The Queen's Agent: Francis Walsingham at the Court of Elizabeth I (34 page)

‘The cipher I had from Thomas Morgan in France.’ Thus Francis Throckmorton described to his interrogators how he had been communicating with Mary Stuart and her English supporters in Paris. William Parry had also consorted with Morgan during his time in France. According to Holinshed’s
Chronicles
, Parry confessed to having bragged to Morgan of his resolve to do the Catholic Church some great service, even to kill the greatest subject in England. Morgan’s reply was blunt: why not the queen herself? We have glimpsed Morgan before, trailed by rival Walsingham and Stafford surveillance operations in Paris, and we shall soon meet him again as a cog in Anthony Babington’s treason of 1586. Morgan it was who recruited the Douai student and double agent Gilbert Gifford to be a courier for the Queen of Scots. Remarkably, he played his part in the Babington plot from within the walls of the Bastille, where he had been shut up at Queen Elizabeth’s request following the Parry plot. In France as in England, a prison could be an effective headquarters for an aspiring revolutionary.

Thomas Morgan was the
éminence grise
of the expatriate Catholic resistance during the 1580s. Like Parry he was a Welshman from a minor gentry family, and was forced by 
dwindling finances to shift for himself. Early in Elizabeth’s reign he worked as a scrivener or clerk for William Alley, Bishop of Exeter, and then as secretary to Archbishop Young of York. These were odd connections for a man who later claimed to be born of Catholic parents, since both Alley and Young were paragons of the Protestant establishment. Stranger still, Thomas Morgan’s life was the reverse of that of his own brother Rowland, who began his career as a Catholic seminary priest and ended up in the Church of England. For Thomas, the moment of epiphany was his meeting with the Queen of Scots when he joined the Earl of Shrewsbury’s household in 1568. His favours to Mary – tip-offs when her rooms were to be searched, help in hiding suspect papers – were sensed by the government at the time of the Ridolfi plot, and Morgan spent much of 1572 in the Tower. On his release, bail set at £10,000 could not prevent him from returning to the Scottish queen’s service. In 1575 Morgan was named in connection with Henry Cockyn, a London stationer and bookseller who had been using his shop as a post office for Mary’s letters, and Walsingham ordered his arrest once again. But this time he was more agile, and slipped away to Paris. In 1581 Mary began paying him a pension and found him a place in the household of Archbishop Beaton, her ambassador in France.
⁷

Beaton put Thomas Morgan to work as a cipher clerk. Queen Mary used her Paris embassy to petition sympathetic powers abroad, and the bulk of this correspondence flowed through Morgan’s hands. Given the precariousness of Mary’s position in English custody, security of communication was paramount. Letters carried obvious risks. Couriers could be arrested and pressed in the Tower to reveal what they knew. Royal searchers at the south-coast ports knew where to look for papers concealed in imported bales of cloth, or bottles of wine, or the bindings of books. Once intercepted and digested, incriminating documents 
were often released to their intended destination to encourage plotters to betray themselves. Walsingham’s servants included the forger Arthur Gregorye, who could restore the wax seals on letters ‘that no man could judge they had been opened’.

Expecting to be searched, agents developed methods of concealing the secrets that they carried. In a treatise dating from the mid-sixteenth-century, the scientist and occultist Giovanni Battista della Porta of Naples explained how a solution of an ounce of alum and a pint of vinegar could be used to write a message on the inside of an egg, invisible until the shell was broken. Alum was a familiar compound to the cloth trade as a mordant or dye-fixer, and quantities were imported to feed the English wool industry. It was available to the imprisoned Queen of Scots, who excitably forwarded a recipe for invisible ink to the French ambassador in London: ‘the paper must be dipped in a basin of water, and then held to the fire; the secret writing then appears white, and may easily be read until the paper dries’. We know this because the mole in Castelnau’s household passed Mary’s letter on to Walsingham, where it remains in his archive. In February 1586 Arthur Gregorye reported to Walsingham ‘from my poor house, half blind’ on his own investigations into the chemistry of secret writing. Alum, he explained, could indeed be ‘discovered’ by fire and water, but best of all with coal dust. At the foot of his letter, Gregorye gave his master an exercise to do. ‘If your honour rub this powder within the black line,’ he wrote, ‘the letters will appear white’. Walsingham did as he was bid: four lines of Latin appear palely through a smudge of soot, no longer legible even under ultraviolet light.
⁸

