Chapter III: Changes in the Marshall Organization

With its reduced staff G-1 consisted of the Officers, Enlisted, and Miscellaneous Branches. A Statistics Branch was added in July 1943 to help develop uniform personnel reporting in the Army. A new Legislative Section merged with the Miscellaneous Branch to form a Legislative and Special Projects Branch. In March 1944 the Office of the Director of the Women's Army Corps was assigned to G-1.

A major reorganization in April 1945 set up a Personnel Group (later called the Policy Group). A Planning Branch was added to it later to deal with personnel readjustment policies and universal military training. Finally in August 1945 a Control Group was set up to include the Statistics Branch, plus a Requirements and Resources Branch and an Allocations Branch responsible for the replacement system generally. Both branches were transferred from G-3. G-1's remaining functions were consolidated into a Special Group, including a Miscellaneous Branch now responsible for personnel and morale services previously performed by The Adjutant General's Office and the Special Services Division of Army Service Forces.7

A major factor complicating G-1's burden of co-ordinating and supervising Army-wide personnel operations was the division of responsibility for personnel functions among a great many different agencies at all levels of the War Department from the Secretary of War's Office on down the chain of command.8

Too large rather than too small a staff created serious management and organization problems for G-2. Its staff more than doubled in size from 1,000 in 1941 to 2,500 at the end of the war.9 In order to separate G-2's staff from its operating functions, the Marshall reorganization had created a new field agency, the Military Intelligence Service (MIS), theoretically outside the department, as an operating command. Almost immediately the distinction between G-2 and the MIS was largely wiped out by appointment of the Deputy Assistant Chief of Staff, G-2, as the Chief of the Military Intelligence Service and the G-2 Executive Officer as Assistant Chief of the Military Intelligence Service for Administration.

Maj. Gen. George V. Strong became the G-2 in May 1942. He, like most other Army officers, thought the whole concept of separating staff and operating functions impractical and recommended the abolition of the Military Intelligence Service as a separate agency. In the two years that he was its chief, G-2 and the MIS underwent four major reorganizations resulting finally in the abolition of the MIS. The principal issue was the function of evaluating intelligence and whether this should be performed by G-2 as a staff function or by the MIS. This version

Secretary Stimson, General Marshall, and General McNarney became progressively dissatisfied with the management and organization of the Army's intelligence operations. This dissatisfaction came to a head after General Strong's departure as chief in February 1944. A special War Department board under Assistant Secretary McCloy, assisted by a working group under Brig. Gen. Elliot D. Cooke from the Inspector General's Office, met to study means of strengthening Army intelligence. The resultant reorganization once again separated G-2 and the Military Intelligence Service, although the latter retained the function of evaluating intelligence. At the same time the MTS was relieved of all other functions except the collection, evaluation, and dissemination of information. Counterintelligence, training, and propaganda operations were removed from MIS and continued under the General Staff supervision of G-2 along with a World War II Historical Section, which had been established in August 1943. The Military Intelligence Service itself was reorganized along functional lines with a Directorate of Information responsible for the collection and dissemination of intelligence, a Directorate of Intelligence responsible for evaluation, and a Directorate of Administration. Co-ordinating and directing the MIS and other intelligence operations within G-2 was a policy staff similarly organized along functional lines.12

These changes, according to General McNarney, created much bitterness and resentment within G-2 and the MIS, but "frankly," he told the Patch Board, "G-2 defeated me. I never got G-2 organized so that I thought it was functioning efficiently." The principal reason, he thought, was the innate conservatism of professional intelligence personnel and their resistance to new ideas. "What I would like to do," he said, "is get rid of anybody who has ever been military attache and start new from the ground up."13

Of all the General Staff divisions G-3 was least affected by the Marshall reorganization. In contrast to the others its organization remained rather stable throughout the war. There were an Organization and Mobilization Group and a Training Branch, both divided along ground, air, and service forces lines. A Policy Branch was added at the end of the war. At this time also responsibility for the Army's replacement system was transferred to G-1 from G-3.14

G-3 officers like their colleagues found that it was impractical to try to draw a strict line between planning and operating functions. For example, as a policy planning agency G-3 made monthly allocations of training ammunition to AGF troops. In the process it also had to determine the necessity, suitability, and utilization of training facilities before their procurement, all of which were operating functions.15

The fragmentation of responsibility for personnel, aggravated by the manpower shortage, was the principal frustration for G-3 during the war. It was responsible for mobilizing, demobilizing, and training the Army, for determining the overall size or troop basis of the Army, for establishing unit tables of organization and equipment, and for dealing with OPD on allocating troops for overseas shipment. All of these functions depended upon the availability of military manpower.

