Who Commanded History? Sir John Colville,Churchillian Networks,and the ‘Castlerosse Affair’

This article has its origin in a remarkable finding we made at the Churchill Archives Centre, Cambridge, UK. ¹ It was a cassette recording made in 1985 of a discussion between Sir John Colville, formerly Private Secretary to Winston Churchill, and two of the Centre’s staff: Correlli Barnett, then Keeper of the Archives, and another archivist (now deceased) Pat Ackerman. Most of the tape was resolutely unsensational, consisting of Colville’s comments and advice on which other Churchillians might be invited down to Cambridge to share their memories. This probably explains why apparently no-one, if they had ever listened to the cassette at all, had ever made it as far as the second side, towards the end of which he mentioned Denis Kelly, who had been one of Churchill’s literary assistants in his later years. Suddenly switching tone, Colville – fully aware that he was being recorded – revealed that Kelly had done something that he, Colville, regarded as awful. It was, he said, rather scandalous and therefore not to be revealed for a a long period of years. (The archivists never disclosed it.) Kelly had discovered a number of love letters in Churchill’s papers from a woman named Lady Castlerosse, with whom, Colville claimed, Churchill had definitely had a short affair while staying on the Riviera many years previously.

As the Colville family, who hold the copyright in the tape, have not granted us permission to quote from it, we are obliged to give a detailed paraphrase instead. Colville explained that he, Colville, had been having tea with Winston and Clementine Churchill at Chartwell, their house in Kent. Then Kelly appeared and presented Clementine with the letters, saying that he wondered who they were from. Clementine read the correspondence and went pale. According to Colville, she was frightfully anxious about it for months. She used to tell him that she had never previously thought that Winston had been unfaithful to her. Colville, in response, tried to play it down, observing that plenty of generally respectable husbands occasionally strayed on a summer's evening; but in spite of his attempts to reassure her she continued to worry about it terribly. Colville blamed this on Kelly's foolish behaviour in showing her the letters. ²

This tale, which has remained hidden until now, was something in the nature of a bombshell. Although Colville took pains to emphasize that the alleged affair had been a solitary lapse on Churchill’s part, his claim poses a challenge to the historical orthodoxy that it is ‘impossible to imagine’ him engaging in marital infidelity. ³

Was Colville’s allegation about Churchill and Lady Castlerosse correct, however? We examine the evidence briefly below. But that is ancillary to our primary purpose in this article, which is to assess Colville’s own role as curator of Churchill’s memory. For in some ways the most extraordinary thing about Colville’s revelation was that he decided to put it on record at all, given the years that he had previously spent burnishing his former boss’s reputation. In order to make sense of this, we believe, the ‘Colville tape’ must be considered in its entirety – the apparently dull parts as well as the three-and-a-half minutes of dynamite that sit among them. For the tape illustrates the position Colville held as a key player in a network of Churchillians (as Colville labelled them) – former colleagues and employees, friends and relatives – who had significant influence on the way that Churchill scholarship developed, and continues to develop. ⁴ In his acclaimed biography, Roy Jenkins was insistent that Churchill did not engage in concealment, and that it would be ‘almost impossible’ to unearth new facts, nearly all of which, he said, had been provided in Randolph Churchill and Martin Gilbert’s official biography. ⁵ Many subsequent works have shown that there is, in fact, much more that can be learned, but the availability today of the digitized version of the approximately 800,000 pages of the Churchill papers (together with huge numbers of other sources in the public domain) might tend to suggest that the record is both complete and wholly available. That, however, is not quite the case. What scholars can see and the conditions under which they can see it are a legacy of decisions taken after Churchill’s death, strongly influenced by the Churchill family and a range of other intermediary figures and institutions in addition to Colville himself. Whether or not they are aware of it, scholars still deal with the legacy of this process, even 50 years after Churchill’s death.

