The Dorabella Cipher Competition
Edward Elgar sent a cryptic note to Dora Penny, dated 14 July 1897, enclosed in a letter from Alice Elgar to the Reverend Penny and his wife.
The Elgars had stayed at the Wolverhampton Rectory for a few days and Edward Elgar had become friendly with Dora Penny, who was twenty years his junior.
The Elgar Society Competition 2007
The Elgar Society is familiar with the background to the Dorabella Cipher. Edward Elgar sent an enigmatic note to Dora Penny on 14 July 1897. It survived in a bottom drawer, and, forty years later, Dora (née Penny) Powell published a photograph of the note in her memoirs, saying that she had never been able to read it. And, so far, no one has determined the meaning of the message.
Eric Sams produced a translation of Dorabella in 1970 . But, although he made a number of reasonable points, his method of attack is convoluted and hard, or impossible, to follow. His interpretation of the message was this:
STARTS: LARKS! IT'S CHAOTIC, BUT A CLOAK OBSCURES MY NEW LETTERS, A, B [alpha, beta, ie Greek letters or alphabet] BELOW: I OWN THE DARK MAKES E. E. SIGH WHEN YOU ARE TOO LONG GONE.
The length of Sams'text is 109 letters (ignoring the parenthetic note on Greek), whereas the original text contains only 87 or 88 characters. Where do the extra 22 letters come from? Sams claims they are implied by phonetic shorthand.
Having doubts in Sams, members of The Elgar Society who have a special interst in ciphers declared a Dorabella Competition in 2007.
Accept that the writer uses the words code and cipher interchangeably; yet, the assumption should be that the message is possibly a cipher – and a form of double encipherment to boot? To claim that ‘it is written in a Code’ is not a valid assertion simply because it looks Arabesque and mysterious. In fact, the symbols do not resemble Arabic which is written from right to left and links its letters. Greek is a slightly better candidate as an alphabetical model, but, simply, the double-arched cursive in Elgar’s signature is a likely contender for the basic shape. Furthermore, the contention that it is a Code waiting to be broken makes it sound more of a task for the difference engines of Bletchley Park than the pencil and paper potterings of an amateur cryptographer.
Dorabella is from the Nineteenth century world of Classical, amateur cryptography. It conforms to a popular view of a cipher. It looks mysterious; and that may be its only purpose – to be a mystery for the sake of mystery.
The Elgar Birthplace Museum preserves four copies of The Pall Mall Gazette from early 1896 which carried a series of articles on cryptography called Secrets In Cipher. Elgar kept these four issues and, presumably, read them before his trip to Wolverhampton in July 1897. The spelling of the word cipher or cypher was arbitrary at the time. Certainly the word was taken to mean a piece of secret writing; but that was not its only sense.
British manufactured good were stamped with ciphers throughout the reign of Queen Victoria. Nowadays they are often called diamond registration marks, but they are referred to as ‘ciphers’ in the Registration of Design Acts. All British glassware, metalware and ceramics were stamped with a cipher. It was a maker’s mark and a seal of approval. Elgar saw the registration stamps on his bottles and plates and cups and saucers and knives and forks and spoons every day.
The glyphs of Dorabella resemble the marks of Renaissance nomenclator ciphers , but they belong to the class of diagrammatics because they are formed from a regular geometrical shape. These glyphs are Elgar’s personal cipher in two senses. First, he used them for private notes, in which case they are a form of secret writing; and, second, another older meaning of cipher is a monogram. For centuries, Kings and Queens of England have had a personal cipher or monogram. Victoria was the first to put hers on the newly invented post boxes.
The printed form of the Royal Cipher appears on the reigning monarch’s props dependent on the crown. Elgar had seen plenty of VR ciphers, and Dorabella was written only weeks after Victoria’s Diamond Jubilee. (She was VRI by then). Why not have one of your own? A man who could devise a palindromic telegraph address SIROMORIS (Knight, Order of Merit) would surely think up a distinctive domain name for himself in these internet days. Moreover, as a publicist, Elgar would want a logo; and that logo could be the Dorabella Cipher itself.
