A null cipher, also known as concealment cipher, is an ancient form of encryption where the plaintext is mixed with a large amount of non-cipher material. […]
In classical cryptography, a null is intended to confuse the cryptanalyst. In a null cipher, the plaintext is included within the ciphertext and one needs to discard certain characters in order to decrypt the message. Most characters in such a cryptogram are nulls, only some are significant, and some others can be used as pointers to the significant ones.
https://en.wikipedia.org/wiki/Null_cipher
My hypothesis here is that z340 is a null cipher where plaintext is hidden in plain sight as a set of meandering phrases, each delimited at the beginning and end by a pair of control characters. Phrases are composed of symbols meant to be interpreted literally, as the letter they are or resemble.
In the following solution for this hypothetical scenario, you will find:
– 9 coherent phrases (numbered 1-9), mutually exclusive and non-intersecting, composed of 170 plaintext characters (exactly 50% of the total cipher).
– A clear explanation of the presence of the corrected "K" symbol (see phrase #2).
– A partial explanation of the presence of pivots.
– A partial explanation of the presence of box corners.
In the edited z340 below, I use certain colors as follow:
– Green: phrase #1, where the nature of the cipher is described.
– Blue and Red: all other phrases.
– Yellow: control symbol pairs, marking the beginning or end of a phrase.
Here are all the 9 symbol phrases found, accompanied by their corresponding textual equivalent (I added minimal punctuation in light grey for clarity):
Here are some thoughts, numbered according to the corresponding phrase:
#2. I find this phrase of particular interest as it corroborates the nature of the cipher by explaining the presence of the corrected "K" symbol in the cipher.
#3. They are indeed 3 victims by gun.
#4. IDN = Identification. 9mm and .22, the two types of gun calibers used by zodiac AFAIK.
#7. Robert Colby: http://www.mysteryfile.com/Colby/Tribute.html. An author of the zodiac’s era. He coincidentally wrote a book partly on zodiac in 1973.
R Tol = R Tolkien??
#8. Seems to anticipate what a future book on him would look like.
As for pivots and box corners, they are all partially formed by either a control symbol pair or by the looped structure of the phrases.
Thoughts?
_pi
Hey _pi,
My thoughts are that a solution should be verifiable.
If you apply the 408 key to the 340 it actually says "NULL" in the middle of the cipher:
TEGILKIBIPTETOACE ELELOMNNNEUSRTSHY LLDAHEIEAEOYTMSTF ALLOIKATBDLNEEGSC RRHESN?ET?SIE??P? LAGISKN?TEVPSICNO M?ESFN?CIAOABSSTT OACFH?MNEREUSE?TR ERHE?EEGCSLALW??S ?SKIBEIFENLORSLL? IEG??NETNULL?OHSN <--- CRAKGFTT?O?HSE?LS S?RALMEET?SLAO?AG KASEIH?C??SAB?NEE LLOETD?RVEINIE??E TASDSLSONLISHHFAC GAOETE?VRESKELTYT EEYEVSEAINATO?OYL TSPEERONLRUNLT?VA IHNTELPASENAWTS?E
One of the things I like about this idea is that it would give the author enough flexibility to hide little easter eggs and false leads in the cipher, like the "signature" at the bottom, and many of the other little idiosyncrasies that have been pored over so much.
If you apply the 408 key to the 340 it actually says "NULL" in the middle of the cipher:
That’s an interesting quirk! Are you certain of this result? I get something a bit different (NUL) when applying the 408 key to the 340 with zkdecrypto.
One of the things I like about this idea is that it would give the author enough flexibility to hide little easter eggs and false leads in the cipher
Totally agree.
Although I like such type of ‘encryption’: Pretty sure it’s a dead end.
QT
*ZODIACHRONOLOGY*
If you apply the 408 key to the 340 it actually says "NULL" in the middle of the cipher:
That’s an interesting quirk! Are you certain of this result? I get something a bit different (NUL) when applying the 408 key to the 340 with zkdecrypto.
Well I am not sure now. Largo generated it a while ago.
Very intriguing work _pi
Sentence 1 and 2 are very compelling, both for their clarity and their self-referential nature to the cipher! I also like that this scheme doesn’t require any quadrant flipping or other complex rearrangements.
I’d be curious to know what doranchak (or someone else who has thought a lot about the period data) thinks of this. Is there a way to quantify how much this explains the P19 data?
Now that there’s no flipping or quadrant shifts, I could imagine an explanation for P19 being as simple as "the z340 is 17 characters wide, so bigram shoot up at P19 because that happens to mix diagonally-linked characters from nearby lines more fruitfully than other periods". It might still be hard to explain why P18, P16, and other adjacency-mixing periods don’t show as much of a peak, but I’m encouraged enough by this result that I would be curious to see some hard analysis of this.
