Ever wonder what fractionation is?
I have.
It basically means taking the plaintext character and enciphering it with a cipher that generates a multi-character replacement for the symbol.
Sort of like X = YZ.
The value comes in being able to manipulate this ciphertext with another cipher method such as transposition. Move, separate and jumble Y and Z by some set method and it will be hard to recover the plaintext.
Fractionation is little more than a category of substitution ciphers, the ones that give you more ciphertext than plaintext. That’s it. It’s only really effective when used in combination with something else.
This is important for Kryptos because it is very possible that this sort of fractionation/cipher combination was used in K4. One drawback is that it implies that the final message of K4 is less than 97 letters which is short enough already. An argument could be made that K1-3 have used methods that allowed 1 character of ciphertext to decipher to 1 character of plaintext and that it is unlikely that Sanborn would change that. It’s also plausible to argue that it’s hard to make a message of 40-60 letters that will fit into the whole sculpture and provide impetus to solving the final message.
As a relatively common and widely used phenomenon, it would be foolhardy to not consider it and possible applications of fractionation that might have been used in K4.
I was wondering this same thing, actually. What if the pairs are related to the table in some way or the PT or CT of one of the previous sections?? The PRNG has bothered me for a long time. I think it tends to stand out to people who have been making love to Kryptos for a long time, which is what we’re doing.
I kicked it around as I went to get lunch and I was thinking that fractionation can cut both ways, if each ciphertext letter is a syllable then you could get a much larger message. Taking that thought and yours, you could make the argument that you can use the CT/PT as a field to draw the plaintext out of according to each letter of K4. I realize however that I am edging towards the fringe of what I think is likely with this.
What the heck is up with the PRNG anyways? It’s in K3 so it’s just transposed plaintext. No way it’s a coincidence to find it in K4.
This idea fits something that has been mentioned four times: The Navajo language. In fact, according to testimony, they very much considered using it for k4, but opted out in favor of staying with the English language. They said Navajo would be unfairly hard. What ways could they adapt this? They could use an OTP in collaboration with a prominent fiction writer — or just reuse some of the stuff already in Kryptos. I think that latter idea has been mentioned more than once by Sanborn too. Again, I hate to rely on their interviews. It’s as good as misdirection, but at the same time, they’re human, and any human has that child-like attitude, “Lookie what I did!” I’m merely saying that if we know their thought patterns and rejected ideas, it may lend some value in predicting what they could have done as an alternate. I have no problem believing Scheidt when he says that Navajo would have been unfairly difficult. It would ultimately be no better than an alphabet soup cipher. The pronunciation idea I like. It’s artistic, consistent with ciphers, and relatively immune to traditional computer attacks. I’m always reminded about that TELEFON booth in the Scheidt interview with Nova ScienceNow. I know it alludes to a spy movie, but I think there are subtle connections here and there that are meant to make us feel all the more silly once it’s solved. Like — it took you this long to get THAT?! *smug*
This wasn’t what I thought it was
Who says K4 has to be 97 letters? He only ever says that there are 97 letters left to solve.