If SD [Viet Cong Self-Defense forces] and SSD [VC Secret Self-Defense forces] are included in the overall enemy strength, the figure will total 420,000 to 431,000. . . . This is in sharp contrast to the current overall strength figure of about 299,000 given to the press here. . . . We have been projecting an image of success over the recent months. . . . Now, when we release the figure of 420,000-431,000, the newsmen will . . . [draw] an erroneous and gloomy conclusion as to the meaning of the increase. . . . In our view the strength figures for the SD and SSD should be omitted entirely from the enemy strength figures in the forthcoming NIE.
- Gen. Creighton Abrams, Deputy Commander, MACV, 20 August 1967(1)
As we have seen, in the first decade of direct US involvement in Vietnam, dating from the French surrender at Dien Bien Phu in 1954, policymakers seeking good news had encouraged optimistic reporting and ignored or complained about intelligence analysis that failed to support their expectations. The bliss of ignorance had several times cost the US war effort dearly, but worse was in store at the end of January 1968, when a misreading of the enemy's intentions and a calculated understating of his strength left the nation and its political leaders wide open to the shock of the Communists' unprecedentedly massive spring military campaign, the "Tet (Spring) Offensive." This episode portrays the role CIA played in the related episodes of the MACV order-of-battle (O/B) controversy and the runup to the Tet offensive. We will see that CIA's estimates of the enemy's strength were considerably more accurate than those turned out elsewhere; that CIA's Saigon Station accurately warned that a Tet-like general offensive was coming; that CIA Headquarters did not share that warning; and that senior policymakers, in any event, both overrode CIA's insistence that MACV's estimates of the enemy's order of battle were much too low, and ignored Saigon Station's warning that an unprecedented enemy offensive was at hand.
The O/B Controversy
So far, our mission frustratingly unproductive since MACV stone-walling, obviously under orders. . . . [the] inescapable conclusion [is] that General Westmoreland (with [CORDS Chief] Komer's encouragement) has given instruction tantamount to direct order that VC total strength will not exceed 300,000 ceiling. Rationale seems to be that any higher figure would generate unacceptable level of criticism from the press. This order obviously makes it impossible for MACV to engage in serious or meaningful discussion of evidence.
- George A. Carver (DCI's Special Assistant for Vietnam Affairs),10 September 1967(2)
For years US military chiefs in Vietnam had estimated the enemy's military and guerrilla forces at much lower levels than CIA analysts in Saigon and Washington thought justified.(3) Ultimately, it would take the Communist Tet Offensive of 1968 to rip away the paper backdrop MACV's order-of-battle staff had erected behind the "command estimate" of enemy strength. The last of several attempts to resolve the disparity ended abruptly some four months before the Tet Offensive when emissaries of the Intelligence Community went head-to-head with General Westmoreland and his immediate MACV subordinates. Despite the Viet Cong's demonstrated persistence and strength, and in the face of evidence that Communist regulars and irregulars might total half a million, MACV insisted that enemy forces in South Vietnam could not be numbered at more than 300,000. The Intelligence Community, not without working-level protests, so reported to the President. MACV and the White House continued to use these lower figures in their public pronouncements.
MACV stuck to its lower O/B estimates for several reasons. First, the MACV staff had been claiming for some time that the enemy was suffering great losses in Vietnam, and in mid-1967 predicted that a "crossover" would soon occur when losses would exceed the replacement capacity; an accounting correction in the O/B would muddy the arithmetic behind this claim. Second, as CIA files show, MACV based its O/B estimates heavily on South Vietnamese (GVN) sources.(4) Third, MACV used mainly Confidential-level documents and prisoner interrogation reports, and, in contrast with CIA's practice, did not generally use data derived from intercepted enemy radio signals, or SIGINT.(5) And MACV's position rested on an O/B estimation process whose flaws its own officers could (and, later and privately, did) point out.
But the most important regulator of the MACV O/B estimates was the fact that General Westmoreland and his immediate staff were under a strong obligation to keep demonstrating "progress" against the Communist forces in Vietnam. After years of escalating US investments of lives, equipment, and money, of monthly increases in MACV's tally of enemy casualties, and of vague but constant predictions of impending victory, it would be politically disastrous, they felt, suddenly to admit, even on the basis of new or better evidence, that the enemy's strength was in fact substantially greater than MACV's original or current estimates.
Working-level intelligence analysts were less affected by this MACV concern, and by early 1967 even some senior Pentagon officials, both military and civilian, had become uneasy with challenges about the accuracy of MACV's judgments on a number of questions, among them the enemy's O/B and the overall progress of the war. In January of that year Gen. Earle Wheeler, Chairman of the Joint Chiefs of Staff (JCS), registered his dissatisfaction with "the contradictory order of battle and infiltration statistics which are contained in numerous documents now being circulated."(6) Responsibility for solving the problem was thereupon given to CINCPAC, the armed services' commander-in-chief in the Pacific theater, who convened representatives from his own headquarters in Honolulu and from MACV, the Defense Intelligence Agency (DIA), the National Security Agency (NSA), and CIA. The conferees were directed "to standardize and agree upon definitions, methodology and reporting procedures" on some 15 questions concerning the enemy's military strength, and in due course they agreed to adopt a number of positive new measures. In subsequent practice, however, MACV did not budge from its previous O/B estimates.
Meanwhile, the principal Pentagon skeptic about MACV's judgments had become none other than Secretary of Defense Robert McNamara. CIA's George Allen had received intimations of this as far back as 1964, when during a trip to Saigon, McNamara had confided to him that the situation in Vietnam was "far worse" than the Pentagon at large realized.(7) By 1966, following a year of US airstrikes against North Vietnam, McNamara's continued concern about MACV's claims of "progress" had impelled him to ask CIA's Directorate of Intelligence for a private assessment of the enemy's will and ability to continue the war. The study, delivered to McNamara on 26 August 1966 under the title "The Vietnamese Communists' Will to Persist," concluded that planned US measures were not likely to deter the North Vietnamese. According to historian Gen. Bruce Palmer, Jr., this CIA assessment made a deep impression on Secretary McNamara and "no doubt had much to do with changing his views about the war."(8) In any event, in April 1967, according to CIA files, McNamara took the extraordinary step of asking CIA, a nonmilitary agency, to provide him periodic, independent assessments of enemy O/B, as well as of the effectiveness of US air operations against the North and the progress of pacification efforts in the South.(9) It was two months later, in June 1967, that McNamara quietly directed a small group of staff officers (military and civilian) to compile the massive record of the US war effort that became known as The Pentagon Papers.
Until McNamara made his unusual request of the Agency, there had been no centralized or systemized CIA work on the Vietnamese Communist O/B, mostly because of a general appreciation within the Agency that this chore was the proper responsibility of MACV, not a civilian office. Prior to McNamara's tasking, certain officers in various offices of the Agency--SAVA, O/NE, OCI, and ORR (Office of Research and Reports)--had dug into the O/B problem informally from time to time. Though differing in approach and degree, they reached a common conclusion: MACV's estimates of enemy O/B were much too low.
For instance, when CIA analysts focused on the mostly civilian and irregular components of the O/B as a legitimate object of their analysis, they found many problems. As veteran military analyst George Allen later wrote, MACV's order-of-battle holdings had long been "misleadingly low. . . . They had done almost no real research on the guerrilla-militia forces; their estimate remained at the 'guesstimate' my [DIA] team had come up with in Saigon early in l962."(10)Allen's boss at CIA, Special Assistant George Carver, told a White House military aide in September 1966 that MACV's estimate of 100,000 to 120,000 Viet Cong irregulars "may be extremely low."(11) In January 1967, O/NE observed that documentary evidence suggested that the enemy's irregular strength in South Vietnam had reached 250,000 to 300,000 by the end of 1965, whereas MACV was still sticking to its 100,000 to 120,000 estimate.(12) In May 1967, shortly after McNamara's tasking of CIA and at a time when MACV was carrying a total enemy O/B in South Vietnam of 292,000, CIA responded to an inquiry by Under Secretary of State Nicholas Katzenbach that the enemy's paramilitary and political organization in South Vietnam "is still probably far larger than official US order of battle statistics indicate," and thus that the total enemy O/B there "is probably in the 500,000 range and may even be higher."(13)
CIA's most diligent researcher into the O/B problem and Washington's boldest champion of a higher O/B figure was Sam Adams, a brilliant, energetic OCI and (later) SAVA officer. By December l966, for example, after examining the situation firsthand in Vietnam and digesting stacks of raw reports at CIA Headquarters, Adams concluded that the total number of enemy forces in South Vietnam was 600,000.(14)
Contrary to his later claims, Adams was not a lone voice crying in the wilderness. Through mid-1967 at least, his boss, George Carver, and many officers elsewhere in CIA agreed with his general argument that the enemy O/B figures, especially those for VC irregular forces, should be much higher than what MACV was accepting. In January 1967, Carver advised the DDI that MACV's O/B total "should be raised, perhaps doubled."(15) In April, Carver took Adams's conclusions to NSC staffer Robert Komer--soon to be the President's deputy for civil operations and rural development support (CORDS) under General Westmoreland in Saigon. He told Komer that while O/B data on North Vietnamese and Viet Cong Main Force and Local Force units were fairly reliable, the accepted number of irregular forces was being substantially underestimated and might number "more than 300,000," making the total O/B "as high as 500,000."(16) In May, drawing on Adams's work, Carver drafted the report CIA gave Under Secretary of State Katzenbach, referred to above, which held that the total enemy O/B was about 500,000 and might be even higher. And in June, Carver cabled Komer in Saigon that "the fundamental problem at this time is to overcome the longstanding cultural lag in MACV's holdings which have never reflected the substantial growth in non-combat, non-main force strength which occurred between 1962 and 1966."(17)
CIA's formal estimators, the Office of National Estimates (O/NE), did not accept Adams's highest O/B estimates, but did give him several hearings and did agree that MACV was significantly underestimating enemy O/B. In January 1967, O/NE told DCI Helms that there "is now documentary evidence which strongly suggests that at the beginning of 1965, [VC] irregular strength was about 200,000," that is, twice the going estimate accepted by MACV, and that during 1966 the number of irregulars had probably grown to some 250,000 to 300,000.(18) Adams's strongest backer in CIA was George Allen, who had been studying the Vietnamese "people's war" in Indochina since the early 1950s. When, along with Adams, he expressed certainty that MACV was grossly underestimating the enemy's irregular forces, he was drawing not only on years of immersion in Vietnamese military affairs, service as both a US Army and a DIA analyst, and careful study of available intelligence, but also on his reading of Hanoi commander Vo Nguyen Giap's doctrine that the irregular troops, guerrillas, and militia constituted "core forces" which had an "extremely important strategic role" to play in the war effort.(19)
Coincident with these developments at CIA, a few military officers had begun to agonize about MACV's O/B totals. According to George Allen, Col. Gaines Hawkins (chief of MACV's O/B section) confided to him in July 1967 his belief that the proper O/B figure ought to be much higher, but said "our hands are tied; this is a command position; we have to stay within a total figure of 300,000; I personally share your 500,000 estimate, but we cannot accept it."(20) Field reports sent to the Pentagon by Hawkins's boss, Gen. Joseph A. McChristian (MACV J-2), had contributed to a growing belief among working-level officers of the Defense Intelligence Agency that General Westmoreland's official O/B estimates were too low.(21) Even though McChristian and Hawkins did not succeed in getting General Westmoreland to revise his "commander's estimate" of enemy forces, their reports from the field, together with CIA's document-based O/B assessments, generated enough concern at the top of the Pentagon to cause JCS Chairman Wheeler to convoke the Honolulu conference and Secretary of Defense McNamara to commission regular independent O/B assessments from the CIA.
Responsibility for resolving the now-open dispute among the O/B analysts rested ultimately with the Chairman of the United States Intelligence Board, DCI Richard Helms. Helms was a reluctant adjudicator. As early as January 1967, he foresaw that what he termed "the Vietnam numbers game" would be played "with ever increasing heat and political overtones" during the year.(22) By May, at least, he had recognized the disruptive potential of the commission McNamara had given the CIA, a civilian office, to report to him regularly, without DIA or MACV coordination, on its O/B figures(23) So Helms told CIA and the military intelligence agencies to come up with an agreed figure, if at all possible, in a definitive Special National Intelligence Estimate (SNIE 14.3-67) in which the question of the enemy's O/B could not be ducked. When the military and civilian analysts deadlocked on their disparate bookkeeping of enemy strength figures, Helms personally called an interagency meeting on the draft SNIE late in June and told those present that the O/B question had become "the most important disagreement about the war," that "we've got to come to an agreement," and that the disputants should go back and work out an answer to the problem.(24)
The impasse nevertheless persisted, with DIA insisting on MACV's "official" figures, even though some of these had not been changed in several years of combat and seemed to lack evidentiary basis. Nor did it help that MACV sent some of its intelligence staffers to Washington to join the discussions; the problem, as Colonel Hawkins (above) revealed to George Allen, was that the MACV representatives were under orders not to yield on their "commander's estimate" of fewer than 300,000 men. Helms, as USIB Chairman, then commissioned his Special Assistant for Vietnam, Carver, to take a team of Intelligence Community analysts to Saigon and grind out directly with MACV an agreement on the estimated strengths of the enemy's several armed components.
The Saigon conference ensued, Carver heading a Washington team that included Sam Adams; William Hyland, chief of O/NE's Far East staff; the chief of the North Vietnam branch in CIA's Office of Current Intelligence; and George Fowler, DIA's senior civilian analyst on Vietnam. In Saigon Carver ran into a MACV brick wall. Although he and his experts poked gaping holes in the evidentiary basis for MACV's O/B estimates, Carver could not budge Col. Daniel O. Graham and his MACV O/B estimates team. Cabling Helms on 10 September, Carver characterized his mission as "frustratingly unproductive since MACV stonewalling, obviously under orders." MACV's officers will not accept any O/B total larger than 298,000, said Carver, and "the inescapable conclusion" must be drawn that Westmoreland "with Komer's encouragement has given instructions tantamount to direct order that VC strength total will not exceed 300,000 ceiling." Carver added that he hoped to see Westmoreland and Komer the next day and would "endeavor to loosen this straitjacket. Unless I can, we are wasting our time."(25)
According to Carver, he and his Washington colleagues found some of the top Saigon officers not only adamant but also personally insulting. Reporting to Helms on the 12th, Carver described his meeting the previous day with Komer and Maj. Gen. Phillip Davidson (McChristian's successor as Westmoreland's J-2) in these terms: the meeting ended in an "impasse"; at one point "I was frequently and sometimes tendentiously interrupted by Davidson . . . [who] angrily accused me of impugning his integrity," and who stated that the figures MACV had tabled were its "final offer, not subject to discussion. We should take or leave it."(26)
Carver found Komer similarly difficult. He reported that at their 11 September meeting, Komer "launched into an hour-long monologue, " stressing MACV's inability to convince the press and the US public of "the great progress being made" in the war effort and of "the paramount importance of saying nothing that would detract from the image of progress" being made in Vietnam. Komer criticized the SNIE draft under way in Washington, "faulting the quality of its prose and its analysis and calling it a sloppy, thin and altogether disappointing piece of work." Komer then derided the Agency's entire analytic effort on Vietnam: in his view, CIA had "only a small number of analysts working on Vietnam, none of whom know much about it," and CIA's analysis consequently "could not expect to compete in depth and quality to that of MACV." Komer concluded that there must not be any quantifying of the enemy's irregular forces, on the grounds that so doing "would produce a politically unacceptable total over 400,000."(27)
Carver had no better success in budging Westmoreland's officers from their insistence that no figure for enemy irregulars could or should be included in the total estimated O/B. The MACV officers based their insistence on several arguments: the difficulty of categorizing such an amorphous and differing body of part-time enemy forces; the paucity of hard intelligence on such forces; and the argument that, at least according to MACV, such irregulars were not very important to the enemy war effort anyway. At the forefront of MACV's adamant position was its J-2 estimates chief, Col. Daniel O. Graham. Having for some time predicted the imminent arrival of the "crossover point," in Saigon (according to George Allen), Graham now challenged Carver's presentation of Washington's O/B analysis, disparaging the irregular forces as having no military significance. Years later Graham admitted to George Allen that "of course" he had not believed MACV's 300,000 figure but had defended it because it was "the command position."(28)
Carver's task in Saigon that September was further complicated by the fact that his own team was not of one mind. According to Bob Layton, an analyst then serving at Saigon Station, O/NE's William Hyland told Sam Adams that the particular O/B numbers didn't make much difference, while Adams raged against what he termed the "rug bazaar bargaining" by which MACV proposed to reach an agreed estimate.(29) And Carver's DIA team member, George Fowler, took positions midway between those of Carver and MACV.(30) Moreover, Saigon Station officers had for some time tended to defend MACV's O/B ceilings because of their own concern about the political embarrassment that might ensue if they were radically expanded.(31)
In any event, after three days of heated exchanges with MACV, George Carver suddenly changed course, agreeing (over Sam Adams's outraged protests(32)) to a major compromise which essentially accepted MACV's position. On 14 September, all parties--MACV, DIA, INR, and CIA (represented by Carver)--agreed (1) that the total enemy O/B figure should be 249,000; (2) that the total enemy political cadre O/B should be 85,000; and (3) that no quantified estimate should be given for "irregulars," that is, enemy self-defense forces, secret self-defense forces, assault youth, and so on. This agreement was reached, moreover, in an atmosphere entirely different (or ostensibly so) from that of the conference's stormy beginning. Upon returning to Washington, Carver now asked DCI Helms to express gratitude to General Davidson and "all of his able, most impressive staff, the thanks of the entire Washington delegation for their effective, comprehensive briefings and other invaluable contributions to the success of our joint endeavor."(33) Helms was able to report to Under Secretary of State Katzenbach that agreement had been reached on the O/B figures, that Ambassador Bunker was consequently "most pleased," and that Bunker and General Westmoreland were "very complimentary about [the] Washington delegation mission."(34)
The key question, then and now, is what pressures prompted Carver suddenly to agree to a virtual 180-degree turnaround on figures and mood. The available documentary record is not clear and the recollections of participants differ. There appear to have been several reasons why Carver backed off from his initial anger at MACV officers' stone-walling and peremptory behavior. As we have seen, Carver's team was itself not of one mind, nor had all CIA's Washington analysts agreed with the initial Carver-Allen-Adams positions. And even though he initially argued in Saigon for much higher O/B figures, Carver's freedom of action was circumscribed by the general requirement DCI Helms had given him to reach an agreed figure with MACV.
