Experiment 259
F13.4h (Alexis, 2026-10-28, #101: chase it rather than note it). 0237 put K4's best reading at B2 (the sculpture's table with a key of 2, then a move; best move 'two turns cw 70,19 inverse', a width-19 read) above all 100 shuffles; 0238 replicated it (1 of 300 at or above); 0239's blind judge put the reading at the edge of chance. The working guess: a property of the carved order at width 19 (like 0056's width-7 doublets), not a message. Here: 0238's frozen search (imported, unchanged) at B1-B2 and A1-A2 on K4 and 1,000 fresh shuffles, run three times per text with the move library cut to (a) every move, (b) width-19 reads only, (c) every move but width-19 reads; a statistic of the moved text itself, without any search (the two period-2 classes' index of coincidence under the flagged move, and the best over width-19 moves); 8 plants (B2 under a width-19 move) through the same cuts; the 24 known letters counted on K4's readings only after the comparison.
What the method does with the 24 known letters.
One earlier search found that K4the carved letters The 97 letters cut into the copper at the CIA, shown at the top of every experiment page., put through the sculpture's cipher table with a two-letter key and then read out of a grida grid Writing the message into rows of some width and reading it out in another order, column by column for instance, which shuffles the letters without changing them. 19 letters wide, gives a more English-looking jumble than almost any scrambled copy of K4 does. The reading itself is gibberish. We check whether this comes from the way K4's letters happen to sit at width 19, rather than from a hidden message: we repeat the search with only width-19 grids and without them, against 1,000 new scrambled copiesa shuffled control The same experiment run on the carved letters shuffled into a random order., and we measure that width-19 pattern directly.
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.
F13.4h: 0237-0239 found K4 reading better than its shuffles under the table with a two-letter key, then a width-19 move; the guess was a pattern of the carved letters at width 19.
1,000 fresh shuffles; the search repeated with width-19 moves only and without them; a direct measure of the pattern under the flagged move.
K4 above 999 of 1,000 again, and above 996 even without width-19 moves; the simple pattern measure is ordinary. The guess is wrong as stated; the excess is real and not yet explained (F13.4i).
What this experiment was built on, from the lineage page: 14 experiments before it, back to the root of its line. Its number is in bold.
This experiment produced no candidate: the search ruled everything out before a full text was built. Its product is the bounded negative under Result and bounds below.
Redo the method on a few letters with pencil and paper.
The same steps run by the code on the real letters.
F13.4h: is K4's English-like jumble at B2 a pattern of the carved order at width 19?
K4 --(two turns cw 70, 19, inverse)--> moved text
moved text: odd places | even places -> index of coincidence of each half
same for 1,000 scrambled copies; K4's rank says whether the width-19 pattern is unusual,
and the search repeated with and without width-19 moves says whether it carries the excess.
Family, key space, the constraint under test, and the clues it reads.
F13.4d's excess: K4 under order B (periodic tableau key of length 2 on the sculpture's 8 table classes, then one of 0129's 27,852 moves; ciphertext only, MIX8m) reads above 299 of 300 shuffles. Under order B with key length 2 the reading is the moved text through two fixed substitutions (odd and even places), so a moved text whose two classes are unusually 'monoalphabetic' (high index of coincidence) lets the climb make English-like pairs without any message. If the excess lives only in width-19 reads and the moved text's period-2 IC is as extreme as the B2 score, the excess is a fingerprint of the carved order at width 19, like the width-21 repeats and the width-7 doublet column (04-structure.md), and not a reading. Known letters: the search never reads them (program.md, F13); after the comparison, K4's B2 readings are checked for the 24 letters in message order and in carved order (A2). Clue entries C1(a) (a matrix move then the table, in the other order here), C25 (the width-7/21 statistics).
Written before the run; the result decides it.
The B2 excess replicates in (a) at p below 0.05 and in (b) width-19 reads only, and vanishes in (c) without them; the flagged move's period-2 index of coincidence on K4 is above at least 95% of the shuffles under the same move. K4's B2 readings put at most 4 of the 24 known letters in place either way. Plants found in (a) and (b), lost in (c). If so, the excess is recorded as a property of the carved order at width 19 and F13.4h is closed.
The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.
F13.4h: K4 at B2 p all 0.002, width-19 only 0.001, without width 19 0.004 (1,000 shuffles); period-2 IC under the flagged move: 137 of 1,000 at or above; plants found all 8/8, w19 8/8, not19 5/8
What was searched, as the harness recorded it.
LDFCTSBAMCWZEIEDAYSHEEATINQLLVRDJEWINKWORNYDFCHUTVISMNBULOOBLNVCONTRYCFXPAIXHIWCJIPZADKJVBPREJIHGHASCIPLYNELEDMLOPNEPPDEKWSIXOROHWFMJWSDAFREYTJORNTCUEJDUFNRMLZIPERUKIYUZPHXWILSHKESTLYQBVUSHOWGYFLDFCTSBAMCWZEIEDAYSHEEATINQLLVRDJEWINKWORNYDFCHUTVISMNBULOOBLNVCONTRYCFXPAIXHIWCJIPZADKJVBPREJIHGNBENDJZERANDBCKNFILLBAADWRONNWIIHLGARTNRHHKORCULLHCNCSIDEPXMEDFCKRHALORKYXQTRGROWDOJOWSMSHDVNLYATTIJWYNNOSYJVHADXNGLTZVAHOLLEJUAARCCIAFZXKROOTWINLEEFEBDPCBTQKNICHCCRPIYIDWFEJBEINDDWLSMSHRVIPVBUMFJTYBTIBCUSOFWAHINPSENGUUDAYAXTWCTISHEUMWOFNDRNOZGZHEXNLGCMIDAWHUHABATCSHANICORNTMEJVGQDBGRDSXYZKTIJWYNNOSYJVHADXNGLTZVAHOLLEJUAARCCIAFZXKROOTWINLEEFEBDPCBTQKNICHCCRPIYIDWFEJBEINDDWLSMSHRVIPVBUMNBENDJZERANDBCKNFILLBAADWRONNWIIHLGARTNRHHKORCULLHCNCSIDEPXMEDFCKRHALORKYXQTRGROWDOJOWSMSHDVNLYATLDFCTSBAMCWZEIEDAYSHEEATINQLLVRDJEWINKWORNYDFCHUTVISMNBULOOBLNVCONTRYCFXPAIXHIWCJIPZADKJVBPREJIHGHASCIPLYNELEDMLOPNEPPDEKWSIXOROHWFMJWSDAFREYTJORNTCUEJDUFNRMLZIPERUKIYUZPHXWILSHKESTLYQBVUSHOWGYFLDFCTSBAMCWZEIEDAYSHEEATINQLLVRDJEWINKWORNYDFCHUTVISMNBULOOBLNVCONTRYCFXPAIXHIWCJIPZADKJVBPREJIHGTZUODASJFPKQIDPPPEYTXEVTAKSVARDODUCAGOXFGXVSNFMTRNKBOMZINGFEBIUPLEOPERHAGHMYGRUZQUXYARRORLOPLEOFVHow this experiment sits among the others is on the lineage page; its family's open hypotheses are on the families page.