Even the most squalid governments have secrets worth grabbing; and after the war, there was a scramble to salvage such treasure from the ruins of Nazi rule. Most famously, there was breathless urgency among the Americans to take control of, and to master, the German advances in rocket technology. For GCHQ and the post-war codebreakers, there were also prizes to be found amid the ugliness and dumb brutalisation. To secure them, and bring them back to Britain, would require experts more used to confining themselves to laboratories.
Nor were these men the world’s most obviously qualified secret agents. One such was Alan Turing. At the time, he had been seconded to a hugely secretive research establishment called Hanslope Park, in Northamptonshire. This institute – centred on a handsome white stuccoed 19th-century house, surrounded with Nissen huts, and all within carefully fenced-off and frequently patrolled parkland – was an offshoot of the main codebreakers’ operation. It was set up by the ebullient Brigadier Gambier ‘Pop’ Parry, under the umbrella of the Radio Security Service. Hanslope Park was on the face of it quite a straightforward military establishment and part of its remit was to explore and push the boundaries of interception technology. But there was also much work being done on new technological means to devise encryption methods.
The work was very much tilted towards the coming age of the computer – the age that Turing himself had done so much to pitch the world towards. Here, in these secure Northamptonshire laboratories, were vast electronic machines, the size of wardrobes, linked with labyrinths of wire and cable. The air was thick with the smell of oil and the heat off thermionic valves. Turing – not in uniform, unlike so many around him – was working on projects such as the ‘Delilah’, an intended means of scrambling and encyphering voice transmissions so that, say, Churchill might be able to talk to Roosevelt on a line that would sound – to interceptors – like the anonymous hiss of white noise. The reason for the name ‘Delilah’ was that it was named after the Biblical figure – the ‘deceiver of men’.
At Hanslope, there was none of the day-to-day intense pressure of Bletchley. Turing was able to work in calm (and rather congenial) surroundings, with his young assistants Robin Gandy and Don Bayley.
The Delilah device had some competition, for the Post Office Research Laboratories at Dollis Hill in north-west London were working at the same problem: that of encoding the human voice. The presiding genius of Dollis Hill, Dr Tommy Flowers – a supremely gifted engineer and also not an obvious choice of secret agent – had hailed from one of the poorer districts of London’s East End. As a young man, he had struggled and worked intensely hard to put himself through night school to pursue his passion for engineering; this is perhaps why some of the Cambridge-educated codebreakers had sometimes treated his ideas with scepticism and occasionally open disdain. Dr Flowers vaulted above such snobbery, and he achieved stewardship at Dollis Hill at a time when telephone technology was starting to evolve dizzyingly fast.
Flowers and Turing had met often throughout the war years, and came to admire one another. Flowers had a genius for giving solid structure to the most abstruse of mathematical theories; a genius that Turing lacked. In 1944, Flowers’s code-smashing Colossus had proved enormously successful and more were built. As the war in Europe ended, the East End engineer and the public-school mathematician found themselves called upon to make an unexpected journey together.
In the summer of 1945, both Flowers and Turing were quietly summoned from their respective establishments to travel to Paris; there, with several other British colleagues, they met with a small team of American scientists. Their onward mission – handled in part by codebreaking veteran Lieutenant Colonel Patrick Marr-Johnson – was one of the most intense secrecy. From the relative lightness of Paris, they then travelled far into the dark ruins of Germany. There was a strong perception of risk: English voices moving through some of the obscurer corners of a traumatised, nerve-shattered population, sometimes deep in the country and far from the safety of the occupied cities, may have been regarded as a provocation to desperate vanquished communities. Not everyone in those remote rural populations would have been willing to welcome such figures. But Turing and Flowers formed part of what were to become known as TICOM expeditions. These were to prove beyond value.
TICOM was the Targeted Intelligence Committee, a joint effort between the Americans and the British to send experts into the depths of former Nazi territory to acquire any scientific or technological secrets that the retreating forces had not wilfully destroyed. Different teams were sent to Peenemünde, to Innsbruck, to Heidelberg, to salvage what they could of Nazi missile technology, as well as advances in such areas as liquid oxygen, infra-red instruments and anti-radar equipment.
Turing and Flowers had their own specific mission: to examine how far the Nazis had got in terms of computerised cryptology. The areas that their team started off with – Frankfurt, Bayreuth – were under American control and there was a great deal of nerve-grating bureaucracy as the scientists had continually to prove who they were. But in that uncertain landscape – there was a prevalent and perfectly understandable fear that many German young men would not relinquish their Nazi loyalty – Turing, Flowers and the Americans travelled far up into the mountains, to a secret radio research establishment, and they made their sleeping quarters in an old hospital. The abandonment of these establishments high up in the wild hills must have lent them a slightly eerie and unsettling feel.
