My Web page about Cryptology
The Louis XIV's codebooks
|
A cipher letter from 1685In Catinat's memoirs, a letter dated 1685 seems to be the oldest example of the use of a two part codebook. Note: the date is not very legible, but taking into account the previous and following dated documents, the date 1685 seems to be the most likely. In any case, it is only the last digit that is suspect (the 5). Here is the letter Here is an exerpt: Si vous jugez à propos de ne pas plus fatiguer de garde la garnison de la ville que n'est celle de la citadelle, le roi approuve que vous le fassiez: mais il régler cela de manière qu'il n'y aipas moins de quate cents soldats fran- çais qui montent la garde à la ville, et les distri- X X X qu o u i buer de manière 401. 421. 218. 1. 6. 60. 57. l y a u ra u n so l da t 90. 124. 202. 69. 195. 134. 18. 70. 114. 356. 41. i ta li e n i l y e n a i 112. 20. 132. 43. 22. 57. 90. 178. 43. 18. 202. 112. t to u s io u r s de u x 123. 84. 177. 16. 308. 60. 5. 59. 105.134. 127. f ra n co i s x x 253. 317. 18. 148. 112. 68. 429. 186., et comme présentement que vous devez être en possession du château, les postes que vous teniez de ce ...English translation: If you deem it advisable not to tire the garrison of the city more than that of the citadel, the king approves of your doing it: but he will regulate it in such a way that there are no less four hundred French soldiers who mount guard in the town, and distribute them in such a way that where there is an Italian soldier there will be two Frenchmen every day as a presentment that you must be in possession of the castle, the posts which you hold of this... It should first be noted that only part of the letter is encrypted. At the time, it was a common practice although reducing security. Bazeries' assertion that Louis XIV's codes and their use were clearly superior to Napoleon's is arguably overstated. Encrypted lines are overcome by their group-to-group translation. It was therefore easy for me to reconstruct the part of the code corresponding to the encrypted groups used. Nuls are represented in the translation by capitalized X letters. The code used
a 202,162 i 57,112 r 5,31 c 231 io 308 ra 195,317,379 ca 179 l 90,114 re 64,75,47,3 Casal 413 la 94 s 16,59,68,168,204 ce 37 le 78,325 sa 25 ci 223 li 265,132 se 160 co 148,272 ma 146 si 8 cu 382 me 13,75 so 70 da 356 M.Mantoue 431 su 225 de 105,120,29,82 mo 263 t 41,123,99 du 105 n 18,22,194,113,173 ta 20,23 e 43,164,111 ne 91,286 te 15,35,62 et 158 ni 328 ti 67,161 f 253 o 6,32 to 84,171 fe 361 p 10 u 60,69,134,177 fi 29 pe 53 uo 71,46,73 fu 374 qu 1 x 107,185,127,76,401, g 206 que 27 y 124,178If we compare this codebook to that of two part from 1676, we immediately see that it has a similar size. The value of the highest numbered group is 431 (M. de Mantoue). On the other hand, the redundancy is stronger. The letters N and S each offer five alternative encryption possibilities (n: 18, 22, 194, 113, 173). Because we have very short ciphertext, it is possible that we also have at least 5 alternatives (instead of 3) for each letter. We also note the presence of the coding of proper names: the city of Cazal and Mr of Mantua. The corresponding codes being respectively 413 and 431, i.e. at the end of the code; suggesting that these proper names were added at the end. If we inspect the letter, we notice that the use of nulls makes it possible to hide the true beginning and the true end of the encrypted parts. References
|