Dorabella Cipher

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?

dorabella-cipher-image

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:-

BLTACEIARWUNISNFNNELLHSYWYDUO
INIEYARQATNNTEDMINUNEHOMSYRRYUO
TOEHO’TSHGDOTNEHMOSALDOEADYA

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:-

There is also much more of interest in the moderated Yahoo group Elgar-Cipher.

101 thoughts on “Dorabella Cipher

  1. mark stahley on November 10, 2012 at 4:33 pm said:

    I think the symbols are sounds not letters. It’s difficult for me 2012 American, to know the exact pronunciation but I believe the first line starts ” Am I your Enigma?”
    and to make it harder, Enigma is spelled out in a musical way
    Dora stuttered so possibly that is how she said it without stuttering(?)
    am eI uer Iy-E,N-e, aA-e, ge-e, em-me,a-I ?
    Am I your enigma? in p-
    riory Y B I,” E E” eh. A Rene is PA Y Y in P-
    apa’s P-As? PS a P-A-I- M> Pa I? oui
    PA is Public Assylum where he gave weekly concerts before he was famous.

  2. Narissa Andrews on November 29, 2012 at 1:22 pm said:

    This has been solved by that Australian guy Tim Roberts and it makes total sense and Id accept it as the correct solution.

  3. Narissa: I wish it were so, but strongly suspect that it’s not – http://ciphermysteries.com/2009/09/18/dorabella-latest-news

  4. Narissa Andrews on November 29, 2012 at 1:57 pm said:

    Ill hold it as the best solution upto yet but ive got a uk composer looking at it at the moment with fresh eyes so see what he thinks.

  5. I think it’s easy- they aren’t sounds, or even letters. They are 19th century icons… Each symbol represents one of them and/or an action: I met you, you turned away, you came back, we fell in love, it was wrong, we parted ways.

  6. Pati: Any evidence for this apart from it sounding right to you?

  7. Hello, Im currently working on a solution which is based on secret cryptic system I had as a kid. Back in 1897, the old alphabet was used, which only used 24 characters (J & W are not included). That suggests three sets of 8, which correlates with the 1, 2 and 3 squiggle symbols. Each of these than point in a certain direction, North, South, East, West, NE, NW, SE and SW. If we organize the letters in three squares of 3 by 3 leaving the center square empty. Depending on the number of squiggles we select the square, and according to the direction we choose the letter. The Question that remains, is in what order were the letters arranged in the squares?

  8. Andrej: what you’re describing is very similar to what is known as “pigpen cipher” or “masonic cipher”, and people have pointed out the similarity between this and the Dorabella Cipher’s alphabet many times before. Having said that, for me the biggest mystery about the Dorabella Cipher is that there’s no obvious reason it should have been non-trivial – as I recall, in July 1897 Elgar had only just met Dorabella, so why would he send her enciphered in a ridiculously arcane way? Similarly, the two had only just got to know each other, so all the so-called “decrypts” I’ve seen containing bizarre phrases in some alleged ‘private language’ make no sense to me.
    http://ciphermysteries.com/2009/03/19/is-dorabella-a-rotating-pigpen
    http://ciphermysteries.com/2008/01/15/pigpen-cipher-gravestone

    In my opinion, the answer is likely to be simple, playful, and obvious once you see it: really, we’re probably all looking straight through it!

  9. Soroush Hoseinpour on February 13, 2013 at 11:35 am said:

    He has sent a letter to his love and why it shouldn’t be a poem?
    I/LOVE/LOVE/YOU/I/LOVE/LOVE/YOU/MY/LOVE/MY/LOVE/MY/LOVE/I/WANT/TO/BE/WITH/YOU/I/WANT/TO/BE/WITH/YOU/I/WANT/I/WANT/YOU/
    Underlined words (love&love) in the original letter are same.
    These two (T)s are same too.

    what do you think?

  10. Soroush: the letter that the Dorabella Cipher was folded into was from Alice Elgar (Edward Elgar’s mother) to the Penny family, but I’ve never seen a copy of it, let alone seen any underlined words in it. Do you have a copy or a scan of this that I can see? Otherwise I can’t really make any comment on what you propose, sorry. 🙁

  11. Thomas Upton on February 26, 2013 at 5:46 pm said:

    This is very close to J S Bach’s shorthand for musical notation. Elgar uses curves where Bach uses zigzags. It is a code, in the sense that longer work is hidden, but this is a quicker way than staff and notes. Look up “shorthand for musical notation.

  12. Tom: do you mean basso continuo?

  13. Decrypter X on March 6, 2013 at 5:44 am said:

    Guys: I think I just got it (or something similar) the first 3 “squigglies” mean “why”. Not saying anything else right now SUPER PUMPED!!!!!

    Signed,
    Decoder X

  14. Decrypter X on March 6, 2013 at 6:01 am said:

    Wait nevermind. My “solution has multiple outcomes. Sorry for false alarm. But I did find a way to write like this that I believe is almost virtually impossible to decode( unless you know how, of course). Not sure yet if I should release it. Let me know.