Ultimately, however, Carver had to cave in because he was up against a hostile united front of superior officers who feared the adverse political consequences of a suddenly enlarged O/B total. Heading that opposition was not only MACV commander Westmoreland but also Ambassador Ellsworth Bunker. Even after the conference had ended to the Saigon establishment's complete satisfaction, Bunker the next month sent an Eyes Only cable to the White House warning that there would be a "devastating" result "if it should leak out . . . that despite all our successes in grinding down VC/VNA here," statistics showed that "they are really much stronger than ever."(35) Gen. Creighton Abrams, the MACV deputy to Westmoreland, had made the point explicit when he reminded JCS Chairman Wheeler in August 1967 that "We have been projecting an image of success" over recent months, and warned that, if a much higher O/B figure were released, newsmen would draw "an erroneous and gloomy conclusion as to the meaning of the increase."(36) National Security Adviser Walt Rostow repeated the argument in Washington, advising President Johnson that "the danger is press will latch on to previous underestimate [of O/B] and revive credibility gap talk."(37)
These were the arguments given Carver by the immediate point man for Saigon's defense of its numbers, Ambassador Robert Komer--who was not only chief of CORDS but a close aide of LBJ's, a friend of Carver's, and a very forceful personality known appropriately as Blowtorch Bob. In June 1967 Komer had cabled Carver from Saigon that any upward revision of O/B would make it appear that the United States had not been doing a good job of whittling down enemy strength.(38) When Carver visited Saigon the next month, Komer told him that the release of increased O/B figures would cause political problems for MACV because this would come at a time when General Westmoreland was asking for more US troops.(39) And after Carver returned to Washington in August, Komer cabled him that he could not see the case for including "low-grade part-time hamlet self-defense forces" in a new O/B; doing so would create a "ruckus" that would further widen the credibility gap at the very time "when in fact we are moving toward much more valid estimates."(40) Then, according to George Allen, during the September conference in Saigon, Komer told Carver over dinner that "You guys simply have to back off. Whatever the true O/B figure is, is beside the point." If a much larger figure should be published, said Komer, within hours "some dove in State will leak it to the press; that will create a public disaster and undo everything we've been trying to accomplish out here."(41)
An important question raised about Carver's sudden cave-in to MACV is whether DCI Helms ordered it in a cable to Carver during the Saigon conference. Here the record is contradictory. A few participants in the O/B controversy maintain that Helms did send such an explicit cable. For example: Robert Komer: "Why did George Carver cave in and compromise with MACV on the O/B question? Because that's what Helms told him to do."(42) Thomas Powers: "The deadlock was finally broken on September 11, 1967, when the CIA station in Saigon received a cable for FUNARO (Carver's official CIA pseudonym) from KNIGHT (Helms's pseudo) which directly ordered Carver to reach agreement."(43) Sam Adams: "The Saigon conference was in its third day when we received a cable from Helms that, for all its euphemisms, gave us no choice but to accept the military's numbers."(44)
The testimony of other observers is less clear as to whether Helms directly ordered Carver to back down. Bob Layton: "I heard long ago that Helms sent such a cable but I have never seen it; my position in the Saigon Station at the time was one where I would have been unlikely to have been cut in on such a sensitive directive."(45) R. Jack Smith recalls that it was Carver who suggested the compromise and that Helms then "instructed Carver to proceed according to his own best judgment."(46) John Ranelagh: "At the end of the Saigon meeting on September 13 Carver cabled to Helms that he had made a major concession in not quantifying the irregular forces, because this had been MACV's major sticking point."(47) On 13 September 1967, Carver cabled Helms that he, Carver, had worked out an O/B compromise with Westmoreland, subject to Helms's concurrence.(48) Carver later recalled, "I saw Westy and suggested the deal . . . I then told Dick what we were going to do; he did not give me any orders."(49) Helms said in 1992: "I have no recollection of having cabled George in Saigon, ordering him to strike a bargain. He already knew my basic views: that because of broader considerations we had to come up with agreed figures, that we had to get this O/B question off the board, and that it didn't mean a damn what particular figures were agreed to."(50)
Although the record is ambiguous, it is clear that Carver and the facts of intelligence were outgunned. Whatever the O/B evidence, the governing considerations were political, especially the need to protect a beleaguered administration against a new, crucially damaging embarrassment. Certainly there were broader considerations underlying Helms's decision to direct (or to accept) a major compromise with MACV. In the view of this author, the DCI may have believed that CIA's future analytic credibility, as well as his own continuing entree to policymakers, could be better served if he did not arbitrarily marry the Agency to particular O/B estimates that other officials considered extreme. In any case, Dick Helms had numerous equities in maintaining good relations with the military members of the Intelligence Community on other questions. He also was still in a sharp struggle with the military and the White House on a crucially important question: CIA's support of Secretary McNamara's view at the time that US bombing campaigns in Vietnam were not materially hampering the enemy's war effort. As biographer Thomas Powers has phrased it, that bitter bureaucratic struggle "made the OB fight look like a mild disagreement," and Helms gave in to MACV's O/B position "because he just did not want to fight about the OB along with everything else."(51)
Whatever the reasons, reactions to Carver's Saigon concessions varied widely. George Allen recalls his fury upon learning of the suddenly struck bargain: "I had never been so angry in my life, and I toyed with the idea of resigning from CIA."(52) Similarly, at the Westmoreland vs. CBS trial in 1984, Allen termed "unprincipled" the O/B positions MACV had taken in 1967, a "prostitution of intelligence"; the CIA "had sacrificed its integrity on the altar of public relations and political expediency."(53) Understandably, Sam Adams was also dismayed: "I left the Saigon conference extremely angry," and when he was asked thereafter by a member of the Board of National Estimates whether CIA had "gone beyond the bounds of reasonable dishonesty," he replied that that had occurred even before the Saigon conference.(54) Adams told Carver that General Westmoreland's O/B figures had been "a monument of deceit," and that the Agency's retreat had been an "acquiescence to MACV half-truths, distortions, and sometimes outright falsehoods."(55) Even Saigon Station observed after the conference that MACV was still "officially carrying the ridiculous [O/B] figure of 112,760 irregulars, unchanged for over a year and a half."(56)
Not all CIA officers were critical of the compromise struck in Saigon. George Carver continued to defend the bargain he struck, holding that the enemy's later inability to follow up on its Tet Offensive showed that it did not have the requisite strength; consequently, in his view, the O/B figures that had been agreed to in Saigon proved "essentially in the right ball park."(57) R. Jack Smith, the DDI in 1967, who had to sign off on the SNIE draft, took a similar position: the agreement worked out by Carver was "a highly enlightened formulation," and it would have been "simplistic and intellectually dishonest" to have insisted on the higher CIA O/B figure, "based as it was on 'spongy' evidence and a complex methodology."(58)
Endorsements of the Saigon conference compromise were undercut by military officers' later admissions that they had known at the time that General Westmoreland's insistence on an O/B total of no more than 300,000 was an artificial position dictated by political considerations, and that the true number of enemy forces had almost certainly been much higher. In addition to General Graham's admission, discussed above, former MACV J-2 officers General McChristian and Col. Gaines Hawkins so testified at the CBS trial.(59)
Even so, having concluded the September 1967 Saigon bargain with MACV, when CIA's intelligence managers moved to coordinate the long-fought SNIE they found that their analysts still had to battle separate MACV estimates and characterizations of the enemy forces. The coordination process took two more months, during which time CIA and the military clashed repeatedly over markedly differing estimates of the enemy's irregular strength. Those differences surfaced almost immediately after the Saigon conference, according to Carver, when the military backed away sharply from the agreement struck with him there. In Washington, the Pentagon's Assistant Secretary for Public Affairs, Philip Goulding, circulated a draft press statement announcing that enemy irregular strength in Vietnam was now down. Carver protested angrily, telling Goulding that what he proposed did not jibe with the Saigon negotiations and, furthermore, that "evidence continues to come in showing that the VC make considerable use of these 'irregulars' and not infrequently assign them actual combat tasks."(60)
MACV, according to CIA files, also prepared a press statement on O/B that deserted the positions it had just hammered out with CIA. The MACV statement inflamed a senior Office of Economic Research (OER) officer who had been helping coordinate the SNIE (and whose office soon came to have primary responsibility within CIA for following enemy O/B). He alerted Carver that the statement implied a coincidence of views between MACV and CIA that did not exist; the draft even suggested that "we have overestimated guerrilla forces" and, in sum, constituted "one of the greatest snow jobs since Potemkin constructed his village."(61) Despite objections by CIA officers, MACV proceeded to announce that the total North Vietnamese/Viet Cong order of battle in South Vietnam had dropped from an estimated 285,000 to 242,000, a decline of 43,000.(62) CIA files show that MACV's classified estimate, produced by Col. Graham, was even more optimistic, holding that enemy strength in South Vietnam had dropped from an estimated 285,000 to 235,000, a decline of 50,000.(63)
As work on the SNIE ground forward into November, Saigon Station offered up an imaginative suggestion for a wholly new approach to quantifying VC irregular strength. The Station asserted that the VC had access to between 2 1/2 and 3 million people in territories they controlled, plus more in disputed and GVN-held territory. The VC considered all these people to be "resources" to be used in war, and "a large part of them are directly employed on a sporadic, part-time, or full-time basis in military and quasi-military activities. . . .What they do for the Viet Cong is largely done by uniformed full-time troops in allied forces, which makes comparison of numbers misleading."(64)
Such efforts to persuade the military analysts to rethink their artificially constructed tallies of irregular forces proved unavailing, and the finished SNIE,(65) published on 13 November 1967, represented a rout of CIA's yearlong efforts to show that the enemy in Vietnam was far more numerous than MACV had been estimating. The SNIE's Conclusions, outlined at the beginning of the Estimate, and a table accompanying the text, stated that enemy regular force strength there was 118,000 and its guerrilla strength 70,000 to 90,000, for a total of 208,000 at most. This was substantially less than Carver's team and MACV had agreed to in Saigon just two months previously. Even more remarkable, to those readers familiar with it, was the contrast with the previous year's National Intelligence Estimate of the enemy's O/B in South Vietnam, which in July 1966 had judged that the enemy had some 285,000 to 305,000 troops.(66) The new SNIE reduced that total by close to 100,000.
SNIE 14.3-67 did explain that its statistical categories differed from previous NIEs, and in its Discussion section it did admit that the enemy's total strength, counting his entire military and political organization, was "of course considerably greater than the figure given for the Military Force." Prose caveats buried deep in the SNIE, however, could not compete among senior readers with the impression created by the tabulation of ostensibly hard numbers up front in the Conclusions section.(67) The SNIE, moreover, repeatedly stressed that the enemy's strength in South Vietnam was declining and his guerrillas had "suffered a substantial reduction." There was "a fairly good chance" that the overall strength and effectiveness of his military forces and political infrastructure would "continue to decline." Thus, said SNIE 14.3-67, the enemy has been reduced to carrying out a protracted war of attrition: he would still have the capability to continue "some forms of struggle--though at greatly reduced levels."
Of the many and various reactions in CIA that the SNIE evoked, the following three fairly represent the spectrum:
- On the eve of the SNIE's publication, Sam Adams sent DCI Helms some blistering comments on the draft. In Adams's view the Estimate was ill formed and incoherent, less than candid, and unwise: it did not come to grips with "the probability that the number of Viet Cong, as currently defined, is something over half a million. Thus it makes canyons of gaps, and encourages self delusion."(68)
- Sherman Kent's successor as head of O/NE, Abbot Smith, told Adams that the SNIE's managers and the Board of National Estimates "had had no choice: Helms had agreed to accept the military's figure, it was his paper ultimately, what could they do?"(69)
- Helms, who recognized that the SNIE's disputed O/B figures could prove very sensitive politically, told President Johnson the day after the completed SNIE had been disseminated that he had considered not publishing it at all.(70)
In such manner the intelligence and policy communities entered the new year of 1968 with MACV, the CIA, and an authoritative SNIE backing up the perception that the enemy did not have the capability to launch major operations. Most important, the publishing of the Estimate coincided with a White House-orchestrated public relations campaign that emphasized the bright developments supposedly taking place in the war effort. Typical of such claims was General Westmoreland's speech to the National Press Club in Washington on 21 November 1967, just a week after the SNIE had been published, in which he said, "I am absolutely certain that whereas in 1965 the enemy was winning, today he is certainly losing. . . . It is significant that the enemy has not won a major battle in more than a year. . . . [He] has many problems: he is losing control of the scattered population under his influence. . . . He sees the strength of his forces steadily declining."(71)
And so, as the 1968 Tet holiday approached in Vietnam, optimism was about to be revealed as self-delusion.
These same [enemy] documents call for all-out, coordinated attacks throughout South Vietnam utilizing both military and political means to achieve "ultimate victory" in the near future. . . . VC/NVA strategy toward the war appears to have reached a crucial phase in which changes in the tempo and scale of the war are envisioned. . . . In sum, the one conclusion that can be drawn from all of this is that the war is probably nearing a turning point and that the outcome of the 1967-1968 winter-spring offensive will in all likelihood determine the future direction of the war.
- CIA Saigon Station Dispatch, 8 December 1967(72)
[This field study of 8 December quoted above] should not be read as the considered view of this Agency. . . . [The Station's assessment has been] predicated on certain assumptions whose validity seems questionable from our perspective here in Washington.
