Woodrow Wilson had declared that America entered World War I to make the world safe for democracy. The effort had not succeeded. Four great empires had collapsed in Central and Eastern Europe, creating a revolutionary battleground for many disparate and competing political ideologies. Novelist John Dos Passos later described the year 1919 as a time for "machinegunfire and arson, starvation, lice, cholera, and typhus."1 The peace process went awry. The victors at Versailles saddled the new republican government of Germany with great territorial losses and reparations payments while allowing it to remain the strongest and most industrialized power on the European Continent. Meanwhile, the growing problems posed by the widening revolution in Russia went unaddressed, and animosities developed rapidly among the victors. Wilson had hoped that the peace settlement, however flawed, would be redeemed by the establishment of a collective security organization, the League of Nations. But his own Senate refused to ratify the measure as America retreated toward isolationism.
Within the United States, there was also unrest as the bonds that had held the nation together during the war began to dissolve. Motivated by wartime idealism, pundits had spoken vaguely of a postwar era of "reconstruction" that would lay the basis for an "industrial democracy" in which the American economic cornucopia would be shared more equitably by all.2 Instead, the end of the war brought unemployment, inflation, and labor unrest. Alarmed by revolution abroad and radical labor agitators at home, the country was swept by a "Red Scare" leading to thousands of arrests. In the end, a fatigued American people, weary of wartime idealism and stress, voted for Warren G. Harding and a return to "normalcy."
Military Intelligence was intimately involved in these events, both at home and abroad. Twenty MID officers accompanied President Woodrow Wilson and
42
the American delegation to the Paris Peace Conference. Sixty CIP agents, directed by Van Deman, provided security for the American party, and Maj. Herbert O. Yardley of MI-8 furnished cryptologic support.3 Meanwhile, Army intelligence personnel accompanied the new Third Army the American occupation force that marched into the Rhineland under provisions of the armistice. On the domestic front, Bolshevik agents replaced German spies as the focus for MID's counterintelligence efforts. In the fall of 1919 a naive MID officer warned that "the situation in the United States [was] . . . verging on revolution."4 However exaggerated the estimate, it accurately reflected the fears of many Americans who found the world changing too fast for America ever to recover her lost innocence.
Military Intelligence at Peace
Normalcy meant a return to a peacetime Army, a process that had begun almost as soon as the fighting ended. The vast conscript armies that had won the war were hastily demobilized, and a new volunteer force enlisted. Subsequently, the National Defense Act of 1920 provided for a Regular Army of 280,000, backed up by Organized Reserves and a 475,000-man National Guard. The command and control of this force would be exercised through nine corps areas in the continental United States and through three overseas departments in Hawaii, the Philippines, and Panama. Under the terms of this legislation, MID retained its place as one of the four principal divisions within the Army Staff. Among its numerous duties were administering the attache system, supervising military drawings and maps, writing regulations for tactical intelligence personnel, performing liaison with other intelligence agencies, approving codes and ciphers, and planning censorship operations. As one of the four assistants to the chief of staff authorized by law, MID's director was given the rank of brigadier general.
