French aspirations and Anglo-Saxon suspicions: France, signals intelligence and the UKUSA agreement at the dawn of the Cold War

Abstract

Although France ranked among the victors in 1945 and set about creating a world-class Sigint apparatus as she rose from her ashes, Paris was not invited to sign the 1946 UKUSA Agreement – nor did French leaders want to join the AUSCANUKUS community. Determining how this situation came about is not as straightforward as it seems. France had played a pivotal role in early peacetime Sigint cooperation, the breaking the Enigma and collaboration with Great Britain continued even after the fall of France in 1940. Nevertheless, post-war French ambitions and Anglo-Saxon suspicions prevailed at the dawn of the Cold War, resulting in a unique dynamic between these three powers.

Keywords:

• Signal Intelligence
• France
• United States
• Great Britain
• Second World War
• Cold War
• 1946 UKUSA Agreement

The year 2010 marked the long-awaited declassification of the British–US Communication Intelligence Agreement. ¹ Signed on 5 March 1946, and more commonly known as the UKUSA Agreement, this covenant grew out of the Second World War BRUSA Agreement of 1943 governing cooperation between these two countries in matters concerned with signals intelligence (SIGINT). The 1946 accord would expand over the years to include the dominions of Canada, Australia and New Zealand, a circle of trusted countries sometimes referred to as the ‘Five Eyes’ or AUSCANZUKUS, a community of interest and mutual support that would last through the Cold War to this day.

France has maintained a most ambiguous stance vis-à-vis these countries, most particularly Great Britain and the United States. Although de Gaulle's Free French participated in the defeat of the Axis and France recovered remarkably well in the immediate aftermath of the Second World War (gaining a seat at the United Nations Security Council, securing an occupation zone in Germany and recovering her colonial empire), Paris would not be part of the special relationship established between her two closest allies. Nor did she want to be. Accords such as the UKUSA Agreement involved dependencies and compromises over sovereignty that appeared as unacceptable to de Gaulle as to the governments that would succeed him under the Fourth Republic.

The literature is sparse regarding this conundrum as applied to intelligence sharing, especially in terms of SIGINT. It may appear to offer little room for exploration when reduced to this simplest form: the English-speaking powers did not trust the French and France did not want in. Nevertheless, it is of interest to determine how this situation came about as this process was not as straightforward as it seems. France had been a major intelligence power over the centuries. Her cryptographers had collaborated with those of Poland and Great Britain to break German codes in the late 1930s and such SIGINT cooperation continued even after the fall of France in 1940. This relationship did not survive the war although France, Britain and the United States became closely aligned in facing the communist threat at the dawn of the Cold War and greater trust between these three powers, at least in the secretive realm of SIGINT, could have generated positive synergies in a variety of ways.

Setting the pattern of reluctant cooperation

Early French superiority in the realm of cryptography was in evidence in the years leading to the First World War as the Bureau du Chiffre, under the Ministry of Foreign Affairs, routinely decrypted diplomatic traffic between Germany's Foreign Office in Berlin and its embassy in Paris. ² The French Army's 2e Bureau had also established intercept and direction finding stations along the border to monitor German wireless, developing an extensive knowledge of the peacetime communications of the Kaiserreichsheer (the Imperial German Army) through both traffic analysis and cryptanalysis. ³ This expertise, unique among the allied powers, proved most valuable during the initial stage of the war as the Germans advanced into Belgium and northern France, forcing them to rely increasingly on wireless as the Allies destroyed the local telegraph and telephone infrastructure as they withdrew. Although the French advantage did not deter the German offensive, it did contribute to stopping it during the ‘Miracle on the Marne’ and facilitated the draw that resulted from the following ‘Race to the Sea.’ ⁴

This initial phase of the war was marked by an unprecedented level of cooperation between French and British forces. This resulted from secret military staff talks that had followed the Entente cordiale of 1904 between Paris and London and the eventual rise of the Triple Entente in the wake of the Anglo–Russian Convention of 1907. ⁵ Although such discussions included some elements of coordination in the realm of communications, namely the promulgation of an English–French codebook that became effective in 1913, no measures were taken to actively exchange military intelligence, while SIGINT did not even exist as a concept at the time. ⁶ Nevertheless, the hostilities gave rise to some initial collaboration as the French seemed intent on sharing early successes in decrypting German military and naval codes, as well as those of Austro-Hungary, with England's newly formed Room 40. Such favours appeared to have been rarely returned by the British codebreakers who did not deign leveraging French expertise: ‘Our Allies were regarded as untrustworthy, or at least liable to indiscretions. In fact both the French and the Italians had broken an Austrian code and sent a copy to the Admiralty where it lay untouched for two years!’ ⁷

This disdain by the British was unwarranted. While the success of Room 40 in decrypting German naval codes remains recognized to this day, one must also appreciate that French cryptographers continued to dominate the field on the Western Front, regularly breaking the Kaiserreichsheer's ciphers while denying such success to the Germans. ⁸ This advantage continued throughout the war and contributed in part to the defeat of Ludendorff's offensives in the spring of 1918, including that in front of Compiège in June when Captain George Painvin succeeded in breaking the most recent ADFGVX code. ⁹ Victory, however, bred smug confidence. Although the Foreign Ministry's Bureau du Chiffre and the General Staff's 2e Bureau continued monitoring German communications through the Armistice and into the 1920s, the number and quality of dedicated cryptanalysts rapidly decreased in France while any inclination towards cooperation with former allies in the field of SIGINT vanished. ¹⁰

The situation was dramatically different in defeated Germany. Both the Foreign and War ministries maintained large cryptographic services in Berlin. Focusing on interception and cryptanalysis, the ‘Cipher Bureau of the Army High Command’ gained practical experience monitoring military communications on its eastern border through the Polish–Soviet War of 1919–1921 and the continued fighting within the borders of the former imperial Russia until the mid 1920s. ¹¹ An inquiry into the weakness of German codes and ciphers during the war also led to a momentous decision in 1925. Convinced of the value of providing military and government communications with the utmost security, Berlin adopted a militarized version of the commercially manufactured ENIGMA machine to encrypt radio transmissions. Within a year, German communications became protected ‘with the most secure system of cryptography in the world,’ a system that would only be defeated through a collaborative effort involving several countries – one of them, France, playing a pivotal role. ¹²

Fragile alliance against ENIGMA

Although German communications gradually became unreadable as ENIGMA machines were distributed to larger segments of the armed services and other branches of government after 1926, this development did not seem to raise as much concern in the West as one could have expected. Cryptographic services in France and Great Britain eventually realized that the garbled traffic being intercepted was the result of encryption through a complex rotor machine. Both quickly abandoned their desultory deciphering efforts as it was widely believed that such a mechanism was unbreakable. ¹³ One nation could not afford such a cavalier attitude. Poland, her borders threatened on all sides and wary of a resurgent Germany, immediately set about cracking the ENIGMA. The Polish Cipher Bureau – Biuro Szyfró – had experienced much success from its earlier days, contributing decisively to defeating the Soviets in 1919–1921 but its cryptanalysts were initially as baffled by ENIGMA as those of the other powers. ¹⁴ This situation was turned around in two stages. First, the Poles decidedly altered their approach to cryptanalysis from one rooted in languages to one based on mathematics. As a result, the Biuro Szyfró approached the University of Poznan in 1929 to discuss cryptography with a select group of students and formally hired three of them in September 1932. ¹⁵

These mathematicians made some important advances in figuring out ENIGMA settings and operating procedures but they were still well short of breaking enciphered intercepts. The next big break instead came through the more traditional means of espionage and treason. In June 1931, Hans-Thilo Schmidt, a civilian employee at the Reischswehr Chiffrierstelle (the German cryptographic agency), approached the French Embassy in Berlin and offered to exchange secret documents for money. After some months spent checking his background and evaluating this potential agent, the French Secret Service commenced taking regular delivery of packages through a handler codenamed Rex. ¹⁶ These documents were eventually relayed to the 2e Bureau's Major Gustave Bertrand who quickly realized their value as they related to ENIGMA. ¹⁷ However, neither his French colleagues nor British authorities, once they were allowed to view this material the following spring, could put them to use so minimal was their understanding of the German cipher machine.