Turing and Flowers were, throughout, implacably silent with their colleagues about all the work that they had done, all the triumphs that they had enjoyed back at Bletchley Park. They also said nothing to those Germans that they met.
Flowers and Turing had the opportunity to meet a German cryptographer who, despite defeat, was intensely proud of the technology that he had worked with. They listened with quiet respect as he gave them a demonstration of the machine that had produced the Tunny code. It was perfectly unbreakable, the German scientist averred. He pointed to the countless millions of different combinations and permutations, the limitless complexities of its codes. Turing responded – with discretion and politeness – with a raised eyebrow and an exclamation of surprise. Neither he nor Flowers were about to tell the scientist that his unbreakable codes had in fact been shattered, and on a spectacular scale.
Elsewhere in Germany, another Allied codebreaker – an American mathematical prodigy who had been brought over to England for the war and who was to later help construct the US codebreaking efforts throughout the Cold War – was engaged in an even hairier mission. Arthur Levenson was seconded to another TICOM unit, set up to grab as much cryptological machinery and expertise out of Germany and Austria as they could. He was face to face with Nazi soldier prisoners and indeed Nazi civilians who might turn on American soldiers without any warning.
‘We wanted the Tunny [encryption] machines,’ recalled Arthur Levenson some years later. ‘And then we had some idea of who the people were and we interrogated them and mostly they were very co-operative. Occasionally you’d get a guy and then we told him if he doesn’t want to answer questions we’re turning him over to the Russians. And they talked. Then when we took this group back to England with us they thought they were very lucky indeed because we passed along the way what they called dustbins. These’, he explained, ‘were just curious enclosures, all out in the open, teeming with prisoners, and they were much better off being with us. So Major Tester and I took a dozen of them back to England.’1
Given the horror of what the Allied troops had uncovered deep in the forests of Eastern Europe, how did Levenson, who was himself Jewish, manage to stay so apparently at ease with his prisoners? A distinction had been drawn, if subconsciously; for these prisoners were in essence cryptographers too. ‘Then we became very friendly,’ Levenson said. ‘These were not combat troops, these were communications… They were obviously not going to run anywhere. In fact, they felt much better off with us than anywhere else, including escape. Because they’d only be picked up and terrible things would happen… And then we landed in London. Then one of them, a nice little old man, said he had always wanted to go to London. Just before the war broke out, he had finally got enough money and then the war came… I said “well, now you’re getting a free trip.”’
This extraordinary TICOM convoy, comprising six trucks, invaluable top-secret equipment and German prisoners of war, wound its way to Buckinghamshire, to a site not too far from Bletchley Park at Beaconsfield. It was time for the prisoners to demonstrate the machinery, and to unlock the last of the German encryption secrets. Because clearly, the principles of such technology would still be in use in a variety of other places.
‘About 100 yards [90 metres] apart, we set up the antenna, they broadcast to each other, and they showed us how they set the patterns, what the whole procedure was,’ recalled Levenson. ‘They told us there was the death penalty for sending depths but they had sent depths [a series of messages encyphered with the same code machine setting, more vulnerable to breaking than single messages encyphered with different settings]… They were still alive.’ And the much-desired Tunny machines did not disappoint. ‘They were good, they were well constructed. They had been built when things were good… they had no ersatz material. They all used steel.’2
Levenson had his military training; but the German interlude, this exploration of a country in the throes of a nervous breakdown, must have been altogether more disconcerting for the civilians Tommy Flowers and Alan Turing. As it happened, on the days they were up in those eerie mountains in the secret German codebreaking station, Hiroshima and Nagasaki were wiped out in a flash of pure white; the military apotheosis of scientific research, the atomic bomb, had now opened up vistas of annihilation of which no-one before could have dreamed, even in the darkness of Nazi Germany.
But for both men, away from all this horror, the age held other rather more exhilarating opportunities. Turing had been wondering whether to return to his academic career at King’s College, Cambridge after a six-year absence; Tommy Flowers, in the meantime, having been so instrumental in bringing the computer age into being, had a great deal more work to do in his north-west-London laboratory.