Unfortunately for the Army, the economy-minded Congress never provided the necessary appropriations to maintain the authorized force. By 1929 the Regular Army had thus shrunk to a strength of 137,000, with four skeleton combat divisions in the continental United States and three more in the overseas departments. That part of the force assigned to the continental United States was scattered over a multitude of tiny posts inherited from the Indian Wars. The subsequent Great Depression made Congress even more parsimonious, forcing the Army to make do with the large but increasingly obsolescent stockpile of weapons left over from World War I. The Military Intelligence Division was directly affected by these policies, dwindling from a peak strength of 80 officers
43
and about 160 civilians in 1920 to a cadre of 20 officers and fewer than 50 civilians by 1934, and undergoing repeated internal realignments.5
The reorganization of the Army Staff under General John J. Pershing also weakened Military Intelligence. After becoming chief of staff in 1921, Pershing imposed the AEFs wartime model of a five-part General Staff on the Army, directing that separate divisions be established for personnel, intelligence, operations and training, supply, and war plans. However, congressional legislation had provided for only four brigadier generals on the staff, and in the savage competition for general officer slots the intelligence staff often lost. Of the seven directors of the Military Intelligence Division who served between 1922 and 1939, only two were brigadier generals.6 Military Intelligence thus became a second-class citizen, since its lesser position on the Army Staff was reflected throughout the rest of the Army Implicitly, U.S. military leaders seem to have accepted the dictum of British Field Marshal Sir Douglas Haig that "intelligence is rather a special kind of work and has a very small place in the army in peacetime."7
This trend was reinforced by Pershing's insistence that G-2 undertake public relations duties as it had done in the AEF Under peacetime conditions, these responsibilities became one of the Military Intelligence Division's principal functions. Intelligence officers at corps level and below also spent much of their time in this secondary area. As a result, intelligence work often became a dumping ground for officers incapable of performing any more demanding activities, and astute officers regarded intelligence assignments as detrimental to their military careers. According to intelligence historian Thomas Troy, "intelligence was neither a profession [n] or a career; at best, it was a one-time activity in an army or navy officer's service. Hence, when closely scrutinized, the intelligence services [were] small, weak stepchildren of their parent organizations. "8
The weakness and lack of professionalism of Army Intelligence during this period was reinforced by the decline in its capabilities for intelligence collection. The attache system remained MID's principal means for collecting foreign intelligence, irregularly supplemented by arrangements made with American businessmen working abroad. General George Marshall would write later that interwar military intelligence was "little more than what a military attache could
44
learn . . _ at a dinner, more or less, over the coffee cups. "9 Even here the limitations were severe. Postwar economies forced the abandonment of a number of minor attache posts, and the scarcity of funds dictated that attache assignments be restricted to officers with private means of support, irrespective of their professional qualifications. Finally, the interwar years saw the rise of totalitarian governments and controlled and closed societies. Military attaches were allowed increasingly less freedom to collect useful intelligence in precisely those countries that posed the most dangerous threat to American security. 10
Army Intelligence in the domestic arena was equally weak. This was partially the result of a general public reaction to some of the excessive actions of the Wilson administration against dissidents, both in World War I and during the postwar "Red Scare." Fortunately the impoverished peacetime volunteer Army had little to worry about in the way of threats posed by espionage, sabotage, and subversion. Thus the Army discontinued its countersubversive system in 1920 and recalled all MID regulations on the subject from the field. With only six agents on duty, the Corps of Intelligence Police narrowly escaped extinction when wartime emergency legislation expired that same year, but MID managed to keep the organization alive by detailing personnel to CIP duties from the Army's Detached Enlisted Men's List. The Negative Branch of MID, which had supervised the War Department's counterintelligence work, was less fortunate and closed shop in 1921.
