In 1897, the composer Edward Elgar sent a short enciphered letter to his (much younger) lady-friend Dora Penny – because his nickname for her was “Dorabella”, this note has acquired the name “The Dorabella Cipher“. What it actually says is doubtless merely a trifle from one close friend to another (it’s hardly a Zimmerman Telegram, let’s be honest), but its inability to be deciphered has led to its status as an enduring and popular cipher mystery.
Elgar was fascinated by secret writing, even cracking a supposedly ‘uncrackable’ cipher published in Pall Mall Magazine: and part of his enduring fame arises from the way he concealed identities of various friends (as well as a well-known melody in counterpoint) in his famous Enigma Variations (which the German Enigma machine was named after, in homage to Elgar). Musicologists have managed to decrypt most of the secrets of the Enigma Variation: but what of his short enciphered letter?
Curiously, precisely the same cipherbet (‘cipher alphabet’) used here appears elsewhere in Elgar’s notes (which are riddled with cryptograms, puns, and deliberate misspellings). The letter-shapes are formed from a simple 24-letter symmetrical key (a pigpen cipher variant using 3 versions each of 8 rotated E-shapes, almost certainly a visual pun on Edward Elgar’s initials) – exactly the kind of thing a cryptologist would expect of a self-constructed cipher. The problem is that applying this (logical and apparently correct) key to the Dorabella Cipher produces an apparently nonsensical cleartext:-
The mystery of the Dorabella Cipher, then, is neither a whodunnit (because Elgar signed and dated it), nor even a howdunnit (because it seems that we already have the key to the cipher), but more like a “whodunnwhat” – though we can apparently decipher its text, we don’t know what it means, or even how to try to read it.
Currently, perhaps the most persuasive reading is that of Tony Gaffney (A.K.A. “Jean Palmer”), who proposes that the not-so-cleartext (above) was a written version of Elgar’s and Dora Penny’s shared private language, and so could only be read as a tricky combination of backslang, abbreviations, contractions, in-jokes, puns etc. Having said that, Tony’s attempt to reconstruct what it says remains somewhat tortuous: and so the mystery continues.
All of which talk of private language brings to mind the distinction the Annales historian Marc Bloch drew (in his posthumous book “The Historian’s Craft”) between intentional evidence (intended to influence others) and unintentional evidence (intended for an audience of one or less). I think Bloch’s idea was that the reliability of the evidence differed according to the use language was put to; and hence that some famous ciphertexts probably remain uncracked because the text they contain is unintentional evidence, too personalized a shorthand for anyone but an audience of one to read.
Will the Dorabella Cipher ever be cracked? Right now, I’d say probably no, simply for the reason that it is too short a text to do significant statistical analysis on, if (as I think likely) the plaintext was written in Elgar’s and Dora Penny’s shared private language. And that’s the difficult challenge I believe posed by many cipher mysteries: while enciphered intentional evidence can be too trivial, enciphered unintentional evidence can be simply too hard… not unlike trying to achieve triple-jump distance with a single leap.
Dorabella Cipher Links:-
- A page from the 2007 BBC Proms website on the Dorabella Cipher
- Kingston University musicologist Kevin Jones’ view of the cipher
- Cipher Mysteries: post discussing Eric Sams’ “STARTS: LARKS!” decryption
- Cipher Mysteries: post discussing Kevin Jones’ page and various other pages
- Cipher Mysteries: post discussing Tony Gaffney’s Dorabella decryption
- Cipher Mysteries: post discussing Tim Roberts’ (2009) claimed decryption
There is also much more of interest in the moderated Yahoo group Elgar-Cipher.