THE CODEBREAKERS

by DAVID KAHN

  242 second de Viaris book: L’art de chiffrer et déchiffrer les dépêches secrètes (Paris: Gauthier-Villars, 1893).

  242 Valério: “De la cryptographie,” Journal des Sciences militaires, 9th series. XLVIII (December, 1892), 385-402; IL (January, 1893), 37-49; (February, 1893), 244-260; L (April, 1893), 75-97; LI (July, 1893), 102-116; LII (November, 1893), 248-276; LIII (March, 1894),443-468; LVII (January, 1895), 124-152; LVIII (April, 1895), 127-142; (May, 1895), 285-300. Valério also served as a handwriting “expert”—who made a false identification—in the Dreyfus affair; for an analysis of this work, see Edmond Locard, Les Faux en Ecriture et leur Expertise (Paris: Payot, 1959), 131-134.

  242 Delastelle biography: William Maxwell Bowers, “F. Delastelle—Cryptologist,” The Cryptogram, XXX (March-April, 1963), 79-82, 85; (May-June, 1963), 101, 106-109. This is based on documents in the Mairie of Saint-Malo and on recollections of Delastelle’s niece. His first book was Cryptographie nouvelle … (Paris: P. Dubreuil, 1893).

  242 “only catalogues”: Delastelle, 2.

  243 Playfair invention: Delastelle, 72-82.

  243 bifid: Delastelle, 86-93. For methods of solution, see Friedman, IV, ch. x; Sacco, §103A; William Maxwell Bowers, The Bifid Cipher, Practical Cryptanalysis, II (American Cryptogram Association, 1960).

  243 trifid: Delastelle, 101-106. For methods of solution, see Sacco, §103B; William Maxwell Bowers, The Trifid Cipher, Practical Cryptanalysis, III (American Cryptogram Association, 1961).

  243 slide dispositions: Delastelle, 52-63; Sacco, §31; Friedman, II, ch. ii and appendix 1.

  244 Bazeries biography: Dictionnaire de Biographie Française; Musée National de la Légion d’Honneur, dossier of Bazeries; Pierre Sourbès, “Le Commandant Bazeries: l’Homme Qui ‘Cassait’ les Codes,” Le Miroir de l’Histoire, No. 153 (September, 1962) 282-289. The Sourbès article is based on an interview with Mme. Jean Yon, Bazeries’ daughter, but is dangerously unreliable in relating Bazeries’ cryptologic career. As just one instance of many, Sourbès gives 1876 as the date of an incident that Bazeries himself says occurred in 1890. In addition, the conclusion—that Bazeries solved cryptograms enabling the French to place their forces to halt the Germans at the Battle of the Marne—is false. These cryptograms were not solved until after the battle, according to the chief of the cryptologic bureau at G.H.Q., Givierge, vi. Some additional personal and professional details come from my interviews with Mme. Yon, then 98, at Perpignan, July 15 and 19, 1966.

  244 Nantes solution: Bazeries, 34-35.

  245 Bord: Bazeries, 121-127.

  245 ciphers solved: Bazeries, 200, 128-139, 151-184. étienne Bazeries, Les ‘Chiffres’ de Napoleon Ier Pendant la Campagne de 1813 (Fontainebleau. Maurice Bourges, 1896); étienne Bazeries et Emile Burgaud, Le Masque de Fer (Paris: Firmin-Didot, 1893), 257-272.

  245 anarchists: Bazeries, 111-114.

  245 Gronsfeld: Gaspar Schott, Magia universalis … (Nuremberg, 1659), IV, 33.

  246 Orléans: Bazeries, 114-119. False repetitions, 243-247. “Merde,” 248-249. Joseph Reinach, Histoire de l’Affaire Dreyfus (Paris: Charpentier et Fasquelle, 1905), V, 6, translated in Rosario Candela, The Military Cipher of Commandant Bazeries (New York: Cardanus Press, 1938), 3.

  247 train-ride solution: Bazeries, 201.

  247 two proposed systems: Bazeries, 203-207.

  247 “knocking his brains out”: Bazeries, 207.