Other agents experimented with crushed onions, or citric acid. Walter Williams reported from Rye prison in 1582 on ‘the plan of the traitors for conveying intelligence by secret writing with orange juice’. When nothing else was available, they used their own urine. Meanwhile Walsingham’s constant battle 
against ill-health meant the hiring of physicians and druggists, and these could have briefed him on the chemistry of secret writing. One likely candidate is Roderigo Lopez, son of a Portuguese
marrano
or forcibly Christianised Jew, who treated Walsingham’s kidney stones in Paris in 1571 and went on to become physician to the queen. Lopez milked his cousins within the
marrano
merchant community for news that he could pass on to Walsingham, although he also gained a reputation as a poisoner. Another source of advice may have been the alchemist John Dee, whom Walsingham called ‘my very loving friend’ on the basis of their discussions about the Gregorian calendar and the exploration of a north-west passage to China.
⁹

The standard formulae for invisible ink were widely known, and offered limited defence against discovery. Secret writing was far more effective if encrypted by code or cipher. Italy led the field, attributable to the precocious diplomacy of its jostling city-states. The Venetian republic had three cipher secretaries in the 1540s, with an office located in the doge’s palace and a school to develop their skills. The pope appointed a cipher secretary from 1555, and the duchies of Florence and Milan employed cryptographers of their own. Francis Walsingham’s activities in Padua in the 1550s are thinly documented, but his Italian exile could well have offered training in espionage as well as the civil law. Meanwhile Philip of Spain, who married Mary Tudor in 1554, may have introduced sophisticated Spanish ciphers into England. In general, however, the English seem to have lagged behind Italy, France and the Habsburg states in their development of encrypted communication until Walsingham spotted its potential.
¹⁰

Innovation was forced by the arrival of the English Catholic mission in the mid-1570s. If priests and conspirators against the Elizabethan state were communicating by cipher, then those ciphers would have to be broken. Sometimes the Protestant 
powers of Europe shared intelligence among themselves. In summer 1577 the Prince of Orange informed the English diplomat Daniel Rogers about the plans of Don John of Austria to land a Spanish army in England under cover of taking refuge from a storm. The information had been captured by a Huguenot general and deciphered by Orange’s adviser the Baron de St Aldegonde. Rogers forwarded the news to Walsingham, who had co-operated closely with St Aldegonde during the latter’s embassy to England in 1576. In March 1578 Walsingham turned to St Aldegonde once again to decode a letter from the Portuguese ambassador. It was revealed to be a lengthy, and all too plausible, complaint that Queen Elizabeth was avoiding an audience by feigning illness. Relying on sympathetic foreign statesmen was not a long-term solution, however, and so Walsingham cast around for an English talent whom he could nurture. He found him in Thomas Phelippes.
¹¹

We can build up a fuller biography of Phelippes than some of Walsingham’s operatives, partly because his higher social rank has left more traces in the records. The son of a London cloth merchant, Phelippes may have studied at Trinity College, Cambridge in the early 1570s while maintaining his father’s financial interest in the customs house. He inherited a property in Leadenhall Street and owned others in Chiswick, Holborn and beyond. Phelippes was urban well-to-do rather than landed gentry. The royal court was a natural place for him to seek employment. In class terms he was closer to Christopher Marlowe than to Walsingham’s field agents like the ex-soldier Maliverny Catlyn or the gentleman’s servant Nicholas Berden. A Thomas Phillips was nominated as burgess for Hastings in the 1584 and 1586 sessions of Parliament, and this may well be the decipherer. Unusually we have contemporary accounts of Phelippes’s temperament (Elizabethans would have said ‘humour’) and appearance. The first, from his father, comments on his ‘staid 
and secret nature’. The second comes from Mary, Queen of Scots, who met Phelippes when he came to stay at Chartley and tried to bribe him. She describes his slight build, his blond hair and beard, his shortsightedness and the smallpox marks that ravaged his face. Perhaps to compensate for his looks, he had taught himself to be a skilled mimic.

