The problem is all the more puzzling because the Chinese forces, apparently so formidable, had glaring weaknesses that UN forces could (and in the case of the marines, did) exploit.26 They had few if any tanks and very little field artillery: For the most part they attacked using infantry mortars, light machine guns, and hand grenades. Short of motor and even animal transport, their divisions lacked staying power. After a few days’ fighting they had to return to their rear areas to replenish ammunition and food, and their wounded had little prospect of recovery save through individual resilience. The bitter cold, particularly in eastern Korea, struck at them far harder than at the well-clothed and -equipped Americans: Quilted cotton jackets were no match for Siberian winds and snows. And even under less-than-arctic conditions—wet rather than dry cold—their clothing did not shield thousands of them from death by exposure.
The second Chinese offensive, beginning on November 26–27 hurled into battle some 300,000 PLA soldiers in eighteen lightly armed infantry divisions against the Eighth Army and another dozen against X Corps. Opposing them stood seventeen Eighth Army and X Corps divisions (seven of them American) plus forces worth at least another division of fine British, Commonwealth, and Turkish troops. Counting the remaining NKPA forces, who lent some support to the Chinese, the opposing forces were roughly equal in size—some 400,000 men each. To be sure, the Chinese had enormous reserves behind them in Manchuria, and the ROK units (at this stage of the war at any rate) demonstrated a terror of the Chinese at odds with their more creditable performance against the NKPA. On the other hand, UN forces had overwhelming superiority in matériel of every kind—from artillery to food, radios to medical facilities—and air superiority only sporadically challenged by Chinese (and possibly Soviet-piloted) MIG-15s flying out of Manchuria.
UN Command and its field forces expected to meet stiff resistance from Chinese forces (though not a full counterattack) as they advanced. When they did so they faced no gross disparity in numbers and possessed large advantages in firepower. Nonetheless, they—and in particular the Eighth Army—suffered a costly and humiliating defeat. Herein lies the failure we will examine below.
THE ROOTS OF DISASTER
Intelligence Unaware
Students and even practitioners of intelligence often suggest that it has two chief functions: warning (predicting what an opponent is about to do) and order of battle (information about the who, what, and where of an enemy’s forces).27 In respect to both kinds of intelligence the United States fell short in November 1950. The failure to predict large-scale Chinese intervention is perhaps the more understandable, since there appears to have been some considerable debate within China about the wisdom of a full-fledged offensive in Korea.28 American analysts devalued what appears in retrospect the best sources of intelligence—Chinese Nationalists reporting back to Taiwan. Reports from these sources warned that the Chinese Communists “intend to throw [the] book at UNO forces in Korea.”29 UN intelligence took the possibility of such an intervention seriously, and, as we have mentioned, MacArthur acted on it, most notably by ordering an intensified air campaign in North Korea.30 Indeed, one reason for the seemingly cavalier reception accorded Panikkar’s warning was that American officials believed that the Chinese already had begun to intervene in Korea—the only question was one of scale.31
One of the Allies’ senior intelligence officers in World War II remarked that intelligence can only tell a commander that “the stage is set and that the circumstances and conditions appear to be propitious for the other side to do this or that.”32 And in fact, American intelligence, and particularly G-2, FEC, did allow that a vigorous Chinese attack could occur, but did not discuss what form it might take or what it might mean.33 Both FEC and Eighth Army intelligence acknowledged that as UN forces neared the Yalu, warning time would decrease, and that rapid reinforcement from China remained possible. If we accept as appropriate the more modest objective of letting the commander know that “the stage is set,” we must conclude that American intelligence did not fail in November 1950.
On order of battle the picture is more mixed, for FEC and Eighth Army intelligence certainly underestimated the number of Chinese troops actually in North Korea. On November 16, for example it estimated that the Chinese had twelve divisions in North Korea, although it vitiated that estimate by declaring that these units numbered only four to six thousand men each, rather than the 10,000 men per division that was their real strength. At the same time, however, intelligence took note of the enormous reserve strength in Manchuria—more than 460,000 regulars, plus 370,000 district troops.34 Eighth Army intelligence was even more pessimistic, placing opposing Chinese forces at somewhere between 100,000 and 150,000 men. This underestimated Chinese forces by something like a factor of three, an error partly excused by the acknowledged difficulty of identifying Chinese units as they crossed the border.
The failures—or more accurately, semifailures—in warning and order of battle intelligence have received a good deal of attention from students of the Chinese surprise attack in 1950. Another more serious and generally ignored type of intelligence failure occurred, however: failure to gauge the enemy’s way of war, his methods, strengths, and weaknesses.35 It is in the picture of the enemy held by U.S. forces in the Far East that we find one of the chief sources of the failure of the winter of 1950.