  247 cylindrical cryptograph: Bazeries,207-212,250-261; M.le Capitaine Bazeries, “Cryptograph à 20 rondelles-alphabets (25 lettres par alphabet),” Compte rendu de la 20e session de l’Association Française pour l’Avancement des Sciences (Paris: Au secrétariat de l’Association, 1892), 160-165.

  247 de Viaris solution: L’art de chiffrer …, 100-109. For an exposition in English, see [William F. Friedman], Several Machine Ciphers and Methods for Their Solution, Riverbank Publication No. 20 (Geneva, Illinois: Riverbank Laboratories, 1918), 37-58.

  249 U.S. Army adopts: United States [War Department], Chief Signal Officer, Instructions for Using the Cipher Device Type M-94, February, 1922 (Washington, D.C.: U.S. Government Printing Office, 1922); Harris, 335.

  249 pencil and paper cipher: Bazeries, 262-274. See Candela for solution.

  250 comments in Bazeries’ book: “revelation,” 2-3; “willful blindness,” 3; “retreated,” 34; “routine,” 214; “public danger,” 33; survey of current literature, 56-80; “to abandon,” 34.

  250 Spanish Army cipher: Carmona (Madrid: Est. Tip. Sucesores de Rivadeneyra), 99-117.

  251 Martí: Las Claves de Martí y el Plan de Alzamiento para Cuba, deciphered by Dr. Rebeca Rosell Planas, Publicaciones del Archivo Nacional de Cuba, XVI (La Habana, 1948), frontispiece, 3, 65-71.

  251 Ethiopia: L. Zehnder, “Geheimaltung drahtloser Telegramme,” Prometheus: Illustrierte Wochenschrift über die Fortschritte in Gewerbe, Industrie und Wissenschaft, XXIII (May 18, 1912), 524. I am indebted to Maurits de Vries for this reference.

  251 codes for diplomats: for example, Mexico, Diccionario Telegrafia (México: Imprenta Imperial, 1866); Portugal, Ministerio da Marinha e ultramar, Diccionario cryptographico (Lisbon: Typ. do Instituto Geográphico Portuguez, 1890).

  251 superencipherment of codenumbers into letters: [W. Clausen-Thue], The ABC Fifth Edition Universal Commercial Electric Telegraphic Code [London, 1901], vii-ix.

  251 Sittler transposition: F.-J. Sittler, Dictionnaire Abréviatif Chiffré, 4th ed. (Paris: Imprimerie Lefebre, 1879), section “Correspondance Secrète” at back of volume.

  252 additive superencipherment: one of nine methods proposed in [United States, War Department] Telegraphic Code to Insure Secrecy in the Transmission of Telegrams, J. F. Gregory, compiler (Washington, D.C.: U.S. Government Printing Office, 1886).

  252 Navy cryptographic responsibility: Captain L. S. Howeth, U.S.N. (Ret.), History of Communications-Electronics in the United States Navy (Washington, D.C: U.S. Government Printing Office, 1963), 7, 8, 9, 233.

  252 David Porter message: Letters to Officers, Ships of War, VIII, 486, 510, NA, RG 45.

  252 1877 Vigenère: message of March 12, 1877, from Commander T.V. McNair, commanding Kearsarge at Nagasaki, to Rear Admiral William Reynold, commanding U.S. Naval Force on Asiatic Station, NA, RG 45.

  254 Dewey message: “Reception of the Report,” New York Tribune (May 8, 1898), 1, 2; Dewey to Long, May 1, 1898, Naval Records Collection of the Office of Naval Records and Library, Area 10 File, 1798-1910, January 1898 to May 10, 1898, NA, RG 45, for codetext. Conclusion about apparent superencipherment stems from the lack of a fixed relationship between the codewords in this and other code messages in that file and their plaintexts as given in NA, RG 45, Ciphers Sent, October 27, 1888, to May 31, 1898.

