Spymasters

Secret Writing: A History of Codes, Ciphers and Codebreaking

Course 1 · Chapter 6 · Spymasters

Chapter 5 showed a cipher that was strong on paper and broken in the end. This chapter is about what ciphers meant to the people who used them: kings, queens, plotters and the officials whose job was to read other people's letters. A cipher is only part of a secret; the rest is the people carrying the letters and the people waiting to read them.

A Queen in a Cage

Mary, Queen of Scots, fled to England in 1568 and spent almost twenty years as a prisoner of her cousin Elizabeth I. English Catholics saw her as the rightful heir, and several plots aimed to free her or to put her on the throne. Mary needed to write to friends and supporters abroad, so she needed a cipher and a way to send letters out. Her enemies needed exactly the opposite: to read all of it.

Key figures Mary, Queen of Scots (1542 to 1587): imprisoned in England from 1568 and executed in 1587.
Sir Francis Walsingham (about 1532 to 1590): Elizabeth I's principal secretary, who ran a network of agents and informers.
Thomas Phelippes (about 1556 to 1625): Walsingham's cipher secretary, skilled at breaking ciphers and at copying handwriting.
Anthony Babington (1561 to 1586): a young Catholic gentleman who led the plot named after him.
Gilbert Gifford: a double agent who carried letters for Mary while reporting to Walsingham.
Not Yet Verified The names, dates and roles above are written from general knowledge and were not checked against a source for this chapter. Birth years in particular are approximate or uncertain for several of these people. Treat them as pointers for your own search.

The Babington Plot, 1586

In 1586 Walsingham arranged a trap. Mary's letters were smuggled in and out of her prison in watertight packages hidden in a beer barrel. She believed the route was safe; in fact every letter was handed to Walsingham's men, copied, and sent on. Mary and Babington then exchanged enciphered letters in which, according to the usual account, plans for rescuing her and for killing Elizabeth were discussed.

Mary used a nomenclator (Chapter 4): a table of cipher symbols for letters, plus symbols for common words and names, plus a few meaningless "nulls" to confuse the counting. Phelippes's skill was to read such a cipher without the table, using the methods of Chapter 3: count the symbols, guess the commonest as E and the other frequent letters, and use probable words such as names and greetings to anchor the rest. It was this chain of skill, rather than any clever machine, that exposed the plot.

A nomenclator, in outline (invented symbols, not Mary's real table) Letters : A = 7 B = 3 C = 9 D = 1 ... (one symbol per letter) Words : 14 = the queen 27 = letter 31 = send 52 = ship Nulls : 08, 00 = meaningless, ignore when reading Sending a short word costs one symbol; spelling a name costs one per letter.
Not Yet Verified Popular accounts say Phelippes added a forged postscript to one of Babington's letters, asking Mary to name the men who would carry out the plot. I did not check a source here, and some historians dispute how reliable the surviving evidence is. Present it as the traditional story, not as settled fact.

The Price of a Broken Cipher

Mary was tried in 1586 and executed in February 1587. The lesson for secret writers was harsh: a cipher can be used correctly and still fail if the person who relies on it trusts the wrong courier, or does not know that someone is reading along. A broken cipher is not only a technical failure, it is a failure of the whole system around it.

Ciphers in the English Civil War

Ciphers were not just for conspirators. During the English Civil War (1642 to 1651), both sides enciphered their letters. Royalist letters captured after the Battle of Naseby in 1645 are often mentioned as an example, and the mathematician John Wallis became known for deciphering such messages for Parliament. Wallis's career shows something new: breaking ciphers was becoming a recognised skill that governments paid for, and mathematicians began to see it as a field in its own right.

Not Yet Verified The Naseby detail and the exact extent of Wallis's work are written from general knowledge and were not checked against a source. Search for "Wallis cipher Parliament" if you want the details.

The Great Cipher of Louis XIV

In the late 1600s the French court used a cipher so elaborate that it was known as the Great Cipher. It was devised by the Rossignol family of cryptographers (Antoine Rossignol and his son Bonaventure). Instead of one number per letter, it used several hundred numbers, and many of them stood for whole syllables, with a few for common words and some nulls.