Bertrand then lobbied his superiors to transfer these documents to the Polish Cipher Bureau, which was authorized in December 1932. Cryptographs at the Biuro Szyfró immediately realized the outstanding potential of such files as they included instructions on the use and keying of the ENIGMA as well as key tables for the previous months of September and October. These manuals and past settings allowed Poland to make considerable headway in breaking the ENIGMA and it was estimated that, by January 1938, Warsaw could decipher up to 75% of German intercepts within days, if not hours. ¹⁸ Such successes, however, came to an abrupt halt in September of that year when German authorities increased the machine's security features by amending enciphering procedures as well as delivering two additional wheels from which operators could choose to set their machine by. This greatly increased the number of possible keys and placed the decryption effort well beyond the Polish capability. ¹⁹ To further complicate matters, the traitor Schmidt broke contact with Rex in 1938, leaving both French and Polish authorities in the dark. ²⁰

This episode demonstrates both the extent and limits of cooperation between France and friendly powers in the 1930s, and reveals some of the traits that would mark the field of SIGINT cooperation in the future. First, Major Bertrand was not authorized to share the material obtained from Schmidt until it was realized that French cryptologists could not use it themselves. Only then did he approach Great Britain, France's closest ally at the time, and, again, it was only the failure of British code breakers to exploit that material that resulted in his approach to Poland. While the cooperation with Great Britain's Government Code and Cypher School (GC&CS) occurred rather informally, the approach to the Biuro Szyfró took place within the formal framework of the Franco–Polish alliance of 1921, a defensive pact aimed at containing Germany and deterring Soviet expansion in Central Europe. ²¹ As the Poles had already expressed their interest in material related to ENIGMA, the French eventually passed them such copies, which allowed them to attack that specific system in an early example of specialization between allied cryptographic services. ²²

The limits of such cooperation, though, were also demonstrated as Poland would not reveal the real extent of her success against ENIGMA until cornered by recent setbacks and the growing German menace. Poland had relayed to France intelligence summaries based on ENIGMA decrypts through the 1932–1938 period but without indicating the source of such information. As late as January 1939, on the occasion of a meeting between British, French and Polish cryptanalysts, the latter did not reveal that they had been reading German communications until recent months. ²³ Nevertheless, as Hitler renounced the German–Polish Non-Aggression Act of 1934 while making growing claims about extraterritorial rights on a corridor to Danzig, Poland's General Staff authorized the Biuro Szyfró to disclose to France and Great Britain the full extent of its knowledge of ENIGMA in a desperate bid for assistance to regain the upper hand against Germany's codes. French and British representatives were invited to Warsaw in July 1939. Not only were they briefed on the Poles' achievements over the last decade but also provided with replicas of the ENIGMA and blueprints for the ‘bombes’ required to simulate the machine's innumerable key settings, essential tools given the mechanization of cryptography. ²⁴

As French and British code breakers set about exploiting these products, the German blitzkrieg swept across Poland and key Biuro Szyfró personnel escaped the onslaught through Romania, eventually finding refuge in France. Integrated in the French wartime cryptographic unit established south of Paris under the designator Poste de commandement Bruno, they made an important contribution to French success in breaking German codes within months of the start of the war. Nevertheless, as PC Bruno was first evacuated to southern France following the collapse of June 1940 – to a new facility codenamed PC Cadix – and then to Great Britain in the wake of the German occupation of Vichy territory in November 1942, the Poles would not be granted access to the British GC&CS establishment at Bletchley Park for the remainder of the war. ²⁵ Thus, gave way the fragile alliance of French spies and Polish mathematicians who had made such a pivotal contribution to cracking ENIGMA as a new relationship rose to challenge France's rapport with her closest allies through World War II and into the Cold War.

Rise of the special relationship

While the British government had formally authorized more extensive intelligence exchanges with France in April 1939, leading to renewed success against ENIGMA before the German offensive in the West, one would have expected such links to be severed after the armistice of June 1940. ²⁶ It was not entirely the case. Although the defeated Vichy regime had agreed to dismantle the General Staff's 2e Bureau and pledged not to conduct intelligence operations against Germany and Italy, the French were allowed to maintain a small radio monitoring service, the Groupement de communications radioélectriques (GCR), to track clandestine radios operated by the resistance and foreign saboteurs. In parallel to this effort, operators clandestinely monitored German communications and soon forwarded intercepts to PC Cadix, where Major Bertrand had relocated PC Bruno's team of French and Polish cryptanalysts. ²⁷ Resuming operations in October 1940, Bertrand was able to contact British authorities the following March to arrange for data to be forwarded from Vichy France to Bletchley Park until Germany moved into the zone libre in November 1942. ²⁸

One must not overestimate the impact of such limited cooperation. For the British, the fall of France meant that the United Kingdom and the dominions stood alone in the west. Winston Churchill, installed as Prime Minister on 10 May 1940, immediately cast an eye towards Washington. Occupying the American presidency since 1933, Franklin Delano Roosevelt had first focused on the economic crisis at home and followed a strictly isolationist policy abroad but he became concerned with the rise of fascism in Europe and in Japan during his second term. He sought to convince an American public largely concerned with domestic issues that these regimes constituted a threat to United States' interests through an ‘education process’ that can be traced back to the Quarantine Speech of 5 October 1937 when he enunciated America's moral commitment to the cause of the democracies. ²⁹ Following the collapse of France, the president announced his intent to rebuild the country's armed forces and readiness to extend material aid to countries resisting Axis aggression. ³⁰

Within months, Roosevelt engineered the Destroyers-for-Bases Agreement to transfer 50 destroyers to the Royal Navy in exchange for basing rights in British territories in the Western Hemisphere. ³¹ While seeking neither Congressional approval nor modification of the Neutrality Acts for this last initiative, the president dramatically increased the defense budget and introduced an unprecedented peacetime draft. ³² Following the presidential election of 1940, Roosevelt continued his education effort by challenging the United States to become the ‘arsenal of democracy’ as stated during one of his radio fireside chats and declaring his commitment to the ‘Four Freedoms.’ ³³ His attention then returned to more practical measures with the introduction of the Lend–Lease Act of 11 March 1941, the deployment of United States forces as a security guarantor to Greenland in April, and taking over British garrisoning duties in Iceland in July. ³⁴ These developments occurred as Roosevelt broadcast on 27 May his intention to declare a state of Unlimited National Emergency and extend the Neutrality Patrol further out into the Atlantic. ³⁵

