History
On Line
A settlement of the truce negotiations or a
continuation of the hot war might have obscured several of the problems that
became important during the winter of 1951-52. But the absence of conclusive
developments either at Panmunjom or on the battlefield focused more attention
upon the flow of affairs in the rear areas. The lack of decision in the debates
and at the front did not obviate the need for decisions behind the scenes.
Regardless of the details of the eventual agreement at Panmunjom, the basic
problem of the Communist threat in the Far East would remain. By November 1951
it was evident that no military decision would be won or even sought. What,
then, would come after the armistice?
Since World War II the United States had
provided the chief opposition to the spread of communism all over the world. The
North Atlantic Treaty Organization, formed in 1949, offered a nucleus for
resisting further Communist aggrandizement in Europe. By the close of 1951 the
United States had built up the American forces on the European continent to six
divisions and was asking the other NATO member nations to increase their
contributions. Progress was slow since rearmament and upkeep of armed forces
were expensive items and the threat of war in Europe did not appear to be
critical. In February 1952, however, an event of
considerable importance for NATO took place when the NATO conference held at
Lisbon approved plans for a fifty-division NATO ground force that would include
German elements for the first time. The news of the rearmament of West Germany
and its future participation in NATO evoked protests from the Soviet Union, but
these were rejected by the United States, the United Kingdom, and France. The
addition of West Germany forged another link in the European defense line, but
whether this link would be a source of strength or a weakening liability was as
yet unknown.
In the Far East, on the other hand, the
United States stood alone. The colonial commitments of Great Britain and France
in Africa and Southeast Asia ruled out hope of major assistance from them in the
near future. Unless the United States wanted to continue to shoulder the burden,
only one practical alternative remained-to tap the manpower potential at hand in
the Far East. To fashion an effective force that would have the training and
equipment as well as the will to fight against Communist encroachments would be
expensive and time consuming, but not as costly as maintaining large numbers of
U.S. troops in the area. Fortunately a start had been made in the Republic of
Korea, Nationalist China, the Philippines, and Japan. Military assistance advisory
groups had begun the long-term tasks of developing national forces to withstand
aggression. The main problem would be to strengthen and accelerate the military
aid program so that ultimately the United States could delegate some of the
responsibility for the defense of the Far East against Communist expansion.
Improving the ROK Army
As long as the war continued, the Republic
of Korea would remain the most critical link in the defense chain. Here lay the
direct threat to a nation sponsored and supported by
the United States- a threat that could not be ignored or evaded without
endangering the entire U.S. position in the Far East. To meet the Communist
challenge the bulk of the U.S. military forces in the Far East had been
committed to the war in 1950 and reinforcements from the United States had
quickly followed with a resultant drain upon the strategic reserve. The only
hope for halting this flow of manpower seemed to rest in the substitution of
Korean troops for U.S. soldiers. But before Korean forces could take over and
successfully defend their own liberties, much remained to be done.
The Army, as has been mentioned previously,
had begun to work on the task.1 Operating through the Korean Military
Advisory Group, active steps were under way to improve the quality and
efficiency of the ROK Army. Both the Secretary of the Army, Mr. Pace, and the
Chief of Staff were personally interested in the progress of the KMAG plan of
action and this helped to remove some of the obstacles that had hampered the
program.2
Basically the chief impediment had been
time. Until the pressure on the battlefield had lessened, it was impossible to
withdraw units for training or refresher courses. Replacements frequently were
rushed up to the front lines with insufficient instruction in tactics and
weapons. It was a wasteful process, but, under the circumstances, necessary.
When the fighting slacked off in July,
General Ridgway and his advisors began to devote more attention to the adequate
preparation of men and units for combat. The raw material supplied by the ROK
was good, although often undernourished. If properly led, the average Korean
youth showed courage, stamina, and a great deal of native patience- all
excellent qualities, especially in a defensive war. Despite the variable performance of the South
Korean forces during the first year of the war, military observers were
convinced that most of their worst moments could be traced directly to poor
leadership and lack of training. It was against these weaknesses that KMAG
launched its main assaults.
In the program outlined by Ridgway in July
1951, the chief objective was to correct the leadership problem by
re-establishing and reinvigorating the South Korean military school system. Now
that time was available, he hoped eventually to create a professionally
competent officer and noncommissioned officer corps.3 This did not
promise to be an easy project. The ROK Army did not pay its officers or enlisted
men more than a pittance, considering the inflationary trend of the South Korean
economy. It was hardly surprising that many of the officers should try to make
ends meet by resorting to questionable expedients, but it was not conducive to
the creation of a good army when these same officers put personal benefits ahead
of military necessity. Despite the continual pressure that KMAG applied upon the
ROK Government to take severe disciplinary measures against corrupt officers,
the problem was likely to remain until the officers received sufficient
compensation to support themselves and their
families.4
Another element in the complex undertaking
of building a capable officer corps was the instilling of confidence at all
levels-confidence in the officers among the soldiers and confidence of the
officers in themselves. The average South Korean officer was young, and in many
cases regiments were commanded by men under thirty. Yet despite the leavening
factor of youth, caution was characteristic. In the absence of higher authority
or direct command, juniors were usually reluctant to act lest they offend their
superiors. The dearth of initiative would not be simple to compensate for. It
was a basic deficiency that arose from the emphasis that the Koreans placed upon
rank and seniorityyou bowed to those above you and bullied those below you. As
long as this condition lasted, few ROK officers would be willing to risk
offending their superiors by taking independent action. To counter this
tendency, KMAG instructors would have to exert skill and patience over a
considerable length of time.5
While KMAG attempted to implant confidence,
initiative, and professional skill in the upper echelons, a Field Training
Command was put into operation behind the lines to bolster the morale of the
soldiers. As each ROK division was rotated through a nineweek course of basic
training, refresher instruction in weapons and tactics helped to weld the
fighting units into better combat teams. The success of the course of training
led to the establishment of three
additional camps - one in each corps area- in September 1951.6
By the beginning of November 1951 considerable progress was made in
the organization of the school and training system. The Replacement Training and
School Command under General Champeny had acquired additional personnel and was
ready to handle large groups.7 To centralize training installations
the RTSC recommended that the Infantry School, Artillery School, and Signal
School all be relocated at Kwangju in southwestern Korea, about 120 miles west of Pusan. The
consolidated school opened in early January and was given a new name the
following month- The Korean Army Training Center. Up to 15,000 troops could be instructed at
one time at this installation.8
At the officer candidate school level the
course was extended from eighteen to twenty-four weeks to provide extra training
for the new company grade officers. And on 1
January 1951 the Korean Military Academy reopened its doors at Chinhae near
Pusan with a full fouryear curriculum patterned after West Point. For field
grade officers a Command and General Staff School was established at Taegu and
officially launched on 11 December 1951.9
In the meantime 150 ROK officers attended the Infantry
School at Fort Benning, Georgia, and another 100
took the course at the Artillery School at Fort Sill,
Oklahoma. From this group trained in U.S. schools the Army hoped to recruit the
future instructors for the ROK Army. As a whole, the officers chosen
to go to the U.S. service schools were the pick of
the crop and did well at their studies. The chief problem, as usual, was that of
language and interpreters had to be sent along with the students, but many of
the latter took advantage of their tour in the United States and learned some
English as well. Since few Americans learned to speak Korean, this was to be of
great value when these officers later returned to Korea. The problem of
communication between Korean and American soldiers was a continuing and
perplexing one, especially when technical exchanges took place. The first group
of ROK officers graduated from the U.S. schools in March 1952 and a second contingent of
250 officers left Korea
the same month to begin the next cycle.10
The growing effort in behalf of the ROK
Army increased the demands upon KMAG and led to a request from Ridgway that the
group be augmented. On 1 November the Department of the Army approved an expansion of over
800 spaces for KMAG, bringing its total strength to over 1,800 officers and
men.11
As the ROK Army began to improve in
quality, Acting Secretary of Defense William C. Foster raised the question of
its ultimate quantity. On to November he requested the JCS views on the mission
and size of the postwar ROK defense force.12 Since both MacArthur and
Ridgway had consistently
favored a ten-division, 250,000-man army the JCS recommended that this figure be
maintained despite the fact that the President and his advisors had decided in
the meantime to increase ROK military strength.13 The Joint Chiefs
informed the Secretary of Defense in late January 1952 that the ROK economy did
not have the capability to sustain a significant expansion of military forces in
the near future. In their opinion, the present ROK units, when properly trained,
equipped, and led, should constitute a sufficient deterrent to further
aggression or, if the occasion demanded, could delay Communist advance until
reinforcements could be brought in.14 The ROK Government was in the
throes of a serious financial crisis as a result of steady inflation and hardly
in a position to assume additional heavy expenses, it is true, but this was but
one facet to the problem. It should not be forgotten that the United States had
made commitments to supply many of the military requirements of its NATO allies
and was about to sponsor the renascence of the Japanese defense forces as well.
