The Pentagon Papers: Excerpt and Links

"Report of the Office of the Secretary of Defense Vietnam Task Force" (7,000 pages in pdf format)
(first made fully available to public in 2011)

[From The Pentagon Papers, Gravel Edition, Volume 4, Chapter 1, "The Air War in Vietnam, 1965-1968"]

2. The July Escalation Debate
The full U.S. entry into the Vietnam War in the spring of 1965--with the launching of air strikes against NVN, the release of U.S. jet aircraft for close support of ARVN troops in SVN, and the deployment to SVN of major U.S. ground forces for combat--did not bring an immediate turnabout in the security situation in SVN. The VCINV A may have been surprised and stunned at first by the U.S. actions, but by the summer of 1965 they had again seized the initiative they held in late 1964 and early 1965 and were again mounting large-scale attacks, hurting AR VN forces badly. In mid-July Assistant Secretary McNaughton described the situation in ominous terms:
The situation is worse than a year ago (when it was worse than a year before that). . . . A hard VC push is on. . . . The US air strikes against the North and US combat-troop deployments have erased any South Vietnamese fears that the US will forsake them; but the government is able to provide security to fewer and fewer people in less and less territory, fewer roads and railroads are usable, the economy is deteriorating, and the government in Saigon continues to turn over. Pacification even in the Hop Tac area is making no progress. The government-to-VC ratio overall is now only 3-to-l, and in combat battalions only 1-to-l; government desertions are at a high rate, and the Vietnamese force build-up is stalled; the VC reportedly are trying to double their combat strength. There are no signs that the VC have been throttled by US/GVN interdiction efforts; indeed, there is evidence of further PAVN build-up in the I and II Corps areas. The DRV/VC seem to believe that SVN is near collapse and show no signs of being interested in settling for less than a complete take-over.
Faced with this gloomy situation, the leading question on the U.S. agenda for Vietnam was a further major escalation of troop commitments, together with a call-up of reserves, extension of military tours, and a general expansion of the armed forces.
The question of intensifying the air war against the North was a subsidiary issue, but it was related to the troop question in several ways. The military view, as reflected in JCS proposals and proposals from the field, was that the war should be intensified on all fronts, in the North no less than in the South. There was political merit in this view as well, since it was difficult to publicly justify sending in masses of troops to slug it out on the ground without at least trying to see whether stronger pressures against NVN would help: On the other hand, there was continued high-level interest in preventing a crisis atmosphere from developing, and in avoiding any over-reaction by NVN and its allies, so that a simultaneous escalation in both the North and the South needed to be handled with care. The bombing of the North, coupled with the deployment of substantial forces should not look like an effort to soften up NVN for an invasion.
During the last days of June with U.S. air operations against North Vietnam well into their fifth month, with U.S. forces in South Vietnam embarking for the first time upon major ground combat operations, and with the President near a decision that would increase American troop strength in Vietnam from 70,000 to over 200,000, Under ­Secretary of State George Ball sent to his colleagues among the small group of Vietnam "principals" in Washington a memorandum warning that the United States was poised on the brink of a military and political disaster. Neither through expanded bombing of the North nor through a substantial increase in U.S. forces in the South would the United States be likely to achieve its objectives, Ball argued. Instead of escalation, he urged, "we should undertake either to extricate ourselves or to reduce our defense perimeters in South Viet-Nam to accord with the capabilities of a limited US deployment."
"This is our last clear chance to make this decision," the Under-Secretary asserted. And in a separate memorandum to the President, he explained why:
The decision you face now, therefore, is crucial. Once large numbers of US troops are committed to direct combat they will begin to take heavy casualties in a war they are ill­-equipped to fight in a non-cooperative if not downright hostile countryside.
Once we suffer large casualties we will have started a well-nigh irreversible process. Our involvement will be so great that we cannot--without national humiliation--stop short of achieving our complete objectives. Of the two possibilities 1 think humiliation would be more likely than the achievement of our objectives--even after we have paid terrible costs.