- George Carver, Memorandum to the White House's Walt Rostow, 15 December 1967(73)
We will see (1) that three intelligence components, only, rang fairly sharp alerts prior to the Tet Offensive--the Army communications intelligence group supporting Maj. Gen. Frederick C. Weyand's 3rd Corps, National Security Agency Headquarters, and CIA's Saigon Station; (2) that their alerts barely registered outside the immediate tactical scene in Vietnam; and (3) that the rest of US intelligence, CIA Headquarters included, did little to prepare policymakers for the fact, scope, or significance of the Tet Offensive. The result was that the sudden, countrywide enemy attack stunned the Johnson administration and the American public and left an unbridgeable credibility gap between them.
Some postmortem judgments of the pre-Tet intelligence performance have been harsh, as witness the evaluation in a West Point textbook published a year later: "The first thing to understand about Giap's Tet Offensive is that it was an allied intelligence failure ranking with Pearl Harbor in 1941 or the Ardennes Offensive in 1944."(74) Or the judgment of former Secretary of Defense Clark Clifford: "The fact is that three months before the offensive both Westmoreland and Ellsworth Bunker . . . loudly proclaimed that enemy strength was decreasing. . . . [Their] telegrams contained not one word of warning about the possibility of large-scale, coordinated attacks in the future. On the contrary, they . . . must rank among the most erroneous assessments ever sent by field commanders."(75)
There were significant external influences on the failure of US intelligence, CINCPAC, MACV, the Saigon Embassy, and the White House to anticipate the 1968 Tet Offensive. Among them were the distractions of near-simultaneous foreign incidents, mostly in East Asia, that demanded the attention of intelligence analysts, diplomatic and military officers, national security strategists, and the President: North Korea's seizure of the USS Pueblo; a North Korean penetration of South Korean President Park Chung Hee's residence; Seoul's subsequent pressures on Washington to permit South Korea to withdraw some of its military units from Vietnam; the Communist capture of a vital outpost in easternmost Laos; serious new pressures on the West Berlin air corridor by Soviet aircraft; and the crash of a US B-52 laden with nuclear weapons. A pervasive and probably more important contribution to the failure was Lyndon Johnson's preoccupation, as the presidential election year approached, with demonstrating success in Vietnam in the face of the sharply rising tide of public opposition to the war.
But the most important cause of American surprise was the deliberately optimistic mindsets key policymakers had adopted and continued to project in the runup to Tet. President Johnson, National Security Adivser Walt Rostow, Secretary of State Rusk, JCS Chairman Wheeler, CINCPAC Adm. Ulysses S. Grant Sharp, Ambassador Bunker, and MACV commander General Westmoreland all appeared confident that American ground and air operations were so grinding down Communist forces in Vietnam that they would not be able to maintain anything more than a limited war of attrition. The pronounced gulf between their beliefs and reality deserves representative highlighting:
Robert Komer, March 1967: Mr. Komer opened [the White House meeting] by exuding optimism on the current trend in Vietnam. . . . [He] expressed consider able disdain for MACV J-2, and particularly what he believes to be its overall underestimate of enemy strength. . . . Concluding, Mr. Komer recognized the possible trip-ups in the overall situation but anticipated that unless they occur, major military operations might gradually fade as the enemy began to fade away or put his emphasis on a protracted guerrilla level war. In either case, he said, the size of the problem in Vietnam will diminish, and fewer U.S. resources will be needed.(76)
Walt Rostow, mid-1967: Chaired by Mr. Rostow . . . the [concern of this White House] group . . . was with opinion manipulation and political persuasion, with the aim of altering perceptions to make them coincide with specific notions, whether those notions were supportable by evidence or not.(77)
Gen. Earle Wheeler, August 1967: In his prepared testimony, General Wheeler stated that the air campaign against North Vietnam is going well. . . . In some instances where he did present intelligence estimates, he made it clear that he did not agree with the conclusions of the Intelligence Community.(78)
Walt Rostow, September 1967: Mr. Rostow . . . commented that he was "outraged" at the intellectual prudishness of the Intelligence Community [concerning its evaluation of the lack of progress in pacification].(79)
Gen. W. C. Westmoreland, November 1967: Infiltration will slow; the Communist infrastructure will be cut up and near collapse; the Vietnamese Government will prove its stability, and the Vietnamese army will show that it can handle the Vietcong; United States units can begin to phase down.(80)
Walt Rostow, January 1968: [Mr. Rostow criticized CIA for being "fixed on certain positions" and urged it to develop new analyses based on] certain totally different hypothetical key facts, e.g., . . that the gentlemen in Hanoi see the equation . . . as tending to indicate that one year from now, they will be in a considerably worse bargaining position than they are today; so that settlement now might be to their advantage.(81)
General Westmoreland, January 1968: The year  ended with the enemy increasingly resorting to desperation tactics in attempting to achieve military/ psychological victory; and he has experienced only failure in these attempts. . . . The friendly picture gives rise to optimism for increased successes in 1968.(82)
A "we are winning" consensus pretty much permeated the Saigon-Washington command circuit; intelligence reports and analyses that deviated from it tended to be discounted. The growing uneasiness about the course of the war expressed sporadically by a handful of senior statesmen(83) had little dampening impact on the pre-Tet convictions and pronouncements of the dominant administration officials.
Prior to the 1968 Tet Offensive, the quality of CIA officers' assessments of the situation in Vietnam was mixed. On certain questions their judgments were more accurate, overall, than those of the dominant policymakers. Those judgments have won kudos from a wide spectrum of observers. Notable among these is Gen. Bruce Palmer, Jr., who later wrote glowingly of "the extraordinarily good performance of the CIA" in its Vietnam analyses.(84) As we have seen, one of the questions working-level CIA officers had right was the enemy's order of battle. Despite the Agency's backdowns on this issue--in Saigon and in SNIE 14.3-67--most CIA analysts working on Vietnam continued to judge that the true totals of enemy forces were much higher than MACV had accepted into its O/B, and that local VC self-defense and irregular forces constituted a significant source of the enemy's effective strength. A representative example, contrasting sharply with the agreed language of the just-completed SNIE, was a December 1967 CIA study which stressed that over and above the accepted enemy O/B, "the Communists make a strong effort to organize much of the total manpower under their control into various work forces and semimilitary organizations. Among the most significant of these organizations are the local 'self-defense' forces."(85)
Another area where CIA's assessments looked good was in the evaluation of allied bombing efforts. Here the themes stressed in studies prepared in 1967-68 for the President, Walt Rostow, and Secretary McNamara were (1) that, although ROLLING THUNDER and other bombing programs were seriously complicating the enemy's war effort, the level of supplies getting through to the Viet Cong was continuing to rise; (2) that US bombing of North Vietnam was not proving a significant limiting factor on enemy operations in South Vietnam; and (3) that the DRV's ability to recuperate from the air attacks was of a high order.(86) CIA's good batting average on these bombing questions has been acknowledged by a wide range of commentators. Among them is David Halberstam, not notably a booster of the Agency, whose view is that Secretary McNamara "pushed the CIA very hard for judgments on how effective the bombing had been and received in return what were considered some of the best reports ever done by the Agency."(87) In 1970, George Carver, still the DCI's Special Assistant for Vietnam, judged that these earlier CIA bombing studies were probably the Agency's "most important contribution" to President Johnson's post-Tet decision (of 3l March 1968) to curtail US bombings of the North.(88)
In the months before Tet, as they had consistently held since mid-1963, CIA officers continued to judge that bombing (no matter how unrestricted) could not render North Vietnam physically incapable of carrying on the struggle and that the Communists would almost certainly try to match any US escalation of the war. And Agency assessments persisted in the view, although not as consistently or clearly, that the enemy was not really interested in negotiating a settlement of the war and would use negotiating tactics only to provide breathers for the next round of warfare and to gain concessions from the US/GVN side.
Meanwhile, despite their routine disregard of the Agency's negative judgments, and in the midst of the Intelligence Community's embroilment with the SNIE on enemy military capabilities, the President and his advisers in 1967 continued to enlist CIA's help and participation on a wide range of Vietnam projects. The White House repeatedly asked O/NE and SAVA to estimate probable enemy reactions to various theoretical US courses of action. Policymakers involved the Agency in programs to help CORDS develop more accurate technical systems for quantifying success in Vietnam, and CIA officers led an NSC interagency task force seeking better ways to judge Vietnam data and trend indicators.
In September and October 1967, George Carver and Richard Lehman (Deputy Chief of the DDI's Office of Current Intelligence) helped the Pentagon's Assistant Secretary for International Security Affairs and his deputy (Morton Halperin) do a special study of Vietnam policy alternatives for Secretary McNamara. Carver later realized that in this study the Pentagon staffers planted the seeds that blossomed into President Johnson's switch to a negotiating strategy in March 1968 (below). Also in October, DCI Helms received a request from Under Secretary of State Katzenbach for an Agency assessment of what GVN reactions might be to various kinds of discussions between Washington and Hanoi.(89) The workload of outside requests in November 1967 was equally heavy: Secretary McNamara asked the Agency to give him "a comprehensive review of where we now stand in Vietnam" to help him prepare for a national television interview.(90) And President Johnson asked for a CIA estimate of Viet Cong losses and casualties, a task that George Carver fielded in concert with Secretary McNamara.(91)
Some of the White House demands on CIA went far beyond usual intelligence matters. One such example was that of roping the Agency into the Johnson administration's wide-ranging effort, begun in the summer of 1967, to stimulate public support of the President's policies and programs in Vietnam. George Carver usually represented CIA in these White House meetings, which were chaired by Walt Rostow--with Carver's deputy, George Allen, subbing for him when he was absent. Allen terms CIA's participation in those gatherings on manipulation of domestic opinion "the most distasteful and depressing meetings of my bureaucratic career."(92)
A second example of White House pressures: in September 1967, Rostow told the Agency that because President Johnson wanted some "useful intelligence on Vietnam for a change," the CIA should prepare a list of positive (only) developments in the war effort. According to George Allen, SAVA refused to prepare such a study; but, at Helms's request, the DDI did prepare one. Helms sent it to Rostow with a cover note protesting the exercise and pointing out that this special, limited study was not a true picture of the war; but Rostow pulled off that cover note and so was finally able to give the President a "good news" study from the CIA.(93) It was also at this time that President Johnson asked CIA to prepare a questionable (and therefore super-sensitive) study on "The International Connections of the US Peace Movement."(94)
Even though CIA's judgments were contributing to Secretary McNamara's change of heart, as we have seen, the White House found many of them so uncongenial that the President, Walt Rostow, and others occasionally growled at CIA officers during these months for not being "members of the team." For example, according to George Allen, Walt Rostow more than once assailed him with such questions as "Didn't I want to win the war? Whose side was I on, anyway? Why didn't I join the team?"(95)
Not all of CIA's judgments were displeasing to the White House, nor did they all prove accurate, nor do they justify any ringing endorsement of CIA's overall analytical performance in the months leading up to the Tet Offensive. One issue on which the Agency's performance can be questioned was the stability of the Government of Vietnam. In December 1966, an NIE commissioned to forecast the GVN's performance had simply catalogued Saigon's areas of strength and weakness, without providing a clear overall message or bottom-line assessment.(96) Three months later, George Carver wrote Rostow that, even though there were still many soft spots and weak areas in the GVN's situation, "the overall progress made in the last twenty-odd months is inescapable and overwhelming."(97) Spurred by Saigon Station reports, however, Carver modified this optimism in mid-January 1968, asking the Station to inform State's Philip Habib, then visiting Vietnam, that concern was rapidly mounting in Washington over the "disquieting air of malaise and lassitude permeating the GVN."(98)
CIA's assessments of the military balance in Vietnam during this period can also be questioned. On 13 January 1967, Carver had recommended that Congress be told that the buildup of friendly forces in Vietnam was proceeding well, that the relative advantage over enemy forces had reached about four to one, and that in terms of combat potential the rate of growth was "even more favorable."(99) Two months later, Carver privately assured Rostow that that "there is a considerably better than even chance that within a reasonable time frame, say eighteen months, the total situation in Vietnam will have improved . . . to the point where all but the willfully obtuse will be able to recognize that the Communist insurgency is failing."(100) According to Clark Clifford (chairman of PFIAB in 1967, later Secretary of Defense), in the spring of 1967 "the CIA's top Vietnam expert, George Carver, told the PFIAB that by the fall of 1968 the situation should be dramatically improved."(101) In July 1967, Helms gave President Johnson an evaluation by C/FE Bill Colby, following an inspection trip to Vietnam, which concluded that even though there were fragile elements present, "it is very clear that my Soviet and Chinese counterparts' reports must exhibit great concern over the Viet Cong's mounting problems and the steady improvement in the ability of both the South Vietnamese and the Americans to fight a people's war."(102) According to Clark Clifford, briefings given by George Carver and JCS Chief Gen. Earle Wheeler to the President's panel of "Wise Men" on 2 November 1967 "set an upbeat and optimistic tone."(103) And a few days later, DDCI Rufus Taylor wrote Helms that he felt strongly that the "great" progress made in the past year "could be emphasized in press interviews and comment by public officials," and that this "might be of considerable help in countering the peaceniks."(104)
As for giving warning that a major enemy offensive was in the making, it will be seen that the best that could be said of CIA was that it sounded a distant trumpet from the field that came to be muted at Headquarters. But, except for the National Security Agency, no other components of the Intelligence Community did any better.
GVN intelligence collected a few indications, beginning about October 1967, that the enemy might launch an unprecedented winter offensive; and just hours before the Tet Offensive, the South Vietnamese produced at least two reports that proved extraordinarily accurate. The first was an intelligence report transmitted on 29 January 1968 to alert South Vietnamese tactical zone commanders that the Viet Cong would take advantage of the Tet Holiday in order to attack a number of provincial cities.(105) The second stemmed from the capture of an enemy soldier at 2100 hours on 30 January. He stated that Communist troops were going to attack central Saigon, Tan Son Nhut Airbase, and other installations in the capital city beginning at 0300 hours the next day--exactly the moment those attacks did start.(106)
These reports came much too late in the game, however, to help very much. For the most part GVN intelligence on enemy intentions prior to Tet was scattered, incomplete, and ambiguous. On the very eve of the enemy's offensive, CIA's Saigon Station Chief observed that the GVN police had a few scattered reports of upcoming enemy operations but nothing which appeared to be very hard.(107) Moreover, according to South Vietnamese security chief Col. Lung, most GVN commanders believed that the enemy was incapable of launching a major nationwide offensive in the near future; he added that most GVN units did not even share their intelligence take with one another.(108) Nor, according to MACV's J-2 at the time, Major General Phillip B. Davidson, did they pass on their reports to MACV. GVN officials were clearly not prepared for this attack on the opening day of Tet, when large numbers of them were celebrating with their families.(109)
MACV--and virtually everyone else--greatly underestimated the scope and intensity of the coming offensive and remained generally unaware of the enemy's overall intentions and timing, even though North Vietnamese newspapers were speaking rather freely of a coming campaign of "historic dimension." Nonetheless, by January 1968, MACV headquarters was persuaded by captured documents and other indicators that major shifts were occurring among many VC units. One of the clearest forecasts they had of a coming offensive was a VC document captured by US forces shortly before Tet which proclaimed that "the opportunity for a general offensive and general uprising is within reach," and that Viet Cong forces should undertake "very strong military attacks in coordination with the uprisings of the local population to take over towns and cities; troops should . . . move toward liberating the capital city, take power and try to rally enemy brigades and regiments to our side one by one."(110)
Enough such indicators reached General Westmoreland to prompt some concern and, almost at the last moment, some precautionary steps. On 25 January he cabled CINCPAC that that date seemed to be "shaping up as a D-Day for widespread pre-Tet offensive action on the part of VC/NVA forces."(111) On 30 January Westmoreland cancelled a previous Tet ceasefire for US troops and ordered that "effective immediately all forces will resume intensified operations, and troops will be placed on maximum alert."(112) Finally, convinced by intelligence alerts given him by his III Corps commander, Maj. Gen. Frederick C. Weyand, Westmoreland reversed the orders he had just given Weyand's 25th Division to undertake offensive sweeps in the countryside: instead, some of its units were brought into and around Saigon, increasing the number of US maneuver battalions protecting the capital to some 27. These precautionary moves doubtless saved Saigon and the US presence there from disaster.
