that could be expected to cause damage to the national security if the requested figures were revealed [Colby Tr. at 50-51].

Following this reasoning most of the information released by the government every day could provide a useful "benchmark" for a foreign analyst, enabling him to "refine" some estimate of some aspect of U.S. intelligence. Mr. Colby himself admitted that "intelligence today is more and more the study of open material" [Colby Tr. at 14 and 49]; and he made reference to the State of the Union message and copies of Congressional appropriations as examples of other "benchmarks" which could be useful to foreign intelligence agencies in the analytical process he had been describing. [Colby Tr. at 59].

According to the "benchmark analysis" argument, a foreign intelligence agency would attempt to pinpoint all the CIA programs, add up their cost, and compare that cost to the budget and expenditure "benchmark" figures. If that cost figure comes to less than the benchmark, the foreign agency would search very hard to find the programs it knows that it has not identified. Once these programs are found, the foreign agency would take steps to thwart their implementation.

Because of the flexibility of the CIA budget, however, a foreign analyst would have difficulty knowing what programs are included in a CIA "confidential purpose" budget or expenditure figures. Even if the analyst did know these programs, it would still be difficult to estimate their size accurately. Assuming, however, that the analyst was successful in getting this far in the process, he would still have to discover the "missing" CIA programs. And even if he identified the other CIA activities, this does not mean the foreign agency could successfully neutralize them.

A foreign analyst would encounter large hurdles at each step of the way in completing the benchmark analysis described by Colby. Furthermore, no one, including the CIA Director, would deny that there is a substantial difference between a foreign analyst "refining" an intelligence estimate and a reasonable expectation of damage to the national security. Mr. Colby himself stated that he did not believe everything useful to a foreign analyst could reasonably be expected to cause damage to the national security. Yet nowhere in his deposition did Colby explain why release of intelligence budget and expenditure total would threaten national security, while the availability of "sensitive" information from public sources would not.

The implications of the blanket secrecy argument are staggering. Are the CIA and other intelligence agencies so dependent on their protective shields of secrecy that they require total insulation from the very public they purport to serve? The Statement and Account Clause is one of the few public accountability provisions in the Constitution, apart from the requirement of popular elections. Is it to be overridden on the mere assertion that intelligence expenditures cannot be publicly accounted for? Is the public also to be deceived by the lying and manipulation which have become institutionalized under the CIA's cloak of secrecy? If not, Congress must reject the argument for budget secrecy.

The constitutional requirement of a full accounting must carry the day. This requirement is based not only on the Statement and Account Clause, but on the public's First Amendment interest in maintaining an "uninhibited, robust and wide-open debate" on public issues. New York Times v. Sullivan, 376 U.S. 254, 260 (1964). As one commentator has noted, without a regular Statement and Account for intelligence expenditures "neither Congress nor the public can determine whether the expenditures comply with the CIA's enabling laws, or with the Constitution, and whether they have been made without waste or corruption. Neither Congress nor the public can weigh CIA spending against that of other agencies; the CIA's internal ordering of priorities cannot be analyzed." Note, "The CIA's Secret Funding and the Constitution," 84 Yale L.J. 606, 619-20 (1975). Since these important, constitutionally protected purposes are frustrated by the secrecy of intelligence funding, it is necessary to turn to the least drastic means of maintaining the confidentiality of certain programs. Under this approach the separate budgets of each intelligence agency should be disclosed in detail, with the exception of those line items certified to relate to statutorily authorized "confidential purposes."

Thank you for the opportunity to appear before the Subcommittee.

TESTIMONY OF JOHN SHATTUCK, DIRECTOR, AMERICAN CIVIL LIBERTIES UNION, WASHINGTON OFFICE

Mr. SHATTUCK. Thank you, Mr. Chairman.

I would like to touch on several of the points in my prepared statement which I believe have not been fully developed by other witnesses, or have been developed in another direction from mine.

First let me say that the American Civil Liberties Union considers the issue of intelligence budget disclosure to be of paramount importance in the larger effort to insure, as Mr. Phillips himself has suggested, that intelligence activities are consistent with the Constitution and our democratic form of government.

I differ with Mr. Philips in the sense that I believe that in many ways, this issue, whether or not and how much the budget of the intelligence agencies should be disclosed, is a test of the will of this committee and of the Congress to conduct vigorous oversight of the intelligence community. It is probably the first such test, and we think the committee is fully prepared to meet it on the facts which support disclosure in a manner which I will set forth in some further detail,

The intelligence abuses which have been brought to light by the Church committee came about in part because the agencies had little built-in accountability to Congress or the public.

This lack of accountability-and I think that is the key term that is at issue throughout these hearings-is nowhere better symbolized than by the secrecy of the intelligence budgets. The first point to be made about this secrecy is that it is flatly inconsistent with the constitutional mandate, in the statement and account clause. However one may dispute the history of that clause and the manner in which it has or has not been applied, the clause itself is flatly inconsistent with the blanket secrecy of the intelligence budgets.