Yet, the curious glyphs are smoke and mirrors. They appear to represent a 24-letter alphabet arranged in a diagrammatic pattern on eight spokes of a wheel – or, if you prefer, along the eight major lines of a compass rose.
It is an old form of compass cipher. When Sams suggested that we look for a clue to Dorabella in a book, he meant that we should look for the book in which Elgar found his pattern . Sams claimed that such an arrangement is described in a cipher manual of 1809 (Klüber, Tübingen, 1809) . Purely by inference, Elgar could have seen it in the British Museum Library (the book is now in The British Library) as did Sams; though Sams read German whereas Elgar did not. Both Sams and Elgar could also have looked at the Royal Manuscript Collection in the British Museum and found a very similar cipher written by King Charles I (the king’s letter to the Earl of Glamorgan, dated 1646, offering concessions to Catholics) .
Where else might Elgar have found his cipher pattern? Sams implies that we should be looking for a cipher manual; but what other works contain patterns? Could it be a work of popular mathematics? Could it be a book of stage sets and designs? This latter idea is a possibility.
The name Dorabella did not attach to the cipher for a long time, and the name was not originally associated with Dora Penny. When the Enigma guessing game of Who’s Who began, Variation 10 was connected with the Dorabella of Cosi fan tutte – given Elgar’s interest in Mozart. And what of Mozart’s last opera, which followed on the heels of Cosi? Historically, the main users of diagrammatic ciphers have been Masons. Mozart and his librettist Schikaneder were Masons and the stage sets of Die Zauberflöte usually reflect Masonic designs.
Before looking at the possible Masonic connection in more detail, perhaps we should examine the stumbling block of why no one has been able to read the Dorabella note.
Attempts to solve the cipher by analysis of the frequency distribution of the characters have failed. (Elgar Society Notice)
Why? It is a one-off, so cannot be compared with another Elgar note of a similar type. There are too few characters for detailed analysis, and the hand-writing is uneven and arguable. The ‘double-encipherment’ of the Elgar Society blurb is very probably a monoalphabetic substitution followed by a transposition – the simplest of amateur cipherers’ tricks – and Elgar was an amateur cipherer. The result of these two operations was then transcribed into the Elgar glyphs.
1. FOUR AND TWENTY BLACKBIRDS BAKED IN A PIE 2. S B H E N A Q G J R A G L O Y N P X O V E Q F O N X R Q V A N C V R 3. H B E S Q A G N A R G J Y O N L O X V P F Q O E R X Q N N A C V R V
The four stages are as follows:
1. Plain text 2. ROT13 substitution – the mirror letter of the alphabet (ROT12 in Elgar’s case) taken from Elgar’s own cipher wheel (illustration right) 3. Transposition – 3241, which is not difficult when the letters are set out in a grid (as will be discussed below) 4. Transcription into the glyphs, which Elgar knew by heart
Dorabella exhibits patterns that suggest it is not an entirely random text, yet, if the underlying cipher alphabet is random and the method of transposition is arbitrary, then it cannot be read back and cannot be repeated.
Could Elgar read the message back again? Probably not. Could Dora read it? No. Could Elgar repeat the cipher? Well, if it was created at random, then, once again, no. So, we have a message that is unreadable, irreversible and unrepeatable. Is it a cipher in the sense of a (secret) text message? No. So, if it cannot be reversed or repeated, then, NO, it is not a message – for the simple reason that no one can read it at all (including its author). A statement from The Elgar Society website contains an article of faith: ‘when the correct solution is found it will be glaringly obvious.’ If that is so, you wonder why no one has seen it before. Perhaps decipherers are looking for the wrong thing: and, no doubt, they would like a solution of the sort ‘Dear Dora… I will make music of you.’ This is not forthcoming.
