by Alastair Denniston- Code-breaking From Room 40 to Berkeley Street
Following the public release of the Zimmermann Telegram, intelligence continued to flow between Britain and the United States. On 31 August 1917, Page informed President Wilson that:
Admiral Hall has given me a number of documents comprising German cipher messages between German diplomatic officers and the Berlin Foreign Office, chiefly relating to the Argentine and definitely implicating the Swedish Government. In view of the negotiations now going on between Germany and the Argentine, the British Government hope that you will immediately publish these telegrams asking that their origins be kept secret as in the case of the Zimmermann Telegram. I have the cipher originals and sending them to you by a trustworthy messenger [Wiseman] who will deliver them into your hands about 12–15 September. These telegrams will prove that Sweden has continuously user her legations and pouches and her code to transmit official information between Berlin and German diplomatic offices.
Hall would share diplomatic and naval intelligence with the Americans when it was to the Allies’ advantage, but he would not share cryptographic expertise or assist them in their own efforts in the field despite representations from the State Department and the Office of Naval Intelligence. The US War Department set up a cipher bureau in June 1917 and recruited Herbert O. Yardley as its head. The US Navy tended to defer to Yardley’s group and Hall supplied an out-of-date HVB book used by the German High Seas Fleet until March 1915. Meanwhile, the US continued through Bell to provide Room 40 with intercepted messages, some old and some new from Stockholm and other neutral countries. By July 1918, Harrison was pressing for complete cooperation and exchange of information. Yardley visited London in August 1918 and later claimed80 that Hall refused to give him the VB code but did give him that of a neutral country and promised to send a two-volume German naval code to Washington for his use. This claim is considered unlikely by most historians.81
***
By 1917, interception was almost 100 per cent complete for North Sea traffic. Later in the war, the British tried to organise more systematic interception. De Grey led a small cryptanalytical team to Italy (Otranto and Rome) to work on Austrian naval codes but the Austrian fleet made little use of W/T. A separate unit was set up in Malta to work on the German U-boat codes in use in the Mediterranean. Room 40 reached its peak towards the end of the war and kept pace with successive innovations and complications introduced by the enemy. Early traffic analysis and direction finding became increasingly important and kept the intelligence flow going while new systems were being overcome.
According to Admiral Sir William James, writing in December 1936:
It is extremely unlikely that we will enjoy all those remarkable advantages in another war. The whole of the movements of our main Fleet and the defence of our trade were to a great extent dependent on this particular form of intelligence which was hatched in Room 40. But a point that is very often forgotten is that … it was during the last year of the War that a system was introduced whereby an intelligence section, thanks to careful study of everything that came their way, were actually able, at one period, I remember, to keep up a flow of intelligence when there was no cryptography which was very little different from that which was being issued when we enjoyed 100 per cent of cryptography.82
The large number of translated decrypts soon swamped Operations Division (OD) and Room 40 began to weed out unimportant communications. This was necessary as OD did not have the capability to thoroughly analyse all of the traffic coming through. Recording and research devolved to Room 40. And from February 1916, their selected and annotated decrypts were supplemented with a daily summary. Room 40 gradually comprised carefully selected, trained and experienced intelligence officers and by the summer of 1918, the organisation was operating effectively. From 1 July 1918, a branch of Room 40’s Intelligence Section ‘was set apart to survey and examine all wireless messages as they arrived and to report from time to time as necessary on the development of the situation’.83 Another unit, divided into parties specialising according to area or type of vessel, carried out intelligence research. It could be said that Ewing’s small cryptanalytic of 1914 had developed by 1918 into a full-blown naval Sigint centre.84
All Sigint organisations have to deal with problems, both personal and political, and Room 40 was no different in this regard. According to Frank Birch, the official historian of British Sigint from 1914 to 1945:
[The] Admiralty’s failings in dealing with Sigint were due in part to Oliver’s overwork and a reluctance to delegate as COS. He slept in the War Room, rarely left it and insisted on drafting reports based on Sigint himself. Beatty complained that Room 40 gave Oliver ‘priceless information which he sits on until it is too late for the Sea Forces to take action. What it amounts to is the War Staff has developed into a One Man Show. The man is not born yet who can run it by himself.’85
The un-naval language of the civilian cryptographers annoyed navy personnel. Even AGD, despite years of teaching German at Osborne, translated an early decrypt as ‘The Fleet will proceed into harbour athartwise’. One of his colleagues produced an equally bizarre translation: ‘Fisherman have reported that a destroyer with a bulwark over the sternpost rammed and sank a submarine.’ Through Hope’s efforts, Room 40 staff increasingly used correct naval jargon.86 Traditional naval men such as Rear Admiral Thomas Jackson (DOO) were annoyed by professors dressed in RNVR uniforms but who forgot to salute, mislaid parts of their uniform or wore their cap back to front.87
William Clarke later wrote of Jackson: ‘Admiral Thomas Jackson displayed supreme contempt for the work of Room 40. He never came into the room during the writer’s time there except on two or three occasions, on one of which he came in to complain that one of the locked boxes in which the information was sent him had cut his hand, and on another to say, at a time when the Germans had introduced a new code book, “Thank God, I shan’t have any more of this damned stuff!”’88
Hope left Room 40 in early 1917 to serve at sea, was awarded the DSO and became a Rear-Admiral after the war. He was replaced by Admiral Sir William James89 who had been Hall’s executive officer aboard the Queen Mary. James decided that all those not already in uniform should receive commissions as lieutenants in the RNVR. While some of the senior civilians in Room 40 were not keen on the idea, AGD was not one of them and duly received his commission as a lieutenant commander on 8 July 1917.90
