clearly understood procedures for executive sessions, prompt review of transcripts, and other procedural aspects of this kind of exchange, the flow of information from the executive to the legislative branch could be very much improved.

This leaves the question of that type of information that relates to the processes of the executive. Broadly speaking, these processes relate to information gathering, decisionmaking, and administration and personnel security. There are other categories in the Justice Department, but these are the categories that relate to the kind of work I used to do. The classic public cases of executive privilege have usually fallen in this last area—that is, administration and personnel security-for example, Secretary of Commerce Herriman's refusal in 1948 to release the loyalty/security file of Dr. Edward Condon, despite a House of Representatives vote directing him to do so. Similarly, Secretary McNamara in 1962 refused to divulge the names of of individual members of the speech-review office in the Pentagon who had been responsible for changes in certain speeches by senior military officers. In both cases, this authority for the privilege was of course that of the President himself, duly invoked, and in the McNamara case specifically sustained by Senator Stennis. Secretary McNamara, as a correlative, assumed full personal responsibility for the "censoring" decisions in question, and was prepared to explain and defend them. You will find an account of that in Henry Trewitt's recent book, "McNamara."

In the case of such matters of administration, personnel security, and staff-level responsibility, I think it is fair to say that liberal Americans in particular have been strong in defense of the claim of executive privilege. But where the executive process to be protected has to do with the gathering of information and the making of decisions, it seems to be the more conservative Americans who are prepared to give these matters great weight, while the liberal is prone to talk about the need for greater openness in government. This of course is always the balance.

Frankly, on this issue, I must align myself with the more conservative view. There is major importance, as well as the principle of separation of powers, involved in protecting the deliberations and staff processes of the executive branch in foreign affairs. Based on my own practical experience, I believe that disclosures of stafflevel deliberations and of particular sources of information can do serious harm to the ability of the government to operate effectively in the area of foreign affairs. If there have been grievious faults in this areas, as I am sure there have, the remedy lies not in the legislative trying to look over the shoulder of the executive more closely as to the process-which simply will not work-but in command responsibility expressed through the political processthat is, elections which change the administration.

Paradoxically, it is even my view that the processes of the executive may require protection against disclosure for a longer period than almost any information that affects the national interest in a direct sense. For information affecting the national interest, a reasonable lapse of time may lower the damage to nil, or at least to an acceptable level balanced against the public need to know. However, if officers of an administration should come to feel that their confidential

advice would be disclosed, short of a period of many years, I do believe that the consequences in terms of honesty, candor, courage, and frankness within the executive branch could be very serious indeed. This is an aspect that I believe any study of the subject would need to examine with particular care. What is at stake, in the last analysis, is the difference between an executive branch that operates frankly and one where everything is fudged or tailored to a least common denominator.

I do not relate these remarks to any specific case or recent example. On the contrary, they stem from reflection over a long period on the specific question, so crucial to historians, of when State Department records of various types should be made public. I think a strong case can be made as Professor Langer of Harvard argued last January in the New York Times Book Review, that the normal period before publication of key documents should be shortened from the present 20 years to perhaps 10 years-that is, in the case of the Department of State. But I would be inclined to limit such automatic release to final documents of decision, action, or communication, and to be more strict about informal and predecision materials. Bluntly, the State Department has trouble enough already with the problems of conformity and lack of candor.

This may seem a digression, but is not. For I believe the Congress and the executive could both improve the flow of information by steps that recognize the distinction, rough as it occasionally is, between material of substance and material related to the processes of decisionmaking and information-gathering. Specifically, the Congress is generally entitled to know:

What the executive branch has decided, or what action decisions it believes imminent or important in the future, and the reasoning related to these actual or potential decisions; it is not entitled to know what persons are involved, what differing views may be being aired, or other matters similar in nature to those that relate to the deliberations of a congressional committee or of the judiciary, prior to decision in each instance.

