Intelligence in the Era of the Sun King Part II

William III painted in the 1690s

William III’s understanding of foreign intelligence, and in particular of black chambers (so called after the French cabinet noir), exceeded that of any other ruler of his time – or any other monarch in British history, with the possible exception of George I. As well as receiving decrypts from Wallis, from 1693 William III was also able to make use of a black chamber established at Celle in Lower Saxony, the official residence until 1705 of one of the two branches of the house of Brunswick-Lüneburg. William III was a personal friend of the last member of the dynasty, Duke George William of Celle, who had probably been inspired to found a black chamber by William’s diplomatic coup after Wallis’s decryption of Louis XIV’s correspondence with his ambassador in Poland. The two rulers had a common interest in monitoring French diplomacy in Germany and Northern Europe. Copies of intercepted French despatches obtained in Celle were forwarded to London. Inspired by the Celle example, the other branch of the house of Brunswick-Lüneburg in Hanover also set up a black chamber, at Nienburg. After Duke George Louis I, the future King of England, became Elector of Hanover in 1698, the two black chambers cooperated closely. By 1701 they even had a joint bonus scheme. When George William died in 1705, George I inherited the duchy of Celle and the two black chambers as well as the two duchies combined. As well as taking over Celle’s codebreakers, George removed the contents of the ducal palace and took them to Hanover. The development of codebreaking in both Celle and Hanover probably owed much to the inspiration of the great mathematician Gottfried Wilhelm von Leibniz, who, though not a codebreaker himself, had a keen theoretical interest in the principles of cryptanalysis.

Despite the achievements of the black chambers in Celle and Hanover, Leibniz deferred to Wallis, even when Wallis was in his early eighties, as the greatest codebreaker of the age. Wallis wrote in 1701:

I have been solicited by . . . Leibnitz, more than once, in behalf of the Elector of Hanover: who is willing to send hither some young men, whom he desires I would instruct therein; leaving it to me to make my own proposals on what terms I would undertake it. To which I have returned answer, That I shal be ready, my selfe to serve his Electoral Highness if there be occasion: but the skill of doing it, being a curiosity which may be of use to my own Prince, I do not think it proper to send it abroad, without his Ma[jes]ty’s leave.

In 1701, two years before his death, Wallis at last secured the regular salary he had sought for the past decade. While training his precocious grandson William Blencowe, who had just graduated from Magdalen College, Oxford, at the age of only eighteen, to succeed him, Wallis was granted £100 a year, backdated to 1699. At Wallis’s request, the salary was paid to him during his lifetime, ‘which is not like to be long (being now in my 85th year), and thenceforth to the young man during his Majesty’s pleasure’.

William III’s most important foreign agent network was at James II’s court in exile at Saint-Germain-en-Laye, near Paris. The agent penetration of Saint-Germain became so notorious that the Abbé Renaudot, its French liaison with Versailles, mistakenly concluded that James’s Secretary of State, Lord Melfort, was in the pay of William III. At Louis XIV’s insistence, James dismissed Melfort in 1694 and banished him to the provinces. William’s agents at Saint-Germain enabled him to feel some confidence that, as was to happen early in 1696, he would receive advance warning of any attempt by James to launch a cross-Channel invasion with French support. But William could not hope to keep the whole Jacobite exile community in France under surveillance. Numbering about 50,000, it was much larger, in proportion to the size of the British population, than the membership of the British Communist Party at any time in the twentieth century, as well as a far more diffuse and elusive target. Until the last minute, there was no intelligence warning of a serious plot to assassinate William which had been devised by a group of Jacobite exiles in 1695 and timed to precede the planned invasion of 1696.

