Are the ciphers gibberish?

Someone on Quora asked, “How do people know that the Zodiac Killer’s cyphers aren’t just gibberish that have no real meaning?” I responded with the following article which I also posted on Quora:

There are four total ciphers: Z408, Z340, Z13 and Z32. Of the four, only Z408 has been solved. Z13 and Z32 are too short, so verifying solutions would rely on extraordinary evidence such as discovering the keys from Zodiac himself, or by finally catching and questioning him if he’s still alive.

As for Z340, there are a few bits of evidence that could be put in the “it’s gibberish” column:

  1. A non-gibberish cipher would probably have been solved by now.
  2. Zodiac liked to mess with people (especially to make law enforcement look bad), so making gibberish ciphers is right up his alley.

But there’s really no good test to conclude once and for all that the ciphers are gibberish, apart from an admission from Zodiac himself. You could easily create a gibberish plaintext yourself and make a cipher from it, and it could be the next famous unsolved cipher.

On the other hand, there have been some interesting papers on the topic of the Beale ciphers that present some evidence that they are hoaxed:

  2. Cryptanalysis of Beale Cipher Number Two

But apparently they are not enough of a “smoking gun” to fully convince everyone of the hoax hypothesis for the Beale ciphers.

It is possible that there is some characteristic of Z340 that could be discovered, and would only be present if the message was gibberish. I’m really curious to know if such characteristics could be discovered from homemade gibberish ciphers made to resemble Z340.

Nevertheless, there are still many things that can be put in the “it’s not gibberish” columns:

  1. Z340 uses an encoding scheme that hasn’t been fully tested yet. Maybe he used a traditional type of pen and paper cipher that hasn’t yet been exhaustively explored. For example, there are so many ways to rearrange the plaintext before assigning symbols. Or maybe he invented his own scheme that no one has guessed yet. When Z408 was cracked, the papers talked about how the solvers took advantage of patterns found in the cipher text. Zodiac could have focused on somehow removing those patterns from Z340. Or maybe he found some obscure idea in a book or one of those pulp fiction detective stories from back in those days (they were popular and often included ciphers)
  2. Zodiac was sloppy in making Z408 (details here), and thus may have made too many mistakes in Z340, which makes it more difficult (or impossible) to solve. (I suppose this could have effectively turned it into gibberish)
  3. There are clues in Z340 that have significance to cryptanalysts. Many of them are detailed here: Encyclopedia of observations For example, would a gibberish cipher really have the periodic ngram bias described there?
  4. On the 6th row the cipher text, he crossed out some symbol and wrote a backwards K above it:

    If the message is gibberish, why did he bother to correct the mistake? Was it just to give the false appearance of a real message? Or was he truly concerned about the integrity of the cipher text?

So, I think gibberish is a real possibility, but I still believe there might be a real message in there, especially since we still haven’t exhausted all the codebreaking possibilities.

On the other hand, if another 50 years goes by without a solution… 🙂

(Note: here’s a similar article I wrote a long time ago about this same topic: Cipher Legitimacy)