There is already a substantial literature on how the public memory of Churchill has been crafted, including by more recent politicians who have deployed their version of him for their own ends. ⁶ That question intersects with the issue of how Churchill himself cultivated his chosen image through the writing of his war memoirs. Building on the pioneering work of both Robin Prior and John Ramsden – and in line with a growing body of work on political reputations – David Reynolds’s magisterial account shows how Churchill’s six volumes on The Second World War (1948–1954) were written with a constant eye on current political necessity, making selective use of documents to which Churchill and his team had privileged access. ⁷ Reynolds’s final remark is striking: ‘In death, as in life, Winston Churchill … remains in command of history.’ ⁸ With the Colville tape in mind, it is this observation that we wish to interrogate. Whereas it is undoubtedly true that Churchill’s writings have always had a great influence on how he has been perceived and portrayed, it is important not to exaggerate the extent of his personal control and posthumous dominion. On the one hand, as the detail of Reynolds’s story shows, Churchill’s accounts of events were by no means free from public challenge within his own lifetime, even if dissenters laboured under important disadvantages. On the other, Churchill sacrificed some of his control even while he was still alive. In May 1955, just weeks after he stepped down as Prime Minister, he expressed an interest in going through his archive, in part because he was thinking of writing a new memoir dealing with the Edwardian period, and in part because ‘he might like to destroy some private correspondence, etc.’ He was told, however, that the terms of the Chartwell Trust (created in 1946 for the financial benefit of Churchill’s heirs) put limits on his freedom, either to destroy things, or to use copyright material in further published writings without payment. The memoir went unwritten and, as far as we know, the private correspondence remained (at least at this stage) undestroyed. ⁹ In this sense, Churchill’s own ability to ‘command history’ was restricted by his own conflicting objectives: financial efficiency versus the demands of privacy and literature.

We suggest, furthermore, that after Churchill’s death a coterie of subordinate figures had an impact on the development of his reputation in a way that was in important ways as powerful as that of the man himself. Ramsden has hinted at the influence of the Churchillians, the ‘inner circle of admirers and close associates’, but he focused on their various publications (which we will also consider) rather than their activities behind the scenes. ¹⁰ The significance of the latter has not been recognized. The Colville tape certainly raises the question of what else about Churchill might have been hidden or suppressed, but we stress that we are not alleging any kind of coordinated conspiracy. The Churchillians did share a generalized wish to honour and preserve Churchill’s memory, in some cases at the expense of historical truth, but it is important to emphasize their strong positive contributions as well as considering attempts to manipulate the record. It is further necessary to emphasize the diversity of these figures’ approaches: there was no single approach on which these ‘reputational entrepreneurs’ were all agreed, and in fact, there were tensions and rivalries between several of the actors. ¹¹ They had differing interests and priorities, and faced a variety of competing pressures, which militated against the production of a single orthodoxy. It is these things that we need to understand as we confront the published and unpublished evidence of Churchill’s past. Writing about colonial archives, Ann Laura Stoler observes that we need to pay attention not only to ‘archival content, but to the principles and practices of governance lodged in particular archival forms’. ¹² The same is true for Churchill: knowledge about him is structured by his archive, but that archive (which in its original state was chaotic) was curated, and (at the margins) perhaps tampered with, by many hands before it arrived in the condition that we find it today. ¹³ Appreciating how it reached its current state casts light on Churchill’s world and the Churchillian networks that survived him.

Although, as will be seen below, some correspondence between Churchill and Lady Castlerosse survives in the Churchill Papers, none could be described as ‘love letters’. Colville implied to Ackerman and Barnett that he hoped and believed that the letters he mentioned had been destroyed, presumably by Clementine. ¹⁴ In the absence of such evidence, what can we say about the relationship between Churchill and Lady Castlerosse? The woman herself is not mentioned in the eight-volume official biography, aside from some passing references to her in the documentary companion volumes. ¹⁵ There are some references to her in a work on Churchill’s paintings, as he portrayed her at least four times, although the book does not comment on the possible significance of this. ¹⁶ Some valuable information can be found in a 1973 biography of Lady Castlerosse’s husband Valentine, a renegade Irish peer who acted as gossip columnist for Lord Beaverbrook’s Express newspapers. The biographer, George Malcolm Thomson, had been close to Beaverbrook during the war, and it seems possible that he was writing out of personal knowledge. He devoted a few pages to the relationship, writing that Churchill had been ‘fascinated’ by Lady Castlerosse and speculated that the pair may have enjoyed a ‘romantic friendship’. But if, as seems possible, he was trying to imply that the two had had an affair, subsequent writers do not appear to have taken the hint. ¹⁷ The author John Pearson is one exception, insofar as he referred to an unsourced rumour that Lady Castlerosse and Churchill had slept together at the Ritz Hotel in Paris. Pearson stated that this was ‘unprovable either way’. ¹⁸ Another is Lyndsy Spence, who in her recent biography of Lady Castlerosse states that there was ‘much-repeated gossip’ to the same effect, and cites Pearson as her source. ¹⁹ On the face of it, Pearson’s and Spence’s claims do not look well supported. However, there survives considerable evidence which suggests that Colville’s claim of an affair was, at least, plausible.

