there were many in Whitehall who focused on the rather more practical applications; not just scientific, but military too. For all those who saw the poetry in storing data in a cathode ray tube – the data actually being visible as dots of light that would dance – there were many others who were thinking defensively. How could they not? Both America and Russia had tacitly agreed to their roles as opposing superpowers, both locked in what they saw as struggles for survival. A key component of the Cold War would be about technological triumph. Later, this would move to Outer Space, as well; in the 1940s and early 1950s, computing was the mark of scientific superiority.

  Thanks to the war, there were those who moved with genuine ease between the scientific and intelligence communities on both sides of the Atlantic. Turing was one, having been sent over to the US in 1943. Another was the Bletchley organisational genius Gordon Welchman who, as we have seen, decided to emigrate to the United States full-time with his family in 1948, to pursue computing and electronics in the private realm. Another figure – who was to prove absolutely key to GCHQ at Eastcote and Cheltenham, as well as then progressing to being a hugely respected academic in America – was Irving J Good, or Jack Good as he was known to all his friends.

  Good had been something of a prodigy. Born in Hackney in London’s East End in 1916, a childhood bout of diphtheria led to his confinement to bed, and also one of those bedazzling moments of intellectual revelation: in his case, discovery of ‘the irrationality of the square root of two’.1 He went to school at Haberdashers’ Askes school in north London, then went up to Jesus College, Cambridge to study mathematics. When war broke out, he was specialising in ‘fractional dimensions’; one reason, perhaps, why he was not considered ideal material for the infantry. Instead, the more esoteric world of cryptology beckoned, and with it came many strong, abiding friendships.

  The post-war dispersal of many of the codebreakers meant, for Good, the acceptance of an offer from Professor Max Newman to come and teach at the University of Manchester. But while teaching mathematics in theory, Good was also by this stage utterly mesmerised by the new field of computation. So while Alan Turing was fighting for funding down at the National Physical Laboratory, Jack Good was contributing ideas and expertise to the Manchester ‘Mark One’ computer.

  And after three years of this, Good was once more drafted southwards, towards the cryptanalytical corridors of Eastcote. It is simply impossible to imagine that this wonderfully witty codebreaker left all his computational knowledge behind him. The premises of Eastcote had – as mentioned before – housed a great many of Turing’s bombe machines throughout the war, all working their way through the different worldwide versions of Enigma. The machinery that replaced them was no less jealously guarded. Jack Good – like everyone else who worked there – was careful to say nothing to anyone.

  But the lines of communication between Manchester and Eastcote were as strong as their US equivalent of Arlington Hall with Harvard or indeed Bell Laboratories. Although he kept his secrets, Good did remark jokily in later years that his post-war work might have had very different outcomes had he chosen to reverse his career path. He said that if he had gone back to being a pure mathematician, ‘he might have discovered something so important that bad people would have done harmful things with it, whereas his current work at GCHQ was pointless but harmless’.2

  Incidentally, another measure of Good’s fantastic fascination with the uses of computers and the boundaries of mathematics came in 1968 when he gave advice to the film director Stanley Kubrick upon the making of the metaphysical science-fiction epic 2001: A Space Odyssey. Like Wittgenstein, Bertrand Russell and of course Turing, Jack Good also had an abiding interest in the pursuit of philosophy; the precision of mathematics was considered important in establishing absolute clarity in terms of philosophical ideas.

  Good and Turing also enjoyed lively correspondence; Good used to refer to him jocularly as ‘the Prof’, even though Turing had not quite attained this status. Just before Good took up his new position with GCHQ in 1948, there was still time for rather more abstract pursuits. ‘Dear Prof,’ wrote Good to Turing in the summer of that year, ‘When I was last in Oxford I met a lecturer in physiology who said that he thought the number of neurons in the brain was only about two million. This seems amazingly little to me even allowing [for] the fact that the number of processes from each neuron is something like 40. I wonder if you could tell me the right answer, with or without a reference.’3 The question was not wholly random, and indeed might be read as a form of code in itself: at this stage, Turing had been doing some research into the structure of the human brain as a means of investigating further the idea of constructing an electronic brain. He had been delving into how the intellect grows from infanthood. It was not enough for a machine simply to obey instructions. It must also, at some points, learn how to make its own decisions. How would such an electronic structure be possible? Good’s question might have been by way of a slanted request for progress, from one brilliant mathematician now heading back to the life of codebreaking to another working in a field that all the codebreakers would have to become completely familiar with. In the same letter, Good went on to impart some discreet job news, having heard of Turing’s imminent departure from Teddington to Manchester.

  ‘I understand that by next October, we will have swapped towns,’ wrote Good. ‘Judging by the international situation I think you’ve had the better of the bargain.’4 It is a teasing reference; Turing must surely have known where Jack Good was going to work and the kind of Cold War climate that the cryptologists faced. But Good ended on a much lighter note. ‘How near were you,’ he asked Turing, ‘to getting into the Olympics?’5 It was not a facile question: Turing was a seriously good long-distance runner, with a marathon time that more than qualified him to take part in the 1948 Games which were held in London. Unfortunately, by that time Turing was carrying a leg injury. Previously, he had managed to run marathons in around two hours and 30 minutes: extraordinarily good for a hobbyist.