Sentence 3 looks pretty compelling too. But some of the others seem to get progressively less plausible IMHO. Sentence 8 in particular is quite an outlier. I struggle to believe that "book lid look hot" was in anyone’s lexicon in 1960 (and may only just be in Lil’ Wayne’s today). This comment of mine is quite possibly the first instance of "book lid look hot" being written anywhere on the internet. Google’s 4-gram data doesn’t have an entry for "book lid look hot" (or even any 4-gram beginning with "book lid") so it’s an exceedingly unique phrase. Given the subjective nature of the loop creation and the letter alignment, I would continue looking for a more plausible phrase here.
And I’m curious how much your rotational key allows in terms of degrees of freedom. It doesn’t strike me as so extreme that we’re at Bible Code territory here (where anything can be found), but from an information theory perspective, it would be interesting to know how easily this technique can create spurious messages. Would we expect to find messages this intelligible in random sets of data 340 char long and arranged this way? I assume the answer is no, but I’d want some more thought put into this. Like, is this 20x more intelligible sentences of this form than normal or only 2x?
I believe it’s a credit to your proposed solution that there’s low overlap between messages and that it covers so much of the cipher.
It might also be interesting to know if there’s a bias in which symbols go unused. Perhaps even if there’s not another message in them, they may point to another parsimonious pattern that this theory could hang its hat on.
And I don’t know the status of the grails that were mailed to SFPD. Do most people think those were real or a hoax at this point? Do they line up with any of the patterns you’ve found? Or are there other circumstantial clues from other known zodiac writing post-z340 to point towards any of this? Are all the repeated Cs on the Halloween card a taunt, showing the control characters in the z340? I’m not saying there has to be a pattern in the loops, but it would be more convincing if there were alignment with the grails or there were at least some typical loop patterns that repeat or have patterns (since even people trying to do "random" things often exhibit a bias of some sort). If the loops are fully random, I’d be more concerned that we were finding a message where none exists.
I can’t find the "center of letter bubble" image that someone made that eliminates all the z340 letters but shows the offset of each letter and where it’s centered. They aren’t as perfectly aligned in columns and rows as they appear at first glance. Is there any evidence from that diagram that perhaps the letters were written out-of-order in any of the loop patterns you’ve identified? Or that maybe the code letters and nulls display a bias on how well-centered they are?
It also has to be noted that this decipherment is so straightforward that it could have potentially been done by hand at any point in the past 50 years by almost anyone (including gifted 12 year old amateurs). So people will be more skeptical of it both initially, and probably indefinitely, even if it is essentially right. It’s sort of the maximal "f*** you" solution to any expert who has spent any amount of time studying the z340.
But I will say that while I’m still skeptical that this solution is completely right (and it’s possibly entirely spurious), it’s still the best original proposed solution to the z340 I’ve seen.
Keep up the great work _pi!
Cheers,
beijinghouse
I’d be curious to know what doranchak (or someone else who has thought a lot about the period data) thinks of this. Is there a way to quantify how much this explains the P19 data?
I see no correlation. P19 in the 340 is either a sign of a more specific kind of transposition or an outlier.
I’d be curious to know what doranchak (or someone else who has thought a lot about the period data) thinks of this. Is there a way to quantify how much this explains the P19 data?
I see no correlation. P19 in the 340 is either a sign of a more specific kind of transposition or an outlier.
These kinds of word finds using ambiguous interpretations of symbols in arbitrary reading directions can be done even on randomized ciphers. You could pull meaningful phrases out of anything this way.
A long time ago I made a word search gadget to explore this idea:
http://www.zodiackillerciphers.com/?p=420
Pi, I think your approach is interesting and clever, but the solutions you produce will be impossible to discern from randomness.
I’d be curious to know what doranchak (or someone else who has thought a lot about the period data) thinks of this. Is there a way to quantify how much this explains the P19 data?
Now that there’s no flipping or quadrant shifts, I could imagine an explanation for P19 being as simple as "the z340 is 17 characters wide, so bigram shoot up at P19 because that happens to mix diagonally-linked characters from nearby lines more fruitfully than other periods". It might still be hard to explain why P18, P16, and other adjacency-mixing periods don’t show as much of a peak, but I’m encouraged enough by this result that I would be curious to see some hard analysis of this.
We were unable to match P19 stats with a knight’s tour style inscription before substitution, which takes the path of the plaintext in the direction of P19, but also other periods as well. So, this is similar because the path takes multiple directions. Most of the path of the plaintext in a transposition + substitution has to be in the direction of P19 to make P19 stats. About 2/3 of the message at least, if not more like 3/4.
Zodiac wrote of collecting slaves in paradice. This has occult undertones to it. Here is some ancient and magical symbols used by witches and such.
I just figured out how to upload this picture. Ive had it for weeks. I tried many times and failed but I tried again this morning and I got lucky.
Who is Bob, L Fay and S Lee?
Bob Graysmith?
The sentence could be "I kill boys Lee, big on toll". Possibly an ALA reference.