General Weyand called his alerts largely on the basis of his analysis of enemy radio traffic and his professional belief that MACV was greatly underestimating the number and military significance of local VC forces.(113) An experienced intelligence officer, Weyand respected CIA officers and thought they were "focused on one of the right ways to defeat the enemy"; but in the case of the Tet offensive he felt that CIA and MACV did not provide any warning intelligence "worth a damn."(114) Former CORDS Ambassador Robert Komer is similarly critical: "neither CIA nor MACV provided any warning at all of the magnitude or the targets of the enemy's Tet Offensive; we were all completely surprised."(115)
The National Security Agency stood alone in issuing the kinds of warnings the US Intelligence Community was designed to provide. The first SIGINT indicators of impending major enemy activity began to appear in the second week of January 1968. In the following days NSA issued a number of alerts, culminating in a major warning it disseminated widely in communications intelligence channels on 25 January, titled "Coordinated Vietnamese Communist Offensive Evidenced in South Vietnam."(116)
In the period 25-30 January, NSA issued a number of followup alerts for specific areas of Vietnam. Even so, as NSA stated later in its review of Tet reporting, SIGINT was unable to provide advance warning of the true nature, size, and targets of the coming offensive. This was due in large measure to the fact that the enemy's local and irregular forces, which played such a large role in the offensive, made only limited use of radio communications.
Beginning in October 1967, CIA's Directorate of Plans made some 15 disseminations prior to Tet which, in hindsight, provided scattered indications that preparations might be under way in individual provinces for possible major enemy offensives of some kind before, during, or after Tet. These disparate reports, by themselves, did not add up to a sharp alert that an unprecedented nationwide attack was in the offing.(117) When the Intelligence Community later conducted a postmortem on its pre-Tet reporting for the President's Foreign Intelligence Advisory Board, it concluded that CIA field reporting "did not . . . reflect the massive character of the preparations under way all over South Vietnam for simultaneous invasions of nearly all major cities and towns. Nor did this reporting impart a sense that 'all hell' was about to break loose."(118) This was still the conclusion in 1975 when a DO/Vietnam branch officer did a new survey of the CIA field reports prepared prior to Tet; in his view, the warning they had given was "zilch."(119)
The current intelligence publications of CIA's Directorate of Intelligence distributed in the two months before Tet provided occasional intimations of impending Communist operations in the contested areas of northern South Vietnam, but no sharp warnings of a countrywide offensive. The treatment of East Asian matters by the Agency's premier publication, the President's Daily Brief (PDB), focused principally on South Vietnamese political developments; North Vietnamese and Communist Bloc attention to antiwar sentiment in the United States; the buildup of North Vietnamese military units just north of the Demilitarized Zone (DMZ); and especially the growing threat to the US outpost at Khe Sanh.
From 23 January onward, North Korea's seizure of the USS Pueblo dominated the PDB's reporting and analysis.(120)
The PDB at that time was primarily a vehicle for summarizing sensitive or late-breaking reports for the White House; lower-level White House officials and other consumers received the more inclusive Current Intelligence Bulletin (CIB) which, between 11 and 24 January, contained some eight reports on enemy activity, all confined to indications of scattered VC and NVA buildups in this or that local area, especially in the northernmost regions of South Vietnam. As Tet drew nearer, current intelligence publications did begin to focus on the possibilities of a large-scale enemy offensive. On 27 and 28 January the CIB replayed NSA's alerting memorandum of 25 January, reporting that communications intelligence had provided evidence of a widespread, coordinated series of attacks to be launched by the Communists. The 28 January CIB undercut that warning, however, by judging that the Communists intended to launch large-scale attacks on one or more fronts soon after Tet, and that it was not yet possible to determine if the enemy was indeed planning an all-out, countrywide offensive during, or just following, the Tet holiday period.(121)
The CIB for 29 January reported that North Vietnamese main force units were completing battle preparations in the western highlands of Pleiku and Kontum Provinces; that well-coordinated large-scale attacks may have been imminent there; and that the often mentioned "N-Day" may have been set for as soon as 30 January. The following day's current intelligence publications carried no reports or assessments of enemy intentions in Vietnam. On the 31st, as Communist assaults began to erupt all over the country, the DDI's published wrap-up of the situation characterized the enemy's attacks on US targets as harassments, and concluded that the enemy's operations to date might be preparatory to or intended to support further attacks in the Khe Sanh/DMZ/northern Quang Tri areas.(122)
The Intelligence Community's later postmortem described Washington's pre-Tet warning performance, overall, in these terms:
"The urgency felt in Saigon was not, however, fully felt in Washington in the immediate preattack period. As a result, finished intelligence disseminated in Washington did not contain the atmosphere of crisis prevalent in Saigon. We do not believe this represents a failure on anyone's part. The information available was transmitted and fully analyzed, but atmosphere is not readily passed over a teletype circuit. Although senior officials in Washington received warnings in the period 25-30 January, they did not receive the full sense of immediacy and intensity which was present in Saigon. On the other hand, with Saigon alerted, virtually nothing further could be done in Washington that late in the game which could affect the outcome."(123)
True, little could have been done in Washington to affect the outcome in Saigon and elsewhere in Vietnam, but an alerted Johnson administration could at least have prepared the public for the sudden turn of events and better eluded the charge that it and the GVN had been taken by surprise. The sum of the Intelligence Community's pre-Tet assessments was clearly insufficient to alert policymakers or the public to what proved to be a devastating political upset.(124)
The CIA's field intelligence analysis prior to Tet was extremely good, but its alerting performance went largely for naught. In November and December 1967, Saigon Station sent in three major assessments, each of which warned that a powerful, nationwide enemy offensive was coming. The second and most substantial of these studies predicted that the impending offensive "would in all likelihood determine the future direction of the war," a judgment Gen. Bruce Palmer later termed "an uncannily accurate forecast!"(125) That assessment and its two companions stand out as the finest predictive performance by any CIA entity in the weeks leading up to Tet. Untainted by the packaged optimism of the MACV reporting channel, and arriving in Washington far ahead of the disturbing but too-late tactical intelligence reports of enemy troop movements, the judgments in these assessments could have made a profound difference -- if only in bracing the administration for the Tet shock and giving it time to prepare the public. But, as we will see, the Saigon Station's assessments failed to shake the personal preconceptions of senior CIA and White House officials.
The three Saigon studies were the work of the Station's small assessments group headed by Bob Layton, an O/NE officer detailed to Saigon in mid-1967. The first two assessments (21 November and 8 December 1967) were produced as an apparently intentional overresponse to a request from the White House's Walt Rostow that the Station simply send in a list of its previous reports dealing with North Vietnamese/VC intentions. The November study included the requested wrap-up, but Layton and his colleagues added their own analytical estimate of the enemy's intentions in 1968 and promised a more thorough assessment in two weeks.(126)
Drawing heavily on prisoner interrogations and captured documents, this first field assessment concluded that the enemy seemed to be preparing an all-out effort to inflict a psychologically crippling defeat on allied forces sometime in 1968. The Station's analytic group called particular attention to numerous reports that enemy special action units had been directed to engage in widespread terrorism and sabotage in South Vietnam's major cities, coordinated with military attacks on the cities from without.(127) The Communists appeared to believe the time was ripe for such an effort, this assessment explained, because the GVN was perceived to be corrupt, unpopular, and incapable of gaining the allegiance of the bulk of South Vietnam's population and because the GVN's armed forces were suffering from serious morale problems and were incapable of advancing or protecting the pacification program. At the same time, according to this assessment, the US administration was becoming increasingly isolated internationally, was facing rising internal dissension, and thus wanted to end the war before the fall of 1968.
The Station's follow-up assessment of 8 December pondered the recent evidence of Communist exhortations for an all-out offensive against US/GVN forces and bases and decided that this represented a deliberate departure from the existing strategy of a patient war of mutual attrition. This thinkpiece began with a careful sifting of the increasing references in North Vietnamese and Viet Cong documents to the necessity to launch "an all-out military and political offensive during the 1967-68 winter-spring campaign [the period beginning around Tet] designed to gain decisive victory." As described in captured enemy documents and in accounts by prisoners of troop indoctrination sessions, the offensive would include both "large-scale continuous coordinated attacks by main force units, primarily in mountainous areas close to border sanctuaries"--a strategy subsequently reflected in the enemy's major attacks on Khe Sanh--and "widespread guerrilla attacks on large US/GVN units in rural and heavily populated areas." All-out attacks by both regular and irregular forces would be launched throughout South Vietnam, designed to occupy some urban centers and isolate others.
Layton concluded that "the VC/NVN . . . appear to have committed themselves to unattainable ends within a very specific and short period of time," which included "a serious effort to inflict unacceptable military and political losses on the Allies regardless of VC casualties during a US election year, in the hope that the US will be forced to yield to resulting domestic and international pressure and withdraw from South Vietnam." The approaching winter-spring campaign was shaping up as a maximum effort, Layton judged, using all current VC/NVN resources "to place maximum pressure on the Allies" for a settlement favorable to the Communists. And if, as was likely, they failed to achieve this maximum goal, Layton reasoned, they would at least have hurt the US/GVN forces, knocked them off balance, and "placed themselves in a better position to continue a long-range struggle with a reduced force." He continued: "If the VC/NVN view the situation in this light, it is probably to their advantage to use their current apparatus to the fullest extent in hopes of fundamentally reversing current trends before attrition renders such an attempt impossible." "In sum," the study's final sentence read, "the one conclusion that can be drawn from all of this is that the war is probably nearing a turning point and that the outcome of the 1967-68 winter-spring campaign will in all likelihood determine the future direction of the war."(128)
The Station's third alerting assessment (19 December) reiterated, with additional evidence, that available indicators showed Viet Cong/North Vietnamese forces were preparing something very much like an all-out push. Layton's group conceded (as Headquarters analysts had argued) that these enemy themes might be only propaganda designed to sustain VC/VNA morale, but the group doubted this. And though the projected offensive would cost staggering losses, the enemy nonetheless was prepared to accept them in order to accelerate what Hanoi believed was a sharp decline in the American will to continue the war.(129)
These remarkably prescient alerts, with their postulation of the enemy's reasoning and probable actions, met an unfortunate fate. Special Assistant George Carver, the senior CIA official in closest constant touch with the White House on Vietnam matters, administered a coup de grace to Layton's warnings. On 15 December Carver sent the Station's second (8 December) warning study to Walt Rostow but distanced himself and CIA from it. In his cover note Carver told Rostow that the attached field assessment "should not be read as the considered opinion of this Agency;" that it omitted reference to "other [unspecified] materials" bearing on the subject; and that the Station's assessment was "predicated on certain assumptions whose validity seems questionable from our perspective here in Washington." Carver questioned the assessment's thesis that the enemy was about to make crucial new decisions on the course of the war, and he told Rostow in effect that the Communists would continue their strategy of a limited war of attrition.(130) It is difficult not to agree with Gen. Bruce Palmer's later conclusion that Carver's throwing of "cold water on the [field's] studies . . . no doubt contributed to the unprepared state of mind in Washington when Tet 1968 hit."(131)
Worse still, Layton and his colleagues were contending against the judgments not only of the influential Carver, but of virtually all the Vietnam analysts then at CIA Headquarters. On 2 December, two weeks before Carver sent his dissenting cover note to Rostow, the Directorate of Intelligence had prepared a quick critique of Layton's preliminary (November) assessment; the analysts held that captured enemy documents did not indicate that the enemy was about to radically change his tactics, and did not suggest that the Communists thought they could really mount a decisive campaign.(132)
Doubly unfortunate for Layton and his colleagues was the timing of the second Headquarters product. On 8 December, the very day these field officers sent off their second (and most substantial) warning assessment, CIA Headquarters had just produced and distributed a major study--coordinated with all the Headquarters analytical offices--which differed sharply with Layton's conclusions. The 8 December Headquarters study told policymakers (1) because the war was not going well for the Communist forces, their present strategy was to hang on militarily and politically; and (2) the evidence suggested that for the present the North Vietnamese and the Viet Cong felt under no compulsion to abandon their basic objectives in the south or the means by which they were seeking to attain them.(133)
Thus, when Carver advised Walt Rostow on 15 December that, contrary to Saigon Station's warnings, the enemy was not likely to launch a sudden nationwide major offensive, he was speaking not only for himself but for CIA Headquarters--whose analysts of North Vietnamese strategy preferred their in-house expectations of rational behavior by Hanoi to radically new assessments from outside their ranks.
There was irony as well in the reception given Layton's warnings before and after Tet. At the beginning of 1968 no one exuded more confidence and less concern about the course of the war than President Johnson and his head cheerleader, Walt Rostow. Both men, however, later cited Layton's 8 December assessment as specific evidence that they had known all along the enemy's nationwide offensive was coming. In his memoirs, published in 1971, ex-President Johnson claimed that he had "agreed heartily with one prophetic report from our Embassy in Saigon [that the war was probably nearing a turning point and the outcome of the 1967-68 winter-spring campaign would in all likelihood determine the future direction of the war]. I was increasingly concerned by reports that the Communists were preparing a maximum military effort and were going to try for a significant tactical victory."(134) Similarly, writing in 1972, Walt Rostow quoted Layton's 8 December thinkpiece at some length, claiming that it indicated both the extent to which the structure of the Tet Offensive "was appreciated as early as December 8 and the kind of data available to Johnson at that time," and the fact that the President "had been receiving regularly and following closely the piecemeal evidence on which this summation was based."(135)
We have known for several months, now, that the Communists planned a massive winter-spring offensive. . . . The biggest fact is that the stated purposes of the general uprising have failed. . . . when the American people know the facts, when the world knows the facts and when the results are laid out for them to examine, I do not believe they will achieve a psychological victory.
- President Lyndon Johnson, 2 February 1968(136)
Gen. George Custer said today in an exclusive interview with this corresponden that the Battle of Little Big Horn had just turned the corner and he could now see the light at the end of the tunnel. . . . "We have the Sioux on the run . . . Of course we will have some cleaning up to do, but the Redskins are hurting badly and it will only be a matter of time before they give in."
- Art Buchwald, 6 February 1968(137)
Confounding the assurances CIA Headquarters had given the White House about the enemy's capabilities and intentions, 30 January 1968 brought the revelation that the Vietnamese Communists had changed their strategy, suddenly and radically. In the first two days of the Tet Offensive, enemy units attacked 39 of South Vietnam's 44 provincial capitals, five of six autonomous cities, 71 of 242 district capitals, some 50 hamlets, virtually every allied airfield, many other military targets, and Saigon itself. In the capital city, some 11 local force VC battalions struck the presidential palace, Tan Son Nhut Airbase, the Vietnamese Joint General Staff compound, and numerous other targets; not least, they penetrated and for a while contested the grounds of the US Embassy.(138) According to CIA files, JCS Chairman Earle Wheeler candidly told President Johnson on 27 February that the enemy's initial attacks "nearly succeeded in a dozen places and the margin of victory--in some places survival--was very very small indeed."(139) Describing Wheeler's report some years later, Clark Clifford recalled that the JCS Chairman had told the President that the Tet Offensive was "a colossal disaster for us," that "we were in real peril," and that Lyndon Johnson consequently was "as worried as I have ever seen him."(140)
In many respects, the primary casualty of the Tet Offensive was President Johnson himself, who just two months later announced he would not stand for reelection. In the intervening weeks of February and March, the enemy's offensive continued in many areas. It included full-scale battles, the largest yet fought by US troops, at Hue and Khe Sanh. At home, the President's political position was weakening under sharpened attacks: Robert Kennedy entered the presidential field against LBJ, and on 12 March the antiwar candidate Eugene McCarthy captured a remarkable 42 percent of the Democratic primary vote in New Hampshire. Then, while many senior administration and military officials kept assuring the country that Tet had been a severe military defeat for the enemy, word leaked out that General Westmoreland was requesting 206,000 more troops.(141) And by this time not only Robert McNamara had come to doubt the President's course, but so also had his successor as Secretary of Defense, Clark Clifford, and a number of other senior defense advisers. Prominent among the latter were DoD's Paul Nitze and Paul Warnke, who in those weeks of February and March made extremely gloomy assessments of the situation. On 1 March, Nitze told Secretary Clifford that the overall national security interests of the United States demanded that we should stop bombing North Vietnam and give General Westmoreland no more than 50,000 additional troops. Warnke has been quoted as later recalling that Tet exposed the fact that "what we had thought was political progress was just so thin as to be illusory," and that the United States could go on winning battles but it would not make any difference because there was no way in which we could "bring about political progress in South Vietnam."(142)
During these weeks Agency officers, often working against impossible deadlines, made substantial inputs to the administration's efforts to devise military and political responses to the Tet-inspired crisis. In addition to providing a steady stream of current intelligence, CIA officers supported executives and strategists with numerous studies and assessments, both requested and volunteered, and participated in senior policy forums. The range of their inputs was wide and at times went considerably beyond strictly intelligence boundaries.
