Foreign Relations of the United States, 1964–1968, Volume IV, Vietnam, 1966

249. Memorandum From the Presidentʼs Special Assistant (Komer) to Secretary of Defense McNamara¹

Washington, September 29, 1966.

After weighing the pros and cons, I concur in your 22 September 1966 memo proposing a unified US pacification management under COMUSMACV.² In fact, despite the many secondary disadvantages which might be entailed, such a step seems to me essential if pacification is to be given the large scale push forward which I strongly believe is central to a satisfactory outcome in Vietnam.

As my small staff and I have progressively steeped ourselves in the Vietnam civil side, we have as you know also been insisting that stepped-up pacification is “central to success, both in ending the war and winning the peace …” We further agree that progress to date in this neglected aspect of the war has been negligible, largely in my view because of the lack of military resources which could be devoted to doing the local security part of the job. In any event, with most other aspects of the Vietnam war going better, pacification has become—as Cabot Lodge says—our chief unfinished business.

Stepping up pacification seems doubly important because it targets directly on what I believe to be the most vulnerable element in the enemy lineup—the southern VC. If we can step up both pressures and inducements targeted on the VC, we can increasingly isolate Hanoi. Moreover, until we can crack the VC infrastructure, the enemy always has the option of reverting to protracted guerrilla warfare.

I. So I favor every sensible measure to accelerate the momentum of pacification. The most important step is to get the GVN and ARVNAF really targeted on it, and adequately organized and motivated to do the job. At present the ARVNAF are not pulling their weight; unless we can get them fully involved in pacification, US forces will have to assume the main burden here too.

On the principle of first things first, straightening out the lines of responsibility on the US side is an essential prerequisite to tackling the GVN and ARVNAF. Split civil/military responsibility has not worked well enough in the past, nor is it likely to do so in the future. Unified management on the US side seems to me essential to giving a real push to pacification.

If such a “single manager” is thus essential, I further agree that he should be military. Indeed, I donʼt see any other realistic option given the simple fact that the bulk of the resources and management assets needed for pacification are military, on both the US and GVN sides:

A. The first prerequisite for successful pacification is adequate local security. This is primarily a military job. Even if the civil side has the whole responsibility, it doesnʼt control the key assets. An estimated year end strength comparison drives home this point.

	GVN Pacification Assets
	Military	Civil
ARVN (the 50% that Gen. Westmoreland proposed to be devoted to pacification)	140,000	35,000	RD Cadre (mostly not-Vung Tau-trained)
RF	145,000	20,000	Police (in provinces)
PF	155,000	_____
Total	390,000	55,000

B. Similarly, on the US side there are far more Americans in uniform advising on pacification than there are civilians. This is entirely separate from the question of how many US forces are or will be engaged in supporting pacification. Again, the arithmetic is instructive, particularly down at the critical district (sub-sector) levels.

	US Advisor Assets for Pacification
	Military	Civil
Corps	480	143	Region
Sector	860	327	Province
Sub-Sector	1200	0	District
Total	2540	470

C. The military are much better set up to manage a huge pacification effort. Quite frankly, I believe that we have made real progress toward strengthening the civilian management side of pacification since Ambassador Porter took this over last February. Progress has also been made toward better coordination between the military and civil sides of pacification. But coordination in my view is not enough to produce the needed results. The alternative of unified management under civilian control falls down because most of the assets involved are military, and because only the US military staff and advisory resources in Vietnam are big enough to manage pacification on the scale we seek. It would take at least eighteen months in my judgment to get a civilian management structure which might handle pacification as effectively as MACV could today.

In sum, if 60–70% of the real job of pacification is providing local security, this job can be done by the military, and without it the civil side of pacification cannot really get off the ground, the advantages of providing military management seem over-riding.

II. Clarification of your Charts would help. Your proposals raise several questions, particularly on US civil-military relationships. I believe that if they were handled in the manner described below, your proposal would be workable.

A. Would the primacy of overall supervision of the pacification process by the Ambassador and Deputy Ambassador be clearly restated? I believe that the Ambassador and his alter ego must retain clear personal authority over pacification as over all other aspects of our operations in Vietnam.

B. What would be included in pacification and put under the new MACV Deputy? We should avoid overburdening him with any more ancillary activities than he can effectively handle. The purpose of the exercise is to get pacification rolling in the countryside, not to get single management of a lot of functions that the manager wonʼt know how to manage. I assume you would put under him only those field activities (and the Saigon field operations staffs) which primarily and directly contribute to winning the village war. This should include Chieu Hoi, refugees, and resource control. It should not include (1) overall economic policy, the CIP, or anti-inflationary programs—these are best left to AID; (2) CIA programs other than Police Special Branch and Cadre; (3) national information and postwar programs other than those directed at the VC guerrilla structure and the rural population; (4) certain national AID project programs which are still best handled separately, e.g., medical, national agricultural activities, or education at other than the village level, etc.