Phelippes had a gift for languages, ancient and modern. As with the crossword-solvers of 1940s Bletchley Park, this quickness with words must have been the foundation for his skills in cryptography. In June 1578 Principal Secretary Thomas Wilson wrote to Walsingham enclosing a ciphered letter with the advice that if St Aldegonde could not crack it, he should forward it ‘to your servant young Philips, who is with our ambassador at Paris’. The ambassador was Sir Amyas Paulet, whose Puritan outlook and profound suspicion of the Queen of Scots matched Walsingham’s own. When Paulet was recalled from the Paris embassy in 1579, Phelippes stayed on with his successor Sir Henry Cobham. In July 1582 he wrote to Walsingham from the university town of Bourges in central France, where he was toiling over an encrypted letter that was something to do with the Jesuits, ‘against whose practices the Lord defend us’. The cipher was causing him a lot of trouble: ‘these imperfect lines have been worn out of the hard rock. I have had to do, as you know, with many ciphers, but I never lit upon any wherewith I was more cumbered, nor wherein the observations which I serve myself of in these occasions did more fail me.’ Phelippes admitted he was proceeding by guesswork, and could only recommend that Walsingham force the messenger to ‘tell more plainly what is meant’.
¹²

In explaining why he found this commission so challenging, Phelippes provides a clue to his method of working. The Jesuit letter was written in Latin. Of itself this was no obstacle, since Phelippes could read and write Latin fluently. The problem, for
him, lay in the original rather than the ciphered version. Whoever wrote the document had a shaky grasp of Latin, and had made so many mistakes that Phelippes was thrown off the scent. Or perhaps the author was more subtle, deliberately misspelling words in order to frustrate an enemy intercepting his letter; this was a common tactic.

What were the ‘observations’ which normally served Phelippes so well? The answer qualifies him as perhaps the first English cryptanalyst, as distinct from cryptographer: someone with both the mathematical ability and the linguistic dexterity to apply frequency analysis to the decrypting of texts. The technique of counting the occurrence of individual letters or symbols in order to crack a cipher was known to the Arab-speaking world in the tenth century, but it took another five hundred years to percolate into Europe. By the time that Phelippes was working for Walsingham and Thomas Morgan for the Queen of Scots, cryptanalysis was becoming a science of its own, informed by parallel developments in algebra and linguistics. The makers of ciphers learned to change them regularly – Morgan is credited with forty separate alphabets for the Queen of Scots – while old ciphers were archived, and could be sent for if necessary.
¹³

Phelippes shared his fascination with the shape of language with another graduate of Trinity College, Cambridge, the physician Timothy Bright. Walsingham had sheltered Bright and other Protestants in the Paris embassy during the carnage of St Bartholomew in 1572, and Bright dedicated his abridged edition of Foxe’s
Acts and Monuments
to him. In 1585 a letter from Walsingham helped Bright secure a lucrative post at St Bartholomew’s Hospital in London, which gave him the security to pursue his research into alphabets and cryptograms. While neglecting his patients, Bright devised an ingenious system of shorthand writing based on a system of eighteen symbols with a series of hooks, loops and lines to vary their meaning, and a list 
of over five hundred ‘charactericall’ words to be memorised. In 1587 he translated sections of
De Furtivis Literarum Notis
, a study of cryptography by Giovanni Battista della Porta, on instructions from Walsingham’s assistant secretary of state William Davison. Bright may have served Walsingham as a personal physician, as the Portuguese Jewish intelligencer Roderigo Lopez had done.
¹⁴

At its edges, the study of secret writing shaded into the mystical. Della Porta found his researches being scrutinised by the Inquisition. Some seventy years earlier, the Benedictine abbot Johannes Trithemius had apparently set out a system for communicating across great distances by means of spirit helpers. His
Polygraphia
, published posthumously in 1518, dealt straightforwardly with the subject of cryptology. His earlier
Steganographia
, by contrast, was hotly pursued by philosophers and occultists because it was said to have been delivered by a spirit in a dream. Trithemius had known the original Dr Faustus, and wrote treatises on alchemy and witchcraft. When he tracked down a manuscript of
Steganographia
in Antwerp, John Dee spent a purseful of money and a frenzied ten days in a lodging house transcribing endless tables and bewildering lists of names. For Dee, secret writing meant the realm of angels and demons. His ‘skryer’ or spirit medium Edward Kelley duly provided him with a fragment of a language taught by God to Adam. What no one realised until Trithemius’s code was finally cracked in the 1990s is that the mysterious incantations taught in
Steganographia
are in fact codes and pangrams: wordplay rather than magic.
¹⁵