"Humiliation" was much on the minds of those involved in the making of American policy for Vietnam during the spring and summer of 1965. The word, or phrases meaning the same thing, appears in countless memoranda. No one put it as starkly as Assistant Secretary of Defense John McNaughton, who in late March assigned relative weights to various American objectives in Vietnam. In McNaughton's view the principal U.S. aim was "to avoid a humiliating US defeat (to our reputation as a guarantor)." To this he assigned the weight of 70%. Second, but far less important at only 20% was "to keep SVN (and then adjacent) territory from Chinese hands." And a minor third, at but 10%, was "to permit the people of SVN to enjoy a better, freer way of life."
Where Ball differed from all the others was in his willingness to incur "humiliation" that was certain--but also limited and short-term--by withdrawing American forces in order to avoid the uncertain but not unlikely prospect of a military defeat at a higher level of involvement. Thus he entitled his memorandum "Cutting Our Losses in South Viet­-Nam." In it and in his companion memorandum to the President ("A Compromise Solution for South Viet-Nam") he went on to outline a program, first, of placing a ceiling on U.S. deployments at present authorized levels (72,000 men) and sharply restricting their combat roles, and, second, of beginning negotiations with Hanoi for a cessation of hostilities and the formation in Saigon of a "government of National Union" that would include representatives of the National Liberation Front. Ball's argument was based upon his sense of relative priorities. As he told his colleagues:
The position taken in this memorandum does not suggest that the United States should abdicate leadership in the cold war. But any prudent military commander carefully selects the terrain on which to stand and fight, and no great captain has ever been blamed for a successful tactical withdrawal.
From our point of view, the terrain in South Viet-Nam could not be worse. Jungles and rice paddies are not designed for modern arms and, from a military point of view, this is clearly what General de Gaulle described to me as a "rotten country."
Politically, South Viet-Nam is a lost cause. The country is bled white from twenty years of war and the people are sick of it. The Viet Cong-as is shown by the Rand Corporation Motivation and Morale Study-are deeply committed.
Hanoi has a Government and a purpose and a discipline. The "government" in Saigon is a travesty. In a very real sense, South Viet-Nam is a country with an army and no government.
In my view, a deep commitment of United States forces in a land war in South Viet-Nam would be a catastrophic error. If ever there was an occasion for a tactical withdrawal, this is it.
Ball's argument was perhaps most antithetic to one being put forward at the same time by Secretary of State Rusk. In a memorandum he wrote on 1 July, Rusk stated bluntly: "The central objective of the United States in South VietNam must be to insure that North Viet-Nam not succeed in taking over or determining the future of South Viet-Nam by force. We must accomplish this objective without a general war if possible." Here was a statement that the American commitment to the Vietnam war was, in effect, absolute, even to the point of risking general war. The Secretary went on to explain why he felt that an absolute commitment was necessary:
The integrity of the U.S. commitment is the principal pillar of peace throughout the world. If that commitment becomes unreliable, the communist world would draw conclusions that would lead to our ruin and almost certainly to a catastrophic war. So long as the South Vietnamese are prepared to fight for themselves, we cannot abandon them without disaster to peace and to our interests throughout the world.
In short, if "the U.S. commitment" were once seen to be unreliable, the risk of the outbreak of general war would vastly increase. Therefore, prudence would dictate risking general war, if necessary, in order to demonstrate that the United States would meet its commitments. In either case, some risk would be involved, but in the latter case the risk would be lower. The task of the statesman is to choose among unpalatable alternatives. For the Under-Secretary of State, this meant an early withdrawal from Vietnam. For the Secretary, it meant an open-ended commitment.
