List of Abbreviations

PART ONE

EARLY DAYS

1

The beginnings of the Service

SIS began in a curiously understated way. On 7 October 1909 Commander Mansfield Cumming, the founding Chief of the Service, spent his first full day at work. ‘Went to the office’, he wrote in his diary, ‘and remained all day, but saw no one, nor was there anything to do there.’1 Indeed, for about a month Cumming had little to do, until he and Captain Vernon Kell, who together had been appointed to run a Secret Service Bureau, were able to sort out the duties of their new organisation. Part of the delay in getting started stemmed from the very novelty of the enterprise. Its interdepartmental nature also held things up, entailing some delicate manoeuvres over the relative roles of the sponsoring departments - Foreign Office, Admiralty and War Office - a problem which was intermittently to recur during SIS’s first forty years. The profound secrecy of the new Bureau - another continuing feature - also made it difficult for Cumming to get going as quickly as he wished. By the end of 1909, nevertheless, he had successfully established an embryonic organisation devoted to the clandestine collection of foreign intelligence, which in form and function was recognisably the forebear of the Secret Intelligence Service, as it was eventually to become known.

Foreign threats, spy fever and the Secret Service Bureau

The Secret Service Bureau was established at a time of heightened and intensifying international rivalries when British strategic policy-makers were becoming especially concerned about the challenge of an aggressive, ambitious, imperial Germany. For most of the nineteenth century, the United Kingdom had been by far the most powerful country in the world, possessing the greatest empire ever seen, and Britain’s leaders had been able to pursue a policy of so-called splendid isolation, largely impervious to any serious threat from other countries. But by the end of the century Britain’s economic lead over the rest of the world was beginning to be eroded, and as rival countries started to catch up, the very extent of British power - what the historian Paul Kennedy has called ‘imperial overstretch’ - came to be regarded as a potential weakness. In 1906 a Foreign Office official characterised the British Empire as being like ‘some gouty giant’, with fingers and toes spread across the world, which could not be approached ‘without eliciting a scream’. In a series of strategic reassessments in the first decade of the twentieth century Britain sought to ease its international position by coming to terms with potential Great Power rivals. Over a five-year period between 1902 and 1907 agreements were made with Japan, France and Russia which eased British naval commitments in the Pacific and Mediterranean, and (temporarily at least) removed the appalling prospect of having to defend the great British imperial possessions in the Indian subcontinent against Russian aggression. At the same time it was effectively assumed in London that there would never now be a war against the United States, thus further easing the burden of defending Britain’s worldwide empire.2

One major challenge remained, that of imperial Germany, which, not apparently content with being the strongest economic and military power in Continental Europe, by the early 1900s, in evident emulation of Great Britain, had begun to construct a first-class navy and seemed set on carving out a global imperial role. With Britain aligned to Germany’s Continental rivals, France and Russia, in what became known as the Triple Entente, policy-makers and public opinion began to worry about the direct threat that might be posed by Germany. Sensational stories of German spies and underground organisations ready to spring into action in the event of a German attack (or ‘bolt from the blue’) were fuelled by alarmist ‘invasion scare’ books such as William Le Queux’s bestsellers, The Invasion of 1910 (1906) and Spies of the Kaiser (1909), which reinforced widespread concerns about British vulnerability among public and government alike. In the War Office department responsible for army intelligence matters, the Director of Military Operations himself, General John Spencer Ewart, and his colleagues Colonel James Edmonds and Colonel George Macdonogh were all convinced that their opposite numbers in the German General Staff were actively targeting Britain. As Nicholas Hiley and Christopher Andrew have shown, however, the fears of German clandestine networks in Britain were wildly overblown - fantastic even; there were no legions of German spies and saboteurs. Yet they seemed to hit a Zeitgeist in Britain where generalised (and well-founded) concerns about a growing relative international weakness readily fuelled fevered speculations about foreign agents flooding the country and working towards its destruction.3

Such was the strength of public opinion that in March 1909 the Prime Minister, Herbert Asquith, responded to the spy fever by appointing a high-powered sub-committee of the Committee of Imperial Defence (the main British defence policy-making body) to consider ‘the question of foreign espionage in the United Kingdom’. Chaired by Richard Burdon Haldane, Secretary of State for War, the committee included the First Lord of the Admiralty, the Home Secretary and representatives of the Foreign Office and Treasury, along with Spencer Ewart and his Admiralty counterpart, Admiral Alexander Bethell (Director of Naval Intelligence). As well as assessing the danger arising from espionage in Britain, the sub-committee was charged with considering whether any alteration was ‘desirable in the system at present in force in the Admiralty and War Office for obtaining information from abroad’.

‘System’ was in fact putting it rather strongly, since the existing arrangements for acquiring foreign intelligence were notably haphazard and unsystematic. British army and navy requirements fell into two clear categories: first was primarily technical information about new weapons developments and German military capabilities generally; second was the establishment of some reliable system to give early warning of a German attack. In 1903 William Melville, the Kerry-born former head of the Special Branch in the Metropolitan Police, had been taken on by the Directorate of Military Operations primarily to tackle German espionage in Britain, but he also sent his assistant, Henry Dale Long, on missions to Germany under commercial cover apparently to investigate naval construction. From time to time foreign nationals offered to sell information to the British. Army officers also did some of their own intelligence work. In 1905 James Grierson, Ewart’s predecessor as Director of Military Operations, himself visited the Franco-Belgian frontier, and between 1908 and 1911 Ewart’s successor, Henry Wilson, accompanied by fellow officers, cycled up and down both sides of France’s eastern frontier with Belgium and Germany, exploring possible lines of attack for a German invasion as well as noting (among other things) German railway construction close to the Belgian border.4

Between March and July 1909 the Committee of Imperial Defence sub-committee met three times. It heard Edmonds describe how both the French and the Germans had well-organised secret services. His evidence ‘left no doubt in the minds of the Sub-Committee that an extensive system of German espionage exists in this country’ and that Britain had ‘no organisation for keeping in touch with that espionage and for accurately determining its extent or objectives’. The committee were also told that Britain’s organisation for acquiring information about developments in foreign ports and dockyards was ‘defective’, particularly regarding Germany, ‘where it is difficult to obtain accurate information’. Both the Admiralty and the War Office observed that they were ‘in a difficult position when dealing with foreign spies who may have information to sell, since their dealings have to be direct and not through intermediaries’. At the committee’s second meeting (on 20 April) Ewart asked ‘whether a small secret service bureau could not be established’, and a further sub-committee, chaired by Sir Charles Hardinge (Permanent Under-Secretary at the Foreign Office) and comprising Ewart, Bethell, Sir Edward Henry (Commissioner of the Metropolitan Police) and Archibald Murray (Director of Military Training) was deputed to look into the matter.

On 28 April 1909 Hardinge’s sub-committee submitted a report which ‘in order to ensure secrecy’ was not printed ‘and only one copy was in existence’. Their proposals effectively constitute the founding charter of the modern British intelligence community. They recommended that an independent ‘secret service bureau’ be established which ‘must at the same time be in close touch with the Admiralty, the War Office and the Home Office’. It should have three objects. It would first ‘serve as a screen between the Admiralty and the War Office and foreign spies who may have information that they wish to sell to the Government’. Second, it would ‘send agents to various parts of Great Britain and keep [in] touch with the country police with a view to ascertaining the nature and scope of the espionage that is being carried on by foreign agents’; and third, it would ‘act as an intermediate agent between the Admiralty and the War Office and a permanent foreign agent who should be established abroad, with the view of obtaining information in foreign countries’. The committee thought that this individual could be located at the Belgian capital, Brussels, and could be ‘the medium through which other British foreign agents sent in their reports, such a course being less likely to excite suspicion than if these agents communicated with Great Britain direct’. It was proposed that the Bureau should include ‘two ex-naval and military officers’, with ‘a knowledge of foreign languages’. On the recommendation of Sir Edward Henry, it was agreed both to employ a firm of private detectives for the work and ‘that a specially competent agent should be sent out . . . to get in touch with men in various German ports who would be willing to send us information particularly in time of strained relations’. The overall cost of the Bureau was estimated at something over £2,000 a year (the equivalent of about £150,000 in current money), to be met, at least in part, out of ‘the present secret service vote’.