Until the end of the war when these functions were transferred to G-1, G-3 was responsible for maintaining statistics on the availability of troops and units for deployment overseas and for bulk allocation of military personnel to the three major commands. G-3 correlated statistics reported periodically by Army Ground Forces, Army Air Forces, and Army Service Forces. Using these statistics as a base OPD would then determine what units or troops were to be sent overseas in response to forecasts or requests from theater commanders. Since the basic statistics were prepared by the major commands and the decisions on deployment of troops overseas were made by OPD, G-3 in practice was little more than an intermediate co-ordinating staff layer. Its difculties were increased by the

Other problems aggravated the manpower shortage, particularly the distribution of troops between combat and support elements. General Somervell and his staff were firmly convinced that throughout the war overseas commanders, OPD, and AGF continually underestimated the need for service troops overseas. This problem was particularly acute in the year following Pearl Harbor and frequently required General Somervell's personal attention and intervention at the highest levels of command.16

While division of responsibility created serious problems for G-1, the reverse was true in the case of G-4. Its major problem was the deliberate centralization of responsibility for supply and supply planning in General Somervell and ASF by General Marshall. For most of the war his staff rather than the G-4 staff dealt with OPD and the various joint and combined committees on logistical planning. When General Somervell attempted to obtain formal recognition of his status as General Marshall's supply adviser instead of G-4 in mid-1943, his proposal backfired. As a result G-4's formal functions and its staff were increased.17 The assignment to G-4 of officers unfamiliar with the Army's supply system created additional problems.

Under the Marshall reorganization, G-4 at first consisted of the Planning, Supply, and New Weapons and Equipment Branches. After a reorganization in October 1943, the investigation of overseas supply problems by a board under Maj. Gen. Frank R. McCoy, and further reorganizations in July and October 1944, G-4 consisted of three branches, Planning, Policy, and Programs. Theoretically the Planning Branch prepared long-range plans. Looking forward as far as the next war the Programs Branch was to translate long-range plans into

supply programs covering the next year or two, while the Policy Branch made "policy" decisions on current matters. As a practical matter it was still the ASF Planning Division under General Lutes that performed the detailed logistical planning for current and projected overseas operations in conjunction with strategic plans developed by OPD.

The Planning Branch had a Theater Section which supposedly developed broad policies and directives for the use of the Army's logistical forces both overseas and in the zone of interior. It had special responsibilities for hospitalization and evacuation. Another mission was to develop a uniform, coordinated set of supply regulations out of the welter of conflicting directives on the subject issued by various agencies at all levels of command.

An Organization Section studied, reviewed, and revised the Army's logistical organizations. A Special Projects Section studied logistical doctrine, supervised management of Army logistics, and was responsible for logistical aspects of mobilization, demobilization, and postwar planning.

The Programs Branch was responsible for balancing military requirements with the resources available and for approving new equipment and materiel. Its Equipment Section dealt with new weapons and equipment. A Requirements Section developed the Army's supply requirements. After July 1944, it also prepared the supply section of the Army's Victory Program Troop Basis and the Overseas Troop Basis and coordinated the Army Supply Program generally. All three functions had been previously performed by OPD. An Allowances Section analyzed and approved standard as well as special allowances of equipment for Army combat units and other organizations. An Installations Section determined supply plans and policies as they applied specifically to posts, camps, stations, and other facilities under the Army Installations Program.

