A Kryptos Primer - From the eyes of a noob

I'm a KRYPTOS noob. I've always known about it and all - but always considered it kind of "out of my range" because its all CIA and stuff and has remained unsolved. Undoubtedly - many smart, smart cryptographers have worked on this right? Right. So many that this must be some crazy crypto that will need either a team of people or a savant to solve, right?

That's what I'm not so sure about. - And why I was inspired to write this little paper. After looking at the solves that have come and the methods that people have devised to solve, I believe that KRYPTOS might be a bit easier than people are thinking. I know... 'easy' when used to describe a puzzle is rather subjective, but if you played ATT, that is a good analogy. When we first cracked the ATT book, we all were basically stumped for 6 or so months. Think of the JBAIRDSP solve. How easy can you get besides not spelling it out on the paper? This eluded ALL of our eyes for way to long. Why? Because we weren't looking at what was in front of us. We were looking at what we had created in front of us. The idea of a puzzle that MUST be hard to solve right?

Maybe not so hard after all....

As has been mentioned here on the forums, most of the people so far that have attempted to solve KRYPTOS have backgrounds in cryptography. This is a natural skill to think is nessecary, knowing this is a CIA puzzle and seeing the first 3 ciphers. What I will attempt to do here is actually weaken the case that this is intense crypto and strengthen the case that this is actually a puzzle. I hope that by showing this, maybe I can persuade some of you - the ones who haven't taken this puzzle on - to take a look at it. We may not realize it but we (tweleve) have collectively become a very efficient puzzle cracker.

I think KRYPTOS is a puzzle and I think we can break it.

Let me give those who haven't really heard much about KRYPTOS a quick fact sheet then i will begin:

~You will hear people refer to K1, K2, K3 and K4. These are the 4 main text portions of KRYPTOS.

~K1, K2 and K3 have been solved leaving K4.

~Designed by James Sanborn.

~It is said that he is the only one with the "full" solution referred to as "philosophical". It is rumored that 1-2 other people actually know the text translation to K4.

~Dedicated in 1990.

~On the grounds of the CIA in Langley, Virginia

~Made of red granite, red and green slate, white quartz, petrified wood, lodestone and copper. Its basically a large art installation including landscaping.

~The 'main' portion of the puzzle is a large copper wall, that 'scrolls' out of a piece of petrified wood. This contains the text, cut out of the metal.


Now some important facts to support my case:

~James Sanborn has said that you do not need to be on the grounds to solve. This is important as it is VERY hard to make your way into the courtyard. Official business only and even then they generally don't allow pictures. The ones that you can find are few and far between.

~He also said it could be solved with text transcript alone.

~Sanborn expected K1, K2 and K3 to be solved within 3 weeks. It took 9 years.

~James Sanborn wasn't a cryptographer. He was trained in cryptography by Ed Scheidt who was the director at the time of the installation. This training lasted approximately 6 months.

~When Sanborn was installing the art, it was known it would contain a code. For this reason he had to be very very careful who he talked to and what he said. Some people were arrested for taking photos of the installation in creation.

~The last fact is important because Sanborn was so secretive - he even claims that Schiedt doesn't know the full solution. He kept it from everybody.

~Sanborn has said the installation was themed with "light and shadow" - themed to the CIA basically - secrecy - spies, etc.

~The K1 Vigniere Cipher keys were KRYPTOS and PALIMPSEST.

~The K2 Vigniere Cipher keys were KRYPTOS and ABSCISSA.

~The K3 was a simple Transposition Cipher.


Ok. Now that we have that out of the way let me see if I can better explain myself. Lots of these things are naturally my personal opinion so understand I don't profess this to be true. I am just letting you know what I'm thinking. I don't think its too illogical.

Think of being in Sanborn's shoes. He gets hired to do this installation, wants to do something with 'code' cause hes at the CIA, gets trained for a few months at best by the director, and sets to work. People start paying attention to him. He gets paranoid. Starts being secretive. He does the best he can coming up with the puzzles and this is what he finishes with:

1.) A Vigniere Cipher for K1. He makes the keys KRYPTOS (super obvious) and PALIMPSEST (a scroll that has been used, scratched off, and used again. - this imo is actually a genius keyword. Not only is it really really rarely used in literature and spoken word - it is a great theme for hidden messages etc.)

