Between the wars SIS’s limited resources were thinly spread across the world. It was, moreover, never a very large organisation. Of the thirty-three overseas stations in the early 1920s, most were one-man operations (there were no female station heads until after the Second World War), with one or perhaps two secretaries (usually women) for administrative support. By 1938-9, there were sixty-nine officers and 134 ancillary staff in thirty-four overseas stations, and the overseas deployment of the Service amounted only to some 200 personnel, though there was, of course, also an uncountable number of agents. The expansion of the Service reflected the increasing demands being made of it, though this was constantly limited by the lack of funds. In November 1929 Valentine Vivian prepared a memorandum for the Foreign Office highlighting the rather modest resources devoted to secret service in Britain as compared to other countries, observing, moreover, that ‘No Government, except that of Great Britain, would seem to be so ingenuous as openly to budget for, and accurately to publish, the amount annually intended for Secret Service expenditure.’1 The Italian invasion of Ethiopia in October 1935 inevitably stimulated a sharp rise in demands for information from SIS which, in turn, led Sinclair on 9 October to sign off a strongly worded ‘memorandum on Secret Service funds’, in which the evident frustrations of his dozen years as Chief, trying to meet incessant demands with inadequate resources, fuelled a cri de coeur for increased funding. ‘Since the War,’ he began, ‘the British Secret Intelligence Service has been constantly hampered, by lack of funds, in the performance of its duties.’ The ‘want of money’ had been ‘increasingly felt in recent years, during which other Nations have turned Great Britain’s gesture in unilateral disarmament [under the ten-year rule] to account by seizing the opportunity to re-arm secretly’. In these circumstances, he argued, ‘official sources of information have proved even worse than useless’, with governments routinely dissembling on the issue. ‘The most glaring case in point’, he wrote, ‘has been that of Germany.’ Until the German government had ‘themselves made the facts public, practically the only sources of information on her re-armament were those available to the Secret Service’. It was, moreover, ‘a melancholy fact that the march of events has proved their information correct’.
Sinclair argued that ‘a great deal more could, and would, have been accomplished had the Service been in possession of adequate Funds’. Drawing a characteristically naval analogy, he said that his budget, which was meant to cover the whole world other than the United Kingdom, India and the colonies, ‘only equals that spent every year on the maintenance (not the cost) of one of H.M. Destroyers in Home Waters’, and was much less than the money devoted to secret services by Britain’s foreign rivals. Sinclair asserted that ‘a satisfactory Secret Service’ depended ‘essentially upon maintaining a complete and interwoven scheme of work’, with as wide a geographical presence and coverage as possible. But, ‘owing to lack of funds’, it was ‘not possible to maintain any Service in such countries as Switzerland, Spain, Yugoslavia, Albania and Arabia, and, in consequence, a great deal of information about other and neighbouring countries, which could be obtained in these countries, escapes us’. The ‘necessity for spreading the butter thin’, he observed, ‘has proved particularly unfortunate in the present Italian crisis’. For years, since Italy had ‘been regarded as a friend and ally, intensive S.S. work was not pursued against that country, the main effort being directed against Germany’ (an observation which made sense only when the targeting of Comintern operations emanating from Berlin was taken into account). The crisis had obliged the Service not only to divert funds from other work, but also to spend lavishly ‘on obtaining intelligence of a nature which S.I.S. is not normally expected to supply in peace time’ (tracking, for example, the movements of Italian warships).
Nor was Italy the only problem. Sinclair also remarked on the ‘unsatisfactory position’ regarding the Far East, ‘especially in regard to Japan’. Even ‘by diverting sums being spent, with valuable results, on other countries’, the amount available for recent efforts in the region was ‘wholly insufficient, and only suffices to maintain a skeleton organisation’. Sinclair stressed that, even when adequate funds were available, it took ‘at least 2-3 years, under the most favourable circumstances, to establish a satisfactory S.S. in any country’. Meanwhile SIS, ‘living, as it does, a hand to mouth existence, with vast areas to cover’, was able only ‘to scratch the surface’. To ‘obtain really inside information’, he wrote, ‘means spending big money’. He claimed that opportunities frequently occurred ‘of dealing with individuals in responsible positions’, but ‘the offer to them of the few hundreds a year, which represents the amount usually available, is naturally treated by them with contempt’. ‘Whatever may be the outcome of the present crisis,’ he concluded, ‘it is plainly apparent that, in the future, Germany, Japan and Italy will have to be regarded as potential enemies from without, as well as Soviet Russia from within.’ The situation ‘cannot possibly be covered by the existing S.I.S. Organisation, depending, as it does, upon a limited number of Passport Control Officers and representatives anchored to their posts in the capital, and possessing neither the means nor the mobility for covering the many industrial and strategic posts from which essential information can alone be obtained’. Sinclair asserted that a ‘complete plan’ had ‘been worked out for the establishment of a network of permanent resident agents at the vital points’ in Germany, Italy and Japan, but ‘to put this plan into execution’ would ‘necessitate a large increase in S.S. funds’. The precise amount was unspecified, but, without it, ‘no guarantees’ could be given ‘that the S.I.S. will be able to meet the demands of the Armed Forces Departments in times of crisis’.
Having sent copies of his paper to Hankey, Fisher, Vansittart and the Chiefs of Staff (he was clearly not taking any chances), Sinclair secured an increase in funding. From £180,000 in 1935 (at which level it had been since the late 1920s), the Secret Service Vote (covering MI5 as well as SIS) was increased to £350,000 for both 1936 and 1937, to £450,000 in 1938 and £500,000 in 1939.2 No evidence of the ‘complete plan’ for ‘permanent resident agents’ has survived in the Service archives (though this may have been the genesis of the Z Organisation created in 1936), and the expansion of SIS activity, especially in Western Europe, was incremental rather than dramatic. And while the budget increased very markedly in the later 1930s, there were still limitations on what could be done. Reflecting in November 1939 on SIS work since 1935, Commander Reginald ‘Rex’ Howard concluded that the Service had been ‘handicapped very considerably by lack of money’. Howard, a sailor who had served on submarines before the First World War, had joined the Service in November 1931 and became Sinclair’s chief staff officer in September 1935. Even as late as 1 June 1939, he had been ‘informed by C.S.S. that our activities had been considerably curtailed owing to lack of money, and later, towards the end of July, C.S.S. informed me that the position was even worse, and it was practically impossible to obtain further supplies’. Recruitment, too, was seriously affected ‘owing to lack of money’, wrote Howard, ‘and also to the fact that employment in S.I.S. is non-pensionable’. These grumbles about lack of resources, as well as (from the service departments) the inadequate supply of information demonstrate twin permanent and unchanging truths about intelligence: that, no matter what the circumstances, there is never enough money; and, equally, no matter how much information is provided, there is never enough of it.
Working in the USA
SIS retained a small but active presence under Passport Control cover in New York during the interwar years. From October 1919 until March 1922 Maurice Jeffes was Passport Control Officer, and was succeeded as Cumming’s representative by J. P. Maine. Captain Herbert Bardsley Taylor, who joined SIS directly from the Royal Navy, took over in June 1929 and held the position (symbolised in SIS as ‘48000’) until August 1937, when he was brought back to London to be assistant to G.2, who headed the Americas section. From then until the war Captain Sir James Paget was head of station. The forty-nine-year-old Paget was not an intelligence specialist. A baronet, thanks to his grandfather having been an extremely distinguished doctor, he had been a career naval officer before retiring in 1937. That he had no particular qualifications for the job (though his mother was American) rather confirms the comparative un-importance of the station by the late 1930s. Between the wars, however, part of the work was for the armed services. In June 1921 the War Office asked for information on American ‘inventions of war material, chemical warfare and aircraft developments’. In March 1922 Stewart Menzies told Sir Warren Fisher’s Secret Service Committee that both ‘the Air Ministry and the Admiralty considered it extremely important to keep a close eye on naval and aeronautical developments in the United States’. Indeed, the £85,000 annual budget under discussion, which was otherwise thought more or less acceptable, ‘would not give adequate scope for such work in America. To do properly what was required in that sphere,’ continued Menzies, ‘from £3,000 to £4,000 extra would be needed.’3 It seems that the committee accepted this point, since they found another £5,000 for SIS, but little enough of this appears to have been spent in the USA. The £8,520 allocated to the New York station in April 1923 (of which none was set aside for agents) all came from the Passport Control budget, as did the £5,507 earmarked for 1932-3. In that year, six agents were on the New York books, two unpaid and four paid, and of the latter only one was on a regular retainer (of $100 a month). By 1934-5 his retainer had been increased to $200 a month; one other agent was on $100; and a total of £900 per annum of secret service money was allocated for ‘Agents’.4
Substantial evidence of somewhat uneven reporting from New York on Communist and radical groups in North America has survived in the archive, but there is only a little relating to armed forces intelligence. In 1930-1 an employee in a firm of consulting engineers with American naval contacts was taken on as a ‘paid conscious agent’ who knew that he was working for the British. But the Admiralty was unimpressed with the result. One report, about solidifying gasoline, was described as ‘absolute nonsense’, and another, on steel density, was ‘obvious nonsense, and informant appears to be unacquainted with subject’. By 1932 the agent had been dropped. An undated report sent by New York in the early 1930s indicates the kind of naval intelligence sought (though not, alas, whether any was actually acquired). A British ex-Royal Navy sailor had two American brothers-in-law who were ‘working on one of the new cruisers’ being constructed in Philadelphia. Taylor was assured that the Briton could obtain from them ‘any information required’. In May 1934 London noted that if figures ‘giving the reserve ammunition for the United States Army’ could be acquired, ‘they would be of considerable interest’.
