Despite the machine’s assistance, however, the work was still laborious. “They [the Hut] would put the message to be broken on the table and if you failed to break it, you’d put it on a pile and then somebody else could have a go at it,” says Captain Jerry Roberts. “And if they broke it, well and good.
“But that pile of the non-broken messages, the dead ducks, was never big. I deduced from this that we were breaking certainly better than 90 percent of what we were being given to break.
“It certainly did require patience but also don’t forget if you’re having a high success rate, even if it took you four, five, eight hours to break a day’s traffic, the reward was very great because you could decipher a lot more messages using the information you’d discovered.”
And it could be a dispiriting business. In 1944, a slightly reluctant codebreaker, Roy (later Lord) Jenkins, found himself pulled away from active duty to a job in the Newmanry. There were a few other distinguished figures on that “Tunny” codebreaking crash course at Bedford too, it seems. Lord Jenkins wrote in his memoirs:
I was greeted with news of instructions from the War Office that I was required for special intelligence work and that I was to proceed at the beginning of January to an unspecified “course” at Bedford. I was destined for cryptography at Bletchley Park and work on the messages sent out by the German High Command in Berlin to the various commanders in the field, Rundstedt, Kesselring, Mannstein, Rommel, and several others. A. D. Lindsay had been involved and had decided that the traditional role of Masters of Balliol…of placing Balliol men in what they regarded as appropriate jobs outweighed any irritation with my poor philosophy mark. Why he thought I would be a better cryptographer than a philosopher I do not know, but the fact that he did appeared to be decisive.
Those on the Bedford course gave the impression of having been incongruously gathered in from the hedgerows. There was Charles Buckingham, erudite curator from the British Museum, who wore a private’s uniform…there was Francis Dashwood of West Wycombe Park, who was a civilian recruited straight from school, there was a very unglossy university-educated North Midlands second lieutenant who shared a civilian billet with me, there was a sophisticated Etonian other rank who lived in Sloane Street, and about ten others I cannot remember distinctly.4
As Jenkins recalled in an interview: “You could spend nights in which you got nowhere at all. You didn’t get a single break, you just tried, played around through this long bleak night with total frustration and your brain was literally raw. I remember one night when I made thirteen breaks. But there were an awful lot of nights when I was lucky if I made just one, so it was exhausting.” Quite so, although another veteran recalled that Jenkins, although a first-class mind, was “not the world’s most talented codebreaker.”
But the work—when one hit a successful streak—had its own extraordinary rewards. The messages now being intercepted were invaluable. These were communications from Berlin to Italy, and to German divisions on the eastern front. Captain Jerry Roberts recalls one extraordinary day when he realized just how far into German High Command they had managed to penetrate: “The people the messages were going to and coming from would be given at the beginning of the message. So you would have General so and so sending to Army HQ in Berlin. Most of them were signed by a general. Some were signed by Hitler. I can remember myself deciphering at least one message—he just called himself: ‘Adolf Hitler, Führer.’”
For the brilliant minds who were making the Colossus possible, it was not so much knowledge as mechanical problems that they faced. It was very difficult to keep the teleprinter tapes synchronized. The machine also put a huge amount of strain on electronic valves, which repeatedly failed. But in the course of the many indepth three-way discussions between Professor Newman, Alan Turing, and Tommy Flowers that took place in order to improve the machines, what was debated was the feasibility of what was, in essence, the very first computer.
To be precise, they knew it as an “electronic valve machine”; what made it a step up from the Heath Robinson was that instead of relying on two tapes that had to be run synchronously, it had just one. Flowers also made other modifications, such as replacing the sprocket drive with a set of friction wheels. As a triumph of engineering, and of bringing a theoretical design to life, it was unquestionably his; although the astonishing machine had much of Turing’s logical reasoning at its core, there was also ingenuity in the way that Flowers improved upon the electronic valves, making sure that they were never switched off once they were in service, which hugely improved their reliability. Moreover, the contraption could read 5,000 characters a second, five times faster than the Heath Robinson. Flowers and his team at the Post Office research department developed and built it in a matter of ten months.