  Ensconced among his fellow cryptanalysts at Eastcote, Jack Good wrote again to Turing. There was more chat about the brain and neurons. But there were other developments he was keen to pass on. ‘Dear Prof,’ he wrote, ‘…I visited Oxford last weekend. Donald [Michie] showed me a “chess machine” invented by Shaun [Wylie] and himself. [Michie and Wylie were former Bletchley colleagues.] It suffers from the very serious disadvantage that it does not analyse more than one move ahead. I am convinced that such a machine would play a very poor game, however accurately it scored the position with respect to matter and space. In fact, it could easily be beaten by playing “psychologically” ie by taking into account the main weaknesses of the machine…

  ‘When in Oxford’, Good continued, ‘I succeeded in hypnotising Donald… Would you agree that a very typical property of the brain is the ability to think in analogies? This means taking only a part of the evidence into account… Do you know of any reference to Russian electronic computers?’6

  That blend of intellectual high spirits and sly fishing for intelligence – were there any rumours in the Manchester scientific community of Soviet advances in this field? – blended in with the now reflexive opaqueness about the work he and the others were doing.

  In a curious way, the story of their former colleague Gordon Welchman illustrates the sorts of dilemmas that Turing, Good and, at GCHQ, Hugh Alexander and Joan Clarke faced, as the possibilities of the electronic age widened out. For as the work at Manchester University demonstrated, there was clearly also going to be quite a commercial future for computers. For Welchman, all that lay across the Atlanti
c in large firms where his particular brand of expertise would be very highly valued; his extraordinary security clearance would see to that. Back in Britain, the opportunities might have seemed more cramped. However, the spirit of innovation was still very much there. When Turing joined Professor Max Newman, work on the ‘Mark One’ being carried out by FC (Freddie) Williams and Tom Kilburn was at a very advanced stage: memory was no longer held in cathode ray tubes, but in magnetic drums, with the addition of teleprinter read-outs, a less cumbersome way of programming with a single keyboard.

  In 1948, the government was not wholly blind to the possibilities either: Sir Henry Tizard, who had been behind so many advances in the war, and was now chief scientific adviser in Whitehall, was mesmerised by the possibilities that he had seen in Manchester. Acknowledging that money was tight, he made it clear that development must be carried out ‘speedily’; that Britain should try to ‘maintain the lead’ that it had taken in this new discipline, even as the Americans were working with utmost concentration on their own projects, and that he personally would give full support when it came to ‘supply of material’ and ‘obtaining necessary priorities’.

  There was a certain amount of hard-headed financial realism. A local company called Ferranti was called in to work with the Manchester mathematicians to make the world’s first commercially available computer. Turing was one of the team who sought to make this fantastically complex concept into something that middle managers could use.

  While Alan Turing was interested in the wider intellectual possibilities – an electronic mind that could play chess, and brilliantly parry the serpentine twists and gambits of the organic mind, or even, as he suggested on one occasion, a machine with television cameras attached that could take an interest in food, sex and sport – Ferranti was thinking of the business applications, such as those early punch-card machines favoured by the John Lewis Partnership. In its early advertising literature, Ferranti explains that a man sitting at a desk can only perform so many calculations a day; the Ferranti computer (or the Manchester Mark One) would be able to perform the same number of calculations in two seconds. The machine would also be able to carry out all wage calculations, and other assorted administrative tasks. But beneath this practicality lay an advance that was, in its own way, as significant as nuclear power.

  If Professor Max Newman’s brilliant team at Manchester was developing miracles, then what were the Soviets achieving in Moscow? Jack Good would not have been alone in wondering this. The cryptanalysts at Eastcote, drawing in signals from military bases, from cities, from embassies, also needed to be able to eavesdrop on the work of Soviet scientists.

  Intriguingly, Alan Turing himself could not resist the curiosity: what if the Manchester computer could one day be used to generate completely random numbers? If so, it would become, in principle, a genuinely unbreakable cypher-generating machine. Turing wrote: ‘I have set up on the Manchester computer a small programme using only 1,000 units of storage, whereby the machine supplied with one 16-figure number replies with another within two seconds. I would defy anyone to learn from these supplies sufficient about the programme to be able to predict any replies to untried values.’7

  ‘It was during this period’, wrote Turing’s biographer Andrew Hodges, ‘that he found himself being consulted by GCHQ. It would indeed have been remarkable if they had not consulted the person who knew more about cryptology and the potential of electronic computers than did anyone else. And had he not described cryptanalysis as the most “rewarding” field for the application of programming? Few, however, were in a position to perceive this fact, the subject being more secret than ever.’8

  The friendships and associations and intellectual quests continued. The 1950s brought a new scientific fashion for ‘cybernetics’ and there were meetings of like minds at what was called The Ratio Club; Turing gave a talk on ‘Educating a Digital Computer’ which was attended by a highly interested Jack Good.