    PS when I write a cipher, I sign it “Encoder X” and my buddy who helps me “Encoder Y”.

    Pps when we decode ciphers, we sign them “Decoder X” and Decoder Y.

    Signed,
    Decoder X

  15. Cat Darensbourg on July 14, 2013 at 6:44 am said:

    Thank you for your article pointing out the cipher was pig-pen based, most likely. I did some sketches for how the alphabet might have been constructed and posted it on the “Ancient Cryptography” website if you would like a look. (My other, earlier theory that it might be a pig-pen, then a rail-fence, i have set aside for now, and the other cipher stuff is merely discussion at this point.)

  16. The last 11 letters only have 7 different symbols.
    Say you give each symbol a letter, starting at the NE Triple symbol, we have ABCDBAECBFC

    I dont think thats ever going to spell a real word? A tune maybe?

  17. Narissa: “not long to it”? 😉

  18. Yeah that works 🙂 back to the drawing board.

  19. Allan G. on August 25, 2013 at 7:29 pm said:

    Posted this earlier but under wrong heading. Should be this one – The Dorabella Cipher.

    My decrypt is (with my punct.):

    Forli, Malvern.

    Bon ? (16th char is unresolved),

    A and Dai (=Alice and I) opin’ (=opine)
    met St Swithin eighty six.; wed at Brompton Oratory but owed takc …….. (final 8 chars dodgy due to ambiguous cipher chars)

  20. Allan G. on September 25, 2013 at 3:24 pm said:

    Revised decrypt (chars 1-39) now reads:

    ForlE [=pronounce as Forlee], Malvern Link.

    A. and Dai’s [=Alice and I’s] qk [=quick] o [?opinion?] –
    met St Stephen eighty six etc as above

  21. Allan G. on September 25, 2013 at 9:40 pm said:

    Final 8 chars (after ‘but owed takc’):

    Mogul ob’d[=obliged]

    Full decrypt

    Forle, Malvern Link.

    A and Dai’s qk o. Met St Stephen (e)ighty six. Wed at Brompton Oratory but owed takc. Mogul ob’d. (87 chars)

  22. Šuruppag on November 20, 2013 at 2:47 am said:

    After 30 years he had probably forgotten which signs he had assigned to each letter and so he simply reassigned them differently.

    Considering we found a version of it that is a simple substitution cipher in his notes, and he thought Dora would be able to solve it, I really doubt there is a complex stystem at work.

    So why is it that expert cryptographers cant easily solve it? I guess its brevity just really makes it hard to analyze. Too bad he didn’t write her a sappy wall of text.

  23. Šuruppag: actually, most simple substitution ciphers can be cracked with a 30-letter ciphertext (or even less with a couple of really good guesses), so the Dorabella Cipher should be more than long enough to be cracked. So, the mystery is why it can’t be cracked despite its length, not because of it. 🙂

  24. Allan G. on November 29, 2013 at 8:02 am said:

    Re my decrypt of 25 Sept.
    In a letter to his friend Kilburn, dated Aug 3 1897 (3 weeks after Dorabella), Elgar signed off in his usual way then, unusually, added the following:
    “Somebody has given me a ‘fountain pen’ & it’s not a complete joy – yet.”
    Suggest chars 24-27 (recovered as SQKO) = (abbr.) squeako (the sound of his new pen).
    Oh – and chars 78-79 prob = BS (giving TABS, not TAKC).
    Hey ho.

  25. I keep seeing people stating that Elgar’s signature is at the bottom of the Dorabella Cipher, yet if you compare it to his actual signature the two are nothing alike. I think it is actually a clue to decipher the text. I’m gonna play around with it a bit 🙂

  26. , Rick A. Roberts on February 22, 2014 at 6:14 am said:

    I believe that I have solved the Dorabella Cipher using Elgar’s Cipher Code mentioned in the article. The message reads,” BE IT CLEAR NOW YOU FINALLY KNOW WHY ON WEDNESDAY IN OUR QUAINT IN HOME SERMONS YOU DON’T SHOOT THE GOOSE THAT LAYS A GOLDEN EGG.”
    Edward Elgar was very witty and liked to use puns or play on words. He referred many times about, “Don’t Shoot the Goose That Lays A Golden Egg”. This I believe was mentioned in the, “Unitariun Register” also.
    A man and a woman found a special goose that laid golden eggs. They thought that they would become very rich. However, by cutting the goose, they thought that they could get all of the gold coins inside the goose at once. When they cut open the goose, there were no more golden eggs.
    Elgar was a famous compose who composed, “Pomp And Circumstance”, famous marching songs and other patriotic songs that are famous allover.
    Rick A. Roberts

  27. Anton Alipov on August 16, 2014 at 10:59 pm said:

    A different reproduction (allegedly photographic) of the Dorabella cipher is presented at the Elgar Museum’s website.
    Here’s my brief report: http://athenaea.net/index.php?id=55

    Has this photo (the Museum’s) been previously known? Is it original or faked (for whatever reason)?

