Tuesday, 2 July 2002

Enigma: the battle for the code

Listening to:

Mozart, string quintet in B flat major, KV174. Mozart wrote six string quintets and they are all very beautiful.

Just read:

Hugh Sebag-Montefiore, Enigma: the battle for the code.

This book went over pretty familiar territory for me in many ways. Alan Turing, U-boats, plug-boards, yadda yadda. But it told the story well, and also included interesting material that I hadn't seen before. For example, it devotes a lot of attention to efforts by the Allies to get hold of Enigma material by capturing U-boats and various other German naval vessels. The book's introduction gives the impression that this was going to be the principal focus of the book when it was first conceived, but there is a lot of other material too.

For example, the account is concerned to track the French spies and the Polish mathematicians that made the first steps in breaking Enigma. Their story starts in the 1930s and continues through to the war years too. Incredibly, the Polish cryptographers, who could have told the Germans all about the way in which their supposedly secure system was broken, spent much of the war in a French château. The Nazis did manage to capture one of the Poles, and also the German traitor who sold Enigma information to the French, but there are none so blind that will not see, and the secret stayed safe.

There are interesting sections devoted to the German responses. It does seem as if Dönitz, chief U-boat commander had his concerns about cipher security, but when he took these to the Naval Communications people, they kept telling him there couldn't be a problem. It's also very interesting to read that the Germans were reading some of the Allied codes. I’d really like to read more about this German cryptanalysis effort.

The book puts most of the cryptography into 6 appendices. I skimmed these, feeling that it wasn’t really important to understand all of the details. I feel there are no particularly deep principles there; rather there are complicated fiddles that are tied to the details of the Enigma implementation.

My final moral of the story: don’t put weather stations onto the same cryptographic network as your warships. Have them use a different code, that the warships are all able to read, but which they don't use themselves.

To read next:

Granta 74: confessions of a middle-aged Ecstasy eater.