The Policy Branch was responsible for solving problems arising out of current supply operations. A Distribution Section handled issues affecting the distribution, storage, issue, and maintenance of equipment. A Property Section handled questions concerning the acquisition of land, construction of facilities and installations, and similar housekeeping functions. An Economics Section dealt with issues involving Allied supply

Of the five new War Department Special Staff divisions added after the Marshall reorganization, two of them, the War Department Manpower Board and the Strength Accounting and Reporting Office, concerned personnel; another, the New Developments Division concerned research and development of new weapons and material; a fourth, the Civil Affairs Division, dealt with military government of liberated and occupied territories; a fifth, the Special Planning Division, was responsible for demobilization planning, universal military training, and the postwar organization of the Army. Three former special staff agencies assigned by the Marshall reorganization to Army Service Forces, the Budget Division, the National Guard Bureau, and the Office of the Executive for Reserve and ROTC Affairs, were restored by the end of the war as special staff divisions as-the result of political pressure from Congress. The Information and Education Division, an outgrowth originally of The Adjutant General's Office's responsibilities for personnel and morale services, became a special staff agency in September 1945, when the War Department decided to merge all information services under a Director of Information who reported to the Chief of Staff. The other agencies involved, the Bureau of Public Relations and the Legislative and Liaison Division, were already special staff agencies.

The political consequences of American military operations in liberated and later occupied enemy territory were such that neither Secretary Stimson nor General Marshall could avoid assuming personal responsibility for them. Secretary Stimson centralized War Department responsibility for this function in the Civil Affairs Division created on 1 March 1943 as a special staff division of the War Department General Staff.

Military government policy had become a critical problem shortly after the landings in North Africa at the end of 1942 when Lt. Gen. Dwight D. Eisenhower found himself in political difficulties because of his dealings with Admiral Jean F. L. Darlan as de facto head of the local French administration. Eisenhower requested instructions from the War Departmenton how to deal with the situation.

At that time, following the precedent of World War I, military government was the responsibility of the local overseas theater commander. There was no single agency within the War Department to provide direction on this subject. By default OPD, as the liaison between General Eisenhower and the Joint Chiefs of Staff, was handed the problem.

In March 1942 a military government training school at the University of Virginia in Charlottesville was established under the Provost Marshal General. Efforts to develop military government policy bogged down in disagreement within the administration over whether control over civilian populations in militarily occupied areas should be a military or civilian function. Similarly, efforts to agree on a War Department position on military government were stymied by disagreement within the General Staff until General Eisenhower's request made the problem immediate and urgent.

Secretary Stimson sent Assistant Secretary McCloy overseas to North Africa to investigate and report on the problem. The creation of the Civil Affairs Division (CAD) was the result of recommendations Mr. McCloy made on his return. Now a single staff division was responsible for advising the Secretary and the Chief of Staff on nonmilitary matters "in areas occupied as the result of military operations." Its staff was small, and it had no operating functions. The Provost Marshal General continued to run the Military Government School, and theater commanders carried out policies and instructions issued through the Civil Affairs Division.19

Having created the Civil Affairs Division, Secretary Stimson had also to decide whether the chief should be a military man or a civilian in uniform. Choosing the former, he selected Maj . Gen. John H. Hilldring, an experienced General Staff officer and former G-1, who remained chief of the division throughout the war. The division staff was organized along functional lines based on essential community services, and each functional branch was divided along geographic lines.20

The State-War-Navy Coordinating Committee was formed in December 1944 to co-ordinate foreign and military policies. CAD had a representative on this committee and on the Working Security Committee set up in Washington to assist the European Advisory Commission, working under General Eisenhower in London; on the development of postwar policy toward Germany. Finally CAD had to deal with the United Nations Relief and Rehabilitation Administration (UNRRA). Under

Because no critical manpower shortages developed during World War I, the War Department did not anticipate the problem in World War II. A second complicating factor was the division of responsibility for personnel policy and operations among many agencies within the department. (Chart 9) Centralizing responsibility for this .function in one agency would have required a major reorganization causing dislocation and administrative turmoil throughout the Army.

Responsibility for military personnel operations was divided among G-1, G-3, OPD, the three major commands, the seven technical services, and the administrative services. Responsibility for civilian personnel was divided among the Secretary of War's Civilian Personnel Office, Army Service Forces, and the technical services.