  28. Anton: thanks very much for leaving a comment and passing a link to your page. When I tried to track down Dora Penny’s Elgar donation at the Royal College of Music, an archivist there told me that several boxes of her early Elgar material were somehow lost (in Leeds? I can’t recall fully) en route to London, which presumably included the Dorabella Cipher. So – unless there has been some magical discover of this hitherto-lost material – the reproduction in her book is all we have to work with.

    Hence I suspect that what has happened here is that, in the absence of the real thing, a fake version of the cipher was mocked up to put on display as part of an exhibition… but I shall ask the curators about this, just to be sure. Will let you know what they say! 🙂

  29. Anton Alipov on August 17, 2014 at 2:42 pm said:

    Thank you in advance!

    I also thought about this possibility, but the strange side of it is that, in our computer age, the simplest way to “fake” (or “reproduct”) something is just to make a scan/photo of it. In this case they could just print the image from Wikipedia, frame it beautifully and present to the audience. Why introduce additional labour to “mock” the cipher, moreover – with discrepancies?! One should expect more respect to the relics from a museum.

  30. Anton: well, people do all kinds of things for all kinds of reasons. My guess would simply be that the print-out didn’t look hand-written enough. 🙂

    But I’ve already emailed the curators to ask, so hopefully we’ll get a definitive answer before long…

  31. Vedran on October 3, 2014 at 8:30 pm said:

    Hi Nick.
    So have you ever gotten a reply from them (regarding the photo)?

  32. Pingback: Top 10 Unbreakable Ciphers and Codes • NewsTipedia.com

  33. Anton Alipov on October 21, 2014 at 5:35 pm said:

    Yes, very interesting to know if the museum explained what’s actually the case with this alleged photo…

  34. Pingback: Unsolved: Dorabella Cipher

  35. Pingback: Unsolved: Dorabella Cipher | MotionBump Reader

  36. Rick A. Roberts on November 19, 2014 at 2:47 am said:

    Has anyone read my deciphering of the Dorabella Cipher that I posted on 22FEB2014 ? I believe that it is very accurate, and that it makes logical sense. Check it out. Thank you.

  37. Mark Pitt on March 4, 2015 at 7:02 pm said:

    The key to the Dorabella Cipher is ‘DORA X’.

    Insp. M Pitt

  38. Mark Pitt on March 11, 2015 at 9:54 pm said:

    Using the DORA X key provides the following:

    The first line of the Dorabella Cipher – Having lazy tranquil warm rest and recuperation LC crowd,ed kites…..

  39. I’ve posted a solution showing Running Key with stutter and indicator, explanation of the fullstop at char 65, reference to a mystery news ban etc. Also the name of the author. Link is http://www.gilliemusic.co.uk

  40. Anton Alipov on May 29, 2015 at 10:21 am said:

    Of course I’m not a native English speaker, but, sorry, “Cry rane pfog ufeletharge” does not convince me. Or maybe this is something Irish?

    Considering the story in its entirety, I think that this “cipher” is nothing more than a mystification introduced by Mrs Powell herself.

  41. mark pitt on July 16, 2015 at 8:20 pm said:

    The solution to the Dorabella cipher is based on the dial cipher set at AO and the courage card set sequence. This is encoded into the cipher date. The initial 14 symbols decipher ‘to fix set ao set weekly’, minus most of the vowels. The remainder of the cipher then rotates through the dial cipher every 7 and translates via the music cipher. My full explanation is available if anyone is interested I have posted it on the Elgar cipher yahoo group. I explain in the article my breakdown of the cipher date and the indication of the dial setting AO and show how I made my decipherment. This is all supported by the 1896 Pall Mall Magazine articles regarding ciphers as read by Elgar, which hugely support my decipherment.

  42. mark pitt on July 16, 2015 at 8:35 pm said:

    In short the date contains a pi sign at start, indication of dial cipher. There is Elgorod encoded into the date as explained in my article which indicates the ingenious first set of the dial I.e. AO so to complete both the names Elgar and Dora (backwards). Then a music note (the no. 9) and then the 7(repeat). The main lines in the courage card set match the main lines in the date. AO is also the dial setting in the pall mall magazine (1896). Read my article as posted to the Elgar cipher yahoo group on 14/7/15.

  43. Dan Biddle on July 22, 2015 at 9:25 pm said:

    Place a set of musical staves over the top of each line, every symbol is either a crotchet, quaver or whatever (I haven’t the musical knowledge to work it out) and it will form a composition.
    Elgar was a composer, from my hometown no less, so this seems like the obvious thing that it will be.

  44. Gunnar Johannson on September 8, 2015 at 5:20 am said:

    Hi I don’t know music either but they are the finger placements on the keys of 3 different instruments. I’d love to hear the song. He made a little diddy to that young lady she likely asked him I would love to hear your new stuff and he sent the keys order for her to figure out til he came out with his new music.