Assessing the threat posed by Jacobite conspiracy in England was also difficult. Much public denunciation of William and Mary was simply alehouse talk which posed no significant threat to national security. The usual penalty imposed at the Middlesex assizes for such seditious talk was a fine combined with a sentence to spend an hour at the busiest time of the day in the pillory at Charing Cross, Covent Garden, St James’s Street, St John Street, Bow Street, the Strand or New Palace Yard. Most informers’ reports of Jacobite sympathizers contained nothing of real importance. An informant named James Ormiston wrote, for example, in 1695 to one of the secretaries of state, Sir William Trumbull, to report that a Captain Clifford had mistaken him for a Catholic and told him in an alehouse, under the influence of drink, that preparations for a Jacobite rising were going very well (though he declined to give details) and ‘in a short time we should have home our sovereign King James again’. A more experienced agent told Trumbull that such claims were ‘trumpeted in all [Jacobite] coffee houses’. There were similarly exaggerated reports that many French and other foreign travellers were Jacobite spies. Even John Wallis, not usually an alarmist, feared that many of the ‘great concourse’ of foreign students he saw in Oxford were really Jacobite agents sent ‘to take measure of the inclinations of the Kingdom’.

William III was more concerned with Jacobite sympathies among leading members of his own government and armed forces, some of whom had only recently changed sides in his favour and were quite capable of changing back again. Among them was the man who later emerged as the greatest British general of his generation: John Churchill, Earl (later Duke) of Marlborough. Though Marlborough had defected from James II’s forces a few weeks after William landed in England, in 1691 he began a secret correspondence with both James and the Duke of Berwick, James’s illegitimate son by Marlborough’s sister Arabella, and maintained contact with more than one Jacobite agent. Early in 1692 William dismissed Marlborough without warning. Those close to the King let it be known that Marlborough’s recent correspondence with James II had been intercepted and that he was suspected of revealing a secret plan by William to attack Dunkirk. Marlborough was sent to the Tower of London for six weeks and it took him several years to regain the King’s confidence. His later belief in the importance of ‘getting intelligence of the enemy’s motions and designs’ must have owed something his experience of William’s discovery of his own ‘designs’.

The majority of those with Jacobite connections in the government and armed forces were of what was called the ‘fire insurance’ variety – men who had no personal loyalty to James II but thought it prudent to keep some contact with his supporters in case he unexpectedly regained the throne. Keeping track of more serious Jacobite conspiracies was made more difficult by the diffuse nature of William III’s domestic intelligence system. William himself was partly responsible. He appointed no Scot or Thurloe to coordinate intelligence operations at home and abroad, as had happened during the Interregnum. William’s ministers were guilty of a classic failure to learn from past intelligence experience. In the mid-1650s the exiled Charles II had come to the pessimistic (and exaggerated) conclusion that Cromwell’s spymasters supplied him with ‘perfect intelligence of whatsoever His Majesty resolved to do’. The disorganized surveillance of Jacobites after the Glorious Revolution never approached in efficiency the penetration of royalist opposition by Scot and Thurloe a generation earlier. William’s two (sometimes three) secretaries of state independently ran their own agents, as well as receiving spasmodic reports on suspected Jacobites from local justices and mayors. Two Whig MPs, John Arnold and Henry Colt, ran their own spy networks to track down Jacobites, as did the Earl of Monmouth. Some county lords lieutenant, who led the local militias, took it on themselves to investigate subversive activity and arrest suspects.

