THE CODEBREAKERS
Page 25
A good glimpse into the achievements of the Geheime Kabinets-Kanzlei is afforded by the letters of one of its best directors, Baron Ignaz de Koch, who served from 1749 to 1763 with the cover-title of secretary to Maria Theresa. On September 4, 1751, he sent to the Austrian ambassador in France some cryptanalyzed correspondence which “makes one see more and more the main principles that direct the cabinet in France.” Two weeks later, in referring to some other cryptanalyses, he wrote, “This is the eighteenth cipher that we have got through during the course of the year; … we are regarded, unhappily, as being too able in this art, and this thought makes the courts that fear that we can engross their correspondence change their keys at every instant, so to speak, each time sending ones more difficult and more laborious to decipher.” Among letters solved during its existence were those of Napoleon, Talleyrand, and a host of lesser diplomats. These solutions were often made the basis of Austrian strategy.
England, too, had its black chamber. Its origins may be found in the cryptanalyses of a young man who stumbled into cryptology at the same age and at about the same time that Rossignol did, and who may be considered his counterpart. This was John Wallis, better known as the greatest English mathematician before Newton.
He was born on November 23, 1616, in Ashford, Kent, where his father was rector. He studied at Emanuel College, Cambridge, became a fellow of Queen’s College there, and was ordained a minister. He was known to divert himself with arithmetical problems, and one evening early in 1643, when he was serving as chaplain to the widowed Lady Vere, a gentleman brought him a letter that had been found after the capture of Chichester by a parliamentary army during the Puritan Revolution. Wallis told him that he could not tell whether he could solve it or not, “Adding withall, that if it were nothing but barely a new Alphabet, as at the first Sight it seemed to bee, I thought it might possibly bee done. The Gentleman,” Wallis wrote afterward, “who did not expect such an answer, told mee Hee would leave it with all his Heart, if I had any Thought of reading it: And accordingly did so. After Supper (for it was somewhat late in the Evening when I first saw it) having a while considered what Course to take, I set about it, and within a few Houres (before I slept) I had overcome the Difficulty, and transcribed the Letter in a legible Character. This good Successe upon an easy Cipher (for so it was) made me confident, that I might with the like ease read any other, which was no more intricate than that.”
But the next one, a numerical, was so much more difficult than the first, which was a long monalphabetic with word divisions, that Wallis turned it down. His career would have been nipped before it had budded—except that, soon thereafter, Wallis was somehow prevailed upon to try a cryptogram that had lain about for two years because no one could be found to solve it. The cipher numbers in this letter ranged beyond 700, and, wrote Wallis, as the first cryptogram “was one of the easiest, so this second was one of the hardest that I have ever met with.” Several times he gave it up as “desperate,” but after about three months, “I did at last overcome the Difficulty.”
This feat made his fortune. At the behest of Parliament, he solved in 1643 some of the dispatches of Charles I during the civil war, and was rewarded, first, with the living of St. Gabriel’s Church in London, and then with the place of secretary to the Westminster Assembly. Additional solutions for Parliament led to his appointment in 1649 as Savilian professor of geometry at Oxford at the age of 32. In 1660, Charles II’s negotiations with Presbyterian ministers in London for his Restoration were “made known to mee,” said Thomas Scot, director of intelligence for Parliament, “first by one Mr Harvy, since dead and after by Major Adams, who kept them daily Company here, but very much more by letters intercepted which commonly were every word & syllable in Cypher, and decyphered by a learned gentleman incomparably able that way, Dor Wallis of Oxford (who never concerned himself in the matter, but only in ye art & ingenuity); it is a jewell for a Princes vse and service in that kind.” Though Charles knew of Wallis’ parliamentary services through Scot’s confession, he found him so valuable that soon after he ascended the throne he was employing the man who had recently worked against him. Indeed, so indispensable did Wallis prove that Charles compensated him not only with small sums of bounty but eventually with an appointment as a chaplain to the king.