Ball was, of course, alone among the Vietnam principals in arguing for de-escalation and political "compromise." At the same time that he and Rusk wrote these papers, Assistant Secretary of State William Bundy and Secretary of Defense McNamara also went on record with recommendations for the conduct of the war. Bundy's paper, "A 'Middle Way' Course of Action in South Vietnam," argued for a delay in further U.S. troop commitments and in escalation of the bombing campaign against North Vietnam, but a delay only in order to allow the American public time to digest the fact that the United States was engaged in a land war on the Asian mainland, and for U.S. commanders to make certain that their men were, in fact, capable of fighting effectively in conditions of counterinsurgency warfare without either arousing the hostility of the local population or causing the Vietnamese government and army simply to ease up and allow the Americans to "take over" their war.
For McNamara, however, the military situation in South Vietnam was too serious to allow the luxury of delay. In a memorandum to the President drafted on 1 July and then revised on 20 July, immediately following his return from a week-long visit to Vietnam, he recommended an immediate decision to increase the U.S.-Third Country presence from the current 16 maneuver battalions (15 U.S., one Australian) to 44 (34 U.S., nine Korean, one Australian), and a change in the mission of these forces from one of providing support and reinforcement for the ARVN to one which soon became known as "search and destroy"--as McNamara put it, they were "by aggressive exploitation of superior military forces. . . to gain and hold the initiative. . . pressing the fight against VC/DRV main force units in South Vietnam to run them to ground and destroy them."
At the same time, McNamara argued for a substantial intensification of the air war. The 1 July version of his memorandum recommended a total quarantine of the movement of war supplies into North Vietnam, by sea, rail, and road, through the mining of Haiphong and all other harbors and the destruction of rail and road bridges leading from China to Hanoi; the Secretary also urged the destruction of fighter airfields and SAM sites "as necessary" to accomplish these objectives.
On 2 July the JCS, supporting the views in the DPM, reiterated a recommendation for immediate implementation of an intensified bombing program against NVN, to accompany the additional deployments which were under consideration. The recommendation was for a sharp escalation of the bombing, with the emphasis on interdiction of supplies into as well as out of NVN. Like the DPM, it called for interdicting the movement of "war supplies" into NVN by mining the major ports and cutting the rail and highway bridges on the LOCs from China to Hanoi; mounting intensive armed reconnaissance against all LOCs and LOC facilities within NVN; destroying the "war-making" supplies and facilities of NVN, especially POL; and destroying airfields and SAM sites as necessary to accomplish the other tasks. The JCS estimated that an increase from the then 2000 to about 5000 attack sorties per month would be required to carry out the program.
The elements of greater risk in the JCS proposals were obvious. The recommendation to mine ports and to strike airfields and SAM sites had already been rejected as having special Soviet or Chinese escalatory implications, and even air strikes against LOCs from China were considered dangerous. U.S. intelligence agencies believed that if such strikes occurred the Chinese might deliberately engage U.S. aircraft over NVN from bases in China. CIA thought the chances were "about even" that this would occur; DIA and the Service intelligence agencies thought the chances of this would increase but considered it still unlikely; and State thought the chances "better than even."
Apart from this element of greater risk, however, intelligence agencies held out some hope that an intensified bombing program like that proposed by the JCS (less mining the ports, which they were not asked to consider) would badly hurt the NVN economy, damage NVN's ability to support the effort in SVN, and even lead Hanoi to consider negotiations. An SNIE of 23 July estimated that the extension of air attacks only to military targets in the Hanoi/Haiphong area was not likely to "significantly injure the Viet Cong ability to persevere" or to "persuade the Hanoi government that the price of persisting was unacceptably high." Sustained interdiction of the LOCs from China, in addition, would make the delivery of Soviet and Chinese aid more difficult and costly and would have a serious impact on the NVN economy, but it would still not have a "critical impact" on "the Communist determination to persevere" and would not seriously impair Viet Cong capabilities in SVN, "at least for the short term." However:
If, in addition, POL targets in the Hanoi-Haiphong area were destroyed by air attacks, the DRV's capability to provide transportation for the general economy would be severely reduced. It would also complicate their military logistics. If additional PAVN forces were employed in South Vietnam on a scale sufficient to counter increased US troop strength [which the SNIE said was "almost certain" to happen] this would substantially increase the amount of supplies needed in the South. The Viet Cong also depend on supplies from the North to maintain their present level of large-scale operations. The accumulated strains of a prolonged curtailment of supplies received from North Vietnam would obviously have an impact on the Communist effort in the South. They would certainly inhibit and might even prevent an increase in large-scale Viet Cong military activity, though they would probably not force any significant reduction in Viet Cong terrorist tactics of harassment and sabotage. These strains, particularly if they produced a serious check in the development of Viet Cong capabilities for large-scale (multi-battalion) operations might lead the Viet Cong to consider negotiations.
