Dorabella Cipher
Elgar's unsolved 1897 cipher note, investigated: own transcription, live statistics, every attack with its verdict – and a toolkit to try your own.
In 1939 Alexander d'Agapeyeff ended his book Codes and Ciphers with a challenge: 395 digits that nobody has read since. He is reported to have admitted later that he had forgotten how he enciphered them. This investigation takes the cipher apart one measurement at a time – every number is computed live in your browser or quoted from the solver behind it, and nothing is asserted without a control.
The digits turn out to be 196 two-digit coordinates of a 5 × 5 square. Their repeat profile, which no substitution can change, shows that a transposition happened and that every unit moved on its own. Then every transposition in reach was tried: keyed columnar at every width, digit-level keys by exhaustion, 614 unkeyed routes, nulls, real words and dates as keys, the whole thing run backwards and double transposition in all four directions – 222 million keys, none scoring above −1.32 where real English scores −0.74. What remains is what the published record already suspected – d'Agapeyeff made a mistake – and a 2017 finding, verified here against his own printed example, shows how: he did not understand which direction his transposition ran.
Eleven sections cover the cryptogram, the measurements, the evidence, every attack with its control and raw output, what went wrong, the likely cipher type, the published record, what to do next, how it could still be cracked and the passages of the book that bear on it. The workbench puts five tools on one shared working text; its two solvers run for as long as you let them.
No comments yet.
Log in to write a comment or reply. A moderator checks every comment before it appears – for security and to keep spam and abuse out.