coupled perhaps with the burdens that courts often perceive in performing in camera inspections, especially if the records are voluminous or technical in nature. Finally, in the recent Halperin case, the decision, among other things, emphasized that where material withheld under exemption 1 had not been classified until after a freedom of information request for it was received, the agency will have problems of proof that would not ordinarily be present in exemption 1 cases.

I will not go into the problems raised by the various administrative costs and burdens of processing requests for records with exemption 1 aspects. Such burdens are considerable and sometimes of questionable public interest justification. However, I understand that you have other witnesses who will address those problems in detail.

My appraisal of the substantive impact of the amended exemption 1 on national security is that genuine national security information can probably be adequately protected within the act's overall mandate for openness. In other words, my evaluation of the present situation under exemption 1 is "so far, so good." While there are still real problems and much room for improvement, the agencies with major national security responsibilities are applying greater restraint in the use of classification to deny access, the Justice Department is encouraging these developments, and the courts seem willing to take into account the efforts and responsibilities of these agencies. Our experience with exemption 1 is, for me, a heartening demonstration that a government based on the constitutional separation of powers can effectively realize important and even occasionally conflicting goals, so long as the several branches of government work conscientiously and with due regard for each other's functions.

That, Mr. Chairman, is generally the current situation concerning the exemption for classified national defense and foreign policy records. But, as you know, on June 1 the President issued Presidential Review Memorandum 29 (PRM). This memorandum, entitled a Comprehensive Review of the Classification System, directed the preparation of a new Executive order to replace Executive Order 11652. The stated goal of the undertaking was "to provide for greater openness in government while at the same time effectively protecting sensitive national security information." The proposed Executive order to replace Executive Order 11652 was circulated for comments to departments and agencies this week-thus, we have not had sufficient time to make a detailed analysis of its contents. However, I would like to point out and comment generally on several changes relevant to today's topic.

Before highlighting specific, proposed changes, I would like to note that the committee established to draft the proposed Executive order was directed to consider, inter alia:

How to provide for the maximum release of information to the American public on government activities and policies consistent with the need to protect sensitive national security information;

How to promote increased public access to this information through a more rapid and systematic declassification program.

Mr. Chairman, my initial reaction to the proposed Executive order is that it makes significant progress toward accomplishing those goals.

If one compares Executive Order 11652 with the proposed order, I think the following important changes would stand out:

1. The creation of a new Security Information Oversight Office to replace the Interagency Classification Review Committee (ICRC). While the Oversight Office will assume the functions of the ICRC, it has also been strengthened to provide for more effective oversight and control over the classification process.

2. The administrative sanctions for classification abuse have been strengthened. 3. There are new restrictions and controls on the creation of special access compartments.

4. For the first time there will be comprehensive classification criteria which must be applied by the original classifier before a document can be classified. This new concept should go a long way in reducing overclassification, especially in doubtful situations.

5. The problems associated with the classification and declassification of foreign originated documents have been addressed and the handling of such documents has been clarified.

See exhibit 50, p. 458 of the appendix.

6. The proposed Executive order also expands the prohibitions applicable to the classification process.

7. The new order, as proposed, does not tie the level of classification to the length of time the document must remain classified; declassification will no longer be too dependent upon the level of classification.

8. The departments and agencies will be required under the proposed order to adopt declassification guidelines. This is intended to facilitate the declassification of certain documents in bulk.

9. Mandatory declassification review will occur under the new order at 20 years rather than the current 30 years with an exception for foreign originated documents. They will remain at 30 years.

I think all of these changes, Mr. Chairman, will go a long way in establishing a manageable classification system and one that will provide greater public access to this information.

Senator ABOUREZK. The final witness today is Mr. Mark Lynch an attorney with the ACLU.

Do you solemnly swear that the testimony you are about to give is the truth, the whole truth, and nothing but the truth, so help you God? Mr. LYNCH. I do.

TESTIMONY OF MARK LYNCH, ATTORNEY, AMERICAN CIVIL LIBERTIES UNION, COUNSEL TO THE PROJECT ON NATIONAL SECURITY AND CIVIL LIBERTIES

Mr. LYNCH. Thank you, Mr. Chairman, for inviting me to appear here this morning before this committee.