On November 25, 1943, the first Colossus machine was given a test run. By the beginning of 1944, this machine and its ilk were proving invaluable to Bletchley.
Among those who worked with the machines was Harry Fensom, who recalled:
The Colossi were of course very large, hence their name, and gave off a lot of heat, ducts above them taking some of this away. However, we appreciated this on the cold winter nights, especially about two or three in the morning. When I came in out of the rain, I used to hang my raincoat on the chair in front of the hundreds of valves forming the rotor wheels and it soon dried off.
Of course it was essential that the machines were never switched off, both to avoid damaging the valves and to ensure no loss of codebreaking time. So there was an emergency mains supply in the adjoining bay which took over automatically on mains failure.5
The machines had an element of the mad scientist’s laboratory equipment to them—not merely because of the extraordinarily long message tape that had to be carefully threaded, but also the ever-present hazard of electrical danger. In an echo of a previous incident with one of the bombe machines, Fensom continued:
One day I heard a shout from Gillie Sutton, one of the Wren operators.… I rushed over; she had rested her vanity mirror on a convenient part of the control desk, there was a flash, and it melted away. She had placed it across two terminals, very big brass-looking knobs. Unfortunately this time she had used a metal mirror and the terminals were live.6
It was not fully appreciated at the time how much Tommy Flowers had invested in all this work. And Captain Jerry Roberts still feels that Flowers has never received an ounce of the recognition that he should have done. Although Flowers was discreetly awarded the MBE after the war (and also given an Award to Inventors in the sum of £1,000), Captain Roberts feels that a man of his talents and achievement deserved, and should have received, far more attention.
But there was another dimension to Flowers’s work, and that was the skepticism and occasional hostility with which he was treated by Gordon Welchman.
Thomas Flowers was born in East Ham, the beating heart of London’s East End, in 1905. He was the son of a bricklayer. His story is a powerful and surprising illustration of the power of education, in a period long before the term “social mobility” became common currency.
As a boy, Flowers showed a precocious talent for mechanics and science. Thanks to this, he won a scholarship to a technical college. In the 1920s, this in turn led him to enter an open competition for a job as a trainee telephone engineer, a competition in which he came first. Early telephones came under the aegis of the government-run Post Office, and as it turned out, thanks to technological progress, it was an exciting time to be joining. The telephone network was becoming more automatic—with direct dialing, as opposed to going through an operator—and the Post Office’s engineers would be expected to keep up with the latest innovations.
To do so, Flowers had to take evening courses, for the generosity of his new employers did not extend much beyond giving him a job. So he worked by day and studied theory by night. This industrious—and enthusiastic—approach meant that by 1930, he was seconded to the Post Office Research Establishment at the top of Dolli
s Hill.
The nature of his work was now advanced research, into such matters as how international telephone calls might be made by direct dialing. This led to work with electric circuits, and probing the various uses of electronics, a science very much in its infancy. Several years later, the matter of electronics would also come to fascinate Alan Turing at Trinity College, Cambridge.
The two men met for the first time at Bletchley Park in 1939. Flowers had been asked along by his boss Dr. Gordon Radley, director of the Post Office Research Office. He and Flowers were the first of a very small number of engineers to be entrusted with the secret of Enigma.
In these first meetings, Flowers and Turing found that they had a certain affinity; certainly it appears that Turing found it easier to talk to Flowers than many others, possibly because of the other man’s unaffected enthusiasm, as well as his expertise. Respect was mutual: in addition to Flowers’s visits to Bletchley Park, Turing made trips down to Dollis Hill, to Flowers’s workshop.
In these initial encounters, one idea was that Flowers should look into the idea of building a machine that could decrypt Enigma by means of electromagnetics. Such an idea was not to prove practical at that time—the technology was not ready. But the principle lodged in Flowers’s head.
