154. Memorandum of Conversation1


  • President Ford
  • Members of President’s Foreign Intelligence Advisory Board (see attached list)2
  • Dr. Henry A. Kissinger, Secretary of State and Assistant to the President for National Security Affairs
  • Lt. General Brent Scowcroft, Deputy Assistant to the President for National Security Affairs

Adm. Anderson: The unofficial job of the Board is to comment on intelligence estimates. [1½ lines not declassified] We had problems with the ’74 strategic estimate.3 Here is a paper,4 Mr. President, that is unanimously approved by the Board.

Foster: I would like to make a claim about the estimate, support that claim, and make a recommendation about what can be done.

We read it last November and were struck by statements on almost every page. [He reads a paragraph on long judgments about Soviet unlikelihood of first strike.] In our view this is misleading. It gives the reader an unwarranted complacency. It may be right, but it overstates a single point of view. It makes judgments based on damn few facts. The data is frequently flimsy, conflicting, or nonexistent.

People make decisions on force levels. The Congress makes decisions on the budget. When this document doesn’t agree with the Secretary of Defense’s testimony, it’s like shooting ourselves in the foot.

[Page 692]

I can give three examples which make one suspicious: (1) On the SS–18 and 19, there are no facts on accuracy. There are some data indicating reentry vehicles going astray like Poseidon did. [2 lines not declassified] There are two ways to do it: average the data, or the data could come from instruments that are deliberately offset.

Second, we hadn’t worried too much about Minuteman vulnerability because of Polaris and our feeling is that it’s invulnerable. Last summer the Soviets had exercises in the Philippine Sea. They are now using a combination of techniques. One is to take advantage of the fact that a submarine has to loiter in home port. They practiced some sort of tracking techniques. And when one of our subs transited the area, they picked it up also. [less than 1 line not declassified] The Soviets have staffed a lab with people who were new to the field.

I think the sub is getting hemmed in. We don’t think it is fair to say there is no way for them to get hold of the Polaris force for 10 years.

Third, on bombers, the estimate doesn’t deal with the fact that the bombers are on bases that are subject to interdiction by subs. It is nip and tuck whether the bombers would get off, and if they used cruise missiles we might not ever see them. The estimate also doesn’t deal with the bomber problem as they transit the oceans and can be picked up. On penetration, the average number of Soviet exercises against low level bomber attacks are about 3%, after ’74, 30% were below 1500 feet. We have to deal with a variety of air defense systems, including mobile ones which can be moved in to fill gaps. They appear still to be trying hard to ensure high attrition—and it may be even worse in 10 years.

What are the difficulties? First, they do it on the basis of not much evidence, and second, pressure to say what the analyst knows leads him to insert judgments where facts are lacking. Mostly they are very good, but as they get carried forward, it gradually gains the status of fact.

What the decision maker needs from intelligence is what is fact and what is judgment and what is the range of uncertainty.

Kissinger: To what extent are the judgments the product of service bias?

Foster: Strongly. The Navy has especially a problem. It doesn’t want ASW information to get to the intelligence community.

We have complained about the estimate and can’t seem to get through. They are not honestly trying to distort.

I have some suggestions: On important questions, the community should have two teams doing independent, competitive analysis. The DIA and CIA are not competitors.

President: But 10 years ago we put all of them together to get a single viewpoint.

[Page 693]

Foster: Yes, but this would apply just to key issues.

Next, we need closer interaction between intelligence and user. Perhaps the user needs to ask questions differently so that the information is focused on his decision. The intelligence community should be asked to build the best case both for and against the decision.

Third, how to avoid the appearance of a net assessment, which is what this tends to include.

The right process maybe is to do an NIE, with the improvements of the kind we suggest, then a net assessment, then conduct a critique. When we have done net assessments in the past, we have never critiqued them.

We have tried to persuade the intelligence community to accept these critiques, but they don’t see anything wrong with what they do now. They think their judgments are right.