Such public pronouncements led to the summit of 9–12 August 1941 when Churchill and Roosevelt met on board warships at anchor in Placentia Bay, Newfoundland. These talks resulted in a joint proclamation by the United States and Britain declaring their opposition to the Axis powers and their common vision of a post-war world based on common principles such as personal freedom and democracy, the renunciation of force for territorial gains and the implementation of a permanent system of general security, as well as the right to self-determination and access to markets on equal terms. ³⁶ The declaration, soon dubbed the Atlantic Charter, seemed to point to an unprecedented relationship between great powers, one based on close political, diplomatic, cultural and historical roots. At the time, though, it fell well short of the formal US declaration of war the British leader was hoping for as Great Britain was still facing formidable odds. Commonwealth troops were confronting the Germans and Italians in North Africa and in the Mediterranean after having lost the Balkans and Crete, allied convoys were barely getting through in the Atlantic, Japanese forces seemed ready to take the offensive in the Pacific, and the Red Army had been reeling back ever since the launch of the Operation ‘Barbarossa’ in June. ³⁷

Short of Roosevelt declaring war on the Axis publicly, secret military cooperation between the two powers was already well underway. While few formal coordination mechanisms between the United States and the European allies had been established during the First World War, senior military leaders maintained some level of contact once peace returned, especially those naval officers who participated in the disarmament conferences of the inter-war period. It was also obvious to them that, regardless of the isolationism prevailing in the United States, another global conflict would result in renewed cooperation, at least in the maritime domain. As American planners reviewed war plans in the face of Japanese expansion in China, US Navy Captain Ingersoll was dispatched to London in December 1937 ‘for private and “purely exploratory” conversations at the Admiralty upon those prospects of cooperation which the new (war plan) was definitely to take into consideration.’ ³⁸ A wider delegation of representatives from the US services visited London in August 1940 to participate in the innocuously titled ‘Anglo–American Standardization of Arms Committee’ to discuss both armaments and much larger issues of strategy. ³⁹ These discussions led both sides to realize the importance of more formal and extensive talks. Following preparations throughout the autumn, military representatives secretly met in Washington for the first American–British Conversations (ABC) from 29 January to 27 March 1941. ⁴⁰

These talks proved pivotal in that they laid out the military foundations of the combined effort that would see the two powers united by the special relationship through the war. The secret proceedings were promulgated as the ABC-1 Plan, giving the defeat of Germany top priority and accepting a defensive posture against Japan until the situation in Europe stabilized and allowed for the transfer of resources to the Pacific. For the purpose of this article, it must be noted that the requirement for continuous and extensive liaison was also affirmed as the parties agreed to establish missions in each other's capital cities while the United States proposed to exchange with Canada, Australia, and New Zealand such liaison officers as would be necessary – planting the seed of the AUSCANZUKUS scheme. ⁴¹ The prescience of these secret talks became evident after Pearl Harbor, when Roosevelt and Churchill formally endorsed such plans during their first wartime conference (code named Arcadia) hosted in Washington from 22 December 1941 to 14 January 1942. ⁴² However, concerned with grand strategy as such fora were, they shed little light on the role of intelligence in supporting these schemes and whether the special relationship would be extended to the handling of SIGINT.

SIGINT and the special relationship

By the time Churchill arrived in Washington for the Arcadia Conference, Great Britain was well ahead of the United States in terms of SIGINT. Given the opportunity to leverage the work conducted by the Poles and the French against ENIGMA and collecting the necessary personnel and technical resources at Bletchley Park to conduct cryptography on an industrial scale during two years of concentrated effort in war time, Great Britain's GC&CS had established a commanding lead in the field. ⁴³ The Americans were still hobbled, for their part, by peacetime bureaucratic in-fighting and petty jealousies between a wide variety of government agencies involved in cryptographic work, especially the Navy's OP-20-G and the US Army Signals Intelligence Service (SIS). ⁴⁴ Such a bureaucratic approach was also adopted to assess the proposal put forward by an American representative in Great Britain to initiate the ‘technical exchange of information cryptography’ in the fall of 1940, with both services reacting cautiously at first. ⁴⁵ After several weeks of a three-way negotiation between GC&CS, OP-20-G and SIS, as well as their respective political masters, four American cryptologists visited Bletchley Park in February 1941.

This marked the first official exchange of information, including documents and equipment, whereby the British commenced revealing the extent of their success against ENIGMA while the Americans reciprocated with their knowledge of Japanese codes and enciphering machines. ⁴⁶ Such revelations remained guarded, however, especially on the British side as GC&CS was concerned that the Americans, still at peace, would let out knowledge of the ENIGMA breakthrough. Nevertheless, negotiations continued about greater cooperation, particularly between the respective navies as the USN acquired increasing responsibilities in the Atlantic throughout 1941. As the ABC-1 staff talks had led to the creation of military missions in each other's capitals, it was agreed in April to include intelligence representatives. ⁴⁷ It remained that the exchange of information was limited in these formative months, even once the United States joined the ranks of the belligerents after Pearl Harbor. Whitehall exercised increasing pressure on Washington through 1942 to further refine the mechanism in place and develop a greater specialization between the various agencies involved, but faced much reluctance from a growing American Sigint establishment facing worldwide responsibilities. ⁴⁸

Some of that reluctance also resulted from GC&CS's continued attempt to retain control over the generation of Ultra, as the intelligence obtained from decrypted high-level signals was codenamed. This rear-guard action was bound to fail, though, since Great Britain required American material and technical support to continue breaking the ENIGMA as the Germans kept increasing the machine's complexity. While OP-20-G quickly appropriated the funds necessary to acquire 360 of its own bombes in September 1942, there were still only 49 such devices in operation at Bletchley Park by January 1943. ⁴⁹ Grudgingly, another delegation of officials proceeded to Washington to negotiate the end of the British monopoly in that domain. GC&CS and OP-20-G endorsed the Holden Agreement of 2 October 1942 that stipulated how the two agencies would establish full cooperation against the naval ENIGMA. ⁵⁰ Given the disjointed nature of the American SIGINT apparatus, however, recriminations continued between British and US Army authorities as the Holden Agreement had only been concluded with the agency representing the US Navy. Another round of negotiations was required, this time in Great Britain, leading to the BRUSA Agreement of 17 May 1943. ⁵¹

This last accord was particularly important in that it clearly illustrated the special relationship that existed between the US and Great Britain, as applied to SIGINT, after two years of cooperation and rivalry, collaboration and defiance. By then based on extensive wartime practice in interception and decryption, it adroitly drew the line between competing interests and commonality of purpose. The agreement covered ‘the production, exchange and dissemination of all special intelligence derived by cryptanalysis of the communications of … the Axis powers, including their secret services’ but excluded ‘non-service enemy or neutral sources.’ It provided for ‘complete interchange of technical data and special intelligence from the sources covered … and for the dissemination of such intelligence to all field commanders through special channels.’ It also assigned specific responsibility to the United States for Japanese traffic and to the United Kingdom for that of Germany and Italy. ⁵² In effect for the remainder of the conflict, the Holden and BRUSA agreements laid out the fundamental SIGINT security procedures and sharing protocols that last to this day. It remained to be seen how such an entente would respond to the resurgence of a prospective ally rising from the ashes in the latter stage of the war.