  45. I can’t help but remember back to when my daughter was learning to read music, she would make little “scribbles ” like this to remember which notes and fingers to place .. he was in love with the woman , of course he wrote music for her that she would understand .. Great website you have here by the way

  46. Aymii: well, there you have the paradox. On one fragment, the lettershapes are definitely used as a cipher, while on another (the Liszt fragment), they appear to be used as music. And on the third (the Dorabella Cipher itself)… we just don’t know either way.

    Thanks for the compliment about the website, though it’s not really as neatly organized as I would like. 😮

  47. Rick A. Roberts on November 23, 2015 at 7:36 am said:

    I am still awaiting a response from The Elgar Society to my work that I posted on 22FEB14. So far, I still have not received one.

  48. To fix Salut d’amour set weekly…[MUSIC]…In haste yrs
    E Elgar

  49. Christof on March 1, 2016 at 12:27 pm said:

    I did write a Python script for cipher cryptoanalysis. It sets up multiple variations according to the notes found (rotation, ABC, CBA etc.). Afterwards it searchs for multiple text patterns, e.g. words of length >4. I did search ALL combinations in a row (ABC/DEF etc. into both directions (e.g. FED/ABC/PON/etc..). This leads to 10,321,920 text string variations (8*2*7*2*6*2*5*2*4*2*3*2*2*2*1*2). The dictionary contained at least the most frequent 10,000 words. There was no solution. There was no solution in Latin either. Still have to check out French, as the note is signed on the 100yr jubilee of the storming of the Bastille. Ignoring the rotation ‘method’ of Elgar would result into 26*25*24*23*22*21*20*19*18*17*16*15*14*13*12*11*10*9*8*7 variations (20 symbols) or 560,127 billions x billions variations. Tuanhe-2 supercomputer would take approximately 2000 years to solve that one.

    However: We do have one symbol with a frequency of >12%, which thus might very well represent the letter “E”. It’s occurrence during the cipher is irregular. Also multiple double letters is a bit unusual. Although I think it very well is in English language, it could also be in a different one, too.

  50. Christof on March 1, 2016 at 12:33 pm said:

    One more thing for those interested..the hint with the 100yr. jubilee of the Bastille came from my ex-girlfriend (credits..).

    And another: The search algorithm used is the Aho-Corasick, which is faster in analyzing potential text strings as it sets up the dictionary as a ‘trie’. To perform a run in one language takes approximately 1h only. Actually I did expect a result, but there was none (no multiple words eg. >6 letters or more than 8 words >4 letters).

    So either a different language or EE has indeed chosen the symbols fully accidentially rather than according to any kind of rotation pattern. By the way…in his notes, the cleartext of his encryption tests is “A very old cipher”, “Do you go to London” and “Marco Elgar”. Indeed his dog was named Marco, so the encryption method itself can actually be ‘understood’.

  51. Christof: to my eyes, the biggest problem by far with the Dorabella Cipher is that there seems to be no earthly justification why Elgar would employ any tricky complicated crypto machinery to send a note to Dora – although they subsequently knew each other for decades, at the time he wrote the Dorabella Cipher they hardly knew each other. She was also more than a century too early to use CryptoCrack. 😉

    …and yet even now we cannot decrypt it!

  52. Rick A. Roberts on March 2, 2016 at 2:41 am said:

    I believe that my deciphering of 22 February, 2014 is a very good fit for the message that Edward Elgar sent to Dora. It shows Elgar’s wit and ties back into his weekly meetings that he had at his house . Also, it relates to the ” Unitarian register “, in which he used the ” Goose that Laid the Golden Egg ” reference .

  53. Mark Pitt on July 2, 2016 at 9:39 am said:

    Look at lot 92 sothebys, new Elgar cipher material.

    Mwp

  54. nickpelling on July 2, 2016 at 1:51 pm said:

    Mark: very interesting, thanks! Though… I’ll put up a post with my thoughts on it very shortly. 🙂

  55. Just a Random on October 22, 2016 at 7:48 pm said:

    I have only just found out that the Dorabella Cipher exists. I haven’t done any research or anything so I don’t know if this has ever been brought up. I would hope that it would have been thought of right away but here it goes.

    Has anyone thought to put this on a music sheet and see if it can be cracked that way? I don’t mean like it could be just music but it was the first thing I thought of.
    Like a “G” would be the letter “Q” or something. I don’t know much about music theory either to be honest.
    When I saw the squiggly lines it just reminded me of music… Just a thought.

