At the outbreak of hostilities in Korea in June 1950, U.S. military Korean linguists were almost nonexistent in the FEC. Against an authorization at that time for 158 Korean linguists, there were fourteen available. Of these fourteen, only seven were sufficiently fluent to be usable in all situations. Korean language requirements prior to June 1950 had been met almost entirely by hire of third-nationa1 linguists; the few military linguists were used in situations where, for security reasons the foreign-national linguists were unacceptable.

While the actions listed below were by no means the most desirable solutions, nevertheless they were ones that could be effected with least loss of time and would still permit accomplishment of the assigned intelligence missions. Solutions are shown in order of importance and ultimate value.

1. Use of Japanese as an intermediate language in the interrogation of POWs. As a result of almost forty years of occupation by the Japanese, the majority of Koreans speak Japanese as a second language. This permitted U.S. military Japanese linguists, of whom there was a good supply, to be used to advantage in collection of enemy intelligence. In certain cases where POWs were unwilling or unable to speak Japanese, South Korean nationals and ROKA personnel were used in a three-langauge interrogation—English to Japanese to Korean. A similar system was used in translation of enemy documents wherein Korean


nationals translated documents into Japanese which could then be put into English by U.S. military linguists.

2. Employment of English-speaking Korean nationals secured through co-operation of Korean Mission in Tokyo and local hire of qualified Koreans by Eighth Army.

3. Establishment by AFFE Intelligence School of Korean Language Conversation Course. This course was designed to produce Korean linguists capable of performing Army-level interrogations from strong Japanese military linguists over a six-month period. Course was begun on 1 December 1952, with new classes monthly thereafter.

4. Request to Department of the Army for immediate airlift of thirty Korean linguists to FEC.

5. Request to Department of the Army for expansion of Korean-language classes at Army Language School.

The use of Japanese-speaking U.S. military linguists was the most effective of all solutions in the over-all effort to make up for the almost complete lack of Korean military linguists. In fact, the scarcity of these specialists made necessary the continued use of this method in varying degrees until hostilities ceased in July 1953. However, the use of three languages, Korean, Japanese, and English, in interrogation or translation had definite drawbacks. The most serious of these were: (a) inevitable loss of accuracy which would result from double transposition (Korean to Japanese and Japanese to English and vice versa) and (b) the additional time required for this round-about process.

Use of English-speaking Korean nationals procured in Japan and Korea proved almost as successful as the above method. Supervised by a few military intelligence specialists (both linguist and non-linguist), this group performed outstanding work in interrogation and translation fields. Disadvantages were the comparative scarcity of bilingual Korean nationals and also the security risk involved in the use of such personnel. However, the need outweighed possible


breach of security, and this group was used to good advantage throughout the Korean conflict on a calculated-risk basis.

Continuing shortages of military linguists in FEC made it imperative that steps be taken to supplement by local training the meagre flow of these specialists from ZI. In December 1952, AFFE Intelligence School inaugurated language conversation courses from Japanese to Korean and from Cantonese dialect to Mandarin dialect Chinese. Students were selected both from pipeline and from units in FEC. Prerequisites were:

1. Fluency in Japanese or Cantonese and English.

2. Ability to read Japanese or Chinese and English.

3. GCT of 100 or higher.

4. Twelve months' retainability in FEC. New classes in each language were begun each month, and by 1 December 1953 approximately 120 linguists had been graduated—equally divided between Korean and Mandarin Chinese. Although the goal of the school was to train linguists capable of Army-level interrogation, in numerous instances graduates have been sufficiently fluent to be used as interpreters by various components of UNC Military Armistice Commission.

Unquestionably, intelligence operations in FEC were hampered by the acute and persistent shortage of trained, experienced military linguists. Under pressure of combat requirements, every effort was made to solve the linguist problem in FEC. While this effort has guaranteed the timely exploitation of invaluable intelligence information, it has been characterized to a great extent by resort to improvisations and temporary expedients. For example, the largescale use of non-military, foreign nationals is at best merely a stopgap; at worst, a serious security risk. In the final analysis, the Oriental linguist problem cannot be solved permanently until


provisions have been made for maintenance of an adequate pool of U.S. military linguists, provided through normal training channels.