French SIGINT under De Gaulle

As remarked earlier, the French SIGINT effort, supported by those cryptanalysts who had escaped Poland in 1939, continued in Vichy France and even succeeded in cooperating with Bletchley Park until the dissolution of PC Cadix in November 1942. Meanwhile, as the regime of Philippe Pétain fell into collaborationist disrepute, an increasingly strident voice made itself heard in London, that of a little known general claiming to speak for France. Dispatched to Great Britain as the Blitzkrieg rolled into his homeland and the newly installed government set about negotiating an armistice with the invader, Brigadier-General Charles de Gaulle broadcast his famous radio address to the French nation from London on 18 June 1940. Denouncing the prospect of a humiliating settlement with Germany, he called for resistance to continue alongside the Allies. ⁵³ He then set about building up armed forces, the Forces françaises libres (FFL, or Free French Forces), not as a legion of volunteers that would fight under British orders but rather as national services dedicated to a sovereign state claiming a defined territory – the overseas possessions – in preparation for the liberation of metropolitan France. ⁵⁴

Throughout the summer of 1940, de Gaulle laid the foundations of the FFL, which incorporated an army, a navy and an air force as well as a secret service, the Bureau central de renseignements et d'action (BCRA, or Central Bureau for Intelligence and Action) under a former military engineer, André Dewavrin. ⁵⁵ The BCRA incorporated a division tasked with coding and decoding messages under a non-commissioned officer, George Lecot, who established close links with the GC&CS as well as the SOE, the Special Operations Executive charged with espionage and sabotage behind enemy lines. ⁵⁶ This division was re-designated as the Section du Chiffre in October 1941, still under Lecot who had been promoted to Captain by then. ⁵⁷ The section's mandate was focused on supporting resistance operations in occupied France, but the BCRA was not involved in ENIGMA work nor was there any regular contact with PC Cadix in Vichy France. ⁵⁸ Following the allied landings in French North Africa in November 1942, de Gaulle tasked Colonel Jean Joubert des Ouches with setting up a central cipher and cryptographic department in Algiers, which in turn became the Direction technique des Chiffres (DTC, or Cipher Technical Directorate) in August 1943. ⁵⁹ The DTC was then placed under the Direction générale des services spéciaux (DGSS, the Special Services General Directorate), formed in Algiers that November to amalgamate the secret services of the FFL (the BCRA) and those of Vichy's Armée d'Afrique. ⁶⁰

De Gaulle himself officially became head of state on 3 June 1944 when the Allies endorsed the creation of the Gouvernement provisoire de la République française in preparation for the Normandy landings and the march on Paris, where de Gaulle installed the government on 30 August. ⁶¹ This, in turn, required another reorganization of the secret services in order to integrate personnel from the resistance in metropolitan France as well as those elements that had remained loyal to Vichy in the colonies and other dependencies overseas. The DGSS thus became the Direction générale des études et recherches (DGER) on 6 November 1944, which encompassed the Direction du contrôle technique (DCT), responsible for ciphers and radios. ⁶² The DGER, however, was quickly discredited in France and among the Allies. Expanding too quickly to more than 10,000 personnel and reflecting the disparity of political factions that made up the resistance, the service was divided within months by the same cleavages that affected French society at large. The DCT was suspected of monitoring domestic telephone communications for political purposes rather than contributing to the war effort, while the larger DGER pursued an overzealous hunt against suspected collaborationists. ⁶³

Particularly concerned with the penetration of the DGER by communist elements, de Gaulle ordered the creation of the Service de documentation extérieure et de contre-espionnage (SDECE), directly subordinated to him in his capacity as head of the Gouvernement provisoire. Laying the foundations of France's post-war intelligence establishment, the ministerial decree of 28 December 1945 led to the eventual abolition of the DGER and the creation within SDECE of two branches concerned with SIGINT. ⁶⁴ Service 26 (Service des matériels techniques) was charged with the protection of communications between headquarters and stations overseas while Service 28 (STR or Service technique de recherche) sought to intercept and decipher foreign diplomatic traffic. ⁶⁵ All other SIGINT collection and processing functions – those not related to the SDECE's own communications or diplomatic exchanges – were eventually assigned to a new organization named after a former Vichy agency, the Groupement de communications radioélectriques (GCR). Staffed by a mix of former military personnel and civilian professionals, the GCR rapidly developed a world-wide network of radio intercept and direction-finding stations located in metropolitan France, Corsica, Algeria, Morocco, Tunisia, Senegal, Polynesia and Indochina while the SDECE also maintained a monitoring post in the French sector of West Berlin. ⁶⁶

In addition to the SDECE and GCR, the military services and the civilian agency charged with domestic counterintelligence duties, the Direction de la surveillance du territoire (DST), also established cryptographic services to sustain their particular needs. ⁶⁷ Thus, one can surmise that de Gaulle's Gouvernement provisoire was seized of the importance of SIGINT as it oversaw the formation of such an extensive establishment, a trend which continued despite dire budgetary circumstances after the general abandoned power in January 1946. ⁶⁸ Although hobbled by political and bureaucratic rivalries, all components of France's post-war SIGINT capabilities were in place by the middle of that year, including stations and personnel located throughout the métropole and her vast empire, capable of effectively discharging all operations involved in the collection and exploitation of radio traffic at home and abroad, as well as protecting her own communications. Nevertheless, this vast potential would not be considered by the United States and Great Britain as they were about to set up a new level of cooperation in this field, unprecedented in peacetime.

The 1946 UKUSA Agreement

Flushed with victories in Europe and in the Pacific, not yet committed to a path of military confrontation with the USSR, both Britain and the United States embarked on a dramatic demobilization effort through the fall of 1945 and into the following year, a course which also affected their respective intelligence services. Nevertheless, dedicated to the spirit of the Atlantic Charter, it was understood that Washington would not be able to withdraw from world affairs as it had done after the Great War and that intelligence, particularly SIGINT, would continue to play an important role. Forged in the fires of the Second World War, the pattern of British–American cooperation was likely to continue into peacetime. Its precise shape, though, was yet to be determined as illustrated when President Truman signed an order in September 1945 authorizing the Secretaries of War and the Navy to ‘continue collaboration in the field of communications intelligence between the United States Army and Navy and the British, and to extend, modify or discontinue this collaboration, as determined to be in the best interest of the United States.’ ⁶⁹

These instructions laid the foundations for the UKUSA Agreement of 5 March 1946. Following several months of extensive discussions, delegates of the State–Army–Navy Communications Intelligence Board (representing the US State, Navy and War departments) and the London Signal Intelligence Board (representing the British Foreign Office, Admiralty, War Office and Air Ministry) signed this top secret cooperation mechanism. ⁷⁰ Governing ‘the relations of the above-mentioned parties in Communication Intelligence matters’ ⁷¹ and dictating that it would ‘be contrary to this agreement to reveal its existence to any third party whatever,’ ⁷² the document confirmed that the parties agreed ‘to the exchange of the products of the following operations relating to foreign communications:

1.	collection of traffic
2.	acquisition of communication documents and equipment
3.	traffic analysis
4.	cryptanalysis
5.	decryption and translation
6.	acquisition of information regarding communication organizations, practices, procedures, and equipment.' 73

These provisions reflected the Second World War experience, building up on the BRUSA Agreement of 1943, in encompassing all dimensions of SIGINT that required sharing, not only raw intercepts and decrypted traffic but equipment, technical and procedural elements as well. ⁷⁴ The agreement further stated that foreign communications referred to those of ‘any country, whether or not its government is recognized by the US or the British Empire, excluding only the US, the British Commonwealth of Nations, and the British Empire.’ ⁷⁵ This last excerpt very much carried the political sensitivity of the accord, whereby Washington and London promised not to monitor each other's communications but declared that those of everybody else, including friends and allies such as France, were legitimate targets. Although eventually expanded to include Canada in 1948, as well as Australia and New Zealand in 1956, the agreement remained most secret and exclusive regardless of the partner nations joining a variety of other military and intelligence alliances in the early stages of the Cold War. ⁷⁶

There are no contemporary sources that would lead one to believe that the signatories contemplated extending membership to France or that Paris seriously considered approaching London and Washington to be an active participant. This did not prevent the formulation of bilateral relations between French SIGINT agencies and those of the United States (such as the Armed Forces Security Agency and its successor, the National Security Agency) ⁷⁷ and with Great Britain's Government Communications Headquarters (GCHQ), as GC&CS was renamed in June 1946. ⁷⁸ Those exchanges were quite limited in their nature such that they never involved the transparency underlining the UKUSA Agreement. France and her allies did exchange intelligence derived from communications intercepts but the information would always be limited to what one wanted to divulge and when one would choose to do so. Notwithstanding the value of these ‘unwritten Comint-sharing agreements’ in the words of one historian, their use and interpretation always required a degree of further analysis and, sometimes, skepticism. ⁷⁹

As outlined earlier, France possessed SIGINT assets that one could have deemed of interest to her Anglo-Saxon allies. A former intelligence practitioner turned scholar states that two important reasons for countries to collaborate in this field are the sheer amount of collected information that can easily overwhelm any given agency, and local access, which played a pivotal role in the ability to intercept signals, at least until the arrival of the satellite. ⁸⁰ France dedicated extensive personnel and monetary resources in the immediate aftermath of the Second World War in order to reconstitute a first-class SIGINT establishment in the métropole while her colonial empire allowed setting up intercept stations and supply bases for those ships, aircraft and submarines required to collect intelligence close to the borders and the coastline of any given target of interest. However, these apparent advantages did not fill any specific gap in the UKUSA Agreement as her allies had moved so far ahead of the French agencies in terms of analysis techniques and sheer computing capabilities. Further, the location of France's remaining colonies, other than Indochina perhaps, did not really add to the geographic ‘containment’ of the Soviet Union and China given the combined global coverage of American dependencies and the British Commonwealth. ⁸¹

Beyond these practical concerns, however, it remained that the American and British intelligence authorities could not develop with the French the sense of mutual professional respect and trust they had cultivated among themselves through the war and its immediate aftermath. Although far from harmonious on several occasions, the UKUSA effort grew increasingly close and ever more centralized through these years, especially in the handling of Ultra, which remained out of reach to the Free French. ⁸² Meanwhile, the pre-war French establishment of professional cryptographers and its contribution during the conflict was severely undermined by the divisions that plagued France during the German occupation while the traumatic birth of a Fourth Republic deeply fractured the nation between past Vichy supporters and partisans of de Gaulle, while the conservative right faced an increasingly militant left. These concerns actually led American forces deployed in France to discreetly but actively monitor local communications during the liberation and such effort likely continued in peacetime as the French Parti communiste gathered an increasing following under the Fourth Republic. ⁸³

Beyond these Anglo-Saxon suspicions, however, one must also recognize that French aspirations prevented her adherence to the overly integrated UKUSA Agreement. De Gaulle and his successors set about returning France to prominence on the world stage in a new policy of grandeur through economic development and the promotion of a strong and independent foreign policy. ⁸⁴ They understood that this process would involve some degree of dependency vis-à-vis their allies, as with the Marshall Plan and rebuilding the armed forces using American equipment. ⁸⁵ Nonetheless, they would promote an independent path whenever they could. In the field of SIGINT, this translated into the reconstitution of an indigenous, self-sufficient and world-wide infrastructure at great cost to the national treasury while leveraging informal sharing agreements regarding bilateral intelligence matters when convenient. But joining the Five Eye community would be a step too far, requiring unacceptable sacrifices in terms of autonomy and sovereignty.

Conclusion

De Gaulle's accomplishments by the end of the Second World War were remarkable. He had succeeded in gaining for France a seat at the side of the victors despite the catastrophic defeat of 1940 and the small military contribution made by the FFL to victory in 1945. Although the French Parti communiste was one of the strongest in Western Europe at the end of the conflict, with representatives in government as well as agents in the intelligence apparatus, by 1947 these had been banished from the cabinet and the secret services purged. Indeed, the one element common to the disparate branches of the country's complex national security and intelligence establishment in the wake of the abolition of the DGER was ‘the fact that they were uniformly anticommunist,’ while France firmly established herself in the anti-Soviet camp at the dawn of the Cold War, joining NATO in 1949 and dispatching troops to Korea in 1950. ⁸⁶

Nevertheless, Paris was not invited nor wished to join the UKUSA Agreement negotiated in 1946 between London and Washington. The accord ‘represented a crucial moment in the development of the “special relationship” between the two wartime allies and captured the spirit and practice of the signals intelligence co-operation' which had evolved during the war. ⁸⁷ It gave rise to a unique peacetime alliance dedicated to the sharing of the two powers’ worldwide SIGINT resources against the growing Soviet threat. And yet, France stood aside in a ‘splendid isolation’ strangely reminiscent of that of an earlier Albion perfide. While reflecting de Gaulle's aspirations of grandeur, this stance also exposed one of the many burdens that followed the conflict:

All Allied intelligence services emerged decisively altered from World War II.

But nowhere were the mutations more apparent than in France, where war had bequeathed several legacies to the secret services. The change most obvious was that, on an international level, France was no longer an intelligence power. She had stood on the sidelines of the enormous growth in electronic and scientific intelligence during the war. ⁸⁸

Although the resources France dedicated to cryptography waxed and waned until the Second World War, her cryptanalysts had sought to maintain a commanding position in that field. From the Great War to the cracking of ENIGMA before the fall of France in 1940, they kept abreast of, and often led, the evolution of an intuitive art based in languages to the mechanization of a scientific process resting on mathematics. Even under the collaborationist Vichy regime, PC Cadix continued monitoring Axis communications and contributed to the Allied effort at Bletchley Park. It is ironic that, as de Gaulle's FFL grew in stature in the latter stage of the war, he was never given access to Ultra intelligence and his cryptographic services came to be largely excluded from the larger allied effort.