After the Marshall reorganization, G-1 was supposedly limited to policy planning and co-ordination among the three major commands. But, in practice, as indicated earlier, with its drastically reduced staff it became a co-ordinating agency more concerned with administration than planning.

Army Ground Forces resisted the authority of ASF over military personnel operations, and the Air Forces were busy developing their own separate system of personnel administration. Within ASF both the Personnel Division and The Adjutant General's Office were responsible for Army-wide military personnel operations, including personnel and morale services. The Adjutant General was responsible for the induction, classification, and assignment of military personnel. G-3 prescribed the size and composition of units in the Army through tables of organization, and it allocated military personnel in bulk to the major commands. OPD regulated the flow of units and replace-

The Civilian Personnel Division in the Secretary of War's Office was responsible initially for all War Department civilian personnel operations, while ASF's Industrial Personnel Division took over responsibility for civilian personnel management among technical service installations in the field including labor relations. The Civilian Personnel Division continued to be responsible for civilian personnel management within the War Department itself. The latter's actions frequently conflicted with similar activities in the headquarters of the technical services and, of course, AAF headquarters.22

Lacking centralized responsibility for personnel policy and operations, the only practical alternative for the War Department when the manpower shortage did develop in late 1942 was to create another special agency-the War Department Manpower Board-for dealing with this aspect of the problem. Divided responsibility led to conflict among the various agencies of the Army over just how many men there were in the Army. Another special agency, the Strength Accounting and Reporting Office, was established within the Chief of Staff's Office to co-ordinate and standardize personnel statistics within the Army.

Government leaders, including General Marshall, gradually became aware by the end of 1942 that there was not enough manpower available in the country to meet all the nation's requirements, both civilian and military. The Bureau of the Budget inaugurated a program to conserve manpower within the federal government and was responsible for setting civilian manpower ceilings for each agency. In March 1943 General Marshall, on the recommendation of an emergency committee of the General Staff and the three commands, created the War Department Manpower Board under another former G-1, Maj. Gen. Lorenzo D. Gasser. The board reported directly to the Chief of Staff, recommending specific manpower savings, both civilian and military, on the basis of detailed surveys of War

Department activities and installations within the continental United States. Most of the surveys were conducted by teams located in each of the nine service commands and the Military District of Washington. The activities surveyed were under ASF's jurisdiction. Its Control Division assisted teams, using industrial work measurement, work simplification, and standardization techniques which produced considerable savings in manpower. The Industrial Personnel Division conducted similar surveys. As a result of these combined efforts, the War Department Manpower Board claimed at the end of the war that it had reduced the number of civilian and military employees of the War Department and the Army within the United States by about one-sixth of its wartime peak in June 1943. It said further savings could be obtained if unnecessary duplication of functions among the technical and administrative services were eliminated, particularly in their headquarters.23

Conserving military manpower was harder than conserving civilian manpower. The main problem that developed in this area was to provide an effective replacement system that would meet the needs of overseas commanders. The latters' advance estimates of how many people they would require were generally inaccurate, but the greatest difficulty was the inability of the Army to account accurately for troops "in the pipeline," moving from one organization, station, or area to another, in hospitals, on leave, on detached service, or at school.

Divided responsibility for personnel administration inevitably led to conflicting reports on the number of men actually in the Army which the department could not reconcile. Public ventilation of these discrepancies caused Secretary Stimson and General Marshall acute embarrassment, especially in their relations with Congress.

the creation in May 1944 of a new special staff agency within the Office of the Chief of Staff, the Strength Accounting and Reporting Office, which was to improve and standardize manpower reporting. With the issuance of its first monthly edition of the Strength Report of the Army series in July 1944, this office steadily improved and refined military manpower reporting within the Army. 24

Manpower conservation and improved statistics were not enough. Divided responsibility for control over personnel management was a stubborn obstacle that did not yield to piecemeal solutions. The Army never did succeed in developing a satisfactory replacement system during the war. Only the end of the war and the shift to demobilization removed the problem for the time being. Two Air Force management experts, Drs. Edmund P. Learned and D. T. Smith, appointed specifically to study the Army personnel replacement system reported:

No single agency in the War Department General Staff has adequate responsibility or authority to make an integrated Army-wide personnel system work. There are too many offices . . . in the personnel business; there is some confusion in responsibility and no one place that can be held responsible for a total summary of the situation.25

Of their recommendations for centralizing responsibility for the replacement system, the department acted on only one-to transfer responsibility for allocating replacements from G-3 to G-1. OPD continued to allocate combat replacements and so spread the over-all manpower shortage among the various overseas theaters.26

power shortage would still have been beyond their power to solve. Both Secretary Stimson and General Marshall were frustrated in trying to deal with this problem because neither the President nor Congress was willing to vest in one agency sufficient authority to determine manpower allocations among all the claimants. The Secretary and General Marshall repeatedly urged enactment of compulsory national service legislation similar to the system adopted by the British. This would have meant the conscription of industrial and agricultural labor. Strong opposition by labor unions and farm organizations to this proposal led to its rejection in Congress and within the administration.27

Research and development of new weapons and equipment in the Army suffered from subordination to production throughout the war. Agencies responsible for research and development, whether at the General Staff, ASF, or technical services, were subordinate elements within organizations primarily concerned with production and supply.

Dr. Vannevar Bush, president of the Carnegie Institution of Washington, Director of the Office of Scientific Research and Development, chairman of the joint Committee on New Weapons and Equipment of the Joint Chiefs of Staff, and chairman of the Military Policy Committee of the Manhattan District, told the House Committee on Military Affairs that the armed services did not sufficiently realize the importance of science because military personnel by training and tradition did not appreciate the contribution it could make to national defense. They had not learned as industry had "that it is fatal to place any research organization under production departments. In the services it is still the procurement divisions who maintain the research organizations."

Source: Nelson, National Security and the General Staff, New Developments Division History; McCaskey, The Role of the Army Ground Forces in the Development of Equipment; History of Research and Development Division, ASF; Green Thomson and Roots, Planning Munitions for War; Millett, Army Service Forces; Stewart, Organizing Scientific Research for War; Morison, Turmoil and Tradition; and Hewlett and Anderson, The New World.

Functionally, the issue was again one of planning versus operations where mixing planning with operational responsibilities led to the neglect of planning. A second and more immediately important obstacle was the division of responsibility for the research, design, development, production, testing, procurement, and battlefield deployment of new weapons and equipment among many agencies. (Chart 10) The most serious division and the one which caused the most delay was that between the technical services as producers and the AGF and combat arms as users.

Within the Army the technical services throughout the war were the agencies responsible for nearly all military research and development except for the AAF, which had its own programs. G-4 exercised General Staff supervision over the technical services activities through a Research and Development Section created in 1940. The combat arms were responsible for establishing military requirements and characteristics of new weapons and equipment, for service testing them under simulated combat conditions, and finally for accepting or rejecting them as standard Army equipment. Military requirements for new equipment in turn depended on the development of tactical doctrine. These two functions were under the General Staff supervision of G-8.

Under the Marshall reorganization, Army Service Forces took over responsibility for research and development operations from G-4, which continued to have a Developments Section within its Requirements and Distribution Branch. Throughout the war this function was buried within ASF under the Directorate of Materiel and did not even achieve the status of a separate division until the war's end. This reflected

the fact that the Materiel Directorate's primary interest in this area was in the requirements and specifications of those weapons and equipment already developed and proposed for adoption as standard equipment by the Army. Since technical services were the agencies mainly responsible for the conduct of the Army's research and development efforts, ASF's Research and Development Division was largely a co-ordinating staff between them and AGE There were lengthy delays caused by disagreement between the latter, representing the users, and ASF's research and development staff, representing the producers, over specifications which had to be negotiated. Another mission was to promote the use of common items of supplies, and there were lengthy delays in trying to get the technical services, particularly the Ordnance Department, to change their specifications. The Research and Development Division also assisted the technical services when they had trouble obtaining raw materials, equipment, and facilities for their research and development programs.