First, we need to consider Churchill’s sexuality as well as the mores of the time. In the tape, Colville emphasized that Churchill was not an habitual philanderer. ²⁰ Although Colville only knew him during the last 25 years of the latter’s life, he was undoubtedly right that Churchill was not highly sexed. ²¹ However, it is also worth stating that suggestions that Churchill was gay or bisexual (if only in a passive or repressed way) seem wide of the mark – notwithstanding his private tolerance of the homosexuality of friends such as Eddie Marsh. ²² (He was also uncensorious of male politicians who had heterosexual affairs.) ²³ Moreover, even if he was not especially interested in sex, he was capable of falling passionately in love (as he seems to have done initially with Clementine). In his youth he had a significant romantic relationship with Pamela Plowden and proposed to two other women, one of them the American actress Ethel Barrymore, whom he pursued ‘with relentless vigour and the most honourable intentions, laying his heart and career at her feet’. ²⁴ He may have had more pre-marital relationships than we know about, and it may be most helpful to think about him as being somewhat vulnerable to crushes. ²⁵ However, as has often been noted, he had few close female friends and generally did not treat women as intellectual equals.

Doris Castlerosse, it has to be said, had no aspirations to be considered an intellectual. She was born Jessie Doris Delevingne in London in 1900. Little is known of her early life, but by the late 1920s she had emerged into Mayfair society and was well known for using men to finance her extravagant lifestyle. She is said to have been the model for the fast young widow Iris Storm in Michael Arlen’s The Green Hat and, later, for the tempestuous temptress Amanda in Noel Coward’s Private Lives. ²⁶ In 1928 she wed Lord Castlerosse. The marriage was stormy and was dogged by financial trouble, abuse, and mutual infidelity. It appears that in 1932 Doris had an affair with Churchill’s son Randolph. ²⁷ Whether or not she had yet met the elder Churchill is unclear, but during the late summer of 1933 she and Winston spent a prolonged period in each other’s company at the Chateau de l’Horizon, a villa on the Riviera owned by Churchill’s actress friend Maxine Elliott. On this first occasion, during which Churchill painted Doris for the first time, he was accompanied by Clementine and their daughter Sarah. ²⁸ Churchill returned to the Chateau three summers running. Clementine did not go with him owing to her dislike of the louche atmosphere, but on each occasion Doris was present for at least part of the time. ²⁹ Moreover, there exist pictures of the two together. ³⁰ This certainly indicates that she and Churchill had the opportunity for such an affair as Colville described. As mentioned above, there survives a small quantity of communications between them. Particularly striking is a letter Churchill sent her while en route back to Britain after the 1934 holiday (on which occasion Randolph had also been present, and during which Winston had angrily chased away a paparazzo who had snapped Doris watching him painting). Churchill recalled the fun that the two of them had had at the Chateau and said that her presence had been a delight. He described her as a manifest blessing and said that she had been a ray of sunshine around the swimming pool. He speculated too that he and Doris might meet there again the following year. ³¹ Churchill and Doris met in Britain as well as in France: in September 1934 Doris sent the Churchills a social invitation, which they accepted. ³² That December Churchill sent Doris one of the pictures he had painted of her; she appears to have acquired another one later. ³³ It should be noted that, on the long cruise which she took without Winston in 1935, Clementine ‘fell romantically in love’ with one of her fellow voyagers, Terence Philip. Whereas it seems doubtful that she was reacting to knowledge of an affair between Winston and Doris, the episode could be taken as indicative of a coolness in the Churchill marriage at this time. ³⁴

In a 1936 telegram to Doris, Churchill referred to two letters he had received from her, describing them as ‘most welcome’ and signing off ‘Love W’. ³⁵ Significantly, the two letters he mentioned no longer exist in the Churchill Papers. This lends weight to the notion that correspondence from her, possibly of a compromising nature, was later destroyed, as Colville suggested. The most interesting surviving letter was written the following year. In it, Doris discussed her forthcoming divorce. She also told Churchill ‘I should like to see you. I am not dangerous anymore’. ³⁶ This letter could be read as an indication that the affair was now over, and that Doris did not mean to try to revive it.

After her divorce was finalized in 1938, Doris travelled to the USA and remained there after the outbreak of war. ³⁷ After Churchill became Prime Minister on 10 May 1940 she appears to have sent him a congratulatory message, as five days later he sent her a telegram of thanks. ³⁸ That October he appears to have fulfilled her request for a signed photograph for a New York charity auction in support of Bundles for Britain. (She added another plea: ‘Please stop walking around London during air raid.’) ³⁹ In the spring of 1942, upon hearing that Beaverbrook was in the USA, she wrote to him (Beaverbrook) that she was ‘desperate’ to return to London, something that was difficult to achieve in wartime. ⁴⁰ She was keen to return by air, but places on civilian planes were generally reserved for ‘high priests of diplomacy’ and those contributing to the war effort. ⁴¹

