Experiment 23

“Masks that move the known letters: the message written backwards, or padded with nulls, before a periodic tableau”

New root in family 2. The mask is a known rearrangement or padding of the plaintext applied before a periodic tableau: reversal, a null before every k-th letter, or a run of nulls at the start. The cribs then sit at computable intermediate positions.

revertedstatus
a new rootparents
2 mintime, all phases
0candidates
0/24known letters
1

The experiment method

What the method does with the 24 known letters.

The cipher's designer said he first disguised the Englishthe message The 97 letters of English that Jim Sanborn wrote and enciphered. before enciphering it. Two simple disguises change where 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. sit without changing them: writing the message backwardsa mirror Reading or writing the text from the end to the start., or slipping a meaningless letter in after every few real ones. We try each disguisea mask Something done to the English before the cipher proper, such as writing it backwards or slipping in meaningless letters, that is not itself a cipher but moves or hides the letters., work out where EASTNORTHEAST and BERLINCLOCK would then sit, and ask again whether a short repeating secret word could have produced the carved lettersthe carved letters The 97 letters cut into the copper at the CIA, shown at the top of every experiment page. at those spots.

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 ACCENTone mask: a null letter before every 10th, then the table; the nulls are dropped before readingOBKRUOXOGHULBSOLIFBBWFLRVQQPRNGKSSOTWTQSJQSSEKZZWATJKLUDIAWINFBNYPVTTMZFPKWGDKZXTJCDIGKUHUAUEKCARshortest key length each of the 648 masks allows262728293031323326: no mask lets a key shorter than 26 fit-6-5-4Englishbest forced key fragment on the copper -5.38best on a shuffled copper -4.89OUT: NO READING; MASKS DO NOT SHORTEN THE KEYEASTNORTHEASTBERLINCLOCK
the masks

Sanborn spoke of a masking step. Tried: the message reversed, a null letter at a regular interval, or nulls at the start, 648 masks in all, each followed by the table with a repeating key.

the histogram

For each mask, the shortest key length the known letters still allow. None goes below 26, the same floor as with no mask at all.

the scale

The best key fragment the known letters force under any mask scores -5.38 as English, worse than the best on a shuffled copper. Masks of this kind add nothing.

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. Write the message backwards: the E of EAST at position 22 is now at position 76, and the K of CLOCK at 74 is now at 24.
  2. Under the sculpture's alphabet with Vigenere, the carved letter at 76 and plaintext E force one key letter; the carved letter at 24 and plaintext K force another.
  3. The reversed EASTNORTHEAST occupies 64 to 76 and the reversed BERLINCLOCK 24 to 34.
  4. Positions whose distance is a multiple of the period share a key letter; if their forced letters differ, that period is dead.
  5. Now try padding instead: a null after every third letter pushes position 22 to 29 and position 64 to 85.
  6. Repeat the forcing and the period test for each padding pattern.
the worksheet: the first steps by hand, on squared paper
THE MASK BY HAND: STRIKE A NULL BEFORE EVERY 10TH LETTER, CLOSE THE GAPS, RENUMBEROBKRUOXOGHULBSOLIFBBWFLRVQQPRNGKSSOTWTQSOKRUOXOGHUBSOLIFBBWLRVQQPRNGSSOTWTQSJSSthe first 40 carved letters; the struck ones are nullsafter striking: every later letter moves left, so the known letters land on new placesEASTNORTHEAST now sits over different carvedletters than before. Redo the key-letter checkat every key length; the mask is worth keepingonly if a shorter key survives.None does. Six hundred and forty-eight masks,each a few minutes by hand.
4

Worked example, generated

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

mask: the message written backwards; convention kry/kry/kry Vigenere

plain pos  letter  sits at  carved there  forced key letter
       22    E        76        G            Y
       23    A        75        W            J
       24    S        74        K            Q
       64    B        34        S            X
       65    E        33        S            U
       74    K        24        R            R

mask: a null before every 3rd letter: position 22 sits at 30, position 64 at 86
5

Hypothesis, for a cryptanalyst

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

Scheidt: 'I masked the English language' (C4); the NPR 2005 guesses were nulls and phonetic spelling, and reversal is the simplest Sanborn-style 'visual system' (C3). A mask that only moves letters or inserts nulls is a monotone or mirrored map from plaintext position to intermediate position; after it a periodic tableau with separate alphabets forces key letters at the mapped positions, and the divisor rule gives the live periods exactly as in 0006. Masks: reversal; a null before every k-th letter for k = 2 to 24 at every phase; a prefix of 1 to 24 nulls; and reversal combined with each. For each mask and each of the 24 conventions the smallest live period is found; periods up to 30 have their forced fragments matched against the vocabulary and K1-K3 windows and scored as English, with ten shuffled controls. Bounds: complete for these masks and conventions.