An interesting feature of these recommendations (which were entirely accepted by the main sub-committee at their final meeting on 12 July 1909) is the marked bias towards foreign intelligence-gathering contained in the proposed ‘objects’ of the Bureau, a contrast with the original focus on domestic counter-intelligence. It is tempting to ascribe this to the chairmanship of Sir Charles Hardinge, responsible, among other things, for disbursing Secret Vote money. Under Foreign Office control, the Secret Vote had for many years been used for a wide variety of purposes, including payments to both the War Office and the Admiralty for the intermittent employment of spies, and, although the service ministries had clearly positioned themselves as the primary customers for the proposed new Bureau, we might see Hardinge’s hand both in the ‘external’ emphasis of his sub-committee’s proposals and in the explicit acknowledgement that funding would be provided from the Secret Vote.5 Whatever the explanation, the pattern of armed service engagement with, and Foreign Office control over, the secret service was one that persisted for the next forty years. Another aspect of the proposal was the extreme secrecy within which it was made, and the official ‘deniability’ under which the new Bureau would operate. A précis of the sub-committee’s findings, prepared at the time the first staff were appointed, noted that ‘by means of the Bureau, our N[aval] and M[ilitary] attachés and Government officials would not only be freed from the necessity of dealing with spies, but it would also be impossible to obtain direct evidence that we had any dealings with them at all’. This, too, was to be a central and lasting feature of the Service.

Once the formation of a Secret Service Bureau had been approved by the main Committee of Imperial Defence on 24 July, a group met on 26 August to work out the details. Sir Edward Henry and Ewart attended, along with Edmonds and Macdonogh. Bethell sent a staff officer, Captain Reginald Temple. The meeting accepted Henry’s recommendation that Edward Drew, a former police chief inspector and now a private detective, should be engaged and that the Bureau should begin work as soon as possible in offices leased by him at 64 Victoria Street in Westminster. It was agreed that Long, who had been employed ‘for some years’ by the War Office, should be the foreign agent based on the Continent. Evidently, Long had already been approached as he was ‘willing to accept the appointment’ and had agreed to ‘obtain a commercial agency in Brussels to cloak his activities’. It was further noted that an agent had been ‘employed in Germany by the Admiralty’ to cover German ports as suggested by Hardinge’s sub-committee. The 26 August meeting was also told that the War Office and Admiralty had officers in mind to staff the Bureau. The War Office proposed Captain Vernon Kell, ‘an exceptionally good linguist . . . qualified in French, German, Russian and Chinese’ (and who had previously worked in the War Office as Edmonds’s ‘righthand man in the Far East section’), while the Admiralty nominated Commander Mansfield Smith-Cumming, ‘who is now in charge of the Southampton Boom defence, and who possesses special qualifications for the appointment’. Confirming the continuing senior role of the Foreign Office in the new organisation, on a note of the meeting it was added that the Director of Military Operations (Ewart) ‘spoke to Sir C. Harding[e] on 14 September, and he concurred in the above arrangements’.6

Sorting out practicalities

The selection of Mansfield Cumming (he tended not to use the ‘Smith’) as the Admiralty’s nominee for the new Bureau was a classic and pioneering example of the informal way in which for a long time the Secret Intelligence Service treated the important subject of recruitment. The fifty-year-old Cumming (born on 1 April 1859) had no apparent intelligence experience. Unlike the War Office’s nominee, Vernon Kell (who, in fact, was to take over the domestic side of things), he was not a linguist, and it is not at all clear what ‘special qualifications’ Cumming actually possessed for the work, nor do we know if he was the only candidate considered.

Cumming, whose original name was Mansfield George Smith, came from a moderately prosperous landed and professional family (his father was a distinguished engineer) and entered the navy after going to the Royal Naval College, Dartmouth, at the age of twelve in 1872. He enjoyed an apparently successful, but not especially outstanding, career at sea and in various shore appointments (including a spell at the Royal Naval College Greenwich at the same time as the future King George V), before retiring in December 1885 ‘on Active Half Pay’ due to unspecified ill-health. Over the next decade or so he worked as private secretary to the Earl of Meath, serving for a time as his agent in Ireland. During these years he married twice: first in 1885 to a South African, Dora Cloete, who died in 1887, and secondly to May Cumming (when he adopted her family name), an independently wealthy woman whose family had estates in Morayshire, Scotland. In April 1898 he returned to the navy to ‘superintend the working of the Boom defence at Southampton’. Cumming was a practical man, much enthused by the latest mechanical devices. A keen pioneer motorist, and a hair-raisingly fast driver, he joined the Royal Automobile Club in 1902, and three years later was a founder member (and first Rear Commodore) of its offshoot the Motor Yacht Club (also ‘Royal’ from 1910). In 1906 he was a founder member of the Royal Aero Club, acquiring a pilot’s licence at the age of fifty-four in November 1913.7

There are some hints that Cumming’s involvement with motor-yachting (which he shared with many other naval officers) may have helped put him in the frame for the new venture. In the early years of the century the Admiralty, intensely interested in the potential of new types of marine engines, was kept fully informed about the activities of the Motor Yacht Club, which ran international racing competitions and encouraged the development of high-performance motor boats. Its concern was by no means confined to British developments. According to the memoirs of another motor-yachting pioneer, Montague Grahame-White, in the spring of 1905 Cumming was sent ‘on a tour to study the development of motor propulsion in fishing fleets in Sweden and Holland’, in order to ascertain ‘the reliability of internal combustion engines running on paraffin’. 8 Perhaps referring to this mission, Cumming wrote in his diary in late October 1909 that he would ‘like to get in touch with certain Danes and Swedes - with some of whom I made acquaintance when sent abroad recently by the F.O. [Foreign Office] in connection with Marine Motors’.

With his infectious enthusiasm for high-powered cars and motor boats, Mansfield Cumming was celebrated as a pioneer in ‘motorism’.

So it was that by the time he was selected for his new job, he appears to have had some experience of information-gathering abroad. All we know for certain, however, is that a month after the Committee of Imperial Defence sub-committee agreed on the formation of a Secret Service Bureau, Alexander Bethell wrote on 10 August 1909 to Cumming observing that ‘Boom defence must be getting a bit stale with you’ and that he might ‘therefore perhaps like a new billet’. Bethell had ‘something good’ to offer and he invited Cumming up to London to discuss it. Two days later (as Cumming recorded in his diary), Bethell told him ‘that the appointment he had to offer was that of Chief of the S.S. Service [sic] for the Navy - a new Department about to be formed at the insistence of the I[mperial] D[efence] C[ommittee]’. The work ‘was to be the obtaining and collecting of all information required by his Department. I was to work under him and should have charge of all the Agents employed by him and by the W[ar] D[epartment].’ Bethell also told him that he would have a ‘junior’ colleague (Kell was fourteen years younger and of lower equivalent rank), and that the new service was being set up with the agreement of the Director of Military Operations.9

While little in Cumming’s background seemed particularly suited to secret service matters, he undoubtedly grew into the work, and was certainly attracted by the prospect from the start. ‘The offer of the work is most tempting,’ he wrote to Bethell on 17 August, ‘and I should like very much to undertake it,’ but he also made it clear that he was by no means tired of boom defence, and during August and early September managed to establish that he could continue nominally to be in charge of that work while taking on his new duties. Since Hardinge wanted the Bureau to come into existence on 1 October, Drew’s office was rented from that date and towards the end of September there was a flurry of activity sorting out arrangements. On 23 September, highlighting difficulties about the precise division of responsibilities which were to bedevil the early months of the organisation, Cumming was ‘disappointed’ to learn from Bethell that he was ‘not to be Chief of the whole Bureau’, but that Kell ‘was to work with me on equal terms’. Bethell also told him that ‘no recognition of our work would be possible, as we were to be dissociated from the authorities entirely, and not recognised by them except secretly’. More positively, however, Cumming learned that Hardinge ‘had promised that there should be no stint of money to pay our own Agents &c’.

The first formal meeting of the Secret Service Bureau took place in the War Office on the morning of Monday 4 October 1909 when Edmonds and Macdonogh briefed Cumming and Kell about their new responsibilities. They said that they were going to keep Melville (‘the best man we have at present’) in ‘an office of his own’, and that Long, ‘another good man’ who spoke German and French, would be sent to Brussels ‘to act as chief agent there’. Edmonds also referred to some other individuals who had done intelligence work and might be kept on. He gave Kell and Cumming their first instructions about what would become known as ‘tradecraft’. He ‘told us never to keep names and addresses on the same paper’ and ‘never to use paper with a water mark in it’. They should never ‘see any of these scallywags for the first time without M[elville] or someone present’, or use the office as a rendezvous. A private room elsewhere should be rented for the purpose. ‘We were not to address letters from the office or receive letters there, and we were to assume other names.’ Cumming noted in his diary (perhaps this was a joke), ‘K[ell] added a Y to his present name,’ but carefully did not commit his own proposed sobriquet to paper, though he later used the names ‘Captain Currey’ and ‘Captain Spencer’. At the end of the meeting it was settled that as Kell ‘was not free for a fortnight, I should commence work by copying out all the records in M[acdonogh]’s office - as soon as I had procured a Safe in which to keep them . . . I lunched with K and we had a yarn over the future, and agreed to work together for the success of the cause.’