This situation reached a nadir in 1946 as the UKUSA Agreement was negotiated. Distrust in Washington and London resulted from the widespread belief that both the French government and its intelligence services under the early Fourth Republic were penetrated by communists. France's leaders, for their part, aspired to regain for their country a position of prominence on the international scene. This led them to pursue an independent course of action within the larger Atlantic partnership, requiring self-reliance in intelligence matters while maintaining tailored bilateral relationships when appropriate. This very particular combination of French aspirations and Anglo-Saxon suspicions negated the potential for positive synergies in the realm of SIGINT and resulted in dramatically contrasting dynamics between the three Western victors at the dawn of the Cold War.

Notes

1. United States, National Security Agency, NSA Press Release – Declassified UKUSA Signals Intelligence Agreement Documents, http://www.nsa.gov/public_info/press_room/2010/ukusa.shtml (accessed June 24, 2010).

2. Christopher Andrew, ‘Déchiffrement et diplomatie: le cabinet noir du Quai d'Orsay sous la IIIe République,’ Relations internationales 3, no. 5 (1976): 37–64.

3. Wilhem F. Flicke, War Secrets in the Ether (Laguna Hills, CA: Aegean Park Press, 1994), 2–3; Simon Singh, The Code Book – The Secret History of Codes and Codebreaking (London: Fourth Estate, 2000), 106.

4. On the role of signals intelligence on the Western Front in August and September 1914, see Flicke, War Secrets in the Ether, 16–9; John Ferris, ‘The British Army and Signals Intelligence in the Field during the First World War,’ Intelligence and National Security 3, no. 4 (1988), 23–48.

5. For a short but thorough discussion of these agreements, see Henry Kissinger, Diplomacy (New York: Touchstone, 1994), 187–93.

6. David Khan, The Codebreakers – The Story of Secret Writing (New York: MacMillan, 1968), 264.

7. Quoted by Patrick Beasley in Room 40: British Naval Intelligence, 1914–1918 (Oxford: Oxford University Press, 1984), 80 (note 1); see also Khan, The Codebreakers, 277–8.

8. For an extensive discussion of French interception techniques in the field and decryption efforts at higher headquarters, see Khan, The Codebreakers, 299–313. For a rare insight into German code breaking, see Hilmar-Detlef Brückner, ‘Germany's First Cryptanalysis on the Western Front: Decrypting British and French Naval Ciphers in World War I,’ Cryptologia 29, no. 1 (2005): 1–22.

9. Khan, The Codebreakers, 339–47; Sophie de Lastours, La France gagne la guerre des codes secrets, 1914–1918 (Paris: Tallandier, 1998), 241–58.

10. Singh, The Code Book, 143.

11. Flicke, War Secrets, 65–72.

12. Singh, The Code Book, 142; Kahn, The Codebreakers, 420–22; and Friedrich L. Bauer, ‘Rotor Machines and Bombes,’ in The History of Information Security – A Comprehensive Handbook, ed. Karl de Leeuw and Jan Bergstra (Amsterdam: Elsevier, 2007), 381–406.

13. Singh, The Code Book, 143; and John Ferris, ‘The Road to Bletchley Park: The British Experience with Signals Intelligence, 1892–1945,’ Intelligence and National Security 17, no. 1 (2002): 71–5.

14. The Cipher Bureau did not actually take that name until 1931 as a result of the amalgamation under the Army's General Staff of several sections concerned with various aspects of cryptography. The term Biuro Szyfró is used in this article for the whole inter-war period for clarity. On the early successes of Polish cryptographers against the Soviet Union, see Richard Woytak, ‘Colonel Kowalewski and the Origins of Polish Code Breaking and Communication Interception,’ East European Quarterly 21, no. 4 (1987): 497–500.

15. They were Jerzy Róźycki, Henryk Zygalski and Marian Rejewski. While all three played an important role in breaking ENIGMA, the latter was generally recognized as the most gifted. Singh, The Code Book, 149; Bernard Budiansky, Battle of Wits – The Complete Story of Codebreaking in World War II (New York: Touchstone, 2002), 74–76.

16. Singh, The Code Book, 144–5; James Gannon, Stealing Secrets, Telling Lies – How Spies and Codebreakers Helped Shape the Twentieth Century (Dulles VA: Brassey's, 2001), 37.

17. Gustave Bertrand had fought as a private during the First World War. Promoted to officer, he remained in the Army and joined military intelligence in 1926. He began work on German codes in 1930, remaining closely involved with the ENIGMA file right up to an including the Second World War. His memoirs provide a unique insight into the history of Franco–Polish collaboration in breaking the ENIGMA. See Gustave Bertrand, ENIGMA ou la plus grande énigme de la guerre 1939–1945 (Paris: Plon, 1973).

18. F.H. Hinsley et al., British Intelligence in the Second World War – Its Influence on Strategy and Operations, vol. 3, part 2 (London: Her Majesty's Stationary Office, 1988), 949.

19. Singh, The Code Book, 157–8; Gannon, Stealing Secrets, Telling Lies, 40–1.

20. Singh, The Code Book, 158.

21. Although dated, Piotr Stefan Wandycz offers an excellent treatment of the origins of the alliance in France and her Eastern Allies, 1919–1925: French–Czechoslovak–Polish Relations from the Paris Peace Conference to Locarno (Minneapolis: University of Minnesota Press, 1962), 211–38.

22. Singh, The Code Book, 146.

23. Budiansky, Battle of Wits, 93–4; Gannon, Stealing Secrets, Telling Lies, 42.

24. Singh, The Code Book, 159–60; Gannon, Stealing Secrets, Telling Lies, 42–3; Budiansky, Battle of Wits, 94–6. On the Polish ‘bomba,’ later known as ‘bombe’ in both England and France, see Bauer, ‘Rotor Machines and Bombes,’ 420–1.

25. Bauer, ‘Rotor Machines and Bombes,’ 444–5; Budianski, Battle of Wits, 120–1, 276–9; Władysław Kozaczuk, ENIGMA: How the German Machine Cipher Was Broken, and How It Was Read by the Allies in World War Two, ed. Christopher Kasparek (Frederick MD: University Publications of America, 1984), 111–55, 205–9.

26. On the authorization of greater intelligence exchanges between London and Paris, see F. H. Hinsley et al., British Intelligence in the Second World War – Its Influence on Strategy and Operations, vol. 1 (New York: Cambridge University Press, 1979), 488. Exchanges between PC Bruno and Bletchley Park through the winter of 1939–1940 became so extensive as to require the installation of a teletype link between these two locals to exchange keys and decrypts. Budiansky, Battle of Wits, 134.

27. Roger Faligot, ‘France, SIGINT and the Cold War,’ Intelligence and National Security 16, no. 1 (2001): 179–80.

28. Budianski, Battle of Wits, 146. Bertrand's PC Cadix was set up in Uzès, a small commune near the Mediterranean coast. Sources diverge on the exact value and extent of the exchanges between PC Cadix and Bletchley Park but such communications did take place until the evacuation of 9 November 1942.

29. The Quarantine Speech was delivered in Chicago in the wake of the invasion of China by Japan and also addressed the Berlin–Rome Axis announced the preceding year. On Roosevelt's so-called education process through this period, see Kissinger, Diplomacy, 379–88.

30. This policy watershed was announced in Charlottesville, Virginia on 10 June 1940 in what became known as the ‘Stab-in-the-Back Speech’ denouncing Italy's attack against France. Ibid., 386.