With U.S. production not on a full war scale and with heavy demands at home and
abroad to be met, it appeared that ROK Army expansion would have to await a more
opportune moment.
The ROK Government and its most effective
spokesman, President Rhee, did not, of course, agree that an army of ten
divisions would be enough to defend South Korea in the postwar period, but the
matter lay quiescent until late March 1952. During an inspection trip to
Korea, Secretary of the Navy Dan Kimball discovered
that General Van Fleet favored the formation of ten additional ROK divisions.
When he reported this item to the Army Policy Council upon his return there was
considerable consternation. This was the first intimation that the Army had
received of strong support for ROK Army expansion and it was a little
humiliating to have to get the information from the Navy. In any event General
Hull immediately asked Ridgway for an explanation.15
Ridgway was just as surprised as his
superiors had been and forthwith queried Van Fleet. In this roundabout manner he
was finally informed by the Eighth Army commander that the latter did believe in
the expansion of the ROK Army to twenty divisions. Van Fleet maintained that the
ROK had the manpower and the desire to fight and the United States could support
ROK troops in Korea much more economically than American forces. As a conclusion
to a somewhat amazing performance, Van Fleet referred his commander to an
interview he had just had published in the U.S. News and World Report, if
Ridgway desired more information on his views.16
Whatever Ridgway's personal reaction to
this turn of events may have been, he exercised remarkable restraint. He told
Hull that he had not seen Van Fleet's magazine interview, but nevertheless he
flatly disagreed with his subordinate on doubling the ROK Army. Not only was the
ROK economy unable to additional forces, but he thought that the development of the Japanese defense
forces should be given preference at this time. The training program for the ROK
ten-division army was just beginning to bear fruit, he went on, but it would
take another ten months before it was completed. If the United States started to
organize ten additional divisions it would require eigtheen months to prepare
them for action and the United States would also have to furnish all
subsistence, clothing, and pay. Although he had the utmost respect for General
Van Fleet, Ridgway informed Hull that "His outlook, however, in this particular
case is in my opinion quite naturally focused almost exclusively on the Korean
situation, as that situation affects the U.S. I
cannot believe due consideration has been accorded to the inseparable relation
of the Japanese, Chinese Nationalists, and Southeast Asia military programs to
which the United States Government is committed, or which it has under
study."17
General Ridgway's disapproval was enough to
prevent an increase in the ROK ground forces and when he left the Far East
Command in May for a new assignment as Supreme Allied Commander, Europe, no
change had been made in the size of the army. In the matter of ROK air and
marine forces, however, the Far East commander ran into more difficulty. The ROK
Air Force was small
and equipped with propeller-driven planes. In Ridgway's view, a 4,000-man air
force with seventeen obsolescent fighters and twenty-nine miscellaneous craft
could offer no real opposition to a future Communist air sweep and would
probably be wiped out quickly. Maintenance of a tiny, impotent force was
wasteful, Ridgway continued, since in the event of renewed aggression the United
States would still have to provide air support for the ROK. "A second best Air
Force is worse than none," he told the JCS.18 But the U.N. commander
found that it was next to impossible to abolish a service once it gained a firm
foothold. The JCS showed no disposition to tamper with the ROK Air Force and no
action was taken on Ridgway's recommendation.
The same reception met his proposal to
dispense with a separate marine force after the war. To his way of thinking, a
marine division would require separate overhead and support elements that would
duplicate those of the Army and this needless expense would have to be borne by
the U.S. taxpayer.19 The U.S. Navy, however, had already established
both a Naval Advisory Group and a Marine Advisory Group to the ROK in February,
and Ridgway's plea went unheeded.20
Despite the mixed success of Ridgway's
efforts to restrict the size of the ROK armed forces, he and his staff did
manage to effect several internal improvements aimed at bolstering the
efficiency of the ROK troops. In November Ridgway
authorized Van Fleet to increase the strength of the Korean Service Corps to
60,000 men. This would
permit all the laborers and carriers in the combat areas to be organized and
brought under tight control and discipline. It would also assure the fighting
corps of more reliable service support. Eventually Ridgway planned to raise the
ceiling of the Korean Service Corps to 75000.21
The U.N. commander also made efforts to
correct one of the basic weaknesses of the ROK Armythe lack of adequate integral
artillery support. In the past ROK divisions had been forced to rely upon U.S.
artillery support for most of their offensive and defensive missions. Only one
105-mm. howitzer battalion was assigned to each ROK division as opposed to three
105-mm. and one 155-mm.
battalions in each U.S. division. In addition, the latter had tank support and
more heavy mortar companies available to perform its tasks. Previously the
Eighth Army and Far East Command staffs had argued that the rough terrain, lack
of roads, and resupply problems added to the dearth of trained artillerymen and
unavailability of equipment had precluded expanding the ROK artillery. But as
the war lengthened and settled into its static phase, many of these objections
were overcome. In September four ROK 155-mm. howitzer battalions were authorized
for activation before the end of the year. These battalions were trained for
eight weeks by U.S. corps personnel. Three headquarters batteries and six
105-mm. howitzer
battalions were added in November and began their training in January 1952.
Finally in March Ridgway approved a full
complement of three 105-mm. and one 155-mm. howitzer battalions for each of the ten ROK
divisions. In May the Department of the Army sent interim authorization for the
Far East Command to proceed with this program.22
The process of improving the ROK Army was
well on its way by April 1952. Schools and training programs to raise the
leadership level and confidence of the troops had been started and began to
produce demonstrable results. Increased service and combat support to bolster
the ROK forces in combat was being organized and equipped. Given time, the ROK
Army could become one of the better armies in the Far East.
Relations With the ROK
Military affairs were but one aspect of the
problem of conducting a war on the soil of an ally. As the United States had
discovered during the World War II campaigns in China, politics played an
important role that seemed to increase in inverse ratio to the pressures
generated at the front. If the fighting were heavy and external crises dominated
the scene, internal politics might be played down or overshadowed. But a static
war permitted domestic dissensions to come to the surface and frequently
required delicate and diplomatic handling. The situation in South Korea followed
this pattern during the armistice period and was to occasion many a tense moment
for the U.N. Command in its efforts to fight a war and conclude a peace at the
same time.
When the armistice negotiations were
initiated, President Rhee and his government had firmly opposed a compromise
settlement with the Communists. They had no desire to return to the status quo
that had been so unsatisfactory in the prewar period and regarded the time as
ripe for the unification of Korea- under ROK hegemony. As long as the talks
appeared to be making little progress, there was scant reason for vehement
action on their part. During the breakdown of negotiations in August, a ROK
spokesman had frankly welcomed the turn of events and claimed that the
Communists had simply used the discussions as a cover for their military
build-up.23
On 20 September, Rhee set forth his terms
for dealing with the enemy- terms that the Communists could not possibly accept
without admitting defeat. First the Chinese should withdraw from Korea and the
North Koreans should be disarmed. Then the latter would be given full
representation in the Korean National Assembly and presumably this would settle
the whole problem. The ROK President felt that the Communists should be given a
time limit for acceptance; otherwise the negotiations should be concluded. In
his opinion, the enemy was using the talks to humiliate and discredit the United
Nations in the eyes of the Communist world.24
After the negotiations resumed in October,
the ROK Government expressed its dissatisfaction in many ways. A favorite method
was through "spontaneous" demonstrations similar to the one staged by students
in Pusan in early December. Close to 500 students gathered and paraded through
the city carrying signs and placards bearing such anticease-fire
slogans as "No Armistice Without Unification."25
These apparent popular outbursts of indignation
against the truce meetings could be mounted quickly whenever there seemed to be
a possibility of agreement at Panmunjom.