  56. Just A Random: lots have people have tried exactly that, but with no obvious success to date. Oddly, Elgar also used the same set of squiggles for a note in the margin of a piece of music by Liszt (which does definitely look as though it might well be some kind of idiosyncratic music notation), but nobody has yet been able to decrypt that either. In short: if someone manages to decipher the Elgar/Liszt fragment, we might be in business with the Dorabella cipher… but there’s currently no sign of a break happening.

    http://ciphermysteries.com/2015/05/30/elgars-hysterious-other-ciphertext

  57. Andrea R on November 30, 2016 at 4:13 pm said:

    Hello Nick,

    I’m working on the idea that the cipher is a musical piano melody. The symbols might be notations written by Elgar corresponding to musical notes, in fact they strongly remind the well-known musical notes. If you write down a musical note, and just encircle it, you find exactly the Elgar symbols. The Liszt fragment must be as well a sort of Rosetta stone. If you try to map the symbols on the left margin to the musical notes , you can see the same correspondence. Applying this method to Dorabella cipher, I got a musical melody, but I lack the skill of a real piano player, so I don’t know if all the notes are in the right place; if you can help me with some expertise in piano music, maybe we make some progress.
    The cipher is the Elgar secret enigma variations’ main theme? this would explain why the composer sent it to Dorabella, as a way to keep it secret and at the same time to reveal it to future generations. Then he probably tried a personal mapping to an alphabet, by using the circles and ciphering attempts, just for fun, but the core element of the cipher must remain its music knowledge.
    BR/Andrea

  58. Danah Nahari on January 26, 2017 at 8:54 pm said:

    Wasn’t one of Elgar’s hobbies cycling those squiggles look like hills to me maybe those are the hills or mountains they cycled over together?

    Just a thought

  59. lyrabela on March 27, 2017 at 2:50 pm said:

    Is this still active? If so has anyone else heard of the police officer who supposedly solved it?

  60. lyrabela: that would be Cleveland police officer Mark Pitt, feted in the Daily Mail back in 2016 and who was threatening to write a book about his decryption.

    Pitt may possibly have solved it, though there have been so many hallucinated non-solutions put forward by amateur code-breakers before that the odds would seem to be rather against it.

  61. Rick A. Roberts on March 28, 2017 at 1:57 am said:

    Please look at my 22FEB14 deciphering and my message of 02MAR16 that was posted here. Thank you very much. It fits and it makes sense also.

  62. Narissa Andrews on April 1, 2017 at 1:24 pm said:

    Pigpen (or an octothorpe or a hash or whatever you call it) is basically the same as a sharp sign. The sharp sign in music.

    This, in my opinion, is still the likely base of the cipher, looking at the fact it was meant to be easy to crack, the squiggles fit into the sharp sign nicely in all directions and well, Elgar was a composer before the age of computers and WW1/2 code systems.

    I wonder if we have all gone too far with computer programmes and stuff and we should collectively take our brains back to the simplest and possibly most obvious way of cracking it.

    But what and where the letters are placed, who knows!

  63. Louisa on April 2, 2017 at 12:30 pm said:

    As an editor, I am sometimes handed really old historical original text which have letters missing, whole words sometimes, and I have to work out those missing letters.
    I came across this Dorabella cipher and took a look at it. Read all the blogs too and the history and attempts.
    The last squiggles on lines 1 and 2 I feel is the clue but looking at it, it seems to be structured backwards. When looking at frequency pairs and patterns the whole thing just looks backwards to me. Especially as the first letter is not repeated elsewhere so it’s most likely an X. Which people do put at the end of love letters.

    There’s 3 possible palindromes and 3 sections of high frequency arrangements, likely the letters A,R,T and two others.

    I’ll have a go at this, using the same theory that I use to find lost words in old manuscripts.

  64. Louisa: one problem is that the cryptogram is dated immediately below it, so we do have a pretty good idea of which way round it should go. The next big problem is that Elgar never explained his odd writing to Dora Penny, nor even why he sent the cipher to her. And the final big problem is that we have not a jot of evidence that Elgar and Dora had any kind of intimate relationship of the sort that Dorabella cipher decryptors tend to take as a starting point.

    All of which is just to say: take heed of all the blanks you are filling in, not just the words. 🙂

  65. Louisa on April 2, 2017 at 1:36 pm said:

    I’ll start going forward then 🙂

    Anyone searched for 3 line poems of the time? I’m not a huge fan of poems but just wondered.

  66. Louisa: a three line poem?
    He said the words wouldn’t go in
    But he’ll see, I’ll show ‘im!

  67. Bryan on May 14, 2017 at 11:12 am said:

    Unless The images circulating The web are inaccurate, the dorabella cipher is not signed and dated. It only has a date: July 14 97

  68. Bryan: according to Dora Penny, the cipher “came to me enclosed in a letter from the Lady [i.e. Elgar’s wife] to my stepmother. On the back of it is written, ‘Miss Penny’. It followed upon their visit to us at Wolverhampton in July 1897.”

    So while you are technically correct that it is not signed, there is also no obvious doubt whatsoever as to its author.

  69. Rick A. Roberts on May 16, 2017 at 4:10 am said:

    I do believe that my deciphering of the Dorabella Cipher is so far the most fitting that I have seen as to its message. Elgar would have meetings at his house on weekends and invite many well known musicians, philosophers, and scholars to his house to more or less “Brain Storm” and share ideas with one another.

