THE CODEBREAKERS

by DAVID KAHN

  759 varying frequencies of letters: L. F. Brosnahan, The Sounds of Language (Cambridge: W. Heflfer and Sons, Ltd., 1961), chs. 1-3; review by Morris Swadesh, “Pro and Contra Darlington,” Science, CXXXIV (September 1, 1961), 609.

  759 entropy: William Cecil Dampier, A Shorter History of Science (1944, reprinted New York: Meridian Books, 1957), 97-99, 168-169.

  760 Borges: “The Library of Babel,” in Labyrinths (New York: New Directions, 1962), 51-58.

  760 calculation of redundancy: Shannon, “A Mathematical Theory of Communication” and “Prediction and Entropy of Printed English”; George A. Miller and Elizabeth A. Friedman, “The Reconstruction of Mutilated English Texts,” Information and Control, I (September, 1957), 38-55; David Slepian, interview, October 28, 1962.

  Chapter 21 HETEROGENEOUS IMPULSES

  A citation in the form N&Q, 4:5:254, 285, means Notes & Queries, Series 4, Volume 5, pages 254 and 285. References to common works of literature are given by chapter instead of by the page of a particular edition.

  PAGE

  763 Myzskowsk: Cryptographie indéchiffrable (Paris: Société Française d’imprimerie et de librairie, 1902), at 44-46.

  763 Collon: articles in Revue de l’Armée Belge (1899); Sacco, §37.

  763 Rozier: Rosario Candela, The Military Cipher of Commandant Bazeries (New York: Cardanus Press, 1938), 76.

  763 Phillips: Gaines, 185-190.

  763 Amsco: Gaines, 51.

  763 Grandpré: Cryptographie pratique (Paris: Boyveau & Chevillet, 1905).

  763 Schneider: Description d’un système cryptographique à l’usage de l’armée (Paris: L. Fournier, 1912); W. F. Friedman, The Index of Coincidence and Its Application in Cryptography, Riverbank Publication 22 (Geneva, 111.: Riverbank Laboratories, 1922).

  763 Porges: “A Continuing Fraction Cipher,” The American Mathematical Monthly, LIX (April, 1952), 236.

  763 Nicodemus: Gaines, 216.

  763 Mirabeau: Gaines, 209.

  763 Homan: Communication to the author, 1948. This “equifrequency” cipher was broken by Howard T. Oakley; see also his “An Equi-Frequency System,” The Cryptogram (July-August, 1955), 60-62.

  764 cipher mechanisms: In the U.S. Patent Office, inventions dealing with cryptology are classed in the following classes and subclasses—what the Patent Office calls the “field of search for cryptography.” Not every patent in these classes deals with cryptography. Class 35, “Education,” Subclasses 2-4, for cryptography generally; Class 197, “Typewriting Mechanisms,” Subclass 4, for cryptographic typewriters; Class 283, “Printed Matter,” Subclasses 11 and 17, for cryptographic printed matter; Class 178, “Telegraphy,” Subclass 5.1 for secret facsimile and secret television, Subclass 22, for secret telegraphy; Class 179, “Telephony,” Subclass 1.5, for secret telephony; Class 325, “Modulated Carrier Wave Communication Systems,” Subclasses 32-35 and 122, for modulated carrier wave systems involving secrecy; Class 116, “Signals and Indicators,” Subclasses 18-20, for code signaling generally; Class 340, “Communications, Electrical,” Subclasses 345-365, for electrical communications involving code converters and transmitters; Class 234, “Selective Cutting,” Subclasses 69-70, 94-108, for cutting or punching involving coding. A list of the patents in each of these subclasses, as well as the patents themselves, may be purchased from the Patent Office. (U.S. Patent Office, letters, November 21, 1962, and November 17, 1965.) About 1950, Howard T. Oakley compiled a mimeographed List of U.S. Patents Dealing with Cryptography that gives a one-line description of the mechanism of many of the patents. Among the patentees, incidentally, is motion-picture star Hedy Lamarr, who, under her married name of H. K. Markey, patented, with a few others, a “Secret Communication System” to guide torpedoes to their targets by secret radio remote control; this is No. 2,292,287.