  254 Dreyfus: Background information on the case comes largely from Guy Chapman, The Dreyfus Case: A Reassessment (New York: Reynal & Co., 1955), which has a useful bibliography. Material on the Panizzardi telegram comes from Reinach, I, 244-251, and Henri Guillemin, “L’Affaire Dreyfus: Le Télégramme du 2 Novembre,” Mercure de France, CCCXXXIX (August, 1960), 596-616. Because the primary sources in the Dreyfus affair are a chaos of transcripts, depositions, First Revisions, and Second Revisions that terrify all but the professional Dreyfus-case experts, and because the Reinach and Guillemin analyses are keyed almost sentence by sentence back to these original materials, thus permitting any hardy—or foolhardy—soul who wants to spelunk among them to do so, I have thought it better not to burden these notes with point-by-point references to the primary sources. I have checked the primary sources to make certain of the accuracy of all statements in the two articles bearing on the solution of the Panizzardi telegram. Only items not included in either of them are given notes here. To begin with, there is the telegram codetext itself. It is in France, Archives Nationales, BB19 75, dossier 1.

  255 cipher bureau members: France, Ministère des Affaires Étrangères, Annuaire Diplomatique et Consulaire de la République Française pour 1894, new series, XVI (Paris:
Berger-Levrault, 1894), 1, 2, 170, 283, 214, 184, 147, 151, 216. Bazeries: Ministère des Affaires Étrangères, Direction des Archives Diplomatiques, letter, December 15, 1962, says that the archives of the cryptanalytic service were destroyed in 1940.

  256 Baravelli: Published by Ermanno Loescher; reprinted 1896.

  257 Duchess Grazioli: Maurice Paléologue, An Intimate Journal of the Dreyfus Case, trans, by Eric Mosbacher (New York: Criterion Books, 1957), entry for November 10, 1894, 29-32.

  258 Baravelli elements: The fragmentation of Dreyfus and the subsequent one of Schlissenfurt, the reconstruction of the cryptanalysts’ worksteps, and the determination of the superencipherment result from my comparison of the telegram’s encicode with its plaintext and with the codebook.

  260 Matton incident and quotations: La révision du Procès de Rennes, Enquête de la Chambre Criminelle de la Cour de Cassation (Paris: Ligue Française pour la Défence des Droits de l’Homme et du Citoyen, 1908), 240-241. Called the Second Revision in the Dreyfus literature. Page numbers used in my citations are those of the Imprimerie Nationale edition, which are given in the margin of the Ligue republication.

  261 date of Schlissenfurt telegram: Second Revision, 249.

  261 Foreign Office does not know that Army had plaintext: Only Delaroche-Vernet says that the Army had given the Schlissenfurt plaintext to the Foreign Office to help it achieve an accurate solution of the November 2 message, which he says was not fully solved. Paléologue, Matton, and the unnamed chief of the cryptanalytic service all agree that the phony message served only to check the completed solution of the November 2 message. Reinach, I, 249, with all citations. Indices to Reinach indicate that none of the cryptanalysts ever testified.

  261 Munier: Conseil de Guerre de Rennes, Le procès Dreyfus, Compte rendu sténographique in extenso (Paris: P. V. Stock, 1900), II, 228. Munier said that the ciphertext consisted of 20 groups of four-digit numbers, which is not true in the first place. He then said either that the 10th and 17th groups are identical, which is false, or that two 10’s and two 17’s appear in the message, which is also false. On the basis of whatever he meant he alleged that “All the groups No. 10 and No. 17 correspond to interchangeable expressions; now, this condition is realized in Version No. 2. Thus Version No. 1 can apply to the authentic ciphered text.” The episode shows to what lengths the enemies of Dreyfus would go to shore up their false case against him. Munier died before he could be questioned about his conclusions.

  262 French solution of Italian code: Paléologue, entries of November 29, 1897, 93, and August 26, 1899, 267-268.

  262 French solution of German code: Raymond Poincaré, Memoirs, trans, by Sir George Arthur (New York: Doubleday & Co., 1928), III, 251; Gyldén, 17-18.

  262 Commission on Military Cryptography: Gyldén, 10-11; [François Cartier], “Souvenirs du Général Cartier,” Revue des Transmissions, No. 85 (July-August, 1959), 23-39, for part I, No. 87 (November-December, 1959), 13-51, for part II, at I, 23-24, 34, 35.

  263 German cryptology: Gyldén, 14-19.

  263 prewar England and Italy: Gyldén, 19-20, 23; Sacco, §§157-158.

