Refugees in the snow, Korea, 1950

The Changjin Journal is designed to disseminate and solicit information on the Chosin campaign. Comments and brief essays are invited. Subject matter will be limited to history of the Chosin campaign, as well as past or present interpretation of that history.
See End Notes for distribution and other notices. Colonel George A. Rasula, USA-Ret., Chosin Historian,
Byron Sims, Contributing Editor

You can use Ctrl-F to search for words within this page 

Search the Changjin Journal and the entire NYMAS site:


THIS ISSUE of the Changjin Journal is based on Frozen Chosin from a point of view of participating services. In doing so it will be critical of the handling of the Chosin story in past history where omissions in historical fact can misdirect the reader's point of view. In the pamphlet's 132 pages the reader will view about 150 photographs and 11 maps in addition to other illustrations. We regret that many of the maps are not sufficiently detailed nor related to nearby text. When possible this journal will be presented as a filler to that already published in the pamphlet.


The newly published pamphlet by the Marine Corps titled FROZEN CHOSIN: U.S. Marines at the Changjin Reservoir, by retired Marine historian BGen Edwin H. Simmons, is excellent reading. However, based on past historical writings, it does not tell the whole story. As stated in the masthead, this publication is "one in a series devoted to U.S. Marines in the Korean War era, is published for the education and training of part of the DOD observance of the 50th anniversary of that war." We thank General Simmons for including the Changjin Journal in his list of sources, as well as recognition to a select few who offered reviews of portions of the draft document in its early stages.

At Hyesanjin on the Yalu River, left to right, BG Homer Keifer, 7th DivArty commander; BGen Henry Hodes, 7th Div ADC; MGen "Ned" Almond, X Corps commander; MGen David Barr, 7th Div commander; and Col Herbert Powell, commander of the 17th Infantry Regiment which made it to the Yalu River on 21 November. Powell is wearing the standard pullover (reversible) parka shell, the same as worn by other 7th Division troops who went to the Chosin Reservoir. Gen Hodes appears to be wearing a button-down version of a similar reversible parka shell. Keifer on the left is wearing a standard officer's trench coat while Gen Barr is wearing a hooded parka similar to that worn by the Navy and Marines. We leave Gen Almond's garment up to the viewer, a long jacket with fur lined hood. All seem to be wearing shoepacs, although Almond could be wearing well shined combat boots.


(p.2633) Here we read eight pages devoted to the problem of cold during the Chosin campaign. Of interest is the old school that believed socks and insoles could be dried by placing them next to the body, originating from the school of old soldiers who had no experience in winter warfare. There had been a tendency to "fight the cold" rather than understand what one must do to care for himself and his equipment. The belief by some that temperatures ranged from 35 to 40 degrees below zero is well handled by the author by informing the reader that thermometers essential for accurate firing by artillery recorded 20 to 25 below, as noted in the personal memoir of General Smith.

Dealing with the cold at Chosin is a subject that warrants more attention. The problem that existed then was caused by the circumstances of that day in time. Soldiers did not have "winter parkas, shorter and less clumsy than the Navy parkas." This was a World War II 10th Mountain Division reversible parka shell, white on the inside, that served only as a wind breaker and camouflage, relying on field jacket with pile lining and other clothing to be worn beneath. As an outer garment designed for ski troops practicing stem christies on the slopes at Vail, Colorado, it was extremely impractical in the cold at Chosin while being the only outer garment many soldiers had to wear.

One may wonder why bundles of white bed sheets readily available in Japan were not air dropped to provide for improvised camouflage. Even before Chosin, research in cold weather operations was taking place by the Army in Alaska, resulting in far better clothing and equipment, as well as updated concepts of operations in cold and snow, the results of which would be seen in Korea the following winter.

[INSERT Lessons Learned]


"I find it amazing that highly trained professionals with extensive combat experience could have approved and tried to execute the tactical plan of operations for the X Corps in northeast Korea in November 1950. It appears like a pure Map Exercise put on by amateurs, appealing in theory, but utterly ignoring the reality of a huge mountainous terrain, largely devoid of terrestrial communications, and ordered for execution in the face of a fast approaching sub-arctic winter." General Matthew Ridgeway, review comments on MS for Ebb and Flow, 27 Feb 85.

The Gap Map.jpg  Click within the map area above to see a larger image.


