Test a Solution

Cipher

Z13 — the “My name is” cipher, 13 characters Z32 — the map cipher, 32 characters

Try ALFRED E NEUMAN Try DREATATOTPEDO Try DR EAT A TORPEDO

What the test actually proves

Very little, and that is the point. The check tells you whether a proposed plaintext is possible — whether it is consistent with the repeated symbols the Zodiac drew. It cannot tell you whether it is right.

Both ciphers are far below the length at which a solution becomes unique. Ordinary English throws off a passage that satisfies Z32 about once every 3,600 letters. For Z13 there are 49 real first-and-last names that pass, and if you allow a single encoding error the set grows to 3,668,004. So a proposal that clears every constraint has cleared the minimum bar, and nothing more.

What the tool is genuinely good for is the other direction. A proposal that fails the check is refuted, immediately and without argument, because it asks the same drawn symbol to stand for two different letters. That disposes of a great many published claims in about a second.

Reading the result

• Length. A straight substitution reading has to be exactly as long as the ciphertext. Shorter or longer means the claim is doing something else — anagramming, dropping letters, or reading a subset — and the claimant owes you an explanation of what.
• Constraints. Each row is one symbol the Zodiac used more than once, the positions where it appears, and the letters your reading puts there. They must all match.
• Simple or homophonic. For Z13 the tool also reports how many distinct letters your reading uses. Our analysis of his encipherment method indicates Z13 used a small, near one-to-one key, so a reading that needs several symbols to share a letter is working against the evidence.

The famous cases

Try the buttons above. ALFRED E NEUMAN, proposed by Craig Bauer, fails exactly one constraint: positions 3 and 11 carry the same drawn symbol, and the reading puts F at one and M at the other. DR EAT A TORPEDO, in the form Ryan Garlick proposed via the Z340 key, fails one as well — positions 5, 7 and 9 share a symbol and would need T, T, R.

Both are near misses rather than nonsense, and the Zodiac did make encoding mistakes. That is exactly why the one-error allowance matters so much, and why we measured it: permitting a single slip takes the Z13 answer set from 49 to 3,668,004. A near miss is not evidence when three and a half million other names miss by the same margin. The full argument is in Findings.