Conflicts between the technical services and AGF delayed production and procurement of new materiel. Often differences between them could not be resolved short of General Marshall himself. A classic example was the dispute between the AGF in the person of General McNair and the Ordnance Department over the development of a heavy tank. Armored doctrine held that there was no need for a heavy tank because it moved too slowly. Mobility was the vital characteristic, and both armor and firepower should be subordinated to it. One result was the development of a light, half-track armored vehicle known as a tank destroyer which proved unable to cope with heavier German tanks in North Africa. (Later tank destroyers,

There was a tendency among combat officers, the Air Forces excepted, to ignore radically new departures in development of new equipment in favor of tinkering with or improving existing weapons. This conservative tendency stemmed in part from their general unfamiliarity with scientific and technological developments or with production and engineering. Second, the better tended to be the enemy of the good. Developers charged that representatives of the combat arms repeatedly rejected equipment that was not perfect. This often involved redesigning and further delay simply to incorporate some new feature.31

Secretary Stimson was dissatisfied with the slowness of research in the Army, particularly in the field of electronics. His special assistant, Mr. Harvey H. Bundy, was a troubleshooter on scientific problems and acted as liaison with the scientific community. His special task was to oversee the development of the atomic bomb. In the spring of 1942, Mr. Stimson appointed Dr. Edward L. Bowles of MIT as his Expert Consultant to push the development of radar in particular and other improvements in the field of electronics. He had a staff of forty-seven specialists who made frequent trips overseas to obtain firsthand evidence of combat requirements.32

was by virtue of the many key positions he held during the war probably the most influential and the most articulate representative of the scientific community in the defense program. He and Dr. Bowles, acting through Stimson, were responsible for increasing the Army's participation in the development of new weapons and other materiel. They were dissatisfied with the Army's research and development programs. Partly because of their slowness to act in this area, the Chief of Ordnance in 1942 and in 1943 the Chief Signal Officer were replaced. The influence of the Office of Scientific Research and Development (OSRD) and Dr. Bowles on AAF research and development and on the use of operations research techniques has been mentioned previously. The Ground Forces never did make any significant use of the latter during the war.33

Pressure on the department also came from the battlefields. Reports from the Pacific on the unsuitability of existing equipment for jungle or amphibious combat led General Marshall to send a team of experts to that area under Col. William A. Borden to investigate and report directly to him on the kinds of weapons and equipment needed in the area. Colonel Borden, an Ordnance expert with a flair for salesmanship and diplomacy, was then General Somervell's Special Assistant to the Director of Plans and Operations, a cover for his primary function as a troubleshooter.

In October 1943, acting on the recommendations of Bundy, Bush, and Bowles, Stimson created the New Developments Division as a special staff division to expedite production and procurement of new and improved equipment. Under Maj. Gen. Stephen G. Henry the New Developments Division was primarily a troubleshooting agency with a limited staff of about two dozen civilian and military personnel. They tried to bridge the gap between producer and consumer and to hasten delivery of equipment to the battlefield.34

caused by disagreements between the technical services and the combat arms over testing equipment. While the Research and Development Division was flooded with the paper work created by this problem, the staff of the New Developments Division spent more of its time in the field trying to find short cuts around the rigid testing requirements of AGF. This was handled on a case-by-case basis, and with a small staff its success was limited. The problem remained unsolved at the end of the war.35

Another duty assigned the division as the result of the manpower shortage was to provide a pool of technical and scientific specialists drafted into the Army. Induction centers, pressed for combat replacements, generally assigned these individuals to the combat arms. An Army Technical Detachment added to the New Developments Division in October 1944 tried to locate such personnel before they became assigned as combat replacements. In its year of operation the detachment had located and assigned four hundred such specialists to the technical services and other installations performing research and development, but it still had a backlog of over eight hundred unfilled requests.36

The Manhattan Project, organized to supervise the production of the atomic bomb, pioneered in what later became known as project management. The Army took over direction of the atomic program in mid-1942, when scientists working under the Office of Scientific Research and Development had demonstrated that an atomic weapon was technically feasible. Producing the fissionable material required to detonate the bomb involved enormous outlays of men, money, and resources, including huge amounts of electricity and water. The Corps of Engineers was selected to construct and operate the required installations and facilities because of its experience with large-scale public works projects.

