Ouroboros is the archetypical snake eating his tail. It’s a metaphor for infinity, recreation, cycles of birth-death-rebirth, etc.
It’s also a nice metaphor for a reason why people fail to solve the fourth part of Kryptos. It’s not even that obscure. In brief, if you use the 97 letters of K4 to solve K4 then you have fallen victim to petitio principii, also known as begging the question, circular argument, circular reasoning, or circulus in probando. We’ve covered logical fallacies in Part 1 but this deserves it’s own section. Why? Because people do it all the time and it’s a waste.
Begging the question in an argument involves setting up the conclusion as one of the premises. For us, it’s not a perfect example of circular arguments but the concept works well enough. I also can’t very well just say you’re wrong and we all move on. I need sound reasoning to support my point of view.
It’s not impossible that when the four sections of Copperplate ciphertext have been translated that there is some meta-puzzle element that requires us to utilize clues/characteristics of the text. It’s rather impossible to argue that K4 is to be left as is and an over-arching Kryptos solution can be obtainable despite the fact that we don’t know what K4 says.
The meat of the problem is that folks are willing to include some aspect of using the 97 letters of K4 in part of a process whose end result is a translation attempt of K4 itself.
It’s the same thing as defining a word by using the word.
Example: Slow can be defined as an adjective describing something slow.
It just doesn’t work. Yeah, you can do it but it’s rather useless in the end.
Stop using K4 to decipher K4.
This is not to say that you don’t take those 97 letters and apply a decryption method that will replace them with 97 plaintext letters just don’t include them in the decryption process as anything other than input.
A related issue that is of the same cyclical theme is when folks use the decryption methods for K1/2/3/Morse Code on K4. This just doesn’t even make sense, especially when you start custom fitting the method. If you could use the vigeneres then K4 would have been plaintext at the end. K3’s transposition was the same. The only reason that we know there are four sections to the Copperplate is because the original solvers could only get part of it translated at a time and then had to work with the assumption that a different ciphering method was used for the rest. It’s not like they’re clearly delineated in the sculpture, it was more of a “Oh crap, it doesn’t work for the whole thing…”
There are so many cipher methods besides a vigenere and the specific transposition(s) used in K3 that it’s even peculiar to imagine Sanborn using the Vigenere twice. I’m sure he had his reasons, perhaps the idea of having to cut 800+ letters for one cipher wasn’t appealing. We’ve got plenty of traditional options left to try. I would personally petition to have the thing knocked to the ground and melted if it used that few of cipher methods.
The problem with Ouroboros is that it never ends and your search for a solution will never end unless you avoid this pitfall.