When Churchill visited Washington in June 1942, Doris saw her chance to get home. Presumably, Beaverbrook had told Churchill of her situation because on 23 June the Prime Minister rang her in New York. ‘So very many thanks for telephoning me this morning;’ she wrote to him later that day; ‘to hear your voice again gave me such a lift’. ⁴² Thomson’s book gives a detailed description of a meeting between Churchill and Doris ‘In the early weeks of 1942 … in a Washington hotel’. ⁴³ It seems possible that such a meeting did take place although Thomson was perhaps wrong about the timing, given the proven telephone contact between Churchill and Doris in June. Thomson wrote that the pair discussed the risk that Churchill’s ‘paintings of Doris, which were by this time stored in New York, might fall into the hands of an American magazine publisher who could use them to damage the reputation of Britain’s war leader’. It seemed the ‘simplest plan would be for Churchill or a trusted friend [presumably Beaverbrook] to buy the pictures back from Doris’. ⁴⁴ Thomson’s account could be taken to imply that Doris tried to blackmail Churchill with the portraits, for how could they have got into a publisher’s hands without her own active assistance? Nor would they have caused a scandal unless she had attempted to create one. It seems quite clear, though, that no money in fact changed hands. But whether or not this conversation, or one like it, ever took place, Churchill undoubtedly did help Doris to return to Britain, via the good offices of Harry Hopkins, FDR’s eminence grise. In an undated letter to Hopkins, she observed: ‘You kindly phoned me some time ago telling me that Winston Churchill had written you to ask you if you could arrange for my clipper [aeroplane] passage back to England’. ⁴⁵ Hopkins did then help Doris get back to Britain, asking the Chief of the State Department’s Visa Department to expedite her return. ⁴⁶

Upon her arrival in London that autumn, Doris took up living at the Dorchester Hotel. According to Thomson, Churchill’s paintings were still with her. ⁴⁷ In December 1942 she died of an overdose of sleeping pills; the inquest returned an inconclusive verdict. ⁴⁸ An accident seems likely, although her ex-husband assumed suicide. There was an interesting sequel. Although Leonard Mosley, another Castlerosse biographer from the Beaverbrook stable, later tried to obscure the issue, it is clear that the paintings ended up for a time in Beaverbrook’s hands. ⁴⁹ Beaverbrook met with Doris’s brother Dudley twice in the days after her death and we may surmise that Beaverbrook persuaded him to put them in his temporary care. ⁵⁰ This can be seen from a letter Dudley’s lawyers wrote shortly after the war authorising Beaverbrook to ‘hand over to Captain Dudley de Levingne the two paintings in your possession belonging to the late Doris, Viscountess Castlerosse’. ⁵¹

Did Churchill and Lady Castlerosse definitely have an affair, as Colville claimed? The contemporary evidence lends the story credibility. Moreover, Colville certainly believed it and – as one of Churchill's most trusted aides for key periods in the 1940s and 1950s – he would not have made the allegation lightly. The remainder of this article is concerned with Colville’s behaviour as a guardian of Churchill’s reputation in the decades after Kelly’s shocking revelation, and the ways that these intersected with the actions of other Churchillians. Exactly when the incident with the letters happened is unknown. It could have occurred at any point during the years 1948–59, the period that Kelly worked for Churchill, but the later 1950s seems most likely. It may be significant that Colville recorded Clementine’s reaction but not her husband’s; within a few years of his retirement, as his mental faculties weakened, he may not always have been fully aware of what was going on. As his condition deteriorated, his friends and advisers did their best to protect him by carefully regulating his increasingly rare public statements. ⁵²

At the same time, various preparations were being made for publications that would appear after Churchill’s death. In 1960, Churchill appointed his son Randolph to be his official biographer, with the proviso that the work should appear posthumously. ⁵³ Randolph acquired a team of researchers, initially headed by Michael Wolff. Martin Gilbert joined in 1962. ⁵⁴ Randolph’s difficult personality helped produce tensions between some of the Churchillians, but it was not the only factor. Churchill’s friend Violet Bonham Carter (ennobled in 1964 as Lady Asquith) was preparing her own book, on which she had been working for some years. Some delay was caused by the need to reshape it for the US market, at the behest of her US publishers – thus commercial demands acted as one influence on the way that Churchill was portrayed. The book, which in the UK carried the title Winston Churchill As I Knew Him, was on the verge of publication when Churchill died in January 1965 and was then delayed until April. ⁵⁵ This work took the story only up to 1916, and Bonham Carter considered writing a further volume. Although she never did so, she continued (until her own death in 1969) to take a strong interest in Churchill’s reputation, which, on the face of it, she and Randolph had a common interest in defending.

