Frequency Analysis of Possible Secondary Kryptos Ciphertexts

Thank you Simon Singh.  His Black Chamber is frequently useful for Kryptos efforts.

The previous posts generated some possible “secondary” ciphertexts after an attempt to remove an additive/subtractive superencipherment.  These are the frequency analysis histograms generated.

Secondary #1 (3 Q’s but no U’s?)

Secondary #2 (Too many z’s and p’s to be plaintext?)

Secondary #3 (5 Q’s and only 2 U’s?)

Secondary #4 (could be, it really could be.  Probably not because only 2 E’s and 4 Z’s)

Secondary #5 (4 Q’s and only 1 U)

Secondary #6 (6 Z’s?  Unless it has “zany crazy razzmatazz” then it’s not a message I want to read)

Secondary #7 (Maybe?  It’s a Scrabbler’s nightmare with 2 Q’s, 4 X’s and 4 Z’s)

Secondary #8 (4 Z’s but only 1 E?  hard to believe in plaintext frequencies)

So by even the simplest analysis of Q’s to U’s, some of these can be excluded.  Others are maybe but Z’s were all over most of these possible secondary ciphertexts of Kryptos.

Excluded as really unlikely: 1, 2, 3, 5, 6, 8

Maybe but probably not: 4, 7

So not to be too picky but I’m not super convinced by any of them.  The premise was that if you removed the additive then you’d have regular English frequencies.  The justification for this is that Scheidt said that they are masked and that once you remove the masking, you can begin work on the underlying cipher.  It’s possible that any of them could be true but not any more true than many other efforts made by any Kryptos fan.  My verdict is that none of them were close enough to plaintext English letter frequencies to make a convincing case towards going further.  Yeah, those letter frequencies are average but if yours aren’t even close then it’s time to move along.

I may have tried the wrong keywords or orientation.  I still like the idea of 7×14 or two 7×7 grids.  I just may be a little wrong on the masking technique still.

-Kryptosfan