The clause is one of the few provisions in the Constitution, apart from the right of the people to elect their government in popular elections, which makes the government accountable to the people. The clause and the first amendment, which I think reflects some of the same interests, serve the general purpose of accountability with respect to public officials I think the committee has been searching for the interest which the public has in the disclosure of the information in the intelligence budgets, and I would identify it in a way that I don't think has been adequately brought out yet this morning or yesterday, as a right to make judgments, relative judgments about the spending priorities that the public ultimately wants for its money. What is it that the public wants to have its money spent on? How much money should be spent on education? How much on defense, how much on intelligence?

There is no way for the public to make that judgment right now, and even the aggregate budget for the entire community, which we submit is not a sufficient amount to disclose, but even that aggregate amount would permit the public to begin to make those kinds of relative judgments which it has a right to make under the statement and account clause, and under the first amendment.

Now, these judgments and these rights of the public should not be overridden except by a showing of grave and overwhelming national danger. Some views that have been expressed here by some members of the committee have suggested that the opposite should be true-that in fact, if there is any doubt, the Constitution itself should be suspended-but I respectfully submit to the committee that this is the kind of argument that got us into so much difficulty in the Watergate

and the impeachment period, and I am sure that the committee will disregard it.

The resolution that created this committee, Senate Resolution 400, contains an indication of the kind of gravity that would be necessary to override the statement and account clause or any other constitutional provision with respect to secrecy, and that can be found in the certification clause in section 8b (2) where "the threat to the national interests of the United States posed by disclosure is of such gravity that it outweighs any public interest in disclosure," and that certification would have to be made by the President and upheld by the Senate in order to prevent the disclosure of information, including, presumably, budget information.

Now, the second point I want to make, apart from the interest which I have identified that the public has in getting this budget information, relates to the contrast between the budget secrecy practices of the intelligence agencies on the one hand, and other sensitive defense and national security agencies on the other hand. With respect to these other agencies a great deal of budget information is available.

Although there have been instances in our history in which certain types of expenditures have not been publicly accounted for, there is not a single example that I am aware of-and I would like to be corrected if I am wrong-apart from the CIA and the National Security Agency, in which an entire agency has been financed in total secrecy, so that even the aggregate budget items for those agencies cannot be ascertained by the public.

While there are no other instances of blanket secrecy since the beginning of this country, a variety of agencies whose activities are at least as sensitive as the activities of those two agencies, have long followed a "confidential purposes" approach toward budget disclosure, and it is that approach that the American Civil Liberties Union would urge upon the committee for intelligence budget disclosure.

Since 1793, Congress has appropriated funds directly to sensitive agencies, and it has publicly specified the percentage of such funds which are to be used "for confidential purposes," without a detailed accounting. For example, the FBI, the Department of State, the Atomic Energy Commission, and, more to the point, the Defense Intelligence Agency from time to time, and the Nuclear Regulatory Commission are all funded to varying degrees in this manner. Nevertheless, even where intelligence activities by these agencies are treated as a confidential purpose, the total amounts spent for such activities by each separately funded agency are publicly disclosed. For example, the Appendix to the Budget of the U.S. Government for Fiscal Year 1976 includes budget estimates for Army personnel intelligence and communications activities, Navy intelligence operation and communications activities, and Defense Nuclear Agency intelligence and communications activities.

I think the annual publication of this kind of detailed intelligence spending seriously undermines the CIA's contention that its entire. budget must be protected from disclosure.

And let me, without going through the rest of my prepared statement, give you some indication of why we think the confidential purposes approach is appropriate for this committee to adopt in ascertain

ing the degree of information to be published from the intelligence budgets.

A confidential purposes approach to budget secrecy would tie the degree of permissible secrecy directly to the statutory authority of the agency in question. Obviously, one of the principal purposes of budget disclosure is to permit Congress and the public to scrutinize intelligence expenditures closely enough so that intelligence agencies are deterred from conducting activities prohibited by statute or the Constitution.

Now, the American Civil Liberties Union supports legislation that would restrict by statute and apply a charter to the various intelligence agencies in question. The legislation that we support has categories of permissible secrecy that are not unlike some of the categories that Professor Emerson was referring to this morning, but I don't want to get into a discussion of the details of any of the categories of secrecy so much as to commend to the committee the statutory approach toward budget disclosure which I think would be flowing directly from the charter restrictions that would be imposed on the agencies.

The same analytical approach that is taken in H.R. 6051, the charter legislation pending in the House that we support, could be adopted with respect to restrictions and definitions of agency jurisdictions in any legislation that might be developed to control the intelligence agencies.

The point is that this confidential purposes approach would lead to a line by line disclosure of intelligence spending, consistent with the statutory function of the agency in question. It would go beyond the aggregate intelligence community disclosure, and would also go beyond the disclosure of lump sums for individual agencies.

