Given that several living people know exactly what it says, I don’t think it would be entirely right to classify the fourth (unbroken) cipher on the “Kryptos” sculptures outside the CIA building in Langley VA as a “cipher mystery”. From my point of view, it’s not so much a case of “what does it say?” as “when we will find out?”
All the same, here’s a link to a New York Times news item from late 2010 you might possibly have missed, telling how its sculptor Jim Sanborn is (basically) getting tired of waiting for Elonka Dunin to crack its message. 🙂 As a result, he’s now released a few letters from K4’s plaintext: BERLIN. Incidentally, I found the story via Slashdot, whose comment cup brimmeth over – as ever – with Ovaltine references (such as “The mug is round. The jar is round. They should call it Roundtine”, etc).
Regardless, the Kryptos cipher sculpture has four parts: K1 and K2 are both polyalphabetic Vigenere ciphers (keywords: “Kryptos, Palimpsest” and “Kryptos, Abscissa” respectively), while K3 is a transposition cipher. Sanborn has said that the first three parts hold clues to K4: e.g. because K3 is a (slightly wobbly) version of Howard Carter’s account of the opening of Tutankhamun’s tomb, it seems fairly likely to me that the sentence that originally followed it (“wonderful things” or “yes, it is wonderful”) is connected with K3, perhaps as a keyword.
One of the things making K4 tricky to break is that it is a meagre 97 characters long: here it is with the revealed partial plaintext next to it (letters 64 to 69) that Sanborn has given us:-
...........................OBKR ...........................----
UOXOGHULBSOLIFBBWFLRVQQPRNGKSSO -------------------------------
TWTQSJQSSEKZZWATJKLUDIAWINFBNYP ----------------------------BER
VTTMZFPKWGDKZXTJCDIGKUHUAUEKCAR LIN----------------------------
What, you may ask, do I make of all this? Well, the implication would seem to be that this is neither a transposition cipher (such as K3) nor indeed (if there is some kind of progressivist increment from K1 and K2) a straightforward polyalpha. Furthermore, given that Sanborn appears to be expressing some kind of impatience, it seems slightly unlikely to me that there is a further (as yet unrevealed) part of the cipher installed in the CIA’s grounds.
Of course, it’s not a monoalpha (IN –> TT): and though 6 doubled letters (BB QQ SS SS ZZ TT) is slightly above chance for a 97 character text, that’s not hellishly improbable. Really, it does looks a lot like a keyworded Vigenere polyalpha once again… but it apparently is not. So what is it?
I don’t know: but if you think this is something you’d like to try cracking, there’s a (remarkably active) Kryptos newsgroup on Yahoo that would seem to be just perfect for you. Good luck and happy hunting!