But however well he got on with Turing, Flowers found himself attracting the disdain of Gordon Welchman. This first became apparent toward the end of 1941 when circuit expert Charles Wynn-Williams was working on a new decoding machine called Mammoth, one of the forerunners of Colossus. It required electronic valve-sensing units, a task delegated to Flowers. Rather than work to the designs provided by Wynn-Williams, however, Flowers produced his own. These were deemed to be unsatisfactory. The result was friction. Welchman began to refer sarcastically to “Mr. Flowers of Dollis Hill.” He openly showed preference for Wynn-Williams’s expertise over that of Flowers.
This unpleasantness—seemingly over the use of relays as opposed to the more innovative and untested electronic valves—went on for some months, with Welchman leaving Flowers and this Dollis Hill colleagues in the cold, as he instead sought the expertise of the British Tabulating Machine Company and Harold “Doc” Keen, who had been so instrumental in building the first British bombes.
This, it seems, was too much for the pride of the Dollis Hill contingent. At a meeting, Flowers, according to Welchman’s account, declared that Keen should not be “allowed to get away with it.” What he appeared to be referring to was the fact that since those first bombes, BTM had failed to make any steps forward.
The more assertive Flowers grew, the more disdainful Welchman became. In a memo dated June 4, 1943, he wrote:
Dr. Wynn-Williams has found it difficult to get on with the Dollis Hill people, and feels that Mr. Flowers’s idea of cooperation is to run things himself.… It may be that Mr. Flowers honestly thinks that he is better able than Mr. Keen, Dr. Wynn-Williams, and myself to direct the policy of bombe production, but if so, I am quite sure that he is wrong. He is probably very good at his ordinary work, and also very good at designing apparatus for a definite problem that he can understand, but I have found him slow at grasping the complications of our work and his mind seems altogether too inflexible.7
Possibly so. Yet one can also hear a trace, possibly subconscious, of another sort of resistance going on: that of a Cambridge mathematics lecturer to the ideas of a partly self-educated, bumptious East Ender. Out of the senior figures in Bletchley Park, it was Welchman who always seemed most interested in tight control of both research and day-to-day operations; to have this Dollis Hill contingent come in and cast aspersions on Welchman’s own homegrown talent was apparently almost too much for him to bear.
Indeed, Welchman’s ill will toward Flowers grew and grew—he wrote, “the influence of Dr. Radley and Mr. Flowers must be completely removed”—to the extent that he approached the Admiralty supplies department and told them that Flowers was recklessly squandering good valves in the name of his research, and that he and the Dollis Hill team should be excluded altogether from such matters.
The Admiralty chose not to heed him. In due time, Welchman was promoted and set sail for the United States. This might have made things a little easier for the Dollis Hill team. But the remarkable thing, according to historian Paul Gannon, is that the decision to go ahead and develop Colossus properly was taken by the Post Office Research Establishment unilaterally, regardless of the feelings of Bletchley Park. The Park was all set to go with the Heath Robinsons—which worked, despite breakdowns—feeling that the war might be over before a more advanced machine could be put to use; the view of the Dollis Hill contingent was that the war would be going on a little longer than that.
The money, material, and staff for the research into Colossus were therefore provided by the Post Office, though—thanks to wartime bureaucracy, and despite Churchill’s continued bias in favor of the codebreaking work in financial terms—Flowers often made up the financial shortfall for various pieces of equipment himself. The Heath Robinson may have had something of a makeshift genius character about it, but the string and the elastic came from Tommy Flowers’s pocket. Flowers himself said: “It was a feat made possible by the absolute priority they were given to command materials and services and the prodigious efforts of the laboratory staff, many of whom did nothing but work, eat, and sleep for weeks and months on end except for one half day a week…the US also contributed valves and an electric typewriter under the lend-lease.”8
And so this monster, this Colossus, was delivered to Bletchley in January 1944, and with it, many argue, came the dawn of the computer age. For this was more than just a huge, elaborate counting machine; it worked to a program, via electronic valve pulses and delicate, complex circuits, at a rate hitherto unimagined, opening up the Lorenz messages at a terrific rate.