  764 transposition cipher machine: described at Gaines, 13; this appears to be that of Nicoletti, U.S. Patent No. 1,311,457.

  764 Jerdan: William Jerdan, Autobiography (London, 1852), I, 40-44.

  765 Euler: P[ierre]. Speziali, “Le logogriphe d’Euler,” Stultifera navis: Bulletin de la Société suisse de bibliophiles, X (April, 1953), 6-9.

  765 Skeat: N&Q, 8:9:6-7, 33, 58-59.

  765 fractionating cipher: This is described in Givierge with no inventor named. It turned out to be a bad choice to illustrate the ignorance of inventors, because I have since found that in the earliest exposition I know of this system (“Systèmes de Cryptographie,” Revue Scientifique [March 24, 1888]), the author, one Pomey, thoroughly analyzed the structure of the system mathematically. This did not lead him, however, to the solution.

  766 “Whereas complete trial”: Shannon, “Communication Theory of Secrecy Systems.”

  767 Byrne: J. F. Byrne, Silent Years: An Autobiography with Memoirs of James Joyce and Our Ireland (New York: Farrar, Straus, and Young, 1953), 264-307; Book Review Digest, 1953; Lynn Caine, publicity director of Farrar, Straus and Cudahy, telephone interview, May 28, 1962.

  767 Bloom’s cipher: James Joyce, Ulysses (Paris: Shakespeare and Company, 1922), 673. G. Smith, Jr., “The Cryptogram in Joyce’s Ulysses: a Misprint,” PMLA, LXXIII (September, 1958), 446-447, discusses errors in later editions.

  769 Houdini: Houdini on Magic, eds. Walter Gibson and Morris N. Young (1930, 1932; New York: Dover Publications, 1953), 244.

  769 American Cryptogram Association: membership from 1965 Directory.

  770 newspaper cryptograms: The two types are represented by the “Daily Cryptoquote” in Newsday and by the daily cryptogram in the regretted New York Herald Tribune.

  770 Trinity Church epitaph: Meyer Berger, “About New York,” The New York Times (January 2, 1957); Charles L. Wallis, Stories on Stone: A Book of American Epitaphs (New York: Oxford University Press, 1954), 202-203.

  770 church register cryptograms: N& Q, 14:157:134, 214.

  770 Cramer cryptogram: Pierre Speziali, “Une curiosité parmi tant d’autres,” Les Musées de Genève, VI (March, 1949), 1.

  771 Beale treasure: P. B. Innis, “The Beale Fortune,” Argosy, CCCLIX (August, 1964), 70-71, 82-84; photostats of Beale Paper cryptograms, Roanoke Public Library, Roanoke, Virginia; Carl Hammer, letter, November 25, 1964, and print-outs from computer, for which I am most grateful.

  772 Masons: King Solomon and his Followers (Brooklyn: Allen Publishing Co., 1908).

  772 Knights of the Golden Circle: NA.

  772 Phi Beta Kappa: “U Butdz ofn Hpw Tzbu Quhhu, Or, A Table for Phi Beta Kappa,” The Key Reporter, XXIV (July, 1959), 1.

  772 Carbonari: Intermédiate des Chercheurs et Curieux (March, 1954), col. 98, for 1834 cryptogram; Paul Féval, Les Compagnons du Silence (1857; Paris: Michel Levy Frères, 1861), Part 2, ch. 7, for key; Prologue, chs. 5, 8, 9, Part 2, ch. 8, Part 5, chs. 1, 5, 13, for enciphered messages. André Lange et E.-A. Soudart, Traité de cryptographic (Paris: Félix Alcan, 1925), discuss Féval cipher in Annexe X, 322-326. E. T. Bourg [pseud. M. Saint-Edmé], Constitution et Organization des Carbonari (Paris: Corby, 1821), does not give anything about Carbonari cryptology per se, though he does cite letters used as mystic signs (181-182).