As in the latter years of the First World War, SIS’s North American operation continued to concentrate on the activities of Irish republicans and Indian radicals. These were reported to Indian Political Intelligence, Scotland Yard and MI5. Wiseman, Nathan and their associates had been so successful that in a February 1921 report on ‘British espionage in the United States’, the future long-serving director of the Federal Bureau of Investigation, J. Edgar Hoover, claimed that at the end of the war ‘the English were much better informed on radical activities in this country, at least in New York, than the United States Government’. By 1921, he asserted that the British were still very active and ‘must have a very efficient force in operation’. Hoover appended a ‘list of known British agents’, evidently based on extremely circumstantial evidence. It included such improbable characters as the Jamaican-born black activist Marcus Garvey, the Irish labour radical James Larkin and a ‘P. S. Irwin’, the ‘active head of an organization at Miami known as the “Overseas Club”’, an ‘international British Society with headquarters in London’. Irwin’s guilt was apparently due to his being ‘the only white man connected with the branch at Miami all the others being negroes’, and the fact that he appeared ‘to be greatly interested in collecting data concerning both the black and white races, which will interest the British Government’.5
The United States authorities regarded British intelligence activities with some ambivalence and revelations had the potential to be very embarrassing. In June 1921 Jeffes cabled that a former confidential secretary had ‘become violently mad. Detained in public hospital. Talking freely of secret service affairs.’ ‘Do you wish anything done?’ he asked. Cumming gave Jeffes ‘a free hand in the matter’, but little was possible. Warning Norman Thwaites, Desmond Morton remarked: ‘As a matter of fact, presumably so far as you are concerned, all that she can say is ancient history, and as awkward for the Americans as for ourselves,’ a consideration which ‘really applies to our show at large’. In March 1920 Frank L. Polk in the State Department, who had worked closely with Wiseman during the war, asked that the ‘British Secret Service office in New York’ restrict their activities as it ‘had been too well advertised in the past’. Yet, reflecting his passionately anglophile outlook, Polk does not appear to have demanded that SIS cease spying in the USA entirely, merely (as he wrote in his diary) ‘that they try to cover it up as the Irish and others were sure to make trouble’. American records also reveal that in 1925 the State Department certainly knew that British ‘officials’ in New York were targeting Indian nationalists in the USA.6
Much reporting, however, concerned left-wing subversives. In June 1927 Captain Hugh Miller of Scotland Yard forwarded a request from the United States embassy in London for information about a named American citizen who was suspected of having been involved ‘in the transmission of funds for Soviet propaganda in the Philippines’ (then under United States rule). Vivian reported ‘No Trace’, but added that a similarly named person was involved in a ‘reported Communist Centre in Macao’. SIS also volunteered information. In April 1928 Vivian told Miller that the SIS representative in Peking (Beijing) had obtained information from a source in Harbin that the Comintern had ‘ordered the despatch of six agitators from the Far East to the United States, with instructions to work among employees in the textile industries’. They were to be supplied with American passports and ‘chosen from graduates of the Lenin Institute for Propaganda’. Vivian said that the information could be passed on to the Americans ‘provided the source is carefully concealed and no indication whatever’ was given that it ‘emanates from Harbin’.
The SIS representatives in New York supplied a steady stream of reports on Communist activities in the USA, drawn from police and Military Intelligence contacts, as well as their own sources. While SIS regularly circulated this material, there were periodic doubts about its quality. In autumn 1928 Vivian complained that Jeffes’s successor J. P. Maine’s ‘style is peculiar, his views are exaggerated and his perspective is extremely doubtful’. When in June 1930 Maine’s successor, Taylor, asserted that there were ‘at least two million persons enrolled in Communist or Communist-affiliated organisations’, Captain Miller thought the figure was ‘very much exaggerated’. Even after Taylor had defended the figure, quoting an American journalist, Walter S. Steele, Miller still doubted it. Steele, he thought, was ‘a journalist of the excitable order, who is inclined to lump all Radicals as Communists’, and it was ‘quite improbable that a vast organisation as carefully and systematically articulated as that which he describes can exist in the U.S.A.’.
In March 1932, Vivian, responding to a report on the left-leaning League for Independent Political Action, charged Taylor with being too unselective in his reporting. ‘The U.S.A.’, minuted Vivian, ‘is the home of crank societies - most of them semi-social and semi-political - but practically all devoted to the generation and release of hot-air.’ Unless they had ‘a real connexion with international subversion movements’, reports about them could ‘serve no possible purpose here but fill up space in the registry’. Taylor had claimed that the League’s aims were ‘merely the Communist programme re-written’, but Vivian said that was ‘not in itself sufficient’ to identify the organisation ‘with subversive Communist propaganda or organisation to make the report worth recording. After all,’ he added sarcastically, ‘some of J. M. Keynes’ tenets coincide with Communist tenets, but we don’t keep a file of his students of Economics.’ The Chief himself instructed Taylor to be more discriminating. Before he sent a report ‘on any of these unimportant societies’, insisted Sinclair, ‘the test should be applied - “Is there any proof of any connection whatsoever between the Society concerned and organised Communism?” If the answer is “No” the report is not required.’ There were other criticisms, too. In May 1934 London told Taylor (who had been asked in August 1930 to ‘keep a sharp look-out for personnel suitable for the penetration of Japan’) that ‘the information concerning Japan’ in a recent report was ‘hopelessly inaccurate’, and that ‘concerning the United States’ had ‘obviously been taken from public speeches’.7
There were limitations to what SIS could achieve in the USA. In May 1935 Kathleen ‘Jane’ Sissmore of MI5 raised with Valentine Vivian ‘the poverty of your information with regard to the progress of Communism in the United States’, with one notable exception, a network run by Jay Lovestone in New York, which had been comprehensively penetrated by SIS. Lovestone, who led the anti-Stalin Communist Party (Opposition), had worldwide contacts, and London was particularly interested in information about those in Britain and the empire. Canadian names were passed on to the Ottawa authorities, while others, including the Indian Communist M. N. Roy, the Trinidadian ‘rabid Trotskyite’ C. L. R. James and the liberal-Marxist British intellectual Harold Laski (who could hardly be described as a ‘subversive’), were passed on to Scotland Yard, MI5 and Indian Political Intelligence. Flatteringly for SIS, an October 1935 report quoted Lovestone (who was planning a trip to Europe) as saying ‘that the British Intelligence Service was the only thing he had ever been afraid of ’ and he was very fearful of being arrested if he went to England.8
On 30 December 1936, Sinclair, taking a Christmas holiday trip across the Atlantic, inspected the New York office himself. The books and card index, he reported, ‘were well kept and up to date’, but the office, ‘though clean, presents a somewhat dingy appearance, owing to the furniture and fittings being completely worn out, having been in use for over twenty years’. He immediately ordered Taylor to arrange ‘the complete refurnishing of the office’ and interviewed the four members of staff ‘most of whom are elderly married men [who] appeared rather down at heel’. Having satisfied himself that, ‘owing to the extremely high cost of living in New York, they had not really enough to live on’, and in a gesture that showed why his staff liked him so, he ordered supplementary payments to be made to them out of the secret service ‘Other Moneys’.
Generally the British and United States authorities co-operated well, particularly concerning the Bolshevik target, and their intelligence representatives in the Baltic states pooled reports. Between the wars, however, US-UK intelligence liaison was principally handled through the United States embassy in London, originally with Basil Thomson and Scotland Yard and more latterly with MI5. In October 1937, observing that the British had ‘for some time been seriously worried by the development of German Nazi and Italian Fascist organisations within the British Empire’, Guy Liddell (writing from MI5 on behalf of Vernon Kell) proposed to N. D. Borum at the United States embassy that ‘the official exchange of information that has operated between us so successfully over a period of eighteen years on Comintern affairs’ should be extended to cover German and Italian matters. Washington was not keen, distinguishing between the activities of the Comintern, with which the Soviet government had consistently denied any link, and those of German Nazi and Italian Fascist political organisations which were ‘admittedly connected with the political parties controlling the governments of Germany and Italy respectively’. Reflecting that this position was not in practice utterly inflexible, the State Department official John Hickerson nevertheless allowed ‘the possibility of exchanges of information in specific instances where such exchanges appear to be mutually appropriate and advantageous’.9
But Liddell wanted something more formal, to build on MI5’s success in helping the FBI round up an important German spy ring operating in the USA which had been communicating with Germany through a Mrs Jessie Jordan in Perth, Scotland.10 In the spring of 1938 he visited the USA and determined that both the military authorities and the FBI were ‘more than anxious to establish a liaison with us, which could cover not only Soviet, German and Italian activities, but also those of the Japanese’. The difficulty, he observed, was how to do so ‘without causing offence to the State Department, with whom we have been in touch via the Counsellor of the American Embassy here ever since the war’.11 Liddell thought that there was now ‘a unique opportunity’ to capitalise on the ‘existing good relations and reinforce cooperation which might prove of vital importance’ if the liaison developed ‘in future emergency or war’.
