Variable-Length Figure Cipher of Duke of Lorraine (ca.1620?)

A cipher used in a letter to the Duke of Lorraine turned out to employ variable-length symbols.

The letter (BnF Lorraine 377, f.95) signed "Breval" is catalogued as "Lettre chiffrée de M. de Bréval au duc Henri II (Wikipedia)." Breval was a "chargé d'affaires de Lorraine à Paris" (BnF).

The reference to "gouvernement de normandye" in clear after "recompense monsieur de Longueville" suggests the letter is dated from around 1619, when the Duke of Longueville obtained governorship of Normandy in exchange for that of Picardy (Wikipedia). I could not identify monsieur/sieur de Couuongs mentioned twice.

The interlined decipherment allows reconstruction of the cipher as follows.

The cipher employs variable-length symbols, i.e., Arabic figures of one, two, or three digits. Since the figures are written without breaks, the decipherer needs to parse the stream of digits into proper cipher symbols. Thus, the digits "1" and "7" are not used as a single-digit symbol to indicate the beginning of a three- or two-digit symbol. Such variable-length symbols were used in Vatican ciphers in the sixteenth century (see another article) and also in the 1620s (see another article). In contrast, no cipher of this type is known to me among contemporary French ciphers (see another article).

It is yet to be found out how the Vatican-style cipher came to be used in the Duke of Lorraine's correspondence.

©2024 S.Tomokiyo
First posted on 23 March 2024. Last modified on 23 March 2024.
Articles on Historical Cryptography
inserted by FC2 system