Experiment 45

“The earlier messages as a running key, on tables rebuilt from the puzzle's words”

Experiment 4 slid the K1-K3 plaintexts and ciphertexts past K4 under the standard and KRYPTOS alphabets. Here the plain and cipher alphabets are the keyword mixings of every vocabulary word, with the key letters read on the standard or KRYPTOS order.

revertedstatus
experiment 4parents
2 mintime, all phases
0candidates
0/24known letters
1

The experiment method

What the method does with the 24 known letters.

Experiment 4 asked whether the first three messagesthe earlier messages The three solved parts of the sculpture., lined up under the fourth as a long secret word, could explain the known lettersthe known letters Sanborn has confirmed that positions 22 to 34 of the message read EASTNORTHEAST and positions 64 to 74 read BERLINCLOCK., using the sculpture's own table. The tablethe table The right-hand half of the sculpture, 26 rows of the alphabet each shifted one place further, starting from the word KRYPTOS. might have been built from another word, so we rebuild it from each word the puzzle knows and slide the three messages past the fourth again, at every starting point, forwards and backwardsa mirror Reading or writing the text from the end to the start..

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.

OBKRUOXOGHULBSOLIFBBWEASTNORTHEASTOTWTQSJQSSEKZZWATJKLUDIAWINFBBERLINCLOCKWGDKZXTJCDIGKUHUAUEKCARIN: THE COPPER, ALL 97 LETTERS, THE 24 KNOWN PLACES IN ACCENTthe key: the K1 solution, one letter per place from its first letterBETWEENSUBTLESHADINGANDTHEABSENCEOFLIGHTLIESTHENUANCEOFIQLUSIONBETWEENSUBTLESHADINGANDTHEABSENCEOplace 24: the third miss ends this offset0 of 24 known letters at this offset; the best of 146,932,380 trials still misses more than 2OUT: THE READING AT THIS OFFSET ON THE BERLIN AND CLOCK TABLES; HITS IN ACCENT, MISSES OUTLINEDEEGVPYGGOFEPEBVEDBSZWVYYMJPMZGWEMIXDMKFZEDMBMWVJRIAGZKOYVTRTAUSKUWZOOBCMMGHNMWZUHZUAYAGKAULRLPYBL
the key row

A running key is a long text written under the message, one letter per place. Drawn here: the K1 solution, with the BERLIN alphabet for the message and the CLOCK alphabet for the copper.

the rule

The search slides the text one letter at a time along the message, forwards and backwards; three misses among the 24 known places end an offset.

the bound

146,932,380 trials over every offset, both directions, every table pair and three modes: no offset comes within two misses.

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.

  1. Build the alphabet from BERLIN for the message letters and from CLOCK for the carved letters.
  2. Lay the K1 plaintext under the fourth message starting at some offset, so each position has a key letter.
  3. Position 22 is E: find E in the BERLIN alphabet, add the key letter's place in the plain alphabet, and read the result in the CLOCK alphabet; compare with the carved F.
  4. A miss counts one; three misses end this offset.
  5. Slide the offset along the whole text, then reverse the text, then try the next pair of alphabets.
the worksheet: the first steps by hand, on squared paper
RUNNING KEY: THE K1 SOLUTION ON THE BERLIN AND CLOCK TABLES, THE FIRST KNOWN PLACESplacecarvedin CLOCKkeykey no.differenceBERLIN letterknown22F8N1321VE23L1D324YA24R17T1924YS25V21H714MT26Q16E412JN27Q16A016PO28P15B114MR29R17S1825ZT8 misses in the first eight; the third already ends the offset. Thesearch slides the text one letter on and starts again.
4

Worked example, generated

The same steps run by the code on the real letters.

(worked_example failed: IndexError('string index out of range'))
5

Hypothesis, for a cryptanalyst

Family, key space, the constraint under test, and the clues it reads.

Sanborn's 'clues to the last section are contained in previously deciphered parts' (C7) read as a running key was killed under 24 conventions in experiment 4. Under keyword-mixed plain and cipher alphabets (C5, C29) every trial changes, so the sweep is rerun: sources are the K1, K2 (as carved and corrected) and K3 plaintexts, the three ciphertexts, and the concatenations of each set, read forward and backward at every cyclic offset; 63 plain x 63 cipher alphabets x 2 key alphabets x 3 modes; mismatches counted with early exit at three. Chance survival per trial is 26 to the minus 24, with about 180 million trials. Bounds: complete for these texts and alphabets.

6

Prediction

Written before the run; the result decides it.

No text, offset, direction or alphabet pair explains all 24 known letters, and none comes within two mismatches. Zero candidates pass.

7

Method

  1. Build the key texts and the alphabet pool.
  2. For each plain and cipher alphabet, key alphabet and mode, precompute the cipher letter for every (plain, key) pair.
  3. For each text, direction and offset, count mismatches at the 24 cribs, stopping at the third.
  4. Keep trials within two mismatches; decrypt and submit any with zero.
8

Result and bounds

The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.

0candidates returned
0passed every gate
0/24known letters, best
–lowest loss
34champion before

146932380 trials; 0 survivors; 0 within two mismatches.

What was searched, as the harness recorded it.

family
running key from K1-K3 plaintexts and ciphertexts, keyword-mixed alphabets
alphabets
63
key alphabets
std, kry
modes
3
sources
K1 plain fwd
63
K1 plain bwd
63
K2 plain fwd
369
K2 plain bwd
369
K3 plain fwd
336
K3 plain bwd
336
K1+K2+K3 plain fwd
768
K1+K2+K3 plain bwd
768
K1 cipher fwd
63
K1 cipher bwd
63
K2 cipher fwd
369
K2 cipher bwd
369
K3 cipher fwd
336
K3 cipher bwd
336
K1+K2+K3 cipher fwd
768
K1+K2+K3 cipher bwd
768
K2 plain corrected ending fwd
13
K2 plain corrected ending bwd
13
trials
146,932,380
stopped by clock
no
fewest mismatches
more than 2
near misses
none
survivors
0

How this experiment sits among the others is on the lineage page; its family's open hypotheses are on the families page.

Rendered from the record on 2026-10-18 12:26 UTC; updated after every experiment.
view on GitHub