I believe this to be an easy puzzle. Some might disagree that PALIMSEST is such a rare word that it would take ages to try it against the cipher. I believe Sanborn knew he'd have the darn CIA working on his code--in other words some really smart folks. He would have had to assume someone would compile a word list of rare, themed words. PALIMPSEST would be on that list.

2.) A Vigniere Cipher for K2. (again? yep!) SAME METHOD. He makes the keys KRYPTOS (again? yep!) and ABSCISSA (another great key. Its meaning is 'cut off' which is imo the best possible word to use to describe how a spy must feel or how Sanborn was feeling while doing the installation.)

While i believe both PALIMPSEST and ABSCISSA are further used in the "philosophical" solve which has a greater difficulty (imo) , choosing them as keys is not that hard. Again a nice wordlist could surely include these words.

3.) A Transposition Cipher for K3. Check out the solve:


Group into equal rows of 24, rotate, make equal rows of 8, rotate = solved. This took NINE years. I know someone here would have nailed this in a matter of weeks if not days.

4.) The code in K4.

While this is admittedly harder than K1-3, (according to Sanborn) he expected it to be solved long ago. Why? Prolly cause its NOT THAT HARD.


Now there have been some things that have come to my attention from some others working on this - that really make me excited. Its all based on the one portion of the text in KRYPTOS that is different (raised) from the others:

You can find a thread here:


Sanborn has asked about this part - "Hasn't anyone figured this out yet? It's important."

Many people have tried. Many people have failed. I think that suelough may have suggested what is the intended solution. An anagram:


Best part is, this has never been discussed on the Yahoo group claiming to be 'experts' on the Sculpture.

While I will let suelough reveal all of her findings when she finishes her research, I suggest that it will cause many to take a second look at what is going on in this puzzle. To get you started Ill give 2 clues:

I told you how rare the word PALIMPSEST was.

Check some of the images posted to the KRYPTOS forum recently!

-Not only that but let me poke around K4 to further whet your appetite.

This is how it appears on the sculpture. Now this is obviously not English text as the letter frequencies aren't there. What stands out? I'd have to say the double letters. Let's try to group it into a block like K3:

* I should mention here there is a debate about whether K4 has 97 or 98 characters. The reason for this is a "?" symbol in between K3 and K4. It is not known if that ? goes with K3 or K4. For suggestions sake, Ill use it in K4, to make a nice divisible number of characters:98 )

This makes some really interesting patterns in the block of characters. Lots of nice little double letters horizontally, vertically and even diagonally. Here is a scan from my work:

In the secret Tweleve exclusive Elonka interview edit, we hear Elonka say someone who "knew the solve" said to check out "Duress Ciphers". While it is debatable that this is a red herring or whatever (considering the fact it was intended to support disinformation) - it would be interesting in this case if it did apply. Historically a "Duress Code" would be something a spy whose transmissions were being monitored would put in a coded message to let his superiors know he had been compromised. Double letters would do the trick? I've had a hard time coming up with a definition for a "Duress Cipher". The best I can come up with is a cipher spies used when they had no time to use a crypto machine. It would generally be based on a phrase - usually an easy to remember one, that would be used to key the message. Maybe a phrase from one of the books some are interested in?

Now I'm not saying this is anything at all. I'm just saying I'm having fun banging on it and I want you guys to try it too! You don't have to be a crypto nut to play around with it For all we know at this point, its a proprietary method of encryption (it almost surely is after all the cryptographers that have worked it) - in that case, we have an edge!

Now there are other things that are involved. Some people say there are ways to extract the keys from the puzzle. imo these are far fetched. There is even a paper on how to pull the word KRYPTOS from the line spacings of K3 - again way far out imo. There is as well a reason to think that Sanborn might have intended people to solve the anagram first, leading people to the key of K1 PALIMPSEST. This could very well be true. Regardless, this would be a rather simple way of pointing you in the right direction. Another example of something we could have nailed down years ago

There are many things to learn about KRYPTOS but I don't want anyone to get bogged down by its seemingly intense cryptographic nature! I believe you guys could really do some damage to this thing~ (which was Sanborn's whole point!)!


PS - Merry Christmas! :P

PSS - Thanks to Rusty for teaching me a lot of the little tidbits of KRYPTOS info that helped lead me to writing this. Thanks as well to sue for being a fun puzzling partner (and for being wicked smart!)