The confused domestic intelligence system (far less effective than foreign intelligence), combined with widespread fear of Jacobite conspiracy, created new opportunities for confidence tricksters, much as earlier fears of ‘popish plots’ had enabled Titus Oates (undeservedly rehabilitated after the Glorious Revolution) to embark on a career as a celebrity fraudster. The confidence trickster who made the most sensational use of fraudulent intelligence on Jacobite plots after the Glorious Revolution was William Fuller, who claimed to have been brought up as a Catholic and employed both as a servant to James II’s adviser Lord Melfort and as a page to the Queen, Mary of Modena. By his own unreliable account, Fuller then secretly changed sides and worked as a Williamite agent at James’s court in exile at Saint-Germain. After assisting in the arrest of a Jacobite courier, Matthew Crone, in 1690 and giving evidence at his trial, Fuller extracted large sums of money to finance his supposed intelligence operations from William and his government. Among those deceived by Fuller was Queen Mary, who ordered him to be paid £100. Other sums paid to Fuller of which record survives are £180 ‘by his Majesty’s commands’ and £110 from the Earl (later Duke) of Shrewsbury. Fuller’s conspiracy theories, like those of Titus Oates just over a decade earlier, alarmed a nervous House of Commons. During two appearances at the bar of the House in 1691, he claimed that Louis XIV had spies in both the Privy Council and the offices of the secretaries of state. The truth of these and his other allegations of Franco-Jacobite conspiracy, he declared, could be proved by two witnesses who were currently abroad but willing to return if given safe passage and protection. The witnesses, however, failed to turn up as promised by Fuller. Early in 1692, Fuller claimed to be ‘very ill, with great vomiting and looseness’ of the bowels after being poisoned by Jacobites and unable to return to the Commons to give further evidence. Pretending to be on his deathbed, he gave a group of MPs who visited him a sworn statement directing them to the house of an apothecary where the elusive witnesses were allegedly lodging. But the witnesses were, once again, nowhere to be found and Fuller did not die. Embarrassed by their previous gullibility in taking seriously Fuller’s fraudulent inventions, the Commons angrily resolved that he was an impostor and cheat who had ‘scandalized Their Majesties and the government and abused this House and falsely accused several persons of honour and quality’. After prosecution by the Attorney General, he was sentenced to stand in the pillory, like Oates in 1685, and to pay a fine of 200 marks.

The most serious authentic Jacobite conspiracies uncovered by intelligence operations were linked plans to invade England and to assassinate William III in 1696. According to William’s informants at the court of Saint-Germain, James was confident, despite his humiliating flight from England and defeat in Ireland, that, with French assistance, he could recover his throne. In January 1696 he sent his illegitimate son, the Duke of Berwick, secretly to England to prepare an insurrection. Berwick rashly assured Louis XIV, who ordered troops to assemble on the Channel coast to support the insurrection, that ‘King James has a great party both in England and Scotland who will take up arms as soon as they hear he is landed.’ William’s closest adviser, the Dutch-born William Bentinck, first Duke of Portland, received intelligence reports from Flanders ‘that the enemy had collected a great body of troops at Dunkirk and Calais as well as a large number of transport vessels and ships of war, that the troops were either on board or being embarked and that it was well known there that they were assembled for the invasion of England’. Portland later confided to the English diplomat Baron Lexington: ‘We were on the brink of a precipice and ready to fall. When, by a manifest interposition of providence, we were made aware of the danger which threatened us and all Europe.’

What Portland called the ‘interposition of providence’ was intelligence received on the evening of 14 February 1696 that, in addition to the planned French invasion in support of a Jacobite rebellion, a well-prepared attempt would be made next day to assassinate the King as he returned by coach from a hunting expedition down a narrow street. Portland’s informant, Thomas Prendergast, had been chosen as one of a group of eight assassins led by Sir George Barclay, a Jacobite army officer at James’s court in exile who had travelled to London in disguise. Prendergast’s allotted role was to fire a musketoon repeatedly into the royal coach but, he told Portland, his conscience would not allow him to go ahead. With some difficulty, Portland persuaded William to postpone his hunt until 22 February. On the evening of the 21st, Prendergast went to Kensington Palace to report that the assassins had reassembled and intended to kill the King after the hunt next day. His report was confirmed by a double agent, Francis Delarue. Prendergast was shown into the presence of the King and for the first time revealed the names of his Jacobite associates. When the royal hunt was again cancelled, rumours spread that a plot had been discovered and the would-be assassins fled. Though most, including Barclay, eventually escaped to France, eleven of those involved in the plot – some peripherally – were executed or imprisoned. Plans for a French invasion, which would probably have gone ahead if the assassination had succeeded, were cancelled. The exposure of the plot did great damage to the Jacobite cause and strengthened William’s popularity at a time of economic crisis. The Duke of Berwick, who also escaped to France, despite the offer of a £1,000 reward for his capture, later claimed that ‘he came over to stir up rebellion, but knew nothing of the assassination’. He admits in his Memoirs, however, that he was informed of a plot to ‘kidnap’ the King, which he did not countermand. Since no such plot existed, ‘kidnap’ may have been a euphemism for assassination. After his escape, Berwick never returned to Britain.