What the facts are in a given situation, or to be more precise, what the executive branch believes the facts to be-but it is not entitled, in the normal case (that is apart from the committees that look into the Central Intelligence Agency-and other intelligence agencies) to explore the sources of information, to ask for the revelation of confidential sources or to ask for conflicting staff-level judgments how the Executive judges the future, again without being entitled to pry into staff-level views or differences.

To state these points is to frame one of the central requirements for a good flow of information. If Congress is to get what it is entitled to, then the executive branch must be prepared to send competent policy witnesses, usually at assistant secretary or higher level, to take responsibility-in the same sense that Secretary Mcnamara did in the 1962 case I cited-for what has been done, or is being thought or believed. Over and over, the cases of friction in recent years have arisen when lower-level witnesses were asked to testify on matters where a senior man should have been present to take responsibility. Experts of course may be required, but the real responsibility should lie with a senior man. That, as I see

it, is what appointees at the polical level, requiring senatorial confirmation, are for. The burden on their time may at first be greater than it now is, but the ultimate returns in terms of restored confidence and a better informed Congress could be very great indeed. Let me summarize. What I am saying, in essence, is fourfold: (1) The Congress needs much more information than it now receives in the area of foreign affairs.

(2) A substantial quantity of such information could be made available through improved classification and declassification procedures.

(3) A very much greater quantity of information relating to the national interest could be made available if the Congress were prepared to accept appropriate procedures-as of course in a great many cases it now is for maintaining the privacy of what is truly such information at least for significant time periods.

(4) Congressional requests for information could be made very much more productive if they focused on the end result-what the executive branch did, or proposed to do, what the executive branch believes to be the facts of a given situation, or what the executive branch's judgment of future possibilities may be. That is, instead of focusing on processes or personalities, which seem to me the very things most entitled to the invocation of executive privilege, the Congress might profitably focus on the essential elements of action and decision as they have been worked out within the executive branch. The essential corollary is, of course that the executive branch should at all times be prepared to provide, with the greatest speed, responsible officers, preferably at the political level, to testify concerning these end aspects.

Mr. Chairman, these comments are of course personal and general. They are aimed at the broad problem of improving the flow of information rather than at the rather technical provisions of S. 1125. I hope they will be of some modest help in your subcommittee's inquiry.

Thank you very much for the privilege of appearing before you. Senator ERVIN. I would like to observe that while you disclaim any expertise your statement shows a most intelligent comprehension of the questions which arise in this field. In fact, this is an area in which you can't say that everything is black or everything is white. I do not believe it would be possible to make a clearer statement than you have made of what are the essential problems in the area of executive privilege.

Mr. BUNDY. Thank you very much.

Senator ERVIN. I am very much impressed by your suggestion that it would be well to have an impartial commission to study Government classification of information and see if it could come up with some recommendations which would provide very stable and rational guide lines to govern the flow of information between the executive and the legislative branches of the Government.

I am glad to say that Senator Roth of Delaware, who testified before the committee the other day, has introduced a bill to establish such a commission. I have had the privilege of cosponsoring it with him.

Professor Kurland?

Professor KURLAND. Mr. Bundy, I take it that your mention of classification does not suggest that a classification of a document precludes its access by the Congress.

Mr. BUNDY. Indeed not. Indeed, I think I said that the general rule should be that information classified or unclassified should be available to the Congress.

Professor KURLAND. And throughout your statement you talked about matters affecting the national interest. I assume that is a different standard than matters affecting the national security.

Mr. BUNDY. I think one wraps into the other. I think you will find that the term "national security" has grown up in recent years in legislative and administrative regulations. I was using the term "national interest" because my quick reading of the Rogers memorandum and a few other basic legal materials indicated that that was the historical legal phrase, which I take to be all embracing.