From a certain angle, it might be tempting to wonder if Dr Flowers’s post-war projects might have had an element of anti-climax to them. Almost immediately, he fixed his focus on making electronic telephone exchanges a practical possibility. Certainly this was vital work: a new technologically advanced era would need lightning-fast communications between cities, countries, continents. For all sorts of commercial and political – as well as security – reasons, it was in Britain’s dearest interests to keep itself at the forefront of all such developments. This is doubtless how it seemed to Dr Flowers – but how often did he allow himself to think about the leaps in computer science that he had made possible?
At the end of the war, as far as Dr Flowers knew, the Colossus machines – his proud creations – were destroyed for reasons of secrecy. Strictly speaking, Dr Flowers was not even permitted to hold on to the blueprints. He was told that no-one was ever to know of the extraordinary codebreaking feats that he had made possible, after engineering efforts that he had funded out of his own pocket. The government ensured at the end of the war that he was rewarded and reimbursed – he received the MBE, and £1,000, which in those days would have been enough to buy a house – but could the money cushion the intense distress he felt at having his beautiful creations dismantled irretrievably?
Ironically, as we have seen, any distress he suffered was perfectly unnecessary: Commander Travis had quietly decided, amid the gradual closedown and moving of Bletchley Park, that two of the Colossus machines would survive. In the chaos of post-war international politics, who knew who would end up using such technology next?
Visit the Bletchley Park Museum now and you will find an extraordinary re-creation of the Colossus, a great mass of metal, winking lights and fragile-looking white paper tape. It is barely 70 years since these contraptions were so sensitively classified that they had to be transported under shrouds of deepest secrecy from Bletchley to the Eastcote codebreaking station in Betjeman’s Metro-land. And it is very easy to imagine other countries in the 1940s using Lorenz-style technology for their encryptions. For some time after the war, for instance, in the Russian-controlled sector of Germany, there were some Communists who quite blithely and unknowingly continued with Lorenz encryptions.
For Dr Flowers, there might have been the luxury of abstract speculation about such matters, but actually, he was too much of an enthusiastic pragmatist to dwell on what had been; he was drawn to the radon glow of the future. Within a few years, his work of electronic exchanges came to fruition. He can also lay claim – a lighter claim to posterity, this – to have been the genius behind ERNIE. The acronym stood (or rather, still stands) for ‘Electronic Random Number Indicator Equipment’. It was invented specifically for the advent of Premium Bonds in 1956; these government bonds came with the incentive of being eligible for Lottery-style cash prizes. And so it was that Dr Flowers engineered a computer that could generate completely random numbers, to a set number of digits. ERNIE came to be frequently mistaken for a real person. In 1980, the band Madness dedicated an entire song to the machine and its promise of untold riches for ordinary people.
And what of Bletchley’s other computing pioneers? William ‘Bill’ Tutte, the young man who had been so brilliantly instrumental in finding a lever into the hellishly complex Lorenz codes – the breaks that then led to the development of the Colossus – returned immediately to Trinity College, Cambridge after the war. There he gained his doctorate and pursued his work in the realm of pure mathematics and extremely advanced geometry. However, like many of his former colleagues who stepped out into the wider world, the links with the old establishment would prove remarkably tenacious across the years. Indeed, for some, the links were made of purest elastic, and after a few years, they would find themselves twanging back into the secret heart of codebreaking.
Then there were senior figures such as Professor Maxwell ‘Max’ Newman, who had headed up the department known as the ‘Newmanry’. The intellectual excitement of what had been started there could not be quelled. He understood very well just how this computer revolution would unfold and he very much wanted to be at the front of developments in it. He joined Manchester University a little later in 1945 as the Fielden Professor of Mathematics, a post he occupied for many years thereafter. As well as running the department (he was a popular fatherly figure with a very dry wit), Professor Newman was determined that the university should build its own fully functioning computer; this at a time when much of the pioneering work in the field was carried out by the American military.
But Professor Newman was not completely adrift from his old life at Bletchley Park. Very far from it. As well as keeping active contacts open with his old codebreaking friends – and indeed very quietly advising them – in 1945 Newman recruited one of his brightest young colleagues, Irving John (IJ) Good to join him in Manchester for this ground-breaking computer work. The 29-year-old Good became a mathematics lecturer by day, a computer whizz by night. He and Newman and an engineer called Tom Kilburn worked towards the building of a computer called the Manchester Mark One. It was one of the world’s first stored-program computers; a step up from Colossus in the sense that an electronic machine could now be said to have a memory. As we shall see later, the work carried out here was to have a direct impact on the codebreaking efforts in London.
Computer development was far from being an abstracted academic pursuit. In later years Professor Newman’s work would come to be tied up very closely with wider matters of national security and defence. Theory was one thing; but there was a terrific sense of urgency about its application as well.