The death in exile of James II in 1701 was hastened by an English intelligence coup involving his former Secretary of State Lord Melfort, whom James had allowed back to Saint-Germain from provincial exile in 1697 and restored to favour as gentleman of the bedchamber. In 1701 a French messenger mistakenly delivered a letter written by Melfort to his brother Lord Perth to the English court at Whitehall instead. The letter contained secret details about Jacobite supporters in Scotland and discussed the possibility of a French invasion. Seizing this unexpected opportunity to embarrass both James and Louis, William ordered the publication of the letter. Versailles was so furious that it accused Melfort of having deliberately sent the letter to Whitehall in the hope of provoking another war between France and England. James was so shocked when told of the letter’s publication that he suffered a stroke from which he never fully recovered. At Louis XIV’s insistence, the unfortunate Melfort was once again banished to provincial exile.

Within a year, for reasons unconnected with Melfort’s letter, England and France were once again in conflict with the outbreak of the War of the Spanish Succession (1702–14), a further episode in the long-drawn-out struggle to contain the power of Louis XIV’s France that had dominated European politics for almost half a century. For much of the war the armies of England and its allies, under the leadership until 1711 of the Duke of Marlborough, the British commander-in-chief (‘captain-general’), triumphed over those of Louis XIV. The Battle of Blenheim in Bavaria in 1704 was both the first great English land victory on the Continent since the Hundred Years War and Louis XIV’s first decisive defeat. Marlborough was given the royal estate at Woodstock and a blank cheque to construct on it the great palace named after his victory. ‘For a time’, writes the historian Mark Kishlansky, ‘Marlborough was the most famous man in the world.’ He won further decisive victories at Ramillies in 1706, which enabled him to take possession of the Spanish Netherlands in a lightning campaign, and at Oudenarde in 1708, a battle which ended all prospect of a French invasion of the Dutch Republic.

Good intelligence, as well as Marlborough’s inspired generalship, contributed to Allied victories. Unlike British intelligence successes in the Nine Years War, however, those in the War of the Spanish Succession owed nothing to the monarch. Intelligence was beyond the mental horizons of William’s successor, Queen Anne. As her confidante, the Duchess of Marlborough, complained, Anne’s mind was so taken up with ‘ceremonies and customs of courts and such like insignificant trifles’ that her chief topics of conversation were ‘fashions and rules of precedence’. Though only thirty-seven when she became queen, Anne had to be carried to her coronation in a sedan chair, prematurely aged by seventeen pregnancies, all of which had ended tragically in miscarriages, still births or the birth of babies who died in infancy. Her pleasures were limited to dining and gambling.

The moving force in English intelligence during Anne’s reign (1702–14) was the Duke of Marlborough. The latest and best biography of Marlborough, by the military historian Richard Holmes, concludes that ‘The acquisition and analysis of intelligence underlay everything he did.’ When Marlborough assumed command of the Allied army in the Low Countries in April 1702, he appointed William (later General Earl) Cadogan his quartermaster-general; unofficially, he was also Marlborough’s intelligence chief – the first appointed by any English general. Cadogan was an excellent linguist, fluent in French, Dutch and German. While leading a reconnaissance expedition near Tournai in July 1706, he was taken prisoner by a French cavalry patrol. Aware of Marlborough’s high regard for him, the French chivalrously released him in a prisoner exchange. Lord Strafford claimed that the Dutch Grand Pensionary, Heinsius, had told him: ‘If you want to have a Duke of Marlborough, you need a Cadogan.’

