The Easter Offensive of 1972: A Failure to Use Intelligence
W. R. Baker
This article was originally published in the Military Intelligence Professional Bulletin, 1998.
It was not Intelligence (evaluated information of the enemy) that failed. The failure was [that of] the commanders and certain G2s, who did not act on the intelligence they had.
--Colonel Robert S. Allen, on The Battle of the Bulge1
Like the Battle of the Bulge, the 1972 Easter Offensive in Vietnam has often been referred to as an "intelligence failure," mainly because it caught the United States and South Vietnam completely by surprise. A look beneath the surface, however, reveals that U.S. and South Vietnamese combat commanders were aware of significant changes in the posture of the North Vietnamese Army (NVA) and had access to many indicators of an impending NVA offensive. Colonel Allen's assessment that commanders at The Bulge failed to embrace the intelligence available to them holds true when evaluating why American and South Vietnamese forces failed during the Easter Offensive, as well.
Several factors contributed to the success enjoyed by the NVA during the Easter Offensive. First, the U.S. and South Vietnamese commanders failed to use all of the intelligence available to them. Their overconfidence, coupled with command and control (C2), and communications problems, violated the cardinal rule of "never underestimate your enemy."
Also, Allied commanders relied on only two of the three forms of intelligence---imagery intelligence (IMINT) and signals intelligence (SIGINT). In ignoring human intelligence (HUMINT) data, they overlooked the most accurate forecast of enemy intentions leading up to the Easter Offensive.
While most areas of South Vietnam encountered NVA ground force activity during the Easter Offensive of 1972, it was the northern-most portion of the country, the Quang Tri Province, which bore the initial and most severe brunt of the NVA's conventional ground campaign. The offensive began with an extensive artillery bombardment just before noon on Good Friday, 30 March 1972. Only the 1st ARVN (Army of the Republic of Vietnam) (Hue) and 3d ARVN (Quang Tri) Divisions, the 147th and 258th Vietnamese Marine Corps (VNMC) Brigades, and the understrength 20th ARVN Tank Battalion stood in the way of the NVA onslaught into the northern provinces.
The North's plan of attack was relatively simple. The 324B NVA Division, supplemented by two infantry regiments (the 5th and 6th) that had made their way south from North Vietnam, would engage the 1st ARVN Division, widely considered to be the ARVN's best. Operating from its usual B-4 Front staging area in the A Shau Valley west of the Hue, these NVA units would preempt any possibility of relief to the north by applying pressure to the firebases west of Hue and, thereby, to the city of Hue itself. However, the 1st ARVN had conducted offensive operations during March that precluded any B-4 Front operations from achieving surprise before Good Friday. At the same time, the 3d ARVNC comprised of three infantry regiments (the 2d, 56th and 57th), the attached 20th Tank Battalion, and the two VNMC brigades was prepared to face the onslaught of the NVA's B-5 Front, bounding down from the north.
The Easter Offensive - 1972
Nevertheless, the enemy struck with speed and accuracy at the weakest link along the line. The offensive began with the 308th NVA Division and the attached 204th Tank Regiment springing from the western Demilitarized Zone (DMZ). Concurrently, what amounted to (and perhaps was) a division-sized task force composed of three independent infantry regiments (the 27th, 31st, and 126th Sapper) and the 201st Tank Regiment, jumped off from the eastern portion of the DMZ.
Meanwhile, the 304th NVA Division (which had infiltrated from Laos) and the attached 203d Tank Regiment arrived in the Khe Sanh area and attacked eastward, eventually linking up with the 246th and 270th Infantry Regiments on its left flank. This assault from the west allowed the NVA to conduct flanking attacks towards Quang Tri City and, most importantly, enabled some units to become a blocking force against any attempt to relieve the 3d ARVN from the south. Additionally, there was an intelligence report that indicated that the 324B NVA Division detached two of its three regiments (the 29th and the 812th) in an attempt to strike northeastward towards Quang Tri during this period.
Essentially, the NVA forces had achieved a lightning-fast victory that sealed off the 3d ARVN Division from its reinforcements and relief. Striking at the weak link along the Allied line, the NVA completely surprised the Allied forces. Or did they? History tells us that the Allies had prior knowledge of NVA activity in preparation for the attack, but did not use that information to the maximum extent possible.
