Part One: Uncertain Signals
“In 1968, the third year of U.S. combat operations in Southeast Asia, the basic issues of the war remained largely unchanged.” That laconic statement, the opening line in the MACV command history for 1968, stood in stark contrast to the optimistic tone set the year before. In the closing months of 1967 the White House had orchestrated an intense public relations campaign, including another stateside visit by General Westmoreland, aimed at convincing skeptics that there was in fact light at the end of the Vietnamese tunnel. Among those harboring doubts was Secretary of Defense Robert S. McNamara who, all the way back to JFK’s Camelot days, had played such a crucial role in the direction of the war that he took as a compliment references to “McNamara’s War”. But McNamara had become disillusioned by failure of the ROLLING THUNDER bombing campaign and other measures, many of his own making, to bring the war to a successful conclusion. When McNamara's disillusionment began to border on criticism, LBJ essentially fired him, offering in consolation a position as president of the World Bank. McNamara would not officially leave office until 29 February, but Johnson hastened to nominate a successor; long-time political confidant and Washington insider Clark Clifford, whose appointment was confirmed in late January.
McNamara was far from alone in his bleak assessment. Despite the costly, ongoing land battles and megatons of bombs unloaded on North Vietnam, Americans all across the socio-economic spectrum now concluded that the war had ground to a bloody stalemate, with no end in sight. Ironically, observers in Hanoi had reached much the same conclusion. Although details would come to light only gradually, the North Vietnamese had themselves been engaged in a lengthy debate over how best to wage “the resistance war against the Americans to save the nation.” In evaluating results after the insertion of U.S. ground troops in 1965, two factions had emerged. Conservatives argued that U.S. technology and firepower could not be defeated outright. It would be necessary to revert to guerilla tactics—the second phase of Chairman Mao's “people's war.” Through a protracted, relatively low-intensity struggle the Americans would gradually loose heart and agree to a negotiated settlement favorable to the communists. The other side agreed that the war was taking a heavy toll on both the homeland and the troops in the South. But only by continuing to fight large scale, conventional battles could the North deadlock the war and enter a “talking while fighting” stage that would hasten a U.S. withdrawal and end the war. The debate would not be settled until just before the New Year.
Plans: Hanoi and Saigon
As 1968 rolled in, the enemy's annual winter/spring campaign was already in full swing. One objective of the bloody "border battles" of the previous fall had been to test U.S. reaction. There was fear in Hanoi that such attacks might precipitate a counter-invasion of North Vietnam itself. (The concern was not entirely unjustified; Westmoreland's staff had in fact begun planning for just such a contingency, but the notion was quickly squashed by higher authorities.) When no invasion occurred, the communists began to put their own plans in action. The result was a compromise of sorts between Hanoi's hawks and doves. In June, 1967, the Politburo ordered planning to start for a “general offensive-general uprising” in which efforts on all fronts would be applied “to the maximum extent possible in order to secure a decisive victory within a relatively short period of time.”
The plan called for the North Vietnamese and the Viet Cong to “concentrate our military and political forces to launch a simultaneous surprise attack against the enemy’s weakest point: his urban areas. The principal theaters of operations would be Saigon, Hue, and Da Nang.” Main force units would “crush every large puppet army unit" while at the same time “luring away and tying down U.S. mobile forces in the mountain jungle battlefields.” Key to the plan was the “general uprising” half—the assumption that, once communist forces showed themselves in the cities, the oppressed South Vietnamese would welcome them with open arms then actively join in the overthrow of the puppet government. These critical urban operations would largely fall to local VC guerillas and “sleeper” cells within the cities. To increase the chances of success the offensive was set to launch during Tet, when ARVN units were expected to temporarily stand down in celebration of that traditional Vietnamese holiday. For security reasons only a handful of top North Vietnamese officials would know the exact date and time of the attacks.
