Index

World War I:   COMINT in its Infancy 1917-1919

Next

When the First World War broke out in 1914, all the major combatants were using wireless. Unfortunately, the lack of really secure ciphers made wireless transmission risky. If intercepting telegraph traffic was simple, with wireless it was almost unavoidable. Messages were broadcast over the airwaves, and anybody could pick them up. In fact, before wireless, one of the problems in cryptanalysis was intercepting enough messages to permit reasonable analysis. After wireless, the problem was sorting through the flood of messages that poured in over the airwaves. Despite the lack of security, there was often no alternative to wireless, since it allowed governments to communicate with warships at sea and armies on the move.

On the first day of the war, a British ship dragged up Germany's transatlantic telegraph cables and cut them. From that time on, the Germans had to use radio links or telegrams sent through neutral nations, and both methods left them open to interception. As the Germans had advanced into Belgian and French territory where telegraph lines had been cut, it forced them to rely on wireless. The French and British, in contrast, only absolutely needed to use wireless to communicate with ships at sea.

At the outbreak of the war, Britain had no formal code breaking operation. The British Admiralty's intelligence service began to intercept German wireless messages and quickly recognized the need for a formal cryptanalysis organization. Volunteer code breakers were found in the country's naval colleges. The French actually had formal code breaking groups, five of them, in place before the war, but after the outbreak of hostilities the French built up their "Bureau de Chiffre" (Cipher Bureau) in particular. Both the British and French established radio listening and radio direction finding stations. The Germans, in contrast, did not become serious about such matters until 1916, when they set up their own radio-intercept organization, the "Abhorchdienst".

The U.S. had no formal code breaking organization although a few people were engaged in cipher work at Riverbank Lab in Geneva, Illinois. The lab was founded by George Fabyan, a millionaire cotton merchant and honorary Colonel. Originally founded as a genetic research lab, Fabyan also sponsored cipher work by a small staff trying to find and break ciphers supposedly used in Shakespere's plays. When war broke out in Europe Fabyan offered to put his cryptographers and laboratory at the disposal of the government. Riverbank became the "unofficial" cipher bureau until the MI-8 was established. William F. Friedman was part of the staff at Riverbank and began teaching cryptography classes there. In 1918 Friedman was commissioned and joined Pershing's AEF Headquarters where he worked on code breaking as part of the AEF Radio Intelligence Section.

When the U.S. declared war on April 16, 1917 it almost immediately took actions to strengthen its weak intelligence organization. On April 28th the Army Chief of Staff issued a memorandum directing the formation of a code and cipher decryption unit, called the Cable and Telegraph Section, or MI-8, within the Military Information Section (MIS). At this time the MIS was under the War Department General Staff (War College Division) headquartered in Washington D.C. While a gifted Regular Army officer, Major Ralph Van Deman, quickly revitalized the Military Information Section (MIS) , the MI-8 (Cipher Bureau) was organized under the lead of Herbert Osborne Yardley. MI-8 was assigned the dual mission of making and breaking codes. On February 9th 1918 the MIS was redesignated as the Military Intelligence Branch (MIB) and on August 26th 1918 it was redesignated as the Military Intelligence Division (MID). Van Deman's MID was America's first complete, national level intelligence agency. It used HUMINT, SIGINT, and, to a lesser degree, IMINT to collect information. It also had counterintelligence, mapping, and security agencies.

The Signal Corps began training radio intelligence enlisted members at Camp Alfred Vail (founded in 1918 and renamed Fort Monmouth in 1925). Almost all intercept/DF operators enroute to Europe received their training at the Signal Corps School.

While Ralph Van Deman was organizing a workable Army level intelligence unit back in Washington in 1917, our forces arriving in France were facing a while different set of challenges. General Pershing (Commander AEF) and his staff immediately recognized the need for a tactical intelligence system. MAJ Dennis E. Nolan was placed in charge of intelligence for the AEF. Using the French System, the first G2s/S2s were created in the fall of 1917. (Note: Todays INSCOM headquarters is located in the "Nolan" building in Fort Belvoir, VA.)

A Radio Intelligence Subsection was created under the American Expeditionary Force G-2 early in 1917 under Major Frank Moorman, long before the first American fighting forces would arrive. Cooperating with their French and British allies counterparts, they trained and prepared for the coming joint operations.

MAJ Nolan at once realized that it would be impossible to staff the AEF intelligence service with officers from the regular army, there not being enough of them that could be spared. Men were recruited from the Regular Army, the secret service, from policemen, and from training camps among recruits who could speak French or German. With the demand for trained intelligence officers exceeding the supply, U.S. Army officers were sent to the British Intelligence School at Harrow, England, until an American facility could be set up in Langres, France.

Before 1918, there was no technical training for intelligence officers. The American Expeditionary Force in France recognized this deficiency and cabled the War Department to ask that intelligence officers be sent to France ahead of their division's sailing date so that they could attend a special intelligence training course. Initially, the course consisted of a quick visit to the front lines, then enrollment at the AEF General Staff College in Langres, France. The U.S. Army Intelligence School at Langres began its operation on July 25, 1918, with Major Thom Catron as director. The school's faculty was international in flavor, with one British and two French officers on the staff. With about 11 instructors in all, they taught two six week classes and one eight week class, averaging 46 students each, for 46.5 hours a week, Monday through Saturday, and sometimes Sunday. The demand for enrollment far exceeded the number of spaces available because of the demand in the field for trained intelligence officers. (The school was discontinued at the end of the war.)