  773 Lincei-Eck correspondence: Vatican Library, Ms. Vat. lat. 9684, ff. 23-26, 131-3, 140, 144-146.

  773 Galileo and Huygens anagrams: William F. and Elizebeth S. Friedman, The Shakespearean Ciphers Examined (Cambridge: University Press, 1957), 17. John Wallis duped Huygens with a number of anagrams; see Christiaan Huygens, Œuvres Complètes (La Haye: Martinus Nijhoff, 1888), I, 338, 396, 402; II, 306.

  773 Wren: Sir David Brewster, Memoirs of the Life, Writings, and Discoveries of Sir Isaac Newton (Edinburgh: Thomas Constable & Co., 1855), II, 263; N&Q, 5:12:316.

  773 Kinsey: Alfred C. Kinsey et al., Sexual Behavior in the Human Female, Institute for Sex Research (Philadelphia: W. B. Saunders Co., 1953), 59.

  774 Ovid: Artis Amatoriae iii.627-630; i.489 for HIM instead of her; i.137; also Amorum i.4.16ff., ii.5.18.

  774 Winthrop: Harry Andrew Wright, “Those Human Pu
ritans,” Proceedings of the American Antiquarian Society, L (new series) (April 17, 1940), 80-90.

  774 Wetherell: 40 Atlantic Reporter 728, also at 70 Vermont 274; Windham County Clerk, letter, July 25, 1963.

  774 Jonathan Swift: Journal to Stella, ed. Harold Williams (Oxford: Clarendon Press, 1948), I, lv-lix for discussion of the “little language”; 208 for null cipher (in letter of February 24, 1710/11).

  775 Marie Antoinette ciphers: Yves Gyldén, “Le chiffre particulier de Louis XVI et de Marie-Antoinette lors de la fuite de Varennes,” Revue Internationale de Criminalistique, III (1931), 248-256; this ciphered correspondence was reportedly published by Alma Soderhjelm in 1930 and 1934. Princess de Lamballe, Secret Memoirs of the Royal Family of France during the Revolution (Philadelphia: H. C. Carey & I. Lea, 1826), frontispiece, 366, 367.

  775 newspaper advertisements: “Secrets Exposed,” Littell’s Living Age, No. 494 (November 5, 1853), 342-344 (taken from Chambers’ Journal); Babbage Papers, Add. Ms. 37205, ff. 221 for Robert, 68-76 for Flo, 65, 79, 207, 222-224, 227 for others; Wemyss Reid, Memoirs and Correspondence of Lord Lyon Play fair (London: Cassell and Co., 1899), 154-155, for The Times’ editor; Richard J. Cyriax, “The Collinson Cryptograms in The Times’” N&Q, CXCII (July 20, 1947), 322-323.

  776 “I used to resent”: Thelma, Lady Furness, “The Prince and I,” The American Weekly (June 20, 1954), 9.

  777 Erik Brahe: Sven Tunberg, “Riksrådet Erik Brahes chifferkalendarium,” Personhistorisk tidskrift, XX (1918-1919), 37-65; XXIII (1922), 31-38.

  777 William Byrd: The Secret Diary of William Byrd of Westover, 1709-1712, eds. Louis B. Wright and Marion Tinling (Richmond, Va.: Dietz Press, 1941), vii; Another Secret Diary of William Byrd of Westover, 1739-1741, ed. Maude H. Woodfln, trans. Marion Tinling (Richmond, Va.: Dietz Press, 1942), iv.

  777 Bertrand: General H. G. Bertrand, Journal: Cahiers de Sainte-Hélène, ed. Paul Fleuriot de Langle (Paris: Éditions Gulliver, 1949), I, 16.