SIS was keen, also, to improve 48000’s position, as he had hitherto ‘never had a really good working arrangement with the U.S. authorities’. But, as Vivian explained to his colleagues in London, Liddell’s own contacts had been ‘based on the assumption that Great Britain has clean hands, so far as the U.S.A. is concerned, and that we indulge in no espionage activities whatever in the U.S.A., which, if discovered, would undoubtedly destroy mutual confidence and put an end to such liaison’. Before proceeding (and naming 48000 - by now Sir James Paget - as the primary link with the United States authorities), Liddell wanted SIS ‘to make a frank avowal of the Rutland business’, which had preoccupied both SIS and MI5 for some time. Since 1933 it had been known that Frederick Joseph Rutland, a former RAF officer and expert in naval aviation, had been working as a spy for the Japanese. Although he had been based in the USA and was working against American aviation targets, no word of this had been breathed to the Americans. 12
The Rutland case was one thing, but what Liddell did not know was that SIS was currently ‘actually engaged in air and naval espionage against the U.S.A.’. When Vivian told him this, it appears to have shocked him rigid (Liddell, Vivian noted with considerable understatement, ‘was definitely not happy’). It clearly jeopardised his existing contacts and also made it ‘quite impossible for him to sponsor 48000 as the local representative of British-American Intelligence liaison in the United States’. For SIS, therefore, as Vivian put it to his colleagues in May 1938, ‘it is for us to consider whether our Air and Naval work against the U.S.A. is of sufficient importance to maintain against the potential advantages of a satisfactory liaison’. The issue went round to the various sections in Head Office, who (though in some cases slightly grudgingly) agreed that active espionage in the USA might be stopped. The Air Section, for example, reported the Air Ministry’s opinion that collaboration had improved recently and the United States air attaché had been ‘considerably more open’. As a result, they considered that they could obtain through the attaché ‘any information which we should otherwise have got’. They additionally expressed the hope that the enhanced MI5 liaison on counter-intelligence matters might desirably develop into ‘collaboration with the Americans in obtaining information on Germany and Japan’.
With assurances from all the sections that, in effect, it was more productive to be friends with the United States than continue to treat it as an intelligence target, on 7 June 1938 Sinclair ordered that ‘work against America is to cease as soon as possible’. ‘We hope’, wrote Vivian to Liddell, ‘that this will finally clear the way for a valuable liaison, which will not stop at the exchange of information regarding international subversive movements, but will expand into a solid Anglo-American liaison on the German and Japanese activities which threaten the interests of both countries equally.’ In fact matters moved rather slowly thereafter. During 1939 Paget began making direct contacts with the FBI and both the Army and Navy Departments, but as the international situation deteriorated, and especially after the outbreak of war on 3 September, the State Department became, as Liddell put it, ‘anxious to bottleneck everything and not to let the soldiers and policemen get loose on their own’. Accordingly, ‘in view of the delicate political situation’, Paget was ordered to restrict his contacts to the State Department. ‘At the same time, using all your tact’, he was asked if he could ‘see if there is any likelihood of [James] Dunn [head of the Political Relations Department and Paget’s contact at State] receding from the position he has taken up regarding any of the other departments’. Before very long, indeed, there would be a massive expansion of SIS work in North America and greatly enhanced Anglo-American intelligence liaison, managed, however, not by Paget but by William Stephenson, his successor as 48000.
The Far East
During the early twentieth century the United Kingdom had very extensive commercial, imperial and strategic interests across South-East and East Asia. Flanked by Britain’s imperial possessions of India and Burma, and with garrison outposts at Hong Kong in the south and Weihaiwei (Weihei) in the north (the latter held until 1930), China was especially important for British trade and investment. With its weak central government and perennial internal divisions, China presented more of an opportunity than a threat to British interests, while Japan, from the late nineteenth century a rapidly industrialising and ambitious maritime power, represented a very serious potential challenge. While this had been alleviated by the Anglo-Japanese Alliance of 1902, and Japan had fought (though fairly half-heartedly) on the Allied side in the First World War, some British policy-makers continued to regard it as a real threat, particularly after the alliance lapsed in 1923. In London, too, there was an increasing appreciation in the 1920s and 1930s that Britain’s imperial resources were scarcely adequate to meet its responsibilities to defend its widely scattered Far Eastern interests. In such circumstances good and timely intelligence could be of vital assistance.13
From a very early stage Cumming had intentions to develop secret service work in the Far East, but while his postwar plans for the region were quite expansive, they appear to have been of comparatively marginal importance. When in 1920 Winston Churchill presented his Cabinet colleagues with schemes for £125,000 and £65,000 Secret Service budgets, the biggest cut Cumming proposed for the latter was the Far East, from £15,000 to £1,000.14 Bearing in mind the Admiralty’s interest in the region, however, this may have been a tactical suggestion on Cumming’s part. If so, it seems to have been successful, since in 1923 (even after the Secret Service allocation had been reduced to £90,000) the Far Eastern Group retained a total budget of £18,200, of which only £1,200 was Passport Control money. But the reductions which affected all aspects of SIS work had their impact here too. In 1934-5, the next year for which we have reliable figures, only £6,460 was devoted to both China and Japan.
As elsewhere, SIS’s Far Eastern dispositions in the 1920s were primarily directed against the Bolshevik target. In December 1920 Cumming agreed with the India Office jointly to send out Godfrey Denham, who had been Deputy Director of the Delhi Intelligence Bureau (which monitored subversive activities within India), to be ‘head of the organisation’ in the Far East, the India Office ‘paying him his salary & office fund’, and Cumming ‘providing £10,000 a year for 3 years of our work’. The following February Denham went out to Shanghai ‘with a mandate to be in charge of all our work in Japan, China, Tibet and Siberia and Southern Asia’. In June 1921 he produced a forty-five-page paper on ‘Bolshevism and Chinese Communism and Anarchism’ which pulled together all the available information on left-wing personalities, organisations and activities in China. His conclusion was one of balanced gloom: ‘Thus with Bolshevik activity in the North and a growing Chinese Anarchist party in the South, both of which are extremely anti-British in their propaganda, there is every need for care in the future.’ In 1921-2 Denham reported extensively (and expansively) on the political situation in both China and Japan. Had he confined himself to the collection and circulation of secret intelligence he might have ruffled fewer feathers, but his mixture of intelligence and assessment inevitably trespassed on the role which the regular diplomats saw as theirs alone. A long paper, entitled ‘“Dangerous thoughts” in Japan’, warning about the spread of anarchism, was particularly ill received in the Tokyo embassy. Denham’s ‘account of socialistic and other revolutionary movements in this country’, complained the ambassador to the Foreign Secretary in September 1922, ‘does not agree with the other information in my possession’. Denham, for example, had estimated membership of revolutionary societies at some 57,000, while the embassy thought the figure was nearer 7,000.
After the Indian government had withdrawn their support (for reasons of economy), and SIS had proposed reducing his pay, Denham left China in 1923 to become Inspector-General of the Straits Settlements Police in Malaya (and later a successful businessman). At some point in early 1923 a vice consul at Shanghai, Harry Nathaniel Steptoe, was recruited to help plug the gap left by Denham’s departure. The thirty-one-year-old Steptoe had first gone to China in 1912, served in the West African Frontier Force during the First World War and, exploiting a facility for languages, joined the Consular Service in China in 1919.15 Initially given the designation ‘C/33’, he began well and, over nearly twenty years with SIS, embodied both the strengths and, increasingly, the weaknesses of the Service’s Far Eastern work. In October 1923, the SIS representative in Singapore, who seems to have assumed some of the supervisory functions formerly exercised by Denham, hoped Morton would let the Chief know ‘how deeply we are indebted to C/33 for keeping Shanghai going during the last very difficult period of over half a year . . . In fact had it not been for his almost superhuman energy the whole show there would have dwindled to a dangerous degree.’
The Singapore representative’s proposals for what he called the ‘Shanghai Agency’ envisaged a junior role for Steptoe with a view to his eventual long-term employment in SIS. In the meantime Steptoe was fitting in SIS work with his ordinary consular passport and visa duties and, as yet, was not ‘sufficiently qualified to enable him to be a fully satisfactory whole time representative for S.I.S.’. These reservations throw light both on Steptoe and on some of the qualities thought necessary for a ‘whole time’ SIS officer. It was thought that Steptoe was a bit ‘young’ and ‘does not always perhaps control his indignation or his zeal when, for example, hostile criticisms are directed against the S.I.S.’. The Singapore representative had spoken to Steptoe about this, ‘saying that it was only the bitter pangs of experience which had caused me to become a hypocrite when I deemed it necessary, thereby, at times, being able to turn an unfriendly critic into a friendly patron!’. He felt that Steptoe would be able to take over the Shanghai station in about two years and that in the meantime another member of the Consular Service could take Denham’s place.
There were some delays before this individual’s secondment to SIS could be finalised in June 1924. Meanwhile Steptoe, who continued to fill in, was ‘far from fit’ and beginning to show signs of the chronic ill-health that was increasingly to plague him. In January 1924 it was noted that Steptoe was ‘very fully occupied as Passport Officer, etc. all day, and his work for us is done at nights, on Saturdays and on Sundays’. Despite these concerns, and after he had gone home on sick leave, in July Singapore again reported on Steptoe in glowing terms: ‘He has a natural flair for our work, and I hope you will keep your eye on him for future work.’ Once Denham’s successor was installed in Shanghai (where he remained only until August 1925), Steptoe was posted to Peking, with cover as a local vice consul. In June 1925 SIS provided the British minister, Charles Palairet, with evidence that the Soviets were fomenting unrest in China. This was a photograph of a letter signed by the Russian ambassador, ‘procured from a very secret source’, which instructed ‘local committee in Shanghai to prevent strikers from returning to work and to “incite labouring masses by meetings”’. Palairet, in turn, passed it on ‘privately’ to the Chinese government.
