Suppose that Zodiac was an amateur. Where did he came up with using homophones sequentially? Self-invented? Some book? Can this be at least an open question because no one besides me seems to be asking it. I wish we had a "book guy".
is it possible that he did that to make it easier to crack the cipher? as a proven amateur i would have attempted to "randomize" everything i was doing to make the cipher harder, unless i was hoping it would eventually be solved. perhaps the 408 was solved too quickly for his liking, but is there a way to recreate his thought process in the 408 and examine what steps only an amateur would make vs what steps even an amateur would avoid?
That’s another question on mind, why did he use the homophones sequentially? Could have been something he picked up somewhere, but would this encoding be in a detective story? I am not so sure about that. Here’s another question, when was it first realized that the homophones were arranged sequentially? Or was that not a thing back then?
Here’s another question, when was it first realized that the homophones were arranged sequentially? Or was that not a thing back then?
Interesting question…
The FBI files, where they talk about things they tried on the 340, mention "cyclic use of variants", presumably because they knew of their use in the 408 and thus knew to look for them in the 340:
That was around November of 1969.
As for homophonic ciphers in general, King and Bahler ( http://www.oranchak.com/king-homophonic-ciphers.pdf ) note that sequential homophonic ciphers have been around since the 1500s.
it’s trivial to create an unsolvable cipher. if he wanted to play a trick with the 408 he could have done that. instead he chose a substitution cipher, solvable but somewhat more complicated than what you’d consider a simple cipher (caesar shift, etc). he added a second layer by obscuring the character count with homophonic substitution. solving it doesn’t require external tools or complicated equations, it’s simply a puzzle or game. solving it reveals a pointless message – ‘im a bad guy, look how scary i am’. there was no risk for him, it doesn’t reveal anything.
the question is – which part of the above changed for the 340? did he add a damning message that caused him to take more care that it wasn’t solved? did he decide to play a trick and make an unsolvable or meaningless cipher? did he add an extra layer that keeps it solvable but we haven’t figured out yet? do we need extra tools or external clues to solve it?
i don’t see his motivation or risk aversion changing. he wants to create a puzzle game because it causes people to pay attention to him. he doesn’t want to reveal anything about himself so i doubt the message is any more revealing than what we got with the 408. it’s a pointless endeavor if it’s unsolvable so i don’t think it’s garbled noise or an attempt to trick us into foolishly working on it for decades.
it’s either an extra step that he took (maybe he read chapter 2 of his cryptography book of choice) or it requires an external tool or clue. in his other writings he was moving towards that – the map and radians stuff, the halloween card with its puzzling attributes, the bus bomb diagram – using his writings plus something external to solve a puzzle.
jarlve has a thread about higher-order homophonic ciphers. fascinating stuff but there’s no way he did something this complicated. he’s erratic and impatient. he did something simple either by leaving a clue he thought was obvious or by adding a simple step that we haven’t figured out yet. you guys have had some fascinating ideas through the years but i think it would be fruitful to spend a moment thinking about the simplest step an amateur could take to make a solvable cipher hard to decode. what is something in chapter 2 of a beginner’s guide to cryptography that would mess up azdecrypt and all of these fantastic tools?
on the other hand i think those of us who don’t understand a lot of this could undertake an effort to review his letters and think outside the box at the question of ‘if he is leaving us a clue, where is it?’ – not complicated gematric codes or vague references to a poem from the middle ages, or multi-step jaunts through the phone book, but a linear and literal clue in one of his writings – the first one that comes to mind is the dripping pen card/letter that accompanied the cipher. he sent the halloween card with the accompanying paradice/slaves/by rope, knife, gun, fire nearly a year later so that may have been an additional clue since it hadn’t been solved.
i’d bet any one of you a beer that the answer is right there – simple step or simple clue. my 2 cents.
Here’s another question, when was it first realized that the homophones were arranged sequentially? Or was that not a thing back then?
Interesting question…
The FBI files, where they talk about things they tried on the 340, mention "cyclic use of variants", presumably because they knew of their use in the 408 and thus knew to look for them in the 340:
Great find, thanks allot.
i think it would be fruitful to spend a moment thinking about the simplest step an amateur could take to make a solvable cipher hard to decode. what is something in chapter 2 of a beginner’s guide to cryptography that would mess up azdecrypt and all of these fantastic tools?
i’d bet any one of you a beer that the answer is right there – simple step or simple clue. my 2 cents.
Can you find an example? Perhaps we are not thinking simple enough.
Nulls could be a simple explanation. And given enough nulls it becomes very difficult. Another example that comes to mind is the W.B. Tyler 2 cipher, a sequantial homophonic, that went unsolved for some 160 years because of the many errors littering the cipher text.
maybe some kind of mirroring effect like what doranchak and cragle are seeing in this thread. fold the cipher in a certain way and then mirror certain characters (backwards "K" becomes "K" for example). it’s really simple to do but incredibly difficult to figure out, which was his bread and butter. i am currently reading several basic cryptography books looking, as an amateur, for something that would be an "a ha!" moment for me were i pursuing a career in criminal cryptography.
As for homophonic ciphers in general, King and Bahler ( http://www.oranchak.com/king-homophonic-ciphers.pdf ) note that sequential homophonic ciphers have been around since the 1500s.
https://www.slideshare.net/karwanmst/cr … y-41894294
In this slide, authored by Karwan Mustafa Kareem, the homophonic cipher is also dated back to 1500. And is referred to as the "Mantua Homophonic Cipher", googling that does not yield much.
Mantua is a city and comune in Lombardy, Italy, and capital of the province of the same name.
Homophonic Cipher. In this cipher technique, plaintext letters map to more than one cipher text symbol. Usually, the highest-frequency plaintext symbols are given more equivalents than lower frequency letters. In this way, the frequency distribution is flattened, making analysis more difficult. Since more than 26 characters will be required in the cipher text alphabet, various solutions are employed to invent larger alphabets such as a numeric substitution, uppercase, lowercase, upside down, and fanciful symbols etc. A Mantua Homophonic Cipher (15th century, Roman Empire) is an example to this type of cipher.
Them renaissance Italian Dukes were mad for it.
Duke of Mantua:
http://cryptiana.web.fc2.com/code/mantua.htm
http://ciphermysteries.com/2016/07/06/f … yptography
Duke of Parma:
https://www.tandfonline.com/doi/abs/10. … 17.1370038
Seems they also liked adding nomenclatures to their ciphers as well.
In this slide, authored by Karwan Mustafa Kareem, the homophonic cipher is also dated back to 1500. And is referred to as the "Mantua Homophonic Cipher", googling that does not yield much.
Kahn includes an image of it in "The Codebreakers". It split up vowels into multiple symbols:
Thanks doranchak, 1401!