Kissinger: We have found it very difficult to get the intelligence community to put forth competing views. The tendency is to waffle over disputes rather than sharpen them. Second, it is very hard to overcome Service bias. Third, they have a vested interest to support their previous judgments.

Land: Maybe we should institutionalize the process and have competing analyses presented as a matter of course.

Kissinger: If you ask for two views, that will become stylized and compound the problem. For 85% of the issues, nothing difficult is necessary. But for the few cases where they start with different points of view, those should be amplified and fully presented.

Teller: How do you get alternate evaluations?5 I don’t know, but one way would be to get an experienced man—like Foster—to do it. He would have to have access to all the material.

Foster: Maybe you just have to try it—just tell the intelligence community you want a competitive estimate.

President: I doubt you can get that kind of competitive judgment in-house.

Kissinger: But no one outside has the knowledge to make the judgments.

Land: There are people outside who have had access over the years.

[Page 694]

Kissinger: I have great sympathy for the problem. The solution is not so obvious.

Anderson: I would favor a directive to the DCI making it clear this is the kind of change which has to be made in this area. It should come directly from the National Security Council. They can do it for this year’s estimate.

Kissinger: Why don’t you draft one?

Anderson: We must make the intelligence community work the way it should.

President: Draft a directive.

Baker: I want very briefly to review the status of Soviet electronic telephone surveillance. Our interim actions have been effective, but we are very worried about the longer-term actions. [5 lines not declassified] We think we need to establish communication facilities which will be invulnerable—principally by encryption. We think a new directive is needed to establish clear responsibility for getting the job done. We have such a memo prepared for you. It includes a supervisory group under Ed David. Some are domestic and some overseas; [1 line not declassified]

Cherne: Let me add on the economic side that it’s only since June that we are trying to find what they are doing [3½ lines not declassified]

On a different matter, the most recent poll around New York showed strong feeling that the U.S. had to have a strong national intelligence system. While the citizen likes to read about the CIA, he wants a strong one but under your control.

Foster: One ironic point. There has been publicity about American citizens being spied on, and that others are doing it. Why not just tell them to take out all their equipment?

Baker: It might work for a year or two.

President: Would we not be able to detect whether or not they were putting it back?

Baker: They could probably circumvent it.

Teller: You could at least say we would be doing our best.

Baker: We believe you should put out a directive to take steps to minimize our exposure to the Soviets.

President: I thank you very much. We have some tough decisions to make and you are very helpful.

  1. Source: Ford Library, National Security Adviser, Memoranda of Conversations, Box 14, August 8, 1975—Ford, Kissinger, PFIAB. Top Secret. All brackets, except for those included by the editor to indicate omissions in the text, are in the original. The meeting was held in the Cabinet Room of the White House. During a meeting held the morning of August 8, Kissinger briefed Ford about his upcoming afternoon meeting with PFIAB: “The topics will be the weakness of strategic estimates, [less than 1 line not declassified] and more emphasis on economic intelligence. They are overdoing the strategic estimate a bit, but they could be improved. [1 line not declassified].” (Ibid., August 8, 1975—Ford, Kissinger)
  2. According to the list of attendees, attached, but not printed, PFIAB members Anderson (Chairman), Baker, Cherne, Foster, Galvin, Gray, Land, Shultz, Teller, and Byers (Executive Secretary) were present. Luce did not attend. Rumsfeld was also present.
  3. NIE 11–3/8–74 is Document 149.
  4. See Document 155.
  5. On July 21, Teller gave Sonnenfeldt a paper outlining a proposed “alternative NIE” to the intelligence community’s existing estimates of Soviet strategic forces, which Teller claimed underestimated Soviet capabilities and intentions. Sonnenfeldt forwarded Teller’s proposal to Kissinger under a covering memorandum, July 22. (Ford Library, National Security Adviser, NSC Program Analysis Staff Files, Box 25, Subject Series, PFIAB/NIEs, 1975–76 (5))