One of the hurdles cataloguers encounter in deciphering inscriptions in the Cardiff Rare Books Collection is the combination of languages which might appear, both in the actual text of inscriptions and in the names of people and places. Throw some occasionally idiosyncratic handwriting into the equation, and remember that spelling was far from being fixed before the end of the eighteenth century, and the result can sometimes be a challenge.
One such inscription came to light recently, and caused some bewilderment. It appeared on the front endpaper of A Defence of the Thirty Nine Articles of the Church of England (London, 1700). The edition is not particularly rare or otherwise remarkable; there are three copies in the Cardiff Rare Books Collection alone.
The handwriting is unusually neat, with the letters carefully written separately rather than joined up, but the words seemed to make no sense at first glance: “Ryvmnorgu Pbyg ure obbæ ; Ebiynaq Cneel ; Naar Pbyg”.
The Cardiff books originally came from a wide variety of sources, and, as one might expect, many of the inscriptions are in English. Rather fewer are in Latin, and there are also some in Welsh or in a combination of two or more of these languages. An inscription might, for instance, be in English or Latin but include a Welsh place name such as the name of a farm or house.
With this in mind, at first it was thought that this might be Welsh: in particular, there is the ending of the first word, -gu, which is a familiar Welsh suffix, and there are several occurrences of “y”, a commonly used vowel in Welsh. Welsh speakers in the department however were fairly sure that this was no Welsh they had ever come across, and the third person to have a look thought it was more likely to be in code – but how to work it out?
Hastily adding the ability to solve anagrams and crossword puzzles and play Scrabble and the like to the cataloguers’ skillset, I made a note of the inscription and took it home to see if inspiration would strike.
Substitution ciphers, in which letters are substituted for other letters, have a long history. This variety is known as a Caesar cipher, because Julius Caesar is said (by Suetonius) to have used one in his private correspondence, although he did not invent it. In his version letters were shifted three places (so A = D, etc.). There are many variations, and if you are dealing with a piece of English prose there are some clues to help: the letter E, for instance, is the letter which occurs most frequently in English, and there are certain sequences of letters and commonly recurring words which you would or would not expect to find in English. In this case we did not know whether English was the language, and of course an inscription in a book is not the same as a prose passage. Thinking along those lines was however the key to spotting the pattern here, as one phrase which does often appear in lower case book inscriptions is “his book” following a proper name, which fitted the number of letters at the end of the first line. I experimented with “his book”, which didn’t quite work, but looked promising enough, and so I thought I would be a bit more radical and try “her book” instead.
ure obbæ = her book
This produced enough to be able to see what the first name might be and to work out the code =
Ryvmnorgu = E—-be-h = Elizabeth
The code is a simple Caesar cipher, which today would be what is known as ROT13, in which the alphabet is rotated 13 places so that the top half of the alphabet is interchangeable with the bottom half:
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
On seeing the two halves of the alphabet lined up this way, it became apparent that we had all mis-read the letter “x” as “æ”. Using this code to decipher the inscription, we have:
Elizabeth Colt her book
These three names are mentioned in “The Baronetage of England” (page 522, Volume 2, 1771):
“Anne Dutton Colt … died unmarried; and [her sister] Elizabeth, married to the rev. Mr. Rowland Parry, of Letton in Herefordshire”.
Anne and Elizabeth were the daughters of the former MP for Leominster, John Dutton Colt of Dutton House, Leominster (1643-1722), whose career at a troubled time in English politics included a spell of imprisonment. Elizabeth died in 1736 according to a memorial plaque in the church at Letton, and her husband the Rev. Rowland Parry died in 1761. It seems fitting that A Defence of the Thirty Nine Articles of the Church of England should have belonged to a minister’s wife, and the names are unusual enough in combination to feel that we can tentatively identify them with these three.
It seems likely that our coded inscription was a light-hearted piece of fun, and it was entertaining to be able to decode the message at a distance of three hundred years. Clearly this kind of cipher is not a very secure way in which to communicate!
We’ve solved one of this book’s enigmas, but another remains shrouded in mystery: someone has stabbed through the entire text block of the book, leaving a 10-15mm slash through both covers and every single page! We are left to wonder who might have vented their anger on this poor volume and why.