For the time being, and until these statutory restrictions are enacted, there is no question that the aggregate figures for each agency could and must, consistent with the statement and account clause and the accountability of the agencies, be disclosed.

I think I will reserve for any questions you might have the balance of my statement with respect to the arguments of injury that have been made against disclosure during the course of the last 2 days.

Thank you, Mr. Chairman.

Senator HATHAWAY. Thank you very much, Mr. Shattuck.
Mr. Halperin.

Mr. HALPERIN.

[The prepared statement of Morton Halperin follows:]

PREPARED STATEMENT OF MORTON H. HALPERIN, DIRECTOR, PROJECT ON NATIONAL SECURITY AND CIVIL LIBERTIES, CENTER FOR NATIONAL SECURITY STUDIES

Mr. Chairman: I appreciate this opportunity to present my views on whether the budget of the CIA and other intelligence agencies should be made public. Much of what I would want to say on this subject is included in an affidavit I prepared as part of a lawsuit brought by the American Civil Liberties Union under the Freedom of Information Act in an effort to obtain release of the CIA budget figure. The committee has that affidavit and I assume it will be included in the record of these hearings along with other papers from that case (Halperin v. CIA, Civil Action No. 75-0676). If not, I would ask that you do so.

Today, I would simply like to comment briefly on two points: First, the standard that the committee should use in deciding what should be made public and second, the real reasons, as I understand them, for resistance to making the budget figures public.

As this committee is well aware, the Congress has not established any standards to be used by the executive branch in determining what information should be kept secret in the interest of national defense or foreign policy. Until recently, Congress left the President free to decide on procedures and criteria for classification. In amending the Freedom of Information Act in 1974, Congress simply required that the executive branch observe the procedures of the Executive order on classification. That Order, 11652, in its preamble makes reference to the public's right to know but in its operational sections sets criteria for secrecy which relate only to injury to national security from disclosure. Thus, the definition of "confidential" is whether release "could reasonably be expected to cause damage to the national security." There is no requirement that the estimated damage be weighed against the importance of the information for public debate.

I would urge this committee to adopt a different standard in deciding what information it will make public. Over the longer run I would hope that the Congress would legislate a classification system which properly balances the public's right to know against the legitimate requirements of national security.

In the meantime the Congress is, of course, in no way limited by the standards of the Executive order. In deciding whether to release the budget figures the committee should, in my view, explicitly adopt its own standard. Such a standard need not be developed anew. Rather, it can be found in the resolution of the 94th Congress establishing this committee, Senate Resolution 400, in providing that the President can ask this committee to reconsider a decision to release information. In order to do so the President must certify that:

"The threat to the national interest of the United States posed by such disclosure is of such gravity that it outweighs any public interest in the disclosure." Senate Resolution 400, sec. 8(b) (2), 94th Cong., 2d sess.

I would urge the committee to apply this standard in its own deliberations. If it does so, it will have no difficulty reaching the judgment that the budget figures should be made public.

Permit me now to turn briefly to the major arguments against disclosure. Although arguments are made about the adverse consequences of releasing even the single figure, William Colby, with commendable honesty, has admitted that his major concern is that if a single figure is released, there will be pressure to release additional numbers. He cites with horror the detail of the atomic energy budget now made public. The committee may wish to inquire into the question of whether any injury has resulted from the release of those additional figures. James Schlesinger, with equal candor, has suggested that releasing the CIA budget figure would lead to pressure to reduce spending on intelligence matters. I think Messrs. Colby and Schlesinger are both correct, but I draw the opposite conclusion from these predictions. Releasing a single figure will generate pressure to release more, but so it should. As much of the intelligence budget as possible should be released. I would exclude only those details where the threat from disclosure is of "such gravity that it outweighs any public interest in disclosure." Obviously, the lack of public interest in some details should be weighed along with the estimated harm.

If Mr. Schlesinger is right, then that is as it should be. Surely it is unacceptable to keep budget figures secret in order to keep public pressures from influencing how the Federal Government spends our tax dollars.

Let me conclude, if I may, with a procedural suggestion. I would propose that the committee, this year, simply make the total budget figure for each intelligence agency and the overall total public. It should at the same time instruct the intelligence agencies to proceed in the future as other national security agencies do, submitting as much of the budget and its justification as possible in public and the remainder in secret.

Mr. Chairman, I am grateful for the committee's invitation to participate in these hearings and I am, of course, available to answer the committee's questions.

TESTIMONY OF MORTON HALPERIN, DIRECTOR, PROJECT ON NATIONAL SECURITY AND CIVIL LIBERTIES, CENTER FOR NATIONAL SECURITY STUDIES

Mr. HALPERIN. Mr. Chairman, I am painfully aware that your lunch hour is getting shorter

Senator HATHAWAY. Don't worry about it. Go ahead.

90-784 0-77—10