Tommy Flowers was vindicated; the work he did proved utterly invaluable. His nimble engineer’s mind had overcome extraordinary problems. And, of course, he would not be allowed to tell a single living soul.
Adjusting to peace after years of war was extremely difficult for a great many people. It seemed almost cruelly so to Tommy Flowers. It wasn’t just that his boss Gordon Radley was knighted, whereas he simply received that discreet MBE and Award to Inventors. In those days, of course, £1,000 was a very substantial sum, almost enough to cover half the price of a house. But in a way, it was beside the point; for what would the money mean if Dr. Flowers was not permitted to share his extraordinary and innovative electronic knowledge with peers and colleagues?
Worse than this, according to Paul Gannon, was that while a few of Bletchley Park’s decoders could, after the war, decamp to the States and join up with computing projects there, Flowers was stuck in a department where no such transfers would be possible. The Post Office Research Office was extremely respectable and offered the now unimaginable security of a job for life. But it was also excruciatingly limited for a man of Flowers’s talent. And his frustrations grew as the British government’s post-war insistence of keeping every single one of Bletchley’s operations secret further blocked off any advances that he might have made—advances in a budding computer industry to which he had claim to have a greater stake than most.
Not all accept the charge that secrecy held British technological advances back. Harry Fensom concluded a talk given to an Enigma symposium with this thought:
I know some of the Colossi were broken up: we smashed thousands of valves and I believe some panels went with Max Newman to Manchester University. But the know-how remained with a few and the flexibility and modular innovations of Colossus led to the initiation of the British computer industry, such as the work at Manchester and NPL [National Physical Laboratory]. And also of course to the beginning of electronic telephone exchanges. I therefore give my tribute to Dr. Tom Flowers, without whom it would never have happened.
However, Captain Jerry Roberts articulates what he believes that Britain lost, thanks to that insistence on absolute post-war s
ecurity. He is still furious on behalf of Tommy Flowers today: “Dan Brown [author of The Da Vinci Code] wrote a book called Digital Fortress. And he says in that that the computer was invented at Harvard in 1944. That’s the damage that has been done.
“Part of the trouble,” Captain Roberts continues, “was that the Colossus machines were all destroyed, except two which got away. There were ten machines—eight were dismantled and destroyed, and two were kept at Cheltenham at the new GCHQ.
“It was at the orders of Churchill. He didn’t want to reveal anything to the Russians. But this meant—crucially—that Britain couldn’t develop this new computer industry. And I’m sure they could have found some form of cover for the technology—helping building supersonic aircraft or whatever you like to invent.”
When interviewed some years ago, Dr. Flowers himself recalled with some sadness the moment in 1960 when the orders came through to destroy the last two remaining Colossus machines, which had been shipped to GCHQ. “That was a terrible mistake,” said Flowers. “I was instructed to destroy all the records, which I did. I took all the drawings and the plans and all the information about Colossus on paper and put it in the boiler fire. And saw it burn.”
As a postscript to his work, however, there is now, at the Bletchley Park museum, a fully working re-creation of a Colossus machine. It stands, vast and unbelievably complicated, as an enduring testament to an outstandingly brilliant engineer.
25 1944–45: D-Day and the End of the War
The use of Enigma decryption was not confined to intercepted enemy messages, it also played an active role throughout the war in operations designed to deceive. And the most crucial of these was the Pas de Calais gambit of 1944, part of the preparations for D-Day. Indeed, it was reckoned by Bletchley Park veteran and renowned historian Harry Hinsley that, without Bletchley, the D-Day landings might well have been a catastrophic failure and the forces could have been “thrown back into the sea.”
The Secret Lives of Codebreakers Page 27