31. The agreement was signed on 2 September 1940. For an extensive treatment, see Stetson Conn et al., Guarding the United States and Its Outposts (Washington: Center of Military History, 2000), 354–83.

32. George Q. Flynn, The Draft, 1940–1973 (Lawrence: University Press of Kansas, 1993), 11–13.

33. The ‘arsenal for democracy’ radio address took place on 29 December 1940 and the Four Freedoms (freedoms of speech and worships as well as freedoms from want and fear) appeared in the State of the Union Address delivered on 6 January 1941. Kissinger, Diplomacy, 388–9.

34. The Lend–Lease Act effectively circumvented the Neutrality Acts by allowing belligerents to borrow war supplies against the promise or repayment after the war. John Keegan, The Second World War (New York: Viking, 1989), 112. The occupation of Greenland and Iceland are covered extensively in Stetson, Guarding the United States, 442–58, 459–531.

35. For a detailed treatment of the context and shaping of the May 27 announcement, see Eric Larrabee, Commander in Chief: Franklin Delano Roosevelt, His Lieutenants, and Their War (New York: Touchstone, 1988), 40–63. The Neutrality Patrol was initiated by Roosevelt in September 1939 to track and report the movements of belligerents in the waters of the Western Hemisphere. Initially focused on the Caribbean Basin, the zone grew to encompass Greenland, Iceland and the Azores. For a complete treatment, see Brian F. Hussey, ‘The US Navy, the Neutrality Patrol, and Atlantic Fleet Escort Operations, 1939–1941,’ Trident Scholar Report Project No. 180 (Annapolis, MA: United States Naval Academy, 1991).

36. NATO, ‘The Atlantic Charter’ – Declaration of Principles Issued by the President of theUnited States and the Prime Minister of the United Kingdom (14 August 1941), http://www.nato.int/cps/en/SID-41933363-48EBE02D/natolive/official_texts_16912.htm (accessed July 17, 2011).

37. Barbarossa was the German code name for the invasion of the Soviet Union launched on 22 June 1941. For a succinct assessment of the British position on the eve of the Atlantic Conference and Churchill's stated goals, see David Reynolds, ‘The Atlantic “Flop”: British Foreign Policy and the Churchill–Roosevelt Meeting of August 1941,’ in The Atlantic Charter, ed. Douglas Brinkley and David R. Facey-Crowther (New York: St Martin's Press, 1994), 129–50.

38. Mark Skinner Watson, Chief of Staff: Prewar Plans and Preparations (Washington: Center of Military History, 1991), 92–93. This visit was reciprocated when Royal Navy Commander T. C. Hampton proceeded to Washington in June 1939 to hold similar ‘conversations’ with USN senior staff. See Ritchie Ovendale, The English-speaking Alliance: Britain, the United States, the Dominions and the Cold War 1945–51 (London: George Allen and Unwin, 1985), 5.

39. Watson, Chief of Staff, 113–14.

40. On the policy and strategy discussions that preceded the conference throughout the fall of 1940, see ibid., 115–23, 370–3.

41. Ibid., 378.

42. For a detailed account of military aspects of the First Washington Conference, see Maurice Matloff and Edwin M. Snell, Strategic Planning for Coalition Warfare 1941–1942 (Washington: Center for Military History, 1990), 97–119.

43. On the success of Bletchley Park in cracking the ENIGMA during the 1939–94 period, see Singh, The Code Book, 160–89.

44. OP-20-G stood for the ‘Office of the Chief of Naval Operations, 20th Division of the Office of Naval Communications, G Section/Communications Security.’ OP-20-G was mainly focused on the interception and decryption of the Imperial Japanese Navy's communications while SIS dedicated the bulk of its efforts to breaking Japan and Germany's diplomatic traffic in the years leading to the war. On OP-20-G and SIS during the inter-war period, see Frederick D. Parker, Pearl Harbor Revisited: United States Navy Communications Intelligence 1924–1941 (Fort Mead, MA: National Security Agency, 1994), 2–9. On the bureaucratic infighting, see ibid., 12–5.

45. The proposal came from US General George Strong, Chief of Army Intelligence, who was involved in the previously mentioned Anglo–American Standardization of Arms Committee. Budianski, Battle of Wits, 174.

46. Ibid., 175–9; Hinsley et al., British Intelligence (vol. 1), 312.

47. Hinsley et al., British Intelligence (vol. 1), 313–4. The UK mission in the American capital first took the name of ‘Advisers to the British Council’ and later became the British Joint Staff Mission in Washington, eventually supporting the Combined Chiefs of Staff after the latter body was formed in 1942.

48. British urgency resulted in part from the major setback suffered by Bletchley Park when the German Kriegsmarine added a fourth rotor to the ENIGMA machine while its own decryption service, the Beobachtungs dienst or B-dienst, made an important breakthrough into British naval codes. The Allies were, thus, unable to break German naval intercepts from February to December 1942 while Donitz's staff could read 80% of convoying instructions, often 20 to 30 hours in advance. Budianski; Battle of Wits, 234–5; John Keegan, Intelligence in War – Knowledge of the Enemy from Napoleon to Al-Qaeda (Toronto: Key Porter Books, 2003), 270–2.

49. Budianski, Battle of Wits, 238–9.

50. The Holden Agreement also made official the appointment of full-time US liaison officers at Bletchley Park as well as a range of coordination issues regarding both German and Japanese naval traffic. Hinsley et al., British Intelligence (vol. 3, part 2), 56–7; Budianski, Battle of Wits, 238–9; Ralph Erskine, ‘The Holden Agreement on Naval SIGINT: The First BRUSA?,’ Intelligence and National Security 14, no. 2 (1999): 187–97.

51. Budianski, Battle of Wits, 296–9; and Hinsley et al., British Intelligence (vol. 3, part 2), 57–8.

52. Quotes are from the text of the original BRUSA Agreement, which can be found under a cover letter dated 10 June 1943 at National Security Agency, Agreement between British Government Code and Cipher School and US War Department in Regard to Certain ‘Special Intelligence’ (10 June 1943), http://www.nsa.gov/public_info/_files/ukusa/spec_int_10jun43.pdf (accessed July 19, 2010).

53. For an objective summary of the circumstances surrounding the Franco–German Armistice of 22 June 1940 and the growing divide between Pétain and de Gaulle at the time, see Philippe Masson, Histoire de l'armée française de 1914 à nos jours (Paris: Perrin, 1999), 256–9.

54. On the political and legal concerns that fashioned de Gaulle's plan to build up the FFL to serve a sovereign French state, implying the illegitimacy of the Vichy regime, see Éric Roussel, Charles de Gaulle (Paris: Gallimard, 2002), 154–67.

55. BCRA is used for clarity sake but the service actually changed name several times. It was created on 1 July 1940 as the 2e Bureau, then became the Service de renseignements on 15 April 1941 and the Bureau central de renseignements et d'action militaire in January 1942 before taking the BCRA acronym on 1 September of that same year. For a full treatment, see Sébastien Albertelli, Les services secrets du général de Gaulle, le BCRA, 1940–1944 (Paris: Perrin, 2009). On the creation of the FFL's army, navy and air force, see Masson, Histoire de l'armée française, 293–99.