  70. lol, clearly you guys don’t know greek. What a risque letter!

  71. Has the prize been claimed if not what is it?

  72. Sue C: my understanding is that the Dorabella prize was claimed by many but actually given to none.

    As to what it was, I suspect that it would be not entirely unlike the Royal Society Scientific Discovery of the Year award won by the Pirate Captain (and his pet dodo) in “The Pirates! In an Adventure with Scientists”, where the (prestige:gold) ratio was skewed strongly to the left-hand side. 😉

  73. Catherine Mary Darensbourg on August 15, 2018 at 12:50 pm said:

    Dear Nick,

    I think I made progress on this Cipher. It seems it may be simple substitution — only done in reverse. Here is what I got from the letters provided, starting in the lower right hand corner and going right-to-left:

    I am in rm CI or find u rm XI our I nr u-
    r T. Add a O fr I. N.B. T by fm N u B.B. u is DI. Any by
    rt m aka O I cc. N.B. b?b the T k. DI. Y ? N ? I u c ?

    I read this as:

    I am in room 101 or find U room 10-1 our 1 near your cross. Add a circle for 1. N.B. Cross by from number to be. U is 501. Any by route 1000 aka circle 1 to see. N.B. B? The cross k 501 y n ? iuc?

    Do you think this is close?

    Thanks,

    Cat

  74. Rick A. Roberts on August 16, 2018 at 2:47 am said:

    Check out my 22FEB14 deciphering of the Dorabella Cipher. To this date, I believe that it is the closest attempt to solving this enigma.

  75. Shadowwolf63 on May 9, 2019 at 6:04 pm said:

    The Dorabella Cipher has been solved. It is a mono-alphabetic substitution cipher. The plaintext is in Elgar speak which made it much more difficult to figure out. Unlike the many faked solutions this one is a real single alphabet solution with all the words dependent on the correct key.

    https://drive.google.com/open?id=1f4kssBlzrPkp5V1L677npTrbNgVmAsA8

  76. Hi Nick, I haven’t contacted you before but I am wondering if there is a way to contact you other than comments on this website.

    Cheers,
    James

  77. James: my email address – nickpelling at nickpelling dot com – is far from the world’s biggest secret. Feel free to email me there. 🙂

  78. “A WOMAN IS LIKE CHESS, ONE HAS TO MAKE

    MANY SACRIFICES FOR ITS QUEEN. IT IS

    VICTORY SHE COMMANDS, NOT DO BETTER”

    ….Look for the composers broken baton and reconstruct it before attempting to translate something that is totally defragmented….The composers handle(BULB) section is located on the number 7…The last clue….The DOT above the symbol…are you sure there is only one DOT or are there more?…3 DOTS…3 LINES

  79. , Rick A. Roberts on October 24, 2019 at 1:27 am said:

    Again, I must say that my 22FEB14 deciphering of the Dorabella Cipher is the only one that makes sense. Many times Elgar had people of all differing backgrounds come to his house to congregate with him and share ideas. He was a very witty person.

  80. “A WOMAN IS LIKE CHESS ONE HAS TO MAKE MANY SACRIFICES FOR ITS QUEEN IT IS VICTORY SHE COMMANDS NOT DO BETTER”
    There is a 2nd message of one word under the main cipher which reads:
    “RATS”

    There is a monogram of a conductors baton broken into 4 pieces…The reason Sir Edward did this was to leave a clue of how to realign the cipher back to its original state….as you align the baton it changes the sequence of the cipher.

    What everyone missed is the shape of the dorabella…its a conductors podium and the notes found in his journal is the shape of an orchestra.

    Podium, baton & orchestra.

    In short: You move the last 2 symbols of the middle line to the beginning of the bottom line….this gives the perfect podium of 29 symbols over 3 lines.

    The tip of the baton is the dot after the 5th symbol on the bottom line….this dot goes directly up to the top line….the next 2 sections of the baton are the 2 dots near the number 4 on the date….the one above the 4 goes directly up to the middle line….the dot between the 4 and 9 goes directly up to the bottom line….you then move each line one at a time to align with the batons BULB section which is the larger dot on the number 7….as you move each dot you must move the symbols of the lines and as you move the dot one space the end symbol of that line goes to the front of the line.

    Now you have completed the alignment we know that no matter how talented Sir Edward was he is limited to the English language as everyone else is and now working backwards (from the end) you will correctly predict there will be a word somewhere within the cipher ending with “ING or TER etc”….Using his journal as a ref you assume the last 3 letter at the end of the cipher are the last 2 or 3 letters of a common word (in this case it was “TER”….You then cross-check all symbols against the journal alphabet (You will notice the last 3 letters are one step away from the original journal…he had to do this to show the 2nd message “RATS” ….you will now notice parts of words coming to the surface such as “QUEEN, BETTER, CHESS”…Once you have completed step one you then write down what letters are 2 steps away and then 3 steps away….If you have followed the above steps correctly the only 4 letters Sir Edward has not moved is RATS….And that is how to break the Dorabella.