As well as collating tactical intelligence from enemy prisoners and deserters, Cadogan also ran an agent network in France, much of it at the main French ports. The inducements given to his agents were not simply financial. Cadogan told Marlborough’s private secretary, Adam de Cardonnel, in 1705: ‘You will give me leave to remember my good friend the Conseiller Intime. I hope the Tokay and the lady are provided for him as promised.’ Cardonnel also ran an intelligence network through a former private secretary of William III, the Huguenot refugee Jean de Robethon, who had become private secretary and influential adviser to the Elector of Hanover, the future King George I. A few months before Blenheim, Robethon gave him a captured memorandum by Louis XIV’s Secretary of State for War, Michel de Chamillart, which summarized Louis’s instructions to his commanders. ‘We find . . . the utmost designs of the enemy in this memorial’, wrote Cardonnel, ‘and I hope we shall be able to traverse them.’ Marlborough thus knew the French campaign plan – that French commanders were to engage in battle only disunited Allied forces. Before Blenheim, by contrast, the French commander, Marshal Camille de Tallard, professed ‘total ignorance of the strength of the enemy’; he had no idea that the combined forces of Marlborough and his chief ally, the imperial commander-in-chief, Prince Eugène of Savoy (French-born but rejected by Louis XIV for service in the French army), were so close to him. Whereas Marlborough had devised his own campaign plan with the help of good intelligence, Tallard complained to Chamillart that a bad campaign plan had been imposed on him and that his Bavarian allies were impossible to deal with. He also blamed his own forces for his humiliating defeat: ‘The bulk of the cavalry did badly, I say very badly, for they did not break a single enemy squadron.’ Tallard spent the next eight years as prisoner-of-war in relative comfort in Nottingham, teaching his English captors how to make lace, bake proper bread, and grow celery, all skills hitherto unknown to them.

Following Marlborough’s victory at Ramillies and his unopposed entry into Brussels, previously under French control, in May 1706, he and the Dutch acquired a major new intelligence source. In December François Jaupain, who had taken charge of the Brussels post office, offered his services to Marlborough and the Earl of Sunderland, newly appointed Secretary of State. Jaupain gained control of all mail delivery between those parts of the Spanish Netherlands still under French or Bavarian control and Northern Europe. In the summer of 1707 and again in the spring of 1708, he joined Marlborough on his military campaigns, running an intelligence unit which collected information on enemy troop movements and provisioning. Jaupain’s main contribution to the Allied war effort, however, was to provide a regular flow of French intercepted messages to the youthful royal decipherer in England, William Blencowe, who had succeeded his grandfather John Wallis on Wallis’s death in 1703. Blencowe, then a twenty-year-old Fellow of All Souls College, Oxford, was paid the salary of £100 a year negotiated by Wallis two years earlier. He initially made slow progress in breaking French ciphers, partly because of the limited number of intercepts sent to him. On one occasion Marlborough sent him a captured despatch from Chamillart, the French Secretary of State for War, but Blencowe failed to decrypt it in time for Marlborough to make use of it. Blencowe’s ‘French Ministers Letter-book’ contains only three decrypts for the period up to 1706. Thereafter, thanks to the intercepts from Jaupain, their numbers greatly increased. There were also a smaller number of letters from the court of James III, ‘the Old Pretender’, son of James II, to Jacobites in Scotland. Ironically, some of the correspondence intercepted in Brussels contained warnings of the dangers of letter interception in London.

