Cracking the Historical Codes

Times Literary Supplement, 8 Feb. 1980 © The Estate of Eric Sams

   In 1977 Julian Moore and I published (in the TLS of March 4) the only article yet written about how to crack historical codes and cipher, including early shorthand systems. There followed consultations with libraries and researchers about their cryptic manuscripts but the topic itself has remained largely obscure. I feel that the subject should attract those who like myself were wartime code-breakers and many others, especially historians, would find it a valuable pursuit, as I hope to show in this account of my recent practice.

   Of course some archives are likely to remain dark and impenetrable. William Friedman, one of the world’s greatest cryptanalysts, spent many a fruitless hour on the Voynich manuscript, attributed to Roger Bacon, which is fluently written in a natural-looking yet wholly unintelligible language. The British Library has a photocopy, and also owns an original volume of an equally obscure manuscript which begins by saying in plain English that no one will ever unravel the meaning of what follows. So be it; many tracks lead into such caves, but none ever come out; I for one prefer not to enter. The true treasure-chests, in my view, are much more likely to be those which clearly once had real keys, later lost or mislaid. Such cases can often be opened by the simplest tools of cryptanalysis, such as counting and classification. Thus Archbishop Laud in a letter of September 1636, also in the British Library, uses substitution cipher in an artless way. He writes en clair “Pray God some have not a hand in this that you little suspect, for I hear there is”, and he then concludes in cipher, “a successor designed”. The sample word 71 54 33 32 14 72 71 49 70 (“successor”) illustrates both his system and its basic simplicity.

   State ciphers were more sophisticated, but essentially similar and hence vulnerable to systematic scrutiny. The longwindedness of Henrietta Maria and Charles I proved very helpful in breaking their cipher letters. Hers to him of January 1646, now in the Public Record Office, is eloquent of their confusion and despair on the personal as well as the political plane. Her customary and touching salutation remains, “mon cher Coeur”; but she threatens to retire to a convent if her efforts on her husband’s behalf are not received more appreciatively. His letter of January 1643, now in the British Library, turns out to be a fervent appeal to his envoy in Paris for “ten thousand good armes such as wee shall chose there to be brought and provided for our use forthwith freely and without any hinderance lett or trouble to be brought over hither to such a place in this our kingdom as wee shall direct”, and so forth. However, these negotiations seem to have moved at an even more deliberate pace than the royal prose.

   In later centuries private diarists continued to devise their own personal cipher-systems, which provided effective cover then and later. Thus the National Library of Scotland has a diary with previously unread cipher-entries alluding to, for example, the Duke of Newcastle's appointment to the Treasury in 1754 and the disquieting rumour that “Mr Fox wanted to have some of the secret service money and to see the plans of the elections”. Both applications were rejected. So were the amorous appeals of the London worthy Henry Kirk, whose secret journal c 1818 has long remained untranscribed in the British Library, though its cipher is none too demanding and the excerpts I have decrypted are by no means barren of human or historical interest.

   Both  these journals, as it happens, use cipher-symbols akin to those of contemporary shorthand, perhaps with the aim of baffling any interceptor. Conversely, the early shorthands themselves were quite like some contemporary cipher-systems. One main difficulty in gaining access to the former lies not in the absence of keys but in the presence of a huge and jangling bunch of them. In the last four centuries, at least 300 different shorthands have been published in England alone, and many others will have circulated in manuscript form. Further, each such system could be altered or adapted at will by each user.  Even the printed manuals were often revised by their authors in later editions, or plagiarized by others. So it is usually impossible to begin by identifying the system and consulting the source-book.

   It is however entirely feasible, given enough shorthand text, to de­crypt the system by means of analysis and induction, without ever seeing the published manual or indeed knowing of its existence. The greatest triumph of that approach, namely the elucidation of the Pepys shorthand diaries by William Wyndham, Lord Grenville, in 1818, has still not had its full due, even in the magisterial Latham edition. Credit is often wrongly assigned, for instance, to Grenville’s brother Thomas, who was merely an intermediary, or to the otherwise unknown John Smith, who was clearly the transcriber not the solver of a shorthand writing which had long been obsolete and required reconstruction from first principles.

   That same procedure enabled me to decrypt the shorthand system used by William Clarke as secretary to Cromwell’s army. It too had been left unread, and indeed shelved as unreadable. Yet some of its entries are as lively as anything in Pepys, in a style equally formed by the fluent and familiar brevity of shorthand writing.


Leith, 25 February 1651/2. This day one Wragge who was formerly a sutler in the army having a wife in England and since his coining in Scotland married another here which being discovered with was this day sentenced by a court-martial to be tied down to the gallows and after that to have twenty stripes from the main guard to the sand port and so turned out of town which sentence was executed this day accordingly.


- a vivid glimpse of the disciplines of a Puritan army. The Clarke journals have now been published complete on microfilm, and their extensive shorthand passages await transcription.

   There are many more such discoveries to be made. Success requires only the skills of the cryptic crossword solver; there is no  need for any stenographic or historical expertise. The background story is straightforward enough. A book of so-called Charactterie by Timothy Bright appeared in 1588; and this retains an honoured place as the precursor not only of shorthand, with the rudiments of phonetic spelling and contractions, but also of basic English, thesaurus. classification and artificial language. But it depends first on the memorization  of some 500 separate symbols for individual words, and then on the use of a laborious and ambiguous method for recording their synonyms or antonyms. It seems manifestly impracticable, despite contemporary claims, as a method of taking dictation.