All attempts to collect domestic intelligence ended soon afterward. This was caused by the rash actions of Lt. W D. Long, an intelligence officer at Vancouver Barracks, who sent a circular letter to the county sheriffs of Oregon asking them to maintain a surveillance of suspect organizations. Included on the lieutenant's list were veterans' groups, the Nonpartisan League, and the American Federation of Labor. This list caused a considerable uproar when inevitably it was made public. The secretary of war ordered that all intelligence posts in the Army not authorized by tables of organization be discontinued, and that intelligence officers confine themselves to instructing troops in combat intelligence techniques. This left most of the forty-five CIP agents authorized in 1920 with little to do, and the force was repeatedly scaled back. By 1934 the Corps of Intelligence Police consisted of just sixteen noncommissioned officers, and a subsequent survey found that most of them were used as classified file clerks rather than as investigators. Only in the overseas departments and in the Eighth Corps Area on the Mexican border did the Corps of Intelligence Police still provide useful services. 11
45
If the Military Intelligence Division was collecting little foreign and less domestic intelligence over most of this period, it was not overly active in other areas. Although one section of MID was charged with intelligence training, it had no statutory responsibility to supervise such work, and the assistant chief of staff for operations and training refused to acknowledge its authority. As a result, in 1931 the head of the section was forced to confess that "the state and extent of combat intelligence training in the Army is not known to this branch, as it makes no inspections and receives no training reports."12
Another section of MID was charged with supervising the Military Intelligence Officers Reserve Corps (MIORC), which had been formed in 1921 to use the services of the large number of officers who had served in intelligence positions throughout the wartime Army. In theory this organization of reservists should have given the Army a trained and experienced mobilization base, but many reserve officers were in fact journalists who had served G-2 in public affairs positions, and by the time they were ultimately needed, they were too old and too high ranking to fill the positions the Army required. The younger reservists in MIORC had never served on active duty. Perhaps the only lasting contribution of MIORC was the adoption of the sphinx as its insignia, an action which had the effect of permanently associating this somewhat bizarre heraldic item with the field of Military Intelligence.13
Not everything was completely bleak in the intelligence field between the wars. Although hobbled by budgetary restraints, the Army began to take advantage of new technological developments that expanded the possibilities of intelligence collection. Advances in motorized transport significantly enhanced the Army's ground reconnaissance. By 1931 an Experimental Mechanized Force had been created-its scout element had armored cars and radio-equipped vehicles-and the cavalry began to field mechanized units in addition to its traditional horse troops. The Army Signal Corps experimented with methods of detecting aircraft and ships through thermal and electronic means, and by 1937 the Army had a pilot model of a mobile radar set. Although limited to shortrange targeting functions, the principles it embodied would apply to more powerful sets that could perform an early-warning function, filling a critical gap in Army capabilities after the increased speed of aircraft had rendered acoustical techniques almost useless. 14
The growth of the Army Air Corps also affected intelligence. Air Corps officers were incorporated into MID, and Air attaches supplemented the work of the regular military attaches abroad. At the technical level, aircraft development
46
was steady, since this was the one part of the Army that Congress was willing to fund. The Air Corps made significant advances in aerial photography, including night photography, and in aerial mapping techniques, all of which potentially enhanced intelligence gathering. Much of the pioneering work in this area was done by Lt. George Goddard, who would ultimately go on to become an Air Force brigadier general. 15
In practice, the technological advancement of the Army Air Corps was a two-edged sword. The observation squadrons of World War I had flown in flimsy biplanes, and these aircraft could take off from grassy strips and work in close support of tactical commanders. But the sophisticated high-speed monoplanes with ample "greenhouse" canopies which the Air Corps developed in the 1930s for observation work, otherwise advanced machines with greatly improved performance, no longer could operate from unimproved airstrips. Furthermore, aeronautical progress encouraged Air Corps leaders to aspire to an independent role. The Air Corps became increasingly interested in fighting an air war of its own, instead of providing ground support. Aerial reconnaissance was thus designed to support a strategic bombing campaign, not to assist the tactical commander on the battlefield. By 1935 most aircraft had been placed under the direct control of General Headquarters, Air Corps. Although observation groups were still assigned to Army corps areas, only eight observation squadrons were active, a trend that threatened to deprive Army commanders of adequate air intelligence support. 16