Blencowe was rewarded for his success in decrypting French despatches by the doubling of his salary to £200 a year. Queen Anne, no doubt at the prompting of her secretaries of state, also took his side in a dispute with his Oxford college. In 1709 Blencowe sought a dispensation to permit him to retain his fellowship at All Souls without, as was customary, taking holy orders. The Warden, Bernard Gardiner, refused and tried to force Blencowe either to take orders or to resign his Fellowship. Queen Anne’s intervention on Blencowe’s behalf led to the abolition of the Warden’s veto on dispensations. Blencowe’s precocious career as royal decipherer came to a tragically early end in 1712 at the age of only twenty-nine. During a bout of insanity following serious illness, Blencowe shot himself. His memorial in All Saints’ Church, Northampton, records his expertise in ‘the art of deciphering letters wherein he excelled, and served the public for ten years’.

Marlborough was so anxious to keep secret Blencowe’s decrypts of the French correspondence intercepted by Jaupain during the War of the Spanish Succession that he did not share them with his Dutch allies. He had good reason to doubt whether the Dutch could keep the secret. Following the death of the ‘Stadholder-King’, William III, day-to-day management of Dutch foreign policy had passed to Heinsius. Heinsius, in turn, had to answer to a supposedly secret committee of the States-General consisting of representatives of all the provinces, who on important matters had to consult their provincial States. ‘This meant, in effect’, writes the Dutch historian Karl de Leeuw, ‘that in the Dutch Republic more people were involved in the making of foreign policy than anywhere in the world and that it was extremely difficult to conduct any form of secret diplomacy.’

Marlborough was unaware that Jaupain was sending copies of French intercepts to Heinsius, as well as to himself, and that Heinsius was sending them on to the Elector George I’s black chamber in Hanover. From 1707 to 1711 he paid Hanover’s leading codebreaker, Ludwig Neubourg, 1,000 guilders a year. ‘This’, Neubourg told Heinsius in 1707, ‘will greatly enhance my enthusiasm.’ Like Heinsius, George I’s mother, Sophia of Hanover, heiress presumptive to Queen Anne under the 1701 English Act of Settlement, was an enthusiastic admirer of Neubourg’s codebreaking expertise, calling him ‘one of the wonders of the century’. George, however, changed his mind about Neubourg’s cooperation with the Dutch after the leak in The Hague in 1711 of a French diplomatic decrypt which personally embarrassed him. The decrypt revealed a scheme by the French court to cause dissension among the Allies by promoting the election of the Elector of Hanover as Holy Roman Emperor after the death of the Emperor Joseph I in April. During a visit to The Hague two months later, George’s influential personal secretary, Jean de Robethon, reported to him that rumours were circulating about the contents of the French decrypt: ‘It is out of the question that these rumours originate from the Grand Pensionary, but one cannot be sure of the deputies of the secret committee of the States-General, who get the intercepts to read as well, after our people [in Hanover] have decoded them.’ The Hanover black chamber ceased work for Heinsius soon afterwards. However, Heinsius’s private secretary, Abel Tasien d’Alonne, believed to be an illegitimate son of the Stadholder William II, had begun to set up a Dutch black chamber, which grew in importance over subsequent decades.

Blencowe’s decrypts provided Marlborough, on occasion, with valuable intelligence about his Dutch allies as well as his French enemies. Correspondence intercepted by Marlborough’s agents in 1709 revealed secret peace negotiations between Heinsius and Torcy, Louis XIV’s Secretary of State. Cadogan provided confirmation in May that Torcy had passed through Brussels, and sent Sunderland as well as Marlborough details of the peace terms which Torcy had secretly offered the Dutch. British pressure on Heinsius helped to ensure the failure of the Franco-Dutch negotiations. Faced with attempts by Louis XIV to break up the Grand Alliance by offering generous concessions to the Dutch and Germans, Britain publicly pledged to make no separate peace.