There were certain reservations with respect to the above estimate. The State and Army intelligence representatives on USIB registered a dissent, stating that even under heavier attack the LOC capacities in NVN and Laos were sufficient to support the war in SVN at the scale envisaged in the estimate. They also pointed out that it was impossible to do irreparable damage to the LOCs, that the Communists had demonstrated considerable logistic resourcefulness and considerable ability to move large amounts of war material long distances over difficult terrain by primitive means, and that in addition it was difficult to detect, let alone stop, sea infiltration. On balance, however, the SNIE came close to predicting that intensified interdiction attacks would have a beneficial effect on the war in the South.
Facing a decision with these kinds of implications, the President wanted more information and asked McNamara to go on another fact-gathering trip to Vietnam before submitting his final recommendations on a course of action. In anticipation of the trip, McNaughton prepared a memo summarizing his assessment of the problem. McNaughton wrote that "meaningful negotiations" were unlikely until the situation began to look gloomier for the VC, and that even with 200,000400,000 U.S. troops in SVN the chances of a "win" by 1968 (i.e., in the next 2½, years) were only 50-50. But he recommended that the infiltration routes be hit hard, "at least to put a 'ceiling' on what can be infiltrated;" and he recommended that the limit on targets be 'Just short" of population targets, the China border, and special targets like SAM sites which might trigger Soviet or Chinese reactions.
McNamara left for Vietnam on July 14 and returned a week later with a revised version of his July 1st DPM ready to be sent to the President as a final recommendation. The impact of the visit was to soften considerably the position he had apparently earlier taken. His 20 July memorandum backed off from the 1 July recommendations--perhaps, although it is impossible to tell from the available materials--because of intimations that such drastic escalation would be unacceptable to the President. Instead of mining North Vietnam's harbors as a quarantine measure, the Secretary recommended it as a possible "severe reprisal should the VC or DRV commit a particularly damaging or horhendous act" such as "interdiction of the Saigon river." But he recommended a gradual increase in the number of strike sorties against North Vietnam from the existing 2,500 per month to 4,000 "or more," still "avoiding striking population and industrial targets not closely related to the DRV's supply of war material to the VC."
