Chapter IX: Readable ciphers
For my 10th birthday my father sent me a delightful book on cryptography called Codes and Ciphers -- A History of Cryptography by Alexander D'Agapeyeff. Only 144 pages long, I devoured it avidly. Happily the book is still in print.
I decided I needed a code system of my own to put down things I didn't want others to read and which I could use for the rest of my life. First, I wanted it to be "readable" meaning that I could look at the coded text and translate it in my head at a fair speed without the need for pencil and paper -- even many years later. Second, it needed to be "redundant" meaning that there would be several ways to encrypt common letters like E and T. This would protect it from frequency analysis. I succeeded in my task and have been using my code ever since. My code used exactly eight symbols. Now, for example, if you use eight symbols in pairs you get 64 different pairs. Then, if you use wish to code for just 26 letters and some punctuation there is plenty of redundancy. Designing it so you can easily read it without using pencil and paper requires some ingenuity. I am not going to tell you how I did it -- that is still my secret. |
Codes and Ciphers
|
When I told my father and his friend logician Georg Kreisel about it they confidently said that if I gave them a page of coded text they would crack it. Wittgenstein once declared Kreisel to be "the most able philosopher he had ever met who was also a mathematician." I figured that if the two of them couldn't break it, it would be the perfect test of my code's security.
What they didn't know was that I understood redundancy and had designed a code to withstand frequency analysis. The two of them wracked their brains for two days before they gave up and asked me how it worked. When I told them it was secret they were very chagrined. Too bad, it might have helped my father in his future work. |
Georg Kreisel
|