Two cipher letters: one read, one half read
A human-led research project by Thomas Leeming, with Claude Opus 5.5.
In October 2026 we worked on two letters written in cipher for which we found no published reading. We read one in full and about half of the other. Thomas led the project, chose the letters, reviewed the work and approved its publication. Claude, an AI model made by Anthropic, did the transcription runs, the analysis and the drafting. This site has both readings and a full account of how we worked.
The Lisbon letter, December 1558
The French ambassador in Lisbon, Michel de Seure, put twenty lines of a letter to the secretary of state in a cipher of drawn signs. We found no decipherment of them beside the letter or in print. We broke the cipher in its main lines, using passages that a clerk of the period had deciphered, tested the key on text it had not been built from, and read about half of the passage.
I do not wish to stay here much longer … at worst I want only to finish my two years here.
The Amsterdam letter
A sheet of 35 lines of numbers, addressed to a man in Amsterdam and kept in the Swedish National Archives, is listed in the DECODE database as non-decrypted. It is a short political letter in Dutch. The key is the alphabet in order, starting at 30, and one misread digit had hidden it.
I rest neither night nor day, whether I am sick or not, and shun no danger.
How we did it
The full account: the rules we set, the false start, how each cipher gave way, what went wrong, and who did what.
What to know before you rely on this
- The transcriptions and readings were made by AI models and reviewed by Thomas. No specialist in early modern Dutch or in French hands of the 1550s has checked them yet.
- Every doubtful place is marked on the letter pages. About half of the Lisbon passage is unread, and the people in the Amsterdam letter are not identified.
- We found no earlier reading of either letter in about a day of searching. For the Lisbon letter, four works were not available to us; one of them, an article of 1969, is reported to print despatches from the same manuscript.
- Corrections are welcome: write to rightaboutnothing@gmail.com with CORRECTIONS in the subject line.
Credit
The Amsterdam letter is held by Riksarkivet, the Swedish National Archives, and its images were published by the DECODE database. The Lisbon letters are held by the Bibliothèque nationale de France (source gallica.bnf.fr / BnF). D. Bourdeau's catalogue of unsolved ciphers pointed to both letters, and Bourdeau's transcription was the starting point for the Amsterdam one.