One of the broadest tasks requested of CIA intelligence producers was a list of cosmic questions Vice President Humphrey asked the Agency on 20 February to answer, which added up to assessing whether the United States should abandon its basic strategy in Vietnam. This proved too touchy a question even for George Carver; he advised DCI Helms to duck Humphrey's request. Carver argued that it was not the Agency's business "to tell the Vice President what U.S. strategy in Vietnam is or ought to be," and that he (Carver) was "edgy" about providing a CIA answer to the Vice President's staff "which is not noted for its discretion or good security practices."(143) Carver and other CIA officers did, however, respond to the numerous other difficult questions asked them.
The Department of Defense levied several requests on CIA at this time, including some cosmic questions of its own: what the likely course of events in South Vietnam would be over the rest of 1968; what the enemy's military strategy would be; how North Vietnam, China, the Soviet Union, and "other key countries" might respond to a US troop increase in Vietnam of 200,000; and what terms Hanoi would be willing to accept if the United States halted its bombings of North Vietnam. According to CIA files, Paul Warnke's Office of International Security Affairs dropped such blockbusters on George Carver on 29 February, requesting that CIA reply "if possible, by tomorrow noon."(144) CIA was hit at the very same time by somewhat similar questions from the President's "Clifford Group," a set of senior advisers then headed by the President's personal counselor, Clark Clifford. The group included Maxwell Taylor, Robert McNamara, Paul Nitze, Walt Rostow, Phil Habib, Paul Warnke, and Nicholas Katzenbach. In three studies prepared on 29 February, 1 March, and 2 March, the Agency's answers fed directly into the efforts of these senior counsellors--sharply divided among themselves--to advise President Johnson on what basic post-Tet courses he should adopt. Among the specific questions the Clifford group asked CIA were (1) whether Hanoi had concluded that it must abandon its strategy of protracted conflict and risk all-out offensive efforts; (2) what Hanoi's basic views were toward negotiations; and (3) what the course of political/military events was likely to be in South Vietnam, assuming no change in US policy or force levels. CIA's responses to these requests were, in sum, that the enemy had no serious interest in negotiations except as a tactical means to gain temporary relief and reciprocal advantages; that it was not likely to gamble on an all-out try for military victory; and that the most likely outlook in Vietnam was continued stalemate.(145)
The State Department, too, asked for quick CIA answers to very difficult questions. CIA files record that on 8 March Assistant Secretary William Bundy sent DCI Helms a "no dissemination" memo asking the Agency to estimate the effects of four different bombing options, several possible increased US ground force packages, and possible US mining of North Vietnam's Haiphong harbor. Bundy's memo asked for immediate answers and explained that he wanted the Agency's views alone, without bringing DIA into the exercise.(146)
O/NE answered this request on 13 March, judging (1) that the enemy would probably persist and seek to match any US escalations of the war; (2) that increased US bombings of North Vietnam would have no significant effect on North Vietnamese capabilities or determination, given the assumption that this US course did not include major urban attacks on Hanoi and Haiphong or the mining of the latter; (3) that intensified bombing would not be likely to cause China or the USSR to intervene directly in the war; and (4) that US mining of Haiphong would be the action most likely to cause the Soviets to prompt serious retaliations against the United States, though probably in areas outside Southeast Asia.(147)
The President himself asked CIA for several studies during these weeks. One concerned "Future Communist Military Strategy in Vietnam"; another, "Communist China's Troubles and Prospects."(148) In March President Johnson asked the Agency for an assessment of North Vietnamese strength and infiltration into South Vietnam. George Carver fielded that request, writing Walt Rostow that in the past three or four months there had been a "dramatic increase" in the movement of regular NVA units into the South, that the net increase totaled some 35,000 to 40,000 troops, and that available evidence suggested that even more such deployments were under way.(149)
The most notable of the White House requests was for a governmentwide postmortem covering prior indicators of the Tet Offensive, warnings given, and the military response, which was levied through the President's Foreign Intelligence Advisory Board (PFIAB). The PFIAB Chairman, Gen. Maxwell Taylor, asked DCI Helms for a response by 1 April.(150) Helms tasked his Deputy Director for Intelligence, R. Jack Smith, with heading a combined effort by representatives from CIA, DIA, INR, NSA, and the Joint Chiefs of Staff. After studying the record and sending an interagency examining team to Saigon which, in conjunction with CINCPAC and MACV, interviewed a large number of senior US and South Vietnamese officials, Smith's group submitted an interim report to PFIAB on 3 April and a final report on 7 June.
The principal findings of this postmortem were:
- That there had been evidence (especially communications intelligence) that some attacks might occur during Tet, as a result of which field military commanders took various actions that reduced the impact of the enemy offensive.
- That better intelligence performance was harmed by (a) the lack of high-level clandestine penetrations of the Communist hierarchy; (b) the "blare of background noise"; (c) the great pains the enemy had taken to conceal his intentions; (d) the fact that few US or GVN officials believed the enemy would attack during Tet; and (e) the view of most commanders and intelligence officers "at all levels" that the enemy was incapable of accomplishing the objectives stated in propaganda and in captured documents.
- That "prevailing estimates of attrition, infiltration, and local recruitment, reports of low morale, and a long series of defeats had degraded our image of the enemy."
- And that the urgency felt in Saigon was not fully felt in Washington, where finished intelligence "did not contain the atmosphere of crisis present in Saigon."(151)
Attached to the postmortem were several detailed annexes. These included Saigon Station's (Layton's) warning of 8 December that the outcome of the enemy's winter-spring offensive would in all likelihood determine the future direction of the war. The postmortem, however, made no mention of other key elements that had been added to the intelligence equation by CIA: (1) George Carver's 15 December cover memo to Walt Rostow belittling Layton's warning assessment; (2) the DDI-O/NE study of 8 December which held that North Vietnam and the Viet Cong felt under no compulsion to abandon their basic objectives in the south or the means by which they were seeking to attain them; (3) the DDI's 2 December disagreement with Layton's first alerting assessment in November; and (4) Saigon Station's 19 December warning that something like an all-out push was in the offing.
As observed above, it is difficult not to conclude that, even though this postmortem made some criticisms of the pre-Tet US intelligence performance, it was not complete or fully candid, and that had it been, its tone and judgments would have been more severe. Indeed, even before PFIAB requested its community postmortem, SAVA had made an in-house assessment that was more candid than the postmortem turned out to be. On 15 February SAVA told Director Helms (1) that MACV's method of bookkeeping on enemy O/B had "unfortunately been designed more to maximize the appearance of progress than to give a complete picture of total enemy resources," and thus the "very system has a built-in bias for a persistent underestimate of enemy capabilities"; and (2) that "the US Intelligence Community believed the Communists were wedded to a protracted war strategy--which indeed they were, until they changed it, probably sometime late last summer."(152)
Many of the assessments CIA officers volunteered to policymakers at this time took a dark view of the Vietnam situation and clearly went beyond strictly intelligence matters. One of the first such offerings following Tet, dated 2 February 1968, was a thinkpiece, "Operation Shock," which volunteered the policy recommendation that the GVN be told to shape up, or else. Its authors, William Colby (then Chief of the DDO Far East Division), George Carver, and former Saigon Station Chief John Hart, concluded (1) that the Tet Offensive had "forcefully demonstrated" that the GVN lacked some of the principal attributes of sovereignty because it could not "defend its frontiers without a half million U.S. troops"; (2) that there had been a lack of popular resolution to fight the Viet Cong; (3) that the GVN must make a number of immediate reforms, including the dismissal of key officials; and (4) that if "positive results" were not shown within 100 days, President Thieu should then "be advised that the United States will reserve its position" with regard to the GVN.(153)
For his part, DCI Helms gave the President CIA's collective opinion that the United States lacked an integrated strategic plan for Vietnam and should develop one. The genesis of this voluntary excursion beyond intelligence was a wide-ranging examination conducted by Helms on 11 February with a dozen of the Agency's Vietnam specialists (from SAVA, DDI, O/NE, and DDP). With one notable exception these officers took a markedly pessimistic view of the post-Tet situation. Among their assertions (according to Helms's handwritten notes):
- The United States had no strategic concept or coherent policy in Vietnam.
- The outlook there was for five or six more years of continuing war.
- GVN troops had performed poorly during Tet.
- The United States should avoid further set-piece battles like Khe Sanh.
- It would be pointless for Washington to send more US troops to Vietnam.
- If the GVN proved unable to make its way, the United States "ought to get out."(154)
The sole dissenter in this case was George Carver, who complained to Helms afterward about the discussion's atmosphere of malaise and general disquiet, in which participants expressed opinions that contained "many more adjectives than nouns" and "ranged from despondency to despair." He distanced himself from the group's views, writing Helms that "I am apparently very much out of phase with the current thinking of most of my colleagues."(155) Despite Carver's dissent, and acknowledging that it was not the Agency's function to advise on military or political strategy, Helms nonetheless proposed to President Johnson in late February that "an appropriate task force" be urgently formed "to develop a comprehensive U.S. strategy to guide us during the weeks and months ahead."(156)
During the Johnson administration's post-Tet search for new policies to meet the changed situation, CIA volunteered a number of additional studies, including an assessment of the continuing Communist military threat in the northern regions of South Vietnam;(157) a study (jointly with DIA) which found that available manpower was not a limiting factor on the DRV's ability to continue the war;(158) an R&D proposal designed to inhibit the enemy's use of tunnel warfare;(159) a proposal that CIA-directed paramilitary units in South Vietnam be re-equipped with a new assault rifle being developed by CIA;(160) and a number of O/NE assessments of the enemy. Two years later the Office of National Estimates judged that, whereas prior to Tet its estimates of enemy capabilities and intentions had been overly sanguine, after Tet its judgments were more accurate.(161)
The Agency's most significant contribution to post-Tet policy formulation, however, was doubtless the face-to-face counsel provided to the President and his chief strategists by senior Agency officers, notably the Director and his Special Assistant for Vietnam Affairs, George Carver, but also by William Colby (the Far East operations chief) and R. Jack Smith, (the Deputy Director for Intelligence). Of these, the most active player was Carver, who during February and March met almost daily with senior administration officials and consultants and ultimately came to be credited with directly influencing President Johnson's decision, announced at the end of March, to abandon the effort to win in Vietnam.(162) Carver's influential role of course predated these weeks; as this study makes clear, he had enjoyed remarkable access to senior Vietnam decisionmakers for some time.
Carver had been briefing Secretary McNamara on Vietnam every Monday since 1966, meeting with Walt Rostow almost as frequently, and had been representing or accompanying the DCI at major interagency meetings on Vietnam since the beginning of Helms's tenure as DCI. During the late fall of 1967, Carver and Helms participated in informal discussions of a group of distinguished administration officials and consultants, including Dean Rusk, Nicholas Katzenbach, Robert McNamara, Walt Rostow, McGeorge Bundy, William Bundy, Clark Clifford, Cyrus Vance, Dean Acheson, George Ball, Douglas Dillon, Arthur Dean, Robert Murphy, Omar Bradley, Matthew Ridgway, Maxwell Taylor, Abe Fortas, and occasionally others. Carver typically briefed this group on enemy capabilities and intentions, with the Chairman of the JCS presenting the overall military situation and an Assistant Secretary of State covering the political dimension.
Following the Tet debacle, Helms detailed Carver to join a smaller panel at former Secretary of State Dean Acheson's home for "a frank, full discussion inventorying post-Tet 68 positions, problems and prospects" to help Acheson in advising the President. According to Carver's later account, the participants represented "the entire spectrum of informed official opinion" and "the arguments got pretty brisk." Then in early March, Lyndon Johnson convened a panel termed his "Wise Men" under the chairmanship of Clark Clifford, McNamara's successor as Secretary of Defense (whom Carver had prepped for his Senate confirmation hearings). The President directed them to review the entire Vietnam situation de novo and develop their own policy recommendations. Helms and Carver participated in two of that body's early meetings, on 2 and 3 March 1968.
