Bletchley Park

Our visit to Bletchley Park on Sunday was really good, despite my poor state of health. I had booked and paid for it six months ago, long before the exam timetable was confirmed, so it was rather disappointing to find that my first exam was scheduled for 9 a.m. the next day. Having decided that we would go, it would be OK and "if I haven't learned it by then it will be too late", we went, it wasn't, I hadn't, and I felt like living death.

Unusually for May, it was a really hot day. The schedule started with an hour's lecture, then there was lunch, a tour, and another lecture. Run by the Cambridge University Alumni Association, the lectures were delivered by eminent and knowledgeable people, and the intellectual level was pitched pretty high - this was not tabloid stuff. The rooms where we sat were hot and full to the brim, but it was utterly fascinating.

The first lecture we heard was a fairly lighthearted affair, though. There can't be anyone more qualified to talk about the work of Bletchley Park and MI5 than Professor Christopher Andrew, but he was having fun with an audience comprising Cambridge alumni by making pointed gags at the expense of Oxford, which provided spies only to one side in the war, rather than Cambridge's more even-handed supply to both sides.

Professor Andrew has mixed with the highest and mightiest, including Prime Ministers and Russian defectors, and holds a number of very lofty posts himself. But he celebrated eccentricity, citing the downfall of City finance organisations as evidence that recruiting team players leads to mediocrity and the suppression rather than the expression of excellence.

The tour wasn't very long and it is a huge site, so we actually saw very little. We were taken to see the Colossus machine, which has been rebuilt over the course of a decade. This is a considerable achievement since all hardware, plans, blueprints and circuit diagrams were ordered to be destroyed when the extant machines were decommissioned. The reconstruction team worked from a few illegally retained diagrams and six black and white photos.

Colossus wasn't involved in breaking the famous 'Enigma' codes, but instead it dealt with 'Lorenz' codes, which were used between Hitler and the German High Command. The ability to break the codes hinged on just one mistake, a chance in a million, when the same message was sent (and intercepted) twice using the same key settings. Without this, there would have been little chance of breaking the code.

You may recall that I was trying to read a book about the work of Bletchley Park some time ago, but it was much too dry to finish. One of the things I did pick up from the book was that many WRNS were recruited to do the donkey work, and spent days, weeks, months and years doing the most tedious chores without being able to tell anyone how vital it was to the Intelligence services. In the reconstructed telegraph intercept station I saw what this involved: the intercepted messages were output in ink onto long ribbons, and the WRNS had to convert these to punched tape. The ink trace was hardly a centimetre wide; it would have been mind-numbing work if that was all you did for eight hours a day , and they were allowed no more than 6 mistakes per 1000 characters.

The second lecture was brilliant too. Frank Carter works for the Bletchley Park Trust, and he actually described some of the theory thought up by Alan Turing and Gordon Welchman that was used to decode German messages encrypted using the Enigma machine. He had a dummy box for demonstration purpose, with a keyboard for input and a row of 26 LEDs for output.

By concentrating hard, Mr A and I were both with him as he went through the theory of taking a 'crib', which was a first guess at the meaning of a line or two of text, and turning it into a 'map' showing how the Enigma machine might be encoding it. At this point, the decryption process would show that the guess was correct only by pressing the right letter out of 26 on the keyboard to complete the circuit and lighting one lamp - no other keypresses were useful.

Turing's genius was to focus on 'wrong' circuits rather than the right one, so that 25 out of 26 keypresses provided a good deal of evidence towards which was the 1 out of 26 that was correct. We were nearly lost at the last corner, when we heard how Welchman came up with the 'diagonal board' which made the answer almost definitive with any single keypress. Using this theory, decryption machines called 'Bombes' were designed that ran through the war, and they've reconstructed one of these as well. By the time we'd been to see this in operation, I wasn't fit for any more, and we went home. I'd love to go back, perhaps over the summer. It's only just over an hour's drive away.

So how was the exam? I'd been sleeping really badly because of the heat and having a blocked nose and sinus pain and coughing and the rest, so when I woke up at 5 a.m. on Monday I thought I might as well drive to the campus. I was planning to go stupidly early anyway, because any blockage on the motorways would be a serious problem. So I did a couple more hours revision first thing, and it was fine. I dropped in on my tutor just to make sure that if I got any more ill he would support an application for Extenuating Circumstances. Luckily, that was the worst it got, and I've been improving over the last two days. Next exam is tomorrow, and I've spent quite enough time on writing this post.