The so-called 30 May Massacre at the International Settlement in Shanghai in 1925, when Chinese demonstrators were shot by British and British-Indian police, was followed by a wave of strikes and anti-British protests in southern China and Hong Kong. Perhaps slightly defensively (but also evidently to flag up SIS’s capabilities in the region), Sinclair told Nevile Bland at the Foreign Office that ‘in view of the present trouble at Shanghai, it may be of interest to recall that we gave advance information of this in April last, which has already been confirmed up to the hilt by what has actually happened’. In April, for example, SIS had circulated ‘a translation of a very secret despatch dated 26 February, 1925, from the Executive Committee of the 3rd International to its centre at Vladivostock’, containing ‘clear proof of the implication of the 3rd International in the present strike movement in Shanghai’. On 25 June Sinclair sent Tyrrell at the Foreign Office a three-page ‘recapitulation of information, mainly documentary, and obtained from a number of sources’, which ‘clearly shows that the unrest is very largely due to the intrigues of the Soviet Government, and has been very cleverly organised by them’. SIS also acquired copies of the correspondence of the Soviet ambassador, L. M. Karakhan, through an employee of the Soviet consulate-general in Shanghai, despatches which were so inflammatory that they thought ‘he must have been drunk when he wrote them’.
Steptoe had other interests, too. In an intriguing signal to London in June 1928 he proposed paying $2,000 to an American naval rating stationed on the Philippines ‘to supply detailed information on Corregidor defences’. In July 1929 he took on a Chinese man in Shanghai (for ‘$100 a month, plus reasonable expenses’). ‘In addition to knowing the leading Chinese, both bankers, officials and merchants, he also knows the leading members of the present Nanking [Nanjing] Government.’ Unusually, Steptoe had arranged to see this agent personally, ‘since it must be remembered that Chinese agents are very [?often] in possession of information which they will not, repeat not, commit to paper, and which can only be abstracted from them by careful cross-questioning’. But, he reassured Sinclair, ‘I have taken reasonable precautions to see that this agent does not know my real identity either personally, or in my official capacity’. Another target was Formosa (Taiwan), then a Japanese possession. Steptoe found a British merchant in Amoy (Xiamen) on the south-east China coast who undertook ‘to find suitable man either in Amoy or Formosa for penetration Pescadores’, a strategically important group of fortified islands off Formosa’s west coast where there was a Japanese naval base.
By 1930 Steptoe had become ‘28,000’, the senior SIS representative in China, but something of a one-man band. Not only were there no funds for any extensive organisation, but his own success in developing intelligence work in the late 1920s was accompanied by a conviction of the need for a personal relationship with his Chinese agents, which made it very difficult to find even a temporary replacement for him. In dealing with agents, he told London, ‘especially when they are orientals, it is necessary not only to use tact but above all to exercise considerable ingenuity in putting to them certain questions in order to draw from them more information than they are usually disposed to put on paper’. The need for back-up intensified as his health became more precarious. In 1926 he contracted amoebic dysentery, which necessitated six months’ home leave. In July 1930, after another health scare, Sinclair approved his immediate return to England, while instructing that it was ‘imperative that reports should continue on the Chinese political situation not only in the North, but in the Centre and South China’. Steptoe found a temporary local replacement, though Sinclair still warned that he ‘should not leave for home’ until he was ‘satisfied that the existing Chinese agents will work for him, as it is most important that there is no decrease of information during your absence’. It is clear from this signal that Steptoe was producing intelligence which at SIS headquarters, at least, was considered sufficiently useful for the prospect of a diminished flow to be viewed with concern. The following year Steptoe assisted with the intelligence exploitation of a vast haul of Comintern documents recovered by the Shanghai Municipal Police after the arrest of Hilaire Noulens of the Comintern’s clandestine Far East Bureau in June 1931. Although Noulens’s capture was a worldwide sensation, Vivian in London cautiously concluded that it did ‘little more than administer a temporary and partial check to Communist-inspired centres of revolt or disaffection’.16
During 1931, however, things began to go wrong. Steptoe, always rather a difficult character, became more opinionated and started, dangerously, to luxuriate in his secret service role. One sympathetic observer described him in September 1931 as ‘an interesting and quite pleasant sort of fellow’, but one who had struck him as ‘being afflicted with a weakness that I have noticed in so many other “hush hush” men. He loves to weave a veil of mystery over his doings and whisper strange warnings. No doubt he has to be careful of what he does and says, but this pose is apt to defeat its purpose.’17 He also began to offend British diplomats in China. When he accused the editor of the English-language North China Daily News of being ‘rather too subservient to the wishes of the legation’, the offended editor complained to Sir Miles Lampson (later Lord Killearn), the British minister. In December 1931 Lampson wrote to Vansittart at the Foreign Office suggesting that Steptoe (and hence SIS) was a waste of resources: ‘In these days of strict economy are Steptoe’s elaborate telegraphic reports justified?. . . Much that he sends is to my mind unnecessary. Indeed we could abolish him altogether so far as Legation is concerned without impairment of efficiency of our Service.’ Steptoe defended himself to Sinclair, but, once again, his health was breaking down. ‘It appears’, he wrote alarmingly in January 1932, ‘that the whole of my internal mechanism is functioning at a high rate of speed: no digestion takes place but merely fermentation with all its attendant discomfort.’ He was again advised to take sick leave.
Sinclair was sympathetic, telling Steptoe that Vansittart had ‘just expressed his appreciation of your work during crisis’ (the Japanese invasion of Manchuria in September 1931, during which Steptoe had been posted to Peking ‘on special duty’). But he was also very concerned about the Service’s work in what he rightly regarded as a vitally important theatre. ‘There is no intention of replacing you’, he cabled in February 1932, ‘as long as you can stand the strain, but you must realise the serious danger of your organisation breaking down if there is no-one trained and immediately available to take your place if your health gives way. Understudy will take months to train and must be found, as Far East will be of paramount importance for years.’ The breakdown of relations with the British mission, exacerbated by Steptoe’s overstepping of the mark, and the way in which he had evidently begun to report widely on political matters (admittedly following the pattern set by Denham), highlighted a perennial tension between diplomats and secret intelligence personnel. The former, whose legitimate role was to report on the situation in their host country, were often apprehensive about the presence of the latter, whose rather more specialised function of supplying secret intelligence secretly acquired could, if discovered, at the least cause diplomatic embarrassment and, at the worst, a serious international incident. Some ambassadors and ministers, too, were especially jealous of their position and insisted on a strict control over what was reported home. Even Steptoe recognised that Lampson was ‘absolutely opposed to any expression of opinion on the China situation unless he is the spokesman for that opinion’, and ‘that while he may agree with most of the opinions I hold concerning the situation in this country, he does not approve of my expressing them to you [Sinclair] & the Foreign Office’.
One matter increasingly of concern to SIS in the 1930s was that of communications. In general the Service favoured supplying stations with wireless sets which would provide secure (given that reliable cyphers were employed), autonomous communications, not dependent on the world-wide (albeit British-dominated) cable network. But wireless sets could not be introduced without the permission of the British diplomatic mission in the country concerned. When SIS proposed to install a set in the Shanghai consulate-general in July 1933, Lampson turned it down flat, taking the opportunity again to complain about the Service in general (and presumably Steptoe in particular). They should keep ‘to their own province’ and report ‘on such things as communist activities, Indian movements, drug traffic etc instead of encroaching on the political side’. Sinclair thought Lampson’s attitude ‘extremely unfair’ and that he had ‘completely misunderstood the functions of S.I.S.’, which were ‘primarily to supply information to the Foreign Office and the Fighting Forces which cannot be obtained through official channels’. The British government, he added, was ‘hardly likely to expend some £100,000 a year on obtaining information solely about such comparatively unimportant matters as drug traffic, or even Communism’.
The clear priority which Sinclair gave here to ‘Foreign Office’ and ‘Fighting Force’ information echoed a paper on SIS’s position in the Far East which he had circulated to the three armed service Intelligence Directors in March 1928. The ‘primary Far East intelligence target’, he said, was Japan. After five years of repeated failures and the expenditure of large sums of money, all that had been achieved was the establishment of a skeleton organisation which aimed to give warning of Japanese mobilisation. In April 1923 Cumming had sent out a representative, ‘CT/60’, working under business cover to report on Japanese naval and air matters. After two years he had a network of some dozen contacts located in several locations, including Tokyo, Kobe and Nagasaki, but claimed that he could only proceed very slowly because of ‘the complete mistrust that these people (from the highest to the lowest) have of their fellow beings’. Evidently frustrated by the continued lack of progress (and having spent some £5,000 on the organisation), in April 1928 Sinclair bluntly asked the service intelligence chiefs if they considered further expenditure on the Japanese intelligence target was justified. The Admiralty said ‘No’, the War Office and Air Ministry ‘Yes’, but the Foreign Office decided against any further expenditure, noting that ‘valuable information’ was already provided from time to time from an unspecified ‘Japanese source’, though this was presumably signals intelligence from GC&CS, which had had some success with Japanese codes.
The arrival in May 1933 of a new naval Commander-in-Chief, Far East Station, Admiral Sir Frederic Dreyer, stirred things up. The forceful and ambitious Dreyer had been Deputy Chief of the Naval Staff and had been expected to rise to the very top of the navy, but his career had been cut short by the Invergordon Mutiny of 1931 for which the Staff had to take some collective responsibility, and so he had to be content with the China station. Convinced that war with Japan was inevitable, he complained to Sinclair in September about the attitude of Admiral Gerald Dickens, the Director of Naval Intelligence: ‘How in the name of hell can he expect the powers that be to put their shoulders to the wheel and give me the ships, men, harbour defence gear, etc., of which I am so lamentably short, if he, the D.N.I., tells me that such a war is “REMOTE”!!’ Dreyer formed a good first impression of Steptoe (‘I like him very much and think he is very intelligent’), but the navy’s relations with SIS in the Far East began to sour towards the end of 1933 over the Service’s lack of progress in obtaining intelligence on Japan. Tasked by Sinclair on the matter, Steptoe explained the particular problems presented by the Japanese target. ‘Every effort’, he claimed, had been made ‘to find a person who will take over our Organization in [Japan] as a whole time job’. But ‘to do the work successfully’ it was ‘imperative’ that the person had fluent Japanese and, in any case, the growing climate of xenophobia and suspicion in Japan made any initiative extremely risky: ‘to attempt to build up an organisation now, when every foreigner is so closely watched, and the spy scare so prevalent is, if not an impossible task, an exceedingly delicate one’. A further difficulty was the attitude of the British ambassador, Sir Francis Lindley, who strongly opposed the use of the embassy for SIS work or communications.