Yet things were more complicated than that. To begin with, Bonham Carter was if anything more concerned with the reputation of her father, Liberal Prime Minister H.H. Asquith, than she was with Churchill’s. ⁵⁶ She was also concerned with her own reputation. She denied one author permission to quote from a letter which she had written to Times editor Geoffrey Dawson at the time of the Abdication, which was strongly critical of Churchill. To publish it, she claimed, ‘might give a false impression of my attitude towards Geoffrey Dawson’, who by the 1960s was considered a notorious advocate of Appeasement. ⁵⁷ Furthermore, neither she nor Randolph was much inclined to help the other. They wrangled over granting each other access to materials that they respectively controlled. Bonham Carter – who seems to have once had a romantic interest in Churchill – revealed to Randolph, whom she plainly did not trust, that ‘with infinite regret and sorrow I have destroyed some of the most intimate and (politically) interesting’ of Winston’s letters ‘which were of a strictly confidential nature’. ⁵⁸ (Intriguingly, she also appears to have destroyed part of a letter from Colville around this time.) ⁵⁹ At one point, Martin Gilbert wrote to her that suggesting that Randolph resented the fact that his mother had given her blessing to a rival work. Gilbert promised to help her out, however, by supplying her with material from the Churchill Papers on a ‘quite private and unofficial basis’. ⁶⁰

Moreover, Bonham Carter’s efforts to defend Churchill sometimes ran up against inconvenient facts. In 1968 she was upset when Charles de Gaulle publicly alleged that the British had backed a Free French revolt against his authority in 1942. She turned to Colville, who himself suggested approaching Desmond Morton, Churchill's wartime assistant, who had liaised closely with the French. Colville thought he might be encouraged to write to The Times. ⁶¹ By this stage, however, Morton was a rather disillusioned Churchillian. He had provided covert assistance to R.W. Johnson in his critical 1963 biography of Churchill (a book that Randolph considered ‘particularly meretricious’). ⁶² In his view, moreover, what de Gaulle had said was broadly true: ‘there was certainly a period when Winston, who was kept informed of all that was going on, would have given an eye-tooth, if he had any left, to depose de Gaulle.’ ⁶³ Colville took this information, as Churchill would have said, ‘with philosophy’. He commented that Morton's letter was extremely interesting, adding that its author was very well-informed. ⁶⁴ He thus showed himself to be genuinely concerned with historical evidence and was no mere slavish Churchill devotee.

He was, however, more than prepared to rally round when the occasion demanded. In 1966, a few months before the publication of the first volume of the official Life, Churchill’s doctor, Lord Moran, produced his own account. ⁶⁵ Prior to publication he made some ‘radical omissions’ from the last section of the book, as well as removing or toning down references to Churchill family members and other individuals who were still living. ⁶⁶ The Churchill family was nevertheless outraged at what it perceived as a serious breach of Moran’s duty of medical confidentiality. ⁶⁷ The broader issue at stake was whether or not Churchill had been fully physically and mentally competent throughout his two premierships. In response to Moran’s perceived betrayal, a group of former wartime officials produced a volume called Action This Day. Colville organized its writing at the request of Clementine Churchill. ⁶⁸ In many ways, the book can be seen as the successor to two similar, rather uncritical volumes of essays published within Churchill’s own lifetime. ⁶⁹ However, this type of publication by former civil servants was considered ‘an unprecedented step’; the contributors were pushing back the boundaries of official secrecy even as they criticized Moran for his lack of discretion. ⁷⁰ Ramsden observes of the book: ‘Few notes of criticism were permitted to intrude on the traditional picture, and in some cases this involved not just putting a favourable spin on events but departures from the strict truth.’ ⁷¹ Colville, though, was quite honest about Churchill’s inconsiderate behaviour towards his wartime staff and about his tendency to talk too long in Cabinet meetings; and Clementine approved the chapter. There was no sense in writing about such topics unless one was totally honest, or even indiscreet, Colville told Anthony Eden. However, he stated that he would leave out a section about Churchill's real reasons for supporting Edward VIII at the time of the Abdication. Colville intended the book less as a refutation of Moran than as an effort to put Churchill in a correct perspective. ⁷² Colville seems to have been particularly concerned about Moran’s claim that Churchill suffered from the ‘black dog’ of depression, but his efforts to prevent this becoming the conventional story were at best partially successful: the evidence about Churchill’s mental health continues to be debated. ⁷³