Alan Turing was, at this time, also thinking hard about the future of computing. Having theorised about a ‘universal machine’, he was now set on creating an electronic brain. And so it was that he left Hanslope Park to join a rather more open civilian outfit. As well as the GPO, and Newman’s team at Manchester, there was one other site that was to be dedicated to this new realm of computer research: the NPL, or National Physical Laboratory in Teddington, in the south-western corner of London. But in those straitened days, there were also financial limits and increased competition for funding – and as it happened, Professor Newman had managed to win a Royal Society grant for his Manchester department to plough extra investment into the development of their technology.
Turing joined the National Physical Laboratory in 1946; by that time, his mind was a chain reaction of possibilities that led to the concept of ACE – the Automatic Computing Engine. This would be a machine that could calculate algebraic equations, then drop the task and turn instantly to playing chess; this would be a machine that could perform standard mathematical functions, but then be switched to codebreaking.
The machine would even, to a very limited extent, ‘think’ for itself it would have an ability to extend and extrapolate its own programs. The very idea of ACE initially provoked interest in departments far removed from the laboratory: there were industrialists in Whitehall who understood very quickly how such a thinking machine might change the nature of manufacturing. And there were also – as always – rather less public figures in the realm of defence and the War Office who also saw its potential. This meant particularly the chance of turning such a thinking machine to problems of high explosives, and new sorts of bombs.
The elastic that bound Turing to his old codebreaking colleagues was unusually strong, too. The masterminds at Eastcote would later try all manner of enticements to get him to return full-time to the fold of cryptography. In the meantime, vitally, Alan Turing, along with Max Newman, held on to his prized security clearance throughout these immediate post-war months and years. As indeed did Dr Tommy Flowers of the General Post Office research department. They never stopped being in on the latest cryptanalytical developments, whispered on the most unimaginably secret of grapevines.
As to the dawn of the computer age: in the early days of discussions and costings, the head of the National Physical Laboratory Sir Charles Darwin (grandson of the more famous namesake) wrote to the relevant room in Whitehall about the possibility of borrowing the Dollis Hill team and specifically Dr Flowers to bring the NPL’s new computing project to life. In so doing, Sir Charles coyly referred to Tommy Flowers’s previous triumph at Bletchley Park.
‘Very broadly it [Turing’s machine] works using principles developed by your staff during the war for a certain Foreign Office project,’ Sir Charles wrote, ‘and we want to be able to take advantage of this enlisting the help… in particular of Mr Flowers who has had much experience working out the electronic side of it.’3
Turing’s ACE computer creation was finely detailed and exquisitely wrought – but only in blueprint form. There were huge obstacles to bringing it into the world. Partly a question of complexity, and partly a question of money – an enormous outlay would be needed – it also seemed that the competition from Professor Newman’s department in Manchester might stifle it, on the grounds that Max Newman was felt to have a shade more practicality than the supremely unworldly Alan Turing. It has been noted by Turing’s biographer Dr Andrew Hodges that when Professor Newman’s department was awarded the Royal Society grant for its own computer development, it was a rare occasion when the machines were not expected to be turned to immediate military or espionage use.
Yet the thread seemed always to lead back to cryptology. Another of those who left – and yet got snagged back into the currents of codebreaking – was the brilliant mathematician Shaun Wylie. Another of the young men who had worked alongside Alan Turing in Hut 8 during the war, Wylie had also enjoyed great success with the teleprinter cypher Tunny codes that Max Newman had given so much thought to. Thirty-two years old when the war ended, and already a foremost figure in the abstruse mathematical realm of linear algebra, Wylie married his Bletchley Park love Odette Murray and then went back to Trinity Hall, Cambridge. At one point, he was to give some helpful advice to Watson and Crick who were in the midst of researching the fundamentals of DNA; Wylie made suggestions about the way that the double helix might lose its structure.
Naturally, Wylie said nothing to his colleagues of his wartime years (though it must have helped that among those colleagues were other codebreakers, such as Peter Hilton – together, they might almost have revelled in the enforced silence). Equally, though, the post-war codebreaking operation was careful to keep up with Wylie; to the extent that, in the early 1950s, he was lured away from Cambridge to go back into the espionage business full-time once more, this time as GCHQ’s chief mathematician.