The problem did not stem from a lack of basic data. Intelligence in both Tokyo and the field knew fairly well the table of organization and equipment of the average Chinese division, which it explicitly compared with that of the NKPA divisions with whom UN forces had fought for nearly half a year.36 They noted that NKPA units had a full suite of modern Soviet equipment, where the Chinese had a meager hodgepodge of captured Japanese and American weapons. The average NKPA division, for example, had a total of forty artillery pieces—including medium-caliber howitzers—per division. The Chinese had barely nine pieces, and those were light 76-mm. howitzers. NKPA divisions had four times as many heavy machine guns and more antitank weapons than their Chinese counterparts. Chinese divisions were superior in personnel (by all of thirty men), the number of light machine guns and light (60-mm.) mortars, but drastically inferior in transport and support—the average NKPA division had some two hundred vehicles as opposed to none for the Chinese. Intelligence, therefore, came to view Chinese forces as a kind of inferior version of the North Koreans:
The quality of the Chinese Communist fighting man is probably similar to that of the well-trained Korean soldier in mid-campaign. However, it is to be recognized that most of the CCF troops have had no significant experience in combat operations against a major combat power. In addition, their training, like that of the original North Korean forces, has been greatly handicapped by the lack of uniform equipment and assured stocks of munitions.37
Intelligence failed to grasp the crucial fact that UN forces faced not two varieties of the same opponent, but two completely different kinds of enemy altogether. The PLA was not simply a watered-down version of the NKPA but a different army, with unique strengths, weaknesses, tactics, and operational preferences. The vehicle-heavy NKPA, for example, moved and deployed like the conventional forces they were. The extreme “lightness” of Chinese divisions, on the other hand, meant that they could hide in and infiltrate through the forests and villages of North Korea, subsisting on provisions requisitioned from local peasants or carried by their own troops. No large supply trains clogged the roads in North Korea, although American aircraft did notice large movements in Manchuria.
The Soviet-equipped NKPA fought for the most part as Soviet divisions would have, using combined arms and depending on conventional means of resupply.38 Their tactics centered on the frontal assault backed by tanks and artillery, supported by powerful flank attacks, and accompanied by infiltration behind enemy lines. Early on, at any rate, the NKPA fought by day and in the open. They maintained an unremitting offensive, even when their situation might have suggested that defense would be a more prudent pos
ture. Even after the Eighth Army had consolidated its position in the Pusan perimeter in late August, North Korean forces continued to batter their heads against forces superior not only in firepower, but sheer numbers as well.
The Chinese did not fight this way. They attacked mainly by night, using large quantities of hand grenades, light machine gun and mortar fire (which, it will be remembered, were the weapons they had in greater supply than their North Korean counterparts) from very close ranges.39 They usually approached from the rear, after drawing enemy fire by sniping and bugle or pipe music. Operationally, the Chinese had a more subtle approach than the North Koreans: feinting, probing, or withdrawing (as they did after the First Phase Offensive) in order to test enemy reactions or to confuse and intimidate them. They followed thereby the maxims of Sun Tzu, from whom Mao Zedong had derived much of the PLA’s doctrine:
All warfare is based on deception.
Therefore, when capable, feign incapacity; when active, inactivity. When near, make it appear that you are far away; when far away, that you are near.
Offer the enemy a bait to lure him; feign disorder and strike him.
When he concentrates, prepare against him; where he is strong avoid him. . . .
Pretend inferiority and encourage his arrogance.
Keep him under a strain and wear him down.
When he is united, divide him.
Attack where he is unprepared; sally out when he does not expect you.40
This way of war helps us make sense of the First Phase Offensive, so often misunderstood as a diplomatic signal rather than as what it was—an important piece of campaign strategy designed to gain information, test tactics, and bewilder the opponent. As the Chinese commander later wrote, “We employed the tactics of purposely showing ourselves to be weak, increasing the arrogance of the enemy, letting him run amuck, and luring him deep into our areas.”41
Only toward the end of December did Far East Command publish extensive discussions of Chinese tactics that suggested that the Chinese had a distinctive tactical and operational approach to war.42 Before then such information appeared only sporadically and in odd corners of the Daily Intelligence Summaries. Far East Command did describe Chinese and North Korean guerrilla tactics, and the Eighth Army’s IX Corps wrote one short note on Chinese “Hachi Shiki” (inverted-V) tactics.43 As we have mentioned, intelligence had access to order-of-battle information from which might have been inferred (though no one did) that they might use novel tactics. Until fairly late in December 1950, however, one finds in intelligence reports no explicit consideration of what qualitative change in fighting might occur if the Chinese entered the fray. Rather, intelligence assumed that the Chinese would demonstrate (though in lesser degree) the strengths and weaknesses of the NKPA.
