My thoughts on the Z340 Cipher
I'll be honest, I have little knowledge of the Zodiac murder case. The closest I have come was when I watched a film about it, fictionalised by Hollywood to the usual point where 'based on a true story' is a stretch. Despite the fact that the case took place long ago, before I was born and in another country, it really doesn't interest me to think about grisly things and horrid people. I spend most of my time either writing code or on my various other hobbies. I don't need to think about murder cases or events that to be honest terrify me and keep me awake at night.
The idea of encrypting messages always intrigued me, and delighted me as a child. The simple replacement ciphers I used to send notes in in school to my friends I have no doubt could be easily broken. Fast forward to today, and it still does delight me. I don't have a use for it anymore, as I do not have any secrets to keep! But I do like to look at other peoples, either very complex or very simple. And of course the holy grail of vintage encryption is the famous Z340 cipher.
It's the 1960s in San Francisco. A killer is on the loose, and taunting the police with letters sent to them about his crimes. Not unlike Jack the Ripper, I imagine what was going on in his head made sense to him. Letter after letter arrived, the killer leaving tantalising clues as to his identity. He then did something strange. Began to send Ciphers. The first, was a simple replacement cipher that was broken relatively quickly given the tools we had back then.
The second was not so easy. In fact, it has baffled scientists for decades. Various people have claimed over the years to have found solutions, but on inspection none of these can be verified as correct.
The Z340 cipher, can't be thought of as a clue in a murder case. In fact the people who have claimed to have broken it are all very much amateur crime fighters. Knowing too much about the case can mean that we see things that are not there. I can read anything I like from the cipher, and provide a key for it. That's not a breakthrough, that is me looking for an answer, by thinking I know the answer to begin with.
If we are going to break this we must first ask three key questions-
Is the solution in English?
Will the solution make sense?
Is there a solution?
The first question can be assumed as yes. There is no reason to think it is not, but you never know. The second and third questions have the same answer. This person was deeply unhinged, and may have believed the solution made sense but in fact to the more clear thinking is total nonsense. A deeply unhinged person may also send a cipher that is random and has no solution, simply to have people like me looking at it and wondering how clever they are.
How to begin with a solution? Let's start by building on what we know- after all this cipher has been around for a long time. We know that the original cipher was cracked, and turned out to be a fairly simple replacement cipher. Let's assume this is not the case, as I think we would know by now given its notoriety and the countless analysis hours that have been put into it. Also, technology has come on a fair bit since 1969, with the invention of computers and software that can crack complex ciphers.
We know that it was written by hand so we should be able to assume that there is a key, or more likely keys. The best way to complicate a handwritten cipher is to make it so the cipher must be reordered before it can be broken. I doubt it is a simple left to right read, as if that was the case we would be able to spot the pattern with the eye, and let's face it after decades of research we would know.
It is clear to me that the sequence needs to be reordered. Did the writer leave a clue? Maybe. When I look at this there is a very obvious pattern, nearly bang in the middle.
Let's look at line 10 of 20.
Is it me or does that look like it breaks the cipher into quarters? the + being the divider, with the two - at the end. I think that maybe significant.
We know from previous ciphers from the same author that the actual symbols may be meaningless, in fact we know they are. They are taken from various media, some from our own alphabet. That being said, we are making an assumption there. The only way to really be sure, is to find the answer!
Before we begin to write our program, like any program we need to do some analysis. That is going to be line by line. A laborious task as it is not only line read left to right, but also
- Right to left
- Down
- Up
- Each of its diagonals
Each of the above, if our theory is correct regarding line ten will have to be repeated for each quadrant, in each order.
The next think we need to do is to somehow digitise the cipher, so we can programatically read it as a string. Then we can begin to divide it up better. Below is a digital representation of the cipher.
Good old Excel, where would we be without you! I have highlighted in yellow the line that may or may not be the breaking line. Here is the representative key-
Now we can begin to reorder, and try to interpret the cipher using code. I have only begun to try this. I have begun by reading in the digitised version, and will attempt to recursivley manipulate the data until it makes sense. I'll keep you all posted!!