The "Black Chamber" and the Signal Intelligence Service
Army Intelligence made its greatest advances in the field of cryptology during the years between the wars. Following World War I, MI-8's previously comprehensive responsibilities in this field were realigned. The Signal Corps took over the responsibility for communications security, employing William F Friedman, a former AEF officer and Riverbank Laboratories cryptanalyst, as a one-man code compilation bureau. The Adjutant General's Office was tasked with printing and distributing the codes. Officially, MI-8's responsibilities were reduced to approving cryptosystems for Army-wide use and establishing regulations for their employment. In reality, however, MID continued to be deeply
47
involved in cryptanalysis. In the fall of 1919 retired Maj. Herbert O. Yardley, wartime chief of MI-8, set up a clandestine government cryptanalytic unit in a brownstone house in New York City. Jointly funded by MID and the State Department, Yardley's bureau continued to work on diplomatic code-breaking, a task that MI-8 had initiated in World War I. Using material provided secretly by some of the major U.S. cable companies, Yardley and his small civilian staff achieved several notable successes, the most important of which was breaking the Japanese diplomatic code in time to give American diplomats a key negotiating edge during the Washington Peace Conference of 1921-1922.17
Despite the success of Yardley's bureau in producing hard diplomatic intelligence, it still ran into difficulties. Peacetime economies sharply reduced its funding. By 1929 the organization consisted of Yardley himself and a handful of assistants, and its functions seemed increasingly less relevant to both its sponsors. The United States had just signed the Kellogg-Briand Pact purportedly outlawing war, the international order seemed stable, and the State Department saw little need for secret intelligence in a world without enemies. On the other hand, the Army had no direct peacetime interest in the diplomatic cryptosystems Yardley's group exploited, while the clandestine and civilian nature of his bureau prevented its serving as a training vehicle for Army Intelligence reservists. By the spring of 1929 there was already a move to centralize all Army cryptologic functions under the Signal Corps, using Friedman's office as a nucleus. When Henry L. Stimson, a rather excessively upright statesman of the old school, became secretary of state that year, Yardley's fate was sealed. Discovering that the State Department had obtained access to decoded diplomatic messages, Stimson withdrew funding from Yardley's bureau. His attitude was later described: "Gentlemen do not read each other's mail."18
Loss of funding led to the termination of Yardley's organization. Yardley and his colleagues were offered employment in the new Signal Corps cryptologic agency, but they declined. The Civil Service pay scale could not match salaries subsidized by MID confidential funds. Instead, Yardley attempted to recoup his fortunes by writing The American Black Chamber, which publicly exposed America's code-breaking activities for the first time. The work was a bestseller, but proved to be a major diplomatic embarrassment for the United States and only further damaged American intelligence efforts. Yardley would later work as a cryptanalyst for Chiang Kai-shek in China and for the Canadian government, but he would never again be allowed to hold a position in any U.S. cryptologic organization.
48
Upon the demise of Yardley's bureau, Army cryptanalysis and Army cryptography were both integrated into a new Signal Corps element, the Signal Intelligence Service (SIS), which also assumed responsibility for secret inks. The new unit was set up within the Signal Corps rather than the Military Intelligence Division to reduce its visibility and better meet the technical requirements of signals intelligence. After 1934, when it took over the printing and distribution of codes from the Adjutant General's Office, the SIS became the focal point for all Army cryptologic activity
Friedman, who headed the Signal Intelligence Service until an Army officer became director in 1935, quickly recruited a small but talented staff. The four principal members of his initial group-mathematicians Solomon Kullback, Frank Rowlett, and Abraham Sinkov and Japanese linguist John Hurt-would all become prominent figures in U.S. Army cryptology in World War II. At first the thrust of the work was oriented towards theory, training, and the development of advanced cryptographic systems. The Army had fought World War I using codes. In 1922 the Army adopted a simple cylindrical cipher device, the M94, for tactical operations.19 Friedman introduced a more secure but cumbersome strip cipher device, the M138. More importantly, he began developmental work on machine ciphers. The SIS provided the Army with the M134 and M134A "converters" to protect top-level communications. These were electromechanical cipher machines of great sophistication and security 2o Meanwhile, in 1936 Friedman and Rowlett hit upon the cryptographic principles of an even more advanced machine cipher device. Unfortunately, funding was not immediately available to put the prototype into production.