In Saigon there was reason for cautious optimism if one chose to look for it. By MACV standards, ground operations in 1967 could be judged mostly successful in that they had driven VC and NVA main force units—the principle targets of Westmoreland’s big unit war—into the mountains and jungles away from populated areas, and more of the South Vietnamese countryside was now reasonably secure. But the cost had been high. In the year just ended Free World Military Forces lost some 23,000 troops killed in action; more than 93,000 others had been wounded. The joint U.S.-RVN Combined Campaign Plan for 1968 placed renewed emphasis on pacification, with offensive operations to be mounted primarily as a shield to that activity. Officially, “search and destroy” was to be replaced by less gruesome terminology, but from a grunt’s-eye perspective, not much had changed.
On 2 January, a Marine patrol outside Khe Sanh combat base killed five members of a six-man enemy team evidently on a reconnaissance mission. The dead proved to be high-ranking NVA officers. Weeks earlier SIGINT had intercepted communications between regiments of the 304th NVA Division, which counted Dien Bien Phu among its battle honors, as these elite units moved from North Vietnam along Route 9 in Laos. By mid-December the division headquarters had been fixed near Tchepone, a major waypoint along the Ho Chi Minh Trail. Other enemy units, including elements of a second NVA division, the 320th, were also located, almost certainly by EC-47 crews. Shortly thereafter the 325C NVA Division was fixed in the same vicinity. SIGINT and other intelligence then confirmed the activation of a new enemy "front" command controlling the three NVA divisions around Khe Sanh. With Route 9 cut, the Marines were effectively surrounded. Further sustenance could come only by air, and flying weather did not look good.
The clues were obvious and ominous, causing particular distress in Washington, where the inevitable comparisons to Dien Bien Phu had already begun. COMUSMACV and CINCPAC had long since recognized the significance of Khe Sanh and the repercussions, both military and political, that would arise should the enemy overrun the outpost. There was talk of abandoning the base while there was still time, but Westmoreland nixed the idea on both psychological and logistical grounds. Khe Sanh would be held at all costs.
Meanwhile, SIGINT units further south began to reveal "Allied units being targeted, and position reports that pinpointed [enemy] units as they moved into new positions.” Unusual communication activity was noted in the tri-border region, the long-time bailiwick of the NVA B-3 [Central Highlands] Front and its attached units. Another major headquarters was soon located in the Pleiku-Kontum area. Particularly worrisome was evidence, again most likely uncovered by Electric Goon crews, that yet another NVA command was communicating with three subordinate regiments a mere 10 kilometers from Hue. Down in III Corps enemy contact had fallen off recently, but intercept operators "were hearing increased chatter between COSVN and its main force units.” According to an NSA history, “the ‘accumulation of SIGINT data’ indicated that a 'coordinated offensive' would be conducted throughout several areas in South Vietnam." Various sources also hinted that "N-day" for the upcoming offensive would occur at or around the time of the Tet holidays. All intelligence indicators clearly pointed towards something big—"the question was what …?”
COMUSMACV, confident that Khe Sanh remained the primary target of whatever the enemy offensive turned out to be, set about arranging a warm reception when the attack finally occurred. Patterned after the SLAM tactics recently employed at Con Thien, Westmoreland initiated an operation he named NIAGARA, thereby conjuring "an image of cascading bombs and shells.” Beginning on 6 January the first phase, a “comprehensive intelligence collection effort” (NIAGARA I) was kicked off, its primary aim being development of ARC LIGHT targets. A special "all source" Niagara Intelligence Control Center (ICC), operating 24/7, was set up at Tan Son Nhut. The ICC would be supported by “the most intensive tactical reconnaissance program initiated to date," to include “all available USN, USMC, and USAF… reconnaissance and electronic warfare aircraft resources.” The two SENTINEL SARA (ex-DRILL PRESS) and two Z-equipped EC-47s were specifically fragged to provide increased SIGINT coverage in the DMZ area. One of each type aircraft would be based at Phu Bai, while the other pair flew from Pleiku.