At Bletchley, Wylie had not only been a formidable cryptographer; he was also noted for his skill in acting and drama, and his very active role in the performing societies there. There was one other man – neither a mathematician nor a scientist – who agreed to stay on after the war, while at the same time leading quite literally a theatrical double life: Frank Birch. The almost preposterously colourful Birch had been one of the grand old codebreaking veterans of the First World War, who had played an active role in the cryptanalytical efforts of Room 40 in Whitehall. By the end of the Second World War in 1945, he was 55 years old and juggling careers.
An Old Etonian with a strongly academic side (and the purple-faced temper of a true martinet), Frank Birch’s career had by now bifurcated to an almost absurd degree. Some years before, just after the First World War – in the aftermath of which Birch wrote an intensive history of Room 40’s successes and methods for internal use only – he had become a fellow of King’s College, Cambridge, and a popular history lecturer. But in 1928, he succumbed to his more artistic side, left the college, and moved into theatre. Indeed, between the wars, right the way through the 1930s, Frank Birch pursued an extraordinarily immersive acting and producing career, spanning theatre and film. He staged farces, which were enormously popular; he also took supporting roles in everything from detective dramas to drawing-room comedies and adaptations of classics.
The Second World War had drawn Birch back in full-time to codebreaking duties at Bletchley. Indeed, beforehand, in 1938 (just as he was producing a French farce for the London stage), Birch was advising the then director Alastair Denniston of the sorts of young people they would have to think about recruiting. As the war progressed, Birch’s influence grew; he was increasingly involved with Hut 8 and the desperate fight to break the German Naval Enigma. Although Birch came from a different codebreaking generation to that of young Alan Turing, he saw the organisational difficulties that the Hut 8 team were up against, and how their efforts were stuttering partly because they did not get enough time with the proto-computer bombe machines (other departments made even louder demands). It has been suggested by historian Ralph Erskine that it was Frank Birch who was the man ultimately responsible for Alastair Denniston being moved sideways from Bletchley and Commander Travis taking over.
And most notably, it was Frank Birch – the gifted theatrical farceur – who was instrumental in starting the historically unprecedented intelligence-sharing accord between Britain and America. He went to Washington DC with Commander Travis in 1942 and the talks they had with their codebreaking counterparts there led to the Holden agreement; this was specifically to do with the sharing of naval codebreaking of the German Kriegsmarine cyphers. This set down an extraordinary level of trust between two allies (the trust being not so much to do with intentions as with security concerns, and the crucial necessity of the secret being kept on both sides). Birch – as skilled a diplomat as he was an actor – later built on this agreement. Out of the conflicting demands and needs of the Americans and those British codebreakers working in Ceylon at HMS Anderson came the first seeds of the BRUSA (Britain–United States of America) agreement, which led to even more sharing of resources, knowledge and intelligence.
And now, the war over, Birch was prevailed upon to remain a codebreaker; he undertook the job of writing up a detailed history of naval signals intelligence. And yet, this man – known as an unusually stern authority figure at Bletchley – was at the same time poised to spring back under the studio arc lights in an extraordinary range of roles. Film was one thing: Birch had also spotted the exciting new medium of television, and as the late 1940s wore on, he won parts in a variety of screen productions such as Rotten Row, Operation Stocking, Charles and Mary (a drama about Charles and Mary Lamb) and an adaptation of Thérèse Raquin. He was also in the cast of a serialisation of Dickens’s The Pickwick Papers.
Perhaps the most startling of Frank Birch’s forays into the world of showbusiness – while he continued his top-secret work within the world of cryptography – was a role in the 1953 film comedy Will Any Gentleman…? in which he starred alongside George Cole and Sid James. It is almost impossible to imagine, by means of comparison, any of Birch’s heavyweight American codebreaking counterparts making appearances in Abbott and Costello films. Birch, who was married to the Hon Vera Benedicta, also contrived to live in some theatrical splendour in the swish Montpelier Walk in Knightsbridge, London. He was one of the great examples of codebreakers having unexpectedly rich hinterlands.
More importantly, Birch was also a great example of how the esoteric world of codebreaking had a particular lure of its own; that even those who left to embark upon the most wildly divergent projects could never quite resist the chance to return. It was that combination of addictive intellectual rigour and the satisfaction of being privy to the gravest of all national secrets.
On this level, indeed, the lives of the cryptographers were many times more satisfying than those of agents for MI5 and MI6. In the case of the latter, work was often grindingly humdrum, and lacking any real intellectual or even moral challenge. The point about codebreaking, however, was that one always knew that one was making the most direct and concrete kind of difference. On top of this, the codebreakers of GCHQ were wielding global clout at a time when Britain’s power as a whole was starting to recede sharply. Increasingly, the codebreakers would find themselves being pushed to the front line of the nation’s defences.