The navy’s dissatisfaction with SIS was reinforced by Captain W. E. C. Tait, the Deputy Director of Naval Intelligence, who, in the light of Japan’s increasingly aggressive ‘policy of expansion in the Far East’, had been sent out at the end of 1933 to report on the naval intelligence position. Tait, who proposed that a new regional inter-service intelligence office be established in Hong Kong (which later emerged as the Far East Combined Bureau), was very critical of SIS, especially in relation to the lack of information supplied on Japan, Formosa and the Pescadores. He complained about Steptoe’s apparently cultivated air of secretiveness, which had served to undermine his reputation. ‘Secretiveness and mystery’, wrote Tait, ‘are accepted only as long as people are ready to believe there is something behind them. Once there creeps a doubt into this frame of mind, the very secretive precautions that are taken only increase suspicions, until it is honestly believed by some that there is a lot of mystery about C.X. [Steptoe] and very little else.’ Steptoe, as a result, was ‘to a certain extent discredited’.18
By April 1934 Dreyer, too, had lost confidence in Steptoe. While he conceded that Steptoe was ‘an adept’ and had ‘considerable flair’ for ‘Chinese political forecasts’, what Dreyer wanted was ‘actual facts concerning fortifications’ and ‘warnings of definite preparations for hostilities in Japan or any of her possessions’. Steptoe ‘talks too much’, Dreyer told Sinclair, and his ‘indiscreet and boastful talk has left no doubt on people’s minds generally as to what his real functions are. Perhaps’, he added facetiously, ‘Steptoe is only a dummy and you have a real S.I.S. working quite independent of him.’ Dreyer suggested that, so well known was Steptoe’s real role, his ‘presence alone would be quite enough to destroy any S.I.S. organisation which may be built up, just when we need it most. Or perhaps our little yellow friends will push him into the Whangpoa [Huangpu River, which runs through Shanghai] without further ado when their time comes.’
Having become aware of Dreyer’s criticisms, Steptoe offered to resign, but Sinclair loyally backed his man. ‘You still retain my complete confidence and the fact that you do not agree with senior officials belonging to other departments is no reason for your resigning from the S.I.S. If this were so,’ he continued, ‘I should have to resign from my position several times every week.’ All the same, the Service’s failure to obtain ‘Japanese Naval and Military information’ was recognised in Broadway and, on Vivian’s suggestion, in June 1934 Sinclair decided to appoint a new representative to be based at Hong Kong to take over the Japanese target from Steptoe.
The failure of SIS and Steptoe in Japan in the early 1930s stemmed from a combination of scarce resources, which (among other reasons) obliged SIS to persevere with 28000’s one-man-band operation; progressively increasing demands for information, especially from the armed services in the face of a growing threat from Japan; and personality problems, particularly as they affected the unfortunate and chronically unwell Steptoe. Even if he had been better placed in his relationships with his military, naval and diplomatic colleagues, it is doubtful that Steptoe’s progress in gathering intelligence on Japan would have been any better. It was unrealistic to expect him, however well versed in Communism and Chinese politics, to take on a very different target and one that lay at geographical arm’s length. Sinclair was right when he wrote to Dreyer in July 1934: ‘I do not think that, even now, you or the G.O.C. realise the extreme difficulty of obtaining secret information about Japan or Formosa.’ Sinclair was sending out a fresh individual ‘to take over the Naval and Military Intelligence’, but he also remarked ‘that there appears to be no prospect whatsoever of obtaining any additional funds with which to finance this undertaking, which makes the task even more difficult than it otherwise would be’. A note by a source who had assisted Steptoe since 1928 addressed the fighting services’ criticism about the lack of Japanese information. ‘If they require plans stolen,’ he wrote in August 1934, ‘serving officers bribed etc (which unquestionably is the most efficient method of obtaining what they require) they must provide far greater sums of money than are at present available.’
SIS’s new man in Hong Kong was Charles Drage, a retired naval lieutenant commander who had served in China in 1923-6 aboard the sloop HMS Bluebell. Working under business cover, he had to begin from scratch as the embryonic network established in the 1920s had completely collapsed by the early 1930s. Drage was assisted by a South African who had been recruited by the Berlin station in 1923 and had later served in Mukden (Shenyang). This assistant had a flair for recruiting what was called the ‘mechanised type of agent’: Chinese seamen who visited ports in Japan and Formosa. At the beginning of 1936 it was reported in Broadway that four ports in Japan (Yokosuka, Kobe, Asaka and Hiroshima) and two in Formosa (Keelung and Taiku) were being watched. Formosa was an easier nut to crack than Japan itself and in October 1938 Drage boasted that he controlled ‘the only secret intelligence organization in Formosa’ and therefore had ‘a monopoly of reliable information from that Island’. But the return was not very great. In April 1938 after Naval Intelligence had complained about the paucity of intelligence from the Far East, Sinclair was supplied with a list of seventy-two SIS agents and contacts working in the region, of whom twenty-nine were thought ‘likely to obtain Japanese Naval information’. Though this represented a fair spread of assets, a lack of ability to communicate rapidly severely limited the timeliness with which any information could be delivered. In 1935 SIS agents in Osaka and Tokyo had experimented with sending coded messages through regular telegraph channels and a courier service had been tried from Kobe. It was not altogether reliable and in September 1936 Sinclair, when reaffirming SIS’s responsibility ‘for maintaining a Coast Watching Organisation, in order to give advance information of troop movements or possible mobilisation’, went on to grumble to Drage that ‘the breakdowns which are frequently occurring with this organisation coupled with misleading reports such as you have furnished on this matter, give rise to the gravest anxiety here’. There were losses of agents, too. Early in 1937 an agent was arrested and tortured ‘shortly after handing in coded troop movement telegram for Hong Kong’. Fortunately a fellow sub-agent secretly engaged a lawyer and at a cost of $500 secured his colleague’s release.
By the late 1930s SIS was still having difficulties in meeting the increasing requirements of the armed services, though there were occasional successes, such as Sinclair’s report in January 1938, ‘from an unimpeachable source’ (which may have been signals intelligence), ‘that the new Japanese battleships are to be of 46,000 tons (gross) with an armament of twelve 16” guns’. Hankey was sufficiently impressed to congratulate him warmly ‘on obtaining such an important and remarkable piece of information’. Drage, apparently producing some good information from Formosa, was much less successful with Japan. Steptoe continued to report primarily on political matters from Shanghai, while Frank Liot Hill, appointed head of station in Peking in the early 1930s with an ambitious brief to penetrate eastern Siberia, Manchuria and Korea, as well as north China, does not seem to have been producing very much. Hill was still endeavouring to establish viable networks in late 1939, and his list of agents shows that some of his reporting was based on one or two Chinese sources strategically placed at the important Fengtai railway junction just south-west of Peking. In theory they should have been able to provide detailed intelligence on Japanese order of battle as troop trains moved through the junction from the coast and onwards to central China.
The Jonny Case
The case of Johann (‘Jonny’) Heinrich de Graff, a German Communist and agent of both the Comintern and the Fourth Department of the Red Army Staff (later the GRU, Soviet Military Intelligence), was one of the most striking SIS successes of the 1930s. It provided, through a rare penetration lasting for five years, a unique insight into the working practices and personalities of the Comintern. It enabled SIS to forestall a planned Communist revolution in Brazil and was thus an early case of clandestine political action by the Service in Britain’s interests. Geographically, the case spanned Moscow, the United Kingdom, Germany, China and Latin America, and it provided a fuller understanding than hitherto of the extent of Soviet policies and methods of interference in other countries in pursuit of both its political revolutionary and its paramilitary aims. The information from de Graff was confirmed and complemented by two additional sources: another early Soviet defector, an NKVD agent who also started talking to SIS in 1933; and ‘Mask’ intercept material from Comintern communications provided by GC&CS.19
In modern parlance de Graff was a walk-in. The first contact with him was made by Frank Foley, SIS head of station in Berlin since 1923. On 13 February 1933 Foley reported to Sinclair that he was in touch with one Ludwig Dinkelmeyer (de Graff’s alias), ‘German born May 11th ’94’, a ‘Member of Executive of Communist International and Secretary General of illegal Red Front Fighters Union here’. The German, who said he had visited England twice in 1931-2 ‘to report to Moscow on Communist Party of Great Britain’, had, Foley’s signal continued, ‘offered to become agent for me. He states he can give me complete information about Communist propaganda amongst British Armed Forces and continue to keep me informed of most Communist work in England arranged from here.’ De Graff wanted money: 2,000 Reichsmarks down and 500 per month (about £144 and £36; in modern prices something over £7,300 and £1,800 respectively). ‘Consider this most important contact I have yet made and convinced his genuineness. May I continue negotiations?’ asked Foley.