GHQ, FEC had anticipated the possible intervention by the CCF, and in November 1950, G1 FEC was asked by G2 to requisition from Department of the Army thirty Mandarin Chinese linguists against possible future need. As of December 1950, only nine had been received from ZI, none of whom were considered qualified. By June 1951, seventy-five potential Mandarin Chinese linguists had arrived from ZI, of whom only sixteen were capable of performing language duties. An attempt to procure Japanese national Chinese linguists in November 1950 was unsuccessful.

Since the procurement of Mandarin linguists from local sources and from the ZI had proved unsuccessful and unsatisfactory, recruiting of qualified linguists from Formosa remained as the only practicable solution. By March 1951, seventy-five fluent Mandarin Chinese linguists had been received from Taipei and had been passed on to UNC units. Since that time additional Nationalist Chinese have been obtained from Formosa by UNC for linguist assignments with such organizations as Psychological Warfare, CI&E, and ASA.

Of the two conversation courses begun at AFFE Intelligence School in December 1952, the Chinese conversion course from Cantonese to Mandarin dialect was the more successful. Its graduates played a large part in alleviating the chronic shortage of Mandarin Chinese linguists in the theater. The number of usable linguists produced by this training has exceeded the number of usable military Chinese linguists received from ZI since August 1950.

The Far East Command Intelligence School (later to become United States Army Forces Far East Intelligence School) was activated in November 1951. The first classes produced much needed Order of Battle and Photo Interpreter Specialists for use by Eighth Army.


Since that time AFFE Intelligence School has graduated, depending upon the need and the supply obtainable from CONUS, both officer and enlisted intelligence personnel, not only for Army units, but for Navy, Marines, and Air Force as well. On 2 December 1951 MISG/FE (later 500th MI Svc Gr), as an operating agency of G2, GHQ, FEC, was made responsible for the supervision and control of military intelligence specialists in FEC.

The solution provided by the establishment of the AFFE Intelligence School has been eminently successful in supplying intelligence specialists to the command when such specialists have been unobtainable from the United States. The program of the School has been kept sufficiently flexible so that its output could be increased or decreased to meet changing requirements of units in the theater. The designation of a single agency of G2, responsible for proper utilization and assignment of intelligence specialists, has reduced to a minimum the malassignment and pipeline losses of this personnel scheduled for assignment in the Far East Command.



Although many enemy soldiers infiltrated our lines as refugees in the early stages of the war, those who were picked up for interrogation did not present too much of a problem as to identification. They apparently had not been instructed on methods of resisting interrogation or the safeguarding of military information and, consequently, co-operated with the interrogator. Early reports on file at 500th MI Svc Gr indicate that the POWs gave unit identification and any other information they might have had quite willingly. However, when ADVATIS switched to strategic interrogation on 28 October 1950 and the tempo of the attack was stepped up to the point where the number of POWs was multiplied by thousands, the entire system of handling POWs fell apart. Approximately two per cent only of the POWs arriving at Pusan were properly tagged. Some even did not halt at the forward POWEs but were moved directly to Pusan by LST. With the POWs not being tagged prior to arrival at the ComZone POWE, it was impossible to check back in an attempt to identify them. This condition of not being able to identify or even locate a POW who was desired for further interrogation lasted to the time of the repatriation program in 1953.

The lack of identification of POWs resulted in the following:

1. Inability of the interrogation personnel to locate specific POWs for interrogation when preliminary reports indicated the POW to be a prime source of strategic information.

2. Interchange of identity cards between POWs or the complete discarding of cards. This allowed some knowledgeable sources (particularly officers) to assume the name and POW number of an enlisted man and thus avoid interrogation.

3. Confusion in the records of the POW camp personnel to the degree that such records were of little or no value to interrogation personnel. At times, ten to fifteen POWs had to be screened to determine the one being sought.


Actually, no segregation of POWs, which was desirable from the viewpoint of intelligence personnel, existed until the G-2 target camp was constructed and began operations on 6 February 1952. As a matter of fact Enclosure 10 (formerly Enclosure 1 and later POW Camp 2) was under the control of the 14th Field Hospital; POWs were segregated on a medical basis and no consideration was given to proper segregation as outlined in FM 19-10. However, the other POW camps fared no better as far as segregation was concerned, and all of them were greatly overcrowded. One compound at Enclosure 10 contained over two thousand POWs, not segregated by rank, nationality, or ideological group. Those who were earmarked for detailed interrogation were eventually segregated from the rest of the camp. The officers were separated from the enlisted men, but ALL went to the same spot for their food. Communication between compounds throughout the camp was no problem as the camp was not constructed to preclude this practice.