The continual ROK agitation and hints from
Rhee that the government might not observe the terms of an armistice disturbed
General Ridgway. It may be remembered that the UNC control over the ROK armed
forces rested upon Rhee's July 1950 letter to MacArthur assigning command to the
latter and anyone he delegated for the duration of the hostilities only. In the
months that had followed the ROK Government had faithfully observed this pledge
and it had not been considered necessary to seek a firm written understanding on
the matter. But by early 1952 Ridgway felt that a formal agreement covering the
armistice period should be secured to forestall independent action by the
Republic of Korea in opposition to the truce stipulations. Unless the ROK
military forces remained under UNC control after the truce was concluded, there
was a distinct possibility that the truce would be short-lived. Under the
circumstances, Ridgway urged a high-level governmental approach to secure a
written commitment on armed forces and, at the same time, to stop the ROK
antiarmistice campaign.26
While the U.S. political and military
leaders recognized the danger, they doubted that the proper moment had
arrived to negotiate with the ROK Government on the
future control of its military power. To reach an understanding while ROK
emotions were running high might result in the imposition of conditions
unacceptable to the U.N. Command and jeopardize the achievement of an armistice.
Therefore they preferred to work out the terms of the truce first and use the
presence of UNC forces in Korea and the supply and training of South Korean
troops as persuasive points to gain ROK compliance later.27
They were more sympathetic to the
suggestion that President Truman might make an appeal to Rhee to halt the
massive ROK assault on the armistice. On 4 March the President informed Rhee of
the concern of the United States over the ROK attitude toward the truce. He
pointed out that the U.N. unity of purpose in Korea must be maintained at all
costs since divergencies might threaten the support of the U.N. and then issued
a note of warning:
The degree of assistance which your
Government and the people of Korea will continue to receive in repelling the
aggression, in seeking a just political settlement, and in repairing the ravages
of that aggression will inevitably be influenced by the sense of responsibility
demonstrated by your Government, its ability to maintain the unity of the Korean
people, and its devotion to democratic ideals.28
The stress that the President laid on the
relationship between ROK actions and U.N. assistance could not but have its
effect upon President Rhee and his staff. From the close of World War II down
to the outbreak of the Korean War the United States had made substantial
contributions to the South Korean economy. When the war began in 1950, again it
was the United States which had taken the lead in sending military and economic
aid. Food, clothing, and supplies for the thousands who were displaced and for
the sick and wounded were provided not only for humanitarian motives, but also
with the realization that unrest and disease within the UNC area would
complicate the military operations then under way.
The United States also had long-range plans for relief and rehabilitation that
it intended to carry out under U.N. auspices as soon as the war was over. It had
taken the lead in proposing and supporting the formation of the United Nations
Korean Reconstruction Agency (UNKRA) that was established on 1 December 1950 and had provided
the new agency the bulk of its funds.29
Although the
prolongation of the war delayed the effective functioning of UNKRA, the U.N.
Command set up the U.N. Civil Assistance Command in Korea under the Eighth Army
in early 1951 to prevent disease and unrest. Designed to safeguard the security
of the rear areas, UNCACK engaged primarily in relief work, providing consumer
goods to meet the immediate needs of the civilian population. The comparative
inactivity at the front during the truce negotiations permitted reconstruction
and rehabilitation to begin while the fighting was still going on. By the end
of 1951, Ridgway and the UNKRA officials had
fashioned a working agreement that allowed UNKRA to start on a limited
reconstruction program subject to the approval of the UNC.30
Thus the actual control over relief and
economic assistance to South Korea remained under UNC control as long as
military operations continued and for a six-month period after an armistice was
concluded. This, of course, strengthened the hand of Ridgway in his dealings
with the ROK Government. But the exigencies of war and the pouring into Korea of
U.S. money, goods, and services led to a repetition of the U.S. experience in
China in World War II. The undeveloped economy of the ROK, disrupted by war and
essentially agricultural, could not absorb the added purchasing power that large
military expenditures brought into being. While the ROK Government resorted to
the printing press to meet the demands for more currency in connection with
military operations, it could not siphon off the
growing supply of money in circulation by increasing industrial production or by
larger imports of consumer goods. U.S. aid helped somewhat, but the
$150,000,000 that had been expended by 15
September 1951, plus fifty million dollars' worth of
services and ten million dollars in raw materials could not stem the tide of
inflation.31
By January 1952 the ROK financial
situation had become critical. Although the deficit spending indulged in by the
ROK Government and the bank credit expansion practices that were permitted
contributed to the inflationary trend, the ROK officials placed the principal
blame upon the advances in Korean won made to the UNC for military requirements.
They charged that the U.N. Command had failed to settle in dollars for the won
issued and intimated that they would not be able to provide
more currency to the UNC after January.32
By an agreement signed on 28 July
1950 the ROK Government had pledged itself to supply the
currency needed by the UNC and to defer the settlement of claims arising from
this procedure until a time satisfactory to both parties. The hint that the ROK
might not meet its obligation worried Ridgway. He had no objection to making
monthly settlements in dollars for the won
advances as long as the UNC retained some control
over ROK foreign exchange. To help counteract inflation he proposed that the UNC
secure ROK currency by sale of imported commodities to the Korean people and by
purchasing won at
the best rate through any legal source.33 In addition, Ridgway
believed that by making book settlement for UNC services, with no actual use of
money, the amount of currency in circulation could be held
down.34
But the ROK Government balked at permitting
the U.N. Command to maintain control of its foreign exchange and negotiations between the two came to a halt in February. In the
meantime the UNC had drawn eight million dollars' worth of won in January as opposed to
only six million dollars' worth in December and Ridgway asked Van Fleet to give
his personal attention to the problem of holding down expenditures involving the
use of won.35 The gravity of the spiralling inflation can be easily seen in the
increase of currency in circulation between 1
July 1951 and 1 March 1952 - from 122
billion to 812 billion won.36 The impasse
in the financial negotiations and the ever-rising inflation coupled with the ROK
attitude toward the armistice and domestic complications in the ROK Government
prompted Ridgway and Ambassador Muccio to suggest in early March that a
high-level mission be sent from Washington to reach an understanding on the
entire field of ROK-UNC relations.37 Impressed by the urgency of
Ridgway's request, the Department of the Army moved quickly to prepare for the
dispatch of a mission. Defense and State Department approval was soon obtained.
On 28 March the President named Clarence E. Meyer,
head of the Mutual Security Administration mission to Austria, as chief of the
delegation. The Department of State agreed to act as monitor since the mission
was given a broad directive to negotiate "financial, economic and other
appropriate agreements between the United States or the Unified Command and the
Republic of Korea."38
The end result was an agreement signed
on 24 May between the Unified Command and the ROK.
Considering the political turmoil that was rampant in South Korea and the strong
feelings expressed about national sovereignty, the Meyer understanding
represented a fair compromise. The most important
provision established a Combined Economic Board with one Unified Command and one
ROK member to promote effective economic co-ordination. The board would make
recommendations that would be binding on the use of all foreign exchange and
integrate it with the UNC assistance programs. As for the UNC won
advances, the Unified Command agreed to settle up for all advances made between
1 January
1952 and 31 May 1952 at the 6,000-won-to-a-dollar rate.
Claims for 1950-51 would be deferred until a later
date and claims for future months would be paid for at a more realistic rate
than 6,000 to 1. Ten percent of the amount advanced each month would be written
off by the ROK Government as its contribution to the war effort. In addition,
the ROK Government agreed to take internal measures to control inflation and the
Unified Command would attempt to draw won from the market by bringing in
as many salable goods as possible.39
If both sides made sincere efforts to carry
out the terms of this agreement, the economic situation in Korea could improve
considerably in the near future. Whether this might also have a favorable
influence upon the political and armistice problems was another matter. By
May 1952 the armistice negotiations had again reached
a stalemate and ROK agitation against the truce had subsided, but President
Rhee's internal conflict with his fellow politicians threatened to build up into
another crisis. In any case the U.N. Command
might only have adjusted the economic differences in time to be dragged into the
political arena. But, at least, one thorn in ROK-UNC relations had been amicably
removed.
The Japanese Take a Hand
All of General Ridgway's problems behind
the lines did not involve the Republic of Korea directly, but many had an
influence upon events taking place on the peninsula. Across the Sea of ,Japan
new and complicating elements were introduced in late 1951 and early 1952. From
the outset of the war the United States had used the islands of Japan as a huge
supply and staging base for the UNC forces fighting in Korea. In his role as
Supreme Commander, Allied Powers, Ridgway could employ the facilities available
in Japan as he saw fit to support the UNC effort. The signing of the peace
treaty in September 1951, however, foreshadowed a period of change as the
military government closed out its regime and the Japanese civil authorities
once more assumed control of their nation's affairs. In the interim,
arrangements had to be made defining the relationship between the U.S. military
and civil representatives and the Japanese Government and provision had to be
made for the defense of Japan.