For some time, in fact, Cumming remained underemployed. Even before the 4 October meeting he had sketched out plans, but had not as yet shared them with anyone else. The main focus was on Germany and he considered that he ought to have agents in the major German naval ports (such as Wilhelmshaven, Hamburg and Kiel) ‘who could be thoroughly trusted to report extraordinary activity’. He also wanted ‘at least one travelling agent’. ‘Cover’ for the Bureau and its activities was a problem from the start. Cumming argued that he should have ‘some Official nominal post - for such instance as “work in the N[aval] I[ntelligence] Dept in connection with suspected persons in the Dockyards”. Such a post (not publicly announced)’, he thought, ‘would give me some pretext for trying to get help outside.’ Meanwhile he found it difficult even to get the Bureau started. Finding himself sitting in the new Victoria Street office with nothing to do, he began to learn German to fill the time. When he went to the War Office ‘to take away the first batch of Records to copy’, Macdonogh ‘would not allow them to be taken out of the place’, and later wrote to Cumming to say that he ‘proposed to hand all the W[ar] O[ffice] work over to K[ell] and to communicate with him alone on such matters’.

Frustrated, Cumming complained to Bethell he had been told ‘that I am not to have letters addressed to the office, am not to see anyone there, nor address letters from there, so that I can not see what possible use it will be . . . Surely’, he continued, ‘we can not be expected to sit in the office month by month doing absolutely nothing.’ His ‘only object’ was ‘to make a first class success out of a new thing’. Noting that Macdonogh wanted to work through Kell alone, Cumming felt ‘that any exclusion of myself in favour of K would be fatal to my success’. He argued that it would be ‘best to keep the WO and Naval work separate’. He would be ‘quite content to have charge of the latter and leave all the former to K, but I must have an equal chance to carry out my part and I can only secure that by being put au courant with all that has been done and with the organisation at our disposal up to the present time’. Cumming also grumbled about what he regarded as the excessive degree of secrecy surrounding the new organisation. It was, he thought, ‘only necessary for us to keep our connection with those who pay us a secret’. He wanted to be able to ‘let it be known’ that he was ‘open to receive information’, while his own identity would be carefully concealed - ‘a matter easily arranged’ - and that any ‘connection at all with any authority that could be traced in any way’ should be suppressed. ‘Half the secrecy that we are maintaining’, he wrote, ‘is of no practical good.’

At last, on 21 October, at a meeting held at the War Office to sort out the working of the Bureau, the separate functions of what became the Security Service and SIS were first formally established. Macdonogh proposed that Kell ‘should undertake the whole of the Home work - both Naval and Military, “espionage and contre espionage”’, and that Cumming ‘should have charge of all the foreign - bothN&M’. As Cumming noted in his diary, the ‘S.S. Bureau’ was to have four ‘duties’: ‘1. Act as screens to Ad[miralty] & W.O. [War Office]. 2. Conduct investigations. 3. Correspond with all paid agents and persons desirous of selling secrets. 4. Act as representatives of Ad. & W.O.’ Cumming’s ‘espionage’ task was to ‘Organise an efficient system by which German progress in Armaments and Naval construction can be watched, being careful in doing so that every thing which would point to concentration should be reported’, while Kell’s ‘contre espionage’ was to ‘counteract all measures hostile to G.B. taken by foreign Governments’. Kell was to have Melville and Drew, though, in answer to a direct question, Macdonogh said he was to ‘avoid employing’ the latter if possible. The existing foreign agents were to be kept on, ‘and there would be about £2700 for them’. The first ‘object’ of the Bureau was to ‘obtain information of any movement indicating an attack upon this country’. Other tasks were to ‘watch all suspected persons - such as foreigners residing in British territory’ and ‘counteract’ the ‘formation of demolition centres’ in Britain. Finally, the Bureau had to ‘organise a scheme of permanent correspondents both at home and abroad, who will furnish information from within the enemies [sic] lines in time of war’. Cumming felt that he was at last making progress, though he thought his was ‘the most difficult part’ and that since Macdonogh proposed to hand both Melville and Long over to Kell, he would be left with ‘no one of any tried value’.

On 28 October a meeting (to which neither Cumming nor Kell was invited) chaired by Hardinge at the Foreign Office confirmed the division of responsibilities within the Bureau. Bethell afterwards told Cumming that the foreign agents were to ‘remain as at present’, but that there was ‘no more money’ and that Cumming was to see what he could ‘get done voluntarily’. With his actual duties defined, Cumming now turned to the practical arrangements. He thought that there were ‘several disadvantages’ in sharing an office with Kell and Drew. Although it was ‘a large place’ it was of ‘very little use’ and for his purposes quite insecure. He did not think that he could ‘create and develop the elaborate organization of which I am to be given charge, from the Office, as I am not to address letters from there, receive them there or see anyone there’. He therefore proposed renting a flat, which would include an office, where he would be available at all times of the day and night. ‘A separate Office - such as the present one - ’, he remarked, ‘immediately suggests a business, and invites interest and curiosity, but a private dwelling calls for no comment. ’ At home he could organise his work without raising suspicion or attracting attention, and he could meet agents and others in rooms hired for the purpose elsewhere. He would also set up a ‘photographic copying plant’ at the flat, where it could be ‘arranged so that no one will know of its existence’.

Reflecting on the sorts of people he would have to deal with in order to get information (and demonstrating that he had been thinking productively about the whole business of intelligence), Cumming contrasted his position with that of Kell. In Britain, he observed, ‘every third man one meets would be glad to help his country’, but ‘abroad the case is entirely different’. The Consular Service of official British representatives overseas was ‘expressly barred’, and it was ‘useless as a rule to approach natives and invite them to betray their country. Some of the lowest class may consent, but they will of course in turn betray us without scruple if it serves their purpose, and cannot be relied upon for the more important work.’ As to ‘Englishmen living abroad’, they would be ‘reluctant to do anything to damage the country they are living in’ and would ‘be fully aware of the risk they will run - to their business, or even their liberty - if found out’. Cumming wanted ‘free scope to make enquiries, sound every person likely to be of use’ and ‘be able to offer substantial retaining fees and rewards for valuable information’. While he had not so far ‘had time to think out any scheme for forming a system of look outs who will give us instant warning of the movements of ships, transports, concentration of foods and stores &c’, this was ‘the most important work of all’, and ‘the agents selected must be of reliable character and position, and will have to be paid in proportion to these advantages’.

But Cumming still worried about his prospects. On 3 November he jotted down some despondent reflections: ‘cannot do any work in Office. Been there 5 weeks, not yet signed my name. Absolutely cut off from everyone while there, as can not give my address, or be telephoned to under own name.’ Kell had ‘done more in one day than I have in the whole time’. The system had been ‘organized by the Military, who have just had control of my destinies long enough to take away all the work I could do, hand me over by far the most difficult part of the work (for which their own man [Kell] is obviously better suited) and take away all the facilities for doing it. Am firmly convinced that K will oust me altogether before long.’ Bethell came to the rescue. After Cumming had poured his heart out over dinner, Bethell assured him ‘that I need not do anything to justify my appointment. I must wait patiently for work to come. That I need not sit idle in the Office, but could go about and learn. That I should not be watched, and that he had every confidence.’ Bethell told Cumming he could say he was employed in the Naval Intelligence Department, but ‘must use great discretion in doing this’, and that he could rent a flat (‘at my own expense of course’) and work there. If the experiment was successful, the Foreign Office might take on the cost. Bethell finally told Cumming that the War Office was ‘to have nothing to say to my work, which is to be managed entirely by me, under him (the D.N.I.)’.