Wallis seems to have been largely self-taught. He studied the works of Porta and others, but learned little from them because, he says, they chiefly treated of methods of encipherment. “So that I saw, there was little Help to bee expected from others; but that if I should have further occasions of that Kind, I must trust to my owne Industry and such observations as the present Case should afford. And indeed,” he continues perceptively, “the Nature of the Thing is scarce capable of any other Directions; every new Cipher allmost being contrived in a new Way, which doth not admit any constant Method for the finding of it out.”
The mind that, without aid, could find its way so unerringly through the labyrinth of cryptology could also blaze new trails in the unexplored fields of mathematics. This Wallis did. His Arithmetica Infinitorum arrived at results that Newton used as a springboard to develop the calculus and that contained the germ of the binomial theorem. Wallis invented the symbol ∞ for infinity, and he was the first to give the value of π by interpolation—a term, incidentally, that he coined. In later years, he taught himself to calculate mentally to while away sleepless nights, performing such astounding feats as extracting the square root of a number of 53 digits and dictating the answer (which proved correct) to 27 places.
Hale and vigorous of body, of medium height with a small head, he was set down by that vivid chronicler John Aubrey as “a person of reall worth” who “may stand very gloriously upon his owne basis, and need not be beholding to any man for Fame, yet he is so extremely greedy of glorie, that he steales feathers from others to adorne his own cap; e.g., he lies at watch at Sir Christopher Wren’s discourses, Mr. Robert Hooke’s, Dr. William Holder, &c; putts downe their notions in his Note booke, and then prints it, without owneing the author. This frequently, of which they complain.” Wallis helped found the Royal Society, but despite his accomplishments, Samuel Pepys, who met him on December 16, 1665/6, noted in his diary, “Here was also Dr. Wallis, the famous scholar and mathematician; but he promises little.”
Wallis’ most active period, cryptologically, came late in life, when he was employed as cryptanalyst to William and Mary. In 1689, he reported to William’s Secretary for War, the Earl of Nottingham, that in the past two weeks he had forwarded three packets of solutions and now had lying before him five letters in three or four different ciphers, all new to him. He told Nottingham, who was always pressing him for solutions, that he could not yet give the plaintext of some cryptograms, though, he went on, “I have already employed about seven weeks on them, and have studied hard thereupon eight or ten hours in a day, or more than so very often, which, in a business of this nature, is hard service for one of my years [then more than 70] unless I would crack my brains at it.” Nottingham, in fact, was once so anxious to have a solution to letters from Louvois to one of his generals that he told Wallis that he had ordered the messenger who brought the cryptograms to wait until Wallis had solved them. Wallis managed to break the nomenclator in four days and, on returning the plaintext, tactfully apprised Nottingham of a few cryptologic realities to explain why he had let the messenger go.
John Wallis’ solution of a dispatch of Louis XIV of France of June 9, 1693
His solutions—nearly all nomenclators, a few monalphabetics—had a considerable impact on current events. In the summer of 1689, he solved the correspondence between Louis XIV and his ambassador in Poland. In one dispatch, Louis was caught urging the King of Poland to declare war against Prussia with him; in another, he was discovered promoting a self-serving marriage between the Prince of Poland and the Princess of Hanover. Wallis described the value of this work in a letter asking for a raise: “The deciphering of some of those letters having quite broke a
ll ye French King’s measures in Poland for that time; & caused his Ambassadors to be thence thrust out with disgrace. Which one thing,” he adds pointedly, “was of much greater advantage to his Matie & his Allies, than all that I am like to receive on that account.”
Though Wallis entreated Nottingham not to publicize his solutions for fear France would again change her ciphers, as she had done nine or ten times before (probably under the expert Rossignol tutelage), word of his prowess somehow spread. The King of Prussia gave him a gold chain for solving a cryptogram, and the Elector of Brandenburg a medal for reading 200 or 300 sheets of cipher. The Elector of Hanover, not wanting to depend on a foreign cryptanalyst, got Wallis’ fellow intellectual, Baron Gottfried von Leibnitz, to importune him with lucrative offers to instruct several young men in the art. When Wallis put off Leibnitz’ query as to how he did these amazing things by saying that there was no fixed method, Leibnitz quickly acknowledged it and, hinting that Wallis and the art might die together, pressed his request that he instruct some younger people in it. Wallis finally had to say bluntly that he would be glad to serve the elector if need be, but he could not send his skill abroad without the king’s leave.