Under the Security Treaty signed on 8
September between the United States and Japan, the former was granted the right
to maintain armed forces on the islands until the Japanese could build up
sufficient strength to defend themselves. The conditions governing the
disposition of U.S. troops and the use of Japanese
facilities would be worked out by an administrative agreement between the two
countries.40
Since the end of the war in Korea remained
uncertain and the utilization of Japanese facilities and ports appeared
necessary as long as the conflict continued, the Security Treaty afforded the
legal basis for the continued presence of U.S. forces in Japan. Even under
optimum conditions, it would take considerable time for the Japanese to
organize, train, and equip adequate units to defend Japan on their own. And the
renunciation of war by the Japanese constitution would make the development of
armed forces a delicate matter.
Fortunately, insofar as Japanese defense
forces were concerned, a start had been made in mid1950 shortly after the Korean
War began. When General MacArthur realized that he would have to deploy the
majority of his U.S. units to Korea, he authorized the Japanese officials to set
up a National Police Reserve force of 75,000 men. Although the organization
ostensibly was formed to preserve internal order, the recruits went through a
thirteen-week basic training course during which they became familiar with small
weapons and then moved into an eighteen-week course which stressed small unit
training and used machine guns and rocket launchers. In June 1951 the Police
Reserve engaged in battalion maneuvers. When the armistice negotiations got
under way in July, the force was organized into four infantry divisions of
15,200 men each, but it lacked heavy equipment and had not had sufficient training to qualify for other
than internal security functions.41
Despite these deficiencies, in May 1951
President Truman approved planning and budgeting for sufficient material to
equip ten National Police Reserve Japan (NPRJ) divisions by 1 July 1952. After studying the
political and economic factors involved, Ridgway recommended in September that a
phased expansion to a balanced ten-division force be adopted. The difficult
part, in his opinion, would be the preparation of Japanese public opinion for
training of the NPRJ with heavy equipment and armament. This would have to be
done by the Japanese Government and Ridgway would see Prime Minister Shigeru
Yoshida on this point soon.42
While arrangements for an increased NPRJ
went forward in Tokyo, the Department of the Army came up with some
disconcerting facts. General Collins informed Ridgway in mid-December that a
lack of funds might force the drastic reduction of the NPRJ program. There was
no money available from the Mutual Defense Assistance Program and funds for the
NPRJ had been cut from the Army fiscal year 1953 budget. Under the circumstances
Collins advised Ridgway to revise his plans and endeavor to get along on the
funds already allocated.43
Ridgway was shocked. "It is to me
incredible that from a national defense budget of $57 billion, we cannot find
the relatively meager funds required to support the rapid establishment of a
small Japanese army .... For each dollar expended, it is my
considered opinion that the U.S. can purchase more security through the creation
of Japanese forces than can be purchased by similar expenditures in any other
nation in the world, including the United States." The alternative, Ridgway went
on, would be to maintain U.S. troops at far greater cost in the Far East. He did
not see how he could discuss the NPRJ matter any further with the Japanese until
a firm U.S. policy was forthcoming. Vacillation on the part of the United States
would create a similar response in the Japanese. If the United States desired to
expand the NPRJ more slowly, this might fit in very well with Japanese desires,
but Ridgway regarded the proposed cutback "as nothing less than catastrophic to
the vital interests of our country."44
The fervent plea of the Far East commander
produced a quick reaction in Washington, and by 23 December, Collins was able to
allay Ridgway's apprehensions. Secretary Lovett had approved the inclusion of
three hundred million dollars in the Army budget for the Japanese defense
program.45
Armed with this reassurance, the SCAP staff
reviewed NPRJ planning and recommended on 1
January 1952 that priority during the first stage of
the expansion be accorded to nondivisional combat units, such as antiaircraft,
tank, and artillery, rather than to the formation of new divisions.46 But
further implication from Washington that the UNC program could not be carried
out in its present form led to another round of messages. The upshot of the affair was
that a SCAP delegation, headed by Maj. Gen. William F. Marquat, was sent to the
United States in late January to thresh out the matter. After five weeks of
consultation, the Army and SCAP representatives fashioned a modus operandi acceptable to
both groups. Ridgway would complete the equipping of the four divisions already
established and see to it that they became and remained combat-ready in the
future. During the next fiscal year the NPRJ would be increased to six divisions
with supporting units and the expansion to ten divisions would be phased over
fiscal 1954 and 1955, as funds and equipment became
available.47
Actually the decision to stretch out the
expansion program was not influenced by the lack of money and equipment alone.
As it turned out, the Japanese Government had no desire to move quickly in
rearming the country. Prime Minister Yoshida would not approve an increase of
the NPRJ beyond 110,000 for the 1953 fiscal year. Although SCAP pressed for an
augmentation to between 150-180,000, Yoshida declined to make a commitment until
after general elections were held in early 1953.48 Since Japanese
reluctance to rearm swiftly dovetailed with U.S. financial and production
deficiencies in connection with the program, the slowdown in developing adequate
Japanese defense forces probably represented a workable compromise.
In the meantime ,Japanese public
opinion could be conditioned for the return of Japan
to a position of responsibility in the Far East. The United States intended to
help Japan by assisting the nation to secure access to raw materials and markets
and by encouraging Japanese industry to provide the means by which the country
could once again defend itself. Some offshore procurement of ammunition and
equipment might be arranged to give the Japanese munitions industry a start, but
the Department of the Army felt that the main task had to be accomplished by the
Japanese themselves.49
The re-emergence of Japan as a sovereign
nation spawned a host of other problems as well. To General Ridgway in his role
as Supreme Commander, Allied Powers, one of the most important was the future
status of the U.S. and U.N. military forces in Japan. After the peace treaty was
ratified, the occupation would end and SCAP would be abolished. Before the
latter came to pass, Ridgway wanted to settle the future relationship between
the UNC/FEC and the Japanese Government.
On 18 September he outlined his approach to
the subject to the JCS. Ridgway pointed out that while theoretically Japan would
enjoy full political control, its national security would depend for some time
upon both Japanese and U.S. forces. Since this was a military reality, the Far
East commander felt that he should have adequate authority to counteract any
threat to the security of the U.S. forces under his command and to deal directly
with the Japanese Government on all military matters. His orders should come, as
in the past, from the JCS or their superiors. He would, of
course, coordinate with the chief of the U.S. diplomatic mission to Japan
whenever it was necessary.50
The prospective end of military rule in
Japan and the return of civilian control, however, had a concomitant-the
restoration of normal diplomatic relations and of the pre-eminence of the U.S.
Ambassador in intergovernmental intercourse. On 22
December the Army informed Ridgway that the JCS had
submitted most of his recommendations to the Secretary of Defense, but that the
chief of the diplomatic mission would take precedence over him and be the
channel for all governmental matters except those specifically of a military
nature.51
During February and March, State and
Defense Department representatives worked out further compromises in detail, but
it was not until April that they arrived at an agreement that was satisfactory
to both sides and approved by the President. It provided that the Ambassador
would be responsible for all government relations between the United States and
Japan, but that Ridgway would not be subordinate to him in military matters. The
Far East commander could negotiate directly with the Japanese Government on
security, defense, and military assistance affairs and was authorized to appoint
the U.S. member of the newly formed joint Committee.52
Ridgway had insisted upon receiving his
orders directly from the JCS and their superiors and
this channel of command remained as before. His authority to select the U.S.
representative of the joint Committee also came from the JCS as he had desired.
The Joint Committee stemmed from the Administrative Agreement signed on
28 February 1952 between the United States and
Japan in Tokyo. In the process of establishing the terms under which U.S. forces
would remain in Japan and contribute to Japanese defense, a joint Committee with
one U.S. and one Japanese member was set up for consultation on the
implementation of the agreement. Since such complex matters as the use of ports
and facilities, custom regulations, taxes, postal privileges, and legal
jurisdiction were covered, the joint Committee was held necessary to straighten
out differences of opinion.53
On 28 April the occupation of Japan ended
and U.S. military forces assumed a new and diminished position as guests and
allies rather than conquerors. But since there would be a long period during
which Japanese security would be dependent upon U.S. forces, the Far East
commander and his staff retained considerable prestige. The need for protection
until Japanese defense forces were ready to take over the major responsibility
argued that Ridgway and his successors would wield a goodly measure of influence
in Japanese affairs despite the loss of the bulk of their powers. On the other
hand, civilian ascendancy had been reestablished and the importance of the U.S.