Although Bethell’s assertion was not strictly true, since from the start the Bureau had been conceived as an interdepartmental service (and the War Office was to remain an important ‘customer’), his confidence in Cumming and the reassurance of Admiralty backing was very welcome, and had practical effect three weeks later when arrangements were being made for Cumming to take over ‘B’, an existing War Office agent based on the Continent. The plan was for Cumming to use him to run agents at Hamburg and Wilhelmshaven and ‘one Travelling man’, but when Kell insisted that he would come to the first meeting between Cumming and B in order to pay B’s salary, Cumming put his foot down. With Bethell’s backing he got Ewart and Macdonogh to prevent this and they instructed Kell to hand over the money to Cumming to pay B himself. The meeting with B on 26 November 1909 was Cumming’s first encounter with a real spy. He was introduced by Edmonds, who had previously been running him but told B that henceforth Cumming would ‘deal entirely with him’. Bethell’s staff officer, Reginald Temple, attended the meeting to help with German translation as B (an Austrian) did not speak English. Cumming, who had been taking German lessons at the Berlitz Language School, could ‘follow what was said; but not enough to understand all his [B’s] ideas and opinions’.

The meeting went well. B ‘seemed to think that he should have no difficulty in getting us the information we wanted, as he said all the TRs [Cumming’s diary term for Germans - short for ‘Tariff Reformers’]10 were open to bribes and could not resist the sight of a gold piece’. It was settled that there should be ‘one man in Hanover - to attend primarily to Military matters, one in Wilhelmshaven and one who should travel about, making his headquarters in Stendal or Wittenburg, and visiting all the big [ship] Yards at least once in every three months’. When Cumming raised the question of ‘the 4 Dreadnoughts [battleships], supposed to be about to commence building at Pola and elsewhere in Austria’, B ‘jibbed at this immediately and said he was an Austrian and could do nothing that could hurt his native country’. Cumming thought him ‘an intelligent and bold man’ and that he would ‘probably prove my best aide, but’, he added, ‘the difficulty about his patriotic feeling for Austria will have to be considered’. Bethell was pleased when Cumming reported these arrangements to him. He believed the War Office ‘had evidently recognised that they had made a tactical mistake in dividing the work in the way they had done, and that I [Cumming] had really secured the more important part. He thought all would come right when we had settled down.’

Although Cumming did not finally hand in his keys to the Victoria Street office until March 1910, by the end of November 1909 he had moved into a flat in Ashley Mansions, 254 Vauxhall Bridge Road, and established an independent base there for his section of the Bureau. Early in the New Year he arranged a bogus ‘cover address’ with the Post Office - ‘Messrs Rasen, Falcon Ltd, Box 400, General Post Office, London’ - an alleged firm of ‘Shippers and Exporters’, thus establishing a precedent for the classic ‘import and export’ espionage cover. He had two telegraphic addresses, ‘Sunbonnet, London’ for ‘general use’ and ‘Autumn, London’ for ‘special use’. All correspondence to these addresses was to be redirected to Cumming at Ashley Mansions, any important ‘Autumn, London’ cables ‘by special messenger under a double cover’. Cumming stayed at Ashley Mansions until 1911 when he moved both his flat and his office to 2 Whitehall Court, between Whitehall and Victoria Embankment, adjacent to the War Office and not far from the Admiralty.

The War Office, having conceded responsibility for foreign work, handed over their ‘whole German Intelligence system’, including B (who had been ‘found valuable in Russia and might be required to work there again’), along with ‘the names and addresses of the 5 assistants’. By the end of 1909 Cumming had himself begun to run other sources. On 9 December he met ‘WK’ who had been an Admiralty agent, apparently reporting on German guns. Just before Christmas he briefed ‘FRS’, who was going to Fiume in Austria-Hungary ‘to find out what progress has been made in laying down slips for D[readnought]s’. On New Year’s Eve at the Royal Automobile Club in Piccadilly he met ‘D’, who was based in Hamburg and was one of three agents engaged to warn of likely war. Cumming ‘promised him £500 [an astonishing 25 per cent of his then entire budget] if he could send me accurate news of the imminence of war before any other agent, and at least 24 hours before any declaration or overt act’. Although ‘the best of the three “passive” agents’, D was ‘evidently timid and accepts as a foregone conclusion that at the outbreak of war he will fly (and bring his message with him)’. Cumming felt sure ‘that he has little resource, and would not risk anything at all to get his warning to us’.

Targeting Germany and running agents

In an exchange of notes with Bethell during January 1910 Cumming formally established his responsibilities vis-à-vis the Admiralty. Bethell laid down that Cumming was to hold himself ‘directly responsible to me for all matters connected with your duties’ and was ‘to assume charge of the entire S.S. Intelligence system outside the United Kingdom, the Military Officer who has been appointed as your colleague being responsible for the work at home’. His ‘principal duty’ was ‘to obtain early and reliable information of all important movements of Naval and Military forces’ in order to provide ‘timely warning of impending hostilities against this country on the part of any foreign state within the range of your information’. Cumming was also to meet requests for ‘special items of information required by the Naval and Military authorities’. For his part, Cumming confirmed that providing advance warning of war was ‘by far the most important part of the work’, but he also suggested ‘that a plan of action should be devised which will ensure the sending of information after hostilities have commenced and whilst the war is actually going on’. He recognised that this was a much more difficult task. Any agent (working, say, in Germany), ‘especially if of foreign birth, or if suspected in the smallest degree’, would inevitably be closely shadowed, and it would be ‘extremely difficult for him to get any information through. If caught, he will certainly be shot.’ Cumming also maintained that, while Germany was the principal target, he would like to have agents in neighbouring countries, arguing perceptively that it would ‘often be possible to get information about Germany through another country, and secrets that may be carefully guarded from us [in Germany] may be more readily accessible elsewhere’. Wisely, he also observed that political conditions in different European countries could rapidly change and that he needed back-up systems for the supply of information.

At this very early stage Cumming was already thinking sensibly about the problems of foreign intelligence work, but much of this was ambitiously optimistic, and would remain so for years. Even thirty years on in the Second World War, after dramatic advances in wireless technology, establishing reliable and secure agent communications from behind enemy lines proved very difficult indeed. In the meantime, Cumming’s progress in establishing the new ‘S.S. Bureau’ was embodied in a report he prepared in April 1910 covering its first six months of existence. After a slow start, he said that since December 1909 the work had ‘increased rapidly until at the present time I have as much as I can tackle’. He described how his ‘staff of agents’ was ‘of two kinds’. In the first place were those watching Germany who were simply ‘expected to keep a good look out for any unusual or significant movements or changes - either Naval or Military - and report them. From these agents’, he added, ‘“no news is good news” and in the absence of any evidence to the contrary, it is to be believed that they are doing their duty and are earning the pay they receive.’ The other agents were ‘those who in addition to giving warning of extraordinary activity on the part of those they are deputed to watch, are expected to collect information of all kinds and forward it to me at stated intervals’. As yet, he was unable to make any firm judgment about these men ‘as sufficient time has not elapsed since their appointment to enable me to form an opinion’. The ‘principal Agent’, B, was paid nearly £1,500 (equivalent in modern terms to about £110,000) for himself and three sub-agents, about whom Cumming knew nothing whatsoever. ‘I have never seen them’, he remarked, ‘or heard their names, and I am not by any means certain that they exist at all.’ In April, after a trip to Paris to meet B, Cumming (revealing that the peculiar world of intelligence was beginning to affect his thinking) confided to his diary doubts about the existence of B’s sub-agents: ‘I could not feel absolutely sure that he had three men in his employment at all, although perhaps this is only the suspicion that grows upon one after the first few months of this work.’

B was a problematic asset in other ways too. As Cumming noted in his April 1910 report, he was ‘a foreigner, a potential enemy (for he is an Austrian) and a professional spy’. On the other hand, he was ‘intelligent’ and had evidently ‘rendered good service in the past’. In February he had supplied detailed dimensions of the new 25,000-ton German battleship Thüringen, which B had acquired from a naval engineer in Bremen, and which the Naval Intelligence Department was pleased to have. But, on the whole, the reports supplied from B’s network were ‘very meagre’ and did ‘not up to the present justify the large salaries paid - more than is paid to all the other Agents put together’. B, argued Cumming, had ‘no incentive to send in good reports, as he is paid the same whether they are good or bad’, and he wanted to change the system to one whereby the agent would still be paid ‘a fairly large retaining fee for himself ’ (though clearly less than he was currently receiving) with further sums ‘on a liberal scale for all information supplied and approved’.