    “A WOMAN IS LIKE CHESS ONE HAS TO MAKE
    MANY SACRIFICES FOR ITS QUEEN IT IS
    VICTORY SHE COMMANDS NOT DO BETTER”

    Regards

  81. Anon on May 20, 2020 at 7:07 am said:

    @Mark Pitt The revolving dial cipher is just a simple Caesar Shift Cipher with extra numbers adding additional shifts. If the code was just using a simple Caesar Shift, then it would have been solved within the first 10 years regardless of how much “Elgar Speak” was used in the plaintext message.

  82. Hello Mr P

    Maybe (probably) I am being dense, but I cannot understand anything of what you are saying here. Can you clarify?

    I am not seeing anything like a ‘baton’ here, to say nothing of a podium, or orchestra (in the journal page).

    I tried solving this thing some years ago, and wrote a program to try any/all variants of the key as it is in the journal, and no amount of rotating or re-aligning makes any difference that I can tell.

    Can you give specific example of how you are decoding to arrive at your solution? I just don’t see it.

    Thanks in advance!

  83. J. A. M. on December 12, 2020 at 9:35 pm said:

    I am reading Dora Penny’s book Memories of a Variation to get more info about Dora’s relationship with Sir Edward Elgar. Dora was a young friend, a child, an amuser and a fan of Sir Edward Elgar. The note has a message that concerns Dora for sure.

  84. J. A. M. on December 15, 2020 at 6:44 pm said:

    Hi everyone,
    Here is my solution for Dorabella cipher. I apologize for my handwriting, I hope you can read the labels on the sketches. Please read my explanation to understand what the gestures conveys. Please comment your thoughts. Thank you!
    https://youtu.be/y6C3-MNwI8M

  85. Simon on March 24, 2021 at 2:23 pm said:

    Not got enough time to look in detail right now but just a casual observation having stumbled across this page… The symbols being written quite often show the stroke start (the blot from the ink). People not breaking old habits may have a tendency to “rotate the paper” if noting directional “hop” based methods.

    There are many potential approaches that could be taken, but one example would be a step-based decoding method, with the “n” – “mm” representation showing how many “hops” take / directionally on the grid with the left/right alignment etc indicating direction. It’s quite a good idea, all the more so if paired and used on 2 play-fairs. I’d personally be inclined to have a default “base” step assumed before applying any other translations to break up consistency.

  86. Hi Everyone.

    I’m new to cryptography and I have a solution to the Dorabella to propose. A few things that I believe are interesting about this potential solution:

    1. The arrangement of Elgar’s symbols into a rose-shaped key is elegant, and may explain the circular diagrams in his notes from the 1920’s.
    2. The message seems to fit in and illuminate Elgar’s work. I believe it may foretell his writing of the “Enigma Variations”
    3. The rose key points to a solution to his Enigma – the hidden melody of the “Enigma Variations” – his own 1892 song “Like to the Damask Rose”
    4. This solution to the Enigma meets all of Julian Rushton’s and Patrick Turner’s criteria.
    5. The key seems to work for the Liszt Fragment and the “Microcryptogram”.
    6. I do not believe he made any mistakes in the encryption.

    A few things I’m still uncertain about:
    1. My British English is a bit rusty since it’s been a couple centuries 🙂
    2. The decryption process led me to a layer in addition to the MASC that uses “u” for the word “you”, “w” for the word “with”, “ay” for the word “ever”, and that encodes the digraph “ae” and the diacritical é with their own slots on the key. I would have rather had a clean MASC, of course.
    3. I believe there is a second dot that is less visible than the famous one, that has been lost to most online images. It’s fairly clear when scanned from Dora’s book.

    I’m not ready to claim that the solution is 100% correct, but I think the coherent picture that emerges indicates that I’m onto something. If anyone is interested you can read the solution on Medium, or get in touch with me if you prefer a PDF. https://zbelanger.medium.com/solutions-to-the-dorabella-cipher-and-edward-elgars-enigma-65583d2506c5

  87. Anthony on September 9, 2021 at 3:24 am said:

    Hello,
    I created this video. I’m not sure if I solved this puzzle. Anyway, I hope you enjoy this presentation.

    https://www.youtube.com/watch?v=pERLcFENOpc

  88. I found something interesting. I was manually inspecting a (relatively) small space of possible solutions, and I noticed that every time the last line started to look like English, the top line looked nothing like English. And vice versa. I don’t mean coherent words, but arrangements of consonants and vowels. I visually inspected maybe 10,000 different results over the period of a couple months, and the pattern continued: every time the one line became more English-y, the other was less so. I started to think that applying a hill climber to only part of the code might offer an opening, that maybe there was something about a part of the text that was throwing off the algorithms, and a partial solution might lend insight. So I started looking at hill climbing results for each line. Much more coherent but with a lot of partial words at the boundaries. I then shifted to splitting the Dorabella into three parts demarcated by the two dots – the well-known 3rd line 5th character dot, and the less well-known 2nd line 11th character dot, which doesn’t appear in most online images but which appears in Dora’s 1949 book.