The urgency which infused McNamara's recommendations stemmed from his estimate that "the situation in South Vietnam is worse than a year ago (when it was worse than a year before that)." The VC had launched a drive "to dismember the nation and maul the army"; since 1 June the GVN had been forced to abandon six district capitals and had only retaken one. Transport and communications lines throughout the country were being cut, isolating the towns and cities and causing sharp deterioration of the already shaky domestic economy. Air Marshal Ky presided over a government of generals which had little prospect of being able to unite or energize the country. In such a situation, U.S. air and ground actions thus far had put to rest Vietnamese fears that they might be abandoned, but they had not decisively affected the course of the war. Therefore, McNamara recommended escalation. His specific recommendations, he noted, were concurred in by General Wheeler and Ambassador-designate Lodge, who accompanied him on his trip to Vietnam, and by Ambassador Taylor, Ambassador Johnson, Admiral Sharp, and General Westmoreland, with whom he conferred there. The rationale for his decisions was supplied by the CIA, whose assessment he quoted with approval in concluding the 1 July version of his memorandum. It stated:
Over the longer term we doubt if the Communists are likely to change their basic strategy in Vietnam (i.e., aggressive and steadily mounting insurgency) unless and until two conditions prevail: (l) they are forced to accept a situation in the war in the South which offers them no prospect of an early victory and no grounds for hope that they can simply outlast the US and (2) North Vietnam itself is under continuing and increasingly damaging punitive attack. So long as the Communists think they scent the possibility of an early victory (which is probably now the case), we believe that they will persevere and accept extremely severe damage to the North. Conversely, if North Vietnam itself is not hurting, Hanoi's doctrinaire leaders will probably be ready to carry on the Southern struggle almost indefinitely. If, however, both of the conditions outlined above should be brought to pass, we believe Hanoi probably would, at least for a period of time, alter its basic strategy and course of action in South Vietnam.
McNamara's memorandum of 20 July did not include this quotation, although many of these points were made elsewhere in the paper. Instead, it concluded with an optimistic forecast:
The overall evaluation is that the course of action recommended in this memorandum--if the military and political moves are properly integrated and executed with continuing vigor and visible determination--stands a good chance of achieving an acceptable outcome within a reasonable time in Vietnam.
Never again while he was Secretary of Defense would McNamara make so optimistic a statement about Vietnam--except in public.
This concluding paragraph of McNamara's memorandum spoke of political, as well as military, "vigor" and "determination." Earlier in the paper, under the heading "Expanded political moves," he had elaborated on this point, writing:
Together with the above military moves, we should take political initiatives in order to lay a groundwork for a favorable political settlement by clarifying our objectives and establishing channels of communications. At the same time as we are taking steps to turn the tide in South Vietnam, we would make quiet moves through diplomatic channels (a) to open a dialogue with Moscow and Hanoi, and perhaps the VC, looking first toward disabusing them of any misconceptions as to our goals and second toward laying the groundwork for a settlement when the time is ripe; (b) to keep the Soviet Union from deepening its military in the world until the time when settlement can be achieved; and (c) to cement support for US policy by the US public, allies and friends, and to keep international opposition at a manageable level. Our efforts may be unproductive until the tide begins to turn, but nevertheless they should be made.
Here was scarcely a program for drastic political action. McNamara's essentially procedural (as opposed to substantive) recommendations amounted to little more than saying that the United States should provide channels for the enemy's discrete and relatively face-saving surrender when he decided that the game had grown too costly. This was, in fact, what official Washington (again with the exception of Ball) meant in mid-1965 when it spoke of a "political settlement." (As McNamara noted in a footnote, even this went too far for Ambassador-designate Lodge, whose view was that "'any further initiative by us now [before we are strong] would simply harden the Communist resolve not to stop fighting.'" In this view Ambassadors Taylor and Johnson concurred, except that they would maintain "discreet contacts with the Soviets.")
McNamara's concluding paragraph spoke of "an acceptable outcome." Previously in his paper he had listed "nine fundamental elements" of a favorable outcome. These were:
(a) VC stop attacks and drastically reduce incidents of terror and sabotage.
(b) DRV reduces infiltration to a trickle, with some reasonably reliable method of our obtaining confirmation of this fact.
(c) US/GVN stop bombing of North Vietnam.
(d) GVN stays independent (hopefully pro-US, but possibly genuinely neutral).
(e) GVN exercises governmental functions over substantially all of South Vietnam.
(f) Communists remain quiescent in Laos and Thailand.
(g) DRV withdraws PAVN forces and other North Vietnamese infiltrators (not regroupees) from South Vietnam.
(h) VC/NLF transform from a military to a purely political organization.
(i) US combat forces (not advisors or AID) withdraw.