  777 Pepys: John Eglington Bailey, “On the Cipher of Pepys’s ‘Diary,’ ” Papers of the Manchester Literary Club, II (1876), 130-137; W. Matthews, “Samuel Pepys, Tachygraphist,” Modern Language Review, XXIX (October, 1934), 397-404; W. Matthews, “Pepys’s Transcribers,” Journal of English and Germanic Philology, XXXIV (April, 1935), 213-224; Encyclopaedia Britannica; DNB. Thomas Grenville as first solver of the diary is theory of D. Pepys Whitely, Pepys Librarian at Magdalene College, Cambridge (letter, November 19, 1965). His second choice is Grenville’s older brother, William Wyndham Grenville, Baron Grenville. Both were uncles to Richard Griffin Neville, 3rd Baron Braybrooke, first editor of the diary and, as holder of the barony of Braybrooke, hereditary visitor of Magdalene College. None of their DNB biographies shed any light on the question.

  779 Rabelais: Pantagruel (1532), ch. 24, trans. Sir Thomas Urquhart (1653; Philadelphia: J. B. Lippincott Co., 1912), I, 270-271. Rabelais’ Third Book, ch. 20, has an episode involving symbolism. Notes by ed. Jean Plattard in the edition Les Textes Français (Paris: Editions Fernand Roches, 1929) say that Zoroaster never wrote anything on letters hard to understand and that if Bossus, a grammarian contemporary with Domitian, ever wrote the book cited, it is unknown. Notes by eds. Jacques Boulenger and Lucien Scheler in the edition of the Bibliothèque de la Pléiade (Paris: Gallimard, 1955), say that Zoroaster is a grammarian of the time of Domitian and that Bossus and his alleged book are unknown. I can find nothing about ciphers of Zoroaster’s and nothing at all about Bossus. N&Q, 9:3:128, refers to a “Cyphral Dispatch” at end of one of translator Urquhart’s tracts.

  781 Shakespeare: The Life of King Henry the Fifth, II.ii.6-7, 71. William F. Friedman, “Shakespeare, Secret Intelligence, and Statecraft,” Proceedings of the American Philosophical Society, CVI (October, 1962), 401-411, demonstrates, in my opinion, that the proofs of guilt that Henry shows the traitors are indeed the intercepted letters. But I think Friedman has fallen victim to an unnecessary interpretation when he argues that the letters had been enciphered and that Henry’s agents had cryptanalyzed them. His arguments are purely hypothetical and without basis in the play. One might as well suppose that the letters were also in invisible ink. The original documents had no effect on Shakespeare’s play, of course, but M. McGuinness of the Public Record Office wrote on May 9, 1963: “A number of documents (other than those printed in Rymer’s Foedera IX, p. 300 et seq., and in Rotuli Parliamentorum IV, pp. 64-67) relating to the conspiracy against Henry V were found in Miscellanea of the Exchequer (E.163/7/7) and are printed in the Forty Third Report of the Deputy Keeper of the Public Records, 1882, p. 582 et seq. The original documents are written en clair and I have not found any evidence to suggest that they were originally enciphered. Documents produced for and at the trial of the conspirators at Southampton would normally have been filed among Ancient Indictments (KB 9) but I have been unable to trace any at the appropriate date; nor any entry on the Plea Roll for Michaelmas 1415 (KB 27/618).”

  781 Balzac: La Physiologie du Manage (Paris, 1829), Meditation XXV, §1, at II, 207-210 and 347 for erratum. In the Notes to the edition of the Bibliothèque de la Pléiade (Paris: Gallimard, 1960), which was based on the last text reviewed by Balzac himself, editor Marcel Bouteron says, at 895, that “One need not seek any sense in the text, designedly indecipherable, on page 835. Balzac wanted to hide his opinion on religion and confession from us; he managed it by a typographic jest in the fashion of his favorite author, the English humorist [Laurence] Sterne, in having printed letters assembled by chance.” Bazeries, 90-98; Lange & Soudart, Annexe VI, 305-307; N&Q, 10:3:368.