The ‘Jonny Case’, as it became known in SIS, entered the Service folklore, and the story - certainly a good one - undoubtedly grew with the telling. Years afterwards, Vivian said that, after he had been urgently summoned to his office by Sinclair and shown Foley’s telegram, the Chief said he was immediately to go to Berlin. When Vivian protested that he had no overnight luggage, Sinclair said, ‘I’ve done all that, your wife will be up at 11.00 with a suitcase packed for a week and here are your reservations. ’ The archive record more prosaically shows, rather, that Sinclair sent a signal to Berlin summoning Foley home at once. While Foley, and others, helped run the agent, Vivian became Jonny’s principal case-officer, writing up the reports and providing information against which Jonny’s reporting was checked. But Sinclair also took an extremely close interest, so much so that the matter was described on Head Office documents as ‘a CSS only case’.
Jonny’s first report (delivered in March 1933) was a twenty-eight-page description of ‘Communist disintegration work among H.M. [His Majesty’s] Forces’, which Sinclair (after checking corroborative details with MI5) judged to be ‘genuine and of the utmost value’. Communication with Jonny was problematic. SIS had no way of contacting him and, after Vivian had provided him with a ‘dead letterbox’ address, it was left that he could meet his SIS handlers only (as Vivian recalled) ‘if he was able to give us an address where he could meet him in comparative safety’. In March 1933 Nazi pressure forced Jonny out of Germany and Foley had to meet him in Prague. Three months later Jonny was sent under business cover as a ‘soybean merchant’ to review a clandestine Soviet mission in Manchuria (then occupied by the Japanese and known as Manchukuo). On his return to Europe in December he was debriefed by Vivian in Copenhagen over a period of several days. Jonny’s next Comintern assignment was to Shanghai in February-July 1934 to revive their Far East organisation, which had been disrupted in the wake of Hilaire Noulens’s arrest. There he was handled by Harry Steptoe, who transmitted reports back to London on the weakness of the Comintern presence in China and stating, reassuringly, that the subversive threat to British forces stationed in Shanghai was negligible, a conclusion the War Office described as ‘quite interesting’. A later report from Jonny after his return from Shanghai in autumn 1934 stressed that the connection between Comintern activities and Soviet foreign policy was so close that it was impossible to imagine them apart from one another.
The main value of Jonny’s reporting, however, lay not so much in the political aspects of Comintern activity as in what Vivian called the ‘technical aspects’: the aliases, passport details, physical descriptions and travel plans of other Comintern agents that Jonny met, about which Vivian commented extensively. In this respect the damage done to the Comintern by Jonny was considerable, and he was right to be mindful of his own security and to remind SIS, as he did, of evidence of Communist penetration of Scotland Yard.20 What was Jonny’s motivation? Although financial considerations remained important, by late 1934 Vivian thought there were additional personal factors. Foley, for example, was able to help with the welfare of his wife while Jonny was away. What is clear is that Jonny was brilliantly exploited and handled by Foley, Vivian and even the much maligned Steptoe.
The most dramatic part of the Jonny case was the role he played in disrupting a planned Communist revolution in South America. Briefed in Paris in November 1934, Vivian learned that Jonny was being sent to Brazil. With special operations and explosives training, his task was to assist a Comintern operation in support of Luís Carlos Prestes’s revolutionary movement which aimed to topple the right-wing government of President Getúlio Vargas, who had seized power in 1930. There was no SIS station in Brazil, and thus no one to handle Jonny, or to provide communications, or to deal with the embassy in a situation which might develop disastrously for British interests - at a time, moreover, when the United Kingdom had very substantial investments there. Vivian, therefore, with a warm recommendation provided by Vansittart at the Foreign Office to the ambassador, Sir William Seeds, went out himself to make the appropriate arrangements. By the time he arrived in Rio de Janeiro on 11 February 1935, Vivian had an address for Jonny but no rendezvous. In his version of the story he found the house and rang the doorbell: ‘a negro servant opened the door, screamed and shut it in my face’. Vivian rang again, but there was no answer, so he went round the side and found an open, though high, window. ‘As I scrambled in and fell down the other side, an enormous arm protruded from a curtain levelling a 500 automatic Colt [sic] at me.’ The arm belonged to Jonny, who had expected less friendly visitors.
Matters improved thereafter, and Vivian had several meetings with Jonny, though one, on Copacabana beach, left him badly sunburned. The ambassador, meanwhile, was typically difficult. Seeds told Vivian that he ‘would do a great deal for Sir Robert Vansittart’, but would not let any of his officials ‘act as an intermediary’. Nor would he ‘take the part that Vansittart thinks I can of warning the Brazilian Government if and when a revolution is to take place. They will think we have got spies in the country and I won’t do it.’ Before he returned to London, Vivian found a local British businessman to provide a link with Jonny, who over the next few months reported regularly on the progress of the conspiracy. By June things appeared to be coming to a head. Sinclair told Vansittart on 12 June that the plot had ‘made alarming progress’. The ‘first act’ of the revolutionary government, he warned, ‘would be to take possession of all the British undertakings in Brazil, and to deport all British subjects connected with them’. He told the Director of Naval Intelligence that it was ‘almost certain that a revolution will break out before the end of the year, which may necessitate naval action being taken in order to protect British interests’. Faced with this intelligence, Vansittart decided to instruct Seeds to warn President Vargas, which he did on 20 June, though, ‘as zero hour of plot is still distant’, Seeds was ‘anxious that Brazilian authorities should not take precipitate action’.
Vargas (who had been ‘grateful and interested’ in response to Seeds’s warning) applied some counter-measures, but the Comintern conspirators remained at large, and a distinct possibility remained that some leftist army units might mutiny, which indeed occurred four months later. Alerted by Jonny on the evening of 25 November that things would happen in Rio within twenty-four to forty-eight hours, SIS’s local businessman contact in turn warned the ambassador. At 10 p.m. on 26 November, the businessman was called to a rendezvous by Jonny and told that Prestes and his military friends were going to strike that night. At 3 a.m. the SIS man told the ambassador, but, more effectively, he also warned the general manager of the Canadian-owned Brazilian Traction, Light and Power Company (colloquially known as the Light), the principal Rio utilities company, who was able to hinder and frustrate the revolutionaries by cutting off the ‘power supply from revolting barracks, who were thus without light and could not operate their radio stations’. After some heavy fighting, the rising collapsed. Further information from Jonny led the Brazilian authorities to the chief Comintern representative and a mass of incriminating documents, which subsequently enabled the destruction of the revolutionaries’ South American apparatus.
During the general round-up of conspirators in January 1936 Jonny himself was arrested, but, with the businessman’s help, was released and escaped to the Argentine. He was eventually recalled to Moscow in December and, despite Vivian’s misgivings, insisted on going. He survived a Comintern inquiry and was sent back to Brazil to assemble a team intended to take over a Communist military network in Japan. After his wireless operator failed to turn up, this appears to have been abandoned and Jonny stayed on, enjoying the (by now very considerable) financial fruits of his espionage endeavours. From 1938 SIS had a representative in the region (based in Montevideo) who employed Jonny ‘as a sort of Nazi agent-provocateur’, a role he was not very good at playing. As Vivian recalled, ‘after the very sensational work that he had done for years it was a small meat’. In November 1939 he was arrested for ‘espionage’ and very roughly handled by the Brazilian police. After the businessman managed to alert London, the Foreign Office protested to the Brazilian ambassador, and Jonny was released. He was brought back to Britain and later settled in Canada, where some years later he told stories of his work as a British intelligence agent to the press.21
Opportunities and difficulties in Europe
Under Harold Gibson, head of station in Bucharest from December 1922, Romania was regarded as an important location for work on the Soviet Union. Gibson had originally gone to Romania as a correspondent for the Morning Post, and from 1924, assisted by his younger brother Archie (who also operated under journalistic cover) and drawing extensively on White Russian contacts from his time in Turkey (including his agent HV/109), he assembled a sizeable network of sources. During the 1920s he ran some seventy individuals, though not all necessarily at the same time, on the Romanian and Soviet sides of the frontier. One of Gibson’s groups was run by a clerk (‘313’) in the Sevastopol naval base. He worked for a Romanian secret service officer who was also passing information to Gibson. Agent 313 had sub-sources who provided an ‘abundance of useful material’ on the Soviet Black Sea Fleet, before one of them betrayed him in 1930. There were also Red Army sources in the Ukraine and as far afield as Irkutsk in Central Asia. In 1930, the heads of both the Naval and Military Sections at Broadway - Russell and Menzies respectively - commented favourably on the reports produced (none of which has survived). Menzies declared that they ‘have proved accurate time after time’. After Harold Gibson’s departure for Riga in March 1931 (taking HV/109 with him), his successor, Major Montagu ‘Monty’ Chidson, and Archie Gibson established new networks, based in part on Ukrainian nationalist groups with sources in Soviet military and industrial organisations. Although SIS forged quite close relations with both the Romanian security police and military intelligence - with the latter greatly facilitating the movement of agents and couriers across the Romanian-Soviet frontier - Archie Gibson claimed afterwards that (unlike the position in the Baltic states) nearly all of the Bucharest station’s Soviet reports came from its own sources.