Sams wrote in the notes to his Musical Times article:
It seems in any event incredible that Elgar should have expected anyone to decipher his message with no kind of key or clue.
Why send an unreadable message to the Pennies? The message went to the whole family as it was inside Alice’s thank-you letter, written the day after the Elgars’ return from Wolverhampton to Great Malvern. The Reverend and Mrs Penny saw the note as well as Dora. It must have had a meaning or significance for them, connected with Elgar’s stay; and, presumably, it appeared to be more curious than sinister. (Secret messages are objects of suspicion). Dora kept it for forty years and sent it to her publishers to be photographed. Subsequently, the original note was lost.
Professor Kevin Jones makes an interesting point on this question, regarding Elgar’s visit to the Wolverhampton Rectory in the second week of July 1897:
Dora's father had just returned from Melanesia where he had been a missionary for many years. Fascinated by local language and culture, he possessed a few traditional talismans decorated with arcane glyphs. Perhaps such an item surfaced as a conversation piece during the Elgar's week in Wolverhampton? And if Dora recalled this when writing her memoirs, it might account for the fact the coded message was referred to as an 'inscription' when communicating with the director of SOAS many years later.
If the message has no literal or textual meaning, can we find any significance in it at all? Yes, the numbers remain when we count the parts.
Is it a 3 x 29 letter pattern or is it in two parts? I believe that the text length is actually 88 characters with a hiatus at position 66. This is easily constructed in an 11 x 8 grid.
Although a text underlies Dorabella, the text does not need to make sense when the length of text counts rather than the words. When constructing a message in such a grid, the text is either padded or abbreviated to fit. Elgar did it in half an hour, or less. Decipherers who midnight vigils keep in efforts to unravel the text will be at it until the lowing herd wind slowly o’er the lea. Any old rope would do, and Elgar was not at a loss for words to fill the gaps.
Why a text length of 88 characters? First of all, it’s a nice number: it is the number of keys on a piano and Elgar was writing a bagatelle. Did anyone wonder why the Chinese began the Olympic Games in Beijing on 08-08-08? It’s a fun and fortune cookie number. But Elgar split that number into two parts – a 65 (a figurate octagonal number), a hiatus, and a 22 (a figurate pentagonal number) . It is a numbers game.
Where do these numbers come from, what do they signify, and had anyone done anything like it before? Yes, bearing in mind that Dorabella is possibly a cipher in the sense of a mark or seal, there is a very good historical example of such a construction based on numbers – it’s on every US dollar bill, and the man primarily responsible appears on the $100 dollar note (a Benjamin).
On 4 July 1776, The Continental Congress in Philadelphia conducted more than one bit of business. First and famously they agreed on a final draft of a Declaration of Independence – leaving out awkward matters such as the abolition of slavery. Then, having signed it, the Congress appointed a commission to devise a new cipher or Great Seal for the United States. The first commission was given to Benjamin Franklin, Thomas Jefferson and John Adams. Six years and three committees later, the present Seal was adopted in 1783. It is a graphic and significant declaration of Sovereignty. Recollect that James II was deemed to have abdicated when he threw the Great Seal of the Realm into the River Thames in 1688. It meant something – rather a lot – to a new government. The US Seal is a construction based on the number of Colonies at that time and Masonic symbolism – the Eye of Providence is set in an equilateral triangle, with twelve spokes of light, and sits on a pyramid of thirteen steps. It is a mark of Providential Design.
Earlier, Benjamin Franklin’s first draft had been rejected as cumbersome and elaborate. He took a woodcut from the Geneva Bible of 1596 depicting the host of Pharaoh (George III), the Israelites, Moses, the Pillar of Fire, and the East wind parting the Red Sea. This is an illustration of Exodus Chapter 14, Verses 19, 20, and 21; and these three verses are curious for their numbering in the original Masoretic Text: each verse is composed of 72 Hebrew letters. And how does a scribe make sure that his text is exactly 72 letters long? He uses a 9 x 8 grid, in the same way that Elgar probably used an 11 x 8 for Dorabella.