This latter proved to be a crucial point, because intelligence believed that it had discovered the North. Koreans’ critical vulnerability. Only weeks before the large-scale Chinese entry into the war, intelligence confirmed what American commanders had long believed: American airpower had paralyzed the NKPA. In-depth interviews of two thousand NKPA prisoners of war revealed that over half the NKPA’s equipment losses and a third of its personnel losses stemmed from aircraft—twice as much damage to equipment and the same damage to personnel as inflicted by artillery.44 FEC Intelligence concluded that:
the net effect of such tactical air support was the greatest single factor contributing to the successful conduct of UN ground operations against the North Korean Communist invaders. . . . Based entirely upon PW [prisoner of war] reactions this study indicates that there can be no doubt that the impact of UN air efforts in the tactical support of UN ground forces probably has been the greatest single factor contributing to the overall success of the UN Ground Force scheme of maneuver.45
Air power, not the Inchon landing, had blocked the success and weakened the grip of the North Koreans investing the Pusan perimeter. General Walker, commander of the Eighth Army, put it bluntly: “I will gladly lay my cards on the table and state that if it had not been for the air support that we received from the Fifth Air Force we would not have been able to stay in Korea.”46 In proportion to its size the Eighth Army had behind it more fighter-bomber support than did Omar Bradley’s Army Group storming across Europe in 1944.47 It thought very highly of that support, and analysis proved it right to do so.
FEC Intelligence judged that UN air power would force the Chinese “to campaign under the same psychological and physical handicaps as those borne by the North Koreans.”48 American commanders judged—quite rightly—that those handicaps had been extreme and possibly decisive. Small wonder, then, that at the Wake Island conference of October 15, 1950, General MacArthur had assured President Truman that because of American air power “if the Chinese tried to get down to Pyongyang there would be the greatest slaughter.”49 Small wonder too that he fought so hard—at times almost hysterically—to get permission from the Joint Chiefs of Staff for an immediate and unrestricted air campaign right up the Yalu River and against the bridges across it.50 MacArthur did not simply discount Chinese intervention, but he thought he had the antidote to it, in the form of broken bridges, strafed roads and tracks, and if necessary, incinerated villages and towns.
The Chinese did indeed respect American air power, but their organization and tactics allowed them to minimize its impact. By operating off the roads, at night, and close-in against UN forces, they negated much of the advantage of enemy air superiority. Their superlative camouflage and march discipline (soldiers stood stock still when an enemy airplane came into view, and officers had authority to shoot those who moved) hid them from aerial observation. In the short term, at any rate, and so long as they did not have long supply lines to maintain, the Chinese could conduct an offensive against American and ROK forces that the more conventional NKPA could not. They made of their material weakness, in other words, an operational strength.
The deficiencies of American intelligence stemmed in large part from analytical assumptions, but we cannot divorce these from the sources that helped shape those assumptions.51 We know that prisoner-of-war interrogation yielded a great deal of information, and indeed often provided the most useful intelligence concerning the Chinese. A shortage of Chinese linguists, however, coupled with insufficient appreciation on the part of lower echelons of the importance of prompt evacuation of prisoners of war to higher levels slowed the transmissions of this kind of intelligence. Unfortunately, American intelligence treated prisoner-of-war information with reserve, failing to believe that Chinese enlisted men would possibly have as much knowledge of their organizations and overall strategy as they claimed. In this they failed again to understand the peculiar nature of their opponents, who stressed the importance of explaining their mission in great detail to their men.52
If it undervalued POW intelligence, G-2 overestimated the thoroughness with which aerial reconnaissance could keep track of the enemy. Desperately short not only of photo-interpreters but of photo-reconnaissance aircraft, Far East Command concentrated its resources on inspecting the bomb damage from its deep interdiction raids, particularly strikes against the Yalu bridges.53 Virtually no careful photo-reconnaissance of other areas in North Korea occurred. S. L. A. Marshall discovered in November 1950 that as many as ten days would frequently pass between a request from the Eighth Army for aerial photographs and G-2’s receipt of them—and often the Eighth Army received no more than three or four such reports a day.54 Given Chinese abilities to move under cover and at night, and to lie up during the day, photographic intelligence, which had proved invaluable in World War II, fell short of its promise.
We know little about two other major sources of information, espionage and signals intelligence. From various sources it appears that such spy networks as UN forces had at their disposal crumpled once the North Koreans invaded the south, and despite strenuous efforts were never fully replaced. Chinese Nationalist reporting and occasional contacts through Hong Kong provided some limited but use
ful information on operational and strategic matters. Signal intelligence is a closed book (the army official historical monograph on the subject does not even mention it) and will remain so until National Security Agency files on the subject are opened in the year 2000. Though we may make several inferences from information in the FEC intelligence reports, these are no more than guesses. We know that the Chinese made relatively little use of radio, preferring (for good reason) to rely on intrinsically more secure land lines. At the lower levels (from regiment on down) Chinese units had no radios at all—a weakness in terms of tactical flexibility but a strength in terms of security. On the other hand, it is difficult to see how American intelligence could have tracked the movement of dozens of divisions into Manchuria and the border area during September, October, and November of 1950 if not by interception of radio signals.
Military Misfortunes Page 23