Cracking a code in a pre-modern setting

Try The Code Book: The Science of Secrecy from Ancient Egypt to Quantum Cryptography by Simon Singh. A nice intro to the history of codes.

Layla Nahar said:

In my WIP I have three students trying to figure out a mystery. The setting is something like 16th century France. One of the students talks to an older student, something like a post-doc, and the post-doc finds some mysterious entries in some 500 year old ducal chronicles. (fwiw - I've had her find four chronicle entries that were made up of ten-word sentences, distributed in an intriguing way in the months of one year.) The post-doc is going to propose they breakup the task of trying to figure out what the code is. One of the three is good at classic languages, the other two are not good with the languages, but they are merchants and are good at calculations. I want to have the post-doc split up the work of solving the code. (I am open to changing the 'ten-word sentences')

I don't know much about this kind of code stuff - I have to admit I'm intimidated by it because it always makes me feel dumb when I think about it.

Does anybody have any ideas about how the 'post-doc' might split up such work? And/or any ideas about how people might hide/encode things in a pre-modern setting?

Click to expand...

Substitution codes were popular; John Dee used code routinely, as did Queen Elizabeth, and her "star chamber"; so did John Donne and Mary Queen of Scots.

http://cryptiana.web.fc2.com/code/elizabeth.htm

There are some things you're going to need in this scenario -- First off, is this a code or a cipher?

Codes are inherently harder to break then ciphers which is why most diplomats spend a great deal of time trying to "acquire" code books from other embassies/consulates/ships/etc. Some codes have alphabetic entries to spell out uncommon words (like towns/names) while others will send a null code group like 1111 or 9999 to indicate start/stop using the first letter of the code word to spell and go back to using the actual code meaning.

If it's a cypher (or cipher . . . different people spell it different ways), then you need a sufficient amount of enciphered text to begin an attack. Four ten-word sentences, unless there's a significant amount of overlap, is unlikely to be sufficient unless it's an extremely simple code. As a former cryptanalyst, I could buy four enciphered paragraphs (say, 100 words per), as that would allow for possible duplication of words or phrases -- this is usually the most direct way to break into a cipher. Again, the more complex the cipher, the more recoveries using the exact same system are required to get the first break which will allow your cryptanalysts to have a chance to recover not only the message but the system being used. After that, it gets down to boring stuff like letter frequency use, identifying simple words, etc. (and this is assuming you know what the original language of the cipher is to begin with). If each of the four passages is set with a different key, (1st being A=R, then A=Z, etc.), then I'm not sure even a paragraph is going to be sufficient.

Sorry if I'm getting too far down in the weeds, but if you're wanting to make this fairly realistic, you'll need to keep things like that in mind. If you want more help, I'd be happy to assist.

Many letter substitution codes have a key document, often something common like The Bible or a specific document like the Magna Carta. Finding the key could be an intriguing part of the story.

Jeff

I could be completely wrong here but I wonder if this issue could be solved backwards. In other words, instead of creating a code and then putting the message into that code, start at the end.

What is it you want the codebreakers to discover? What is the solution? What is the message that has been encoded?

Then see if that message can be twisted and turned backwards till it gets lost.

As I said at the beginning- I don't know, I'm just thinking about how I might approach the same thing myself, were it my story problem.

Good luck.