by Sue Black
I also had another email from Viv Parry, who had been so helpful earlier in sending me names and contact details for several journalists. This time, after congratulating me on getting great media coverage for the story, she suggested that I get in touch with Simon Singh, the well-known mathematician and author, who might be interested in what I was doing. I found Simon’s contact email on his website and sent him a message.
Simon very kindly got back to me the next day, congratulating me on the media coverage and offering to help. He included his home phone number and suggested that I give him a ring the next day to discuss things further. We had a great chat about how important Bletchley Park was. We were both in agreement that it needed to be preserved for the nation, and we spoke about the best way to make this happen. I told him about my ideas around the joint venture between the Science Museum and the Imperial War Museum and my conversation with Sir X. Simon gave me some campaigning advice and lots of encouragement and told me to get back in touch if he could help in the future.
Saving Bletchley Park
From: Dr Sue Black
To: Simon Singh
Sent: 29 July 2008
Dear Simon,
You may have seen our letter to the Times last week and some of the resulting publicity in the press and on TV and radio.
As you are a supporter of Bletchley Park and, I have heard, as passionate as I am about the place :-) I am writing to ask for your help in our (currently at an early stage) campaign to save Bletchley Park. Anything that you can do towards the cause would be extremely welcome.
Best regards
Sue
Meanwhile, I had been approached by Maggie Holland from IT Pro as she wanted to write an article about Bletchley. Maggie had written several articles about women in computing so we had talked previously, particularly as she had covered our BCSWomen Bletchley Park oral history project. This time Maggie wrote the great piece entitled “SOS Bletchley Park” – a rallying call for support, which asked people to sign the petition and help raise awareness of its financial state and also highlighted the contribution of the women that had worked there. I had previously sent Maggie some of the audio files from the BCSWomen oral history project and she had obviously listened to them. There was a nice mention of my colleague John Turner as well; after all, if he hadn’t had the idea to write a letter to The Times . . . would any of this have happened?
Maggie’s article was accompanied by the photo of Hut 6 that I had taken at the reception on July 1st. Reading the article I couldn’t believe that it had been only four weeks previously that I had stood there in front of Hut 6 with a tear in my eye, listening to an account of all the amazing achievements that happened there during World War II. So much had happened since then.
SOS Bletchley Park
Maggie Holland
ITPRO
29 July 2008
Is the hospital/house you were born in still standing? How would you feel if it was there but not there – a decrepit shadow of its former self? Not good, one would imagine. Yet that’s exactly what is happening to the birthplace of computing and the place that played home to the code breakers during World War II.
Bletchley Park is an absolute goldmine of historical significance. It is incredibly exciting – even nail biting – to think of what went on within its walls during the war, and it is attracting visitors from around the world. The gems on offer to those visiting are akin to those found in many of Britain’s great museums. Indeed, the site now serves as home to the National Museum of Computing.
But, unlike the majority of museums, Bletchley Park receives no government funding. As a result, there is a real danger that this piece of history could be consigned to just that: history.
06
Enigma
“Hut 6 would never have gotten off the ground if we had not learned from the Poles, in the nick of time, the details both of the German military version of the commercial Enigma machine, and of the operating procedures that were in use.”
—Gordon Welchman
The Enigma machine was a fiendishly clever device. It encrypted messages by substituting one letter for another, as most ciphers do. However, the machine used a series of rotors to run a series of substitutions for every letter it coded. Therefore, unless you knew the starting positions for each rotor, the possible combination of substituted letters ran into the hundreds of thousands. But that was just the start; Enigma was far more complicated than even that.
Depressing a letter on the keyboard – let’s say the letter “P” – sent an electrical signal to Rotor I. The signal then passed from one side of the rotor to the other, but the two sides wired each input letter to a different output letter. So the input letter “P” may have been wired to the output letter “G”, for example. The output letter “G” then passed to Rotor II where it was output as, let’s say, an “S”. The “S” passed to Rotor III which output it as a “Q”. After leaving Rotor III, the signal would hit a component called the Reflector which changed the letter again – a letter “H” for these purposes – and then the character would travel back through the three rotors. However, the return journey would be different to the outward journey because the rotors, as their name suggests, rotated. Rotor I was called the fast rotor and would advance by one position after each key depression. Consequently, if you pressed a single key – our “P” for example – several times, each keystroke would generate a different output letter. Rotor II (the middle rotor) also rotated one position after the fast rotor had completed a full circuit of 26 positions. In addition, a sliding ring bearing the letters of the alphabet could be set at any of the 26 positions on each rotor. If the letter “A” on Rotor I was turned into an “R” by Rotor II, then re-setting the ring by one position on Rotor I would mean that “B” would code to an “R”. You could also decide which letter position would trip Rotor II and set the position on Rotor II that would trip Rotor III after Rotor II had completed a circuit. Finally, the finished cipher signal would light up one of 26 letters on the face of the machine, displaying the letter that your initial letter “P” had now been coded into.