Ambassador was certain to increase as military dependence upon the United States
lessened.
The Far East commander meanwhile had the
problem of finding out where the Japanese forces would fit into the overall
defense picture. Would they fight as separate units or be integrated with U.S.
troops if war broke out? Would they come under U.S. supreme command or remain
under their own leadership? As yet, no intergovernmental arrangement on the
control of Japanese security forces had been reached and the Administrative
Agreement merely provided for consultation between the two governments if
hostilities threatened. These questions would have to be settled definitively
and quickly, General Ridgway believed, and the development of the security
forces of Japan closely correlated with those of the Republic of Korea lest they
get out of proper balance.54 Since the United States was sponsoring
both nations and bad feeling existed between them, the formation of formidable
military forces on one side might eventually lead to an unstable situation
unless it were matched by a similar development on the other. It seemed apparent
by the end of April that although the Japanese were now officially in the game
on their own, the United States would be supplying the stakes with which they
would play. It would be part of the U.S. task to make sure that the Japanese
played along with and not against the other members of the team.
Ammunition Shortages
The complexities of dealing with the ROK
and Japanese Governments seem quite simple when
compared to the perplexing and tortuous labyrinth of ammunition shortages. In
the spring of 1953 a Senate subcommittee conducted a lengthy investigation of
the matter and heard from Van Fleet and Lt. Gen. Edward M. Almond as well as
Washington defense officials. The testimony given revealed the confusion that
existed at the time on the causes and effects of the shortages.55
Much of the confusion stemmed from the lack of background information on the
subject.56
At the end of World War II, the United
States had a tremendous inventory of ammunition on hand, but unfortunately it
was not a balanced stock. There were enormous quantities of some types of
ammunition and only small amounts of others. The hasty demobilization that
followed stripped the Ordnance Department of the military and civilian personnel
that might have properly assessed and cared for the huge inventories of
ammunition in its custody. During the years preceding the Korean War, powder
packed in cotton bags and fuzes made of substitute metals deteriorated. The Army
drew freely upon the big stockpile for training purposes yet made no real effort
to replace consumption or to balance the items in stock. Lack of personnel to
take a complete inventory and the drive for economy among the Armed Forces
contributed to this oversight. Ammunition was expensive and the amounts on hand seemed
adequate for years to come under peacetime conditions.
The lack of postwar orders sent the
ammunition industry into eclipse. Manufacturers converted to civilian goods and
purchased available surplus machine tools to service the booming demands for
consumer items that the war had held in leash. When the United States entered
the Korean struggle so suddenly in 1950, ammunition facilities and plants were
at a low ebb and the prosperity then prevalent made businessmen reluctant to
reconvert their factories to wartime products. Another element that restrained a
shift to the immediate production of ammunition was the prevalent belief that
the Korean War would be short and did not warrant a sizable dislocation of the
U.S. industrial effort. Even after this fallacy was shattered by the entry of
the Chinese into the war in late 1950, the policy of butter and guns continued
and no large-scale mobilization of industry took place.
The sense of complacency that pervaded the
nation during the early phase of the Korean War cost dearly, for valuable time
was lost in getting the languishing munitions industry back on its feet. Under
optimum conditions it took from eighteen to twenty-four months after funds were
voted to produce finished ammunition in quantity. Since Congress did not approve
the first large appropriation for ammunition until early January 1951, this meant that even under optimum
conditions the end products of this money could not arrive on the scene until
late 1952 or early 1953.
In the meantime the U.S. and ROK forces in
Korea had to live off the stockpile. Fortunately, in addition to the
supply of finished rounds of ammunition, there were
also large quantities of component parts available that could be used. Since the
first months of the war were characterized by a high degree of mobility that
required less artillery expenditure, it appeared that the shells on hand and
those that could be readily finished were sufficient to carry the United States
and its allies through the war.57
As the war ground to a slower pace in
mid-1951, artillery assumed a new importance. Static warfare required more
artillery missions to harass and interdict the enemy. This meant that the day of
supply- the average number of rounds that a gun was expected to fire daily over
a considerable period of time-had to be raised.58 Since the day of
supply in turn determined the number of shells that were held in reserve in the
Far East Command, an expansion in reserve stocks followed.59 The
increased demands upon the stockpiles and the knowledge that there was no
possibility of replenishing the heavy consumption of artillery rounds until at
least late 1952 formed the backdrop to the events of the fall of 1951.
Concern over the theater artillery situation began to
arise during the battle for Bloody Ridge in August-September 1951. 2d Division
artillerymen fired over 153,000 rounds during the fight and the 1 5th Field Artillery Battalion
set a new record
for light battalions by firing 14,425 rounds in twenty-four hours. By the end of
the action artillery supplies in the theater reserve were greatly reduced but no
rationing was introduced except for illuminating shells which were in very short
supply.60
Despite denials from the divisions that
ammunition was wasted or misused, thousands of rounds of 105-mm. howitzer
ammunition were hurled fruitlessly against enemy bunkers on Bloody Ridge. The
high trajectory of their fire lessened the chances of direct hits upon the
Communist strongpoints and reduced their penetrating power. The job of knocking
out the pillboxes and bunkers had to be done by the heavier and more accurate
8-inch howitzers with concrete-piercing shells set for delayed firing and by
flat trajectory gun fire. It is interesting to note that after the battle Van
Fleet issued a warning against waste of 8-inch and 105-mm. howitzer ammunition
since these were then in short supply and at the same time the Eighth Army
commander advocated the use of 155-mm. ammunition
instead.61
During the assault on Heartbreak Ridge,
however, Van Fleet imposed no restrictions upon the 2d Division artillery. But
because of the heavy expenditures, local deficits appeared. For example,
4.2-inch mortars had to be used when 81-mm. ammunition ran low and air and rail
shipments to the front had to be made to keep the 4.2-inch ammunition on
hand.62 It was
not surprising that the sustained barrages quickly
consumed the supplies on hand in the firing units since original plans for the
taking of Heartbreak Ridge envisioned the task as a relatively short and simple
one.
The I Corps COMMANDO operation in October
demonstrated another phase in the ammunition saga. When the Communists massed
their artillery against this advance, UNC guns depleted the stores at ammunition
supply points and I Corps had to place restrictions on its artillery units. As
it pointed out later, the I Corps did this not only to replenish the supply
points, but also to compel units to use up the ammunition they were stockpiling
in excess of what they were normally allowed to have on hand.63
Stockpiling was a long-established practice to guard against sudden emergencies
and to provide a cushion in case supplies were temporarily cut off.
Although the experiences during the
August-October period had to do with local and temporary shortages that were due
to a high volume of daily fire, General Ridgway decided to bring the matter to
the attention of the JCS. The withdrawals had left the theater artillery reserve
in a weakened condition and, in Ridgway's opinion, had revealed the danger in
accepting World War II rates of daily fire for the Korean War. World War II
corps had far more artillery battalions assigned to them than did the corps in
Korea and could maintain a lesser rate of fire per gun each day to carry out
comparable missions successfully. With relatively fewer guns and Communist
artillery strength constantly mounting, the U.S. artillery units in Korea had to
fire more frequently. Ridgway argued earnestly for an increase in the
day of supply for his 8-inch, 105-mm., and 155-mm. howitzers and for his 155-mm.
guns, pointing out the grim relationship between artillery and casualties:
Whatever may have been the impression of
our operations in Korea to date, artillery has been and remains the great killer
of Communists. It remains the great saver of soldiers, American and Allied.
There is a direct relation between the piles of shells in the Ammunition supply
points and the piles of corpses in the graves registration collecting points.
The bigger the former, the smaller the latter and vice
versa.64
The increase for his heavy caliber
howitzers and guns were but one part of Ridgway's request. If they were granted,
he wanted to raise the reserve of these shells from 75 to 90 days as quickly as
possible. He in turn would augment the supplies in Korea from 30 to 40 days and
keep 20 days' supply in
the pipeline leaving only 30 days' reserve in Japan.65 But even as
Ridgway sent off his request, he informed Van Fleet that the Eighth Army would
have to live within its ammunition income in November. Since it would take
considerable time to build up the theater reserve again, "There must be no
mental reservation that regardless of disapproval of subordinate commanders
wishes for ammunition that such ammunition will be supplied in case stocks get
low. Your ammunition resources, present and predicted, are as stated above.