But it was Carver's later briefing of the "Wise Men" on 25 March and of the President himself on 27 March that has been cited as CIA's most direct and telling contribution to President Johnson's decision to seek negotiations with Hanoi and retire from office, which he announced on 31 March.(163) On the 25th, Carver, Philip Habib from State, and Gen. William E. DePuy briefed the "Wise Men." When DePuy, leading off, asserted that the enemy had suffered a crushing military defeat, he ran into a buzzsaw. Pointing out the numerical contradiction between MACV's understated enemy order of battle on the one hand, and its claims of enemy killed and wounded on the other, in order to demonstrate that there could be few if any NVA/VC troops left, senior US jurist and diplomat Arthur Goldberg asked DePuy, "Who, then, are we fighting?"(164)
Phil Habib gave what Clark Clifford later called "the most important briefing," full of "hard facts and honest opinions." It was a grim presentation of the Tet Offensive's destructive effects on the GVN political-administrative structure, the pacification program, and rural development. Clifford elicited from Habib the opinion that the war could not be won under present circumstances and that the best US course would be to stop the bombing and negotiate.(165)
Carver, speaking from notes, dwelt first on the origins of the shift in Communist strategy represented by the offensive in an analysis very similar to Layton's pre-Tet reasoning. He observed that, although the Communists now controlled much of the countryside and ringed most of the cities and major towns, they had expended a great deal of their human, material, and psychological assets without achieving their maximum objectives of defeating the ARVN, discrediting the GVN, and forcing a US withdrawal. He thought there would be more major attacks and continuing pressure on allied forces; the Communists, meanwhile, would exploit the political and economic advantages of their widened control in the countryside. Noting the major setback the pacification program had suffered, the enemy's unchanged objectives, and the GVN's fragility, he concluded that the next three months were likely to be decisive, with the primary burden of coping with the enemy falling on the GVN; its response would be the prime determinant of the war's eventual outcome.(166)
Following discussions with Clark Clifford and among themselves the next day, the "Wise Men" met with the President, and most of them now advised him that the United States should stop the bombing of North Vietnam and seek negotiations. The President, shocked by the sudden turnaround of men who had previously shared his views and supported his strategy, demanded to know if the briefers had "poisoned the well." The authors of The Pentagon Papers attribute Johnson's shock to the fact that "throughout much of 1967 he had discounted 'negative analyses' of the US strategy by the Central Intelligence Agency and the Pentagon offices of International Security Affairs and Systems Analysis," and had instead "seized upon the 'optimistic reports' from General Westmoreland to counteract what many Pentagon civilians sensed was a growing public disillusionment with the war."(167)
President Johnson's own account is similar: "The net effect [of the daily reports received during March 1968] was positive and this was one reason for my growing confidence in the situation. These reports help account for the surprise I felt every time I encountered otherwise knowledgeable people who seemed to be sunk in gloom." The President's account is also strange: "I think the explanation [for the presumably gloomy assessments given the "Wise Men" on 25 March] was in part that the briefers, in passing on some judgments about Vietnam, especially concerning the situation in rural areas, had used outdated information."(168)
President Johnson then asked that the three briefers meet personally with him and repeat their presentations. Phil Habib, who appears to have been the most gloomy of the three, was out of town (in Canada), but Gen. DePuy and George Carver met with the President on Wednesday afternoon, 27 March, in the Cabinet Room of the White House. Also present were Vice President Humphrey, Walt Rostow, Gen. Creighton Abrams, Gen. Earle Wheeler, and DCI Helms. DePuy and Carver repeated the briefings they had given the "Wise Men" two days before. According to Carver's later account of that meeting, the President paid close attention to his briefing (even though interrupted repeatedly by phone calls), asked Carver a number of questions, and, shaking off the lengthy interruptions "with a grin," told him to complete his full presentation. Then, as Carver later recalled:
If President Johnson was upset or distressed, he certainly did not show it. In fact he started to walk out of the room then turned to walk its full length to where I was standing, pumped my hand, thanked me warmly for my presentation, and made some very flattering and gracious remarks about my overall work and contribution to the national effort.(169)
It is apparent that the tone of Carver's briefings of 25 and 27 March were markedly different from the generally more optimistic views of the situation he had been expounding right up to that time.(170) As we have seen, a few weeks before Tet he had told Rostow that he did not share Saigon's concern that the enemy might be about to launch a nationwide offensive. Not long after the offensive, he had told Helms that he did not share the more pessimistic views of most of his CIA colleagues. Three weeks later, commenting on a bleak characterization of the GVN by Saigon Station, Carver told Helms that "We get the impression, perhaps unfairly, that our colleagues (like the rest of the US Mission) are tired and a trifle defensive in their response to Washington's needles."(171)
Indeed, Carver had long been viewed as a "true believer," one who tended to emphasize the positive in his assessments and go along with the administration's "we are winning" philosophy. Journalist David Halberstam summed it up this way: " . . . in savvy Washington circles it was said that there were two CIAs: a George Carver CIA, which was the CIA at the top, generally optimistic in its reporting to Rostow; and the rest of the CIA, which was far more pessimistic."(172) Helms's biographer Thomas Powers: "Perhaps the clearest expression of the CIA view came from George Carver, who remarked that 'intelligence is not written for history; it's written for an audience'--meaning that it's useless if the audience for whom it's written refuses to read it. If the White House absolutely insists on an enemy OB under 300,000, that is what it is going to get."(173) One of the sharpest such criticisms is voiced by a former NSC staff officer: "Within a few weeks after Carver became head of SAVA he had changed from an independent analyst into a courtier . . . I felt that as long as Carver held the SAVA job, we'd never get the right picture of the war."(174)
Then, suddenly, in his White House briefings on the 25th and the 27th of March, Carver, in the words of his SAVA deputy, George Allen, "uncharacteristically leveled with the 'Wise Men' and President Johnson."(175) Others also thought that Carver's candor had a significant effect on the Wise Men. Former Assistant Secretary of State William Bundy believed that Carver's late-hour shift "undoubtedly had a tremendous influence on his hearers because they knew his usual optimism."(176) No one made the point about Carver's impact more explicitly, however, than Vice President Humphrey. Writing Carver a letter of thanks on 19 April 1968, Humphrey congratulated him for "holding your ground and telling us about the situation as you saw it in Vietnam. It was a brutally frank and forthright analysis. The President's speech of March 31 indicated that your briefings had a profound effect on the course of U.S. policy in Vietnam."(177)
Whatever the degree to which Carver deserted his previous more optimistic assessments, his--and the CIA's--influence upon President Johnson was clearly less than that of many other forces above and beyond the inputs of CIA's intelligence: the shock of the Tet offensive itself; the sharply rising tide of antiwar sentiment among the Congress and the public; the candid, very grim post-Tet assessments given by JCS Chairman Earle Wheeler, Paul Nitze, and Paul Warnke; and the sudden defections of Clark Clifford and most of the other "Wise Men" who had previously backed Johnson's war effort. Nonetheless, to these causes of the President's change of heart must be added the late-March assessments given him by State and CIA officers.
In preparing this episode of CIA performance, the author subscribes to the thesis that the outcome of the Tet Offensive was a military defeat for the enemy: Communist forces suffered crippling losses; contrary to the apparent expectations of their leaders, the South Vietnamese countryside did not rise in their support; and it took the Communists seven more years to gain victory in Vietnam.
Nonetheless, the author shares the preponderant view of historians that the Tet Offensive was an overwhelming political victory for the enemy. The psychological shock of the offensive, which swept away the remaining optimism about the war that the White House and MACV had been at such pains to generate, helped destroy the Johnson administration and was instrumental in causing Presidents Johnson and Nixon to begin the process of negotiating the best US backdown in Vietnam that they could.
Recent events indicate that we should reopen the question of excluding from numerical military order of battle holdings all Communist components other than main and local force . . . We strongly suspect that much of recent urban excitement was caused by personnel drawn from secret self-defense components, perhaps the assault youth, and other elements currently written out of the record by J-2 MACV on the grounds that they "have no military significance."
- George Carver, 13 February 1968 (178)
The Central Intelligence Agency has concluded that the enemy's strength in South Vietnam at the beginning of the winter-spring offensive was significantly greater than U.S. officials thought at the time . . . the two categories excluded by Gen. Westmoreland, the political cadres and the hamlet-level irregulars . . . played a major role in the assault on the cities, military and civilian sources say.
The New York Times, 19 March 1968
Tragically, the Tet Offensive validated three significant judgments CIA officers had previously tried to sell their superiors. The enemy's sudden nationwide offensive made manifest the November-December assessments Saigon Station had sent in, warning that the NVA/VC were about to launch just such attacks. The offensive's scope validated the long-held certainty of most CIA analysts that the enemy's total O/B was substantially greater than MACV's intelligence managers and commanders--and the unfortunate November NIE--had been willing to admit. And the enemy's commitment of substantial numbers of irregular forces to the Tet Offensive brought home the truth of many CIA analysts' earlier arguments that such forces were militarily significant and justifiably part of a total O/B. That these judgments had not been bought by top intelligence and policymaking officers can legitimately be termed an additional casualty of the Tet Offensive.
Following the Tet Offensive, the sharply different O/B assessments CIA and MACV officers had championed resurfaced with new vigor. Less than a month afterward, CIA began deserting the compromise O/B positions it had agreed to with MACV in November. On 21 February a CIA Intelligence Memorandum stated that there was now sufficient evidence to support a judgment that in his offensive the enemy had committed numerous irregular forces, of various types.(179) Two days later, in response to a query from the White House, DCI Helms reported that available evidence did not support the US military's claim of an enemy decimated by Tet.(180) On 1 March OCI and OER sharply questioned MACV's continuing claims that the enemy had suffered a very high percentage of losses: "the dilemma with respect to the casualties arises when the reported enemy KIA (38,600) is considered against the total offensive force estimated [by MACV] to have been involved (77,000). Taken at face value, this means that approximately one half of the attacking force was killed in the offensive and its aftermath."(181) This OCI-OER study concluded that these figures were exceedingly difficult to accept, given the continuing current high level of enemy activity throughout the country.(182) An OER officer shortly thereafter ridiculed MACV's claims, pointing out that if the 1.5 to 1 ratio of wounded to killed in action were applied, the resultant casualty total exceeded the forces committed.(183)
Nevertheless, US military officers clung to their previous O/B estimates despite the contradiction created with their claims of enemy losses. Negotiations over the enemy strength estimates between CIA and the military, reminiscent of the previous year's work on the NIE, dragged on for weeks but failed to bridge the gap. In April DCI Helms had to admit to Gen. Maxwell Taylor in the White House that he had "become increasingly concerned that the strength of enemy forces was underestimated and that there were serious errors in the way the forces were characterized and in the way attrition was handled."(184)
Finally, in May 1968, Helms, who six months before had accepted the contested MACV O/B figures rather than send a split estimate to the President, told the White House that the two sides had thus far been unable to reach agreement: MACV, DIA, and CINCPAC still held enemy strength in South Vietnam to be between 280,000 and 330,000, he reported, whereas CIA now believed the figure to be somewhere between 450,000 and 600,000. Helms added that, of those totals, CIA accepted some 90,000 to 140,000 enemy irregulars, whereas MACV and CINCPAC still maintained that such forces could not and should not be quantified.(185) And there the matter rested.
Clearly, MACV's O/B estimates remained much too low. It is also clear that MACV's reluctance to accept higher enemy O/B figures all along had a corrupting effect on the conduct of US intelligence analysis and presentation. Perhaps the premier exemplar of such flaws was one of MACV's chief intelligence officers, later DIA Director Lt. Gen. Daniel O. Graham, who testified in 1975 to Congress that MACV had not been surprised by Tet; that all the previous MACV, DIA, and CIA estimates of the enemy's O/B had been "too high in terms of total VC combat strength available; and that the worst estimate around was Mr. [Sam] Adams's 600,000."(186) As noted earlier, in 1967 then Col. Graham had confided to George Allen that "of course" he did not believe MACV's lower O/B figures, but "it's the command position and I'm sticking with it."(187)
- Senior intelligence and policymaking officers and military leaders erred on two principal scores: for having let concern for possible political embarrassment derail objective assessments of the enemy order of battle, and for ignoring NSA's alerts and Saigon Station's warnings that did not accord with their previous evaluations of probable enemy strategy.
- The least astute performance, clearly, was MACV's. Its O/B positions misled planners and policymakers, distorted intelligence reporting and analysis, contributed directly to the psychological shock the Tet Offensive inflicted on the public and the White House, and thus caused serious damage to the national interest. One military element did work well, at the last moment, when General Westmoreland responded to field communications intelligence alerts and approved certain changes in US troop dispositions that limited the scope of the enemy's depredations in Saigon and General Weyand's III Corps sector.
- The best Tet Offensive alerts were those provided by NSA and by General Weyand's units cited above. Communications intelligence often afforded a better reading of the enemy's strength and intentions (and was better heeded by command elements) than did agent reports, prisoner interrogations, captured documents, or the analytic conclusions derived from them. But in Washington the SIGINT alerts apparently made little impression on senior intelligence officers and policymakers.
- CIA's performance in the O/B and Tet episodes was mixed--better than MACV's in the former, less perceptive than NSA's in the latter. The best CIA performances were by a few working-level officers who tried to sell their judgments that the enemy had thousands of irregular forces that were militarily significant, and that the enemy was about to launch a major nationwide offensive. But the Agency's most senior Vietnam intelligence officers gave in to MACV's stonewall defense of its O/B estimates, enshrined that position in a definitive NIE, downplayed Saigon Station officers' warnings, and so left administration officials unprepared for the shattering of their illusions of progress in Vietnam. To the intelligence managers' credit, after Tet they did level with the President on the facts of the situation, abandoned their earlier O/B compromises with MACV, and acknowledged to the White House that the Intelligence Community's assessments of the enemy's numbers, capabilities, and intentions had been in error.
(1) Abrams cable to Gen. Earle G. Wheeler (Chairman, Joint Chiefs of Staff). Attachment to letter, Congressman Paul N. McCloskey (R-CA) to DCI William Colby, 6 June 1975. CIA Files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 9.
(3) See, for example, John M. Newman, JFK and Vietnam (New York: Warner Books, l992). Newman, making use of Kennedy Library materials, discusses at length the battles over O/B that preceded President Kennedy's decision in late 1961 to expand the level and nature of US military support in South Vietnam. See also James J. Wirtz, The Tet Offensive: Intelligence Failure in War (Cornell University Press, 1991), which draws on original cables and documents declassified for the Westmoreland v. CBS trial.
(4) "MACV's present estimates on the strength of Viet Cong irregulars are derived from estimates provided by GVN province chiefs." Report of the Honolulu Intelligence Conference to Standardize Methods for Developing and Presenting Statistics on Order of Battle, Infiltration Trends and Estimates--6-12 February 1967, p. 10. (C). CIA files, Job No. 78T02095R, Box 1, O/DDI, Folder 2. Former CORDS chief Robert Komer's view is that as of 1967, "MACV's intelligence for the most part was terrible." Komer, to author, 21 May 1990. According to George Allen, MACV had lowered the number of estimated enemy irregulars in 1965, "although no one had really done much research in depth on the question." Allen, to author, 11 April 1990.
(5) Discussed in some detail in CIA, Response to the Questions on Vietnam Posed in National Security Study Memorandum Number 1, 7 February 1969. (S/Sensitive). CIA files, Job No. 78-906, DDO-ISS IP, Box 2, Folder: "Vietnam-National Security Council." A March 1968 MACV cable states that the only new element in the O/B picture at the time is "the correct assertion that MACV OB figures do not always reflect all-source intelligence." MACV 05301 (exact date illegible). (S/Eyes Only). CIA files, Job No. 78T02095R, O/DDI, Box 4, Folder l3.
(7) Allen, to author, 11 April 1990. As of 1964, Allen was a CIA/OCI (Office of Current Intelligence) officer stationed in Saigon. Earlier an officer of the Army's G-2 and DIA, Allen subsequently became deputy chief of SAVA.
(8) Palmer, "US Intelligence and Vietnam," p.47. At the time he wrote this study, General Palmer was a member of the DCI's Senior Review Panel. He had earlier been General Westmoreland's deputy in Vietnam, and then Army Vice Chief of Staff. General Palmer is also the author of a highly regarded unclassified history, The 25-Year War: America's Military Role in Vietnam (New York: Simon & Shuster, 1984).
(9) Richard Helms, Memorandum for Robert S. McNamara, "Servicing of Vietnam Assessment Requests," 27 April 1967, (S/Sensitive/Eyes Only). CIA files, Job No. 80B01285A, DCI, Box 11, "DCI Helms," Folder 3. According to Hans Heymann, who was one of the compilers of The Pentagon Papers and who had wide access to McNamara's private papers, by l967 the Secretary had become so scornful of military reporting from Vietnam that he began scribbling "I don't believe it"-type marginalia on some of those reports, in one instance writing, "This is a lot of crap." Heymann, to author, 26 September 1991. Heymann had earlier been a RAND Corporation officer concerned with Vietnam questions, and in the early 1980s became the National Intelligence Council's National Intelligence Officer (NIO) for Economics.
(11) Carver, Memorandum for Lt. Col. Robert M. Montague, Military Assistant to Mr. Komer, "Calculation of Viet Cong Irregular Strength," 28 September 1966, (S). CIA files, Job No. 80B01721R, O/D/NFAC, "Substantive Policy Files, DDI Vietnam Files," Folder 7.
(14) Adams, "Vietnam Cover-up: Playing War with Numbers," Harpers, May 1975, passim. The author of this History Staff study knew Adams, subsequent to the events discussed in this study. Adams was a prodigious researcher whose findings were groundbreaking and generally more accurate than those prepared by other officers; he repeatedly hurt his case, however, by overstatement and self-defeating conduct. Numerous observers have discussed the Sam Adams phenomenon. See, for example, a rejoinder letter to the editor by James C. Graham (a former member of the Board of National Estimates), Harpers, June 1975, pp.15-16; Eleanor Randolph, The Washington Post, 8 November and 3 December 1984, 10 January, 30 January, 13 February, and 18 February 1985; Bob Brewin and Sydney Shaw, Vietnam on Trial: Westmoreland vs. CBS (New York: Atheneum, 1987); T. L. Cubbage II, "Westmoreland vs. CBS: Was Intelligence Corrupted by Policy Demands?", Intelligence and National Security, July 1988, pp.118-180; Thomas Powers, The Man Who Kept the Secrets: Richard Helms and the CIA (New York: Knopf, 1987); Renata Adler, Reckless Disregard: Westmoreland vs.CBS et al.: (New York: Knopf, 1986); George Allen, "The Indochina Wars;" John Ranelagh, The Agency: The Rise and Decline of the CIA (New York: Simon & Shuster, 1986), pp.457-465; and Sam Adams, War of Numbers, (South Royalton, Vermont: Steerforth Press, 1994) passim.