This committee, of course, has played an historic role in the development of information law. Those of us on the outside of the Government are delighted to see the committee undertaking this kind of oversight hearing at this time and considering the possibilities of any legislative changes that might be appropriate and necessary.

Senator ABOUREZK. We have heard testimony from Government agencies in the past few days that they are delighted with the Freedom of Information Act.

Mr. LYNCH. I am prepared to testify on two different matters this morning. First, is the operation of the Freedom of Information Act under the (b) (1) exemption, with respect to classified national security information. Second, is with respect to the draft Executive order.1 Things have changed in Washington apparently. I got my copy of the draft order on Wednesday afternoon. Mr. Harmon did not get his until last night. And Mr. Feldman has not gotten his at all.

I am somewhat better prepared, although not completely prepared. We will be submitting a detailed commentary on the draft order. We are not sure to whom to submit it yet, but we will certainly submit it to this committee, among other bodies.2

In the past day and a half, I have done the best I can to consult with several of my colleagues and a few of my adversaries to get a reaction to the order.

First, I think perhaps I ought to comment briefly on the way the act is working with respect to classified information. In doing so, I think I will be commenting on a couple of matters that Mr. Blake raised and a couple of matters that Mr. Harmon raised, and Mr. Feldman as well.

1 See exhibit 50, p. 458 of the appendix.

See p. 469 of the appendix.

First of all, the act has been spectacularly successful in disgorging classified information; that is to say, a great deal of information that has been declassified as a result of freedom of information requests.1 We find that when people make initial requests, they very frequently get some of what they are after. They then appeal to a higher level in the agency, and they get somewhat more information. In fact, this varies drastically from agency to agency. For example, the appeals practice of the Department of State-based on my experience handling a lot of cases; I do not have an empirical, statistical basis-is virtually useless. They almost always affirm the decision of the initial authority if it was to withhold information.

The CIA, on the other hand, has an appeals process which is extremely beneficial. The appellate level-and I am not altogether clear on who they are seems willing to release more information than the initial authority. I suspect that this is because the initial decision is made by the originating component. Frequently the information that many people are after-some of the most interesting information—is information that has been originated by the Directorate of Operations, the clandestine service.

Obviously, as Mr. Blake pointed out, these people have been trained in the arts of secrecy for many, many years. This has been a traumatic experience for them. I can understand and appreciate that.

At the higher level, other people, with a more dispassionate point of view, get into the act. Their decision is frequently to release more information. Then we find that if we go to court, the agencies release still more information. I think Mr. Blake did not represent the situation entirely accurately when he said that the CIA has not lost any cases under the (b) (1) exemption. There are many cases where people have filed complaints in court, and the CIA has released very substantial amounts of information once the complaint is filed. This is well in addition to what they released at the appellate level. Additionally, as the litigation progresses, it has been our experience that even more information is apt to be released.

That is because of a case that no one has discussed yet this morning, Vaughn v. Rosen,3 484 Fed. 2d 820 (D.C. Cir. 1973). Vaughn v. Rosen is the central procedural case with respect to FOIA cases. It sets up a series of procedures whereby an agency must file affidavits with the court in which it indexes the documents that are being withheld and provides a detailed justification as to why those documents are being withheld, under whichever exemptions are being claimed for the documents or portions of documents.

There initially was an attempt after the act was amended-that decision was rendered prior to the 1974 amendments. There was an initial attempt on the part of the Justice Department representing its clients in the national security community to try to avoid the requirements of Vaughn v. Rosen in (b) (1) cases. While they met some success in some district court cases, the Vaughn requirements have been held to be applicable to (b) (1) cases. That is one of the holdings of the Phillippi case, which involved Glomar Explorer materials.

1 See p. 473 of the appendix.

2 See pp. 69, 71, 88 of the hearing text.

3 See exhibit 69, p. 618 of the appendix.

Phillippi v. CIA, 546 F. 2d 1009 (D.C. Cir. 1976). See p. 606 of the appendix.

I might also point out that the chairman raised a very good question as to why, after 18 months, somebody changed their minds and decided to disclose Glomar Explorer documents.1

I do not know who changed their mind. I think it would be a very good question to find out. I do not know if it was the Attorney General's people. I do not know if it was the CIA. I do not know if it was Mr. Brzezinski at the National Security Council.