C. Would the civilian components merely come under the new Deputyʼs command control or would MACV take them over in toto? The new Deputy and his corps and province Pacification Directors should assume operational control over the existing civilian components, but these should remain in civilian clothes, and be paid, administered, and logistically supported as at present. This would lead to certain “dual responsibility” problems, but be less disruptive than blanketing all the civil echelons functioning in the field into MACV.

D. What should a new Deputyʼs responsibility be for developing plans, programs and requirements for pacification? Clearly he must have such management tools at his disposal. Yet he shouldnʼt be burdened with the laborious tasks of preparing the CAP, programming personnel replacements, etc. Basically, he should be responsible only for those plans and programs that directly bear on field operations, and draw on the existing agencies for such support.

E. Should the pacification organization have logistic responsibilities? These should remain with AID, CIA, JUSPAO, and other elements of MACV, which would together service the new organization. To do otherwise would be disruptive and cause major delays, when the purpose of the exercise is to get up more momentum now.

III. Likely Disadvantages. Thereʼs no doubt that weʼd have to pay a price for unified US military management of pacification, but some of these seem to me to be more apparent than real.

A. The image problem. At a time when weʼre trying to emphasize the non-military aspects of the war, and when the GVN itself is civilianizing, there will be criticism of any apparently opposite trend. But this must be offset against another, even more important image which we seek—that pacification is finally moving forward, that the NVA forces are being increasingly isolated as the southern VC fall away, and that the population is at long last coming our way in ever larger members.

B. Is an essentially military US management the best means by which to influence the GVN? Since our policy is to utilize mainly Vietnamese for pacification rather than US forces, the interface between the US and GVN pacification structures is important. But the fact remains that the bulk of GVN pacification assets are under military control. Even the RD Ministry is largely military. So are the province chiefs. Military officers are in most of the key slots at region, province and district levels. The ARVN, RF, PF, and really the police are under military management, and likely to remain so. In practical terms, I suspect that a US military management could most effectively motivate and advise them.

C. An essentially military pacification structure might put us in an awkward position in case of negotiations. This would be the case if any truce arrangement called for a stand-down of organized military forces on both sides, which could leave the GVN/US without adequate instruments to deal with the VC infrastructure. For this reason, some have argued that we should civilianize the GVN paramilitary forces and put them under a civilian ministry. But I doubt as a practical matter that the Directory would countenance this, even after another election.

D. The proposed structure is too corps-heavy, when many argue that corps is an impediment to pacification and should be more decentralized. I agree, but doubt that it will be politically feasible to push the ARVN corps commanders out of the picture for a while. If so, letʼs use them, not deplore them.

E. Problems will arise from putting US civilians under the US military. Will many of them refuse to serve in this manner, including some quite senior people? Many bureaucratic sensitivities will be affected. Would this result in a major loss in efficiency on the increasingly important US civilian side of pacification? I believe that keeping civilian personnel in their own agencies, even though under a new Deputyʼs command control, would mitigate many of these problems, especially after a shakedown period. The mixed civilian-military staff you contemplate at all levels would help further.

F. Some dual responsibility and administrative confusion would result. I believe that this could be minimized under the leadership of Ambassadors Lodge and Porter, who would remain in policy control of all US activities in Vietnam. What is proposed is essentially to unify management responsibility for pacification in the field. I cannot see that such a reorganization would create more dual responsibility and administrative confusion than exists today. Simply having a US team chief in each province would be a major step forward; collaboration is no substitute for management control.

IV. Conclusion. Despite the disadvantages cited above, and perhaps others, they seem to me to be outweighed by the positive thrust which would be generated by unified US management. To be perfectly candid, I regard your proposal as basically a means of bringing the military fully into the pacification process rather than of putting civilians under the military. Without minimizing the growing civil side contribution to pacification, the US and GVN military have the bulk of the assets needed to provide the essential local security input and the management structure needed to make pacification work. So assigning responsibility to the Deputy COMUSMACV is the best way to get pacification moving in Vietnam. I donʼt think that we can achieve this—soon enough or well enough—without such a step.³

R.W. Komer

1. Source: Washington National Records Center, RG 330, McNamara Files: FRC 71 A 3470, SVN Trip, October 1966. Secret. Copies were sent to Rusk, Gaud, Marks, and Helms.
2. Document 245.
3. Carver forwarded a copy of this memorandum to Colby under cover of a memorandum, October 3, stating that “this office finds little it can agree with in Mr. Komerʼs memorandum.” (Central Intelligence Agency, DCI-Executive Registry, Job 80–B0158R, McNamara Project)