When the American First Army arrived in France, a three man Radio Intelligence Subsection was formed on 12 June 1918 with Direction Finding and intercept sections. Commanded by First Lieutenant Charles H. Matz, it was enlarged to three officers and eight men by the armistice. The First Army Radio Intelligence Subsection was responsible for analyzing and translating communications intercepted by the Signal Corps radio intelligence operators, and locating enemy radio stations based on bearings plotted by Signal Corps direction finding operators. These Signal Corps radio intelligence personnel had arrived in France in December 1917 and had undergone training enabling them to intercept messages at the rate of 25 words per minute and to translate 15 words per minute from the German. The goniometric teams used the portable SCR-83 radio receiving sets with six-foot-square antennas. Two stations could triangulate signals transmitted by enemy radios and pinpoint their locations. By analyzing traffic and combining that information with direction-finding, they could determine the depth of the enemy echelons and compile a daily order of battle.

The intelligence structure created by MAJ Nolan for the AEF included:

• At the lowest level, each battalion would have an intelligence officer and a detachment of 28 enlisted men, including 15 scouts, eleven observers and two chief snipers. The scouts would go out with patrols to take prisoners or observe enemy activity. Others would man observation posts in forward areas. Nolan realized that it was the frontline intelligence detachments that were vital to the system as a whole. It was an intelligence patrol that marked the first American action reported in G2 communiques. On 12 May 1918, three intelligence service scouts came upon a German fortification manned by 18 soldiers. In the ensuing fire fight, they shot four of whom one appeared to be an officer, secured valuable papers and retired under heavy fire.
• At regimental level an intelligence officer and eight enlisted observers would use listening-in sets, especially where trench warfare guaranteed the time to devote to their installation and operation. They would pass their intelligence information along to an artillery liaison officer.
• There was no intelligence organization prescribed at brigade level, although many brigade commanders improvised ad hoc intelligence organizations.
• In a division headquarters, an Assistant Chief of Staff, G2, had the responsibility for Military Information and the conduct of Combat Intelligence. He was assisted in this by an assistant for Military Information, a topographical officer and one commissioned interpreter, along with the requisite number of clerks. He had his own observation posts and access to the observation balloons and observation posts of the artillery information service of the divisional artillery from both his own division, neighboring divisions and higher headquarters. It was at the division that prisoners and captured documents were gathered for the first time after their capture. Prisoners were interrogated for tactical information that might affect the division and then sent on to corps.
• At the corps level, the scope of intelligence gathering was expanded, adding the tools of aerial reconnaissance and photography, corps observation posts, the corps artillery information service, balloons and flash and sound ranging sections. Prisoners were interrogated in more depth than at division. Relying on aerial photography, they made and updated topographical maps. The corps was responsible for collecting information from their immediate front to a depth of five miles. Beyond that would be the job of the Army headquarters.
• In the Army headquarters, the intelligence section devoted itself to strategy rather than tactics and concentrated more on interpreting the mass of information sent in from the lower levels. Here estimates were prepared of the enemy's situation, plans and intentions. Intelligence personnel looked into economic, political and morale aspects of the enemy, as well as questioning prisoners at greater length. The Army headquarters had a Radio Intelligence Section working on decoding and translating enemy messages. Intercept was done by a Signal Corps radio section at GHQ in Chaumont, using a combination of direction-finding equipment, listening posts, and induction coils placed near enemy ground lines. Communications security was undertaken by the Signal Corps.

  The AEF Radio Intelligence Section did solve the German code in a portion of the line they were holding. In March 1918, the Germans changed their code and the Americans braced to solve it. But one of the outstations sent a message saying that it couldn't read the new code, so send messages in the old code. After receiving messages in both the old and new code the Americans were reading the German messages before the Germans were even familiar with the new code.

  The U.S. Army radio intelligence activity operated three types of field collection stations: (1) radio intercept stations, (2) radio direction finding stations, and (3) "listening stations". Listening post operators had three duties: intercepting and decoding German ground telegraph messages, intercepting German trench telephone communications, and monitoring U.S. trench telephone conversations. Ground telegraphy was the process of transmitting morse via a short landline to a metal rod inserted in the ground. The ground was used as a conductor and one could send and receive over short distances. Intercepting these communications required placement of a rod and landline in proximity to a suspected German ground telegraphy station or laying a large induction loop in proximity. Because of this, the listening stations were placed in the most forward trenches, closest to the German frontline. The enemy messages accessed in this manner were generally were company to battalion-level. Tapping enemy landlines was also an option if the opportunity presented itself.

  In 1918, the U.S. Army Signal Corps established its first long range intercept station in Houlton (Gillian Farm) Maine, to listen to Europe. Lt Arthur E. Boeder was in command of the station.

  By the end of the war MI-8 (Cipher Bureau) had set up its own intercept sites along the Mexican border (the 14 radio tractors were eventually replaced by permanent sites) (In 1918 a Radio Intelligence Service Section was created for operations along the Mexican border) and the Military Intelligence Division (MID) had become a full division of the General Staff and was headed by a brigadier general, while its activities had expanded to include code breaking,conducting counterintelligence operations, and operating the military censorship organization. The MID had offices in seven major U.S. cities and managed over 400 intelligence officers at Army posts and other locations throughout the CONUS. MID also watched for threats from the civilian community and assisted volunteers from the 200,000-man strong American Protective Association. Finally, MID also carried out collection and analysis, directed the activities of military attaches, and prepared official reports on Germany and its allies.

  INDEX

  NEXT