To go into the symbolism and significance of numbers is too much for a short article. The Pythagoreans played number games and Euclid set them down in his Elements. Hebrew scholars played with numbers and magic squares; and, in written Hebrew, letters are numbers and numbers are letters. The game was played over two and a half millennia before computers – one example is the relation between the Divine Name (Tetragrammaton) in the Torah and the Pythagorean Tetraktys. These are games that Masons play and more than a few Americans have been Masons: one 32 degree Mason (Franklin Delano Roosevelt) approved the construction of The Pentagon in Washington; and what are the internal angles of a regular pentagon?
At bottom, the name of the game is turning a square into a circle, and the Babylonians, anciently, knew the factors at play in the number 360. English Masons based their own diagrammatic cipher – known to schoolboys as PigPen – on the 3 x 3 magic square which adds to 15 in all directions and has the number 5 at its centre. They took the idea from the Knights Templar who used a diagrammatic cipher for their internal secrets and transactions. The Templars based their cipher on a 5 x 5 magic square which sums to 65 in all directions and has the number 13 at its centre. And, of course, 5 x 13 = 65. The number 5 was the second most significant number to the Pythagoreans after the sacred 10. The letter E happens to be the fifth letter of the English alphabet and double ‘e’ may count as 10 – and represent the EE of Edward Elgar. Such games, however, wander into the province of Dan Brown and the Da Vinci Code. The Pythagoreans took number patterns to be proof of divine architecture.
Benjamin Franklin played with both squares and circles and came up with a magic circle composed of 65 sections that can be made to add to 360 in lumps and clumps and clusters of 180.
The symbol on the door of Freemasons’ Hall is a representation of the first step in turning a square into a circle; and it is the compass pattern on which Elgar based his diagrammatic letters – for which he could equally have used a straight Latin E, in a manner similar to the cipher of Charles I.
Elgar may have picked up a smattering of Masonic symbolism from Mozart whose Magic Flute stage sets plaster graphical arcana around their temple. In the opera, Sarastro is the High Priest of sweetness and enlightenment – a Master Mason like Franklin and the Masons of Philadelphia – opposed to tyrannous Night.
No one can know exactly why Elgar sent the note to Dora. It is likely to have been in response to something that was seen or said in Wolverhampton a few days beforehand. He had the pattern in his mind or in a book. It is also a dry run for “The enigma I will not explain - its 'dark saying' must be left unguessed ...” Balzac had pulled a similar publicity stunt with a hoax cryptogram in 1829 which left the punters guessing . Decipherers blamed themselves for their own lack of acuity. Balzac had a good laugh. Dorabella, too, is an Elgar joke. It is a cipher, yes; but it is not a cryptogram. Elgar was making his mark.
As a last curiosity, the Variations on an Original Theme follow the pattern of two book ends with a dozen identifiable friends and a mystery guest – Franklin’s mysterious Masonic 13th.