One aspect of Colville’s influence lay in his efforts on behalf of Churchill College. He had been a moving spirit in its creation in the late 1950s. ⁷⁴ In the 1960s he set about establishing a collection of archives. ⁷⁵ Unless Randolph Churchill changed his mind, he informed Bonham Carter in 1966, all of Winston's papers would eventually find a home at Churchill College. Colville’s ambition was to collect there as many papers as possible related to Churchill and his era. An excellent start had been made with the papers of Clement Attlee, Ernest Bevin, and many others. However, he had a competitor in Beaverbrook’s son Max Aitken, who was also collecting papers. ⁷⁶ Aitken’s holdings soon found a home at the Beaverbrook Library in London, which opened in 1967. With A.J.P. Taylor as Director, the Library might have established itself in the long-term as a rival powerhouse to the Churchill Archives Centre (which got its own purpose-built building in 1973, although it was not actually the technical owner of the pre-1945 Chartwell Trust Papers). However, after the Beaverbrook Foundation withdrew funding, the Library shut in 1975. ⁷⁷ The importance of Colville’s Establishment, charitable, and City connections (he was now working as an investment banker) should not be underrated, as they helped establish the Archives on a sound footing. It is significant that the bulk of the discussion on the Colville tape focused on ways of raising money.

In June 1968 Randolph died. The problem of finding a successor to him as the author of the official biography occupied a lot of Colville's and other people's time, the problem being that there were too many candidates rather than too few. ⁷⁸ The biography was being financed by a company called C & T, which comprised the British and US publishers of the book, together with the Daily Telegraph, which held the serial rights. This gave an effective veto on the choice of biographer to Lord Hartwell, owner of the Telegraph, and it was only after numerous meetings between C & T and the representatives of the Chartwell trustees (Colville and Sir Leslie Rowan, another former Churchill private secretary) that Martin Gilbert was selected. This was part of a compromise solution suggested by Gilbert himself whereby a parallel single volume biography was to be written by Lord Birkenhead, although this book was never completed on account of Birkenhead’s early death. ⁷⁹

Gilbert, then, was certainly not the inevitable candidate, and the official biography might have been a rather different beast if written by another author. However, he brought superb qualities to the role. Whereas the finished product, which is something in the nature of a chronicle, has often been criticized for being insufficiently analytical, it should be recognized that Gilbert was indefatigable in his completion of a daunting and not especially well-paid task. ⁸⁰ The job, as he interpreted it, required him not merely to make use of material already in the archives but to track down documents and to source recollections from surviving Churchillians. In his memoir, Gilbert admitted ‘to having been somewhat afraid of Colville’, even though he did not doubt that he was well intentioned and wanted to assist: ‘He so frequently expressed to me his low opinion of historians (“you people almost never get it right”) that I was sometimes inhibited even from asking questions, lest they revealed the abyss of my ignorance.’ ⁸¹ Colville’s impact on the published biography was noticeable, but Gilbert’s regard for him may not have been fully reciprocated, and during a crisis in Gilbert’s relations with the trustees the possibility of recruiting an alternative biographer was mooted. ⁸²

Colville’s comments to Gilbert about historians are telling. He actually wrote history himself, starting with a biography of Lord Gort published in 1972, but felt that many scholars failed to understand ‘the conditions, the prejudices, [and] the accepted ethics’ of the times about which they wrote. This observation was closely linked to Colville’s own Tory politics. The detractors of Churchill were, in his view, little people who talked down Britain, were critical of its former Empire, and urged the destruction of everything that Britain had stood for in past times. There was, of course, a certain amount of distortion that was a cause for legitimate concern. Colville was particularly exercised by Rolf Hochmuth’s Soldiers, a play staged in London in 1968, which alleged Churchill’s complicity in the death of General Sikorski. Although Colville claimed to be sure that Churchill’s reputation was proof against such attacks in the long run, it is clear that he nonetheless felt some anxiety on that score. ⁸³

Public discussion of Churchill was not monopolized by the official Life, then. When researching his biography, the Cambridge historian Henry Pelling consciously chose not to seek help from Churchill’s family or former colleagues, in the belief that this would allow him to write ‘with more freedom’. ⁸⁴ Nevertheless, Churchillian networks remained influential. In 1970, Robert Rhodes James published a study of Churchill’s life in the period 1900–39, subtitled A Study in Failure. To begin with, Randolph had ‘warmly encouraged’ this project, but the two men fell out over questions of interpretation; and Gilbert was angered when Rhodes James publicly questioned the ‘massive length’ of the official biography. ⁸⁵ However, Colville had a high opinion of the Study, which may help explain why Rhodes James was permitted to edit Churchill’s Complete Speeches, which were published in eight volumes in 1974. ⁸⁶ Churchillians also published numerous books of their own. Notably, Colville gave assistance to Mary Soames, Churchill’s youngest daughter, on her biography of Clementine (who died in 1977) and to Winston S. Churchill on his life of Randolph. ⁸⁷ Writing to young Winston, Colville encouraged him to be more forthright about Randolph’s staggering rudeness and excessive drinking. ⁸⁸