The Signal Intelligence Service also began to set up an intercept organization, which Yardley's bureau, with its dependence on cooperative cable companies, had never attempted. A Provisional Radio Intelligence Detachment was organized at Fort Monmouth, New Jersey, in 1933. We may discern something of the flavor of the times from the fact that its first commander found none of the twelve men in his unit present for duty when he assumed command-all had been detailed to President Franklin D. Roosevelt's Civilian Conservation Corps. This detachment was later expanded to become a provisional company, and the 1st Radio Intelligence Company was formally activated at Fort Monmouth in 1938. Meanwhile, in September 1936 the Signal Intelligence Service had set up a chain of numbered monitoring stations in the overseas departments and in the Eighth and Ninth Corps Areas, creating SIS
49
detachments in five signal service companies. Since the Federal Communication Act of 1934 had made it illegal to divulge foreign communications, this had to be accomplished under tight security.21
These initial arrangements for providing the SIS with an intercept capability were not completely satisfactory. Tasking the 1st Radio Intelligence Company with an operational assignment interfered with the unit's primary mission, training for deployment in the field to support tactical elements. In addition, the intermingling of intercept and regular communications personnel in the existing signal service companies not only posed a security threat, but also worked against effective personnel management. Under Army regulations, trained intercept personnel working abroad were automatically returned to the general Signal Corps pool when they returned to the continental United States and thus were lost to the SIS. To solve these problems, a centralized signals intelligence unit, the 2d Signal Service Company, was set up at Fort Monmouth on 1 January 1939 to control all Signal Corps personnel at the permanent monitoring installations. The result was an efficiently functioning Signal Intelligence Service and an intercept organization that would represent the Army's principal strength in the intelligence field.22
The SIS paid less attention to intelligence collection at the tactical level. Under the Army's Protective Mobilization Plan, a World War I-style general headquarters was to be fielded in the event of any crisis demanding troop mobilization, and available troops would be concentrated as needed under one of the four field armies into which the Army had been divided in 1933. The actual field army selected would depend on the direction from which the threat was expected. These plans also specified that the existing 1st (and only) Radio Intelligence Company be placed in support of GHQ. To provide signals intelligence at field army level, the Signal Corps relied on the National Guard to organize two additional radio intelligence companies, one on each coast.23 These modest preparations would soon be overtaken by the rush of events.
1 John Dos Passos, U.S.A.: Nineteen Nineteen (Cambridge: Houghton Mifflin, 1946), p. 281.
2 For example, see William Leavitt Stoddard, "The Shop Committee- Some Implications," The Dial 67 (12 July 1919): 7-8; "Reconstruction Miscellany," The Survey 42 (31 May 1919): 375.
3 Some of the CIP agents in Paris questioned the utility of their assignments. Apart from the men detailed to guard President Wilson, one wrote, they were employed as "a species of bell boys, ladies maids, and hallmen." Gilbert Elliott Ms, p. 7, RG 319, NARA.
4 Memo, Col C. H. Mason for Brig Gen Marlborough Churchill, 31 Oct 19, sub: Sinister Inertia in Present United States Situation, MID Documents.
5 From 1920 on, MID strength was affected by the provisions of the National Defense Act of 1920, which limited the number of General Staff officers on duty in Washington, D.C., to 93. This meant that MID would be allotted 14 to 16 General Staff officers. When the law took effect, Brig. Gen. Dennis Nolan, assistant chief of staff, G-2, hoped that 31 more officers could be detailed from the line to keep MID up to what he regarded as minimum strength. In this, he was disappointed. Memo, Nolan for Chief of Staff, 28 Jul 21, sub: Relief of Thirteen Line Officers on Duty with Military Intelligence Division To Date, MID Documents.
6 Memo for Chief of Staff, 19 Jun 37, sub: Increase in Number of Assistants to the Chief of Staff, MID Documents.
7 Maj. Gen. Sir Kenneth Strong, Men of Intelligence (New York: St. Martin's Press, 1972), p. 34.
8 Thomas F. Troy, Donovan and the CIA: A History of the Establishment of the Central Intelligence Agency (Frederick, Md.: University Publications of America, 1984), p. 6.
