by Alastair Denniston- Code-breaking From Room 40 to Berkeley Street
5. DENN 1/2, Churchill College Archive, Cambridge.
6. This view was taken by historians because MI1(b)’s contribution to diplomatic Sigint had been forgotten.
7. This is probably Admiral Sir Sydney Fremantle, who was assigned to the Signal Division in the Admiralty and responsible for compiling secure code and cipher books. See Fremantle, p. 172.
8. See Ewing, The Man of Room 40.
9. Oliver had been Director of the Intelligence Division (DID), the senior Division of the Naval Staff. His taciturn manner resulted in the nickname ‘Dummy’ and he was apparently very poorly dressed. He was a hard worker and former Admiral of the Fleet who lived to the age of 100. His principal deputy was Captain Thomas Jackson, Director of the Operations Division (DOD).
10. Admiral Sir Arthur Wilson, a former First Sea Lord and before that, Commander-in-Chief of the Channel Fleet. Retired in 1912 at 70 but returned in an unofficial and unpaid capacity in November 1914 to help his friend, Oliver. Dedicated and tough, he was known on the Lower Deck as ‘Old ‘Ard ‘Art’.
11. See Churchill, Vol. 1, p. 361.
12. Hall was born at Britford, Wiltshire, on 28 June 1870. He was the elder son of Captain William Henry Hall, RN, the first Director of the Intelligence Division at the Admiralty. The Admiralty Board decided to set up a section under Hall Sr. in 1882 to collect and sift reports from abroad and to provide them with up-to-date information about foreign fleets in the event of potential conflict. Hall Jr signed on in 1884 and after rising through the ranks, was promoted to Commander in 1901 and appointed to HMS Magnificent, flagship of the Second-in-Command, Channel Fleet. He was promoted to captain in 1905 and installed by the First Sea Lord, Sir John Fisher, as Inspecting Captain of the new Mechanical Training Establishment. In 1913 he was appointed captain of the new battlecruiser Queen Mary. The British naval historian Patrick Beesly met Hall once and provides a colourful description of him. See Beesly, p. 36.
13. Ewing had been made a Commander of the Bath in 1906 and Knight Commander of the Order in 1911.
14. GCHQ is in possession of two 1907 reports from Simla dealing with Russian ciphers which provide a useful insight into modern Sigint. They were distributed to headquarters in India and via the India Office to the War Office and possibly the Admiralty.
15. This system was probably the German training cipher known from the discriminant used in exercise traffic as ÜBCHI (Übungschiffrierung). MO5 had obtained a German army staff manual before the war and it was studied for ‘many weary hours’ by Major Walter Kirke, Macdonagh’s assistant in the ‘Special Section’.
16. John Ferris, ‘The British Army and Signals Intelligence in the Field during the First World War,’ Intelligence and National Security, 3/4 (1988).
17. Intelligence E (c) to G.S.I.e, 10.10.18, TNA ADM 137/4701.
18. The only primary sources available are: three folders of MI1(b)’s own papers, a few MI1(b) diplomatic decrypts and information on its 1919 work and staffing recorded in the course of the discussions about establishing GC&CS; History of MI1(b), HW7/35 written between November 1919 and January 1923 (it is quoted in a War Office paper of that date ‘History of the Cryptographic and Wireless Intelligence Organisation’, HW 3/39) possibly by Maj. G.L. Brooke-Hunt); Hay’s notes and his wife’s book.
19. The War Office destroyed the relevant records after the end of the war. It is likely that as in Aden, cryptanalysis was used to detect attempts to evade censorship.
20. This was the so-called ADFGVX cipher, a complicated combination of substitution and transposition with frequently-changing keys.
21. Macdonagh would later serve as DMI from January 1916 until September 1918.
22. The MI1(b) files were destroyed during the Blitz, along with the rest of the War Office’s intelligence archives, and those of other arms of service as well.
23. Hay would later contest with AGD for a new post which would head a combined army/navy intelligence section. He would also go on to be an eminent historian, publishing a number of major historical works. Hay was admired by his staff and considered to be a good chief and was given charge of constructing codes and ciphers for British forces early in 1917.
24. A ‘crib’ is a guessed word or phrase contained within an encrypted message.
25. See Alice Hay.
26. History of MI1(b), TNA HW 7/35, written between November 1919 and January 1923 (it is quoted in a War Office paper of that date (History of the Cryptographic and Wireless Intelligence Organisation, TNA HW 3/39)).