These "fundamental elements," McNamara said, could evolve with or without express agreement and, indeed, except for what might be negotiated incidental to a cease-fire they were more likely to evolve without an explicit agreement than with one. So far as the difference between a "favorable" and an "acceptable" outcome was concerned, he continued, there was no need for the present to address the question of whether the United States should "ultimately settle for something less than the nine fundamentals," because the force deployments recommended in the memorandum would be prerequisite to the achievement of any acceptable settlement; "a decision can be made later, when bargaining becomes a reality, whether to compromise in any particular."
In summary, then, McNamara's program consisted of first substantially increasing the pressure on the enemy by every means short of those, such as the bombing of population centers in the North, that would run sizeable risks of precipitating Soviet or Chinese direct intervention in the war, and then seeking a de facto political settlement essentially on US/GVN terms.
The July 20 memo to the President was followed up by two others on specific aspects of the problem before the end of July. On July 28, he replied to a series of eighteen points made by Senator Mansfield with respect to the Vietnam war. In so doing, Secretary McNamara informed the President of his doubts that even a "greatly expanded program" could be expected to produce significant NVN interest in a negotiated settlement "until they have been disappointed in their hopes for a quick military success in the South." Meanwhile he favored "strikes at infiltration routes" to impose a ceiling on what NVN could pour into SVN, "thereby putting a ceiling on the size of war that the enemy can wage there." He warned that a greatly increased program would create even more serious risks of "confrontations" with the Soviet Union and China.
McNamara stated that the current bombing program was on the way to accomplishing its purposes and should be continued. The future program, he said, should:
a. Emphasize the threat. It should be structured to capitalize on fear of future attacks. At any time, "pressure" on the DRV depends not upon the current level of bombing but rather upon the credible threat of future destruction which can be avoided by agreeing to negotiate or agreeing to some settlement in negotiations.
b. Minimize the loss of DRV "face." The program should be designed to make it politically easy for the DRV to enter negotiations and to make concessions during negotiations. It may be politically easier for North Vietnam to accept negotiations and/or to make concessions at a time when bombing of their territory is not currently taking place.
c. Optimize interdiction vs. political costs. Interdiction should be carried out so as to maximize effectiveness and to minimize the political repercussions from the methods used. Physically, it makes no difference whether a rifle is interdicted on its way into North Vietnam, on its way out of North Vietnam, in Laos or in South Vietnam. But different amounts of effort and different political prices may be paid depending on how and where it is done. The critical variables in this regard are (1) the type of targets struck, (e.g., port facilities involving civilian casualties vs. isolated bridges), (2) types of aircraft (e.g., B-52s vs. F-105s), (3) kinds of weapons (e.g., napalm vs. ordinary bombs), (4) location of target (e.g., in Hanoi vs. Laotian border area), and (5) the accompanying declaratory policy (e.g., unlimited vs. a defined interdiction zone).
d. Coordinate with other influences on the DRV. So long as full victory in the South appears likely, the effect of the bombing program in promoting negotiations or a settlement will probably be small. The bombing program now and later should be designed for its influence on the DRV at that unknown time when the DRV becomes more optimistic about what they can achieve in a settlement acceptable to us than about what they can achieve by continuation of the war.
e. Avoid undue risks and costs. The program should avoid bombing which runs a high risk of escalation into war with the Soviets or China and which is likely to appall allies and friends.
3. Incremental Escalation
Secretary McNamara's 5 principles prevailed. The bombing continued to expand and intensify, but there was no abrupt switch in bombing policy and no sudden escalation. The high-value targets in the Hanoi/Haiphong area were kept off limits, so as not to "kill the hostage." Interdiction remained the chief criterion for target selection, and caution continued to be exercised with respect to sensitive targets. The idea of a possible bombing pause, longer than the last, was kept alive. The Secretary refused to approve an overall JCS concept for fighting the Vietnam War which included much heavier ROLLING THUNDER strikes against key military and economic targets coordinated with a blockade and mining attack on NVN ports, and he also continued to veto JCS proposals for dramatic attacks on major POL depots, power plants, airfields, and other "lucrative" targets.
