Foreign Relations of the United States, 1964–1968, Volume III, Vietnam, June–December 1965

142. Memorandum From Secretary of Defense McNamara to the Chairman of the Joint Chiefs of Staff (Wheeler)¹

Washington, September 15, 1965.

SUBJECT

• Air Strikes Against North Vietnam

We have considered the Joint Chiefs of Staff’s proposed Rolling Thunder 32/33 program for 17-30 September. That program would implement immediately a major, concentrated effort of air strikes against North Vietnam’s Phuc Yen and four other Hanoi/Haiphong airfields, against rail and highway routes and traffic between Hanoi and Haiphong, against rail and highway arteries and traffic between Hanoi/Haiphong and China, against four thermal power plants generating nearly 50% of the DRV’s thermal power, and against SAM installations which defend the above areas.

At this date I am not persuaded by the reasoning of JCSM 670-65² that the military advantages the Joint Chiefs of Staff state would flow from the proposed strike effort outweigh the military and political risks involved in implementing the proposal.

There is doubt whether the proposed program will mean a net advantage to US/GVN operations within South Vietnam. For example,while the LOC targets proposed in northeast North Vietnam all relate to interdiction of war materials into North Vietnam and therefore add to our efforts to interdict the flow of materials out of North Vietnam into South Vietnam, the intelligence estimates have been that such US strikes may well lead to a more vigorous effort by the DRV in support of the VC in South Vietnam. And the estimates have been that strikes of the kind proposed by the Joint Chiefs would not at this time significantly injure the VC ability to persevere in the South or persuade the Hanoi Government that the price of persisting was unacceptably high.

More important is the risk of a US-Chinese confrontation, which could well be increased by a program of the kind proposed. I am acutely aware that it is possible that US/GVN forces at Da Nang may be subjected to air attack by North Vietnam’s eight IL-28s (and perhaps MIG 15/17s). The intelligence judgment, however, has been that such an attack is unlikely, largely because of the fear of widespread and large-scale retaliation on North Vietnam by the United States. The judgment has been that such a DRV attempt to strike in the South might be triggered by a US/GVN strike against the Hanoi/Haiphong area, and that there is a substantial risk that a strike program of the weight recommended would induce the Chinese Communists to intervene in the air from Chinese bases.

It has been decided therefore that:

1. The Rolling Thunder program proposed by the Joint Chiefs of Staff for the period 17-30 September will not be approved at this time.

2. A new up-to-date Special National Intelligence Estimate will be obtained to determine the likely DRV, Chinese and Soviet military reactions to a program of the magnitude, timing, and scope recommended by the Joint Chiefs.³ This SNIE also should evaluate the effect of the strike program on the effort within South Vietnam and should take into account any new element introduced into the equation since earlier SNIEs were prepared, such as the Pakistan-India war.

3. The Joint Chiefs of Staff should make recommendations with respect to any further intensification of air defense and early warning capability required to properly defend South Vietnam and our forces there.

Robert S. McNamara⁴

1. Source: Washington National Records Center, RG 330, OSD/ADMIN Files: FRC 70 A 1265, Vietnam 381. Top Secret.
2. In this September 2 memorandum, the Joint Chiefs recommended the program summarized in the preceding paragraph. (Ibid.) Deputy Secretary of Defense Vance’s comments on JCSM 670-65, September 4, are contained in a memorandum to the JCS. (Ibid.)
3. SNIE 10-11-65, September 22, “Probable Communist Reaction to a US Course of Action,” concluded that the Democratic Republic of Vietnam would probably be more susceptible to a U.S. program of bombing attacks on certain new targets than it had been in the past. The Bureau of Intelligence and Research of the Department of State dissented fundamentally from this estimate, and its dissent is included in the SNIE. (Department of State, S/P Files: Lot 72 D 139, SNIE 10-11-65)
4. Printed from a copy that bears this typed signature.