A syllable cipher in outline (invented numbers) Table : 18 = SE 42 = CRET 63 = KEY 77 = IS 00 = null Cipher : 18 42 63 77 00 Plain : SE CRET KEY IS Letter counting finds nothing useful: no number stands for "E" or "T" alone.

Because the numbers stood for syllables, ordinary letter-frequency analysis (Chapter 3) failed. The cipher resisted attack for about two centuries and, when the table was lost, the messages protected by it became unreadable. Around 1890 the French cryptanalyst Etienne Bazeries worked on a set of the old messages and reportedly succeeded after realising that the numbers represented syllables, not letters. One famous result was a letter that appeared to reveal something about a prisoner of the king.

Not Yet Verified The Rossignol names, the date of Bazeries's break, and the identity of the prisoner mentioned in popular accounts are written from general knowledge and were not checked against a source. Some of the stories about the break are disputed or romanticised. Treat them as leads for your own search.

Comparing the Tools

CipherUsed byStrengthWeakness
Mary's nomenclatorMary, Queen of ScotsLetters, words and nulls mixedFrequency analysis plus probable words; a compromised courier
Civil War ciphersRoyalists and ParliamentQuick to use in the fieldCaptured letters; skilled mathematicians
Great CipherLouis XIV's courtSyllables defeat letter countingOnce guessed, the syllable idea opens the whole table

What Spymasters Teach

Three themes run through this chapter. First, the weakest link is rarely the mathematics: it is the courier, the habit or the assumption that the cipher is safe. Second, cryptanalysis was turning professional, with Phelippes, Wallis and later the Rossignols showing that governments would pay for people who could do it. Third, complexity is not the same as strength: the Great Cipher was complicated and held for two centuries, but it fell once someone guessed its underlying idea.

Try it yourself: a shifted letter A courier carried this Caesar-shifted message (Chapter 2). The shift is 3 letters, so to decipher you move each letter 3 places back in the alphabet:

KLGH WKH OHWWHU LQ WKH EDUUHO

What does it say?
Show the worked answer

K moves back three to H, L to I, G to D, H to E: KLGH is HIDE. WKH is THE, OHWWHU is LETTER, LQ is IN, WKH is THE and EDUUHO is BARREL. The message is HIDE THE LETTER IN THE BARREL. A single shift like this is far weaker than Mary's nomenclator, which shows why serious users moved beyond it.

Where This Leaves the Secret Writer

By 1700 cipher making and breaking were professional trades practised by governments. The next chapter moves to the nineteenth and early twentieth centuries, where the telegraph made messages travel faster and made secrets harder to keep.

Exercises

Exercise 1

Use this nomenclator list: 14 = the queen, 27 = letter, 31 = send, 45 = the plotters, 52 = ship, 60 = tonight, 08 = null (ignore). Read the message 31 14 08 27 60 08 52 and write out what it says. Then explain what the nulls are for.

Exercise 2

Use this syllable list: 18 = SE, 42 = CRET, 63 = KEY, 77 = IS, 90 = HID, 35 = DEN, 00 = null. Read 18 42 00 63 77 90 00 35. Then say in a sentence why counting the commonest number would not give you the letter E.

Exercise 3

Here is a Caesar-shifted message: AOL SLAALY NVLZ PU AOL IHYYLS VU MYPKHF. (a) Guess the plain word for AOL and work out the shift. (b) Check the guess against PU and VU. (c) Decipher the full message.

Quick Reference

  • Nomenclator: symbols for letters plus symbols for common words and names, with nulls added
  • Babington plot (1586): Mary's letters were read in transit by Walsingham's team
  • Phelippes: broke the cipher with frequency counting and probable words
  • Civil War: codebreaking became a paid, recognised skill
  • Great Cipher: numbers stood for syllables, which defeated letter counting until the idea was guessed
  • Check: names, dates, the forged postscript story and the Great Cipher break

Next chapter: Telegraph, Wire and Codebooks, with commercial codebooks, the Zimmermann Telegram and the weakness of reuse.