Reviewing his other agents, Cumming recommended that WK should be retained at a salary of £200 (equivalent to £15,000), ‘to rise to £240, if he gives satisfaction’. Normally based in Germany, WK had reported in December 1909 about torpedo-boat trials in the Baltic and off Wilhelmshaven, and on naval construction, including submarine work of which there had been a particular concentration so that ‘this branch of the Navy’ might ‘render valuable service in time of war’. In April 1910 Cumming sent him to the Austro-Hungarian ports of Trieste and Pola (now Pula) to investigate naval shipbuilding. WK’s report, limited somewhat by the fact that he had ‘no technical knowledge’ of warships and ‘had been much troubled by the police at P[ola]’, nevertheless partially confirmed a report which the agent FRS had delivered in January of clandestine warship construction for the Austro-Hungarian navy, which had provoked great interest in the Admiralty and which Bethell asserted was the best job of intelligence work done ‘since he came to the office’. Possibly with FRS’s report in mind, Cumming declared that his ‘most valuable information’ had been ‘procured by a man who was sent abroad to ascertain certain definite facts’. This contrasted with his three ‘passive agents’ on whom he depended for ‘early information of war’ and whose performance was ‘difficult to appraise as to their value, as they send in no reports’. Indeed, one of these agents, ‘U’, seemed particularly feeble. Although he lived for part of the year in Kiel and socialised constantly with German naval officers, he told Cumming that ‘he never asked any of them any questions, for fear of arousing suspicion - in fact, when the conversation turned upon Armaments &c, he always asked them to change it’. He assured Cumming, however, that should he ‘smell war in the air’, he would ‘at once hurry across the Dutch frontier and send us a telegram’, and even ‘if necessary follow it himself ’.

For Cumming, a more valuable source, and ‘one which it is hoped will be greatly extended’, was that of ‘voluntary help’ provided by British people ‘whose business or profession gives them special facilities for finding out what is going on abroad’. One such individual (called ‘Mr Queer’ in the diary) was a director of a British armaments company and reported that the German firm Krupp were buying up stocks ofnickel-tungsten steel for the manufacture of small guns. On returning in January 1910 from a business trip to an unnamed foreign government, he gave Cumming the specification of a heavy gun (‘to throw a projectile weighing over 500 kilos’) which the government in question had themselves got from Krupp. Among other things this revealed that ‘for their larger guns’ Krupp were using ‘Nickel-Chrome steel, entirely Oil-hardened and tempered’. Queer, who was evidently very well plugged into the European armaments industry, also reported that the Skoda Works in Bohemia had received orders from the Austrian government to manufacture big guns ‘for two ships of the “Dreadnought” class’. Cumming concluded his first report by stressing how little he and Kell had in common, and on security grounds urged that the home and foreign sides of the work should be completely autonomous. He further thought it a ‘pity that the S.S. Agent [that is, himself ] should be obliged to tell even his Chiefs what he is doing - certainly he should not tell more than one person’. Indeed, he concluded that ‘it would be far better if he could keep his work, his methods and all knowledge of his assistants entirely to himself ’.

One thing which Cumming omitted in his report was any mention of his own intelligence-gathering role. Over five days in February 1910, accompanied by Captain Cyrus H. ‘Roy’ Regnart, a Royal Marine who was Bethell’s assistant in the Naval Intelligence Department, he went twice to Antwerp to meet an agent who failed to appear. He had a more adventurous time in April when (again with Regnart) he went first to Paris to meet his agent B, then on to Liège where they were to meet ‘JR’, who had promised to provide intelligence on German airship construction and show them a new type of portable weapon he was to smuggle out of Germany. A firearms expert was brought from London, and Cumming organised a professional photographer to join them en route in Brussels. Unfortunately these elaborate - perhaps over-elaborate - plans broke down. Although JR arrived from Berlin, Cumming and the photographer got separated from Regnart and the expert, neither of whom made the rendezvous. JR (who perhaps had more experience than Cumming in these matters) altered the arrangements at the last minute, not bringing the weapon to Liège but keeping it at a previously undisclosed location about an hour’s drive away. Cumming and the photographer got there in mid-afternoon, and in failing light quickly had to set to work. JR produced the device only after Cumming had paid him ‘the 25 [pounds] agreed on, and also a further 10 for the answers to the questions sent by the airship people’. As JR claimed he was in a hurry to get away and take the weapon back to Germany, Cumming ‘had only a few minutes in which to handle the thing, make such measurements as I could, and make a thumbnail sketch. I did my best, but was surprised afterwards to find how much I had managed to leave out.’ All he could ascertain was that it had sights ‘marked from 100 to 700 M and from 800 to 1900 on the raised part’ and that it ‘fired 5 shots’. It all seems to have been in vain. Back in London, both the weapons expert and Macdonogh concluded that the device was of no special interest, and Cumming even offered to pay back the £25 he had given to JR. Macdonogh reasonably ‘said that it was not fair to make me responsible for all the failures - a certain proportion of which must occur - and that not many offices could afford to make good such losses’. Cumming, who was still at this stage meeting the rent for his combined flat and office out of his own pocket, mused ‘that no officer who had not private means could take my work at all, as I reckoned it cost me the whole of my pay to keep it going’.

On 9 May 1910 Cumming’s report was considered at a meeting of himself, Kell, Bethell, Ewart and Macdonogh. The division of the work into two quite separate offices was confirmed and the private detective Drew’s services dispensed with. On Macdonogh’s suggestion it was agreed that the budget (now £6,200) would in future be divided equally between Cumming and Kell. Two days later these proposals were broadly approved by Sir Charles Hardinge. When Macdonogh declared that the money allowed for the Bureau’s work was ‘not nearly enough for the purpose’, Hardinge said ‘that if the work required it, the amount must be increased’. Cumming was allowed to lay off some non-producing agents, reduce B’s retainer and pay him by results in future. Hardinge further agreed to cover the cost of Cumming’s separate office and increase the budget for his necessary travelling expenses, once again assuring him (and Kell) that ‘there was no wish to restrict the work in any way, and if the money already granted was not sufficient, more would have to be found from elsewhere. All he wanted was to make sure that it was spent wisely.’

The work of the Bureau

The typescript version of Mansfield Cumming’s diary, the most important single source for the early days of the Secret Service Bureau, is tantalisingly incomplete from the end of August 1910 until the beginning of 1914. There are no entries from 1 September to 21 November 1910. Only 6-18 January survives for 1911, and 1912 contains only a few days in January, March and December. For 1913 we have 1 January to 27 May; 26-27 June; 31 July and 10-31 December. From 1 January 1914 the original, handwritten desk diaries are available, though even these have frustrating gaps when Cumming, without any particular explanation, simply seems to have stopped writing it up. We know, for example, from Vernon Kell’s diary (itself a pretty sketchy document which covers only June 1910 to July 1911) that there were meetings to review the first year’s work of the Bureau in November 1910, and that Kell and Cumming both submitted formal reports, but neither appears to have survived.11 The growth and development of the Secret Service, however, can be followed in outline from the scanty minutes of five meetings of the committee which supervised the Bureau’s work, which met at six-monthly intervals from November 1910 to May 1913, chaired by Sir Arthur Nicolson, who succeeded Hardinge as Permanent Under-Secretary at the Foreign Office in November 1910.

Recognising that the Bureau’s work was growing rapidly, the first of these meetings (on 16 November) agreed the appointment of full-time assistants for both Cumming and Kell. Backed by the War Office, Cumming also asked for funds to base an ‘officer-agent’ (Macdonogh’s term) at Copenhagen who would develop a network of sources in Germany, primarily to report on ‘German naval construction and armaments’, and, if war became a possibility, provide information on ‘German naval mobilization & concentration, & the assembly of transports in German harbours & also regarding the movement of troops, especially either towards the north coast or towards Holland & Belgium’. Copenhagen was suggested not only because it was especially well located ‘for receiving information from the German Baltic & North Sea ports’, but also (as Macdonogh argued, though without providing any specific evidence) because ‘Danes in many ways make the best S.S. agents for employment’. Nicolson recognised that the decision to appoint a British intelligence officer permanently in a foreign country (who, in later SIS parlance, would have been the first overseas ‘head of station’) was not one to be taken lightly. This was as much a new departure as establishing a permanent Secret Service Bureau in the first place, and as General Wilson, who had become Director of Military Operations in August (and was to be very supportive of Cumming), noted, the intention was that ‘this appointment should be merely the start of a wider system’, with further ‘Branch agents’ appointed elsewhere if ‘this one proved a success’. Nevertheless, since from the start one of the chief purposes of the Bureau had been to distance the British government from the problematic business of secret intelligence-gathering, the Foreign Office clearly wanted to be reassured that no hint of official involvement might accompany this new development. Nicolson insisted that the matter be submitted to the Foreign Secretary, Sir Edward Grey, himself. Grey approved the proposal ‘in principle’, but wanted ‘a great deal more detailed information’ about both the circumstances and the individual suggested for the post.