On 16 April 1952, the MISG/FE (ADV) potentials were screened and segregated into two groups: Those desiring repatriation were segregated from those who did not.

This was a step in the right direction, but was still not segregation to the degree desired by interrogation personnel. There were several glaring weaknesses in this method: First, no consideration was given to the ideological leanings of the POW. Second, those desiring repatriation for compassionate reasons were placed in a compound with those who desired repatriation for ideological reasons.

The procurement of POWs for interrogation presented problems for several reasons: The prime concern of the POW camp commander was the administration of the camp. The lack of adequate records made it difficult, sometimes impossible, to locate the POW's compound; some compounds were entirely controlled by communist elements, and neither the camp commander nor the interrogators could procure a POW from those compounds. Logistical problems were involved in moving the POWs to the interrogation center.


It was the policy to ship all POWs to the island of Koje-Do for quarantine. Since conditions on Koje-Do were not conducive to good interrogations, POWs had to be shipped back to Enclosure 10 for interrogation. Many delays were encountered. It was decided on 5 June 1952 to retain the POWs selected for interrogation in Enclosure 10 until their interrogation was completed and then ship them to Koje-Do.

This agreement solved the procurement problem except for the withdrawal for administrative functions of POWs undergoing interrogation. Hospital cases could not be screened until released from the hospital. As segregation was based on a purely medical foundation, this resulted in collusion among POWs and, in some cases, POWs possessing valuable information had been intimidated.

Until the G-2 Target Camp was placed in operation, facilities for handling POW intelligence potentials and interrogation quarters were inadequate. The lack of good communication facilities and the existing political situation within the camps at Koje-Do precluded operations on the island.

At Enclosure 10 the interrogation unit had its interrogation compound at one end of the camp. Interrogations were carried on in winterized tents which, during the frequent power failures, had to be opened continually to admit sufficient light. Poor light, swarms of insects in warm weather, and shivering interrogators and POWs were not conducive to a maximum effort in collecting intelligence information. Many hundreds of man hours of interrogation time were lost due to the above factors. As strategic interrogations usually lasted several weeks, and frequently several months, the POWs soon lost interest under the prevailing conditions.

The lack of proper segregation had a detrimental effect on the strategic interrogations being conducted. Those who were awaiting interrogation were exposed to communist agitators and to those who had been or were still being interrogated. Frequently an internee who apparently had only meager information one day was able to give


detailed information the following morning, a clear indication that he had been briefed during the night. There were also occasions where an apparently knowledgeable source would "clam up" and refuse to divulge additional information—the result of intimidation and/or indoctrination. Since POWs were segregated according to their desires for repatriation and not on ideological grounds, marlin man hours of interrogation time were lost as some of the POWs were complaining that they were not communists but were forced to live in a communist compound.

On many occasions copies of interrogation reports of forward elements reached ADVATIS (MISG/FE (ADV) 511th MI Svc Co) too late to be of any value. The lack of understanding of the "Phase" system of interrogation resulted in a duplication of information in the various reports. Personal notebooks and other documents which could have supplemented the interrogations never reached the rear areas where strategic interrogations were conducted.

The emphasis on personnel assigned to interrogation units seem to have been placed on the ability of the individual to speak a foreign language rather than on his general background. Many had little or no military background or knowledge of training, equipment, or operations. The inability to differentiate between positive and negative military information and to recognize valuable prisoners led to many hours wasted in interrogating useless prisoners and brought about a lack of enthusiasm on the part of the interrogator. In some instances officers appointed as team captains lacked aggressive leadership and the over-all ability to command troops. There were many instances of enlisted personnel being drafted, given basic training, and assigned to strategic interrogation on the basis of their ability to speak the Japanese language. Due to their youth—often nineteen and twenty years of age—and lack of military experience, these personnel not only were not qualified for strategic interrogations but were also disinterested in the entire program.


The 511th MI Svc Co, augmented by the attachment of Detachment 5, 6004th AISS, was activated as the final outcome of numerous conferences and letters to set up a strategic interrogation unit. Both the 511th and Detachment 5 were further augmented by DACs and DAFCs to maintain the adequate working strength needed to conduct interrogations of POWs and refugees. By gearing the criteria of selection of interrogees to the flow of POWs, adequate coverage of intelligence information and a constant state of training was maintained.