Maintaining independent lines of information like this was valued in London. In January 1936 Menzies’s deputy at Head Office insisted that it was ‘most important’ that SIS’s organisation in Romania ‘should be maintained in one form or another; otherwise we shall be entirely dependent on the S.I.S. posts in the Baltic states, which are themselves practically dependent on information provided by the General Staffs of those states’. But by this stage SIS’s Romanian operations had been fatally compromised by the Flemmer affair. In October 1935 the head of the Romanian Military Secret Service, Major Mihail Muruzov, informed Chidson that he had arrested two brothers, Mikhail and Alexander Flemmer, who had been working both for the Romanians and for Harold Gibson’s agent 109 (though not for SIS). Muruzov had discovered that the Flemmers were OGPU agents. He said that any public trial would be most undesirable - especially as it might reveal the close association between the Romanians and SIS - and that he planned to shoot the two men, having first interrogated them. Chidson strongly agreed with the need for secrecy and accepted Muruzov’s offer to question the Flemmers himself in order to ascertain the extent to which SIS’s organisation in Romania and its courier lines in the Soviet Union might have been compromised. After interrogating the unfortunate brothers and 109, Muruzov concluded that the latter was innocent, but that the OGPU had thoroughly penetrated both the Romanian and SIS anti-Soviet networks. Chidson, while conceding the possibility of some damage, refused to panic and wanted to carry on as usual until he had firm evidence to the contrary. As it turned out, this was a serious error of judgment.
Muruzov, meanwhile, had arrested one of 109’s sub-agents, ‘109/18’, and satisfied himself that he, too, was working for the Soviets, though he admitted to Chidson in December 1935 that it ‘was difficult to pin anything definite on him’. Nevertheless, as with the Flemmers, he proposed arranging his ‘quiet disposal’. Chidson, vehemently opposed to the killing of one of his own agents, got Muruzov to release him, but (presumably to be on the safe side) had him dismissed from SIS employment. The affair clearly rocked the Service. Reviewing the situation at Broadway in January 1936, some officers felt that the Bucharest station could no longer be relied upon for Soviet information. Menzies was ‘profoundly disturbed’ about the extent of OGPU penetration in the Romanian networks, but he also maintained that SIS could not prove 109/18’s guilt. His impression, indeed, was that ‘much valuable material’ had come ‘from this station during the regime of the present 14000 [Chidson]’. The War Office had cast no serious doubt upon the reliability of SIS’s sources and (straining perhaps to make the best of the situation) he felt it ‘almost impossible that the O.G.P.U. should have fed us with genuine material for years’. In Bucharest Chidson was equally reluctant to accept that his entire operation was compromised. During February, in an effort to contact two of 313’s sources and resume work, he sent a trusted agent into the Soviet Union. Neither he nor 313’s contacts were heard of again. Presumed betrayed to the OGPU through 109/18, by April 1936 they had been reported as casualties. This was, in effect, the end of SIS’s Soviet operations in Romania.
Chidson was subsequently transferred to The Hague, and in August 1936 Archie Gibson took over as head of station at Bucharest. Shortly afterwards he reported Muruzov’s annoyance at SIS’s attempts to try to sustain an organisation in Romania, which the latter correctly presumed could not operate against the Soviet Union and which he therefore assumed was either working or preparing to work against Romania. Muruzov, wrote Gibson, had in fact been very helpful to the British, ‘putting his own men at our disposal’, providing facilities for frontier-crossing, ‘to say nothing of information which came from his sources and many other favours. From us,’ reported Gibson, ‘he has had a gold cigarette case, an occasional gift of cigarettes and, from time to time, some information.’ But the Bucharest station (and the Anglo-Romanian intelligence relationship as a whole) never recovered from the Flemmer debacle. The whole affair amply demonstrated a recurring dilemma whereby continuing and apparently well-founded intelligence operations are suddenly undermined by the discovery of possible treachery. Perhaps hoping against hope and clearly reluctant to abandon years of careful intelligence work, neither Chidson nor Menzies was initially willing to concede that SIS’s Romania operations might have been thoroughly compromised. While in one sense (and in marked contrast to Muruzov’s much more hard-nosed and ruthless attitude) this embodied a commendable faith in their existing agents, if not also (and less admirably) in their own good judgment, in the end it appears to have had fatal consequences for the network Chidson tried to revive in the spring of 1936. Chidson’s perhaps overly trusting attitude was also to have disastrous consequences at his next posting, where Folkert van Koutrik, the assistant to SIS’s head agent at The Hague station, was turned by the Abwehr and betrayed leading German agents of both SIS and MI5.
In contrast to Romania, which had taken the Allied side during the First World War, Bulgaria had been an enemy power, and between the wars the SIS station in Sofia had little or no official contact with the local security services. Here, indeed, the generally applied prohibition on working against the country of residence did not apply and at the start the station concentrated on looking out for breaches of the Neuilly peace treaty, which the Bulgarians had signed in November 1919. Subversive activities of Communists and the Internal Macedonian Revolutionary Organisation (a violent extreme nationalist group with ambitions to add Greek and Yugoslav territory to Bulgaria) were important targets. Throughout the interwar years Sofia set its net wider and began to gather intelligence on Yugoslavia, Romania, Turkey, Italy, the Soviet Union and, as the Second World War approached, Germany. The head agent of one SIS network, a journalist with good contacts among top Yugoslav officials, supplied political and economic information on Yugoslavia, Italy and Germany. Another head agent, based in Belgrade, had a source who supplied weekly lists of goods carried on Danube barges between Yugoslavia, Germany, Romania and Bulgaria. A third, a White Russian living in Sofia, ran sub-groups which supplied Bulgarian military material information on arms shipments to Bulgaria and illegal immigrants to Palestine. One of his agents, who had sources in the Turkish army and naval dockyards, was described as having ‘pro-Communist leanings’, being ‘against the Kemalist regime’ and also needing ‘money for his mistress’.
Between the wars the Sofia station (‘11000’) had an unusually high turnover of staff. From 1920 until the war there were six heads of station, all of whom operated under Passport Control cover. They included an ex-actor, three sailors, including Lieutenant Commander Leonard Hamilton Stokes (April 1924 to December 1929) and Engineer Captain Charles Limpenny (1933 to July 1935), one soldier, Captain (later Major) William Mackinnon Gray (August 1935 to September 1937), and an airman. The most intriguing of these appointments was that of the fifty-one-year-old Limpenny, a submariner and presumably longstanding acquaintance of Sinclair’s. One of a talented and creative West Country family - the concert pianist Moura Lympany was his niece - he had been appointed an ADC to the King in December 1932, and on his formal retirement from the Royal Navy in April 1933 was promoted rear admiral.22 What the diplomatic corps in a small Balkan capital (let alone the local security authorities) made of such a distinguished personage occupying the comparatively modest position of Passport Control Officer is anyone’s guess.
Limpenny evidently made a success of his posting because, after brief sojourns in Athens and Stockholm, he was brought back to Broadway in December 1936 to take over as head of the Economic Section, where he stayed until June 1946. The same could not be said of his successor, Mackinnon Gray, who had been introduced to SIS in 1932, and had spent some time as a ‘learner’ at Istanbul and an assistant at Jerusalem before his appointment to Sofia. Gray had a problematic personal life. In 1936 his wife sued for divorce, and there was concern that his SIS appointment might get mentioned in the proceedings. Situations like this, when crises in the private lives of officers (and, indeed, agents) might jeopardise cover or security generally, were not uncommon, though in this instance - and to his evident relief - Gray’s real role in Sofia remained secret.23 During 1937, however, concerns were raised about financial irregularities in the payment of an agent. Sinclair ordered an inspection of the Sofia office and sent Limpenny out to investigate the matter. The results of these enquiries revealed Gray to have been guilty of carelessness and slipshod record-keeping rather than peculation, but they also throw revealing light on the mechanisms by which a local SIS station ran its finances. While there were strict rules that any local expenditure had to be approved from London (which Gray had failed to follow), funds appear routinely to have been transferred from Head Office to the head of station’s personal bank account and from there to a second offshore account. In Sofia local currency for paying agents was acquired clandestinely on the ‘Black Bourse’ by the wife of a junior member of the Passport Office staff.
In September 1937 Gray simply abandoned his post, returned home without authorisation and resigned from the Service. In Sofia Limpenny made good some bounced cheques and reported that there was no cause for anxiety regarding the functioning of the local SIS organisation. Thus the SIS senior management might have hoped that the Mackinnon Gray case was closed. In 1941, however, after he murdered his second wife and attempted suicide in Chilcompton, near Bath in Somerset, there was a fresh alarm that his secret service work might be exposed. This time Valentine Vivian was despatched for damage limitation and successfully arranged with the Somerset Chief Constable to leave any reference to Passport Control out of the legal proceedings. Gray, who recovered from his self-inflicted wound, was sentenced to death at Winchester in July 1941, but was reprieved and subsequently jailed for life.24 He evidently had a difficult and troubled private life. There is no way of estimating the extent to which (if at all) the pressures of secret intelligence work may have exacerbated his personal problems, though one may reasonably suppose that the strains of having to maintain a secret professional existence may not have helped. What is clear from this case, however, is that domestic crises, of whatever sort, had an unsettling potential to go public and endanger operations. Thus the Service took these kinds of problems very seriously indeed.