Cumming’s candidate was Regnart, ‘an excellent linguist, speaking Danish, German & other languages & . . . very keen on S.S. work’. Interviewed about the appointment by Nicolson’s private secretary, Lord Errington, Cumming told him that Regnart had private means (which because of the poor rates of pay offered to full-time members of the Bureau was widely thought to be essential), ‘did not care for Society’ and ‘was prepared to sink his identity altogether - even to the extent of taking a shop under a trade name and working it as a bona fide business, to cover his real objects’. Errington ‘said that the FO did not wish to place themselves under any obligation to the officer, so that he could come to them in a year or twos time and say that he had lost say £5000, and want it good’. He was also most anxious about maintaining the secrecy of the matter and ensuring that there should not be the remotest possibility of the government being associated with the proposed intelligence work. How was the officer to ‘sink his identity’, he asked? ‘Did the attaches know him? If his letters went astray or were intercepted would he not be traced as having been at his present office?’ After Cumming had ‘reassured him on all these points’, Errington ‘appeared satisfied on the whole’, though he continued to quiz Cumming about the chances of disclosure. Cumming’s carefully prepared scheme was upset in November 1910 by the trial for espionage in Germany of two British officers, Lieutenant Vivian Brandon and Captain Bernard Trench, who had been caught red-handed in August with maps, notes and photographs of defence and naval installations on the German North Sea coast and along the Kiel Canal. The two men had been working primarily for Regnart (though Cumming had agreed to provide £10 for any ‘extra expenses’), and there were fears that his involvement might become known to the Germans.

In the end the appointment was not made and Regnart remained in the Admiralty. In May 1911 the next six-monthly Secret Service Bureau meeting agreed to reallocate ‘the £700 for a man at Copenhagen, who had been approved but never appointed’, in order to pay for work in Brussels. Reflecting continuing uncertainty about which department was actually in charge, together with a slight misapprehension about how Cumming’s organisation would work, Sir Arthur Nicolson referred to this as ‘an Admiralty agency in Brussels’. Nicolson had some sense of the interdepartmental nature of the Bureau and also ‘understood’ that ‘both the Admiralty and War Office would be directly under C who would arrange co-ordination of their work’. The committee were very satisfied with its work as a whole. On Wilson’s proposal they approved salary increases (from £500 to £600) for Cumming and Kell, for having ‘done excellent work’. Bethell also ‘spoke in high terms of their work’ and remarked that Cumming ‘had spent considerable sums out of his own pocket . . . and that he was most economical, travelling second class though he was entitled to first’.12

Six months on, the committee thought Cumming and Kell were still doing well: ‘Sir A. Nicolson, Admiral Bethell and General Wilson concurring, expressed satisfaction at the excellent work being done by both branches.’ Yet the evidence of the intelligence agency’s performance during the Agadir Crisis over the summer of 1911 is rather mixed. The crisis was set off by the arrival of the German gunboat Panther at the Moroccan port of Agadir which the French regarded as being within their own exclusive sphere of influence, and a sharp deterioration in Franco-German relations for a while seemed to threaten a war between the two countries into which Britain might be drawn. The Secret Service Bureau had been formed to provide intelligence during precisely this kind of situation. Towards the end of July, The Times reported that the German High Sea Fleet had begun its ‘annual summer cruise’ and that one of the German squadrons had passed through the Kiel Canal to the North Sea, demonstrating that the canal was ‘ready for war’. In London there were worries about the location of the German warships, and even that the British fleet might be attacked. During the evening of 26 July Macdonogh sought out Henry Wilson and told him ‘that our Admiralty have lost the German Fleet & have asked us to find them. Macdonogh sent [Bertrand] Stewart off to Brussels to see L. [probably Long] & send him round the German Ports.’ The whole thing’, wrote Wilson in his diary, ‘is a Pantomime.’13

The ‘pantomime’ turned into a disaster when a week later Stewart was arrested in Bremen and charged with espionage. By one account he was arrested ‘in bed at 1 a.m.’; by another he was in a public lavatory, attempting to destroy a code-book planted on him by a German double-agent. ‘The Hun police broke open the door of the privy, [and] he was arrested with the corpus delicti on him.’ Stewart, a thirty-nine-year-old London solicitor and officer in the part-time West Kent Yeomanry, was an enthusiastic amateur, who (perhaps disingenuously) claimed during his subsequent prosecution that he ‘only knew enough German to obtain his meals and to make himself understood in hotels and on railways’. On Macdonogh’s orders, in fact, Stewart was working directly under Cumming, who had sent him in the first instance to Nijmegen in the Netherlands to contact an agent called Verrue who was working in Germany. Inadvisedly accompanying Verrue across the frontier, Stewart had visited Hamburg, Cuxhaven and Bremerhaven before being arrested. When he returned to England after more than two years in a German prison, Stewart claimed £12,500 compensation, blaming Cumming, Macdonogh and Wilson for his predicament.14 Cumming hazarded that Stewart had been shopped to the Germans by his ‘passive’ agent ‘U’ (evidently Verrue), whom Bethell, with the wisdom of hindsight, argued might have ‘been a decoy all through’. It is ‘annoying’, he told Cumming, ‘but we must expect drawbacks such as these in this kind of business’.

During September 1911 Cumming’s ‘early-warning system’ also brought reports of threatening developments in Germany. On 4 September Wilson noted a report from an agent in Belgium that two German divisions were concentrating in Malmédy, just across the frontier, which, combined with other indications, seemed so ‘ominous’ that he briefed Winston Churchill (Home Secretary since February 1910) and Sir Edward Grey personally about the matter. Later in the month Wilson recorded several similar warnings, including on 18 September alone ‘no less than four reports of our S.S. from the frontier saying German troops were massing along Belgian frontier’. Alarmist reports of German preparations were circulated to senior ministers, including the Prime Minister himself, and, although they all came to nothing, this did not appear in any way to affect the reputation of Cumming’s Bureau (as the minutes of the November 1911 committee meeting confirm, and when it was given a ‘special grant’ of £500 ‘because of the crisis’).15

The minutes of the Secret Service Bureau committee meetings for November 1912 and May 1913 show continued support for Cumming’s expanding work. At the latter meeting a combined estimate was approved of £16,212 for both branches, as was Cumming’s scheme to develop a network of agents in Norway and Denmark reporting on German naval matters, especially ship movements through the entrance to the Baltic Sea. Sir Arthur Nicolson asked that a proposal in November 1912 to station permanent agents ‘in four continental ports’ (at a total annual cost of £1,600) be included on the 1913-14 estimates ‘and he would then consider it favourably’. Six months later, a slightly scaled-down scheme, costing £1,200, was approved.16 Late in 1912 Lord Onslow (who had succeeded Errington as Nicolson’s private secretary in May 1911) allowed £1,000 ‘for miscellaneous payments and contingencies’, which Cumming called ‘my special fund’.

Cumming also secured permission to expand his operations in Belgium, and he proposed that Regnart should be appointed as ‘Branch Agent’ in Brussels. This led to an extraordinary public disagreement between Cumming and Captain Thomas Jackson, who had succeeded Bethell as Director of Naval Intelligence (but with the new title, which prevailed for the next six years, of Director of the Intelligence Division) in January 1912. When Cumming explained to the May 1913 committee meeting that he wanted ‘to employ a certain Marine officer, who possessed special qualifications’, Jackson interjected that ‘he personally did not consider the man whom C wanted was suitable. He did not consider him either hardworking, clever or tactful, nor that he would be loyal to C, but that was C’s affair.’ Cumming persevered, asserting that ‘the officer in question was well fitted’. Henry Wilson supported him. ‘It was’, he said, ‘impossible to get a perfect man for the appointment, but he knew the officer referred to was keen on his work, a good linguist, and an artist in Secret Service.’ Jackson then changed tack and argued ‘that no officer should be selected for this work while still on the active list. In fact he should not even be offered it until he had retired and it should not be possible for him to say that he had left the service [the Royal Navy] in order to take up the job and thus establish a claim for compensation in case of discharge.’ No one disagreed with this, and in the end Cumming got his man.17 Nicolson ‘finally said that if & as soon as Roy [Regnart] retired, I could have him. McJ [Jackson] said we should all regret it, but it was decided that as I had to work with him, I should be allowed to try him.’ Cumming had no illusions that Regnart would make a congenial colleague. When considering him for the Copenhagen job he had reflected in his diary that he was ‘a very difficult man to work with, as he plays an independent game and will not submit to control - I shall find him a constant thorn in my side’. Yet he also believed that he was ‘the best man for the post’, and ‘I would rather risk a certain amount of personal discomfort and worry than have a secondrate man as my Chief Branch Agent’.