Since no photographs or fingerprints were taken in the forward area, no positive means of identification were available by the time the POW reached the strategic interrogation unit. The 511th MI Svc Co set up a fingerprinting section when the G-2 target camp was established. Each POW was also given a camp identification number. His meant that it was possible to make a positive identification of the POW from that point on. As the preliminary interrogation reports accompanied the POWs to the G-2 Target Camp, it was possible to identify each POW with his interrogation report. In cases of doubt, agents planted in the compounds usually found out the true identity of the POW. This was an improvement over the system previously employed at the POW camps. No photographic equipment was available.

The initial segregation of POWs according to their desires for repatriation was a step in the right direction, but did not prevent contamination. G-2 AFFE, recognizing that the construction of POW camps did not provide the degree of segregation required for intelligence-gathering purposes, approved construction of a special camp and interrogation center which was known as the G-2 Target Camp. The mission of operating the camp was given to the CO, 511th MI Svc Co. In addition to the regularly assigned interrogation and administrative personnel, Group Headquarters attached a supplementary group to provide for the internal security and the administration of


the Target camp. These additional personnel were given intelligence training and schooled in the most desirable methods of segregating and handling POWs.

A new method of screening was used as a basis for segregation. First, a POW was screened to determine his ideological leanings, all incoming POWs being divided into two groups: Communist and non-Communist. Then the Chinese were separated from the Koreans and the officers from the enlisted men. The above were further divided into those who were pending interrogation, those undergoing interrogation, and those who had been interrogated. Those who were not selected for interrogation were shipped out through normal POW channels. AT NO TIME, HOWEVER, WAS THE POW QUESTIONED AS TO HIS DESIRES CONCERNING REPATRIATION; for, had segregation been made on this basis, the same situation would have existed as did in the POW camps. With segregation being made according to the ideological leanings of the POWs, however, there was no mixing of communist and anti-communist elements and the problems previously experienced ceased to exist.

Since the administration of the G-2 Target Camp was keyed to meet the requirements of the interrogation center and since, with the exception of the sick and wounded, all POWs were channeled through the Target Camp, the problem of procuring POWs for interrogation was solved. The logistical problem which existed earlier was solved by holding all POWs scheduled for detailed interrogation at the Target Camp until the completion of such interrogation. Sick call for POWs was held prior to the arrival of the interrogators, and no POWs were withheld for work details or administrative purposes. Consequently, POW targets were always available for interrogation.

Facilities for the interrogation of POWs were ideal. The interrogation center consisted of ten prefabricated buildings, each subdivided into ten interrogation booths and having a large space at the end for the team captain's desk and for storage of reference


material and maps. Latrine facilities, as well as a waiting room, were provided in another building so that it was unnecessary for either interrogators or interrogees to leave the interrogation compound. Still another building housed the operations and administration personnel, the reference library and a conference-schoolroom. All buildings were well lighted and contained additional outlets for lamps. Generators were kept on a stand-by basis in the event of failure of the local power sources. All buildings were well ventilated and contained sufficient space heaters to warm the rooms during the coldest days. In addition, since camp personnel fired the heaters in the early morning hours, the buildings were warm when interrogations were begun and no interrogation time was lost. Messing facilities, showers, dispensary, and recreation areas were available for both interrogation personnel and POWs.

With segregation made on the basis of ideological screenings, the problem of indoctrination of POWs continued but on a small scale, and only in a manner that was an aid rather than a detriment to the interrogation effort. Indoctrination by anti-Communist POWs of those internees who were ideologically undecided or indifferent frequently resulted in a more co-operative attitude on the part of the POW and in a saving of interrogation time and an increase in information obtained. Moreover, such indoctrination was strictly controlled by the camp and interrogation personnel through the use of agents and informers within the compounds. On the other hand, since the number of Communist POWs being captured was small, normal segregation by rank and nationality so diminished the number in each compound that routine surveillance prevented procommunist POWs from indoctrinating or coercing the others.

A natural tendency to impress the interrogator favorably resulted in the briefing of POWs under interrogation by other POWs in the same compound. However, this problem was overcome by several methods. First, the interrogee was made to understand that only his


personal knowledge or qualified hearsay was of value to the UNC. Second, POWs were not allowed to mingle while within the interrogation compound. Third, the compound leader was instructed to prohibit briefings within his compound, and violations were reported by agents and informers.