  And the cat’s cradle of codebreaking links grew denser. At Manchester, Turing was eventually joined by the young mathematician Peter Hilton who had worked with him in Hut 8. Hilton was recruited to the university from Oxford by Max Newman. There was little sense that Manchester was a provincial come-down from the dreaming spires; rather, there was intellectual excitement combined with the sense of a new kind of country being formed out of the convulsions of war. It was a country prefiguring Harold Wilson’s famous phrase ‘white heat of technological revolution’ some 15 years before he coined it. Turing gathered Hilton into the further developments in the computer department – and both men were linked by the silent secrecy of the proto-computers that they had worked with at Bletchley.

  Like Turing, Peter Hilton was gripped not only by the applications of mathematical logic, but also by the underlying philosophy. Bletchley’s graduates might have scattered wide, but the bonds they had forged were strong. Hilton, in turn, was good friends with Shaun Wylie, a mathematician reckoned by many of his contemporaries to have been one of the best all-rounders at Bletchley Park: gifted at cryptanalysis, a youngster who made a big impact with his work on Tunny, but who was also a dedicated fan of amateur dramatics (a little like Nigel de Grey and Frank Birch) and who understood the importance of immersing oneself in pursuits other than codebreaking. After the war, Wylie went to Cambridge where he was made a Fellow of Trinity Hall; he was fondly remembered as a brilliant teacher by students – but he could not stay away from the pursuit of cryptology for long. Eventually, in the late 1950s, Wylie relented and re-joined GCHQ, now as their chief mathematician. Along the way, he and Peter Hilton co-authored a book on aspects of topology. The discipline of codebreaking was, for all of these people and many others, an aptitude that made them constantly in demand. The secrets of Bletchley could never be unlearned; and these gifted people had a talent that was now needed in a war that could conceivably stretch out without end.

  In the Manchester mathematics community, the proximity to the manufacturers Ferranti was intriguing for another reason. It represented a further cat’s cradle intersection of incredibly sensitive work – not merely on computers, but also on military technology, such as the development of automated guided missiles. All of which was perfectly natural; the government, and foreign secretary Ernest Bevin in particular, were anxious that Britain should be able to stand alone in terms of defence. It was around this time, too, that – elsewhere – Britain’s first atomic weaponry was ready for testing. This was about independence: a proud nation that should not have to look across the Atlantic for its nuclear protection.

  Alan Turing, his young colleagues recognised, was both an unearthly genius and a rather slovenly engineer. This was not laziness on his part; he was forever attempting to construct lash-ups himself, both in the laboratory and also at his house in Wilmslow. In domestic terms, he always resisted the idea of getting workmen in; if a new path had to be laid, then he wanted to be the one to lay it. He just couldn’t do it very well. And so, too, with the delicate mechanics of these vast new computing machines. The young men and women around him were rather startled to find, one day in the early 1950s, that Turing had been made a Fellow of the Royal Society. One colleague confessed that he had never really seen him as Royal Society material.

  Turing also inhabited the twilight life of Manchester. At Bletchley, he had been relatively open about his sexuality with colleagues. Yet homosexuality was still very much illegal – Turing mistakenly believed that a government commission was looking into the possibility of decriminalising sexual acts between men – and the atmosphere was growing steadily less tolerant. Manchester had some streets in which gay men, who had come to understand the code, could meet up: furtive, watchful encounters in darkened, smoggy alleys, as steam engines screamed on viaducts above. Turing – so hesitant and curiously unable to engage in other circumstances – knew these codes well. In time, he came fatefully to meet the eye of a young man called Arnold Murray. This was the start of the affair that would lead to the tragically e
arly end of Turing’s life.

  A little time before this, the codebreakers at Eastcote had made a serious effort to employ Turing’s services full-time once more. Thanks to his continued friendship with Hugh Alexander, Turing was readily available to consult, and his advice was always considered extremely valuable. So valuable in fact that Alexander offered Turing a salary of £5,000 a year to come back to work on cryptanalysis at Eastcote full-time. To put it in perspective, that sum would now be roughly equivalent to £100,000: a substantial amount for a cash-strapped department to offer for an invisible civil service position.

  We might extrapolate a little further than this, it would not only be the British who would be benefiting from Turing’s codebreaking genius: his admirers in America would also have been very pleased to have such a mind back in this enclosed community. Certainly, the level of the financial offer was most unusual at a time when every other economic transaction in Britain was gaunt with austerity.

  Yet the climate was changing. The earthquake of paranoia produced not only by the Venona decrypt revelations, but also by a wider, almost unconscious national sense of an enemy within, would go on to make men like Turing outcasts. In other ways too, the world was becoming daily less forgiving and more fearful. As the Americans developed their first hydrogen bomb – a weapon so nightmarishly powerful that even Winston Churchill shuddered at it – the game of espionage became ever more deadly serious. And against this, the codebreakers would also find themselves facing the complete disintegration of British power in the Middle East. In many ways, this period was to be a stern test of whether the Eastcote operatives could match up at all to the intense pressures and expectations weighing upon them.