  783 Poe’s interest in cryptology: For other theories on its origin, see DAB and Clarence S. Brigham, Edgar Allan Poe’s Contributions to Alexander’s Weekly Messenger (Worcester, Mass.: American Antiquarian Society, 1943), 11, which reprints Poe’s articles in that paper. This brochure is reprinted from the Society’s Proceedings, LII (April, 1942), 45-125. Reference to Poe’s articles will be made by date to AWM (for Alexander’s Weekly Messenger). I am grateful to W. K. Wimsatt, Jr., for reading my section on Poe and commenting upon it.

  783 The Narrative of Arthur Gordon Pym: ch. 23.

  784 Schuyler Colfax: AWM (April 29, 1840), 2:4.

  784 “ugliest and drollest”: AWM (January 22, 1840), 2:5

  784 “Do people really”: AWM (February 12, 1840), 2:5.

  784 “Were we to engage”: AWM (March 25, 1840), 2:6.

  784 statistics on number of cryptograms solved: W. K. Wimsatt, Jr., “What Poe Knew About Cryptography,” PMLA, LVIII (September, 1943), 754-779 at 755.

  784 “Just let us into,” “Well, what will”: AWM (January 22, 1840), 2:5.

  784 seven alphabets, promise to reveal method: AWM (February 19, 1840), 2:2-3. William F. Friedman, “Edgar Allan Poe, Cryptographer,” American Literature, VIII (November, 1936), 266-280, at 270-272, raises the question whether this seven-alphabet message might have been a keyword polyalphabetic. Wimsatt, 762, dismisses this possibility, and I agree: Had Poe actually solved one such, he would have crowed about it.

  784 “Upon second thought”: AWM (February 26, 1840), 2:4; Wimsatt, 779.

  784 Kulp false cryptogram: AWM (February 26, 1840), 4:3-5.

  785 Cardano grille: AWM (February 26, 1840), 4:3-5.

  785 Lord’s Prayer, “Twelfth Night”: AWM (February 26, 1840), 2:4, (March 11, 1840), 2:3.

  785 “single glance”: AWM (February 26, 1840), 2:4.

  785 unkeyed symbol alphabets: My tabulation, based on cryptograms as printed in AWM. a = 1 alphabet, AWM (April 22, 1840), 2:3.

  785 “Our correspondent will”: AWM (January 15, 1840), 2:4.

  785 errors and omissions: AWM (April 22, 1840), 2:3, for example; Wimsatt, 765.

  785 Poe’s reputation: Brigham, 11; Wimsatt, 765 for “immediately,” 777 for “This cryptograph,” 765 for “much shorter time,” 760 for “in one-fifth,” 778 for “The most profound.”

  786 “far beyond the ordinary,” “native power”: Wimsatt, 778. Wims
att’s article is the definitive study of Poe’s cryptology and stands up extremely well under scrutiny; most of its judgments are valid, and I am indebted to it for the framework of my own study, but I think that here Wimsatt’s lack of familiarity with the practice of cryptanalysis has led him astray. This charge cannot be leveled against William F. Friedman, who concludes his “Poe” article with the judgment that “Had he [Poe] an opportunity to make cryptography a vocation, there is no doubt that he would have gone far in the profession.” But this is a pure guess, not even based on the evidence of the Alexander’s articles, which had not then been discovered, and its force seems to be negated by a nearby sentence (280): “Against his will he [the serious student of cryptography] is driven to the conclusion that Poe was only a dabbler in cryptography.” Even Poe never claimed to be anything else. Moreover, the article is unfair to Poe, evaluating him from a modern and not a contemporary point of view. It belittles what the achievement of a polyalphabetic solution would have been in those pre-Kasiski days and taxes Poe with not knowing some things (about Renaissance cryptology) that no one else in his time knew. The article was, nevertheless, the first study of Poe’s cryptography. In sum, I think that the Wimsatt and Friedman judgments of an exceptional cryptanalytic ability for Poe are unwarranted.