Secretary Stimson with the approval of President Roosevelt placed Brig. Gen. Leslie R. Groves, the man responsible for building the Pentagon, in charge of the project. His organization was known as the Manhattan District of the Corps of Engineers, but the Chief of Engineers was relieved of responsibility for the project shortly after General Groves' appointment. For practical purposes, it was an independent agency. General Groves reported to a Military Policy Committee set up to oversee the project and determine general policy. Dr. Bush was its chairman. On the committee were Maj. Gen. Wilhelm D. Styer, General Somervell's deputy, Admiral William R. Purnell, and Dr. James Bryant Conant, president of Harvard and head of the National Defense Research Advisory Committee of OSRD. Conant and Bush represented the interests of the scientific community. General Groves also reported directly to General Marshall and to Secretary Stimson, usually through Mr. Bundy's office.37

As Chief of Staff of the Army during World War II, General Marshall had two principal missions. He was the Army's chief strategy adviser and also general manager of the department. The increasing size and complexity of the Army's operations as the United States gradually mobilized for war made it physically impossible for Marshall to perform both functions. Since his major function was to advise President Roosevelt on strategy and military operations, he was forced to divorce himself more and more from his administrative functions as general manager of the department.

From Marshall's viewpoint the existing structure and standard procedures of the Army's General Staff made it practically impossible for him to delegate responsibility for administration to the General Staff. Its committees were too slow in reaching collective decisions and could not distinguish between important questions and minor details which they constantly thrust at him for decision.

Passage of the First War Powers Act in December 1941, right after Pearl Harbor, gave Marshall the opportunity to streamline the department's organization. Under the new organization he delegated his administrative responsibilities to a single Deputy Chief of Staff within the department and to three new major field commands, Army Ground Forces, Army Air Forces, and Army Service Forces. At the same time he selected his own principal deputies and subordinates. The reorganization left him free, as he insisted, to concentrate on military strategy and operations aided by the staff of the War Plans Division. Redesignated the Operations Division it became an operating headquarters instead of a planning agency. In effect it became a super general staff, bypassing the other General Staff divisions in the interests of prompt action.

In this manner General Marshall could control departmental operations by decentralizing responsibility for their administration just as the pioneer industrial managers at DuPont, General Motors, and Sears had done in the previous decades. Although Marshall was apparently not familiar with these earlier industrial management reforms, it is not surprising that he, faced with similar problems, came up with similar solutions. Marshall's understanding of the basic principles of management as well as his exceptional judgment of men made him one of the department's most effective administrators. The results of his reorganization were so satisfactory that he strongly recommended applying the same principles in organizing a new department of the armed services after the war.

General McNarney, as Deputy Chief of Staff and general manager, exercised tight control over the department, except for his increase in the functions and personnel of G-4 in mid1943. General Handy, his successor who had previously been Chief of the Operations Division, was more sympathetic to the General Staff, which Marshall and McNarney had largely ignored. Handy was also more critical of Somervell's ASF than McNarney.

The difficulties Marshall and McNarney had with the management of intelligence, personnel functions, and research and development of new weapons indicated that the reorganization had not solved all problems of administration. The relations between the functionally organized ASF headquarters and the

Supervising and co-ordinating the technical services along functional lines which cut across formal channels of command inevitably generated friction. If the offices of the chiefs of the services had been phased out of existence as had been done with the chiefs of the combat arms within AGF, there might have been less friction and ill-feeling. AAF headquarters deliberately created its own integrated supply system from the start and did not have to deal with any technical services with long-established traditions and influence.

ASF might have solved its organizational and management problems by confining its top staff to broad policy planning and co-ordinating functions. The technical services chiefs argued for this alternative, but the experiences of the three major commands led their commanding generals to insist that their headquarters staff must operate in order to exercise effective control over their subordinate agencies and commands.

Although put together in haste, the Marshall reorganization worked as well as it did because General Marshall was the real center of military authority within the department. Both Roosevelt and Secretary Stimson supported him. In turn General Marshall delegated broad responsibility with commensurate authority to Generals McNarney, McNair, Arnold, and Somervell. While the Marshall reorganization lasted only as long as he was Chief of Staff, it was based upon the accepted military principle of unity of command and similar to concepts of administrative management developed by major industrial corporations.