“Every route from 0006 and 0008, then the tableau: does any route make the forced key read as English?”
0006 and 0008 matched forced key fragments against a word list. Here the forced key under every route and live period up to 40 is scored as English, with a shuffled-letter control, as 0009 did for the plain tableau.
Earlier we asked whether 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. scramble followed by 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. could use a short secret word; none could. Now we ask a softer question: for each scramble, 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. pin down 24 letters of the secret wordthe secret word The word or string a cipher needs besides the method., scattered through it. Do those pinned lettersthe forced letter Given a known message letter, its carved letter, and a way of using the table, the one letter of the secret word that would turn the first into the second. ever spell something that looks like English? We check every scramble we have, and we run the same check on the carved lettersthe carved letters The 97 letters cut into the copper at the CIA, shown at the top of every experiment page. shuffled at random, so we know what a meaningless best looks like.
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 routes
The message is first moved by a route (23,930 of them: grids, rotations, the K3 rule with every start and step), then enciphered on the table with a repeating key. Under each route the known letters force key letters; where they sit side by side they form runs.
the runs
Each run was scored as English. The best, UHAVETHE, looks like words but is eight letters long out of 461,692 tries; short English-looking runs turn up by chance.
the verdict
The best run on the copper beats the best on two shuffled coppers by a little, which is what a long search produces. Recorded as chance, not a lead.
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.
Take the route: write in rows of 7, read down the columns.
The known letters at 22 to 34 land at 4, 18, 32, 46, 60, 74, 88, 5, 19, 33, 47, 61, 75; those at 64 to 74 land elsewhere.
Under the sculpture's alphabet with Vigenere, each landing spot has a forced key letter.
Choose period 28: slot is the landing spot minus one, mod 28.
Write the 28 slots in a row and fill in the forced letters; 24 of 28 are filled.
Read the filled stretches.
Score them as English; compare with what random letters would give.
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.
route: columnar width 7, read columns left to right; convention kry/kry/kry Vigenere; period 28
slot: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
key: . . . Q G . . . . L E . . . . . . U N . . . . O K . . .
runs scored:
5
Hypothesis, for a cryptanalyst
Family, key space, the constraint under test, and the clues it reads.
Under K4 = tableau(T(P)) with a periodic key of period p, each route T sends the 24 cribs to 24 intermediate positions whose forced key letters fill 24 slots of the key, most of them distinct when p is 27 or more. If the key is a word or phrase (C5, C6), the filled slots read as English in slot order. Routes: the 2,905 columnar routes of 0006 and the 21,025 K3-rule and double-rotation routes of 0008, under all 24 conventions; for each live period up to 40 the forced slots are collected, runs of five or more consecutive filled slots, at least four quadgrams in all, are scored by quadgram log-probability, and the best (route, convention, period) is compared with the same procedure on two shuffles of the K4 letters. Chance expectation: the best of about a million scores sits where the shuffled best sits. Bounds: complete for these route sets, conventions and periods, for readable keys.
6
Prediction
Written before the run; the result decides it.
The best forced key over all routes, conventions and periods scores within 0.2 of the best from shuffled K4 letters, and the top entries show no recognisable word of five or more letters. Zero candidates pass.