Conferences between the CO, 511th MI Svc Co, the operations officer, 302d MI Svc Co, and the CO, POW Escort Guard Co, KCOMZ resulted in a verbal agreement concerning prompt delivery of POW Document Envelopes. It was agreed that a POW document envelope for each POW shipped be handed over to the POW Escort Guard officer. Each envelope was to contain a copy of each preliminary interrogation report written on a POW, or an envelope was sent with a notation giving the reasons why a report had not been published. Envelopes of suspected communists or agents were marked with a code that had been previously agreed upon. Hospital cases were also coded. These envelopes were turned over to the POW Escort Guard officer, who in turn gave them to the G-2 Target Camp Commander. For the first time documents were arriving at the same time as the POW, a situation which proved extremely useful in the screening process.


With the fall of PYONGYANG imminent in October 1950, it was evident that the North Korean capital would yield a wealth of intelligence from documentary material if such documents could be seized intact. Upon receipt of an order from CG, Eighth United States Army Korea, dated 16 October 1950, and Special Operational Instrnotions, HQ Eighth United States Army Korea, dated 17 October 1950, the CG, 2d Infantry Division, on 16 October directed the Division G-2 to organize and command a company-size task force which was to seize and secure designated targets in PYONGYANG containing documents of intelligence value. This unit was designated Task Force Indianhead.

The task force consisted of one platoon of medium tanks (not committed), one reconnaissance platoon, one infantry company (less one platoon), one engineer squad (demolition experts), aid men as


selected, and G-2 personnel organized into seven task force teams, each of which consisted of one CIC agent (in charge), one assistant CIC agent, one Korean guide, and two Korean linguists. The mission of the Task Force was to enter PYONGYANG with advance elements of I US Corps and/or II ROK Corps and to seize and secure documents relevant to military operational plans (especially those concerning past and expected future aid from foreign countries). Two secondary missions for the Task Force were Target Project Alpha, which was to locate certain personalities in PYONGYANG, and Target Project Beta, which was to locate certain NK organizations within the city.

By 23 October the final exploitation of all primary and secondary targets had been accomplished.

An additional twelve targets of opportunity were exploited by 25 October 1950.

During the period 21 October 1950 through 29 November 1950 a total of 35,393 documents, comprising 3,834,508 pages, was shipped beck to Hq 500th MI Svc Gp in Tokyo. Of this total, 564 documents, comprising 27,261 pages, were classified "A" documents, or documents of theater interest.

The basic plan was sound and practicable; however, a number of problems arose which could have been avoided if more time had been spent in training and planning. The Task Force lacked specialized personnel such as experienced intelligence analysts, personnel experienced in the exploitation of documents, and technical intelligence personnel. Documents and equipment could have been more efficiently classified if such personnel had been attached to the Task Force. Operations would have progressed more smoothly if tactical commanders had been notified of the possible employment in the Task Force in their zones of action. When organizing a Task Force of the size and scope of Indianhead, provision should be made for training of tactical personnel in seizing and scouring targets, house-to-house fighting techniques, and the protection of target papers, books, documents, and material.



The 500th Military Intelligence Service Group (MISG) was the agency responsible for the co-ordination and exploitation of all captured documents taken in the Korean operation. This centralized control was to ensure the maximum utilization of scarce-category linguists and the maximum exploitation of intelligence from captured documents for dissemination to all levels of command.

With the co-operation of G-2 ROK Army, G-2 Eighth Army, and G-2 KMAG, a series of visits was made at appropriate times to G-2 ROKA by personnel from HQ 500th MISG. These visits were made for the purpose of establishing operating procedures whereby all documents would be fully exploited to the mutual advantage of both sides.

An agreement was reached whereby all captured documents, irrespective of capturing units, would be forwarded to Hq 500th MISG through 302d MISG. Documents captured by ROK Army units were to be screened at Hq 500th MISG and then returned to Section V, G-2, ROK Army through 302d MISG within a period of forty-five days. This was considered an improvement over previous arrangements, since it afforded for the first time full exploitation of all captured documents without undue delay; however, it denied UNC Headquarters the right to retain indefinitely such documents as personal identification cards, etc. which were required for covert purposes.

In August 1952 a visit was made to assist G-2 Eighth Army in the exploitation of documents captured by ROK Army and to co-ordinate processing of documents between 302d MISG and Hq 500th MISG. During this visit a total of 11,250 pages of documents not previously reported by ROKA were scanned, of which 2,395 pages were referred back to Hq 500th MISG for full exploitation.


page created 10 January 2002

Return to the Table of Contents

Return to CMH Online