Cyphral Distich: How a 127-Year-Old Royalist Cryptogram Reportedly Fell to Claude Fable 5.1

A two-line, 64-number cryptogram tucked inside Sir Thomas Urquhart's 1652 Logopandecteision sat unsolved since it was first posed as a puzzle in 1899. According to a report from codebreaking-guide.com, Claude Fable 5.1 produced a decoding in 44 minutes, reigniting a debate about what these solves actually prove.

A puzzle that outlived three generations of codebreakers

The Cyphral Distich is small by cryptogram standards: just two lines, 32 numbers each. It appears in Logopandecteision, the eccentric linguistic treatise written by the Scottish Royalist Sir Thomas Urquhart. Notes and Queries put the cipher to its readers as an open problem in 1899, and it stayed open long enough to earn a spot on Klaus Schmeh's list of the top 50 unsolved encrypted messages. Across more than a century, documented attempts leaned on frequency analysis, straightforward substitution, and homophonic substitution — all reasonable approaches for a short numeric cipher, and all apparently insufficient here. The report credits Claude Fable 5.1 with producing a solution after 44 minutes of work, which is the detail that makes this worth examining rather than simply repeating.

What the claimed key actually does

The reported method doesn't treat the Distich as a numeric substitution at all. Instead, it uses Urquhart's own 32 Proquiritations — a structural feature of his book — as the key. Each number in the cipher selects a word position within the matching Proquiritation, and the first letter of that selected word becomes one letter of plaintext. Because there are 32 numbers per line and 32 Proquiritations, the mapping is one-to-one, line by line. The result, per the report, is two lines of exactly 32 letters each. The first line is described as a prayer for Charles II, consistent with Urquhart's documented status as a committed Royalist, and the two lines are said to end in words that rhyme. None of that proves intent or deliberate construction on Urquhart's part — the report simply states what the decoded text contains, and that distinction matters if you're trying to verify the claim rather than just admire it.

The second cryptogram in the same book, and where it gets messy

The Jewel, Urquhart's 1652 publication, contains a second, larger cryptogram called the Cyphral Octastich — 285 numbers rather than 64. Its reported key structure is different: instead of Proquiritations, it uses numbered pages, with each number again selecting a word position whose first letter contributes to the plaintext. The claimed result is a royalist prayer purportedly written in London in March 1652, but the report is explicit that nine letters remain unresolved, including an uncertain sequence that may correspond to the word 'Irish.' A later section of the decoding reportedly requires a one-page offset from the scheme used earlier, and separately, the report flags that one coordinate can shift depending on whether the Latin phrase hinc inde is read as a single unit or two words. These are the kinds of ambiguities that separate a verifiable claim from a tidy press release, and the report doesn't paper over them.

The access problem nobody has solved

Here is the detail that should give any reader pause before treating this as settled: the report states plainly that no free page images of the 1652 Jewel were found during the process. That means the key structure for the Octastich — numbered pages of a 374-year-old book — was apparently reconstructed or sourced without a freely accessible scan to check against. If you want to verify any of this yourself, the practical starting point is the Logopandecteision text itself, since the Proquiritation numbering for the shorter Distich is at least a stated structural feature of the book rather than a page-image dependency. For the Octastich, independent verification would require locating an actual 1652 edition or a reliable facsimile, which the report's own account suggests is not trivial to do for free.

What readers are actually arguing about

The discussion around this report splits in a telling way. Some readers find it a genuinely appealing instance of AI-assisted historical research, and one suggested testing the same model against the Voynich manuscript next, or even the identity of Satoshi Nakamoto. Others are far more skeptical: one asked whether the cipher could be hallucinated outright and asked for an original scan, which lines up with the missing page-image problem above. Another characterized the solve as brute-force persistence dressed up as intelligence, and a different commenter pointed out that models solving hard ciphers and math problems while still failing basic critical-thinking tasks is a pattern worth tracking rather than ignoring. A more structural critique held that some of these 'breakthroughs' may just be previously neglected, low-hanging problems rather than evidence of a genuine leap. There's also a quieter, more philosophical thread asking whether a world where AI solves every remaining puzzle leaves anything for human puzzle-solvers to enjoy — a question the report doesn't answer, but one worth sitting with longer than the 44 minutes it reportedly took to crack the Distich.

Specs

• Distich structure: 2 lines, 32 numbers per line
• Distich key: Corresponding Proquiritation word index; take the word's first letter
• Octastich size: 285 numbers
• Octastich key: Numbered-page word index; take the word's first letter
• Octastich status: Solved except for nine letters

Sources

• Unsolved Cryptograms: The Cyphral Distich and Octastich — codebreaking-guide.com