(15) Carver, Memorandum for the DDI, "Revisiting the Viet Cong Order of Battle," 11 January 1967, (S). CIA files, Job No. 80B01721R, O/D/NFAC, Box 1, "Substantive Policy Files, DDI Vietnam Files," Folder 7.
(16) Carver, Memorandum for The Honorable Robert W. Komer, "Vietcong Desertions," 3 April 1967, (S). CIA files, Job No. 80R01580R, DCI/ER Subject Files, Box 15, Folder 8. The next month, May 1967, President Johnson sent Komer (a former O/NE officer) to Vietnam with ambassadorial rank to head the CORDS program there. As we will see, a few months after taking up his CORDS position, Komer played a central role in influencing the outcome of the Saigon O/B showdown between MACV and CIA.
(19) Allen, Memorandum for William J. Jorden, Senior NSC Staff Member, "Comments on Giap's Speech on the Role of the Militia," 20 April 1967, (S). CIA files, Job No. 80R1720R, O/D/NFAC, Box 5, "GAC Files (SAVA-NIO), Folder 6.
(21) According to George Fowler, DIA's representative to the coordination meetings on SNIE 14.3-67. Fowler, to author (in Taipei, Taiwan), September 1967. Fowler explained that his DIA office was caught in a no-win crossfire at that time between MACV (which was putting pressure on DIA not to budge from MACV's official O/B figures) and CIA (which was berating DIA for ignoring the research and analysis behind the Agency's higher O/B estimates).
(23) "Input to the DCI Audit Project Chaired by Under Secretary of State Katzenbach," 23 May 1967, (S). CIA files, Job No. 85T00268R, DCI/O/DCI, Box 7, "NSC Papers," Folder 7. Former DDI chief R. Jack Smith later emphasized the emotional nature of this unique commission: "Never before had a civilian intelligence organization challenged an army in the field about its orders of battle. . . . But here were a bunch of civilians telling not only the Pentagon but also the forces in the field that the number they were facing was higher. That created a very difficult position: it was their war. They were the ones getting killed. There was a lot of emotion involved in that." Smith, to interviewer John Ranelagh, as cited in the latter's CIA: A History (London: BBC Books, 1992), p. 127.
(24) According to Thomas Powers, The Man Who Kept the Secrets, pp. 186-187. Also, on a memo Carver sent Helms on 1 May 1967, "Order of Battle Note" (S), Helms wrote: "Good--But let's try to simplify these suggestions, tighten up the language, and move to get agreement from the military." CIA files, Job No. 80T01719R, O/D/NFAC, Box 3, Folder 7.
(25) SAIG 1826, to Helms (only), (S). Attachment to Carver, Memorandum for the DCI, "1967 Order of Battle Cables," 26 November 1975, (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 9.
(26) SAIG 1826, to Helms (only), as above. Many CIA officers found General Davidson more difficult to work with than they had McChristian. Even after the Saigon conference and apparent agreement had been reached on how the upcoming SNIE should present its O/B figures, Carver described what he termed "MACV J-2 childishness": CIA's files show that its Saigon Station officers were finding it necessary to get prior written permission from MACV Headquarters in order to visit working-level J-2 officers, "as if CIA officers were acting as agents of potentially hostile foreign power." (Carver drafted) DIR 49100 to Saigon, 3 November 1967, (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 5, "GAC Files (SAVA-NIO)," Folder 8. Former CORDS chief Komer, however, holds that Davidson was "a great improvement over McChristian--much more able, and much more willing to defend his views when questioned by Westmoreland." Komer, to author, 21 May 1990.
(27) SAIG 1926, 12 September, (S). CIA file as SAIG 1826, above. The author of the present study has for years been a close friend and associate of both Bob Komer and CIA's Vietnam analysts of the period under review. In his 11 September 1967 meeting with Carver, Komer, characteristically given to brash overstatement, seriously understated the quality of CIA's analytic talent. Carver, George Allen, Bob Layton (O/NE), and other CIA officers had not only been following Vietnam affairs in depth for years, but also had racked up high batting averages in their assessments and estimates. By contrast, in addition to being pressured by their seniors not to admit the existence of more than 300,000 enemy forces, total, in South Vietnam, MACV's intelligence officers were in the main much less experienced and acute.
(28) Allen, "The Indochina Wars," pp. 317-318. According to Allen, Graham made this statement in 1984 at the time of the Westmoreland v. CBS trial. In late January 1968 Graham told the author of this study, then visiting Saigon, that things were fairly quiet there. The next day the enemy launched its Tet Offensive. Graham eventually became a Lieutenant General, the Director of DIA, and, in retirement, the founder of "High Frontier," a foundation championing a US antiballistic missile defense system.
(29) Bob Layton (of CIA's Saigon Station, present on that occasion), to the author, 18 February 1992. Also Adams's highly personal later account, in which he quoted Hyland as saying that the "political climate" would not permit MACV to accept higher O/B figures and that he, Adams, was "living in a dream world." Adams, War of Numbers p.118.
(31) SAIG 0826 (IN 35169), 19 August 1967, (S). CIA files, Job No. 80B01721R, O/D/NFAC, Box 1, "Substantive Policy Files, DDI Vietnam Files," Folder 6. Also, some analysts in CIA had been hesitant, at least initially, to accept the higher O/B figures championed by Carver, Allen, and Adams. In June 1966, for example, the North Vietnam Branch of OCI reported that its analysts "have concluded that current MACV estimates on infiltration into South Vietnam and on Communist order of battle in the South can be used with confidence. ORR analysts concur in this view." David Siegel, Memorandum for the Director of OCI, "The Accuracy of MACV's Reporting on Infiltration and Order of Battle," 27 June 1966, (S, compartmented). CIA files, Job No. 78T02095R, O/DDI, Box 4, Folder 17.
(36) Abrams, cable to Wheeler, 20 August 1967. Attachment to letter, Congressman Paul N. McCloskey (R-CA) to DCI Colby, 6 June 1975. CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 9.
(43) Powers, The Man Who Kept the Secrets, p. 188. A sidelight to the O/B quarrel: In June 1967 Carver had told Helms that the White House Situation Room had recently been "somewhat shaken up looking around for a staffer named Funaro." Carver, Memorandum for the Director, "Komer Correspondence," 27 June 1967, (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 5, "GAC Files (SAVA-NIO)," Folder 7.
(46) Smith manuscript, "Richard Helms and Intelligence Production," August 1983, (S), pp.8-9. On file in CIA History Staff. At the time of the Saigon conference, Smith was CIA's Deputy Director for Intelligence. His then staff aide, Richard Kovar, has the same recollection of the Carver-Helms cables. Kovar to author, 27 November 1995.
(48) Carver, SAIG 1988 (IN 51159), (S). Attachment to Carver, Memorandum for the Director, "1967 Order of Battle Cables," 28 November 1975, (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 9.
(50) Helms, to author, 11 May 1992. Available CIA files do not contain any communication from Helms to Carver either directing him to compromise, or agreeing to a Carver proposal that he should back off from the O/B estimates he had previously championed so strongly.
(51) Powers, The Man Who Kept the Secrets, p. 189. Richard Lehman, at the time deputy chief of the Office of Current Intelligence, recalls that he was "of the impression" that DCI Helms had discussed this O/B question with President Johnson and that the President was aware that the official O/B estimate might be too low. Lehman, to author, 14 March 1995.
(60) Carver, Memorandum for The Honorable Philip Goulding, "Proposed MACV Press Briefing on Enemy Order of Battle," l3 October 1967, (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 3, "GAC Files," Folder 4.
(61) Walsh, Memorandum for Carver, "MACV Press Briefing on Enemy Order of Battle," 11 October 1967, (S/NF). (Emphasis in the original). CIA files, Job No. 80B01721R, O/D/NFAC, Box 1, "Substantive Policy Files, DDI Vietnam Files," Folder 6.
(63) From MACV J-2's Monthly Order of Battle Summary for October 1967, as cited in SAIG 5297 (IN 04823), 3 December 1967, (S). CIA files, Job No. 80B01721R, O/D/NFAC, Box 2, "Substantive Policy Files, DDI Vietnam Files," Folder 5.
(64) SAIG 4140 (IN 85887), 3 November 1967, (S). Attachment to Carver Memorandum for the Director, "More Vietnam Numbers Problems," 3 November 1967, (S). CIA Files, Job No. 80R01720R, O/D/NFAC, Box 2, Folder 11.
(67) The author's judgment that paragraphs of prose relegated to the back of a National Intelligence Estimate cannot compete with hard numbers and crisp Conclusions up front in the Estimate, especially in their impact upon senior policymaking readers, is based on decades of personal first-hand experience in writing, managing, and marketing National Estimates. Gen. Bruce Palmer is similarly critical of SNIE 14.3-67 on these scores: "Although, according to Carver's argument, the total North Vietnamese structure approaching the half million mark in strength is described in the text, the fact remains that the summary of the estimate (all that would probably be read by most busy senior policymakers) cited a total of 188,000-208,000." Palmer, "US Intelligence and Vietnam," p. 51. George Allen: "In the end Carver and the SNIE ended up with no figure at all for the irregulars and only a general prose statement that there might be lots of these guys; to the reader this could not help but look like a complete CIA back off, an admission that their much higher VC irregular force figure had been wrong all along." Allen, to author, 11 April 1990.
(68) Adams, Memorandum for the Record, "Comments on the Current Drafts of the Introductory Note and Text of National Intelligence Estimate 14.3-67," 7 November 1967, (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 9.
(69) Quoted in Powers, The Man Who Kept the Secrets, p.189. The CIA retreat on O/B which the SNIE represented was almost certainly influenced in part by the fact that the Board of National Estimates was now headed by Abbot Smith. The author of this study was for years a colleague and close friend of the late Dr. Smith. He was a brilliant and perceptive officer, a man of great integrity. He was a scholar, however, not a bureaucratic scrapper. Moreover, his philosophy of estimating was that the most important service an NIE or SNIE could perform was to present the greatest degree of agreed judgments, not sharp alternative views.
(70) From Helms's statement to the President: "The new estimate is sensitive and potentially controversial primarily because the new strength figures are at variance with our former holdings. . . . I have considered not issuing this Estimate and after considerable consultation, believe this would be a mistake. . . . In short, the charge of bad faith or unwillingness to face the facts would be more generally damaging than the issuance of this document which can stand on its own feet." Helms, Memorandum for the President, 14 November 1967, (TS). CIA files, Job No. 80B01285A, DCI, Box 11, "DCI Helms," Folder 4.
(76) William E. Colby, Memorandum for the Record, "VIC Meeting, 1 March 1967," 3 March 1967 (S). CIA (DO) files, Job No. 78-646, Box 1, Folder 3, "Vietnam Interagency Committee (Komer Meetings), April 1966 - May 1967." (Emphasis added).
(78) CIA memorandum (unsigned), commenting on General Wheeler's closed-door testimony to the Senate Armed Services Committee, 16 August 1967 (TS). CIA files, Job. No. 787S02149R, O/DDI, Box 3, Folder 20, "Stennis Committee (The McNamara Testimony on Air War in North Vietnam)."
(83) The most significant such critic was now Robert McNamara, formerly a staunch supporter and key architect of the Johnson war effort, who was about to be replaced as Secretary of Defense. Among other senior critics at this time were Vice President Humphrey, George Ball, Clark Clifford, McGeorge Bundy, Robert Kennedy, Eugene McCarthy, some senior DoD civilians, and a scattering of doubters in the Congress.
(84) Palmer, Memorandum for DD/NFAC, "A Look at US Intelligence Assessments re SE Asia, 1965-1975," 5 October 1975, (S). Copy on file in History Staff. In this document General Palmer recommended that the CIA should "commission a special historical effort that would describe and objectively evaluate the Agency's performance (analytical side, not operational) during the Vietnam war." In his view, such an undertaking would "not only enhance the reputation of the Agency but also boost the pride and esprit of CIA personnel." The desired result, he wrote, "would be a fairly short publication, well documented, and if at all possible, unclassified." (Emphasis in the original).
(86) As we have seen, on 21 April 1967 Secretary McNamara had asked CIA to give him periodic assessments of the effectiveness of US air operations, as well as of the enemy's O/B and the progress of pacification. By mid-1967 the Agency and DIA were jointly preparing monthly bombing assessments. A few weeks before Tet, one of the customers of these reports, the JASON division of the Office of the Secretary of Defense, issued its own, similarly pessimistic assessment of the bombing programs, an assessment the authors of The Pentagon Papers later termed "probably the most categorical rejection of bombing as a tool of our policy in Southeast Asia to be made before or since by an official or semiofficial group." (Gravel ed.), Vol. IV, p. 222.
(89) Katzenbach, letter to Helms, 16 October 1967, (TS/Sensitive). DDI prepared CIA's response: Memorandum, "The South Vietnamese View of Negotiations: Problems and Prospects," 27 October 1967, (TS/Sensitive). CIA files, Job No. 80B01721R, O/D/NFAC, Box 2, "Substantive Policy Files, DDI Vietnam Files," Folder 5.
(90) Carver, Memorandum for R. J. Smith, et al., "Request from Secretary McNamara," 21 November 1967, (S), attachment to Phil G. Goulding (Assistant Secretary of Defense for Public Affairs), Memorandum for George Carver, 21 November 1967, (S). CIA files, Job No. 80R01580R, DCI/ER Subject Files, Box 15, Folder 7. CIA's response to McNamara's 67 specific questions took the form of an Intelligence Report, "Questions and Answers Relating to Vietnam," 8 December 1967, (TS/Compartmented/NF). CIA files, Job No. 82S00205R, DDI/O/DDI, Box 3, Folder 19.
(92) Allen later wrote that he was outraged by Rostow's "dishonesty" in misleading the President, and that the exercise "was an element of the public opinion campaign which was designed to peak with the visits to Washington in November  of Ambassador Bunker and General Westmoreland, who were to pull out all stops . . . in beating the drum for the 'light at the end of the tunnel.'" Allen, "The Indochina Wars," pp. 302-304.
(94) On 15 November DCI Helms replied that the CIA could find no evidence linking the peace movement to foreign support. See Helms cover memorandum (S/Eyes Only, later declassified and given to Congressional investigative groups in 1975). CIA files, Job No. 80B01285A, DCI, Box 11, "DCI Helms," Folder 3.
(95) Allen, p. 301. The same went for newsmen. Following the Tet Offensive, when asked by John Scali whether there had been an intelligence failure, Secretary of State Rusk is reported as having exploded, "There gets to be a point when the question is whose side are you on." Marvin Kalb and Elie Abel, Roots of Involvement: The US in Asia--1784-1971 (New York: W. W. Norton, 1971), pp. 206-207.
(96) NIE 53-66, "Problems of Political Development in South Vietnam Over the Next Year or So," 15 December 1966, (S/Controlled Dissem; declassified in November 1975). Copy on file with CIA's History Staff.
(97) Carver, letter to Rostow, 2 March 1967, (S/Sensitive), attachment to Carver, Memorandum for the Director, "Rostow-requested Memorandum, 3 March 1967, (S/Sensitive). CIA files, Job No. 80R01580R, DCI/ER Subject Files, Box 15, Folder 8.
(102) Colby, Memorandum, "Review of the Activities of the CIA's Vietnam Station," 25 July 1967 (S/Sensitive), attachment to Helms, Memorandum for the President, "Transmittal of Vietnam Report," 27 July 1967, (S/Sensitive). CIA files, Job No. 80R01720R, O/D/NFAC, Box 5, "GAC Files (SAVA - NIO)," Folder 7.
(107) As cited in Carver, Memorandum for the Honorable Walt W. Rostow, "31 January Telephone Conversation with Saigon Station," 31 January 1968, (S/Sensitive). CIA files, Job No. 80R01720R, O/D/NFAC, Box 5, "GAC Files (SAVA-NIO)," Folder 9.
(110) As cited in Lt. Col. Dave Richard Palmer, Summons of the Trumpet: U.S.-Vietnam in Perspective (San Francisco: Presidio Press, 1978), pp.178-179. See also Clark Clifford, The New Yorker, 13 May 1991, p. 48.