Cadogan ended 1709 with a strikingly confident intelligence assessment: ‘Great numbers of deserters come in daily, they are half starved and quite naked, and give such an account of the misery the French troops are in as could not be believed were it not confirmed by the reports and letters from all their garrison towns on the frontier.’ Cadogan was too optimistic. Unlike Marlborough’s previous triumphs at Blenheim, Ramillies and Oudenarde, the Battle of Malplaquet in September 1709 was a Pyrrhic victory which left him with double the casualties of the French. For the first time he received no congratulations from Queen Anne. Marshal Claude Louis Hector de Villars, the last great French general of the Louis XIV era, wrote to the King: ‘If God will have the goodness to lose us another battle like this one, Your Majesty can count on the destruction of all our enemies.’ In the aftermath of Malplaquet, a war-weary British government was ready for peace negotiations.

The transformation of the English government in London into a British government by the Act of Union in 1707 had owed much to a domestic intelligence operation run by Queen Anne’s future First Minister Robert Harley, then Northern Secretary of State, who a few years earlier had claimed that he ‘knew no more of Scotch business than of Japan’. During the negotiations which preceded the passage of the Act, Harley had sent secret agents to Scotland to report on, and seek to influence, Scottish opinion. Chief among them was the great writer, polemicist and unscrupulous businessman Daniel Defoe, whom Harley had already used as an agent in England. Defoe, who arrived in Edinburgh in September 1706, was later described by a contemporary as ‘a Spy amongst us, but not known to be such, otherways the Mob of Edinburgh had pulled him to pieces’. When the draft Act of Union was published in October, a ‘villainous and outragious mobb’ threatened members of the Scottish Parliament and judiciary. Despite the street protests, Defoe acquired sources in the Scottish Parliament, the Church of Scotland, and major business and civic groups, as well as gaining control of all the newspapers. Though Defoe publicly denied being a spy, writes the Cambridge literary scholar John Kerrigan, ‘he so liked to cut a dash in coffee houses that he couldn’t resist hinting at his role. This mixture of concealment and showing off is typical of the man . . .’ Economic ‘inducements’ certainly played a major part in winning a majority for the Union in the Scottish Parliament. Harley said cynically of the negotiations which won Scottish support for the Act of Union: ‘We bought them.’

Harley’s most dangerous encounter with intelligence work came in 1711, when, to quote the usually authoritative Oxford Dictionary of National Biography, he narrowly ‘survived an assassination attempt by a French spy, the marquis de Guiscard’. Antoine de Guiscard, Abbé de la Bourlie, marquis de Guiscard, was, in reality, a talented fantasist who spent more of his fraudulent career as an English than as a French agent. While living in Lausanne in 1704 Guiscard met and greatly impressed the English diplomat Richard Hill, who reported to the Earl of Nottingham, Secretary of State, that he ‘would engage to raise a revolt in Dauphine and Languedoc among the Catholics, if I would promise him such a protection and assurance as was absolutely necessary to begin the work. I liked the character of the man, and his temper so much, a man of figure, and family, very well known, that I promised him every thing.’ Guiscard claimed to have access to many French secrets and to carry with him a vial of poison in case he was caught by the agents of Louis XIV who were allegedly pursuing him.

Guiscard’s bizarre career is evidence of the vulnerability of Queen Anne’s governments (and, to a lesser extent, William’s) to intelligence fraud and the high-level access which a plausible impostor such as Guis-card could achieve. At the height of his influence, in 1706, he enjoyed the strong support of the Secretary for War, Henry St John, received 600 guineas from Queen Anne and a pension from the Dutch, and was given command of a regiment to land on the French coast and foment the uprising he had promised Hill two years earlier. The landing never took place and Guiscard’s credibility and income steadily declined over the next few years. In 1711 he sought to transform himself into a double agent working for Louis XIV, promising Torcy that, in so doing, he would ‘expiate his crimes towards [His Majesty] and towards his fatherland’. Though aware of Guiscard’s villainous reputation, Torcy was so impressed by his past high-level access to the British government that he sent an agent to England to make contact with him.