The disagreement over Regnart’s appointment illustrates both Cumming’s increasing confidence in his own judgment and a preparedness not to defer automatically to higher authority, as well as a shrewd appreciation of the variable range of personalities he had to deal with in intelligence work. One such was ‘Major H.L.B.’ whom he met in January 1911. He was ‘a curious looking man with a large hawk like nose, brown eyes and brown hair. Medium height - about 5.8 - rather showy dressed, with an enormous pearl pin.’ He had ‘a shifty look about him’ and his nationality was ‘very difficult to fix’. He claimed to speak many languages, but always travelled ‘on the Continent as a Spaniard speaking Spanish and Portuguese’. HLB told Cumming that he had enjoyed a cosmopolitan career as a soldier in Africa, Chief of the Secret Police in Bolivia, a French secret service agent and an arms dealer. He added that ‘the smartest Agents in the German service’ were in Britain, and one of them, called von Gessler, ‘may be known anywhere by his having 4 rows of teeth’. He said he (HLB) had a Peruvian ring which contained ‘an Indian poison’ that would ‘knock a man down in three seconds’. It was ‘a bit “risky” to use, as a microscopic overdose would make a strong man a hopeless idiot for life and beyond any chance of cure’. Cumming thought the man was ‘no doubt a blackguard, but is probably a clever one, and I should like to get something out of him. All my staff are blackguards,’ he continued, ‘but they are incapable ones, and a man with a little ingenuity and brains would be a change, even if not an agreeable one.’

Another thing which struck Cumming was the consistently high expectations of potential agents. In August 1910, one man whom he wanted to send on a tour of Germany told him that ‘he must have an allowance for Champagne’. In December 1910 the part-Russian, Dutch and English divorced wife of a German officer (whose conduct ‘had certainly been reprehensible’) offered to get information from a male admirer on the German Naval Staff. She said ‘she was not going to do this work for money, but to revenge the slight upon her honour and for the sake of her children’. She agreed, however, to accept £20 a month for ‘expenses’ and ‘was rather insistent upon a “guarantie” [sic] which’, noted Cumming with the benefit of not much more than a year in the job and an air of tired cynicism, ‘all spies ask for. They explain that it is a guarantee of good faith which is due to them in exchange for the compromising gift of their names and addresses, but my experience teaches me that it means an advance of payment followed by an unbroken silence.’ Later the same day Bethell told Cumming about a man who had ‘access to Krupps Yard’ and might be worth cultivating. Cumming thought that the only way of getting hold of him was ‘to ask him to eat and drink, and that all these people without exception make a strong point of doing this in the best style at the most expensive restaurants’. Two Danes who turned up in London, also at the end of 1910, offered Cumming an abundant selection of material, including enlarged maps of German naval bases, sketches of a torpedo mechanism and several signal codes as ‘used by Searchlights, Wireless from Submarines &c’. They asked for £5,000 (equivalent to £350,000 in modern prices) which in Cumming’s opinion ‘was not to be thought of ’. Even ‘if we had accepted their plans as genuine, we should have offered £200 for the lot’. In the end he paid them nothing other than £10 between them for their travelling expenses, although in conversation (and reflecting his longstanding interest in harbour defences) he had obtained ‘a good deal of information’ about underground tunnels in the various harbours and the existence of ‘land mines laid from a central firing station to different salient points’.

Over the first few years of the Bureau, we can see Cumming working on tradecraft. He appears to have enjoyed using disguise when meeting agents. For a rendezvous in Paris in July 1910 he ‘was slightly disguised (toupee & moustache) and had on a rather peculiar costume’. In preparation for meeting (in January 1911) a man he called ‘Ironmould’, an engineer who was offering to go to Trieste to report on Austrian naval shipbuilding, Cumming was made up at William Berry Clarkson’s famous theatrical costume shop in Wardour Street in Soho.18 The disguise was ‘perfect . . . its existence not being noticeable even in a good light’. Cumming ‘then went to a photographer and had a photograph taken of the disguise as it is necessary to give the dresser something to go by, if it is desired to repeat a disguise exactly - as in my case’. Cumming also developed other techniques to avoid being identified. In January 1913, meeting a contact called ‘Ruffian’ - ‘a plausible chap, but rather oily looking’ who claimed his brother-in-law worked as a foreman in a German gun factory - Cumming arrived by taxi, using a method which he claimed was ‘the best I have tried, as it is almost impossible for the man - if a rascal - to point you out to his friends, who may be waiting with a camera &c.’. The trick was to ‘drive past the rendezvous on the opposite side’ and when the target was spotted, ‘drive up close to him, open the door and invite him in. I lean back the moment I have caught his eye, and from then onwards do not show myself at all.’ After the conversation was finished the contact was deposited ‘at some point well away from the place of meeting . . . Of course,’ added Cumming, ‘this plan is too expensive for a prolonged meeting, but for a short one it is very good.’

Cumming brought his enthusiasm for the latest technology to bear on some of the intelligence challenges he had to face. Although wireless communications technology was in its infancy, he thought it offered a possible solution to the problem of getting rapid messages through during the period of political tension which it was assumed would precede any German invasion. In the spring of 1912 he discussed with French intelligence colleagues the development of a ‘mobile wireless station’ in a motor car, with a range of 250 miles, which would be based in Belgium. Although the French thought the Belgian authorities would not allow such a thing, they offered to see if ‘the Car &c’ could be purchased ‘thro’ a Belgian Agent (a Govt. official whom they knew of)’, but Cumming had his doubts about ‘the soundness of the idea’. Contemplating a scheme in January 1913 for a father-and-son team to gather information along the Danish and Norwegian coast using a thirty-ton motor pilot boat, Cumming reflected that ‘the knotty point about the transmission of news remains unsolved’. Late in 1913 Cumming was working on a scheme to base an aeroplane in France, which could be used to keep watch along the country’s eastern border, but Macdonogh baulked at the prospective cost - Cumming thought £1,700, the air experts £3,000 or more - and Wilson said it could be done only in co-operation with the French, which he promised to help with.

The approach of war

General Wilson as Director of Military Operations was a particularly active promoter of close Anglo-French relations, and his championing of the alliance with France was part of a wider and growing closeness between the two countries which was reflected in the intelligence world. In January 1910 Wilson’s predecessor, John Spencer Ewart, had told Cumming that he ‘did not wish any espionnage [sic] work done in France just now, as our present excellent relations might be disturbed thereby’. Cumming’s earliest liaison contacts with a foreign intelligence service came in March 1912 when he met officers of French Military Intelligence to discuss matters of common interest, inevitably concentrating on gathering information about German capabilities and intentions. The most senior French officer involved, Colonel Charles-Édouard Dupont (head of the Deuxième Bureau of the French General Staff from 1913 to 1918), ‘was quite disposed to be frank and friendly and so were the other officers we met subsequently’, but it was evident that they were ‘a little nervous about telling to strangers of another nation the matters they have kept secret for so long’. Dupont, however, was ‘very strong indeed as to the vital necessity of our meeting each other and deciding now at once upon a plan of concerted action to be taken when the crisis came’.

By 1913 the French and British were exchanging intelligence material and Macdonogh was directing requests to the French for specific information through Cumming. In January, for example, he asked about the composition of German armies on their western frontier and ‘a new powder for smallarms’. On 6 March Cumming noted that ‘a large packet of valuable stuff came in from our friends’, which he brought to Wilson, ‘who thought it extremely valuable and is to show it to Sir J[ohn] F[rench, the Chief of the Imperial General Staff] at once’. Wilson ‘talked to me about the state of affairs and said that it was an extremely good move to get hold of these people, and hoped that we were doing something for them’. Liaison with the French remained close. There is a hint in Cumming’s diary that by January 1914 he had an officer actually posted to the French headquarters. In July, on Macdonogh’s recommendation, Cumming took on Captain Edward Louis Spiers (‘a very good man & interpreter Fr., Ger. & Ital.’) and toyed with the idea of sending him to Paris.19 In February, with Henry Wilson’s help, he got French agreement for the aeroplane scheme. He purchased a machine and was making arrangements for its base in France in July, but the declaration of war at the beginning of August seems to have come before any practical use had been made of it.

