John Churchill, Duke of Marlborough since late 1702, led a coalition successfully in the campaigns of the War of the Spanish Succession from 1702 to 1711. During his absence on the continent, he kept correspondence with Sidney Godolphin, his personal friend and political ally, and his beloved Sarah, Duchess of Marlborough.
In terms of cryptography, his ciphers are not of much interest. They merely assign numbers to words, mostly proper names. Although the original ciphers are not preserved, Ciphers A-G reconstructed from the existing papers are printed in The Marlborough-Godolphin Correspondence.
From 1701 to 1707, a new cipher was used every year.
Cipher A used in 1701 includes code numbers such as 1 (King William III), 6 (Godolphin), 19 (Marlborough), etc. In 1701, Marlborough was trusted with negotiations in The Hague in forming a coalition.
Cipher B used in 1702 has 62 (Marlborough), 75 (Godolphin), 79 (Queen Anne), etc.
Cipher C used in 1703 has 26 (Godolphin), 51 (Marlborough), 85 (Queen Anne), etc.
Cipher D used in 1704 and early in 1705 has 1 (Queen Anne), 16 (Godolphin), 86 (Marlborough), etc.
Cipher E used in 1705 from 23 April/4 May, when Marlborough left The Hague and set on a campaign of the year, has 2 (peace), 71 (Marlborough), 72 (Godolphin), 79 (Queen Anne), etc.
Cipher F used in 1706 has 4 (war), 5 (peace), 83 (Queen Anne), 90 (Marlborough), 91 (Godolphin), etc.
Cipher G used from 1707 to 1711 has 38 (Godolphin), 39 (Marlborough), 42 (Queen Anne), 80 (war), 81 (peace), etc.
Somehow, yearly change of cipher ceased after 1707 and Cipher G was used till 1711, when Marlborough was dismissed from his offices.
Cipher G was not only used for multiple years but also given to other correspondents such as Arthur Maynwaring, Robert Walpole, Horatio Walpole, James Craggs the elder, and Thomas, Lord Coningsby.
Use of cipher was not only for protection against foreign enemies but also for security from political enemies in Westminster.
Apparently, the Duchess of Marlborough prepared a private cipher for use only between her and her husband and gave it to the Duke when he embarked at Margate.
Such a private cipher was never used and the cipher used in this letter was Cipher G.
On the other hand, Queen Anne appears to have had a special cipher with Marlborough. A letter of Queen Anne to Marlborough dated 25 August 1707, printed by Lever, uses the following numbers: 4 (Harley), 10 (Godolphin), 17 (the Queen), 18 (the Whigs), 19 (the Tories), 40 (Marlborough), 41 (Duchess of Marlborough). These are different from Cipher G used in this year.
A letter in 18 June 1708 from Queen Anne to Marlborough also uses 10 (Godolphin).
While use of cipher between the Marlboroughs and Godolphin are modest with only some names and a few words are written in figures, a letter from Marlborough to Prince Eugene in 11 June 1705 is known to be "principally in cipher" (Coxe i, p.280).
Need for cipher by no means ceased when Marlborough was dismissed from offices in late 1711. Coxe prints a letter from the Duchess of Marlborough to Mrs. Clayton (her friend in England) dated 13 September 1713 (OS), when the Marlboroughs were in exile in Antwerp. Coxe guesses the numbers as follows: 7 (the Queen), 8 (the Electress of Sophia), 9 (electral prince), 11 (the pretender), 12 (the pope), 17 (Lord Oxford), 18 (Lord Bolingbroke), 19 (Lord Chancellor Harcourt), 36 (Lady Masham), 59 (England), 88 (the Tories), 89 (the Whigs), 140 (the king of France).
The author thanks his son, then yet to be two years old, for reminding me of these ciphers by picking up a volume of Trevelyan from the bookshelf and opening the plate of Marlborough's letter with cipher (above).
Henry S. Snyder (ed.) (1975), The Marlborough-Godolphin Correspondence (3 vols.)
William Coxe (1847), Memoirs of the Duke of Marlborough (Vol. 1, Vol. 2, Vol. 3)
Sir Tresham Lever (1952), Godolphin, His Life and Times
Trevelyan, George Macaulay (3 vols.)(1930, 32, 34): England Under Queen Anne