Their increase is beyond the capability of this
theater."66
The approval of Ridgway's requests on
20 October did not, of
course, produce an immediate improvement in the ammunition situation in the
FEC.67 But the end of the fall campaign and the negotiation of the
line of demarcation stabilized the battle line and lowered the intensity of the
fighting. The possibility that an armistice might be concluded soon led Van
Fleet to secure Ridgway's permission in early December to bring his ammunition
level up to a forty-five day reserve rather than thirty. Van Fleet feared that
the Communists might succeed in getting a clause freezing ammunition stocks at
their current level written into the armistice and preferred to bolster his own
before this happened.68
At the end of 1951, the ammunition tale
took a new twist. The records of ammunition expenditures during the summer and
fall campaigns evidently were brought to Ridgway's attention and disturbed him
deeply. Although it was too late to do anything about the ammunition already
spent, the Far East commander decided that the phenomenal rates of fire were due
to "either extravagant waste or expenditure of ammunition, or misuse of
artillery, or both." Since excessive use of artillery shells imposed heavier
demands upon U.S. industry and drained raw materials, Ridgway told Van Fleet to
maintain constant supervision lest the performance be repeated.69
This was a serious charge and Van Fleet was
not inclined to let it pass unchallenged. He did not believe that there had been
either waste or misuse of artillery and warned against rigid comparisons of
World War II firings with those in Korea. "Based on World War II European
standards," he went on, "I estimate that the Eighth Army is short approximately
70 battalions of field artillery. Hence, the greatly reduced intensity of field
artillery battalions per mile of front, has required more rounds per individual
tube to achieve the effectiveness required. The effectiveness of one volley from
four battalions is far greater than four volleys from one battalion." He
admitted using his artillery freely to kill the enemy during the offensives,
but, taking a leaf from Ridgway's own book, he reminded the Far East commander
that if he had tried to take the objectives with limited artillery fire, the
casualty lists of the Eighth Army would have been materially higher. In closing,
Van Fleet maintained that he kept a watchful eye on the ammunition level and
that he had conserved a considerable amount of shells during the static
October-December period.70
As the Ridgway-Van Fleet exchange mirrored
the increasing concern in the Far East Command over the situation, supply
officials in Washington offered little hope that there would be improvement in
the calibers that were short until late in 1952.71 Mortar ammunition
and shells for 8-inch guns and 155-mm. howitzers became less plentiful during
the winter months and there was no prospect of relief
in the heavy shell category in the near future.72
The time lag between obligating funds for
ammunition production and the delivery of the finished shells was emphasized
during early 1952. Despite the fact that billions of dollars of contracts had
been let, the end result in many cases was still six months or more in the
offing. In the meantime shortages in the Far East Command became more difficult
to explain to Congress and the U.S. public. Although little was happening at the
front in Korea and efforts were made to restrict nonessential artillery
missions, General Collins felt that expenditures were still too heavy. Pointing
out that two and a half billion dollars of the three and a half requested for
Army procurement in fiscal year 1953 must be spent for ammunition, Collins asked
Ridgway on to March to see whether major reductions should not be made at once
and retained unless large-scale fighting resumed. Ridgway in turn assigned the
problem to Van Fleet.73
Considering the small number of casualties
inflicted upon the enemy during the early part of 1952, the Eighth Army's
expenditures of artillery ammunition appeared rather high.74 But Van
Fleet was quick to remind Ridgway that during the winter months, Eighth Army had used
less than 60 percent of the ammunition allocated it and at present rates would
expend about three-quarters of a billion dollars worth in 1952. Heavy mortar
ammunition was under rigid allocation already and could not be reduced further,
Van Fleet continued. If savings were mandatory, the only category that he could
afford to reduce was interdictory fire. Since 66 percent of Eighth Army's
missions were interdictory, as against 19 percent for counterbattery and 15
percent for meeting enemy actions, Van Fleet was ordering his corps commanders
to cut interdictory fire by 20 percent, but this was as far as he could
go.75
When Ridgway replied to Collins on 9 April,
he had increased the estimate of ammunition costs for 1952 to slightly over one
billion dollars, but after reporting the 20-percent reduction contemplated by
Van Fleet in interdictory fire, Ridgway struck at the heart of the matter:
It still seems to me that the most
fundamental factors in this problem are the ones most frequently obscured by the
search for economies. Those factors are that we are at war in Korea, and that
ammunition must be provided to meet essential requirements, both of expenditures
and stock levels. Provided these requirements are reasonable, economy ceases to
be a factor. The only alternative is to effect savings of dollars by expenditure
of lives.76
By the end of April several facts were
readily apparent. The huge ammunition stockpile left over from World War
II had been a blessing and a curse. For while it had
provided a substantial backlog on which the United States could draw to meet the
demands of Korea, the imbalance in its stocks had gone unnoticed and the very
mass of the stockpile had introduced a dangerous sense of complacency. The
expectation of a short war had fostered this complacency and permitted the
rebuilding of the defunct ammunition industry to be delayed. Compounding the
situation, the lack of industrial mobilization that followed the outbreak of the
war led to further setbacks in the battle for ammunition production. In the
meantime the imbalances had come to light and, as it happened, many of these
were in mortar and howitzer ammunition that were most in demand for the
artillery war that set in from mid-1951on. The tremendous costs of the
ammunition program that were cited in late 1951 and early 1952 reflected the
decelerated pace of the war and served as an excuse for reducing the rate of
expenditure of ammunition. A lower rate of daily fire in turn would help
alleviate the problem of dwindling ammunition reserves in the essential
categories. On the other hand, restrictions in the number of rounds that could
be used each day caused the man at the front to complain and brought the whole
matter to the attention of Congress and the public.
Despite the charges and countercharges in
the ammunition free-for-all, the principal enemy was time. Until production
could begin on a scale that would replenish stocks as well as current needs, the
ammunition crises would go on. The rationing which was adopted in the winter and
spring of 1952 was a temporary expedient to bridge the gap between the decreasing stockpile and new
production, but until the transition was complete, shortages and expedients
would be the rule.
The disadvantages of fighting even a
small war without an adequate production base in being or
capable of quick expansion are readily discernible in the ammunition
situation of 1951 and 1952. Feeding the hungry maw of
the Far East Command drained the reserves in the United States and led to
reductions in allocations for the Army units in Europe. An expansion of the war
might well have been catastrophic for no amount of money or effort could buy the
most priceless commodity-time.
Fortunately, the Communists matched the UNC
in their disinclination to press the fight on the battlefield or to broaden the
war. It appeared that as long as the moderate pace of the conflict continued,
U.S. ammunition supplies would be sufficient to maintain the status quo until
new production took up the slack.
Propaganda Assault
The general indisposition toward
combat in early 1952 confined itself wholly to the
front and did not extend to the battle behind the lines for world opinion. Since
words had proven themselves effective in the matter of the incidents during the
summer and fall of 1951, the Communists began once again to increase the flow and intensity
of their propaganda. As events at the conference table at Panmunjom revealed the
basic differences in approach to the problem still outstanding, the enemy fell
back upon its tried and tested method of exerting pressure upon the UNC by means
of a series of new "incidents."
Although there had been several violations
of the neutral zone and of the agreements made between the Communists and the
UNC on convoys to the Panmunjom area, the enemy's reaction to these breaches had
been mild during December and January. A B-26 light
bomber had strafed a truck in the Kaesong sector on 11 December because of the pilot's
navigational error and another pilot had unloaded a bomb on Kaesong on 17
January instead of dropping his pylon fuel tank. On the following day a
prescheduled air strike on a bridge at Hanp'o-ri, some 18 miles north of Kaesong, caught the
Communist convoy to Panmunjom as it approached the bridge and damaged one of the
trucks. The enemy accepted the expressions of regret in each instance and made
no attempt to use the incidents for other purposes.77
As the negotiations began to bog down over
Items 3 and 4 at Panmunjom, indications of a new propaganda campaign were
disclosed in February. In a United Nations meeting Soviet Delegate Jacob Malik
accused the United States of using poison gas in Korea. While this was not the
first time the charge had been leveled, it seemed significant that Malik had
made it himself. It caused a flurry in Washington since it might be a warning
that the Communists were preparing to employ gas warfare themselves. On the
other hand, the enemy may have discovered that Ridgway had ordered all his
commanders to organize, equip, and train their forces to defend themselves
against chemical, biological, and radiological attack and deduced from this that
the UNC was getting ready to introduce new forms of warfare.78 The
Ridgway order was purely routine, but the enemy could not be certain of this.