6

Prediction

Written before the run; the result decides it.

Reversal and every null pattern leave the smallest live period in the 20s or above, as random position maps do. No forced fragment matches a word or beats the shuffled controls. Zero candidates pass.

7

Method

  1. Build the masks as position maps from plaintext index to intermediate index.
  2. For each mask and convention, force the key letter at each crib's intermediate position.
  3. Find live periods by the divisor rule.
  4. For live periods up to 30, match the forced fragment against the vocabulary and K1-K3 windows and score it as English.
  5. Repeat the scoring on ten shuffles of K4.
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
nonechampion before

648 masks x 24 conventions. Smallest live period over all masks: 26. Best fragment -5.378 vs shuffled best -4.888. 0 matches.

What was searched, as the harness recorded it.

family
masked plaintext (reversal, regular nulls, prefix nulls) then periodic tableau
masks
648
conventions
24
smallest live period histogram
26
11
27
41
28
403
29
110
30
36
31
21
32
6
33
6
masks dropping a crib
14
best forced fragments
  1. score
    -5.378
    mask
    reversed, null before every 10th letter, phase 2
    convention
    kry/kry/kry/vig
    period
    30
    fragment
    TALIT.FV..USVICCCUS.KDYZ..QQVY
  2. score
    -5.473
    mask
    null before every 9th letter, phase 7
    convention
    std/kry/std/varbeau
    period
    30
    fragment
    ETBYVOW...CSOLLAS.MQYU.DEYZ.MV
  3. score
    -5.519
    mask
    null before every 10th letter, phase 0
    convention
    std/kry/std/vig
    period
    30
    fragment
    WHZ.BEFL..YIMPPAI.OKCG..SUCKOF
  4. score
    -5.617
    mask
    reversed, 10 nulls first
    convention
    std/kry/kry/beau
    period
    28
    fragment
    IL...JAMBSAHREDA.NIGHAMIMPYY
  5. score
    -5.682
    mask
    reversed, null before every 10th letter, phase 3
    convention
    kry/kry/kry/vig
    period
    30
    fragment
    TALITN.V..USVICCCUSQ.DYZ..QQVY
  6. score
    -5.683
    mask
    reversed, 10 nulls first
    convention
    std/kry/kry/beau
    period
    29
    fragment
    ....JAMBSAHREDANIGHAMIMPYYIL.
  7. score
    -5.704
    mask
    null before every 10th letter, phase 3
    convention
    std/kry/std/varbeau
    period
    30
    fragment
    ETBYVO.P..CSOLLASLPE.U.DE.YQMV
  8. score
    -5.719
    mask
    null before every 13th letter, phase 6
    convention
    kry/kry/std/beau
    period
    29
    fragment
    URORN.KB..MDADSTGE.FIY.MDAYWG
  9. score
    -5.727
    mask
    null before every 10th letter, phase 1
    convention
    std/kry/std/vig
    period
    30
    fragment
    WHZC.EFL..YIMPPAIP.KCG..SUCKOF
  10. score
    -5.731
    mask
    reversed, null before every 11th letter, phase 10
    convention
    kry/kry/std/beau
    period
    30
    fragment
    GPXH.PF.IY.TOOBOPTDELB...UDISC
  11. score
    -5.738
    mask
    reversed, null before every 11th letter, phase 9
    convention
    kry/kry/std/beau
    period
    30
    fragment
    GPX.GPF.I.FTOOBOPTDEL.U..UDISC
  12. score
    -5.766
    mask
    reversed, 10 nulls first
    convention
    std/kry/kry/beau
    period
    27
    fragment
    PYYIL.JAMBSAHREDA..NIGHAMIM
and 3 more
shuffled control best
-5.209, -5.11, -4.894, -5.524, -5.188, -5.528, -5.534, -4.888, -5.255, -5.415
vocab matches
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