Wallis Ciphers John Wallis & the English Civil War · 1640s
Cromwell’s mathematician became the first professional English cryptanalyst — and then served every English government for half a century without changing sides.
Why This Matters
John Wallis (1616–1703) was a Cambridge-trained mathematician, ordained Anglican clergyman, and — from 1649 — Savilian Professor of Geometry at Oxford. He was also one of the founders of the Royal Society. None of this is what made him historically important to cryptography.
In late 1642, at the start of the English Civil War, Wallis was shown a captured Royalist letter by a friend at dinner. He broke it in two hours. Word spread. By 1643 he was Parliament’s house cryptanalyst, reading the King’s correspondence as it was intercepted. After Naseby (1645) the captured King’s cabinet of letters was given to Wallis to decipher; the resulting publication, The King’s Cabinet Opened, devastated Royalist propaganda by exposing Charles I’s secret negotiations with Catholic Ireland.
Wallis kept his post through every regime change. He served Parliament, then Oliver Cromwell, then Charles II at the Restoration (Charles II is said to have grumbled that Wallis had “broken his father’s ciphers” but kept him on anyway), then James II, then William III. He was still actively breaking diplomatic ciphers for the English government when he died at 87. He is the founding figure of professional English state cryptanalysis — the institutional ancestor of the Government Code & Cypher School and ultimately of GCHQ.
The Royalist ciphers Wallis attacked were Argenti-pattern nomenclators of the standard 17th-century kind: a small alphabetic substitution table (often with one or two homophones for common letters) plus a longer codebook of person-and-place codes. Royalist operators were also fond of nulls and of inserting the names of their own family members as agreed cover-words.
The demo above implements this dominant design: a 60-word codebook of English Civil War vocabulary (people, places, military terms) using 3-digit codes from 100, plus a keyed 2-digit homophonic alphabet (codes 10–89) for words not in the codebook — wrapped in sentinel codes 90 and 91 so the decoder can tell which is which. Try encoding KING CHARLES MARCH OXFORD to see how the codebook handles the named entities directly.
Wallis’s attack technique was the standard one of the next two centuries: identify the most-frequent codes, guess the alphabet portion against expected high-frequency English letters, look for repeated codeword patterns that match expected proper nouns (“the King”, “Parliament”, “Oxford”), and bootstrap from there. He was unusually fast at it.
Once Wallis had read a few intercepts in a given system he could identify codewords by their distribution: a code that appeared once or twice per letter, around proper noun positions, was almost always a person or place name. He kept running tables of broken codewords and shared them across the war effort. The Royalists never developed an equivalent counter-cryptography effort.
Royalist agents reused nomenclators across years, mixed plaintext and cipher in the same letter, and sometimes labelled their codebook entries with the person’s actual name in the margin. Captured agents carried codebooks; captured letters carried decryption hints. By 1645 Wallis had a substantial library of broken material.
Because Wallis worked continuously for the English state from 1643 to his death in 1703, he accumulated a body of cryptographic knowledge unmatched in Europe. Continental ciphers he had broken decades earlier were still being modestly evolved by their users. He recorded much of his technique in cipher notes now in the Bodleian Library.
| Wallis lesson | Modern echo |
|---|---|
| Concentrated, persistent, well-funded cryptanalysis beats episodic effort | Modern signals intelligence: NSA, GCHQ — the same institutional pattern, scaled |
| Operational discipline matters as much as algorithm choice | Modern crypto failures are almost always implementation or operational, not algorithmic |
| Cumulative knowledge across years compounds the attacker’s advantage | Why crypto-agility & aggressive deprecation cycles are now standard |
| The defender publishes; the attacker doesn’t | Why open cryptanalysis (NIST competitions, IACR) is so valuable to defenders |
Wallis sits inside the same nomenclator tradition mapped in Cryptiana, especially the archival work on 16th- and 17th-century diplomatic cipher practice. Read him in sequence with Argenti for the design side and with Arnold-André for the later book-cipher branch of state secrecy.
| Cryptanalyst | John Wallis (1616–1703), Savilian Professor of Geometry, Oxford |
| Targets | Royalist nomenclators (Charles I, Prince Rupert, the King’s court) |
| Period | Professional cryptanalyst 1642 – 1703 |
| Mechanism | Nomenclator + 2-digit homophonic alphabet |
| Patrons | Parliament, then Cromwell, then Charles II, then James II, then William III |