Reluctance to Use Intelligence
Indications of some sort of offensive were forecast first during Tet, then in March. The NVA's 324B Division was known to be headed for their usual AO in the A Shau Valley. A slight buildup across the DMZ was detected, but it was not enough to cause any great alarm.
Despite the buildup, most Allied commanders were confident that the NVA could not sustain an offensive so soon after Tet. The South Vietnamese I Corps commander, when presented with the idea of a North Vietnamese attack across the DMZ, bluntly stated, "They cannot." 2 His U.S. counterpart at the First Regional Assistance Command (FRAC) agreed, "His appraisal appeared reasonable and well-founded." 3
Also, despite the known NVA buildup across the DMZ, the 3d ARVN commander, Brigadier General Vu Van Giai, decided to swap the positions of the 2d and 56th Regiments, creating a situation where his forces would be in transition. Unfortunately, they were in the process of switching positions when the Offensive began. In fact, NVA shelling began less than thirty minutes after both regiments' communications shut down. Coincidence? Or, rather, evidence of good enemy use of intelligence?
A South Vietnamese Joint General Staff (JGS) account later claimed that there was a country-wide alert on March 29. If so, the word of the alert never arrived at Military Region-1 (MR-1) or the Military Assistance Command Vietnam (MACV), or it was largely ignored by those commands. Otherwise, U.S. Air Force and Republic of Vietnam Air Force aircraft would have attempted to preempt offensive operations in the days and weeks prior to the offensive. Also, U.S. and South Vietnamese ground units would have been on alert and the two regiments of 3d ARVN would not have been changing AOs.
In Vietnam, C2was problematic. The leadership provided by the South Vietnamese commanders was rarely exemplary. The I Corps commander, Lieutenant General Hoang Xuan Lam, who was more a political general than a military commander, often exhibited inept and indecisive leadership. To compound the problem in I Corps, the Corps' headquarters had "never actually functioned as a field headquarters in combat," as the FRAC commander later admitted.4
However, the most infamous example of the dearth of ARVN leadership was the surrender of the 56th ARVN Regiment by Lieutenant Colonel Phan Van Dinh, its commander, at Camp Carroll after a short fight on Easter Sunday. Many other examples of cowardice occured during the Offensive, but none was so flagrant or damning.
To add to the Allies' C2 woes, there was a distinct lack of cohesion in coordinating the efforts among the army and marine corps units in MR-1. The 147th VNMC Brigade at Mai Loc and the 258th VNMC Brigade at Fire Support Base Nancy were technically attached to the 3d ARVN Division. The 3d did not, however, have operational control (OPCON) over these units the I Corps commander did. The Vietnamese Marines were also considered part of South Vietnam's strategic reserve a reserve whose squandering would ruin their commander's career. As a result of this C2 confusion, the VNMC units located with the 3d ARVN were under-utilized during the Offensive.
Finally, MACV was receiving most of its intelligence third-hand from Commander, U.S. Pacific Fleet (PACFLT) in Hawaii. In addition to providing offshore naval gunfire support, U.S. Navy destroyers, with the naval intelligence liaison element in Da Nang, were presumably providing the PACFLT Commander in Hawaii with more timely tactical situational reporting than the MACV commander, General Creighton Abrams, was receiving from I Corps. The destroyers were not OPCON to MACV or I Corps and, therefore, reported directly to the PACFLT Commander. He, in turn, relayed the reports from Hawaii back to Saigon. "As it turned out, it was up to three days after the enemy attack was launched in MR-1 before it aroused any major concern in Saigon," the JGS Assistant J2 stated years later.5 Across the board, C2 was not very smooth for the Allied forces.
During the Easter Offensive, two of the three pillars of intelligence, IMINT and SIGINT, supplied some indicators of what was about to occur. However, it was the ignoring of the third pillar, HUMINT, that ended up shaping the incorrect opinion among Allied commanders that the NVA offensive was unlikely and, even after it began, that it was doomed to failure.
IMINT provided the first real clues of the NVA buildup north of the DMZ. U.S. photographic intelligence missions detected a large concentration of tanks near Bat Lake, where the borders of North and South Vietnam met the border of Laos.