    I also manually inspected ranges of high-scoring results, not just the local maxima. Long story short, I came up with this, using a different key for each section, demarcated by asterisks:

    PSLOVEROFACTWITHTTESSTINANDCU
    RTRENOFBOLT*ITSOLDINISHALFERRENA
    TASHA*WINDERWASNORETHERSTILT

    With word spacing, caps, and punctuation:

    PS Lover of act with T. Tesstin and curt ren of Bolt.
    It sold Inishal Ferre. Natasha
    Winder was nor ether’s tilt.

    Multiple keys would explain the failure of hill climbers, but also increases the chances that the results are untrustworthy, of course. The dots do occur at the end of the first sentence and after “Natasha”, which is pretty clean, but nowhere near definitive. The names may offer a path to verification, especially if there was a performance of “Ben Bolt” around the time the cipher was written. Another verification path could be connecting the keys with each other, if Elgar systematically derived them from each other, but to me there’s no obvious way to translate between them. I put a (short) writeup here, in case anyone is interested.

    https://zbelanger.medium.com/solutions-to-the-dorabella-cipher-and-edward-elgars-enigma-65583d2506c5

    I still think “Like to the Damask Rose” could be the solution to the Enigma, but if this Dorabella solution gets traction, it wouldn’t hurt to look at “Ben Bolt”, too. The song is about the death of a woman named Alice, which is also Elgar’s wife’s name.

    Is anyone familiar with the term “ether’s tilt”?

  89. Zackery: there is no dot in the second line in the copy of the 1947 edition on my bookshelf, so I’m unable to follow your reasoning here, sorry. 🙁

  90. Zackery Belanger on October 6, 2021 at 4:18 pm said:

    Nick – that’s interesting. Mine is a 1949 copy by Methuen & Co. Ltd., printed in Great Britain. The dot is in contact with the 12th character, at the tip of one of the arms of the squiggle, but it’s clearly there, and about the same size as the dot on the third line. Is yours the same publisher?

  91. Zackery: Geoffrey Cumberlege, Oxford University Press, Second edition 1947. No dot in sight.

  92. Zackery Belanger on October 6, 2021 at 6:12 pm said:

    Fascinating. Would you mind giving an opinion on whether the dot in my copy could be legit? Not from the standpoint of what publisher may have done what, but just graphically using the close-up scan of it at the beginning of my post here:

    https://zbelanger.medium.com/solutions-to-the-dorabella-cipher-and-edward-elgars-enigma-65583d2506c5

    Also let me know if a more zoomed out photo would be helpful.

  93. “Bryan on February 4, 2014 at 4:25 am said:

    I keep seeing people stating that Elgar’s signature is at the bottom of the Dorabella Cipher, yet if you compare it to his actual signature the two are nothing alike. I think it is actually a clue to decipher the text. I’m gonna play around with it a bit”

    It’s not a signature, it’s the date, July 14 ’97

    Should these people really be trying to decode this ‘cipher’ if they can’t even read a date? *shakes head*

  94. James: that’s not the worst, not by light years. 🙁

  95. Zackery Belanger on March 11, 2022 at 9:13 am said:

    I kept digging into the extra dot and it led to an interesting potential solution

    https://zbelanger.medium.com/a-botanical-solution-to-the-dorabella-cipher-8058046cd21f

  96. Anon on April 7, 2022 at 2:24 pm said:

    Recent news on the breaking of a Bellaso/Della Porta cipher from a
    notebook [BL, Sloane MS 1902] to which both Arthur Dee and his famous alchemist, polymath father, John Dee, had contributed.

    https://www.atlasobscura.com/articles/medieval-alchemist-secret-code

  97. James M on April 26, 2022 at 5:55 am said:

    Anon on April 7, 2022 at 2:24 pm said:
    *snip*
    This is spam, has nothing to do with the Dorabella Cipher

  98. Amos G. on July 23, 2023 at 4:53 pm said:

    I see you have a post on Tim Roberts’ decryption, but nothing on Richard Henderson’s claimed decryption which is cited on Wikipedia “whY AM I VERY SAD, BELLE. I SAG AS WE SEE ROSES DO. E.E. IS EVER FOND OF U, DORA. I kNOw I PeN ONE I LOVe. All Of My Affection”. The message seems quite sensible to me, and I would like to know why it’s been rejected by cryptologists as a likely solution.

  99. James M on February 8, 2024 at 10:51 pm said:

    “The message seems quite sensible to me, and I would like to know why it’s been rejected by cryptologists as a likely solution.”
    People continue to start with the assumption that it’s a love letter and work from there. This isn’t how you’re going to solve it as your bias will cause you to create solutions that aren’t real. Also people are using the same methods over and over again which have been tried ever since the cipher was shown in Dora’s diary including people from the NSA. The only way we will ever solve it is if people start thinking outside the box instead of wearing out the same path.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.