Much has been made of the 80mile gap between the X Corps and the Eighth Army as it continues to be discussed among Chosin veterans and historians to this day. Why is "the gap" a problem? Was the much larger gap between the Inchon landing and the Pusan perimeter argued in the same light? No. Two major forces, Eighth Army and Tenth Corps, were pursuing a beaten enemy. The argument about the gap comes from hindsight that followed the introduction of a new threat in the theater of operations. The argument also involves the tendency to search for a scapegoat, someone to blame when the participants are partly to blame for the circumstances that were faced. Once again, one must study a topographic map of that 80mile gap, then answer the question not often asked: who in his right mind would use such terrain, and if so, for what purpose? And finally, in the end, why didn't the Chinese make use of it? There lies the answer.


(p.36) Security was the operative word at Chosin but being overcautious could tip the advantage to the enemy. Ground reconnaissance was the key to security and often neglected during the Chosin operation. Reports were made that company and platoon size patrols were used to seek out the enemy (see Patrols on p.45), when in fact it was not a recon patrol, but a combat patrol. Ground reconnaissance can only be effective during day or night by using small teams relying on the mobility of the infantryman.

When no confirmed information is available about enemy units in the area of operations, a combat patrol will see only that which the enemy wants them to see. Whereas a recon patrol of well-camouflaged men using stealth would have accomplished as much as the North Korean or Chinese scouts were doing when they observed the Americans. Once the enemy unit has been located it is then time to make use of combat patrols to contact and maintain that contact until the commander commits forces to attack and destroy. The action at Sudong during early November is an example of permitting a Chinese division to break contact without being pursued.

The 5th Marines east of Chosin sent patrols to the north and northeast, both too large for reconnaissance purposes. Major Chinese forces were just beyond the next mountain and within artillery range of Marine positions. These Chinese units were on the move late in the day when the last of the 5th Marine battalions withdrew, attacking the arriving Army units within hours.

Security at Chosin was a guessing game. Reports based on very few captured soldiers who identified their division were an indicator that elements of the division were in the area. Little is known about the specific locations and activities of Chinese divisions and their various regiments other than results of interrogations of a few enemy soldiers. After-the-fact enemy documents and publications remain incomplete. Details on attack plans and locations of battalions and companies are needed to tell the complete story.

Security also relates to the disposition of friendly forces, especially securing bases of operations such as Hagaruri which was a key road junction serving units to the north and west. East Hill, actually a mountain, was far too large to be secured by the units available at the time. This was not the case at Kotori more than four miles from the crest of the most critical terrain feature, the Funchilin Pass. Most neglected was the gatehouse bridge in the pass that apparently received some recognition but no action to secure the site. Was this too far to the rear to be of concern?


(p.40) One of the most difficult problems in understanding Chosin is the timing of plans and orders, as well as the viewpoints of the commanders involved. The 1MarDiv had the mission of attacking north and was urged by the corps commander to make haste, to get going and make contact with the enemy. This did not happen in the way envisioned by General Almond.

On the other hand, General Smith was concerned with the dispersion of this regiments, wanting to gather them as close as possible considering the terrain and build up his bases along the Chosin MSR. Today most historians or writers look at Smith's delay as beneficial. This may be so, although this theory comes from hindsight and does not represent the need at the time for movement to accomplish the mission. A war game would reveal that, within the time frame of the movement north, two of the Marine regiments could have been disposed from Hagaruri to the town of Changjin north of the reservoir dam, a gap through which the Chinese moved forces south to the reservoir.

While hastening the movement north, the division's Reconnaissance Company and later the 41 Commando, Royal Marines, both highly trained in reconnaissance, could have been used to find the enemy. Had this happened the disposition of the Marine regiments would not have favored the plan to attack west from Yudam-ni, but would have envisioned the attack west from the town of Changjin, a far greater threat to the Chinese. Bear in mind that the primary route south by the Chinese was down the Changjin river system, and not through the mountains west of Yudamni. The Changjin route followed a valley with a reasonably good road, a narrow gauge railroad and many manmade structures which offered shelter to advancing forces. The Chinese used it. The Americans didn't.

The Marine regiments were eventually disposed near the south end of the Changjin reservoir system when MacArthur came up with his plan and order to attack through the rugged mountain range west of Yudamni at a time when intelligence documents reported the beginning of severe winter weather, also a time when friendly forces were unprepared in both training and equipment to be combat effective in winter warfare. A similar command situation had already developed in China as they too committed unprepared armies through the same terrain to meet the Americans. The Chinese soldier's weapons, limited equipment and logistic system were far less combat effective than his enemy.