During the 1970s, Colville himself was increasingly in the public eye. He contributed to the landmark documentary TV series The World at War (1973–4). He published one volume of memoirs in 1976, and was subsequently persuaded by another publisher to produce a follow-up. ⁸⁹ He was involved in controversy when Lord Boothby, still bitter at his treatment during the Second World War (when a scandal had forced his resignation as a junior minister) claimed in an interview that Churchill had been ‘gaga’ when he returned to office in 1951. ⁹⁰ Colville responded that Churchill had barely set eyes on Boothby at that time and had not cared about his views in the slightest. ⁹¹ Boothby hit back, claiming that Colville had for many years been ‘boasting, and boring everyone in talk and print, about his relationship with Sir Winston Churchill’, yet was in fact ‘of no importance at all’. ⁹² In turn, Mary Soames and Martin Gilbert rushed to Colville’s defence. ⁹³

Colville also liaised regularly with the government, which continued to take an interest in the Churchill Papers, partly because civil servants took the view that some of them were still state property. In the late 1960s, for example, we find him acting as a contact point between the Cabinet Office, the Lord Chancellor’s Department, the trustees, the publishers, and Gilbert, over the question of the physical location of the papers, which were at that time still at Randolph Churchill’s house in Suffolk. ⁹⁴ His diplomatic skills were much in need. In 1989, one official noted:

the inherent contradiction of the Archive Centre; an institution built for the purpose of holding the Churchill Archive, of which the major part was deposited on only a temporary basis, and subject to a Trust drawn for the benefit of the lineal descendants of Winston Churchill.

While he lived, Colville played a key role in smoothing over this unresolved tension. ⁹⁵ Soon after his death, though, the trustees and the College/Archives Centre fell into bitter conflict over the ownership question. This crisis was only resolved in 1995 with the controversial purchase of the papers for the British nation with the assistance of a Heritage Lottery Foundation (HLF) grant. ⁹⁶

Colville’s last major act was to publish his wartime and postwar diary. ⁹⁷ He had previously said that he would not do so – in spite of drawing on it in his memoirs, and making it available to Martin Gilbert – ‘because it contains opinions of people and views of events which in the light of subsequent knowledge I believe to be unjust’. ⁹⁸ Perhaps he felt the time had come for greater frankness, yet he was not, perhaps, completely frank. When the published version is compared with the originals at the Archives Centre, his claim to have eliminated only minor entries of no general significance appears justified. ⁹⁹ However, the section of the diary dealing with the very final stages of Churchill’s period as Prime Minister, and with the Suez Crisis of 1956, is not open to researchers. It is currently listed as being due for review in 2046. ¹⁰⁰ When the book was published another of Churchill’s private secretaries, John Peck, commented: ‘I am filled with amazement at the risks Jock was running in the matter of security, for which he should have been sacked on the spot if he had been caught.’ ¹⁰¹ But by keeping a diary, preserving it, and making it available, Colville wrote (and published) himself into history, inadvertently at the expense of Downing Street colleagues such as Peck and John Martin, whose respective books could not compete with the colour and immediacy of Colville’s. ¹⁰²

The year 1985, when the diary was published, returns us to the Colville tape. The discussion recorded on it was principally concerned with how to meet a fundraising target of £30,000: in the course of it Colville revealed how well connected he was, as he ran through a list of people who might be contacted. These included the Sultan of Brunei and wealthy Americans such as Ross Perot. Colville also cast light on some of his own attitudes to former Churchill associates, such as Clarissa Eden, with whom his relations were strained because his diary had been critical of her, and Harold Macmillan, whom he regarded as voluble but whose recollections he thought not very reliable. He also ran through a list of surviving Churchillians who might be contacted for the Centre’s oral history programme and in this way he came to Denis Kelly. His story about Churchill and Lady Castlerosse thus came out virtually as an afterthought. ¹⁰³ In 1987, Colville died of a heart attack, never apparently having spoken of the matter again.