Guiscard’s correspondence, however, had been intercepted, and by the time Torcy’s agent arrived he was under arrest in London. On 8 March he was questioned by the members of the Cabinet (meeting without the presence of the Queen). Charged with treasonable correspondence with France, Guiscard at first denied it but was then confronted with his intercepted letters, which until that point had been hidden under a hat on the Cabinet table. He then lunged at Harley, head of the government elected in the previous year, and stabbed him with a penknife from his pocket. According to dramatic contemporary accounts, since many times repeated, though Harley was badly wounded, his life was saved by the heavy gold-thread embroidery lovingly sewn on his coat by his devoted sister Abigail, which broke the blade of the knife. Though the gold embroidery may have been a myth, the knife did break during the attack. A second attempt by Guiscard to stab Harley with the broken blade also failed to kill him, though Harley was confined to bed for the next six weeks. Harley’s attempted assassination by a French spy greatly increased his popularity among a public unaware of Guiscard’s earlier career as a fraudulent British agent. Guiscard died in Newgate Prison on 17 March from wounds sustained after his attack on Harley. Public interest in the assassination attempt was so great that the jailer pickled Guiscard’s corpse in a barrel, put it on display and charged a penny for admission. His remains were eventually interred at Newgate by royal command and the jailer, somewhat undeservedly, given £5 ‘to repair the damages done to the floor and ceilings of 2 rooms by the salt water that ran out of [Guiscard’s] cofin’. No double agent in British history has met a stranger end.

In June 1711 Harley, now Earl of Oxford, decided to begin secret negotiations with Louis XIV. On 12 July the poet and diplomat Matthew Prior was sent to France, travelling on a false passport in company with the French secret negotiator, Abbé François Gaultier, also travelling incognito. After ten days of negotiations with Torcy and an audience with Louis XIV, Prior returned to England with another French negotiator, Nicolas Mesnager. On landing in England he was briefly jailed by a customs official, whose suspicions may have been aroused by his false passport. The secret negotiations suddenly ceased to be secret. The Whig opposition to Harley’s Tory government derisively called the preliminary peace treaties signed on 8 October ‘Matt’s peace’.

Marlborough told Harley that ‘there is nothing on earth I wish more than an end to the war’, but disagreed with him on what constituted reasonable terms. His position was weakened by investigation of his military accounts, which uncovered a large shortfall. Marlborough maintained in his defence, not very persuasively, that he had personally received ‘no more than what has been allowed as a perquisite to the general, or commander-in-chief of the army’ and, rather more persuasively, that he had spent much of the money on ‘getting intelligence of the enemy’s motions and design’. On 30 December 1711 he was dismissed as commander-in-chief. His successor, the Duke of Ormond, was instructed to avoid engaging the enemy. On 11 April 1713 France, England, Holland, Portugal, Prussia and Savoy finally signed the Treaty of Utrecht, ending their involvement in the War of the Spanish Succession. The Emperor and his German allies continued for another year but achieved nothing of significance.

Obsessed by his increasingly unsuccessful quest for glory, Louis XIV ruled France less successfully at the end of his reign than at the beginning. In the 1660s, guided by Colbert, he showed a serious interest in understanding his account books and maintaining national solvency. After Colbert’s death he gave up. On his death in 1715 Louis left a bankrupt France with no effective accounting system and little more territory than at the beginning of his personal rule, despite the ruinous cost of his almost continuous wars in human lives as well as money. Half a century earlier, Colbert had had a broader vision of intelligence than any other European statesman of his time. In the Nine Years War, by contrast, William III made far better use of foreign intelligence than Louis or any of his ministers. During the War of the Spanish Succession, no French general matched Marlborough’s intelligence flair. Louis XIV’s grasp of intelligence, as of much else, did not compare with that of Cardinal Richelieu, who had died when he was only four. Under Richelieu, France had been the world leader in intelligence. It has never been so since.