At any rate the Communists evidently were
taking no chances and attempted to forestall the possible use of chemical
warfare. Actually the Far East Command was in no position to launch a gas
attack. The theater was not permitted to stock toxic chemicals, in the first
place, and there was also a shortage of over 50,000
gas masks in the Far East. Because the individual
soldier in World War II had frequently been inclined to discard his gas mask,
none had been issued in Korea. Instead they were stored at depots where they
could be distributed within twenty-four hours.79 The absence of toxic
materials in the FEC and the lack of special preparation within the theater to
wage or defend against chemical warfare belied the Communist charges, but as so
frequently happens, accusations, no matter how false, leave residual damage.
Before the furore over the poison gas had
completely died down, the enemy opened a fullscale attack in another quarter. In
late February radio broadcasts from Moscow, Peiping, and Pyongyang openly
charged the United States of conducting bacteriological warfare in North Korea
and Manchuria. Enemy newspapers picked up the story and related how UNC planes
had dumped infected insects and materials and artillery had fired shells filled
with bacterial agents into Communist areas. Complete
with pictures, one article "proved" that on 17 February a UNC plane had dropped
a weapon north of Pyongyang filled with hideous, infected flies that could live
and fly in snowy weather.80
Intelligence reports estimated that the
Communists were not only trying to discredit the United States through this
campaign but also were attempting to cover up their lack of success in
preventing and controlling epidemics and to whip up new enthusiasm for the
Korean War in China and among Communist sympathizers throughout
Asia.81 In 1951 there had been extensive typhus, cholera, typhoid, and smallpox
outbreaks in North Korea and it was quite possible that the enemy expected
reoccurrences and desired a scapegoat.
Despite strong and immediate denials of the
use of germ warfare by Secretary of State Acheson and other officials in
Washington, there was evidence that some Asian countries were lending credence
to the enemy's claims. Both the State and Defense Departments began to show
concern as the attack grew more intense and instructed Ridgway to do all he
could in the way of categoric disavowals if the subject were brought up at
Panmunjom.82 In the meantime the State Department sent an invitation
to the President of the International Committee of the Red Cross in Geneva
suggesting that the United States would welcome a full investigation of the
Communist charges by a disinterested body to reveal the falsity of the enemy
propaganda.83 The ICRC accepted the U.S. offer in mid-March, but
there was little hope that the Communists would have anything to do with
representatives of a committee that they regarded as an agent of the United
Nations Command and not as a disinterested body.84
On 8 March Chinese Foreign Minister Chou
En-lai had hinted at another facet of the antigerm campaign. In a broadcast he
implied that if the Chinese caught U.S. Air Force personnel engaged in spreading
disease over China, they would be treated as war criminals. The Air Force could
not let this threat go unchallenged and the JCS told Ridgway to issue a strong
statement holding the Communists responsible for proper treatment of prisoners
of war. At the same time he could again deny the accusations and warn the enemy
against using an epidemic to mask ill treatment of prisoners. Army G-3 held that
this would allow the UNC to shift over to the propaganda offensive.85
As Ridgway prepared his statement, the U.N.
World Health Organization volunteered to send technical assistance to North
Korea to help combat disease and epidemics and the United States quickly agreed
that the WHO should communicate directly with the Communists on this matter. If
the enemy refused to receive WHO teams, it would tend to discredit the charges
and reflect badly upon the Communist concern for the welfare of their
people.86
Although the propaganda drive gathered
momentum during March and April with the Communists reporting the dropping of
infected spiders, fleas, beetles carrying anthrax, voles carrying plague, and
even poisoned clams in North Korea and in China, the rejection of the ICRC and
WHO offers to investigate the incidents and to aid in the control of disease did
much to weaken the effect of the later claims.87 Besides, the
Communists were about to be given a far more potent propaganda weapon as the
trouble that had been simmering for months in the UNC prisoner of war camps
reached the boiling stage.
Notes
1 See Chapter
IV, above.
2 Msg, CINCFE to CG Eighth Army, g Sep 51,
in Hq Eighth Army, Opnl Planning Files, Sep 51, Paper 9.
3 Msg, CX 50942,
CINCFE to DA, 16 Sep 51, in Hq Eighth Army, Opnl Planning Files, Sep 5
1, Paper 28.
4 (1) Kenneth W.
Myers, The U.S. Military Advisory Group to the ROK, Part IV, KMAG's Wartime
Experiences, 11 July
1951-27 July 1953, pp. 26-27, 189. MS in OCMH.
(Hereafter cited as Myers, KMAG's Wartime Experiences). (2) Memo, Jenkins for
CofS, 9 Nov 51, sub: To Determine What Can be Done Now to Make Better Use of
Korean Manpower, in G-3 091 Korea, 187/7.
5 Myers, KMAG's
Wartime Experiences, pp. 26-27, 189.
6 Ibid., pp. 134-36.
7 Ibid., pp. 128-31.
8 Ibid.
9 Ibid., pp. 201-04.
10 (1) Ibid., pp. 174-79. (2) Memo, Jenkins for DCofS, 21 Nov 51, sub: ROKA Students .
. . , in G-3 350.2 Korea, 5/16.
11 Memo, Jenkins for CofS, 9 Nov 51, sub: To Determine What Can Be
Done Now to Make Better Use of Korean Manpower . . . , in G-3 091 Korea, 187/7.
12 Memo, Foster for JCS, to Nov 51, sub:
Post-Hostilities Military Forces of the ROK, in G-3 091 Korea, 208.
13 See Chapter VI, above.
14 Memo, Bradley for Secy Defense, 23 Jan
52, sub: Post-Hostilities Military Forces of the ROK.
15 Msg, DA
905814, Hull to Ridgway, 9 Apr 52.
16 Msg, G 5347
TAC, Van Fleet to Ridgway, 9 Apr 52, in UNC/FEC, Comd Rpt, Apr 52, an. 1, incl
18. See also Interv with General Van Fleet, in U.S. News and World Report,
vol. XXXII, No. 13 (March 28, 1952).
17 Msg, CX 66647, Ridgway to Hull, 8 Apr
52, in UNC/FEC, Comd Rpt, Apr 52, an. 1, incl 19.
18 Msg, C 65987, Ridgway to JCS, 27 Mar 52,
DA-IN 121000.
19 Ibid.
20 COMNAVFE, Comd and Hist Rpt, Feb 52, pp.
4-2, 4-3.
21 Msg, CINCFE to DA, 18 Nov 51, DA-IN 354.
22 (1) Msg, DA 909826, G-3 to CINCFE, 27
May 52. (2) Myers, KMAG's Wartime Experience, pp. 87-96.
23 New York
Times, August 24,
1951.
24 New York
Times, September
21, 1951.
25 Msg, 070801
American Embassy, Pusan, to SCAP, 7 Dec 51, in UNC/FEC, Comd Rpt, Nov-Dec 51,
an. 1, incl 45.
26 Msg, CX
64241, Ridgway to JCS, 25 Feb 52, DA-IN 109112
27 Msg, JCS
902158, JCS to CINCFE, 27 Feb 52. This message was drafted by the State
Department and cleared with the JCS, Defense Department, and the President.
28 Msg, DA 902912 Eddleman to CINCFE, 6 Mar 52. This transmitted the Truman message to Rhee.
29 For a discussion of U.S. aid policy in Korea see Gene
M. Lyons, "American Policy and the United Nations Program for Korean
Reconstruction," in International Organization, vol. XII, No. 2 (1958),
pp. 180-92.
30 Memo of Understanding between UNC and U.N. Korean
Reconstruction Agency, 21 Dec 51.
31 Msg, CINCUNC to DA, 20
Sep 51, DA-IN 18653.
32 Msg, C 62218, Ridgway to Collins, 25 Jan 52, DA-IN 4572.
33 The legal exchange rate of 6,000 won to the dollar was not considered to be approximate to actual value
of the won. In
January 1952 a rate of 12,000 won to the dollar would have been closer to
the actual value.
34 (1) Msg, CX 60526, CINCFE
to G-3, 31 Dec 51, DA-IN
15295. (2) Msg, CINCFE to G-3, 24 Jan 52, DA-IN 4192.
35 Msg, C 63175, Ridgway to
Van Fleet, 9 Feb 52, in UNC/FEC, Comd Rpt, Feb 52, an. 1, incl 24.