The shrewd old cryptanalyst, who was frequently asking for more money for his solutions, then used Leibnitz’ arguments to his own advantage in successfully urging the secretaries of state to pay for his tutoring of his grandson in cryptanalysis. They agreed in 1699, but it was not until Wallis wrote to the king in 1701, saying that the young man had made such good progress that he had solved one of the best English ciphers and a very good French one, that they were jointly granted £100 a year, retroactive to 1699.
Wallis’ career thus strikingly parallels Rossignol’s. The two men were approximately contemporaneous (Rossignol was not quite 17 years older). Both made their first cryptanalyses in their late twenties on ciphers stemming from civil warfare in their countries. Both had a mathematical bent, and both were largely self-taught in cryptology. Both owed their worldly success to this unusual talent. Both lived into their eighties. And both became their countries’ Fathers of Cryptology in a literal as well as a figurative sense. There were differences, of course. Rossignol had to assist at the more autocratic French court; Wallis seems to have done most of his work at Oxford and in other country places far from London. Rossignol probably supervised French cryptography, but Wallis apparently prescribed an English cipher only once, and that very informally. It is therefore unlikely that these cryptologic titans of the two most powerful and most contentious countries of Europe ever clashed cryptologically. Thus the problem as to who might have been the better must remain—unlike the cryptograms to which they addressed themselves with such success—forever unsolved.
On Wallis’ death on October 28, 1703, the grandson whom he had tutored, William Blencowe, an undergraduate at Magdalen College, assumed the cryptanalytical duties, though he was only 20 years old. His grandfather’s tutorial fee of £100 reverted to him as Decypherer, and he thus became the first Englishman both to bear that title officially and to be paid a regular salary for cryptanalysis. Blencowe did so well that six years later this salary was doubled, and he stood high in the royal favor: Queen Anne intervened in his behalf during a dispute with All Souls College at Oxford, where he was a fellow. But he shot himself in a fit of temporary insanity during his recovery from a violent fever in 1712. He was succeeded by Dr. John Keill, 50-year-old professor of astronomy at Oxford, who, though a fellow of the Royal Society, proved totally incompetent. On May 14, 1716, Keill was replaced by Edward Willes, a 22-year-old minister at Oriel College, Oxford.
Willes embarked at once upon a career unique in the annals of cryptology and the church. He not only managed to reconcile his religious calling with an activity once condemned by churchly authorities, but also went on to become the only man in history to use cryptanalytic talents to procure ecclesiastical rewards. Within two years, he had been named rector of Barton, Bedfordshire, for solving more than 300 pages of cipher that exposed Sweden’s attempt to foment an uprising in England. He virtually guaranteed his future when he testified before the House of Lords in 1723. Here, Francis Atterbury, Bishop of Rochester, was being tried by his peers for attempting to set a pretender on the English throne.
The pretender’s cause exhorted the allegiance of many in England, and the nation’s attention focused on Atterbury’s trial. Most of the facts about the alleged conspiracy had come from his intercepted correspondence, and the most inculpatory evidence had been extracted from the portions in cipher by Willes and by Anthony Corbiere, a former foreign service official in his mid-thirties who had also been appointed a Decypherer in 1719. The Lords “thought it proper to call the Decypherers before them, in order to their being satisfied of the Truth of the Decyphering.” To demonstrate this, Willes and Corbiere deposed,
That several Letters, written in this Cypher, had been decyphered by them separately, one being many Miles distant in the Country, and the other in Town; and yet their Decyphering agreed;
That Facts, unknown to them and the Government at the Time of their Decyphering, had been verified in every Circumstance by subsequent Discoveries; as, particularly, that of H-----’s Ship coming in Ballast to fetch O----- to England which had been so decyphered by them Two Months before the Government had the least Notice of Halstead’s having left England;
That a Supplement of this Cypher, having been found among Dennis Kelly’s Papers the latter End of July, agreed with the Key they had formed of that Cypher the April before;
That the Decyphering of the Letters signed Jones Illington and 1378, being afterwards applied by them to others written in the same Cypher, did immediately make pertinent Sense, and such as had an evident Connexion and Coherence with the Parts of those Letters that were out of Cypher, though the Words in Cypher were repeated in different Paragraphs, and differently combined.