Irrespective of what may be concluded about Churchill’s alleged affair, the episode at Chartwell recalled by Colville retains some mysterious elements. It is not possible to say for certain why Kelly, who otherwise showed ‘great loyalty’ to Churchill, acted as he did. ¹⁰⁴ His unpublished memoirs and oral history interviews conducted with him by the Archives Centre cast no direct light. (Kelly disliked Clementine but claimed that, as far as he knew, Churchill had never been unfaithful to her; perhaps Kelly’s action was genuinely unwitting.) ¹⁰⁵ Colville’s behaviour is easier to explain, though it was certainly paradoxical. He kept the secret for many years, only to spill the beans almost at the last minute. Although he clearly (and rightly) trusted Barnett and Ackerman not to reveal it, he must have known that the story was likely to come out at some point in the future. The simple solution is that Colville was conflicted over his duty to history. He was a man who kept a wartime diary in spite of the rules, and overcame the questionings of his conscience in order publish it. He disparaged the efforts of historians but also went to considerable efforts to help them with their work and to build up the archives on which they depended. He was dedicated to Churchill’s memory but was not blind to his faults. It seems that in the end, with respect to the Castlerosse affair, his own commitment to historical truth overcame his instinct for discretion. His simple explanation to the archivists of why he was revealing the story of the affair was merely that it was interesting. ¹⁰⁶

Colville, then, exercised considerable power over Churchill’s reputation, even if he was not always quite sure what he ought to do with it. It is not, however, our intention to displace Churchill from his ‘command of history’ merely to substitute Colville in his place. In the battles over Churchill’s legacy there were no ultimate winners, but rather a range of actors who failed and succeeded in different ways. Thus, in spite of the criticisms offered by Colville’s Churchillians and in spite of doubts over whether his diary was genuinely contemporaneous, few Churchill scholars can resist quoting Lord Moran’s book. ¹⁰⁷ Mutual tensions and rivalries, rather than the hegemonic imposition of a single interpretation, are the key to understanding how the historical memory of Churchill was shaped. Contingency and copyright (such the death of Randolph Churchill) also played a part. Therefore, although Colville undoubtedly played a particularly pivotal role, no one individual commanded history.

In recent years there has emerged a new historiography, which attempts to place Churchill in his cultural and social context, in addition to his political one. This implies the need to think more deeply about the role of his archives, how they were shaped, and how they have in turned shaped the memory of him – yet too much Churchill literature continues to treat the archives as uncomplicated sources of factual information. Stephanie Decker has recently suggested that historians need to be more explicit about how and why they use archives in order to help explain their work to people in other disciplines. ¹⁰⁸ Arguably, this would have benefits for the historical discipline too, as archives are often taken too much for granted. They are not neutral repositories of information but are managed organizations with their own institutional priorities. Historians do not sit outside these processes, in splendid isolation. In fact, they often contribute to or influence archival content themselves, by writing letters to historical figures and eliciting responses from them that are then preserved. Some historians create their own archives that sit alongside those of the people they study.

Even today, Churchillian networks have not disappeared (even though the numbers of those who knew Churchill personally are much reduced) and they remain relevant to the ways in which modern understandings of Churchill are shaped. Yet, as we have shown, these networks never had a single, hagiographical objective: mixed motives, conflicting objectives, and personal rivalries all influenced the public memory of Churchill. That memory continues to evolve as new material is made available, which itself can be an exercise of some delicacy. In pursuit of its mission to ‘to preserve and make available the raw material of our recent past’, the Churchill Archives Centre is not alone in having to balance the immediate interests of its researchers with the sensitivities of individuals and descendants whom it hopes to persuade to donate new material in the future. ¹⁰⁹ In other words, in spite of their desire that ‘collections should be open and accessible’ the archivists cannot always release material as soon as historians would wish, given the need to work with depositors and agree in some cases to restrictions on public access or risk the material being destroyed. ¹¹⁰ It should be emphasized that the Archives Centre does a superb job of preserving and presenting the material in its care. But the nature and scope of the materials that they can present today have been shaped in the past by the processes which we have described, and historians need to be aware of this.

Even if the story of Churchill’s affair with Lady Castlerosse happens to be true, the possibility that he had a brief fling is not enormously significant in itself, not least when set against his other achievements. Rather, we highlight it because it highlights how Churchillians sometimes acted to suppress – for a long time, if not necessarily permanently – information which they considered damaging or discreditable. (The role played by such historical intermediaries more generally may well be deserving of further attention.) Historians, naturally enough, are never best pleased by the destruction or withholding of evidence, but we have endeavoured to show here that Churchillian networks also had a very considerable positive influence, especially through the creation of the Archives Centre. The Centre can legitimately claim to be ‘perhaps the closest thing in the UK to an American presidential library’, yet it has largely avoided the heroic glorification of its subject to which those institutions often fall prey. ¹¹¹ We must not imagine, however, that we have yet got to a point where we can say all the relevant information about Churchill is in the public domain. Perhaps researchers, as they make their way upstairs to the reading room at the Churchill Archives, and look down at the Jock Colville Hall beneath them, will be encouraged to reflect on the complex and ambiguous legacy bequeathed them by the former Private Secretary of an eternally surprising human being. Perhaps they will give a thought, too, to the ghost of Doris Castlerosse.