36 Charles N. Henning, Economic and Related Political Factors in Civil Affairs
Operations, Republic of Korea, ORO Study T-211
(Washington: Operations Research Office, Johns Hopkins University, 1952), p.
43.
37 (1) Msg, C
6505, Ridgway to CofS, 10 Mar 52, DA-IN 114192. (2) Msg, C 65121, Ridgway to
CofS, 12 Mar 52, DA-IN 115005.
38 Draft Directive, sub: Terms of Reference
for the Unified Command Mission to the ROK, no date, in G-3 091 Korea, 42/11. A
copy of this directive was sent to Meyer in Japan in April.
39 Ltr, Meyer to Osborn, no
sub, 24 May 1952, to G-3, 091 Korea, 42/16. The United States agreed to pay
$75,000,000 for the January-May 1952 period and an initial payment of
$35,000,000 was made on 29 July.
40 See the text of the treaty in Department
of State Bulletin, vol. XXV, No. 638
(September y, 1951) , pp. 463-65.
41 CINCFE G-3 Presentation to
Asst Secy Army Alexander, no date, in G-3 091 Korea, 187/7.
42 UNC/FEC, Comd Rpt, Sep 51, pp, 35-36.
43 Msg, DA 89795, CofS to CINCFE, 18 Dec 51.
44 Msg, C 59752, Ridgway to JCS, 20 Dec 51, in UNC/FEC,
Comd Rpt, Apr 52, an. 4, incl 1.
45 Msg, DA 90318, CofS to CINCFE, 23 Dec 51.
46 UNC/FEC, Comd Rpt, Dec 51, pp. 43-44.
47 Msg, DA 902603, CofS to
CINCFE, 4 Mar 52.
48 Memo, Civil Affairs Sec SCAP to CofS SCAP, 28 Feb 52,
sub: Conf by SCAP with Prime Minister Yoshida, in UNC/FEC, Comd Rpt, Mar 52, an,
1, incl 11.
49 Msg, DA 902855, DA to SCAP,
7 Mar 52.
50 Msg, C 50742, CINCFE to
JCS, 13 Sep 51, in UNC/FEC, Comd Rpt, Sep 51, an. 4, incl 10.
51 Msg, DA 90317, G-3 to CINCFE, 22 Dec 51.
52 (1) Msg, JCS 905965, JCS to
CINCFE, 10 Apr 52. (2) Msg, JCS 907213, JCS to CINCFE, 25 Apr 52.
53 See text of Administrative Agreement of 28 February
1952, in Dept of State Bulletin, vol.
XXVI, No. 663 (March 10, 1952) , pp. 383ff.
54 (1) Msg, C 66619, CINCFE to
JCS, 9 Apr 52, in UNC/FEC, Comd Rpt, Apr 52,
an. 1, incl 12.
(2) Msg, C 67740, Ridgway to DA, 29 Apr 52,
DA-IN 133087.
55 Hearings
Before the Preparedness Subcommittee No.
2 of the Committee on
Armed Services, U.S. Senate, 83d Congress, 1st session, on Ammunition Shortages
in the Armed Services, 1953.
56 The following summary is based upon the excellent study made by
the former Deputy Assistant Chief of Staff, G-4, Maj. Gen. William O. Reeder,
after the war was over, and entitled: The Korean Ammunition Shortage. Copy in
OCMH files.
57 It should be
noted that small arms ammunition was always plentiful and caused no concern.
58 The day of
supply was based upon World War II experience.
59 The reserve was computed by
multiplying the day of supply by the number of guns on hand and then multiplying
the result by seventyfive days, which was the safety-level factor in case
deliveries should be halted or cut off for a period.
60 Williamson et al., "Bloody Ridge," ch. V
Aug-Sep 51, pp. 27-28.
61 Hq Eighth Army, Comd Rpt,
Sep 51, sec. I, Narrative.
62 Williamson et al.,
Action on "Heartbreak Ridge," p. 32.
63 U.S. I Corps, Comd Rpt, Oct 51, sec. I, pp. 63, 75.
64 Msg, CX 58171, CINCFE to JCS, 17 Oct 51, in FEC G-3
471 Ammunition.
65 Ibid.
66 Msg, CX 58155, CINCFE to CG EUSAK, 17
Oct 51, in Hq Eighth Army Opnl Planning Files, Oct 51, p. 17. The Department of
the Army daily rate authorized was: 50 rounds of 150-mm., 33 rounds of
155-mm., and 20
rounds of 8-inch howitzer shells. Ridgway had asked
that these be raised to 55, 40, and 50, respectively.
67 Msg, DA 84571, DA to CINCFE, 20 Oct 51.
68 UNC/FEC, Comd Rpt, Nov-Dec 51, CofS sec., an.
1, p. 8.
69 Msg, C 60169, Ridgway to Van Fleet, 26 Dec 51, in FEC G-3 471
Ammunition.
70 Msg, G 3789 TAC, Van Fleet to Ridgway, 29 Dec 51, in
UNC/FEC, Comd Rpt, Nov-Dec 51, CinC and CofS sec., an. 1, incl 12.
71 Memo, Magruder (G-4) for ACofS G-3, 28 Dec
51, sub: Augmentation from
FEC, in G-3 320.2 Pacific, 79/1.
72 Memo, Col Davidson, G-3, for Asst CofS G-3, 29 Feb 52,
sub: Rpt of Staff Visits During Period 23 Jan-19 Feb 52, in G-9 333 Pacific, 1.
73 (1) Msg, DA 903815,
Collins to Ridgway, 10 Mar 52. (2) Msg, C 66253, Ridgway to Van Fleet, 2 Apr 52,
in UNC/FEC, Comd Rpt, Apr 52, an, 1, incl 8.
74 In January 1952, FEC expended 57,000
tons of ammunition costing about $114,000,000. Estimated enemy casualties were
about 20,000, so that each enemy casualty on the average took $5,800 worth of
ammunition before he was injured or killed. See Check Sheet, EKW [Wright] for
CofS, 31 Mar 52, FEC G-3 471 Ammunition.
75 Msg, G 5322 TAC, Van Fleet to Ridgway, 7
Apr 52, in UNC/FEC, Comd Rpt, Apr 52, an. 1, incl 9.
76 Msg, C 66608, CINCFE to Collins, 9 Apr
52, in UNC/FEC, Comd Rpt,
Apr 52, an. 1, incl
10.
77 (1) Msg, HNC
535, Joy to CINCUNC, 11 Dec 51, in FEC Msgs, Dec 51. (2) Liaison Officers Mtg at
Panmunjom, 23 Jan 52, in G-g Liaison Officers Mtg at Panmunjom, bk. II, 1952.
78 UNC/FEC, Comd Rpt, Nov 51.
79 Check Sheet, JHR [Col Jacquard H.
Rothschild] to G-3, 9Feb 52, sub: Questions Arising From Statement Made By
Soviet Delegate Malik Before U.N., in FEC G-3 471.6 Bombs, etc.
80 UNC/FEC, Comd Rpt, Mar 52, p. 31. For an
interesting discussion of the germ warfare charges of this period, see John
Clews, The Communists' New Weapon- Germ Warfare
(London: Lincoln Pragers, 1953) . See also the
statement of U.S. Representative to the UN Assembly, Ernest A. Gross, 27 March
1953 and 8 April 1953, in Dept of State Bulletin, vol. XXVIII, No. 722 (April 27. 1953)
81 UNC/FEC, Comd Rpt, Mar 52, p. 32.
82 Msg, JCS 903060, JCS to
CINCFE, 7 Mar 52.
83 (1) Msg, DA 903096, G-3 to CINCFE, 8 Mar
52. (2) Msg, C 64368, CINCFE to DA, 9 Mar 52, DA-IN 114029.
84 Msg, JCS 903547, JCS to CINCFE, 14 Mar
52.
85 (1) Memo,
Eddleman for CofS, 11 Mar 52, sub: Chinese Communist Threat . . . , in G-3 385,
8. (2) Msg, JCS 903686, JCS to CINCFE, 15 Mar 52. This message was drafted by
the Air Force and cleared by the JCS, Defense and
State Departments, and the President.
86 (1) Msg, C 65348, Ridgway to JCS, 16
Mar 52, DA-IN 116709. (2) Msg, JCS 903780, JCS to CINCFE, 17 Mar 52.
87 (1) UNC/FEC,
Comd Rpt, Apr 52, p. 24. (2) Clews, The
Communists' New Weapon- Germ Warfare, pp. 14-24.