The expansion of ROLLING THUNDER during the rest of 1965 followed the previous pattern of step-by-step progression. The approval cycle shifted from one-week to two­-week target packages. New fixed targets from the JCS list of major targets, which grew from 94 to 236 by the end of the year, continued to be selected in Washington. The number of these new targets was kept down to a few per week, most of them LOC-­related. Few strikes were authorized in the vital northeast quadrant, north of 21° N. and east of 106° E., which contained the Hanoi/Haiphong urban complexes, the major port facilities, and the main LOCs to China. In addition, de facto sanctuaries were maintained in the areas within 30 nautical miles from the center of Hanoi, 10 from the center of Haiphong, 30 from the Chinese border in the northwest (to 106° E.), and 25 from the Chinese border in the northeast.
The scope of armed reconnaissance missions was also enlarged but kept within limits. The boundary for such missions was shifted to the north and west of Hanoi up to the Chinese buffer zone, but it was kept back from the northeast quadrant, where only individually approved fixed target strikes were authorized. The operational latitude for armed reconnaissance missions was also widened. They were authorized to strike small pre-briefed fixed military targets not on the JCS list (e.g., minor troop staging areas, warehouses, or depots) in the course of executing their LOC attacks, and to restrike previously authorized JCS targets in order to make and keep them inoperable. An armed reconnaissance sortie ceiling continued in effect. It was lifted to 600 per week by October, but then held there until the end of the year.
By the end of 1965 total ROLLING THUNDER attack sorties had levelled off to about 750 per week and total sorties to a little over 1500 per week. All told, some 55,000 sorties had been flown during the year, nearly half of them on attack (strike and flak suppression) missions, and three-fourths of them as armed reconnaissance rather than JCS-directed fixed target strikes. Altogether, ROLLING THUNDER represented only 30 percent of the U.S. air effort in Southeast Asia during the year, in keeping with the rough priorities set by decision-makers at the outset.
Although bombing NVN had done much to generate, as Secretary McNamara put it, "a new school of criticism among liberals and 'peace' groups," whose activities were reflected in a wave of teach-ins and other demonstrations during 1965, the bombing also drew abundant criticism from more hawkish elements because of its limited nature. As a result, the Secretary and other officials were frequently obliged to defend the bombing restrictions before Congress and the press.
Most of the hawkish criticism of the bombing stemmed from basic disagreement with an air campaign centered upon a tactical interdiction rationale rather than a punitive rationale more in keeping with strategic uses of air power, a campaign in which the apparent target was the infiltration system rather than the economy as a whole, and in which, as one CIA report put it, . . .almost 80 percent of North Vietnam's limited modern industrial economy, 75 percent of the nation's population, and the most lucrative military supply and LOC targets have been effectively insulated from air attack.
This kind of criticism of the bombing concentrated on the most conspicuous aspect of the program, the strikes against fixed targets, and it faulted the program for failing to focus on the kinds of targets which strategic bombing had made familial' in World War IIpower plants, oil depots, harbor facilities, and factories.
Such "strategic" targets had not been entirely exempted from attack, of course, but they had been exempted from attack where they counted most, in the sanctuary areas. This occasioned some embarrassment in the Administration because any attack on such targets seemed inconsistent with a purely interdiction rationale, while failure to attack the most important of them did not satisfy a strategic bombing rationale. Secretary McNamara was pressed hard on these points when he appeared before the Congressional armed services and appropriations committees in August 1965 with a major supplemental budget request for the Vietnam War. Senator Cannon asked:
I know that our policy was to not attack power stations and certain oil depots and so on earlier. But within the past two weeks we have noticed that you have attacked at least one or more power stations. I am wondering if your policy has actually changed now in regard to the targets. In other words, are we stepping up the desirability of certain targets?