56. Faligot, ‘France, SIGINT and the Cold War,’ 180.

57. François Broche, ‘Les secrets du BCRA dévoilés,’ Espoir – Revue de la Fondation Charles de Gaulle 158 (automne 2009), http://www.charles-de-gaulle.org/pages/revue-espoir/articles-comptes-rendus-et-chroniques/les-secrets-du-bcra-devoiles.php (accessed August 3, 2011).

58. Martin Thomas, ‘Signals Intelligence and Vichy France, 1940–44: Intelligence in Defeat,’ Intelligence and National Security 14, no. 1 (1999), 194.

59. Faligot, ‘France, SIGINT and the Cold War,’ 181.

60. On the creation of the DGSS, see Broche, ‘Les secrets du BCRA dévoilés’; Thomas, ‘Signals Intelligence,’ 195.

61. On the period from the formation of the Gouvernement provisoire to its installation in Paris, see Roussel, Charles de Gaulle, 423–51.

62. Faligot, ‘France, SIGINT and the Cold War,’ 182.

63. Roger Faligot and Pascal Krop, La Piscine: The French Secret Service since 1944, translated by W. D. Halls (Oxford: Basil Blackwell, 1989), 10–26; Douglas Porch, The French Secret Services: From the Dreyfus Affair to the Gulf War (London: Macmillan, 1996), 267–74.

64. The DGER and SDECE co-existed for several months of acrimonious bureaucratic competition but the former was officially disbanded in April 1946. Porch, The French Secret Services, 267. The SDECE was re-organized and re-named in April 1982 as the DGSE, the Direction générale de la sécurité extérieure, which remains in existence to this day. Ibid., 434.

65. Faligot and Krop, La Piscine, 39–40; and Faligot, ‘France, SIGINT and the Cold War,’ 183.

66. The GCR was reactivated through a provisional government decree on 15 March 1946. This agency would eventually be placed under the direct authority of the Prime Minister under the Fourth Republic, which constitution was adopted following the referendum of 13 October 1946. Faligot, ‘France, SIGINT and the Cold War,’ 184.

67. Ibid., 186.

68. De Gaulle resigned as head of the Gouvernement provisoire in disagreement with his political allies over budgetary allocations for the armed forces and the proposed constitution that would provide, in his view, insufficient powers to the President under the Fourth Republic. Roussel, De Gaulle, 557–76.

69. Cited by Christopher Andrew in ‘Intelligence Collaboration Between Britain, the United States, and the Commonwealth during World War II,’ in The Intelligence Revolution – A Historical Perspective: Proceedings of the Thirteenth Military History Symposium, US Air Force Academy, ed. Walter T. Hitchcock (Washington: Office of Air Force History, 1991), 119.

70. The text of the agreement can be found in full at United States, National Security Agency, British–US Communication Intelligence Agreement (5 March 1946), http://www.nsa.gov/public_info/_files/ukusa/agreement_outline_5mar46.pdf (accessed July 3, 2010).

71. Ibid., 3.

72. Ibid., 6.

73. Ibid., 4.

74. The 1946 accord also superseded all previous agreement such as the 1943 BRUSA. Ibid., 8.

75. Ibid., 4.

76. The particular status of Canada, already referred to in the 1946 agreement, was considerably expanded in Appendix H (COMINT Communications) reviewed during the US–British Technical Conference that took place in July 1948. The text can be found in full at United States, National Security Agency, Appendices to US–British Communication Intelligence Agreement (15–26 July 1948), http://www.nsa.gov/public_info/_files/ukusa/appendices_jul48.pdf (accessed July 13, 2010). The text of the 1956 agreement, confirming the status of Australia and New Zealand, can be found at United States, National Security Agency, UK–US Communications Intelligence Agreement (10 October 1956), http://www.nsa.gov/public_info/_files/ukusa/new_ukusa_agree_10may55.pdf (accessed July 13, 2010).

77. The AFSA brought together the cryptologic activities of the Army, Navy, and Air Force in June 1949, activities which were subsumed by the NSA in October 1952. Thomas L. Burns, The Origins of the National Security Agency, 1930–1952 (Fort Mead, MA: Center for Cryptologic History, 1990), 59–80.

78. The British retained the wartime philosophy of one central, civilian agency supported by the armed forces to handle SIGINT in establishing GCHQ. Michael Herman, Intelligence Power in Peace and War (Cambridge: Cambridge University Press, 1996), 24.

79. Faligot, ‘France, SIGINT and the Cold War,’ 189. The author mentions several cases of such cooperation in Indochina, Algeria and Vietnam, at least until Washington ceased all cooperation with France on SIGINT in 1964, as a result of de Gaulle's increasingly confrontational stand and fears of infiltration of the French apparatus by the Soviets. Ibid., 193. See also Porch, The French Secret Services, 292. The latter reports that de Gaulle, not the Americans, ordered the cessation of exchanges between the SDECE and the NSA in 1964. In either case, sources agree that such exchanges were not re-established until the early 1970s after de Gaulle had left power.

80. Herman, Intelligence Power, 204–7.

81. On the limits of the French geographic factor, see Michael Herman, ‘Intelligence Effects on the Cold War,’ in Did Intelligence Matter in the Cold War, ed. Michael Herman, J. Kenneth McDonald and Vojtech Mastny (Oslo: Norwegian Institute for Defence Studies, 2006), 12–3.

82. ‘Even when US analysts advocated “close technical co-operation between US and French Sigint organizations,” they stressed that “ULTRA, of course, will be taboo”… Bletchley's cipher experts discovered significant flaws in the main Free French enciphering machine, the B-211, and the cryptanalysts breaking ENIGMA believed the Germans could read its ciphers.’ Quoted in R. A. Ratcliff, Delusions of Intelligence – ENIGMA, ULTRA, and the End of Secure Ciphers (Cambridge: Cambridge University Press, 2006), 111.

83. On the American effort to monitor French communications in 1944, see Faligot, ‘France, SIGINT and the Cold War,’ 182. The Parti communiste leveraged its record of wartime resistance to garner one third of the seat in the first post-war National Assembly, a legitimacy that entitled it to ministerial positions under the Fourth Republic until 1947, when it broke with government over Indochina and the nascent Cold War. On this phenomenon, see Philippe Buton, Les lendemains qui déchantent, le parti communiste français à la Libération (Paris: Presses de la fondation nationale des sciences politiques, 1993), 269–87.

84. On de Gaulle's vision at the end of the Second World War, and the tensions it created in the allied camp, see the dated but still relevant Milton Viorst, Hostile Allies: FDR and Charles de Gaulle (New York: Macmillan, 1965), 183–219; Sean Greenwood, The Alternative Alliance: Anglo–French Relations Before the Coming of NATO, 1944–48 (Montreux: Minerva Press, 1996), 34–49.

85. Masson, Histoire de l'armée française, 337–9.

86. Porch, The French Secret Services, 274.

87. Richard Norton-Taylor, ‘Not So Secret: Deal at the Heart of UK–US Intelligence,’ The Guardian, 25 June 2010, http://www.guardian.co.uk/world/2010/jun/25/intelligence-deal-uk-us-released (accessed July 3, 2010).

88. Porch, The French Secret Services, 265.