Although no information is currently available concerning the Easter Offensive radio SIGINT intercepts prior to March 30 (nor afterwards), it is reasonable to assume that the NVA communications security (COMSEC) measures and traffic minimization were in effect. However, sensor activity along the Ho Chi Minh Trail was unusually active with "movers" during February and March.
Although there has been little written on SIGINT reporting during the Easter Offensive, it can be presumed that there were, in fact, few SIGINT intercepts during that period, especially when one considers the actions not taken by the Allied forces. If SIGINT had indicated the NVA buildup, the MACV commander would probably have returned early from his rest and recuperation in Thailand or perhaps would not have left South Vietnam at all. In addition, a country-wide alert generated by SIGINT intercepts of NVA offensive preparations would have alerted U.S. forces, thus precluding the senior U.S. Army advisor to 3d ARVN from leaving for an Easter visit to Saigon.
While IMINT and SIGINT provided some clues as to the NVA's intentions during March, HUMINT provided the most accurate forecast of enemy intentions before the Offensive. In December 1971, a "usually reliable source" 6 had provided a significant document, a North Vietnamese Politburo policy, which indicated that the Viet Cong/NVA would switch over to "main force" rather than "protracted" warfare operations.
Then, on March 27, a friendly ambush southwest of Firebase Pedro yielded unexpected results. One of the dead NVA soldiers carried a map showing all of the trails, streams, firebases, and units in Quang Tri Province.
A U.S. intelligence unit even predicted the NVA's preliminary objectives in advance of the Offensive. Based on agent reports, the 1st Battalion (Provisional), 525th MI Group, constructed a detailed description of the major NVA units, their commanding officers, and the date the Offensive was to begin. Using the intelligence summary (INTSUM) for the first time in the 525th MI Group's history in Vietnam, their predictive analysis was sent electronically to the 525th MI Group, MACV, U.S. Army, Vietnam (USARV) and 7th Air Force Headquarters. It was carried by courier to the FRAC, naval intelligence liaison, 196th Light Infantry Brigade, 1st MIBARS (Military Intelligence Battalion, Aerial Reconnaissance Squadron), the Central Intelligence Agency, and U.S. Army Special Forces in the Da Nang area before the Easter Offensive began. Their reporting continued for many weeks afterwards.
Despite the reliable HUMINT analysis sent to them, the American command took a lot of time to realize the true intent of the NVA offensive in the north. It was not until 27 April 1972, 28 days after the offensive had started, that the FRAC wrote to the MACV commander:
Reports are fragmentary at this time but intelligence indicates that the objectives are the capture of Fire Support Base Nancy and to establish a blocking force on the Quang Tri/Thua Thien border. Other NVA forces will then assume offensive operations to capture Quang Tri City.7
In fact, the HUMINT INTSUMs from the 525th MI Group had predicted weeks ahead of the attack that Quang Tri City, and not Hue, was the objective of the NVA action.
Obviously, the main lesson here is that neither SIGINT nor IMINT are the sole foundation of intelligence. They should be treated as two of many elements that make up all-source intelligence analysis. During the 1972 Easter Offensive, the lack of SIGINT collected was, in and of itself, a possible indicator. SIGINT cannot (and should not) be the crux or final determining factor in assessing enemy intentions or capabilities.
The Easter Offensive caught the Allies by complete surprise needlessly so. While the indicators of attack were numerous, U.S. and South Vietnamese commanders ignored them in favor of a more reassuring position: that the NVA could not and would not attack before the end of March. Their failure to use HUMINT to the fullest extent possible also contributed to the Allied forces being caught off guard. The "intelligence failure" during the Easter Offensive was less a failure to collect intelligence than it was a failure to exploit obvious indicators.
1. Colonel Robert S. Allen, Lucky Forward (New York: Manor Books, 1965), 157.
2. Dale Andrade, Trial by Fire (New York: Hippocrene Books, 1995), p. 62.
3. Andrade, 62.
4. Major General Frederick J. Kroesen, Quang Tri: The Lost Province (U.S. Army War College document, 16 Jan 74).
5. Colonel Hoang Ngoc Lung, Intelligence (Washington, DC: U.S. Army Center of Military History, date unknown), 155.
6. Andrade, 25.