Stephen Colbourn SIGSec British Mensa Codes&Ciphers SIG
The full text of his entry to the Elgar Dorabella Cipher Competition is at http://tech.groups.yahoo.com/group/Elgar-Cipher/files/HowDidElgarDoIt/Elgar%20Comp/
The Musical Times, Feb., 1970, (pp. 151-154), Elgar’s cipher letter to Dorabella, by Eric Sams Copyright The Estate of Eric Sams http://tech.groups.yahoo.com/group/Elgar-Cipher/
Copies of the microfiche were uploaded to the group in July 2007: Secrets In Cipher – parts I / II / III / IV From The Pall Mall Magazine 1896, January onwards: republished as a compendium volume in January 1897
Copyright of Design Act 1842, superseded by the Copyright of Designs Act 1883. The diamond marks are recorded in the National Archives at www.nationalarchives.gov.uk/ Nomenclator ciphers resembled a form of secret shorthand; they were a courtly device, used by monarchs and diplomats, which mixed ciphers and codes: the most famous example in English is probably the Babington cipher used to correspond with Mary Queen of Scots. Diagrammatic ciphers are based on geometry and are easily reproducible from memory: this cipher form was a favourite of Masons, Rosicrucians and Templars. Illustration: the personal cipher of Victoria Regina – in her own hand (public domain graphic) Sams: (above, para 3) At first sight it looks easy. The basic system of cusps and arcs is described in a cipher manual of 1809. Elgar's pattern of one, two or three arcs at eight possible angles implies a system of 3 x 8 = 24 symbols and hence a simple substitution cipher (ironic technical term for the direct replacement of letters by other letters or by symbols). The most efficient (though least diverting) way of solving such ciphers is by getting hold of the key. So first we look for a book in which Elgar might have found it. But in vain. BL System number 001986180 Author - personal KLUEBER, Johann Ludwig. Title Kryptographik Lehrbuch der Geheimschreibekunst (Chiffrir- und Dechiffrirkunst) en Staats und Privatgeschäften. Mit vier Tabellen und Sechs Kupfertafeln. (Anhang. Literatur der Kryptographik.)
BM Date of acquisition prior to 1836; so could have been seen by Elgar; but Sams read German whereas Elgar did not. The BL copy is noted as ‘wanting tables.’
http://www.royal.gov.uk/output/Page601.asp ROT13 = Rotation 13, or, in a 24-letter alphabet – Rotation 12 Sams endnote 5 http://www.bbc.co.uk/proms/2007/interact/puzzles/dorabellacode.shtml There are several numeric patterns here; the simplest being that 5 and 8 add to 13 and 5 by 13 is 65. The number 13 also happens to be a Fibonacci prime.
The Unicode consortium has included the octagonal 65 in its Arabic section at decimal 1758 ۞ The rub’ alHizb is a quarter part and printed versions of the Holy Qur’aan divide the entire text into 120 sections, so that four sections can be read on each of the 30 days of the Holy Month of Ramadhan.
The Seal of the United States of America (public domain graphic courtesy US.gov) One example is this: the underlying structure of the Inferno, Purgatorio and Paradiso in Dante’s La (Divina) Comedia is numeric – the numbers 3 and 9 recur and the total number of cantos is the Pythagorean 10x10.
Another example of numerical games in the Holy Bible, apart from the number 72 already mentioned, is the sequence formed by addition of the consecutive numbers 1 to 36: 1 3 6 10 15 21 28 36 45 55 66 78 91 105 120 136 153 171 190 210 231 253 276 300 325 351 378 406 435 465 496 528 561 595 630 666. Several of these number have Gnostic significance or curiosity value – the sum of the numbers 1 to 15 (3x3 magic square) is 120 which happens to be the fifteenth term in the series; the seventeenth term is the narcissistic 153 known as The Catch of the Fishes; the thirty sixth term, which is one tenth of the angular number of a circle or four squares, is the so-called Number of the Beast.
No great significance is claimed for these number patterns, but they were known to sixteenth century cryptographers and are repeated in Masonic literature: the University of Bradford Masonic literature collection is at http://www.bradford.ac.uk/webofhiram/ 1911 Encyclopaedia Britannica From the door of Freemasons’ Hall in Great Queen Street, London (public domain graphic) From Elgar’s notes for the first performance of Variations on an Original Theme for orchestra, Op. 36 ("Enigma"), St. James Hall, London, 19 June 1899 Honoré de Balzac, The Physiology of Marriage, 1829 The number 13 appeared to have a Masonic significance for Benjamin Franklin; he refers to magic squares in his autobiography and also to his 13 precepts of virtue: Mémoires de la vie privée de Benjamin Franklin 1791 (first English edition 1793). The trivial and coincidental connection between Elgar and Franklin is that their portraits were printed on bank notes – the twenty pound note and the hundred dollar bill.