1893 text

The making of ciphers was a popular amusement about this time. Pepys made several for Montagu, Downing, and others.

“character”: Private cryptic code D.W.

This text comes from a footnote on a diary entry in the 1893 edition edited by Henry B. Wheatley.

8 Annotations

First Reading

Emilio  •  Link

17th-century 'characters'
[Posted by Susanna for 25 Apr 1660:]

It was a fairly common practice in Pepys' day to use characters that were not part of the standard alphabet in one's ciphers. In addition to a simple substitution scheme (where, for instance, a=l, b=m, c=n, etc.), other original, i.e., non-alphabetical, symbols might be used to replace common words such as 'and', 'the', 'with' etc. There would probably also be nulls (symbols that stood for nothing at all, designed to confuse the enemy cryptologist) and possibly also a dowbleth (a symbol indicating that the next character should be read as a double letter).

I hope for Pepys' sake that his cipher was a good one, although I doubt it was as cleverly nasty as the Great Cipher used by Louis XIV’s spymasters, Antoine and Bonaventure Rossignol (father and son), which after their deaths (they had not passed its secrets on to anyone else) was not broken until the 1890s. (For more fascinating information about the history cryptography, I recommend "The Code Book: the Evolution of Secrecy from Mary Queen of Scots to Quantum Cryptography", by Simon Singh.)

vincent  •  Link

sipher, cyfer, cifer, ciphre, sypher, ziphre, scypher, cyphar, cyphre, ciphar, zifer, cypher. Francis Bacon who wrote about it spelled it as ciphras in Latin
interesting 16/17 century (j bonds galore) spies etc.,
"...science of cryptology was not taken seriously at least on the royalist side until very late..."
"... lines are from Bishop John Wilkins MERCURY the Secret and Swift Messenger 1641 .." should be read

vincent  •  Link

Ciphers from Grahamt 9 may one gets another in put about cyphers
Wallis in 1642 -.".. one evening at supper, a letter in cipher was brought in, relating to the capture of Chichester on 27 December 1642, which Wallis in two hours succeeded in deciphering. The feat made his fortune..."

vincent  •  Link

Tachygraphy has been reprinted in 1970: Thomas Shelton : Modern Reprint by Professor Matthews Pub. Augustain Reprint Society cost $27.50 ISBN 0404701450 available thru Borders usa

Second Reading

San Diego Sarah  •  Link

Perhaps people in Pepys time called ciphers "characters" because:

The History of Shorthand by Anita Kreitzman (excerpted) -- National Court Reporters Association
"Modern Times
... [in] 1588 a revival of shorthand occurred with the publication in London of Dr. Timothie Bright's Characterie. An Arte of Shorte, Swifte, and Secrete Writing by Character. ... Bright's system was not an alphabet, but rather a list of 500 arbitrary signs to be used in place of words. It was John Willis who first published an alphabet shorthand in 1602."

So having established the idea of characters substituting for words, Hebrew characters would be as good as any if you were writing code as Pepys appears to have been doing in January of 1660.

Third Reading

San Diego Sarah  •  Link

In 2013 Dick Wilson shared:

An earlier annotation asked for clarification of codes, ciphers, and characters from annotators who had worked with codes. In my youth I was an Air Force codes officer.
These days, the message encryption process is so highly automated that it bears no resemblance to the processes Pepys used. A few pen and paper ciphers remain, and they are so weak they are changed daily, and are used only to protect information that needs security for a brief period of time. For instance, you might want to tell someone that an aircraft has taken off, while concealing that fact from hostile eavesdroppers long enough for the plane to land.

In Pepys’ day they used nomenclators that were part code, part cipher. A box with a dot in it might mean “The Pope”. Two boxes one atop the other with a dot in the top might mean “The King” and a dot in the bottom might mean “The Duke of York”, and dots in both boxes “the King and Duke of York”.
Special meanings were assigned to Greek letters, and all of these had to have a spelling table to encipher words and phrases for which no symbols were provided.
You can see how Pepys would have trouble alphabetizing the list of entries. Does Theta come after T?
Where do you put this symbol that looks like a backwards R?

If the lists were reasonably short – one or two pages – you could use the same “character” for both enciphering and deciphering messages. If the lists were long, you needed a two-part character, one part with the symbols in order, and the second with the meanings in order.

Again there were problems with alphabetizing the list. Suppose one entry was for the phrase “His Most Christian Majesty King Louis of France”, would you put the entry under H for “His”, L for “Louis” or F for “France”?

These characters were hard to use, slow and cumbersome both to send, and to receive.
A trusted messenger with unenciphered text was often faster, and just as secure. They avoided transposition ciphers, in which the order of letters or words were scrambled, largely because almost any error would render the message gibberish from the point of the error to the end, and errors are very hard to avoid.

Steganography, which has seen a modern resurgence in use, was used in the 17th Century, but often as a one-time message. For example, if someone received a gift of two oranges, it might mean “burn your papers and get out of town, quick!” If the authorities intercepted the message, it would just be a snack.

I recommend David Kahn’s books if anyone is seriously interested in the subject.

Log in to post an annotation.

If you don't have an account, then register here.


Chart showing the number of references in each month of the diary’s entries.