   This applies also to the Writing Schoolmaster of 1590 by Peter Bales, which is essentially the Bright system simplified. Some Shakespeare scholars have contended that the “stoine and surreptitious copies” of which the First Folio complains were procured by such means. But if plays were indeed pirated by shorthand, a far more likely method is that of John Willis, whose Art of Stenography in 1602 introduced that word into the language as well as the art itself into the modern world. All later systems are much indebted to its basic principles, which include an alphabet of letter-symbols and the idea of showing internal vowels by position only. Although still primitive and unwieldly, this shorthand was both viable and durable, as evidenced by its use in a diary c 1625, which is now being transcribed. Its successor, the Brachigraphy of Edmund Willis, 1618, was apparently more widespread and certainly more serviceable. It figures, in fluently cursive use, in a Bodleian manuscript of the period and also as marginalia on a privately owned copy of the Eikon Basilike: The Portraiture of His Sacred Majesty in His Solitudes and Sufferings, 1648.

   That thirty-year gap between source and sample is not uncommon. The shorthand manuals were no doubt handed down within a family; not only the convenience but the commercial advantages of the new idea were readily recognized. Thomas Shelton’s Short Writingc 1630 was popular enough to go through several editions over three decades; it was used by both Pepys and Clarke, as well as by unknown hands in literary and poetic manuscripts now in the Bodleian Library.  One of these includes several verses of Milton’sOde on the Morning of Christ's Nativity, together with less familiar works.

   Other  mid-seventeenth-century shorthands to survive the test of practical usage include the Brachigraphy of Henry Dix, in British Library manuscript correspondence about the marriage of Richard Cromwell, while Shelton ‘s second invention, the Zeiglographia, and theStenography of Thomas Metcalfe, occur as marginalia on printed books. The Charactery of Jeremiah Rich figures in an interest­ing undated manuscript now in private hands. It contains the royal monogram of Charles I, together with details of a secret cipher which he is known to have used in correspondence. But it consists mainly of agonized and self-abasing prayers written in the Rich system. Their text is largely formular and unrevealing; but occasional phrases such as “my people”, “my duty” and “my danger” suggest that the voice of the king himself may perhaps be recorded here, just as it later was (by William Clarke among others) on the scaffold.

   As the practice of shorthand grew and spread down the centuries and across the world, its infinitely variable symbols registered the same constant themes. Prayer and pious meditation are recurring keynotes. When the preacher James Humphreys left his Massa­chusetts home in 1776 to fight for American independence, he took with him his knowledge of James Weston’s Stenography Completed, 1727. That method enabled hint to chronicle his devotions as well as his campaigns, in a manuscript now preserved by the New York Historical Society. A journal kept in the 1780s by the Rev Alexander Ewing, once a rector in Devonshire, has emigrated into the archives of Bermuda; there too, personal prayers are couched in the intimacy of stenography, this time that of John Byrom’s Universal English Shorthand, 1767.

   Again, the Rev James Hawkes in a Bodleian manuscript of the early nineteenth-century notes his homilies and sermons in the Shorthand Unmask'd of Henry Barmby, c 1789. In general, the early systems served especially to express the writer’s most intense and inward thoughts, whether of religion, politics or love. So shorthand writing is predictably used for concealment as well as con­venience. Thus in 1851 the once renowned phrenologist George Combe examined the bumps of Charles Bray, close friend and mentor of George Eliot, and diagnosed Vigorous Amativeness. Bray's sup­porting testimony is discreetly recorded in a shorthand note, which begins “At twelve years of age he was seduced by his father’s cook”. The Combe journals are now in the National Library of Scotland; their occasional passages in the Universal Stenography of William Mayor, 1800, may well reveal further vignettes of contemporary life.

   The most recently written of the shorthands I have explored com­bines many of these typical features. Its chief topics are politics and love; its method, J. Dodge’s Complete System of Stenography, 1823, was still being put to everyday use nearly seventy years later by Sir John S. D. Thompson, Prime Minister of Canada. No doubt both the source-book and the skill had been acquired from his father, a practised shorthand writer. A draft of December 1892 now in the Public Archive in Ottawa begins, with exemplary suavity: “I have always felt  that the most agreeable part of the duties falling upon one who should choose a new Cabinet was the horrible task of parting with former colleagues for whom feelings of respect and confi­dence had grown up the stronger by years of mutual confidence…” and so on. In other words, you're sacked. The same flowing hand pens the tenderest of love-letters with a limpid readability clearly indebted to the shorthand system itself.

   Better safeguards were already available from Pitman and other systems relying on the thickness, slope and length of lines, and detailed phonetic treatment of vowels and diphthongs. Such texts remain hard to read, in every sense. But even these are far from unbreakable, as evidenced by such Cambridge University Library holdings as a schoolboy’s diary of 1849 kept in the Phonography of Isaac Pitman, c 1840, and Lord Kelvin’s notebooks 1879-81 in the much-modified Pitman of c 1870. The encouraging fact is that any and every kind of short or secret writing may be persuaded to disclose thoughts and facts that have long lain hidden by proce­dures that will certainly entertain the amateur cryptanalyst and with results that may well benefit the literary or historical researcher.