  263 Austria: Gyldén, 21-22; Maximilian Ronge, Kriegs- und Industri-espionage (Zurich: Amalthea-Verlag, 1930), trans, by Adrien F. Vochelle and published in a slightly abbreviated edition as Espionage (Paris: Payot, 1932), 29-30. All my citations refer to French edition unless otherwise specified.

  263 blank code, Serbian code, Italian code: A former Austrian code officer, “Ciphers and Cipher Keys,” The Living Age, CCCXXXIII (September 15, 1927), 491-495.

  264 English-French codebook: Cartier, “Souvenirs,” II, 30. See also Barbara Tuchman, The Guns of August (New York: The Macmillan Company, 1963), 55.

  Chapter 9 ROOM 40

  Because of the importance of the Zimmermann telegram, a note on major printed sources might not be inappropriate. Admiral Sir William James, The Eyes of the Navy: A Biographical Study of Admiral Sir William Hall (London: Methuen, 1956), also published in New York the same year by St. Martin’s Press as The Code Breakers of Room 40, based his Zimmermann telegram chapter upon Hall’s own story of it in his uncompleted autobiography and Hall’s papers; James was himself administrative head of Room 40 later in the war. Burton W. Hendrick, The Life and Letters of Walter H. Page (Garden City: Doubleday, Page & Co., 1925), III, ch. 12, “The Zimmermann Telegram,” 331-364, gives much material unobtainable elsewhere, but does not cite any sources and seems in error in a few places. Barbara W. Tuchman, The Zimmermann Telegram (New York: Viking Press, 1958), is a masterly study of the political circumstances surrounding the telegram and its publication. Unfortunately, it was written before the declassification on January 20, 1965, of William F. Friedman and Dr. Charles J. Mendelsohn, The Zimmermann Telegram of January 16, 1917 and its Cryptographie Background, War Department, Office of the Chief Signal Officer (Washington, D.C.: U.S. Government Printing Office, 1938), a very thorough study, detailing the role of Code 0075, based upon examination of the then extant messages in the State Department archives and upon Mendelsohn’s studies of German diplomatic codes; Mendelsohn, a professor at the College of the City of New York, served as a cryptanalyst in Yardley’s M.I. 8 and his American Black Chamber. All citations to “James” refer to the American edition of his work, to “Tuchman” to the above cited book; their other works are cited in full or short title. Among the major nonprinted sources are DSDF; a letter of the German Foreign Office, March 30, 1965, translated by Hardie, in response to my query of November 20, 1964, cited simply as “GFO”; and James’s letters of September 12, October 8, and November 7, 1962, and—after reading a draft of the chapter—of February 1, 4, and 17, 1964, for which I am deeply grateful.

  PAGE

  266 Telconia: Tuchman, 10-11; Hugh Cleland Hoy, 40 O.B., or How the War Was Won (London: Hutchinson & Co., 1932, republished 1935, to which references are cited), 21-22.

  266 Oliver mentions intercepts: “War Work at the Admiralty,” The [London] Times (December 14, 1927), 16, a report of Ewing’s disclosures of his work in Room 40 at a lecture in Edinburgh; Admiral Sir William James, “Room 40,” Edinburgh University Journal, XXII (Spring, 1965), 50-54, at 50.

  266 “futile” mechanism: A. W. Ewing, The Man of Room 40: The Life of Sir Alfred Ewing ([London:] Hutchinson & Co., [1939]), 174.

  266 Ewing’s appearance: Ewing, 178.

  267 life and work: Ewing; Who Was Who, 1929-1940; Columbia Encyclopedia.

  267 Lloyd’s, four friends: Ewing, 174-175; James, letter, February 1, 1964.

  267 Göben message and incident: James, 60; Tuchman, The Guns of August, ch. 10.

  267 “thick of office work”: Ewing, 176.

  268 German codebooks, intercept stations: James, 28-29.

  268 “Ewing Admiralty”: Ewing, 174.

  268 no previous knowledge, exhilarated, October 25: Ewing, 175-177.

  268 Magdeburg: Winston S. Churchill, The World Crisis (New York: Charles Scribner’s Sons, 1923), at I, 503. All Churchill references in this chapter are to this work.