Even though Sinclair and Pay Sykes (in charge of SIS finance) tried to keep close tabs on Service expenditure, the realities of funding intelligence work at station level, regularly requiring untraceable payments in cash or precious goods of various sorts, inevitably meant that officers and agents, the former of whom were not particularly well paid, could be faced with considerable temptations. Not everyone was able to resist. In the early 1920s one of the Riga station’s staff decamped with the Passport Control Officer’s seal, a month’s advance salary and £500 (equivalent to £20,000 today) in diamonds and gold, stolen from a dealer who had smuggled it out of the Soviet Union. During the following decade the sharp increase in the numbers of Jewish refugees seeking to travel to Palestine underpinned the development of a black market in the necessary British visas. During the 1930s there were three cases of visa-trafficking in the Warsaw Passport Control Office. The most serious, in July 1936, threatened to expose SIS’s role in the Passport Control organisation and the British ambassador wanted to prevent Major Shelley, the Passport Control Officer (and SIS head of station in Warsaw), and other diplomats from having to appear in court lest ‘they might be asked questions about the internal affairs of the Passport Control Office which it would be inexpedient for them to answer’.25
The worst financial scandal of the 1930s occurred in the Netherlands. Here Major Ernest Dalton, a First World War MI1(c) veteran, had been head of station since April 1924. But his time at The Hague was overshadowed by the revelation in the mid-1930s that he had got into financial difficulties and embezzled several thousand pounds’ worth of visa fees.26 Dalton’s wife was seriously ill and he himself suffered from ‘blood poisoning’. Efforts to recoup the money through gambling having failed, he collapsed under the strain and committed suicide in July 1936. ‘I have got myself into such a mess’, he wrote, ‘that this is the only way out.’ Sent out to deal with the situation, Rex Howard vetoed a suggestion by one of Dalton’s assistants that his death should be put down to ‘heart failure’. The ambassador sensibly argued that ‘it was better to be quite frank in the first instance, as otherwise it would be thought that there was something to hide’. Traces of the Dalton tragedy remained at the station. Officers subsequently posted to The Hague were solemnly taken to what had been his bedroom and shown the badly repaired mark in the wall caused by the bullet with which he had killed himself.
Further investigation and suspicions of connivance among the station staff in Dalton’s ‘defalcations’ led to some dismissals and the Dutch-speaking Monty Chidson (his wife was Dutch), urgently summoned from Bucharest, became head of station with instructions ‘to take such steps as he may think fit to ensure that all matters in the office are systematised & regularised as soon as possible’. That Chidson’s attitude to security was not all that might have been desired became evident to Howard at dinner one evening. Chidson wanted to introduce his Rotterdam-resident brother-in-law as he ‘had good connections’ and ‘might be useful’. Reporting on the occasion, Howard wrote that ‘about half way through dinner, I remarked to the b-in-law that I supposed he knew from Chidson that the latter had now become P.C.O.’. The brother-in-law replied ‘that he did not know there was such a post & understood Chidson was in the Secret Service & that was all he knew he did’. He went on to talk ‘quite openly & loudly to Chidson about the Secret Service & his connection with it. The people at the next table’, recorded Howard, ‘were obviously taking a great interest in the conversation & I abruptly changed it.’
As in the Low Countries, during the 1930s SIS’s operations in Scandinavia gradually moved away from the Soviet target to focus on Germany. Here, too, local sensitivities about their neutral status, and well-founded fears of antagonising the Germans, influenced the extent of cooperation with local security and intelligence authorities. In Stockholm, sources in Swedish Military Intelligence were prepared to share information about suspected Communist activities in the late 1920s and early 1930s, but were markedly less forthcoming about Germany. Lieutenant-Commander John Martin, who became head of station in October 1937, made sustained efforts to recruit locally based Britons and Swedes whose business connections with Germany offered the possibility of intelligence which might survive the outbreak of war. A Norwegian clerk in a shipping company in Oslo, for example, was taken on in 1939 to report on arms shipments. He in turn recruited a sub-source to visit German ports, but no evidence survives that they produced intelligence of any great value.
The situation in Denmark was rather similar. In January 1927 London had given the longstanding head of station in Copenhagen (in post since 1920) four main objectives in order of priority: Germany, Russia, ‘Norwegian & Swedish air’ and ‘Arms Traffic’. He reported considerable endeavours, but not much substantive progress in any of these areas. He was followed by a retired RAF officer, appointed to Copenhagen in August 1928 after less than a year in the Service, including three months as a learner on the Athens station. His departure from the Service in mid-1936 reflects the continuing hand-to-mouth financial basis on which the interwar organisation was run, depending to a great extent on representatives having a private income to supplement the hairshirt salaries SIS for the most part offered. He resigned after the brusque rejection of a plaintive appeal to the Chief that, after twenty-two years in government service and at the age of forty-six with growing family responsibilities, he might be given established status, perhaps in the RAF, with some pension entitlement.
Both the Copenhagen head of station from 1936 until early 1940 and his predecessor put a good deal of effort into acquiring assets in Schleswig-Holstein, where they hoped ethnic Danes resident on the German side of the frontier could be recruited as agents. In January 1938 Sinclair observed that ‘by its geographical position’ Copenhagen was ‘very favourably situated’ for obtaining information on Germany. The head of station was, therefore, to concentrate almost exclusively on that target, to ‘make every endeavour’ to acquire information on the German armed forces, and especially to try to recruit a naval officer. But Copenhagen was a one-man station, and, with no operational assistants (though he did have two secretaries), this was much easier instructed than implemented. Although there were Danes in Germany who could cross the frontier and report low-level eyewitness information on the German forces in the area, contact with them was necessarily cautious; the Germans were alert, their counter-espionage measures active, and they could easily close the border, as they did at the time of the Munich crisis. No wireless sets were yet available to overcome such a simple but effective break in contact. None of the station’s resident agents had sufficient access to contemplate recruiting a German officer, naval or not. Danish businessmen with reasons to travel to Germany tended to be understandably unwilling to jeopardise their interests by taking more than minimal risks. The traditional Danish neutrality was widely felt to be good for Denmark and vulnerable to German displeasure. Under these circumstances, it was quite creditable that in 1934-7 the station delivered between four and five hundred reports a year, even if most of them were rather run of the mill. But London inevitably wanted more and better information, especially on economic and political matters. In late 1938 Broadway sent Copenhagen the Circulating sections’ collective comments on the station’s recent reporting: no political information of value; ‘slight improvement’ in air reporting; military reporting ‘disappointing’; the ‘location of the German fleet in German harbours efficiently carried out’. It was hardly a ringing endorsement of the station’s efforts.
During most of the interwar years, SIS had only a minimal operation in Norway. Between October 1924 and September 1938 there was no full-time representative. For much of this time, indeed, Biffy Dunderdale, while based in Paris, was also nominally head of the Oslo station, though the SIS representative in Stockholm, John Martin, also had a watching brief. In fact, for much of this period quite a lot of local work was handled, and a very productive liaison maintained with the Norwegian authorities, by one of the unsung heroes of the Service, an extremely capable multilingual female secretary. In the late 1930s the likely importance of Norway (especially on the naval side) brought a reinforcement of the station, though this was not initially without difficulties. Late in 1938 Sinclair decided to appoint a permanent representative to Oslo ‘to obtain information on all types of German activities, particularly naval movements, and the reactions of his neighbouring Scandinavian countries thereto’. The individual selected, Lieutenant Commander Joseph Newill, a retired sailor and fluent Norwegian-speaker, was appointed from 1 January 1939. Although he seemed ideally suited for the job, on 8 May he told Rex Howard at Head Office that he was ‘in a constant state of perplexity’, had made no progress since his arrival and wondered ‘whether I shall ever do so. I doubt’, he added, ‘whether I have the natural guile so essential for this work!’ Later the same month John Martin from Stockholm delivered a savagely damning judgment on Newill, whom he rated ‘a complete flop’, ‘never likely to do any good or be of any use’ and ‘should be got rid of ’. Newill had told him that the job was not what he had thought it would be and was ‘far more hard work’ than he was prepared to do. ‘I am 52,’ he had told Martin, ‘and I am not going to work myself to death at my time of life.’ He had, in any case, no need to work, since he had a private income far in excess of his salary (at the time, of course, normally a positive attribute for work in SIS). Being plunged in on his own had clearly unsettled Newill, but he slowly got to grips with the work. When Sinclair sent Frank Foley to Oslo in August 1939 to review the situation there he reported that Newill was ‘a good man in the right place’. And after Foley had been installed at Oslo to oversee all the Scandinavian stations, he was able to manage Newill closely and professionally, so that by October 1939 he was functioning satisfactorily and confidently, especially on the naval reporting and maritime front.
In the 1930s independent Ireland became an intelligence target for SIS. After the separatist nationalist, Eamon de Valera, came to power at the head of a Fianna Fáil government in March 1932, he set about dismantling the remaining links with the British empire, and in the new 1937 constitution abolished the role of the British monarch as head of state. Hesitating to declare Eire a republic, however, he devised the concept of Ireland’s ‘external association’ with the British Commonwealth. The government in London was anxious to keep tabs on Irish political opinion and MI5 was asked to set up an organisation to report on Eire. Kell, recalled Vivian some years later, ‘refused point blank on the grounds that such a project was too dangerous and that, moreover, he had no suitable personnel or means to carry it out’. The Prime Minister then sent for Sinclair, told him that now Eire had to be regarded as a ‘foreign country’, SIS would have ‘to do what Kell had refused to do’. Sinclair deputed Vivian to provide ‘a periodical survey of the situation in Ireland, with particular reference to extremist republican opinion, the I.R.A. and German penetration’. This task, reflected Vivian, was ‘not within our regular range of duties’, and was ‘undertaken reluctantly as no other Department was willing to take the risk’. Because of the ‘extraordinary delicacy and dangers’ of the operation, he was instructed to report only to Sinclair and not to mention the matter to anyone inside or outside the Service, unless directly involved. Essentially, this was not an intelligence operation, and there was no real network of agents, but, in the words of an MI5 review of its wartime activities in Ireland, it constituted a ‘very restricted information service’, in which Vivian, using Southern Irish contacts principally secured through the Royal Ulster Constabulary, provided ‘a limited cross section of private opinion of current events of political or public interest in Eire’.27 After the outbreak of war, when Ireland remained neutral, there were heightened worries, especially in the Admiralty, that German submarines might re-provision in Ireland or the country be used for intelligence activities against the United Kingdom. There was a modest expansion of SIS’s work but, after MI5 finally engaged with Eire, Britain’s security requirements in this area were met by an eventually close liaison between the Security Service and the Irish authorities.28