“The scrambled text as its own key, on tables rebuilt from the solved messages' words”
Experiment 25 tested the cipher that feeds on its own output on tables rebuilt from the puzzle's vocabulary. Experiment 62 then showed this recipe is the one that makes the copper's width-21 mark. Here the tables come from every word of the solved messages and Sanborn's 2025 letter, at every distance, with scrambled coppers as the control.
In this cipher each letter is hidden using the hidden letter some fixed distance earlier, so once it starts it needs no secret word at all. Each of the 24 known lettersthe known letters Sanborn has confirmed that positions 22 to 34 of the message read EASTNORTHEAST and positions 64 to 74 read BERLINCLOCK. then tells us exactly which carved letter must sit that distance before it. We check every distance from 1 to 95 on tables rebuilt from every word of the solved messages, and count how close any of them comes, compared with scrambled copies of the copper.
Words in the margin are defined on the glossary page.
1.1
The method at work
The carved letters go in at the top; the reading comes out at the bottom.
the autokey
In a ciphertext autokey the key at each place is the carved letter a fixed distance before it, so once the start is past, the copper itself is the key.
the check
Drawn for distance 21 with SHADING for the message and INVISIBLE for the copper: place 22 needs key R, and the carved letter 21 places back is O.
the bound
All 102 x 102 tables, four key alphabets, three modes, distances 1 to 95: never fewer than 3 misses, while shuffled coppers searched the same way reach 2.
2
Best candidate against the known letters
This experiment produced no candidate: the search ruled everything out before a full text was built. Its product is the bounded negative in section 7.
3
Walkthrough, by hand
Redo the method on a few letters with pencil and paper.
Take SHADING for the message alphabet, INVISIBLE for the carved alphabet and KRYPTOS for the key alphabet, with Vigenere.
Place 22 holds E (9 in SHADING) under carved F (10 in INVISIBLE), so the key letter there is the one at index 1 of KRYPTOS: R.
At distance 21 the key letter is the carved letter at place 1, which is O, not R, so place 22 fails at lag 21.
At distance 10 the key letter would be the carved letter at place 12, L, which also fails.
Count the failures over all known places past the starter, for every distance and table.
the worksheet: the first steps by hand, on squared paper
4
Worked example, generated
The same steps run by the code on the real letters.
place plain carved key letter needed (KRYPTOS) carved 21 back
22 E F R O
23 A L P B
24 S R N K
25 T V C R
5
Hypothesis, for a cryptanalyst
Family, key space, the constraint under test, and the clues it reads.
Ciphertext autokey, key letter = carved letter L back read in the key alphabet, lags 1-95, plain and cipher alphabets from the 102-alphabet pool of 0053 (K1-K3 plaintext words, C13/C14 words, standard, KRYPTOS), key alphabet standard, KRYPTOS, the plain alphabet or the cipher alphabet, three modes. Each crib at place i > L requires a specific key letter, hence a specific carved letter at i - L; the count of satisfied cribs per lag is exact and needs no primer. Motivation: 0062 found ciphertext autokey at lag 21 the only generator that raises the width-21 repeats (0056). Report the fewest mismatches over all settings against ten shuffled coppers. Bounds: complete for this pool.
6
Prediction
Written before the run; the result decides it.
No setting satisfies every crib past the starter; the fewest mismatches is about the same on K4 as on the shuffles (several). Zero candidates.
7
Method
Build the alphabet pool.
For each plain, cipher and key alphabet and mode, find the carved letter each crib requires L places back.
For each lag, count satisfied and violated cribs using the positions of each letter on the copper.
Keep the fewest violations.
Repeat on ten shuffles.
8
Result and bounds
The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.