(111) COMUSMACV cable to CINCPAC, info to General Wheeler and Ambassador Bunker, 25 January 1968. Attachment to "Intelligence Warning of the Tet Offensive in South Vietnam," April 1968. (Initially classified, this postmortem was declassified and released to the House of Representatives' Pike Committee in 1975.) CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 8.
(113) General Weyand, to author, 17 April 1991. The officer commanding General Weyand's communications battalion at the time, former Lt. Col. Norman Campbell, supports Weyand's accounts. Campbell, to author, 18 May 1992. For additional accounts of General Weyand's prescience, see Dave Richard Palmer, Summons of the Trumpet, p. 184; Don Oberdorfer, Tet (New York: Da Capo Press, 1984), pp. 137-141; and Neil Sheehan, A Bright Shining Lie: John Paul Vann and America in Vietnam, pp. 701-709. As of January 1968, Vann was attached to General Weyand's command. Weyand, who had previously held numerous military intelligence assignments, based his certainty of a coming major offensive largely on traffic analysis and radio direction finding; former Lt. Col. Campbell claims that their communications units were also reading some of the enemy's traffic. In March 1968, USMC Lt. Gen. Robert Cushman, then I Corps commander (and later DDCI), told visiting OCI officer Richard Lehman that the Marines in I Corps had had "ample forewarning" of the Tet Offensive, even though the enemy's specific targets had remained unknown. SAIG 0191 (IN 733895), 20 March 1968, (S). CIA files, Job No. 80B01721R, O/D/NFAC, Box 2, "Substantive Policy Files, DDI Vietnam Files," Folder 1.
(118) Intelligence Warning of the Tet Offensive in South Vietnam, Section VII-1, "Indications Received in CIA, 15-30 January 1968," (S/later declassified and provided the Pike Committee in 1975), p.1. CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 8, "Intelligence Warning of the Tet Offensive in So. Vietnam." The CIA's DDI, R. Jack Smith, chaired this Intelligence Community postmortem.
(119) Handwritten comment by the Acting Deputy Chief, Vietnam Operations branch of EA Division in a Memorandum for Deputy Chief, East Asia Division, "Vietnam Reporting Prior to Tet 1968 Offensive," 18 September 1975, (S). CIA files, Job No. 80-00088A, Box 1, DO/EA, Folder 8, "Reporting on the Tet Offensive."
(123) Intelligence Warning of the Tet Offensive in South Vietnam, pp. 5-6. This postmortem conclusion is less critical of the US intelligence performance than seems warranted, at least in hindsight. Its muted tone doubtless can be explained by an understandable reluctance to dramatize the shortcomings in the Intelligence Community's own record or to probe deeply the intelligence operations of a military command that was still fighting its way out of the consequences of its errors. When the postmortem evaluators formed a team to examine the performance of CIA, State, and military officers in the field, DCI Helms told the examiners that they should not "rock the boat"; they could be "critical but not inflammatory" in their report. The recollection of the team's chairman, Richard Lehman, to the author, 14 March 1995.
(124) A prime reflection of surprise is this incident related by CIA's George Allen. At CIA Headquarters he was in the process of giving a Vietnam briefing to State's Phil Habib and Nicholas Katzenbach when a CIA officer rushed in to tell them that the Embassy in Saigon was under attack. "Habib chuckled, suggesting that I have my troops knock off their horsing around . . . The officer earnestly persisted, exclaiming in his best 'Pearl Harbor' tones, 'This is no drill, sir; the wire tickers report that the embassy is under attack and the VC have penetrated the compound' . . . Habib's jaw fell, and he turned ashen gray; he realized immediately the significance of this development; that the wind had been taken out of the administration's sails, the 'light at the end of the tunnel' had been turned off, the administration's policies had been derailed from 'the right track.'" Allen, "The Indochina Wars," pp. 323-324.
(133) CIA Memorandum, "A Review of the Situation in Vietnam," 8 December 1967, (TS/Compartmented), prepared jointly by the Office of Current Intelligence, the Office of Economic Research, O/NE, and SAVA. CIA files, Job No. 78T02095R, O/DDI, Box 1, Folder 1.
(134) Johnson, The Vantage Point: Perspectives of the Presidency, 1963-1969, pp. 371-372. According to Robert Johnson, who had been a colleague of Walt Rostow's in State's Policy Planning Staff, Rostow wrote much of the former President's autobiography. Robert Johnson, to author, 13 June 1992.
(138) DDI Intelligence Memorandum, "Communist Units Participating in Attacks During the Tet Offensive: 30 January Through 13 February 1968," (S/NF). CIA files, Job No. 80R01720R, O/D/NFAC, Box 2, Folder 14. The author of the present History Staff study was in Saigon during the first days of the Tet Offensive and witnessed the Embassy compound minutes after it was secured.
(139) Wheeler, Memorandum for the President, "Military Situation and Requirements in South Vietnam," 27 February 1968, (TS). CIA files, Job No. 80R01720R, O/D/NFAC, Box 3, Folder 2, "March 1968 Review Commissioned by Pres. Johnson." In the body of his report, attached to his cover letter to the President (from which the quotations above are taken), General Wheeler described the Tet Offensive as "a very near thing." Report of Chairman, Joint Chiefs of Staff, Gen. Earle G. Wheeler, on the Situation in Vietnam, 27 February 1968, as cited in Gareth Porter (ed.), Vietnam: The Definitive Documentation of Human Decisions, Vol. II, p.502. Wheeler's description, "it was a near thing," is also cited in the Gravel edition of The Pentagon Papers, Vol. IV, p. 547; and by former MACV J-2 Gen. Phillip Davidson, who describes Wheeler's report as "the blackest possible evaluation of the battlefield situation in Vietnam." Davidson, Vietnam at War, p. 504.
(141) The news story was broken by The New York Times on 10 March, presumably leaked by someone who opposed General Westmoreland's request. The manner in which this news item hit the country --with Westmoreland calling for thousands more troops just after having supposedly inflicted a major defeat on the enemy--obscured the particular circumstances of Westmoreland's message. The 206,000 additional troops were a contingency requirement in the event the United States decided to change its basic ground force strategy in Vietnam from defense to offense.
(143) Carver, Memorandum for the Director, "Request from [Vice President Humphrey's aide] George Carroll," 21 February 1968 (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 5, "GAC Files (SAVA-NIO), " Folder 9.
(145) Pentagon Papers (DoD ed.), Book 6-IV-C-7-b, "The Air War in Vietnam, Vol. II, pp. 149-153. Also, O/NE Memorandum, "Questions Concerning the Situation in Vietnam," 1 March 1968, (S). CIA files, Job No. 80R01270R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 6.
(148) DCI Helms: "These two studies were requested by Walt Rostow on behalf of the President. I thought you would be interested in seeing them." Helms, cover note to Secretary of State Rusk, 23 February 1968, (TS). CIA files, Job No. 80B01285A, DCI, Box 11, 'DCI Helms," Folder 5.
(150) Taylor, Letter to Helms, 23 February 1968, (TS/Compartmented). CIA files, Job No. 80R01580R, DCI/ER, Subject Files, Box 16, Folder 6. Taylor: "President Johnson directed the President's Foreign Intelligence Advisory Board of which I was chairman to investigate the charges that American forces had been surprised by the Tet offensive." Taylor, Swords and Plowshares (New York: Norton, 1972), p. 383.
(152) SAVA Memorandum for the Director, "PFIAB Request," 15 February 1968, (S). CIA files, Job No. 80B01710R, O/D/NFAC, Box 1, "Substantive Policy Files, DDI Vietnam Files," Folder 5. (Emphasis added).
(153) Blind memorandum, "Vietnam Operation SHOCK," 2 February 1968, (S), attachment to Helms, Memorandum for Undersecretary of State Nicholas Katzenbach, Deputy Secretary of Defense Paul Nitze, Chairman of the Joint Chiefs Gen. Earle Wheeler, and Special Assistant to the President Walt Rostow, "An Immediate Program for Vietnam," 2 February 1968, (S). CIA files, Job No. 80B01285A, DCI, Box 11, Folder 5. Helms distributed "Operation Shock" widely to key policymakers, and it became the basis for an advisory cable later addressed to Ambassador Bunker by the State Department. CIA's Saigon Station did not fully share these pessimistic views of the GVN. In March the Station observed to Headquarters that the GVN had passed a great trial by fire and might well emerge with a stronger temper, and it emphasized what it felt were some of the positive factors present. SAIG 0357 (IN 86842), 24 March 1968, (S). CIA files, Job No. 80B01721R, O/D/NFAC, Box 2, "Substantive Policy Files, DDI Vietnam Files," Folder 1.
(156) (Carver-drafted) Helms, Memorandum for the President, "Vietnam Strategy," 28 February 1968, (S/Sensitive). CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 6, "Alternative Strategies in Vietnam."
(159) Carver, Memorandum for Maj. Gen. William E. DePuy, "Tunnel Denial by Means of an Electrical Field Emplaced in the Soil," 21 March 1968, (S). CIA files, Job No. 80R01580R, DCI/ER Subject Files, Box 15, Folder 6.
(161) O/NE Staff Memorandum No. 27-70, "National Intelligence Estimates on the Vietnam War Since October 1964," 11 August 1970, (S/Sensitive/Internal ONE Distribution Only). Copy on file in CIA History Staff.
(162) Carver himself dated the roots of the President's decision to the product of a "quiet quartet" commissioned by Defense Secretary McNamara, back in September 1967, to "canvass possible alternative strategies in Vietnam." The Assistant Secretary of Defense for International Security Affairs, Paul Warnke, and his deputy, Morton Halperin, worked through the fall of 1967 with Carver and with Helms's Deputy Director of Current Intelligence, Richard Lehman, to produce "An Alternative Fifteen-Month Program for Vietnam." Carver later recalled that "the paper basically argued for a curtailment, if not suspension, of the bombing plus the opening of negotiations"; "the final effort . . . was much more Warnke and Halperin's paper than it was Lehman's and mine." Although Carver and Lehman thought the paper had "quietly died" after its delivery to McNamara, when Carver re-read his copy in 1970 he found that "its ultimate function becomes clearer. . . . It contains in well developed outline almost all the arguments Warnke--and Halperin--successfully urged on [Clark] Clifford during March 1968." Found in Carver, Memorandum for the Director, "The Bombing Decisions--31 March and 1 November 1968," 31 March 1970, (S/Sensitive). CIA files, Job No. 80R01720R, O/D/NFAC, Box 3, "GAC Files (SAVA-NIO)," Folder 1.
(163) Numerous observers attest to the impact these briefings produced on their hearers. See Lyndon B. Johnson, The Vantage Point, pp. 416-418; Clark Clifford, The New Yorker, 13 May 1991, pp. 75-79; Maxwell Taylor, The Uncertain Trumpet, pp. 390-392; Gen. Phillip Davidson, Vietnam at War, pp. 524-525; George Allen, "The Indochina Wars, pp. 324-325; Pentagon Papers (Gravel ed.), Vol. IV, pp. 266-268, 591-593 (which mistakenly state that Carver gave these briefings on 18 and 19 March); Stuart H. Loory, The Washington Post, 31 May 1968 (which similarly misdates Carver's briefings and was perhaps the source of the Pentagon Papers' later misdating); Henry Brandon, Anatomy of Error: The Inside Story of the Asian War on the Potomac, 1954-1969 (Boston: Gambit Incorporated, 1969), pp. 132-135; Kalb and Abel, Roots of Involvement, pp. 244-248; and Carver, Memorandum for the Director, "The Bombing Decisions--31 March and 1 November 1968," 31 March 1970, (S/Sensitive). CIA files, Job No. 80R01720R, O/D/NFAC, Box 3, "GAC Files (SAVA-NIO)," Folder 1.
(166) Carver, "Notes for Establishment Briefing II - 25 January 1968," (5 pp. typescript, no classification indicated). CIA files, Job No. 80R01720R, O/D/NFAC, Box 5, "GAC Files (SAVA-NIO)," Folder 9. Also "Where We Stand in Vietnam," (apparently a backup for the above notes, typescript, 28 pp., no date or classification). Above CIA file, Box 1, Folder 8.
(168) Johnson, The Vantage Point, pp. 414, 416. It will be recalled that Walt Rostow was the principal screener of the President's daily information on Vietnam at this time and, according to at least one source, the principal drafter of his memoirs.
(169) Carver, Memorandum for the Director, "The Bombing Decisions--31 March and 1 November 1968," 31 March 1970, (S/Sensitive). CIA files, Job No. 80R01720R, O/D/NFAC, Box 3, "GAC Files (SAVA-NIO)," Folder 1
(170) George Allen says he implored Carver, when briefing the President following his session with the Wise Men, "...to resist the temptation to plug his personal views, to give a balanced presentation. He tended, even after Tet, to take a somewhat more hopeful view of the situation than I did." Allen, "The Indochina Wars," p. 334.
(171) SAIG 9460 (IN 62464), 3 March 1968, (S); and Carver, Memorandum for the Director, "Station Comments on GVN Actions and Performance," 4 March 1968 (S). CIA files, Job No. 80R01720R, O/D/NFAC, Box 2, Folder 13, "DIR/SAVA Field Query and Answers, Feb '68;" and Box 3, Folder 9.
(175) Allen, to author, 11 April 1990. Ambassador Robert Komer recalls that when he learned of what Carver had told the "Wise Men" and the President, "He sent me though the roof because he was so pessimistic." Komer, to author, 21 May 1990.
(177) A copy of this letter is in CIA files, Job No. 80R01720R, O/D/NFAC, Box 1, "GAC Files (SAVA-NIO)," Folder 7. Carver later kept the original proudly on display in his office at l8th and K Streets.
(179) "Communist Units Participating in Attacks During the Tet Offensive, 30 January Through 13 February 1968," (S/NF/prepared jointly by SAVA, OCI, and OER). CIA files, Job No. 80R01720R, O/D/NFAC, Box 2, Folder 14.
(180) Helms, Memorandum for Secretary of State Rusk, "Future Communist Strategy in South Vietnam," 23 February 1968 (TS/Compartmented). CIA files, Job No. 80R01580R, DCI/ER Subject Files, Box 15, Folder 2.
(181) George Carver's deputy, George Allen, had just reported from Saigon that "MACV's J-2 is quite unrealistic. Its recent assessment of 'net losses' for the VC since 1 January was absurd; they listed 14,000 VC troops as participating in the Tet Offensive in III Corps, of which they killed 12,000, another 2,050 died of wounds, and 784 were detained--a total loss of 14,838 out of 14,000 committed!" (Here Allen presaged the White House scene, just a month later, when Justice Goldberg asked General DePuy, "Who, then, are we fighting?") From Allen, letter to Carver, 25 February 1968, (U). CIA files, Job No. 80R01720R, O/D/NFAC, Box 5, "GAC Files (SAVA-NIO)," Folder 9.
(182) OCI-OER Memorandum, "The Communists' Ability to Recoup Their Tet Military Losses," 1 March 1968, (S/NF). CIA files, as above. This memorandum also reported that the enemy employed large numbers of civilian irregulars as "shock troops in many of the urban assaults."
(184) Helms, Memorandum for Gen. Maxwell Taylor, Special Consultant to the President, "Status Report on Resolution of Estimates of VC/NVA Strength," 9 April 1968. (S/NF). CIA files, Job No. 80R01580R, DCI/ER Subject Files, Box 15, Folder 2.
(185) Helms, Memorandum for Walt W. Rostow, Special Assistant to the President, "Estimates of Enemy Strength in South Vietnam" 2 May 1968, (S/Compartmented). CIA files, Job No. 78T02095R, O/DDI, Box 1, Folder 2.
(186) Graham (at the time Director of DIA), testimony to Representative Otis Pike's investigating committee, "Intelligence on Enemy Order of Battle at the Time of the Tet Offensive," November 1975 (Emphasis added). Copy on file in CIA History Staff.