Experiment 3
The key letter is now the plaintext letter L back, not the ciphertext letter.
What the method does with the 24 known letters.
This is the other autokeyan autokey cipher A cipher that uses the message itself, or the scrambled text itself, as the secret word, each letter enciphered with the letter a fixed distance before it.: the real message, not the scrambled one, becomes its own key after a short start. We can only test it where two 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. sit the right distance apart, so some distances get a clear no and some get no answer at all. The point is to know exactly which distances are still alive, so no one wastes time on the deaddead Said of a period, grid or method when the known letters contradict it: no secret word of that kind can produce the carved letters. ones.
Words in the margin are defined on the glossary page.
The carved letters go in at the top; the reading comes out at the bottom.
In a plaintext autokey each letter is keyed by the message letter a fixed distance before it. Where two known letters sit that far apart, one must carve the other's place. Drawn for distance 3.
At distance 3, on the sculpture's table, 17 of 18 checks fail. Under the other 23 ways of using the table it fails too.
34 distances die under every way of using the table. Distance 12 passes its single check in one way. Distances with no two known letters that far apart stay open: the cribs say nothing about them.
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.
Redo the method on a few letters with pencil and paper.
The same steps run by the code on the real letters.
plaintext autokey, lag 3, all alphabets KRYPTOS, Vigenere
alphabet: KRYPTOSABCDEFGHIJLMNQUVWXZ
pos plain key=p[pos-3] idx(p)+idx(k) predicted carved
25 T E 4+11 I V x
26 N A 19+7 K Q x
27 O S 5+6 E Q x
28 R T 1+4 O P x
29 T N 4+19 W R x
30 H O 14+5 N N ok
Family, key space, the constraint under test, and the clues it reads.
K4 is a plaintext-autokey cipher: c[i] = f(p[i], p[i-L]) after a primer of length L. This is the NSA's 1992 first guess for the interval-7 property (knowledge/04-structure.md). The cribs decide it only where both p[i] and p[i-L] are known: within EASTNORTHEAST (lags 1 to 12), within BERLINCLOCK (lags 1 to 10), and between the two runs (lags 30 to 52). We test all 24 conventions at every lag 1 to 96, count the checkable pairs, and record three outcomes per lag: excluded (a mismatch), consistent (all checks pass, at least one check), or undecided (no overlapping pair). For consistent lags we cannot decrypt without the primer, so we report them as open and bound the primer space. Clue register C11a.
Written before the run; the result decides it.
Lags 1 to 12 and 30 to 52 are excluded under every convention. Lags 13 to 29 and 53 to 96 stay undecided. No candidates pass the crib gate.
For each convention and lag L: over crib positions i with i-L also a crib position, compare f(p[i], p[i-L]) with c[i]. Tally excluded, consistent, undecided per lag across conventions. No candidates are generated because the primer is unknown; the bounds are the deliverable.
The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.
No candidate can be built without the primer; this experiment exists for its bounds.
What was searched, as the harness recorded it.
How this experiment sits among the others is on the lineage page; its family's open hypotheses are on the families page.