9 Ibid., p. 15.
10 One especially aggressive attache was Maj. Truman Smith. Posted to Nazi Germany, Smith obtained intelligence reports on the Luftwaffe from the American aviator and popular hero Charles Lindbergh, who repeatedly toured German aircraft factories as a guest of Reichsmarshal Hermann Goering and other Nazi top brass. Washington Post, 4 Nov 84, p. A-3.
11 Of the 16 CIP agents surveyed, 10 were acting as confidential clerks for G-2s in the various corps areas, 5 were engaged in investigative work, and 1 was unfit. History of the Counter Intelligence Corps, vol. 4, p. 87.
12 Bidwell, History of the Military Intelligence Division, p. 362.
13 Ltr with Incl, ACSI-DO, 25 Feb 63, Organization Day for Military Intelligence file, INSCOM History Office.
14 Radar developments in the Army up to 1941 are covered in Dulany Terrett, The Signal Corps: The Emergency, United States Army in World War II (Washington, D.C.: U.S. Army Center of Military History, 1956). For early experiments, see pp. 44, 46-47.
15 Goddard has provided a useful autobiography in Overview: A Lifelong Adventure in Aerial Photography (Garden City, N.Y.: Doubleday, 1969).
16 Robert F. Futrell, Command of Observation Aviation: A Study in Control of Tactical Airpower (Maxwell Air Force Base, Ala.: Air University, 1956), p. 2. It should be noted that balloons and dirigibles were in the Army inventory of reconnaissance platforms in the early 1920s. A short but helpful overview of Air Corps support to ground operations between the wars is provided by Weinert, A History of Aviation, Phase I, pp. 2-4. In Army Air Corps Airplanes and Observation, 1935-1941 (St. Louis: U.S. Army Aviation Systems Command, 1990), Howard Butler offers an alternative view to the standard interpretation of Air Corps history, arguing that the Air Corps remained firmly under the thumb of a ground Army-dominated General Staff during this period.
17 Yardley recounted- some say embellished- his accomplishments in his book, The American Black Chamber (New York: Bobbs-Merrill, 1931). A brief biography can be found in David Kahn, Kahn on Codes; Secrets of the New Cryptology (New York: MacMillan Co., 1983), pp. 62-71.
18 Kahn, The Codebreakers, p. 360. A detailed study of the rise and fall of the Yardley organization is contained in SRH 29, A Brief History of the Signal Intelligence Service, pp. 3-12.
19 The M94 was the brainchild of Joseph O. Mauborgne, an Army Signal Corps officer who later became a major general and Chief Signal Officer. It was an exact copy of a device invented by Thomas Jefferson.
20 The early history of the Signal Intelligence Service is described in SRH 131, Expansion of the Signal Intelligence Service. For the use of the term "converters," see SRH 349, Achievements of the Signal Security Agency in World War II, pp. 42-43.
21 The sensitivity of the whole topic of communications intelligence during this period may be inferred from a revealing piece of Signal Corps correspondence on the subject, which ended with the statement "It is suggested that this letter be burned after perusal." Ltr, Lt Col Dawson Olmstead, Executive Officer, Office of the Chief Signal Officer, to Col Joseph O. Mauborgne, 28 Jun 36, sub: Comment on Lieut. Corderman's Report, Army Cryptologic Records.
22 G. R. Thompson and Dixie R. Harris, The Signal Corps: The Outcome, United States Army in World War II (Washington, D.C.: U.S. Army Center of Military History, 1966), pp. 333-35.
23 Ltr, Maj Gen Joseph Mauborgne, Chief Signal Officer, to Chief, National Guard Bureau, 13 Sep 38, sub: Organization of National Guard Radio Intelligence Companies, Army Cryptologic Records.
page created 10 September 2001