Input/Output
Rotor I
Rotor II
Rotor III
Reflector
P
P to G
G to S
S to Q
Q
R
Y to R
B to Y
H to B
H
But that’s still not the full process. Each machine came with three rotors, all wired differently. Plus, each rotor could be placed in any of the three slots in the machine. So apart from needing to know the starting position of the three rotors, you also had to know what slots the three rotors were in, as there were 60 possible combinations (the Naval Enigma added two more rotors to the mix so that you then had to work out which three of five were in use and their positions). In 1930, a plugboard was also added to the military machines whereby any two letters of the alphabet could be randomly wired together in pairs causing them to be reversed. This added another layer of complication to encryption.[11]
This was why Enigma was so efficient as a coding device; every letter in a message was coded using a different cipher to every other letter. To decode a message, you needed to know which three rotors were being used and what positions they occupied (Walzenlage). You also needed to know the initial starting position for each rotor (Grundstellung), the ring settings that dictated which letter was being coded into which (Ringstellung), the trip position that would trigger rotation of the next rotor, and the chosen paired letters on the plugboard (Steckerverbindungen).
The total number of possible ways in which a standard Enigma machine with plugboard could be set up was 60 Walzenlage x 17,576 Grundstellung x 676 Ringstellung x 150,738,274,937,250 Steckerverbindungen. That’s around 158 million million million possible ways to set up a machine. Therefore it was believed, with good reason, that no one would ever be able to crack the Enigma code system. As Tony Sale, the late curator of Bletchley Park, once said: “Unless you had the exact key, you couldn’t get anywhere with it all. This was the major difference from any of the code systems that had existed prior to that. There was no sense of nearness; you were never nearly at a solution. You’d either got the solution or you hadn’t got the solution.” Codebreaker Howard Smith recalls that a colleague described breaking Enigma as “the frenetic equivalent of blindfold three-dimensional chess”.[12]
But someone did break Enigma.
In 1939, shortly before the invasion of Poland, a group of Polish cryptographers revealed to the GC&CS that they had been working on military Enigma machines since 1932 and, after receiving documents from a disgruntled German clerk and from French Military Intelligence, they had worked out the scrambler wiring on the machine’s rotors. This was very useful information indeed. Dilly Knox was already familiar with the commercial Enigma machine but this didn’t have the plugboard, nor were the rotors wired in the same way. But Polish mathematician Marian Rejewski had figured out the differences and, what’s more, had built a replica machine that he called an Enigma Double to test encryptions. He had also built an electromechanical machine called the Bomba Kryptologiczna (Cryptological Bomb) that used a series of six connected Enigma rotors to speed up the working out of possible combinations.
Poland, sandwiched between Russia and Germany, was ripe for invasion and so the Poles agreed to hand over everything they had to the Allies, including all of their research, an Enigma Double and a Bomba. Sadly, changes in the German operator instructions, and the addition of two extra rotor choices, made much of this work obsolete very quickly. But as Peter Calvocoressi, head of the Luftwaffe section in Hut 3, has remarked, “According to the best qualified judges it accelerated the breaking of Enigma by perhaps a year. The British did not adopt Polish techniques but they were enlightened by them.”
Initially, attempts to break into Enigma were made using Zygalski Sheets, named after Henryk Zygalski, a Polish mathematician who worked as a cryptographer with Marian Rejewski and Jerzy Różycki. Zygalski had noticed that Enigma had a curious quirk; occasionally – about one in eight depressions of the same key – it coded a letter into the same letter twice. So rather than a key depression being a random 1/26 chance, there were often repetitions. He called these repetitions samiczki (females), and there were only a relatively small number of Enigma machine configurations that would generate females. This was good news for the codebreakers. Tony Sale again: “Repetitions give codebreakers a toehold. Repetitions are always bad news in cryptography.” Anything non-random meant that patterns could be detected, and if enough females occurred during a day’s message traffic, it might be possible to find the unique configuration from which all of these doubles could be generated.
Zygalski Sheets[13] were sets of 26 large cards, each with a vertical and horizontal axis bearing the letters of the alphabet, rather like a map’s grid reference system. The 26 × 26 matrix represented the 676 possible starting positions of the middle and left (slow) rotors. Each sheet was different, with holes meticulously cut out with razor blades, to represent every rotor configuration that could possibly produce females. By laying these sheets on top of each other, working through each of the six possible rotor orders (1-2-3, 1-3-2, 2-3-1, 2-1-3, 3-1-2, 3-2-1) and then through the 26 possible left (slow) ring letters, you had 156 chances of finding the Enigma configuration that generated the females you’d spotted in a message.
