“Two earlier carved letters as the key at once, with or without a short repeating word”
Experiment 66 used one earlier carved letter as the key. The two fingerprints point at two different distances, 1 and 21, so here the key is the sum or difference of two earlier carved letters at any pair of distances, a letter-for-letter cousin of the Gromark, optionally with a repeating word of up to 30 letters on top.
The copper's two oddities are tied to the distances 1 and 21. A cipher whose key is made from two earlier carved letters, one just before and one 21 back, would leave both marks. We try every pair of distances, addingVigenere The three rules of arithmetic a table can use: add the key letter to the message letter, subtract the message letter from the key letter, or subtract the key letter from the message letter. or subtracting each letter, with or without a short repeating secret word on top. Because every carved letter is visible, 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. gives an exact yes or no. Scrambled coppers through the same test show how many yeses luck gives.
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 cipher
Each carved letter takes two earlier carved letters as its key, here the ones 1 and 21 places back, plus a short repeating word or a constant.
the check
With a constant, places 22 and 23 leave Z and J: they differ, so this pair of distances is dead.
the bound
Every pair of distances, both signs, word lengths to 30: 15 settings pass three checks against 3 to 12 on shuffled coppers. They are scattered with no common distance, and distances 1 and 21 together never pass more than one.
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.
Count letters along KRYPTOSABCDEFGHIJLMNQUVWXZ from K = 0, and take distances 1 and 21, both added, with no extra word beyond a constant.
Place 22: carved F (12), known E (11), previous carved W (23), carved 21 back O (5): the constant is 12 - 11 - 23 - 5 = -27, which is 25, Z.
Place 23: carved L (17), known A (7), previous carved F (12), carved 21 back B (8): 17 - 7 - 12 - 8 = -10, which is 16, J.
Z and J disagree, so this pair of distances with a constant is dead under this way of using the table.
With a 2-letter repeating word instead of a constant, places 22 and 24 would have to agree, and so on.
Repeat for every pair, sign, word length and way of using the 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.
distances 1 and 21, both added, sculpture's alphabet throughout, Vigenere
place known carved 1 back 21 back constant
22 E F W O Z
23 A L F B J
24 S R L K T
25 T V R R J
5
Hypothesis, for a cryptanalyst
Family, key space, the constraint under test, and the clues it reads.
Family 1 (ciphertext feedback), extending 0066 to two lags: T[i] = s1A(c[i-a]) + s2A(c[i-b]) + k[i mod P], 1 <= a < b <= 74, s1, s2 = +1 or -1, P = 1..30 (P = 1 is a constant shift, which includes no extra key), plain, cipher and key alphabets standard or KRYPTOS, Vigenere or Beaufort (16 classes; variant Beaufort is the all-minus sign). This is a lagged-Fibonacci (Gromark-like) feedback on letters rather than digits; a = 1, b = 21 would produce both the width-7 column mechanism of 0065 and the width-21 repeats of 0062 in one stage. Because the carved text is known, the feedback subtracts out and each crib with i >= b gives an exact equation on the periodic key. 2,701 lag pairs x 4 signs x 16 classes x 30 periods = 5.2 million cases. Survivors are tallied by redundant checks against ten shuffled coppers through the same sweep; forced fragments matched against words. Clue entries: C11a, C4(b) (Scheidt's 'base' remark read as a running numerical key), C22(a) (a Gromark-style generator).
6
Prediction
Written before the run; the result decides it.
No case with three or more redundant checks survives, including a = 1, b = 21 under every sign, class and period. With no extra key (P = 1) nothing survives at all. Survivors with one or two checks occur at the shuffled-copper rate; no fragment matches a word. Zero candidates pass.
7
Method
For each pair of distances, signs and way of using the table, subtract the two earlier carved letters from the total key at each known place past the larger distance.
For each word length up to 30, check that places sharing a slot agree.
Tally consistent cases by redundant checks.
Repeat on ten shuffled coppers.
Match forced fragments against words.
8
Result and bounds
The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.