Secretary McNamara replied:
I would say we are holding primarily to these targets I have outlined. This week's program, for example, includes primarily, I would say, 95 percent of the sorties against fixed targets are against supply depots, ammo depots, barracks . . . but only one or two percent of the sorties directed against [one power plant].
I don't want to mislead you. We are not bombing in the Hanoi . . . or the Haiphong area. There is a very good reason for that. In Haiphong there is a substantial petroleum dump [for example]. First, there is question whether destruction of that dump would influence the level of supply into South Vietnam. Secondly, General Westmoreland believes that an attack on that would lead to an attack on the petroleum dumps outside of Saigon that contain eighty percent of the petroleum storage for SVN. Thirdly, there is the real possibility that an attack on the Haiphong petroleum would substantially increase the risk of Chinese participation . . . for all those reasons it seems unwise at this time . . . to attack that petroleum dump . . .
In defending the policy of not attacking the powerplants and POL sites concentrated in the Hanoi/Haiphong area, the Secretary did not stress the interdiction purposes of the bombing but rather the risks of widening the war. He explained that an attack on the powerplants and POL sites would require also attacking Phuc Yen airfield and the surrounding SAM sites:
I had better not describe how we would handle it but it would be one whale of a big attack . . . this might well trigger, in the view of some, would trigger Chinese intervention on the ground. . . . This is what we wish to avoid.
Before the House Committee on Armed Services two days later, Secretary McNamara stressed both the irrelevance of targets like the POL facilities at Haiphong to infiltration into the South and the risks of Chinese intervention:
At present our bombing program against the North is directed primarily against the military targets that are associated with the infiltration of men and equipment into the South, ammo depots, supply depots, barracks areas, the particular lines of communication over which these move into the South. For that reason, we have not struck in the Hanoi area because the targets are not as directly related to the infiltration of men and equipment as those outside the area. . . . As to the Haiphong POL . . . if we strike that there will be greater pressure on Communist China to undertake military action in support of the North Vietnamese. . . . We want to avoid that if we possibly can.
On other occasions the Secretary put such stress on the limited interdiction purposes of the bombing that it seemed to virtually rule out altogether industrial other "strategic" targets:
. . . we are seeking by our bombing in North Vietnam to reduce and make more costly the movement of men and supplies from North Vietnam into South Vietnam for the support of the Viet Cong operations in South Vietnam. That's our primary military objective, and that requires that we bomb the lines of communication primarily and secondarily, the ammunition and supply depots. . . . The great bulk of our bombing . . . is directed against traffic moving on roads and railroads, and the other portion . . . is directed against specific targets associated with the lines of communication, primarily supply depots and . . . bridges. . . . We think our bombing policy is quite properly associated with the effort to stop the insurgency in South Vietnam. We've said time after time: It is not our objective to destroy the Govel'l1ment of North Vietnam. We're not seeking to widen the war. We do have a limited objective, and that's why our targeting is limited as it is.
When asked whether the U.S. refrained from bombing NVN's more vital installations because it would escalate the war, the Secretary added:
Well, I'm saying that the other installations you're speaking of are not directly related to insurgency in the South, and that's what we're fighting. And that our targeting should be associated with that insurgency . . . our objective is to show them they can't win in the South. Until we do show that to them it's unlikely the insurgency in the South will stop.
The Secretary's arguments had difficult sledding, however. As 1965 ended, the bombing restrictions were still under attack. The U.S. was heavily engaged in the ground war in the South, and a limited bombing campaign in the North did not make much sense to those who wanted to win it. The hawks were very much alive, and there was mounting pressure to put more lightning and thunder into the air war. At that point, in not very propitious circumstances, the Administration halted the bombing entirely, and for 37 days, from 24 December 1965 to 31 January 1966, pursued a vigorous diplomatic offensive to get negotiations started to end the war. 

Pentagon Papers (Daniel Ellsberg) Trial Homepage