27. Hay made notes for much of the war which were published in a biography by his widow in 1971, together with some of her own views. See Alice Hay. The earliest surviving American telegram is from their Ambassador in London to Washington, sent on 23 January 1916. (TNA HW 7/17)
28. History of MI1(b).
29. Ibid.
30. Regrettably, while the Admiralty preserved the archives of Room 40, the Army chose to destroy most of the original intelligence files of MI1(b). Only twenty-five of the 3,330 (at least) files of codebreakers at GHQ France survive. Only four of the 400 weekly reports of the codebreakers in the Middle East on traffic analysis and cryptanalysis survive.
31. DENN 1/3, also DENN 1/2, Churchill College Cambridge Archive Centre (CCAC). Some messages dated between August 1915 and January 1916 are available at TNA (HW 7/5, HW 7/6). Around five of the thirty staff were working on German diplomatic messages (similar to MO6(b)’s effort against American telegrams).
32. TNA HW 3/184.
33. Ibid.
34. See Hay.
35. One of the most skilled MI1(b) cryptanalysts was Captain G.L. Brooke-Hunt. He solved the so-called FürGOD cipher in early 1917 which was used for messages sent around three times a week from a powerful German station at Nauen outside Berlin to German wireless stations with the callsign GOD. A message arranging for a submarine to bring rifles and ammunition to a Moroccan nationalist was passed to Hall at the Admiralty and the submarine was duly intercepted and sunk.
36. A hatted code is arranged in other than numerical (or alphabetical) order. A ‘hat code-book’ is characterized by the fact that when the plain-language terms are arranged in alphabetical order the code groups are not in numerical (or alphabetical) order; i.e. a two-part code. A recipherment conceals the true character and figures or letters of an encoded message by applying a key or subtractor (usually by non-carrying addition or subtraction) or by any system of transposition or substitution.
37. These were standard in format throughout the armies and similar to those still used in WW2, but easily modified locally by changing meanings or by encryptment. They consisted of trinomes or trigraphs together with encipherment tables. Similar codes were used for air-to-ground communications.
38. DENN 1/2, Churchill College Archive, Cambridge.
39. See Appendix 1. GCHQ regard this as their own charter document.
40. Admiral of the Fleet Sir Henry Oliver, ‘Recollections’, National Maritime Museum, OLV 12.
41. The Heligoland Bight is a stretch of water off Germany’s main North Sea naval base of Wilhelmshaven, a coastal town in Lower Saxony.
42. DENN 1/2, Churchill College Archive, Cambridge.
43. Ibid.
44. Churchill, Vol. I, p. 115
45. DENN 1/2, Churchill College Archive, Cambridge.
46. Ibid.
47. ‘Dilly’ Knox was one of four sons of the Bishop of Manchester. Of his brothers, Edmund became editor of Punch magazine, Wilfred was a Church of England theologian and Ronald was a Catholic clergyman who translated the New Testament for Catholics.
48. See Fitzgerald on the Knox brothers.
49. Frank Birch had enlisted in the RNVR at the outbreak of war and served at sea in the Atlantic and at the Dardanelles. He was a Fellow at King’s from 1915 to 1934 and a lecturer in History from 1915 to 1928. In the 1930s he left Cambridge to work in the theatre. He worked in Room 40 from 1916 to 1919 and was Head of the Naval Section at BP during WW2 from 1942 to 1945.
50. Frank Adcock had become a King’s Fellow in 1911 and held the Chair of Ancient History from 1925 to 1951. He came from a Methodist background and excelled academically.
51. Montgomery was 44 and the author of studies of St Augustine and a translator of Albert Schweitzer.
52. De Grey was a favourite of Hall, and went on to take charge of the Mediterranean Section. He had studied languages and was fluent in French and German.
53. Clarke is the source of much of the information about recruits to Room 40. See CLKE 1 ,2 and 3, Churchill College Archive, Cambridge.
54. DENN 1/2, Churchill College Archive, Cambridge.
55. See Churchill, Vol. 1, p. 414.
56. Ibid., p. 500.
57. The Dogger Bank is a large sandbank in a shallow area of the North Sea about 100km off the east coast of England.
58. See Churchill, Vol. 1, p. 559.
59. Ibid., p. 560.
60. Ibid., p. 264.
61. Ibid., p. 175.
62. See Ewing, The Man of Room 40.
63. AGD and Clarke, the two chroniclers of Room 40, never offered a complimentary assessment of Ewing’s contribution and it appears he was not held in the highest regard. He presided over Edinburgh University until his retirement in 1929 and died in 1935.
64. Clarke papers, Churchill College, Cambridge.
65. The Zimmermann Telegram has been written about by numerous authors. James claimed his account was the first, probably followed by Hendrick. Friedman and Mendelsohn’s account is considered by most historians to be close to definitive. Barbara Tuchman wrote an entire book on the subject and more accounts have been provided by Beesly, Gannon and Ramsay. Friedman was still querying aspects of it with AGD in the 1960s.
66. Code 7500 was one of a series of two-part codes used by the Germans. The two parts consist of 1) a set of 10,000 phrases in alphabetic order and numbered from 0000 to 9999, the numbers being entirely mixed up so that they have no numerical sequence; and 2), the same phrases fitted with the same numbers as before, but this time with the numbers in sequence and the phrases mixed up. The first part is used for encrypting, i.e. sending a message, the second for decrypting, i.e. reading a message.
67. This route was revealed during an elaborate investigation by the German Constituent Assembly in 1919-20 into the responsibility for the war and the part played by the Zimmermann Telegram. Among its published documents is a note which reveals the other route used by the German Foreign Office (Vol. II, p. 1337, ‘Official German Documents Relating to the World War’. Translated under the supervision of the Carnegie Endowment for International Peace, Division of International Law). ‘Instructions to Minister v. Eckardt were to be taken by letter by way of Washington by U-Boat on the 15th January; Since the U-Boat Deutschland did not start on her outward trip, these instructions were attached on January 16 to telegram No. 157, and through the offices of the American Embassy in Berlin telegraphed to Count Bernstorff by way of the State Department in Washington’. The German Government frequently used the American State Department to send messages to its Ambassador Bernstorff as it had no cable links to the US. They went through British cable facilities and the British didn’t mind as they could read all of the traffic. The messages should have been handed to the American Embassy in plain text to be enciphered with an American code before transmission. However, President Wilson, who was in effect, his own Secretary of State, allowed messages to be handed in already in code on the basis that they were in aid of peace initiatives. The Zimmermann telegram was attached to a message which did pertain indirectly to peace efforts in a very indirect fashion. The German Government was informing its Ambassador that it ‘promises the early termination of the war and the restoration of that peace which the President has so much at heart’. It proposed to bring about this peace by the institution of unrestricted submarine warfare!! Bernstorff was sent the message on 16 January and told not to inform Washington until 1 February. The Zimmermann telegram was attached to this message (No. 157), put into German cipher and handed to US Ambassador Gerard in Berlin. He telegraphed it to Copenhagen, it then went on to London and finally to the State Department in Washington from where it was delivered to Bernstorff.
68. Code 13040 was an old German diplomatic code which was partially mixed up. The alphabetic vocabulary was broken up into fractions and these in turn into smaller fractions before numerical code groups were attached. However, the original alphabetic sequence of the words and phrases was only partially destroyed. This provided helpful clues when decrypting messages encrypted with the code.
69. A third possible route for the Zimmermann Telegram was by radio between the German station at Nauen and one of two American radio stations at Sayville, Long Island and Tuckerton, New Jersey. See Hendrick, Ewing, James and Tuchman who supported this. While now generally dismissed by historians today, it would be the subject of lengthy discussions between AGD and William Friedman fifty years later.
70. TNA HW 3/177 p. 2
71. HALL 3/6, p. 10, Churchill College Archive, Cambridge.
72. Walter Page was born on 15 August 1855. He worked hard to maintain close relations between the US and Great Britain while the US remained neutral. From an early stage of the war, he urged US intervention on an unwilling President Woodrow Wilson. When the British liner Lusitania was sunk by a German submarine (7 May 1915), with the loss of more than 100 American lives, Page called for a US declaration of war. He insisted then and later that US intervention at that time would have resulted in a swift victory for the Allies. In April 1917, when Wilson did ask Congress to declare war on Germany, he used the arguments that Page had been using for two-and-a-half years. Page became ill in August 1918 and retired. He died shortly after returning to the US on 21 December 1918.
73. Wiseman was a member of the Purchasing Commission of the British Ministry of Munitions. His duties also included intelligence and counter-espionage. (See Hyde, p. 63.)
74. The decoded message actually changes handwriting after a few groups, presumably from Bell’s to that of de Grey’s.
75. Edward Mandell House was an American diplomat and confidential advisor to President Woodrow Wilson. He played a key role in formulating the conditions of peace to end WWI.
76. According to a biological sketch, published by the NSA: ‘William Frederick Friedman (1891–1969), the dean of modern American cryptologists, was the most eminent pioneer in the application of scientific principles to cryptology and laid the foundations for presentday cryptologic concepts. He retired from the National Security Agency in 1955 after thirtyfive years of service with US cryptologic activities.
77. See Kahn’s paper in INS which provides the only source of biographical information on Bell.
78. Harrison was partly educated at Eton and considered to be an anglophile. He was Secretary in the Diplomatic Service and assigned to the State Department. He became Assistant Secretary of State in 1922.
79. Correspondence of Leland Harrison with Edward Bell, 12/154/1916 – 7/8/1918, Department of State, Office of the Counsellor 1909-1919, Record Group 59: General Records of the Department of State, 1763 – 2002.
80. See Yardley.
81. See for example Beesly, Room 40.
82. NID Vol. 26. The memorandum is reproduced in facing p. 18.
83. ‘A Contribution to the History of German Naval Warfare, 1914-1918’, Vol. I, pp. 39, TNA H/W 7/1.
84. ‘Thus was built up in the end a system by which momentary or ‘action’ Intelligence could go hand in hand with cumulative or ‘deferred’ Intelligence, and both be singly or together at the immediate disposal of the authorities.’ (A Contribution to the History of German Naval Warfare, 1914-1918, Vol. I, pp. 48, TNA H/W 7/1).
85. Roskill, pp. 145–6; Churchill, Vol. I, p. 466.
86. CLKE 3, Churchill College Archive, Cambridge.
87. See Fraser.
88. See Clarke.
89. James’ biography of Sir Reginald Hall in 1955 is one of the earliest published
accounts of the activities of Room 40.
90. Navy List for October 1917. Even though the seniority of the rank is listed as temporary, AGD is still listed as a Commander in the RNVR in January 1920.
91. Hall was elected to Parliament in 1919 and took his seat on 13 March as the MP for the West Derby Division of Liverpool. In October 1919 the University of Cambridge conferred on him the degree of LLD and later Oxford University conferred on him the Honorary Degree of DCL. He died on 22 October 1943.
92. See Churchill, Vol. I.
93. See Scheer.
94. Lecture by Major General D.E. Nolan, 20.3.33, Curriculum Archives, 392-A-19, United States Army Military History Institute, Carlisle Barracks.
95. Barker, p. 180.
96. Memorandum by I9e, ‘Enemy Codes and their Solutions’, I.18, TNA ADM 137/4660.
97. Memorandum by Second Army Intelligence, 1.12.18, AIR 1/2268/209/70/200; memorandum for Chief, Intelligence Section, AEF, 1.1.18. A.L. Conger papers.
98. WO 157/164. 3rd Army Intelligence summary, 14 Aug. 1918 (including order by Ludendorff, 19 Dec. 1917).
99. Notes and Records of the Royal Society of London, Vol. 34, No. 1 (July 1979).
Chapter 3: Between the Wars
1. DENN 4/2, Churchill College Archive, Cambridge.
2. An Anglo-French author, journalist and diplomat who wrote a number of books involving international intrigue.
3. Thwaites to Hall, 14 November 1918; Hall to Thwaites, 26 November 1918; Thwaites to Hall, 28 November 1918, WO32/21380.
4. Field Marshal Sir Henry Wilson was Chief of the Imperial General Staff in 1918, the professional head of the British Army.
5. See Alice Hay.
6. Ibid.
7. TNA HW 3/1.
8. Sinclair was a career sailor who had been educated at the Britannia Naval College at Dartmouth, entering the navy as a midshipman in 1888. His service record indicated that his ability and professional knowledge was ‘very good’ and other positive comments included: ‘steady and trustworthy’; ‘zealous & capable’; ‘Excellent tact & temper’; ‘Very discrete & loyal; exceptional powers of administration’ (Sinclair service record, TNA ADM 196/43). He served in the Mobilisation Division of the Admiralty during WW1 and by its conclusion, was chief of staff of the Battlecruiser Fleet. In January 1919, he succeeded Hall as DNI. Hall was pleased and told Sinclair that it was ‘not often given to men that they see their job filled by the only man who can do it’. (Hall to Sinclair, 18 December 1918 and 14 January 1919, Sinclair papers MS 81/091, scrapbook vol. 1). His stay in Naval Intelligence ended after eighteen months when in August 1921 he was appointed to a three-year post as Rear Admiral ‘S’ (commander of the Submarine Service). However, by the late Spring of 1923, it was decided that he should take over from Cumming, whose health was failing, as Chief of SIS in September of that year. Following Cumming’s death on 14 June, he took up his new post on 3 September.