“When the sheets were superimposed and moved in the proper sequence and the proper manner with respect to each other, the number of visible apertures gradually decreased,” wrote Zygalski. “And, if a sufficient quantity of data was available, there finally remained a single aperture, probably corresponding to the solution. From the position of the aperture one could calculate the order of the rotors, the setting of their rings, and, by comparing the letters of the cipher keys with the letters in the machine, likewise the entire cipher key.”[14] “It was like solving a very difficult crossword puzzle”, says Pam Hemsted who worked with codebreaker John Jeffries on the sheets. “But you could actually see it happening. And the triumph when you found it worked! That was fascinating, marvellous. There’s really nothing like seeing a code breaking. That is really absolutely the tops.” And Leslie Yoxall, a cryptographer from the Naval and Japanese sections, recalls: “People were very reluctant to go home at the end of the shift. They wanted to hang in there.”[15]
The Zygalski method worked wonderfully until the Germans changed their operating procedures. For a while, the Allies were back to square one and it seemed that the only way they were likely to get back into Enigma was by capturing enemy code books. At the Admiralty, all kinds of audacious plots were hatched to achieve this goal. Personal Assistant to the Director of Naval Intelligence, Commander Ian Fleming – later to achieve huge success as the author of Chitty Chitty Bang Bang and the James Bond books – came up with the idea of crash-landing a captured German plane into the sea and then overpowering the crew of the ship or U-boat that answered its distress call. The appropriately named Operation Ruthless was never implemented, however.
But then, on a cold evening in February 1940, a young codebreaker and colleague of Dilly Knox called John Herivel had a sudden, brilliant flash of inspiration. In Sinclair McKay’s The Secret life of Bletchley Park, Herivel describes that moment to the author: “Something very strange happened; I may have dozed off before the fire – a dangerous thing to do as I often smoked a pipe and might have burnt a hole in my landlady’s carpet, or worse – and perhaps I woke up with a start and the faint trace of a vanishing dream in my head. Whatever it was, I was left with a distinct picture – imagined of course – in my mind’s eye, of a German Enigma operator. I seem to have taken Aristotle’s advice, that you cannot really understand anything thoroughly unless you see it growing from the beginning.”
By putting himself into the mind of one of the German operators, Herivel imagined the long hours, the fear, the working conditions – maybe not so different from the pressure of work and conditions at BP – and realised that maybe the operators might not be as on the ball as they could be. What if tiredness, or distractedness or even laziness made them do something silly, like not resetting the ring settings on the Enigma at the start of a new day? Acting on his hunch, Herivel worked out exactly how such a mistake could be spotted. What became known as “The Herivel Tip” or “herivelismus” proved to be the next great breakthrough.
Meanwhile, other mistakes had been noticed by Dilly Knox and Gordon Welchman. They christened them “Dilly’s Sillies” or “Cillis” after finding what seemed to be a shortened version of a woman’s name – CIL – as an indicator for the day’s codes. Similar slipshod or lazy operator practice included snippets of poems, proper names, common and non-random typed letter sequences like ABC or QWE, and even swear words.
It’s a chilling thing to consider, but had John Herivel not had his moment of brilliant intuition, and had the German operators displayed a little more of the efficiency that dogs their national stereotype, Enigma would not have been broken for many more months and the course of the war may have gone a very different way.
07
Bletchley Park heroes
In the days after the letter to The Times was published, things started to get back to normal, though there were still a few emails per day coming in from people all over the world who were interested in Bletchley Park. I did a few more interviews with radio stations overseas, including one with an Australian station, ABC Radio Canberra, on the Ross Solly show. I was finding that the non-UK audiences seemed t
o find it incomprehensible that Bletchley Park wasn’t government-funded. I agreed with them. I had used the phrase “ashamed to be British” whilst talking about Bletchley Park and its situation, and though this had now become a phrase my family used regularly to take the piss out of me, I still felt that on some level it was true: I was ashamed that the government hadn’t stepped up to save this important historical site from ruin.
When I went back into the office that week and checked my voicemail, I found that I had a message from someone called Captain Jerry Roberts, asking me to get in touch with him. I immediately looked him up online. He was a Bletchley Park veteran and sounded very distinguished. How exciting! I gave him a ring and we chatted for some time about how the Park needed funding. We were like minds. Jerry had been a linguist and senior codebreaker during the war, and he had served as a shift leader in the Testery, one of the vital components of the code breaking operation at Bletchley. Jerry concluded the conversation by very kindly inviting me over for lunch the next week at his flat in Pimlico.
Lunch with Jerry and his wife Mei was wonderful. They were both such lovely people and obviously keen to tell the world about what had happened at Bletchley and how important it was. Jerry told me a bit about his work at Bletchley, about decoding messages from Hitler himself and about working for months in a room with Bill Tutte, the codebreaker and mathematician who made a crucial advance in cracking the Lorenz cipher, Tunny; Jerry said he’d thought Tutte was not “earning his corn” until his major breakthrough with Tunny, when of course it became apparent that he’d been hard at work.
Captain Jerry Roberts Jerry had so many great stories. I asked him if he had written down all of his experiences and said how great it would be if he published an autobiography. Jerry said that he wanted to, but that due to his age he was finding it difficult. He was nearly 90 years old, and although he was extremely sharp mentally, his body was starting to let him down. His stamina was not what it had been, and he had to have a nap each afternoon.