The two Decypherers appeared before the Lords on several occasions to swear to their solutions. Atterbury twice objected and was twice overruled. But on May 7, as Willes was testifying to the cryptanalysis of the three most incriminatory letters of all, and the bishop felt the noose tightening around him, he persisted in questioning Willes on the validity of the reading though the House had supported Willes’ refusal to answer. He raised such a commotion that he and his counsel were ordered to withdraw, and the Lords voted upon the proposition, “that it is the Opinion of this House that it is not consistent with the public Safety, to ask the Decypherers any Questions, which may tend to discover the Art or Mystery of Decyphering.” It was resolved in the affirmative, the solutions were accepted, and Atterbury, largely on this evidence, was found guilty, deprived of office, and banished from the realm.
Willes, on the other hand, became Canon of Westminster the next year. His salary more than doubled to £500. He succeeded to ever more important posts every four or six years thereafter, and finally, in 1742, when the oldest of his three sons, Edward, Jr., obtained a patent as a Decypherer, he was created Bishop of St. David’s, being translated the next year to the more prestigious see of Bath and Wells. The bishop and his son shared the substantial salary of £1,000 a year. In 1752, he brought another son, William, into the business at an eventual £200, and six years later a third son, Francis, who for some reason served without pay.
Bishop Willes died in 1773 and was buried in Westminster Abbey. His sons Edward, Jr., and Francis inherited a large share of his fortune and landed property and, living as wealthy squires at Barton and Hampstead, continued their cryptanalytic work. Their brother William had retired in 1794, but his three sons, Edward, William, and Francis Willes joined the Decyphering Branch in the 1790s.
Though the Willes family dominated the cryptanalytic branch, others worked in it. Corbiere was paid through such sinecures as his appointment as naval officer at Jamaica, though he never stirred from England, and as Commissioner of Wines Licenses, which sounds like the cushiest of posts. He rose to Under Secretary of the Post Office but continued h
is cryptanalytic work, which ended after 24 years only with his death in 1743, when he was receiving £800. The other cryptanalysts at various times were James Rivers, Frederick Ashfield, John Lampe, George Neubourg, John Bode, Jr., one Scholing, and a Boelstring.
These men received their foreign interceptions from the Secret Office and their domestic ones from the Private Office, both subdivisions of the Post Office. The Secret Office was quartered in three rooms adjoining the Foreign Office and entered privately from Abchurch Lane. Fire and candles burned constantly in one room; the staff lodged in the others. It included men who made their life’s work the specialty of unsealing diplomatic packets with such deftness that they could be resealed without evidence of tampering; one such opener was J. E. Bode, father of John Bode, Jr. He regularly spent three hours on the dispatches of the King of Prussia, opening them and then resealing them with special wax and carefully counterfeited seals. Perhaps surprisingly in a bastion of human rights, its interceptions enjoyed full legality. The statute of 1657 that established the postal service declared outright that the mails were the best means of discovering dangerous and wicked designs against the commonwealth. Leases of 1660 and 1663, confirmed by the Post Office Act of 1711, permitted government officials to open mail under warrants that they themselves issued. They sidestepped this bothersome procedure by promul gating all-inclusive general warrants.* The Secret Office sent interceptions en clair to the king and those in cipher to the cryptanalysts.
They were known collectively as the Decyphering Branch. Unlike the Secret Office, the branch had no specific location. Its tiny staff of experts worked largely at home, receiving their material by special messenger. Nor had it any formal organization, the senior Decypherer being merely first among equals. More secret than the Secret Office, the branch’s funds came from secret-service money issued to the Secretary of the Post Office from Parliament’s surplus revenue. Security was tight—in all of England probably only 30 people knew what diplomatic correspondence was being read at any given moment. Nevertheless, most men of affairs were aware of the practice of opening private letters, and they often enciphered their correspondence or entrusted it to private messengers when secrecy was essential.