The schedule was not based on the need to prepare and move the major units in time to execute a plan, but rather that of wishful thinking. There is a tendency to say the staff didn't properly brief General MacArthur on the terrain and weather involved in such an operation; no, such thoughts were unnecessary because most knew that MacArthur had already made up his mind and nothing could have changed it.

Once again we face various interpretations of the change in X Corps plans when, in fact, the plans originated with MacArthur. The original plan was for the 1MarDiv to attack north on the east side of the reservoir toward the Yalu river, a logical move because to the west were the mountain ranges of "the gap," with limited westward access. At that time the boundary between X Corps and Eighth Army was on the west side of the reservoir with no need for Marine units to be sent to Yudamni.

The "ridiculous plan" (Ridgeway's hindsight) to send the 1MarDiv west did not consider of the terrain nor the time of year. This plan called for the movement of RCT 31 to the east side of the reservoir to relieve the 5th Marines so they could move west with the division. The 7th Infantry Division then taking over the sector east of the reservoir north to the Yalu. As it was, the plan called for urgent relief of the 5th Marines east of the reservoir, executed with such haste that all units of RCT 31 did not reach the Chosin area before the Chinese attacked and closed the MSR.

On the afternoon of 26 November the 2/5 Marines moved to Yudamni to prepare for the attack the following morning, the date ordered by X Corps, leaving Faith's 1/32 as the third infantry battalion for the 5th Marines should they have been needed. On the morning of 27 November the 2/5 Marines initiated the attack west of Yudamni as the remainder of the 5th Marines began withdrawing from positions east of the reservoir. The 2/5 Marines were stopped cold by the dug in and well-camouflaged Chinese, who by their presence announced they were there in strength. Of interest and not mentioned in the pamphlet was the original Chinese plan for attacks in the Chosin area. They had been scheduled to attack on 25 November, the same day the Chinese launched their attack against Eighth Army. Since forces were not ready, the attack was postponed and executed the night of 27 November. Those who believe the Chinese planned to attack two marine regiments at Yudamni must work this fact into "what if?" analysis, realizing had they attacked as originally planned, they would have attacked one marine regiment on each side of the reservoir. The Army RCT planned for Chosin would still have been east of the Fusen Reservoir. With the 5th Marines departing and RCT 31 arriving on 27 November, the Chinese commander had no time to make changes in his plans. He attacked that night.


(p.45-46) The pamphlet's introduction to RCT 31 begins with the arrival of Faith's 1/32 Infantry on 25 November. We note that "a patrol of Taplett's battalion had almost reached the northern end of the reservoir before brushing up against a small party of Chinese," with no explanation. What type of patrol was this, how far was the northern end of the reservoir from the 3/5 positions; in other words, what does "brush up" mean?

Of significance is the next paragraph that states "With the relief of RCT 5 by Faith's battalion, Marine operations east of the reservoir would end." Did they actually believe the presence of 1/32 relieved them of the responsibility for the zone of operation east of Chosin? Of course not, so why say it?

Although "Faith's command relationship to the 1st Marine Division is not clear," we believe there was no question at the time. Faith was in the tactical zone of RCT 5 with no way of getting operational or logistical support other than from the Marines. And, had the Chinese attacked (as planned), Faith's 1/32 would have fought with RCT 5.

We then come to the statement that Murray "did caution Faith not to move farther north without orders from the 7th Division," a statement often misused by historians. This adds confusion to the next statement that "once Murray departed, the only radio link between Faith and the 1st MarDiv would be ... [Stamford's radio]" when, in fact, the RCT commander and his units were about to arrive as stated by BGen Hodes during his visit to Faith on 26 November. The caution not to move further north has been confusing, some believing the intention was not to move into the forward Marine battalion position. Some believe that Faith occupied the northern 3/5 positions on his own. The use of hearsay in describing a situation continues to confuse and create a smoke screen, just as inferring that RCT 31 paid no attention to the enemy threat when they saw the threat just as Murray saw it, for he was their only source of information at that time. Missing is an exchange between Murray and MacLean on 26-27 November.