Professor KURLAND. So, indeed, the difficulty with it, is it not, is that almost any information that the executive branch has can be said to affect the national interest, or at least we would hope so? Mr. BUNDY. I think it ought to be more strict than that, and certainly more strict than that for withholding information from the Congress. I think the information ought to have a very clearclear and present impact is the old phrase-clear and present impact on the national interest to justify withholding it from the Congress. By way of example, I don't think it would have been wise for the Congress to participate, or for the Executive to try to bring the Congress into very sensitive diplomatic maneuvers such as the recent dealings which led to the invitation for the President to go to China. In all bluntness, during the time that that was in process, at any rate, and I would reckon for some time more, that is the kind of thing that the Executive is reasonably entitled to withhold from Congress.

Professor KURLAND. That last kind of material would quite clearly, I suppose, fall within the executive privilege as technically described. Mr. BUNDY. I suppose, if I read the law correctly, the executive privilege could be invoked on anything. You could invoke executive privilege on something which was totally unclassified simply because the Executive felt that this had some bearing on the process.

Professor KURLAND. To the extent it was the President that invoked that privilege?

Mr. BUNDY. Yes. I don't think the law limits this. This is a matter of commonsense and judgment, as the Chairman has already said.

Professor KURLAND. It seems to me that the emphasis you have placed, and some of our earlier witnesses have placed, is neither on the executive privilege nor on the classification of documents system, but rather on a privilege which lawyers know as a confidential communication privilege.

Mr. BUNDY. I think that is half of what executive privilege means. I think it is dual, as I said. Part of it is a question of the national interest, and may in a given case, which I regard as the exception, warrant the Executive withholding from the Congress information on that ground. The more classic case, and the ones that have been actually raised, involve protection of the privacy of the processes of

the Executive, which, as you say, is exactly comparable to the privilege that the Congress has with respect to the privacy of its subcommittee or committee deliberations, and that the judiciary has, as you personally well know, in respect to what the judges say to their clerks and all else before decision.

Professor KURLAND. So that essentially we have really three areas or problems before us: One, the procedure by which the executive privilege can properly be invoked, second, the relevance of the classification system with regard to access by Congress to information, and third, which is perhaps the most difficult, is an executive-privileged confidential communication, which goes beyond the President, I suppose, and down to what you consider politically appointed officials.

Mr. BUNDY. First, I think that is the accurate division. Secondly, as to the privilege going down below, I suppose that is not legally true for a minute. And I would think that procedures ought to be such that you adhere to the basic constitutional point that the President and the President alone is in charge of the executive branch.

Now, the communications the President sees at any given time, even in the most sensitive and important matter of foreign affairs, are only a percentage of what is taking place. It isn't communication directly with him. So that in practice it is assistant secretaries and the like who are involved. But the privilege is to my mind entirely that of the President.

Professor KURLAND. Executive privilege?

Mr. BUNDY. Executive privilege.

Professor KURLAND. But the problem of confidential communications goes below that level?

Mr. BUNDY. Oh, yes, because in essence it all feeds in to the President in the end. No substantial decision will be taken in the executive branch unless it be by the President or in essence with his approval, tacit or expressed.

Professor KURLAND. Thank you, Mr. Bundy.

Senator ERVIN. Professor Winter?

Professor WINTER. One of the difficulties I have, Mr. Bundy, with the formulation of the executive privilege which permits the President to refuse to divulge knowledge simply because he thinks it would in some way harm the national interest is that while the President has the duty to protect the national interest, that is a shared duty. It would seem to me that a valid claim of executive privilege has to be related to some acknowledged power of the Executive, such as his power as Commander in Chief. Surely the power of command includes the power to keep military secrets. His power to represent the United States in dealing with foreign governments surely includes the power to carry on confidential conversations with these other governments. I really have to say candidly that I don't think you can formulate a valid executive privilege which just says something about protecting the national interest and which is not related to an acknowledged executive power.

Let me ask you specifically about S. 1125. The first section of S. 1125 seeks to compel the person who has been requested to appear before a committee to appear even though he intends, or has been instructed, to invoke the executive privilege. Former Secretary Ache