General Scowcroft had been in on the decision to withhold the documents earlier in the litigation. I think it would be very interesting to know where the decision was made to change the position in the Phillipi case. It is the kind of thing I cannot find out in litigation but I think the committee has resources to find it out. It is a very interesting policy question.

One of the reasons that I think the Agency may have changed their or whoever changed-position was that the circuit court said that, if you are going to take this position that you cannot confirm or deny the existence of the records, you have got to make a detailed justification of that position, just as you would under Vaughn v. Rosen, making a detailed justification as to why you were withholding particular documents.

What had happened in the district court was that that justification was made ex parte, in camera with no opportunity to the plaintiff's counsel-I was plaintiff's counsel there-to test or probe the assertions of the Agency. The court of appeals said that, yes, an agency may in some situations properly claim that it cannot confirm or deny the existence of the documents. I can understand that. For example, if someone were to make a request for the file of any CIA agent who is a member of the Politburo in the Kremlin, obviously that file could not be disclosed, or even acknowledged to exist. Even to acknowledge that it existed, for purposes of saying that it could not be disclosed, could jeopardize national security interests. I can understand that in some

cases.

I could not understand it in the Glomar Explorer case, when there was an incredible amount of information already publicly available on the Glomar situation.

But, in any event, the court of appeals said that you will have to justify your position on the public record. The court should go as far as it can on the public record before it accepts any in camera ex parte materials.

That holding in the Phillippi case, which was with respect to the can't-confirm-or-deny position, is equally applicable to all other cases in which they claim that the information cannot be disclosed.

The agency should be obliged, and the court should insist, that the agency go as far as it possibly can in providing the justification on the public record. The plaintiff, in turn, directs discovery-either through interrogatories or depositions-to test and probe those claims and perhaps to construct an alternative theory under which the information might be disclosed.

At that point, with the record as fully developed as possible through vigorous adversarial testing-and all of which is based on a good solid

1 See exhibit 68a. p. 618 of the appendix.

* See Correspondence p. 429-432 of the appendix.

21-636-789

Vaughn affidavit to begin with. At that point, it may be appropriate for the court to look at the documents in camera. If there are a lot of documents involved, it may be possible for the court to look at just a small representative sample.

I do not like seeing my cases decided in camera, but there is really no other way to proceed under the Freedom of Information Act. The chairman said "shooting craps under a blanket"; I think it's a knife fight in the dark.

The only way that the plaintiff can get satisfaction, that the decision to withhold the information has been a sound one based on the exemptions, is to get the court to approve it. But, if you move for in camera inspection too soon-this is the question of litigation strategy-the court is likely to sustain the agency. I think a lot of the cases that have been lost have been lost because of poor litigation strategy. You do not want to dump a lot of documents into the judge's chambers for him to flounder around with, with nothing to guide him but the agency's affidavit.

You want to build your own record in the case so that, when the judge sits down with the documents or perhaps a representative sample of the documents-he has got two different points of view. He has got two different poles of reference before he makes the decision. While it is true that there have not been many cases in which information has been ordered to be disclosed, not many cases which are being litigated in what I regard as a proper manner have not reached a decision yet. So, we really do not know yet how well the act is working with respect to national security cases that are decided by judges on a fully developed record.

A lot of the cases that have been lost are cases where, in my view, people rushed and the courts rushed into in camera inspection.

There is an excellent opinion which I think lays out the way a court should proceed, St. Louis Post Dispatch v. Federal Bureau of Investigation. It was rendered by Judge Green of the district court here. It was decided on June 22. A copy has been furnished to the committee staff.1

There the FBI came on-there are classified documents involvedwith its explanation as to why the documents should not be disclosed. We took discovery. The Government moved for summary judgment, and we opposed with a counter fact statement.

The judge said that, because of the questions the plaintiff has raised about the accuracy of the assertions of the Agency, fairness dictates that the court review a few of these documents, a representative sample, in camera, so it can decide.

I think that that is probably the way matters ought to proceed under the act.

Mr. Harmon suggested that Weissman stands for the proposition that a court is not obliged to engaged in camera inspection. The particular facts of that case were that the plaintiff in the trial court did not attempt to take any discovery. The CIA submitted its affidavits. The plaintiff relied on a purely legal theory, sat back, and did not

1 See p. 626 of the appendix. Weissman v. CIA, -F. 2d. appendix.

(D.C. Cir. 1976) No. 76-1566. See p. 600 of the