(p.48) "Faith, ignoring Murray's caution, received MacLean's permission to move his battalion forward the next morning to the position vacated by Taplett's [3/5] battalion." Without an explanation, the words "ignoring Murray's caution" is once again difficult to understand since Maclean [RCT 31] was ordering [OpO 25] 1/32 to occupy the northern battalion position of RCT 5. The road east of the reservoir was narrow and Faith's movement north would have to wait for the southbound movement of RCT 5 (minus one battalion) which would take most of the morning.

p.5253 Under a subheading "Chinese Order of Battle" the reader learns that "Sung would make the destruction of the 1st Marine Division ... his main effort." Then we read that the "27th Army ... was charged with attacking the two Marine Regiments at Yudamni." This is not correct. However, it does serve to reinforce Chosin hype that Yudamni was the primary objective of the CCF. This does not address the enemy plan to drive south through the Marines east of Chosin and cut off the Marines at Yudamni. The author has elected to follow the Chosin "story" of the past rather than explain the Chinese commander's plan and the execution of that plan which has been known to historians long before the publication of this pamphlet.

The I&R Platoon didn't "roar out of the compound" because "several hundred Chinese had been sighted." The platoon by OpO 25 was given the mission of establishing a screen east of the Inlet, a direction of major concern to MacLean since the RCT left the Fusen area.

Once again we read that the RCT units were "stretched out on the road for 10 miles in seven different positions." If a map with unit dispositons had been provided the reader would have learned that the combat battalions were disposed within supporting distance of each other. The two forward battalions, 1/32 and 3/31, occupied the two forward battalion positions occupied by RCT 5 just a few hours before. Ray Embree's battalion CP occupied the location of Murray's RCT 5 CP.

Faith did not received orders "to attack the next morning." He received OpO 25 in writing delivered personally by RCT 31 liaison officer Lt. Rolin Skilton, an order that stated the RCT would attack "on order," which means be prepared to attack when ordered to attack.

The attack "east of the reservoir" was made by the 80th CCF Division reinforced by a regiment of the 81st CCF Division. On the second day the CCF commander committed the remainder of 81st Division and held the 94th Division in reserve for his main effort down the east side of the reservoir.


p.58-59 "[MacLean] knew little about what had happened south of the inlet." This is not accurate, see Breakout by Hugh Robbins, CJ02.28.04. "Col. MacLean came back from the 1/32 CP about dawn and reported things were pretty much under control and all units of the battalion were holding. The 57FA reported that A Battery was being overrun... . Their CP was under considerable fire and partially surrounded.... the 3/31 reported its command post was under heavy fire and close range attack. In quick succession reports came in that Reilly and Embree had both become casualties, though not killed." As we can see, Col. MacLean knew far more about the Inlet battle than had been reported in the past.

"Stopping to see MacLean, Almond advised him that the previously planned attack would be resumed once the 2/31 joined the regiment. This battalion and Battery C of the 57FA were marooned far south on the clogged MSR."

Battery C/57FA had been detached from RCT 31 and remained in the Puckchong area, while A/31FA (155mm) had been attached to 57FA and never did make it to the Chosin area. See RCT 31 OpO 25.

Noted are the photographs of the Inlet area that are credited to (courtesy of) Norman Strickbine. These same photos can also be found in Appleman's two books credited to either Embree or Miller. The original source is a set of 18 photos by Ivan Long of Hq/31 who had been with MacLean's forward CP. It is believed that many copies of this set were made and found their way into the hands of wounded personnel in Japan. Sgt. Strickbine was a photographer with Hq 13th Engineer Battalion.


P.65 "His comrades buried him [MacLean] by the side of the road." A search for more detail on this subject takes one eventually to a footnote in Appleman's East of Chosin, p.147, which then takes one to p.365 where we find n.35 "I recall reading a newspaper interview with the soldier that told of Col. MacLean's death and burial. The soldier interviewed was probably one of the thousands turned over to American Authorities during the armistice prisoner exchange in 1953. I made a clipping of the article and put it with other Korean War notes. Now, more than 30 years later, I cannot find the clipping and provide the citation. But I am so certain of the facts recounted that I have no hesitation in including them in this narrative." This should excite historians to ask more questions. Who was the source, where did the burial take place, who gave them the tools to dig a grave in the frozen ground, and finally, was there ever a follow-up with a second source to verify the story? Those who have interviewed American prisoners who made those long night marches to the first prison camp at Kanggye find it difficult to accept the story.


Due to the length of the pamphlet being discussed, we will continue our review in the next issue of the Changjin Journal.

END CJ 10.10.04