By the autumn of 1917, the battle with the U-boats was turning in favour of the British as Room 40 continued to locate German submarines and forecast their intentions. It was working smoothly and seemed to be able to handle anything which came its way. Even when the Germans introduced a more frequent change of key and some new ciphers, apart from the odd delay, Room 40 was able to keep a steady flow of intelligence to OD and the Commander-in-Chief up to the end of 1917. In April 1918, it became clear that the Germans were trying to maintain wireless silence. By November, there were indications that the High Seas Fleet was sailing but then Room 40 became aware that the German sailors were refusing to put to sea. Hall’s work was done as the war wound down and from July 1914 until November 1918, he had not had a day off. While some thought of him as a future Foreign Secretary, he had also made enemies. He had crossed swords with the press barons and by-passed regulations to get things done and therefore was not liked by the civil staff who felt they had lost control of the Navy. He had hoped to attend the peace conference as Head of the Intelligence Bureau but was told that he was not going. He also received no postwar honours.91
Writing after the war, Churchill gave his view of the contribution of British naval Sigint:
Our Intelligence service has won and deserved world-wide fame. More than perhaps any other Power, we were successful in the war in penetrating the intentions of the enemy. Again and again the forecasts both of the military and of the naval Intelligence Staffs were vindicated to the wonder of friends and the chagrin of foes. The three successive chiefs of the Naval Intelligence Division, Captain Thomas Jackson, Rear-Admiral Oliver
and lastly, Captain Hall, were all men of mark in the service, and continuously built and extended an efficient and profound organization. There were others – a brilliant confederacy – whose names even now are wrapt [sic] in mystery. Our information about German naval movements was principally obtained (1) from the reports of secret agents in neutral and enemy countries and particularly Germany, (2) from the reports of our submarines, which lay far up in the Heligoland Bight in perilous vigilance, and (3) from a special study we had made of the German wireless. In this we were for a time aided by great good luck.
The Admiralty thus carried to an unrivalled and indeed unapproached degree of perfection our means of fixing the position and, by successive positions, the course of any enemy ship that used its wireless installation.92
German histories after the war concurred with this assessment and according to Vice-Admiral Reinhard Scheer, Commander-in-Chief of the German High Seas Fleet from 18 January 1916:
The English received news through their ‘directional stations’ which they already had in use, but which were only introduced by us at a much later period. In possessing them the English had a very great advantage in the conduct of the war, as they were thus able to obtain quite accurate information of the locality of the enemy as soon as any wireless signals were sent by him. In the case of a large fleet, whose separate units are stationed far apart and communication between them is essential, an absolute cessation of all wireless intercourse would be fatal to any enterprise.93
The use of intercepted traffic by Allied forces during WW1 to determine the enemy’s order of battle, played a vital role in their ultimate success. This early form of traffic analysis, which would prove so effective for Allied Forces in WW2, had a major impact on the land war in Europe during WW1. According to the head of intelligence of the American Expeditionary Force (AEF) in 1917–18, ‘frequently, as many as two-thirds of the identification of enemy divisions along the front became known due to the ability of the Allies to decode intercepted wireless messages’.94
It is hard to determine exactly how many encrypted (using either codes or ciphers) German messages on the Western Front were broken during 1917–18. From the Spring of 1917 until the end of the war, the radio monitoring personnel of each British Army intercepted around 150–200 German messages per week. On occasion, individual Allied stations intercepted 120 German radio messages each day.95 In 1917 alone, British codebreakers solved several hundred encoded German messages.96 In that year according to Friedman, ‘the text of one week was sufficient to break into a new [German] code, and by the end of three weeks, messages were being read by us as quickly and almost as completely by the code-officers as by the enemy’.
The French reconstructed some thirty front-line codes of the German Army during the war. In 1918, the Americans, British and French together continually penetrated one of the two main German trench code systems, had much success with the other, and sometimes solved both of the two main systems used near the front lines. Ferris estimates that the British codebreaking effort probably yielded in terms of text, up to twelve stencilled pages per day.
Between January 1915 and July 1916, the British Army paid little attention to Sigint but from July 1916, it was recognised as an increasingly valuable and reliable form of information. By 1918, the British Second Army noted that the,
special nature of wireless intelligence called for very close liaison between I(e) and the personnel responsible for assessing the enemy’s order of battle. At the same time, according to a well-informed American source, British GHQ regarded signals intelligence on the enemy’s dispositions and order of battle as being ‘always correct and, … by far the most valuable identification of divisional positions and intentions outside of capture of a prisoner. They are even more valuable in a way because they anticipate coming events before prisoners can be obtained’.97
According to Ferris, a reasonably efficient system emerged during WW1 in which Sigint was collected by signals personal who were generally controlled by intelligence officers. It was assessed by the latter before distribution to operations staff and commanders. The British Army handled Sigint as well as any other army during WW1. Sigint staff grew from seventy-five in mid-1916 to 13,300 by the end of the war. In 1917– 18 around 250 men served with listening sets, 600 served in the eleven WOGS in the field and around 200 in the War Office.