During discussions with French opposite numbers in 1912 Cumming had discovered that the two services had some agents in common. One, ‘HCJ’, went to Russia and with Cumming’s permission submitted the same reports to both organisations. In April 1913 Cumming learned that HCJ was also working for the Russians. Over lunch at the Savoy Café in London, he told Cumming that he had been engaged by them to help overhaul their ‘heavy and clumsy’ intelligence organisation. On the same occasion he declared that ‘war was very probable between Russia and Austria as the Russians were prepared in a way we did not expect and would require only the slightest pretext to attack the Austrians’, a prediction which came to pass the following year. In the spring of 1914 Cumming raised the possibility with Admiral Henry Oliver (who, replacing Jackson, had been Director of the Intelligence Division at the Admiralty since November 1913) of working with the Russian secret service to co-operate in running a Danish network to report on Germany. Even if the Russians would not collaborate officially, Cumming thought of placing an agent in St Petersburg ‘to receive & forward telegrams’.

Cumming remained keen to develop official relations with the Russians. In June 1914 a French intelligence officer told him that ‘the new Chief of the Russ. SS’ was coming to Paris ‘& as soon as he knows of his arrival he will let me know & will introduce me to him’, adding (with that avoidance of official channels which characterises much intelligence work), ‘we need not trouble our Attaché in the matter’. The Russian, in fact, came to London, and Cumming met him (apparently along with the Russian military attaché in Paris, Count Ignatieff). No detailed record survives of what was discussed. A scribbled note in Cumming’s hand preserved in his diary says ‘Mobilisation Plan in 8 days. We to photograph’, though whose plan it was and where it came from is not revealed. But the French were involved too. A few days later Cumming noted a conversation with Admiral Oliver, ‘discussing SS matters & the new scheme with Fr. & Russians’. On 2 July Cumming interviewed a potential agent, a ‘good Russian - interpreter - speaks French, German, some Spanish & Hindustani’. He was the ‘agent for [a] patent motor car wheel’ and ‘could live in St. P[etersburg] on 400 [pounds]’. Cumming told him he ‘could promise nothing, but if Russ scheme went through, he might do as Agent there’. Russia was by no means the limit of Cumming’s ambition. During the spring of 1914 Oliver proposed basing an agent in China, at Kiaochow (Jiaozhou) near Tsingtao (Qingdao), where there was a German naval base, and the six-monthly meeting of the Secret Service Bureau committee in May 1914 allotted £200 for the purpose.

Over the first half of 1914 Cumming seems to have spent most of his time working on the deployment of agents along Germany’s western frontier, intended both to give early warning of a German attack and to provide the basis for intelligence reporting after war had started. There were two main networks based in Belgium. The first was run by Roy Regnart in Brussels and concentrated on the eastern frontier with Germany, the ‘Maastricht appendix’ (that part of the Netherlands which protruded south into Belgium and through which the strategists thought a German attack might come), and up into the Netherlands to Venlo and Nijmegen. From here he aimed to watch German ‘military centres’ such as Cologne, Münster and Oldenburg. The second network was run by ‘AC’ with a base at Lille in France and primary responsibility for southern Belgium approximately from Liège eastwards to the Channel coast. Reporting to AC was a local agent, ‘DB’, based at Dinant, south of Namur, who had his own network of sub-agents. Most of the reporting from these networks was practical. In January AC was asked to investigate Dutch railways in the Maastricht appendix and ‘to look out for possible Bridge sites (connected with roads) on the banks of the Meuse’ just north of Liège. In February DB submitted a report on the railway line between Roermond and Maaseik at the northern end of the appendix. Roy Regnart, troublesome as ever, complained to Cumming that he could ‘do nothing without more £ [money]’ and said it was ‘useless to try & get agents at the Ports & elsewhere’ unless he could ‘retain them at once and pay them’. He thought ‘he ought to have at least £500 a year & a free hand’.

Cumming’s networks in Belgium and the Netherlands, as well as shipping-reporting agents based in Denmark, failed completely to provide any advance warning for the assault on France which the Germans launched through Belgium in August 1914. While the broad location of the advance was pretty well predicted (though the Germans avoided the Maastricht appendix), the timing was a complete surprise. The real prewar successes for Cumming’s organisation were in technical reporting, especially on German naval construction. One of Cumming’s best agents (and the highest paid), Hector Bywater, whom he named ‘H2O’, produced a regular series of reports in 1913 and 1914. Details are sparse, but in April 1913 Captain Jackson in the Admiralty ‘remarked that H2O sent a lot of good stuff’. During 1914 one report on German naval guns was thought so significant that H2O was quizzed in person in London on its contents by Admiral Reginald Tupper, who had commanded the naval gunnery school at Plymouth. H2O also reported on aeronautical matters, another Admiralty priority. In January 1914 another agent supplied ‘a big report’ on dirigibles (airships). Macdonogh told Cumming that the Director of the Air Department in the Admiralty, Captain Murray Sueter, thought highly of the agent’s plans ‘& said we ought to pay him all we could afford’. Traces of the technical reporting from Cumming’s agents have survived in Naval Intelligence Division logbooks written up by the head of the German section, Fleet Paymaster Charles Rotter, collating information on German submarine and battleship construction. That Cumming was able to supply the navy with valued intelligence is reflected by an entry in the diary in March 1914: ‘Rotter asked me to get him information about secret building of Submarines. He says they speak of U.21 but may possibly be U.31. He says they have about 50 built & projected.’ Despite the success of Cumming’s agents in collecting technical intelligence, the impact of this work was less than might have been hoped. Nicholas Hiley has remarked on the resistance which much of this reporting encountered in the Admiralty where the preconceived ideas of some experts led them to question intelligence which stressed the great importance of German developments in torpedoes, submarines, mines and aircraft. 20

The growth of the Secret Service Bureau over its first five years is reflected in an Account Book for both branches of the Bureau covering April 1912 until September 1914. Its budget, estimated at £2,000 in 1909, had grown to nearly £11,000 by 1912. Although it was originally envisaged that the money would be divided equally between Kell’s Home and Cumming’s Foreign branch, in December 1912 Kell’s side cost £472 and Cumming’s £810, although this included two months’ payments of £125 to H2O. Cumming’s salary and that of his five home-based staff came to just under £200 a month, less than half the cost of overseas agents. By December 1913 the annual budget for both branches had risen to £15,572, and the proportion of expenditure going to Cumming’s side had increased slightly with Foreign getting £841 per month and Home £428. Although spending on Kell’s branch increased a little during 1914, on the eve of war it was still significantly less than that for Cumming’s. All this was about to change. Over the war years the size of both branches and their budgets would grow exponentially, as perhaps could have been predicted, even in 1914. But how well Cumming’s side would respond to the challenges of the next few years, and even whether it would survive as an autonomous institution at all, remained to be seen.

During the early summer of 1914 there is no sense of the imminence of war in Mansfield Cumming’s diary. The assassination of the Archduke Franz Ferdinand in Sarajevo on 28 June, and the ensuing July Crisis, pass unremarked. Only from the very last day of July is there an impression of unusual activity. That day Cumming had a meeting with a new recruit, Major Cecil Cameron, the agent AC and two others. Taking codes provided by Cumming, Cameron was to cross the Channel that night, meet a contact in Paris and go on to Dinant at once, before basing himself at Givet on the Franco-Belgian frontier. On 2 August he gave more codes to another colleague, who having borrowed a car and driver ‘started off for Brux[elles], via Dover & Ostend’. Intelligence from inside Germany inevitably continued to be a high priority. At the end of July Cumming got Admiral Oliver’s sanction to send a woman agent to Berlin and agreed to pay her £100 for a month’s work there. What precisely she was to do is not recorded, but the mere fact that Cumming was contemplating such an operation at such a time confirms the suddenness with which the First World War came upon Britain. On 3 August the prospective agent assured Cumming that she could get the correct papers to enable her to travel to Germany, but the next day he had to put her off ‘as she cd show no passport’. The last diary entry for that day, 4 August 1914, was ‘War declared against Germany - midnt’.

If you find an error please notify us in the comments. Thank you!