The Marines, meanwhile, beefed up their outposts on the hills north and northwest of Khe Sanh base proper, where increasingly heavy contact with the enemy began on 17 January. Three days later, just as the Marines appeared poised to finish off their antagonists, orders were received to disengage, fall back, and dig in. Earlier in the day, an NVA officer had appeared, waving a white flag, just off the runway. When interrogated he unhesitatingly laid out the enemy plans in detail. The accuracy of his information proved to be "exceeded only by its timeliness"—the anticipated attack would began that very night.
Duly alerted, the Marines suffered only light casualties in beating off a ferocious ground attack on one of the hill outposts. The main base, meanwhile, was blasted by everything from 82mm mortars to 122mm rockets, hundreds of rounds in all. The ammo dump was hit early on, setting off a tremendous conflagration that would sizzle and boom for the better part of two days. In Saigon, the go-ahead was quickly given for phase II of Operation NIAGARA. On day one, 22 January, 245 sorties were flown in support of the Marines. On day two, the total almost tripled, with pilots reporting many secondary explosions. By the end of day five the cumulative sortie count topped 2,400, including 164 by ARC LIGHT B-52s. The enemy would suffer terribly from this Niagara-like torrent of high explosives, but he continued to inflict his own brand of misery, much of it delivered by artillery hidden miles away in the Laotian hills. The deadly routine would continue, virtually without letup, for another two months.
Distractions . . .
Signs of some kind of unprecedented communist offensive in Vietnam was just one of many warning bells clanging in Washington during the last week of January, 1968. Khe Sanh remined at the top of the watch list in both the White House and the Pentagon, but events elsewhere suddenly intruded. As the attacks on the Marine base ramped up, word came that a B-52 armed with four hydrogen bombs had crashed near Thule, Greenland, when the crew bailed out after the aircraft caught fire. The nukes were unaccounted for. The accident was far more serious than initially thought (or admitted), but a much weightier situation developed two days later when North Korea attacked the SIGINT vessel USS Pueblo. By the time help could be summoned, it was too late. The Koreans boarded the ship, towed her to port, and imprisoned the crew. With rescue seemingly not an option, harried military brass and administration officials discussed various retaliatory measures. But the war in Vietnam had stripped U.S. strategic reserves virtually to the bone although once the go-ahead was given, by gathering units from as far away as North Carolina the USAF was able to assemble a very respectable combat air fleet in Korea. But hanging over it all was the ever present specter of Chinese or Soviet reaction and the unpleasant possibility of another war that the U.S. was ill prepared, emotionally or militarily, to fight. In the end, nothing was done. The Navy crew would languish in North Korea jails for another eleven months, at times mistreated to the point of torture. Presumably, the super-sensitive SIGINT equipment and materials left aboard quickly found their way to Moscow and Peking. The extent of the damage done to U.S. intelligence will likely never be completely revealed. A half-century later, the Pueblo remains tied up alongside the Wonsan docks, a galling symbol of the most humiliating international episode in recent U.S. history.
. . . and Diversions
In Washington, signals of an imminent communist offensive throughout Vietnam were largely lost in the noise coming from Korea and Khe Sanh. But at NSA a picture, still not completely clear, had begun to emerge. On 25 January the agency issued the first of several reports on "Coordinated Vietnamese Communist Offensive Evidenced in South Vietnam." Nobody in D.C. paid much attention, but at II Field Force headquarters in Long Binh, LTG Frederick C. Weyand stayed on top of all the intelligence at his disposal. Traffic analysis and radio direction finding bolstered his “professional belief that MACV was greatly underestimating the number and military significance of local VC forces.” This intelligence convinced Weyand that these forces were moving towards Saigon. Weyand laid out his case to MACV deputy commander GEN Creighton Abrams, who agreed with the assessment. With Abrams’ backing, Weyand ultimately convinced Westmoreland to counter previous orders sending IIFFV troops on offensive sweeps on the Cambodian border and instead to move these units back closer to Saigon. In the coming days the wisdom of this last-minute decision would be abundantly demonstrated.
Article by Joe Martin
Revised 06 September 2018
Note: An annex detailing sources for this article is under construction.