  268 October 13, copied, Rotter: James, 29, 56.

  268 German naval code: James, letter of September 12, 1962; French Strother, “German Codes and Ciphers,” The World’s Work (June, 1918), 143-153 at 152-153. Sacco, §110, gives methods of solving superencipherments with known codes. The “German High Fleet code” depicted in Yardley, opposite p. 218, was probably used in the latter part of the war.

  269 three weeks: Churchill, I, 503-504.

  269 crowding, Room 40, I.D. 25: James, 56; James, letters, September 12, 1962, and February 1, 1964; Ewing, 178-179; Francis Toye, letter, March 9, 1963.

  269 trawler: James, 56-57. This code may have been VB 718 (Tuchman, 79).

  269 staff expansion: James, 57,70, 90; Francis Toye, For What We Have Received: An Autobiography (New York: A. A. Knopf, 1948), at 166, 186-187.

  270 direction-finding: “War Work at the Admiralty”; Churchill, I, 504; James, 68.

  270 Hartlepool raid: Churchill, I,
505-513; James, 58; Filson Young, With the Battle Cruisers (London: Cassell & Co., 1921), 97-99.

  271 Dogger Bank: Churchill, II, 124-137 for intercepts and action; James, 128, Ewing, 187, for aftermath.

  271 50 stations: James, letter, September 12, 1962.

  271 new superencipherment: James, 67-68; alphabet reconstructed from Odenwald messages in Strother, 152-153.

  272 key changes and solutions: James, 25, 29, 67-68; “War Work at the Admiralty”; Admiral Sir William James, The Sky Was Always Blue (London: Methuen, 1951), 104.

  272 Jutland: Churchill, III, 114 for preliminary intelligence; Churchill, III, 116, James, 117, and Sir Julian S. Corbett, History of the Great War based on Official Documents: Naval Operations (London: Longmans, Green & Co., 1920-1931), III, 326, for call-sign transfer; Churchill, III, 157, and James, 119, for Regensburg error; Churchill, III, 156, and Corbett, III, 402, for Scheer’s messages and their solution; James, 119, for omission of 9:06 message; Churchill, III, 157, for Jellicoe rejection. Jellicoe does not mention the Admiralty message to him in his The Grand Fleet 1914-1916 (London: Cassell, 1919).

  273 gamma epsilon and gamma u: James, letters, October 8, 1962, February 4, 1964.

  273 German suspicion: Churchill, III, 113.

  273 change of code: James, 115.

  273 L-32: James, 116.

  273 Miller: “A War Secret,” The Saturday Evening Post, CCII (October 23, 1926), 44, 46, 74.

  274 pneumatic tube: James, 129; Ewing, 182; James, letter, September 12, 1962.

  274 15,000: W. R. Hall, affidavit of March 28, 1932, Mixed Claims Commission, U.S.A. on behalf of Lehigh Valley Rr. et al. against Germany, Docket 8103, Exhibit 920. Reprinted in Friedman and Mendelsohn, 30-32. Cited henceforth as “Hall affidavit.” Ewing, in “War Work at the Admiralty,” stated that sometimes 2,000 intercepted messages were dealt with in 24 hours. He does not specify that they were solved, however, as Hall does. Many of the 2,000 might have simply served for direction-finding fixes.

  274 round the clock: James, letter, September 12, 1962.

  274 Zeppelin raids: Hoy, 190; James, letter, February 1, 1964. By reading German messages, Room 40 could warn British defenses of imminent Zeppelin raids—a warning of particular importance because the Zeppelins flew so high that without it the slow-climbing pursuit planes of the day could not attack the airships. Early in the war, Room 40 learned of the raids through Zeppelin messages reporting “Only HVB on board.” HVB, or Handelschiffsverkehrsbuch, was the German mercantile code, a nonconfidential one which was the only one carried over enemy territory. Beginning March 31, 1916, Room 40 was able to solve messages (probably meteorological) indicating raids and to alert defense commands. On August 24, this procedure resulted in the downing of the L-13. See Hoy, chs. 14 and 15; Kenneth Poolman, Zeppelins Against London (New York: John Day, 1961), at 51; Francois Cartier, “Le service d’écoute pendant la guerre,” Radio-Electricité, IV (November 1, 1923), 453-460, (November 15, 1923), 491-498, at 460.