Apart from Room 40 and MI1(b), there was a branch of the Intelligence department of the Indian Army as well as agencies at the GHQs of the army in the field. In December 1917, Hindenburg’s Chief of Staff, Erich von Ludendorff, warned the army commanders that:
From a map issued by the British Intelligence Service (that is, by the British General Headquarters (GHQ) in France), captured at CAMBRAI, it appears that the enemy was completely informed regarding the distribution of our forces in the line and the divisions which had been withdrawn. On the other hand, he was almost completely in the dark regarding the divisions newly brought into back areas of the Army. Judging from the military situation at that time, it is to be supposed that the enemy obtained part of his information from prisoners’ statements. The larger part, however, he undoubtedly obtained from his Intercepting Service. The enemy, and particularly the British, installs his mobile intercepting stations even under the most difficult conditions. The fact that our signal discipline is completely wanting, particularly on battle fronts, plays directly into the hands of the enemy’s Intelligence. The fighting troops must understand the necessity for this discipline, otherwise there is grave danger that the enemy will prematurely learn our intentions, which are calculated on surprise and will be able to take counter-measures accordingly.98
Little is known of AGD’s personal work during WW1. While he would later document the work of Room 40, he always resisted the temptation to describe his own work. There is no record of his activities apart from examples of decrypts bearing his initial. (see Appendix 4). However, as a senior member of the Watch, he would have had a full workload. Despite this, AGD found time for romance. A watchkeeper had been working in Room 40 from 1914 to 1917 as a linguist with good French and German. Dorothy Gilliat was one of five children of a successful Leeds businessman, Arthur Gilliat. Known as ‘D’ to her sisters, she was born in 1891 and read English at Lady Margaret Hall (LMH), Oxford from 1911 and was known to all her friends there as DG. AGD’s sister, Biddy, was one of her year at LMH. She graduated in 1914 and did clerical work in Leeds Military Hospital. She was remembered by friends there as:
Very pretty and attractive with blue eyes, her ready and sympathetic smile, and a quizzical, rather deprecating lift of an eyebrow. She had poise and elegance, she loved social life, and worked the dons hard in those days of the chaperons. But there was another side. D.G. was a fine student who used her excellent brain, but with no parade of industry. Under her charm and a gentle mocking façade was a serious attitude to life. In anything she did she was utterly reliable and competent; she was never shattered by disaster, but undaunted in every crisis. To her friends her sympathy was unfailing.
AGD began courting Dorothy and in 1917 finally asked her to marry him. In a telegram to her parents in Leeds, she sought their approval: ‘Greatly surprised. Deeply thankful. Please may I marry Alastair. Wire reply to D R, 5 War Office Annexe Whitehall Place.’ Their response was positive and AGD and Dorothy were married the same year. The following year, on 7 January 1918, AGD was appointed an Officer of the Order of the British Empire (OBE), an honour which had been created in 1917 by King George V.
At the 11th hour on the 11th day of the 11th month of 1918, WW1 came to an end. At 5.00 am that morning, Germany, now lacking manpower and supplies, facing imminent invasion and revolution at home, signed an armistice agreement with the Allies in a railway carriage outside Compiégne in France. The war left nine million soldiers dead and 21 million wounded, with Germany, Russia, Austria-Hungary, France, and Great Britain each losing nearly a million or more lives. In addition, at least five million civilians died from disease,
starvation or exposure. Vast crowds gathered in London’s Trafalgar Square to celebrate but the joyous mood was shortlived. Post-war Britain was facing a range of political, economic and social problems.
In December 1918, a month after the Armistice, the Room 40 team celebrated their wartime success (some 15,000 German messages decrypted and read) with a concert party in Chelsea. AGD and a colleague composed and sang a duet for their team:
While some say that the boche was not beaten by Foch
But by Winston or Ramsey MacDonald
There are others who claim that the coup de grace came
From the Knoxes (our Dilly and Ronald)
It was Tiarks and Thring who with charts and with string
Gave the U-boats their oily quietus
Yet without the Lord Mayor in his diplomat’s lair
The Huns might have managed to beat us.
There are Zeppelins about, the key isn’t out
And Lord knows what’s afoot in the Bight now
When the tube basket’s crammed and each message is jammed
Operations want all the news right now ….
There was also a performance of a parody of Alice’s Adventures in Wonderland, written by Frank Birch and Dilly Knox. This party marked the end of Room 40 as most of the staff were returning to their pre-war lives. It would be left to AGD to keep Britain’s cryptanalytical capability alive for the next twenty years, as the gathered group would hear at the end of a song near the end of the show: