REPORT OF THE SELECT COMMITTEE ON INTELLIGENCE SUBCOMMITTEE ON SECRECY AND DISCLOSURE
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP81M00980R000600040017-4
Release Decision:
RIFPUB
Original Classification:
K
Document Page Count:
82
Document Creation Date:
December 15, 2016
Document Release Date:
May 18, 2004
Sequence Number:
17
Case Number:
Publication Date:
October 4, 1978
Content Type:
REPORT
File:
Attachment | Size |
---|---|
CIA-RDP81M00980R000600040017-4.pdf | 3.45 MB |
Body:
Approved r Release 2004/05/21 : CIA-RDP81 M00980R00
hews a ease
Snate Select Committee
on Intelligence
REPORT OF
AOLAI E. STEVENSON. ILL.
WILLIAM D. HATHAWAY. MAINE
WALTER D. NUODLESTON. KY.
JOSEPH R. BIDEN. JR., DEL.
ROBERT MORGAN, N.C.
GARY HART, COLO.
DANIEL PATRICK MOYNIHAN. N.Y.
DANIEL K. INOUYE. HAWAII
CLIFFORD P. CASE. N.J.
JAKE DARN, UTAH
CHARLES MCC. MATHIAS, JR., MO.
JAMES S. PEARSON. KANS.
JOHN H. CHAFEE. R.I.
RICHARD G. LUGAR IND.
MALCOLM WALLOP, , WYO.
ROBERT C. BYRD. W. VA.. EX OFFICIO
HOWARD H. BAKER. JR.. TENN.. EX OFFICIO
WILLIAM G. MILLER, STAFF DIRECTOR
EARL D. EISENHOWER, MINORITY STAFF DIRECTOR
SELECT COMMITTEE ON INTELLIGENCE
SUBCOMMITTEE ON SECRECY AND DISCLOSURE
NATIONAL SECURITY SECRETS AND THE
ADMINISTRATION OF JUSTICE
- BARiO
FOR RELEASE Al
OCT-) "1~4110-1!1 I
October 4, 1978
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
TABLE OF CONTENTS
PAGE
I. PREFACE -------------------------------------------- 1
II. SUMMARY --------------------------------------------- 5
III. BACKGROUND OF SECRECY AND DISCLOSURE
SUBCOMMITTEE INQUIRY ------------------------------- 8
IV. "LEAK" AND ESPIONAGE INVESTIGATIONS ----------------- 12
A. "Leak" Investigations ------------------------- 12
B. Espionage Investigations ---------------------- 15
C. Damage by Confirmation Versus Augmentation ---- 17
D. Augmentation of the Damage in Criminal Cases -- 18
E. "Gray Mail": The Price of Failing to
Resolve the Dilemma --------------------------- 21
v. CASES OF "GRAY MAIL" ------------------------------- 23
A. A Case of Bribery ----------------------------- 23
B. The KCIA Case: A More Recent Bribery
Conspiracy ------------------------------------ 25
C. The Khramkhruan Case: Narcotics Trafficking -- 26
D. The Nha Trang Murder -------------------------- 28
E. The Watergate Case.---------------------------- 32
VI. PAST LEGISLATIVE AND ADMINISTRATIVE PROPOSALS IN
RESPONSE TO THE "GRAY MAIL" PHENOMENON ------------- 35
A. Legislative Initiatives: Abortive Efforts
to Enact An Official Secrets Act -------------- 35
B. Administrative Initiatives -------------------- 40
VII. NEW INITIATIVES ------------------------------------ 42
A. Leaks, Espionage, and Current Law ------------- 44
B. Facilitating Enforcement of Existing
Statutes and the Charters --------------------- 50
VIII. RECOMMENDATIONS ------------------------------------ 62
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
i Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
I. PREFACE
For more than one year, the Secrecy and Disclosure Sub-
committee of the Select Committee on Intelligence has studied
the impact of secrecy on the administration of justice in
cases involving the national security. During this period,
the Subcommittee conducted case studies into investigations
and prosecutions where justice has been frustrated by claims
of national security.
The Subcommittee discovered that enforcement of laws
intended to protect national security information often re-
quires disclosure of the very information the laws seek to
protect. Indeed, the more sensitive the information com-
promised, the more difficult it becomes to enforce the laws
that guard our national security. At times then, regardless
of whether the compromise is to a newspaper reporter or
directly to a foreign agent, the government often must choose
between disclosing classified information in a prosecution or
letting the conduct go unpunished. In the words of one
Justice Department official who testified before the Sub-
committee, "To what extent must we harm the national security
in order to protect the national security?"
Evidence of this dilemma has been found in investigations
not only of leaks and espionage but also of bribery, drug
trafficking and murder. Therefore, this dilemma not only
adversely affects national security, but also can pervert
the administration of justice.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21: CIA-RDP81 M00980R000600040017-4
-2-
The balance between accountability to the law and
protection of national security information is a fragile
one. Intelligence agencies through the last several
decades have frequently insisted upon the inviolability
of the "sources and methods" of intelligence gathering
to the exclusion of other concerns. This insistence worked
to preclude many prosecutions involving national security
information.
In the past three years, however, this imbalance of
the past has caused the intelligence community and the
Department of Justice to be especially sensitive to the
importance of prosecuting such crimes. This administration,
much to its credit, has developed ad hoc informal procedures
for resolving this dilemma in many cases.
This Committee desires with the appropriate Executive
Branch agencies to develop permanent and formal procedures
to insure that consideration of the national security should
not in itself defeat the principle of accountability. Of
course, the Committee is especially concerned that the pro-
visions of the proposed intelligence community charters
(S. 2525), which provide for criminal sanctions for egregious
intrusions on the rights of Americans, as well as amendments
to the espionage statutes intended to protect the identity
of our intelligence agents, be enforceable to the fullest
extent possible. If the balance in national security cases
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-3-
in future administrations is skewed once again in favor
of the protection of "sources and methods" and other
classified information, charter provisions may become
unenforceable.
The Committee recognizes the need to confront the
issue of whether a major recasting of the existing espionage
statutes is or is not necessary. The Committee has nonethe-
less found that many practical, legal, and political differences
and difficulties lie in the path of such an undertaking.
Major advances can in the meantime be made in procedures and
practices under current statutes that will permit the resolu-
tion of many of the dilemmas regarding the use of national
security information in the administration of justice.
Although continuing examination of alternatives to the
current statutory scheme is necessary, the Committee at this
time wishes to recommend certain ameliorative steps, short
of any major immediate recasting of the law, because they
can yield effective improvement. By contrast, any substantial
revision of current statutes will occasion months, if not
years, of delay, with no improvement in the meantime.
Furthermore, some of the cases reviewed and testimony
received indicate that even the most radical revision of the
espionage statutes along the lines of the British Official
Secrets Act may not resolve this dilemma. Only the establish-
ment of a secret trial system for these kinds of cases would
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-4-
resolve the problem described in this report--not in our
opinion a very desirable or likely development. Ultimately,
the Congress must decide whether leaks of some national
security information and the exposure of some such informa-
tion in prosecutions are the inevitable cost of constitutional
guarantees of freedom of speech and the press and the constit-
utional right to a public trial.
Joseph R. Biden, Jr., Chairman
Subcommittee on Secrecy
and Disclosure
James Pearson, Vice Chairman
Subcommittee on Secrecy
and Disclosure
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
II. SUMMARY
The Committee's inquiry has led it to the following
conclusions:
(A) There has been a major failure on the part of the
government to take action in leak cases. To date, we have been
unable to identify a single successful prosecution of an individual
who leaked information to a publication. Admittedly, the question
of whether some leaks are punishable under existing statutes is
not altogether clear. The Committee found that leak cases are
uniquely difficult to investigate. But, we found cases where no
action was taken -- investigation or prosecution -- even where a
leak clearly violated an existing statute and caused serious harm
to our national security.
The failure has resulted in part from an impasse between the
Department of Justice and the intelligence community on how to
deal with the futher use of classified information necessary for
investigation and prosecution of these leak cases. Briefly stated,
there is no effective and formal mechanism for investigating these
cases or, in the few cases where the source of the leak is dis-
covered, weighing the risks of additional disclosures against the
benefits of prosecution.
(B) Several immediate steps may be taken to facilitate the
administration of existing laws, while Congress determines the
need for major revision of the espionage statutes. Furthermore,
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
it is possible that improvements in the administration of existin
statutes might affect ultimate decisions on statutory revision.
Present day reality and historical precedent show that numerous
political and practical obstacles would seriously delay any
major new statutes designed to deter leaks. While there is a
wide divergence of views among Committee members about what
changes, if any, should be made in the espionage statutes, a
narrowly drawn provision that would punish disclosure of the
identity of American intelligence agents appears to be necessary.
For the time being, the Committee has recommended several steps
to ensure the removal of obstacles to prosecutions which exist
under current law.
(C) Disagreements over the use of classified information in
prosecutions also impede espionage prosecutions.
(1) The Committee reviewed some espionage cases which
have not proceeded to either investigation or prosecution
for the same reason that leak cases cannot proceed -- concern
about the disclosure of intelligence information in the course
of investigation or prosecution. Furthermore, certain cases
engendered such intense disagreements between the intelligence
community and the Department of Justice that Presidential
intervention to resolve the disagreement was almost required.
(2) However, a resolution of the disagreement over the
use of classified information in espionage prosecutions is
likely for the following reasons:
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
(a) Espionage cases are generally considered more
serious than leak cases.
(b) The federal espionage statutes are more
clearly drawn to cover espionage than most leaks.
(c) Many espionage cases are in effect out of the
control of the intelligence community because the law
enforcement machinery has been engaged by an arrest,
or because the public or officials outside the intelligence
community know of the crime and, therefore, pressure the
intelligence community to provide information necessary
for prosecution.
(d) Usually the constitutional problems (primarily
First Amendment problems) are much less severe in
espionage cases than in leak cases.
(D) The impasse over the use of classified information in
prosecutions occurs in other types of criminal cases and at times
defendants may have placed the Department of Justice at a marked
disadvantage in perjury, narcotics, and possibly even one murder
case.
The Committee has formulated a series of recommendations
designed to alleviate some of the problems faced by the government
in maintaining the secrecy of legitimate national security infor-
mation. These recommendations can be found on pages
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
III. BACKGROUND OF SECRECY AND DISCLOSURE
SUBCOMMITTEE INQUIRY
On April 26, 1977, with the agreement of the full Committee,
the Subcommittee on Secrecy and Disclosure asked the staff to
undertake (1) a review of unauthorized disclosures of intelligence
information and (2) an inquiry into the use of compartmentation --
a procedure to place special limitations on access to information
that is especially sensitive. Although some progress has been made
on the second inquiry, most of the Subcommittee's work has con-
centrated on the first question which will serve as the focus of
this report.
The Subcommittee conducted its inquiry through both interviews
and file searches at the intelligence agencies. Over thirty
interviews and briefings were conducted with officials of the
Departments of Justice and State and the major intelligence agencies
(the Central Intelligence Agency, the National Security Agency, and
the Defense Intelligence Agency). In the course of these briefings
each agency was asked to provide the Subcommittee with ten cases
in which intelligence information had been covertly passed to
foreign powers -- classical espionage cases -- or in which intel-
ligence found its way into the public media -- intentional or
accidental leak cases. We have reviewed over thirty case files
or summaries of case files provided by these agencies. These
files have served as a valuable data base for our survey. They
represent the most comprehensive compilation of such information
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 _fjlA-RDP81 M00980R000600040017-4
in either the Executive branch or Congress. Each file contains
information on an intelligence compromise which has occurred in
the last few years, the action taken -- or not taken, as is
frequently the case -- by the relevant agency or the FBI, and any
disciplinary action taken against the individuals responsible.
In June of last year, after reviewing a summary of the results
of its survey, and based on a number of surprising findings, the
Subcommittee redirected its inquiry. The Subcommittee originally
began on the assumption that the major issue to be addressed would
be evaluating the desirability of additional criminal sanctions for
unauthorized disclosure of information that jeopardized sensitive
foreign intelligence "sources and methods". As the work proceeded,
however, the Subcommittee was soon driven to the conclusion that
no present statute can be effectively enforced against "leaks"
and that it would be a difficult task to draft a constitutional
criminal statute which would solve the enforcement problems. In
fact, the nation's strictest statutory safeguard against unauthorized
disclosure, Section 798 of Title 18, the U.S. espionage statute
which protects communications intelligence "sources and methods"
in a manner similar to that of the British Official Secrets Act,
has been infrequently used despite the large number of leaks of
communications intelligence. The files which the Subcommittee
has studied reveal several cases in which violations of even this
statute were neither investigated nor prosecuted.*
Part of the reason for the reluctance to bring cases under Section
798 is the lack of agreement as to whether courts will require the
prosecution to establish the propriety of classification. If a court
should decide to look behind the classification of a document, then
this would require the public disclosure of additional sensitive
information.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-10-
At the'heart of this failure of enforcement is a very deep-
seated conflict between the concerns of the intelligence community
on the one hand, and the Department of Justice on the other in
enforcing the espionage statutes. The conflict arises over whether
publicly to disclose classified information necessary to conduct
the investigation and to proceed with the prosecution.* Indeed
this question of whether or which classified information is to be
used in a particular judicial proceeding is a pervasive problem
that goes well beyond enforcement of the espionage statutes.
Problems created by classified information have also hampered many
other prosecutions, including perjury, extortion, bribery, narcotics
violations and possibly even one murder case.
.On March 1st, 2nd and 6th, the Subcommittee on Secrecy and
Disclosure conducted public hearings on the matters raised by
our inquiry. The Subcommittee heard from Admiral Stansfield Turner,
the Director of Central Intelligence; Benjamin Civiletti, then
the Acting Deputy Attorney General; Philip Lacovara, formerly of
the Watergate Special Prosecutor's Office; Judge Albert Fletcher,
Chief Judge of the Court of Military Appeals; William Colby, former
Director of Central Intelligence; Lawrence Houston, former CIA
It is common knowledge that the FBI and other counterintel-
ligence agencies do from time to time decide not to prosecute
espionage cases for other reasons such as the desirability of
monitoring a particular spy in order to understand the full
dimensions of a spy network. This report does not address these
kinds of cases but only those where investigation and prosecution
is the preferred approach.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-11-
General Counsel; and Morton Halperin, representing the American
Civil Liberties Union. The purpose of this report is to summarize
the Committee's findings based on these hearings and its year-
long inquriy, and to report its recommendations for legislative
and administrative actions to facilitate administration of certain
statutes related to the national security.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
IV. "LEAK" AND CLASSICAL ESPIONAGE INVESTIGATIONS
A. "Leak" Investigations
The Subcommittee examined thirty recent: cases submitted by
the CIA, NSA and DIA. These cases consisted primarily of
instances of leaks of intelligence information to the newspapers.
Of those thirty cases only three were actually referred to the
Department of Justice for investigation and none of those was
formally investigated. All were recent cases. Almost half of
the cases involved disclosure of communications intelligence,
which could have been prosecuted under Section 798 of Title 18 of
the United States Code (see Appendix). As noted earlier, Section
798 is the only espionage provision currently on the books that
approaches the strict liability criminal standard used by the
British in the Official Secrets Act, the model for recent
proposals to create new criminal sanctions for "leaks."
Many of the "leak" cases have not been investigated by the
FBI because of the Department of Justice's policy of refusing to
investigate unless the intelligence community is willing to
declassify all information related to the case. This policy grew
out of frustration by the Department over the years with
intelligence community reluctance to provide necessary evidence
to prosecute major leak cases after the FBI had invested
considerable time and effort in investigation.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-13-
According to those cases examined by the Subcommittee,
the response to those leaks which are subject to internal
intelligence agency investigations begins with an employee of
an intelligence agency who is familiar with the intelligence
and who identifies the possible leak when it is published.
For example, if the intelligence relates to information gleaned
from communications intelligence, an employee of the unit which
processes that intelligence would probably recognize the
sensitivity of the published information and report it to the
office of security of his agency. Upon receipt of the published
article containing the leak, the office of security of the
concerned intelligence agency would next attempt to determine
the individuals or offices who had access to the information.
This type of investigation is often fruitless because the
leaked information has been disseminated broadly in such inter-
agency classified materials as certain CIA intelligence cables,
the National Intelligence Daily or the Weapons Intelligence
Summary (some of which have circulation in the thousands). The
very information which must be disseminated to policymakers is
frequently the information which requires the greatest protection
from unauthorized disclosure. At the same time that the security
office is attempting to determine the scope of dissemination and
the possible recipients of the information, it is working closely
with the office within the intelligence agency where the.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-14-
information originated in the preparation of a damage
assessment.*
After the damage assessment is completed and a cursory
review of the number of people who might have had access is
finished, the information is forwarded to one of three organi-
zations: to the Security Committee of the Intelligence Community
Staff, to another agency if it is clear that the information must
have been leaked in a publication or from an office or individuals
of that agency, or (in a small fraction of the cases) to the
Department of Justice.
If reference to the Department of Justice is indicated, the
Department's response is pro forma. According to the cases
examined, the Department of Justice does not usually initiate
an investigation. It normally responds with a letter back to
the agency containing what is called "the eleven questions"
(see Appendix). Neither the Department of Justice nor the
FBI will normally proceed further until the eleven questions
are answered. Some of the eleven questions are uncontroversial --
such as whether the compromised information was properly classified
Most of the damage assessments that were reviewed were quite
perfunctory in nature and provided no specific information on the
actual and specific damage caused by the leak.
In fairness to those preparing the damage assessment at such
an early date in the process, it is difficult to assess the
damage because it is not yet clear whether or not a hostile power
has actually responded to the information in the article.
However, damage assessments were rarely updated in the cases
which were reviewed.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-15-
in the first place and whether the article disclosing it was
accurate. In most cases, particularly those of extreme
sensitivity, however, the whole process reaches an impasse
at Question 9, which reads as follows:
Whether the data can be declassified for
the purpose of prosecution and, if so, the name
of the person competent to testify concerning
the declassification.
The intelligence agencies view this as a requirement that
they agree to declassify any and all information in question
before the Department of Justice will agree to investigate the
case. Since the agencies rarely agree to this "up front"
commitment, few cases, if any, are ever actually investigated by
the Department of Justice. Indeed, of the 30 cases provided by
the intelligence agencies, none was investigated by the Department
of Justice.
B. Espionage Investigations
Espionage cases -- secretly passing classified information
to a hostile power -- are taken much more seriously than leaks by
both the Justice Department and the intelligence community. (See
discussion on page 44,) Despite the fact that espionage cases
and "leaks" may both be prosecuted under the same criminal
statutes, the eleven question leak questionnaire is not used in
espionage cases. Indeed in espionage cases a resolution is
almost always reached between the intelligence community and the
Justice Department on how to proceed with investigation.
Although the government is capable of resolving its differences
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-16-
in espionage cases, the decision in the past was often not to
prosecute. Recently, the CIA Office of General Counsel and the
Criminal Division of the Justice Department have -- much to their
credit -- succeeded in surmounting the many obstacles to
prosecution in espionage cases. Therefore the initial impasse
that prevents the opening of investigations in leak cases does
not occur. Nevertheless, even if the decision is to proceed
to trial in an espionage case, it is often a painful and hotly
contested matter causing friction between the Justice Department
and the intelligence community from the grand jury proceedings
through sentencing. The Subcommittee examined cases that did
proceed to prosecution and one case which was subsequently
dropped with no punitive action taken against an individual
who admitted to espionage; in that case the individual was
granted immunity in return for a full confession of what
information he had passed to a foreign nation.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : Cl f RDP81 M00980R000600040017-4
United States v. Moore was the successful prosecution last
year of a former CIA official who tossed classified documents
onto the Russian Embassy lawn here in Washington. United States
v. Boyce and Lee, also successfully prosecuted last year, in-
volves an employee of TRW, a large defense contractor in
California, who passed photographs of documents describing
extremely sensitive intelligence systems to the Russians. Both
cases were the subject of considerable tension between the CIA
and the Department of Justice. Both required protracted nego-
tiations on whether to use individual documents and witnesses
in the trial. In the Moore case disagreements between DCI
George Bush and Attorney General Levi almost required Presi-
dent Ford's intervention on his last day in office.
C. Damage by Confirmation Versus Augmentation
The intelligence agencies' concern about the effect of
investigation or prosecution of a leak or classical espionage
upon the national security falls into two basic categories:
(1) The investigation or prosecution of an espionage
violation can further damage the national security by confirming
the validity of the information disclosed. For example, in
either a covert transmission case or a leak case a hostile
power which discovers information very sensitive to the national
security may discount the information because of questions about
the reliability of the source, whether it be a spy or a newspaper.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIAl AZDP81 M00980R000600040017-4
However, if an indictment is filed against the subject or the
existence of an investigation is disclosed, the hostile intelli-
gence service might tend to interpret that indictment or investi-
gation as confirmation of the accuracy of the information provided.
This particular form of damage to the national security is
practically impossible to remedy because of the constitutional
requirement of a "public" trial -- the defendant has a right to
a public adjudication of the charges against him. This is one
reason why criminal sanctions for even the most serious "leaks"
to newspapers would be a particularly counter-productive remedy.
(2) Investigation or prosecution may augment the
damage to the national security by disclosing either to the
defendant or other interested parties further information
necessary either to investigate the case or to prove the case.
For example, it frequently becomes necessary in the course of
investigation to discuss the facts of the case with a variety
of witnesses who may be associates of the defendant. In a
criminal case there is a plethora of procedures which involve
public discussion of evidence related to the crime. This may
be particularly risky in espionage cases where prosecution may
disclose sophisticated counter-espionage techniques.
D. Augmentation of the Damage in Criminal Cases
This latter problem, augmentation of the damage, may be
easier to resolve than the former. Where the Justice Department
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 :-Cjl~_RDP81 M00980R000600040017-4
has determined to proceed, for example as in the Rosenberg or
Ellsberg case, or in the two major espionage prosecutions last
year, the prosecutors and judges have fashioned ad hoc proce-
dures to protect the national security and at the same time
ensure the administration of justice. These ad hoc procedures
form the focus of the Committee's present efforts.
In a criminal prosecution involving perjury, narcotics
smuggling, organized crime offenses such as extortion, or
espionage, there are a variety of circumstances in the course
of pre-trial or trial procedures in which government attorneys
fear a judge will require disclosure of classified information.
(1) As part of the case against the defendant. In a
typical espionage prosecution, classified information may be
directly relevant in proving the case against the defendant. For
example, in a prosecution under Section 793 of Title 18, it is
necessary to prove that the information passed will actually
damage the national security or be of aid to a foreign govern-
ment. Of course, in some cases the information passed is not
of obvious significance to a foreign government and there is
always the likelihood the foreign government does not understand
the impact of the information passed. In such a criminal trial
it becomes necessary to explain to the jury, and therefore to
the public and to the intended recipient, the significance of
the information passed. For example, in the Moore case the
government had to disclose publicly classified information
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIf13DP81 M00980R000600040017-4
contained in the documents tossed onto the Embassy lawn, but
which in fact were never examined by the :Russians.* Thus, here
the prosecution could have done as much damage to the national
security as the consummated crime.
The Boyce and Lee prosecution earlier this year was one
of the very few prosecutions under Section 798 of Title 18 for
the unauthorized dissemination of communications intelligence.
Even though Section 798 on its face does not require proof of
harm, Boyce and Lee were also charged under other sections of
the criminal code. Thus, it was necessary to prove that the
information was appropriately classified.
(2) As a part of the defendant's affirmative defense.
In the course of any of these prosecutions it is likely that
the defendant will raise an affirmative defense that will require
classified information. For example, an agency official prose-
cuted for deceiving Congress, might offer the affirmative defense
that it was a pattern or practice of Agency officials either to
conceal classified information in Congressional briefings or
even to deceive Congressional committees. In the alternative,
the official might argue that the information he provided the
Committee was indeed truthful. Obviously both of these offers
of proof would have required the disclosure of a considerable
amount of extremely sensitive, classified information. In a
* In this case the Federal judge took the extraordinary step
of sealing a public trial exhibit (consisting of the directory
and other sensitive documents), permitting only limited access
by the jury.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 ---RDP81 M00980R000600040017-4
case of organized crime and narcotics smuggling, a defendant
might allege that a former association with the Agency provides
a putative affirmative defense which would require evidence of
the CIA's relationship to him or similar agency relationships
to other individuals in the underworld.
(3) As part of pre-trial discovery. In every
criminal trial the defendant is entitled under the Constitution,
under statute, or under the Federal Rules of Criminal Procedure*,
to: (a) all materials obtained from or belonging to the defendant;
(b) anything "material to the preparation of his defense"; (c)
information pertaining to the testimony of a government witness;
and, (d) any exculpatory information within. the government's
possession. Frequently the information which must be disclosed
in these pre-trial procedures is classified.
E. "Gray Mail": The Price of Failing
To Resolve the Dilemma
Since the Espionage Act was enacted in 1917, the Federal
Government has been cautious in using the statute because of
the necessity to provide further classified information in the
course of a prosecution. Prosecutors in the Department of
Justice and intelligence community officials have always recog-
nized that the espionage statute is not an effective remedy for
all "leaks" to the newspaper or covert transmission to a foreign
spy because of the counter-productive disclosure of further
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : C_RDP81M00980R000600040017-4
secrets. The Department of Justice is also aware that a
defense counsel, in the course of trial or through pre-trial
discovery, can threaten the government with discovery motions
or a line of questioning that requires the disclosure of
classified information. An internal CIA study of this problem
in 1966 characterizes the dilemma as follows:
Out of this evidentiary difficulty has come
a sort of "gray mail", granted on the immunity
from prosecution (and often civil suit as well)
enjoyed by the thief who limits his trade to
information too sensitive to be revealed.
So long as there is a real threat that prosecution of the
defendant may reveal sensitive information in the course of a
trial, he or she may engage in this "gray mail" to avoid
prosecution.
Philip Lacovara characterized this problem in particularly
strong language.
...Agent 007, had a license to kill, but I think
the testimony and the findings of the Subcommittee
staff... support the judgment that the situation in
real life is even more sweeping than Ian Fleming
wrote of in his fictional novels...People...
connected with intelligence information, whether
they are themselves intelligence officers or other-
wise involved with national security operations, have
by virtue of the immunity from prosecution something
like a license not only to kill, but to lie, steal,
cheat, and spy...
Approved For' Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA- DP81M00980R000600040017-4
23-
V. CASES OF "GRAY MAIL"
The ambiguity of the statutes described in previous sections
and the internal Executive branch procedures for their enforcement
have at times created a legal vacuum -- often tantamount to
immunity -- for people who gain access to secret information.
The dilemma is most often confronted in the leak and espionage
circumstances described earlier, but occurs as well in cases
not usually associated with the national security -- bribery,
extortion, obstruction.of justice or murder.*
The following are actual cases in the public record where
secrecy and concerns about disclosure of sources and methods
actually interfered with the investigation or prosecution of a
serious felony which was not directly related to the national
security. These cases are important because they represent not
only the different kinds of crimes which give rise to this
phenomenon but also the subtlety with which concern about sources
and methods can interfere with the administration of justice.
A. A Case of Bribery
In his book The American Black Chamber published in 1933,
Herbert Yardley, who directed the United States' first signals
* There are no examples of leaks or espionage cases halted
for national security reasons included below because any further
public discussion of these cases might raise the same concerns as
investigations or prosecutions -- further disclosure of legitimate
national secrets.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : GI+RDP81 M00980R000600040017-4
intelligence operation, describes an incident concerning a
message which he intercepted between a foreign Ambassador in
Washington and his home government. The message implicated
the Ambassador in bribery of a high American government offi-
cial and his secretary.
In a subsequent meeting with a high official in the State
Department, Yardley admitted having sent the message to the
Attorney General. The State Department official and the
Secretary were furious that the Attorney General knew the
contents of the intercept even though it pertained to serious
criminal activity by government officials.
Yardley had thought it appropriate to send this message
over because it looked to him like a Justice Department case.
The State Department official was adamant. "The activity of
an Ambassador is never a Justice Department case," he stated.
Yardley himself warned that if the Ambassador were re-
called, "His government will appoint a new ambassador, install
a new code, and one never knows how much difficulty a new code
will cause." Yardley continued:
The new Ambassador will probably engage in the
same sort of activities, but we may not be in a posi-
tion to know just what is going on. Isn't it more
desirable to keep this Ambassador here and know what
he is up to than to have a new one without being
certain that we can check up on his activities?
The State Department official responded:
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA2IDP81 M00980R000600040017-4
Yes we have thought of all that. My impression
is the entire case will be dropped. It is too serious
to meddle with.*
B. The KCIA Case: A More Recent Bribery Conspiracy
In the early summer of 1971, a U.S. intelligence agency
reported to the Department of Justice the details of intensive
KCIA lobbying of the House Foreign Affairs Committee and a
substantial contribution to a U.S. Congressman. This informa-
tion was communicated to C.D. Brennan, FBI Assistant Director
in charge of the Intelligence Division, and to Assistant
Attorney General Mardian in charge of the Department of Justice
Internal Security Division. According to the Justice Depart-
ment records, Mardian promptly contacted the FBI to determine
whether the Bureau was investigating illegal transactions of
government officials with the Korean government. He arranged
for a personal review and an additional review by FBI officials
of the intelligence reports "to determine if any action can be
taken."
A few days later Bureau officials forwarded a summary of
the reports, and the following conclusions and recommendations
to Director Hoover:
We have received no information. regarding this
matter from any other source and there is no data in
Bureau files which would serve as a basis for the
Bureau's conducting any active investigation. We are
precluded from doing this based solely on (sensitive
intelligence reports). Further, even if the allegations
* Yardley, Herbert, The American Black Chamber (1933).
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIf~DP81 M00980R000600040017-4
from these sources could be proven, it is doubtful
that any prosecution could be sustained because of
intended disclosures which would be required in court
proceedings. The most logical action which might be
taken would be in the hands of the Department of
Justice, the Department of State or the White House
and would be in the nature of administration action
(sic) rather than prosecutive action.
Two days after this memorandum was written, FBI Director
Hoover sent a similar memorandum to Attorney General Mitchell
attaching a summary of the intelligence reports. Hoover's
memorandum to the Attorney General reiterates the Bureau
contention that it was precluded from instituting investiga-
tion based solely on such sensitive intelligence reports.
Hoover affirmed the absence of independent material in Bureau
files that could serve as the basis of any active investigation
into the matter. Hoover also expressed his doubts to the
Attorney General that any prosecution could be sustained be-
cause of attendant disclosures during court proceedings. Hoover
then concluded:
Information in the attached memorandum is also
being made available to Dr. Kissinger at the White
House. No further action is contemplated by this
Bureau.
No further action was taken regarding the Korea affair
until 1975.
C. The Khramkhruan Case: Narcotics Trafficking
The following narcotics trafficking case was discussed in
great detail in hearings before a subcommittee of the House
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIf7RDP81 M00980R000600040017-4
Government Operations Committee in 1975.* During those hearings
representatives of the Department of Justice and the CIA dis-
cussed the 1974 dismissal by the Department of Justice of an
indictment against a CIA operative on national security grounds.
In 1973 a CIA operative from Thailand, Puttaporn Khramkhruan,
was indicted for participating in the illegal importation of
25 kilos of raw opium into the United States. According to
the testimony and a subsequent congressional committee report
on the case, the CIA initially cooperated with Customs in
investigating Khramkhruan"s involvement in narcotics trafficking.
Khramkhruan was indicted along with six other individuals in
August of 1973. Originally he was to have been called as a
government witness, and not to have been named as a defendant.
However, Khramkhruan subsequently decided not to cooperate as
a witness and announced that he intended to leave the country.
Khramkhruan was arrested and served a superseding indictment
naming him a defendant. At that point Khramkhruan announced
that part of his defense would be that the CIA knew about his
opium smuggling.
Initially the CIA had promised its cooperation, including
provision of necessary documents and witnesses, to the Depart-
ment of Justice. Indeed it even volunteered to provide a
rebuttal witness to any claim by Khramkhruan that the CIA had
* Hearings before a Subcommittee of the House Government
Operations Committee July 22, 23, 29, 31 and August 1, 1975.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : Cl DP81 M00980R000600040017-4
advance knowledge of his narcotics trafficking.
However, shortly before the trial began the CIA notified
the U.S. Attorney that it would not produce documents necessary
for discovery under the Federal Rules of Criminal Procedures or
pursuant to the ruling in Brady v. Maryland, 373 U.S. 83 (1963),
nor would it provide a rebuttal witness on Khramkhruan's charge
of CIA advance knowledge, nor would it comply with the so-called
Jencks rule (18 U.S.C. 3500) requiring disclosure to the
defendant of prior statements of government witnesses.
According to the testimony of CIA witnesses*, the CIA's
request to the Justice Department for the dismissal of the
indictment was based on the fact that prosecution would lead to
discovery motions by the defendant which, when granted, would
reveal sources and methods of ongoing CIA clandestine operations
in Southeast Asia. The witnesses left unsaid the fact that CIA
would find it embarassing to have one of its operatives found
guilty of narcotics trafficking.
D. The Nha Trang Murder
The Army concedes the existence of a murder prosecution
that was thwarted by national security considerations. However,
the Army's records explain neither the facts leading to the
prosecution nor how national security impinged upon investigation
or prosecution. Because of the incompleteness of the record,
Hearings before the Subcommittee on Government Information
and Individual Rights of the Committee on Government Operations
of the House of Representatives, July 28, 29, 30, 31 and August
1, 1975.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIDP81 M00980R000600040017-4
the Committee was forced to rely primarily on newspaper accounts
and interviews.
In 1968 the Special Forces proposed to conduct an intel-
ligence operation which would employ Vietnamese spies as trail-
watchers operating on both sides of the Cambodian border. The
written operational proposal had to be cleared by the CIA, the
agerzcv charged with coordinating intelligence responsibility
and authority for U.S. forces in Vietnam. The proposal stated
that any agent found to be working for enemy intelligence would
be "terminated with extreme prejudice", a phrase allegedly
interpreted by CIA to mean that the officer would be turned over
to South Vietnamese legal authorities. CIA approved the opera-
tional proposal.
In the late spring of 1969, the Special Forces suspected
that a spy it had employed in the operation was in fact a double
agent who served North Vietnam intelligence. According to a
CIA official interviewed by the Committee, the Special Forces
consulted with the CIA and were advised in the methods of con-
ducting a proper counterintelligence interrogation and
investigation. According to this official, Special Forces person-
nel did interrogate the alleged double agent and, concluding
that he was guilty, killed him, apparently mistakenly relying
on the original operation proposal authority.
General Creighton Abrams learned of the incident and
ordered a preliminary criminal investigation. A month later,
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA- 81 M00980R000600040017-4
eight Special Forces officers were arrested in connection with
the death. Defense counsel for the officers during discovery
proceedings took testimony from a number of U.S. intelligence
community employees. According to the CIA the transcripts of
this testimony contained the details of a large portion of U.S.
intelligence activities in Southeast Asia. Press accounts of
the legal proceeding were extensive and public attention was
focused on the upcoming trial.* A civilian lawyer for three of
the Special Forces soldiers claimed that a representative of the
Agency "hid behind executive privilege."* Based upon the
discovery proceedings, this CIA officer assessed the likelihood
of public disclosure of intelligence sources and methods during
trial as very high.
On October 1, 1969, the New York Times reported that
Secretary of the Army Stanley Resor announced that he had "de-
cided to drop all of the charges in view of the fact that the
Central Intelligence Agency would not permit members of its staff
to testify."
In its recent memorandum to the Intelligence Committee,
the Army Judge Advocate General's office stated,
This office does not have a factual basis to verify
the accuracy of the statement that, for reasons of national
security, the Central Intelligence Agency would not make
available any of its personnel as witnesses at the pending
courts martial. On this basis, however, the charges were
dismissed by the Secretary of the Army on 29 September 1969.
In further explanation, representatives of the Department of the
- New York Times, August 25, 1969.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CT-1DP81M00980R000600040017-4
Army stated that the Army can confirm that the case was dropped
for national security reasons but all records of negotiations
between the Army and CIA over witnesses and documents for use
in the courts martial are no longer available at the Army.
In subsequent discussions with an official of the CIA, the
Committee learned of a meeting between DCI Richard Helms,
Attorney General Mitchell, and Secretary of Defense Laird. Those
principals, with President Nixon's concurrence, decided that the
case could not proceed for national security reasons and in-
structed Secretary Resor to drop the case.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA=F% P81 M00980R000600040017-4
E. The Watergate Case
The claim that intelligence activities must be protected
does not need to be legitimate for it to interfere with
investigations or prosecutions. Nor is it necessary that the
intelligence community make the claim. A prime example of these
two possibilities is the Watergate Case.
Within about a week of the Watergate break-in of June 1972,
FBI investigators discovered evidence linking the burglars to an
individual named Kenneth Dahlberg, and another individual named
Manuel Ogarrio in Mexico City. This was a critical link that
eventually traced the burglars to money in the Nixon reelection
campaign and ultimately to the White House.
According to the House Judiciary Committee Special
Impeachment Task Force report, as soon as the White House
discovered that the Bureau had uncovered the connection,
President Nixon directed Haldeman to meet with CIA Director
Helms, Deputy Director Vernon Walters and John Ehrlichman to
ascertain whether there was any CIA involvement in the Watergate
affair. The Impeachment Task Force's report summarizes the
results of that meeting as follows:
The President directed Haldeman to ask Walters
to meet with Gray to express these concerns and to
coordinate with the FBI, so that the FBI's investigation
would not be expanded into unrelated matters that would
lead to disclosure of the early activities of the Water-
gate principals.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : Q14IRDP81 M00980R000600040017-4
Although Helms had assured Haldeman and Ehrlichman that
there was no CIA involvement in Watergate, he did direct his
deputy to meet with the FBI Director and to remind the FBI of
the agencies' agreement that if either agency appeared to be
running into each other's sensitive operations, that they were
to notify each other and back away.
In a memorandum from Helms to Deputy Director Walters dated
28 June 1972, Helms gave the following directions:
In short at such a meeting (between Walters and
Gray), it is up to the FBI to lay some cards on the
table. Otherwise we are unable to be of help. In
addition we still adhere to the request that they
confine themselves to personalities already arrested
or already under suspicion, that they desist from
expanding this investigation into other areas which
may well eventually run afoul of our operations.
According to Walters' testimony before the Senate Watergate
Committee, Helms again reminded Gray of this arrangement on their
way out of the White House after their meeting with Haldeman and
Ehrlichman.
In a memorandum for the record dated June 28, Walters
summarized his meeting with Acting Director Gray as follows:
I recall that the FBI and the Agency had an
agreement in this respect and that the Bureau had
always scrupulously respected this. Gray said he
was aware of this and understood what I was
conveying to him.
For about a week the FBI did not proceed with the
investigation because it was under the impression that it had
indeed stumbled across a CIA operation and for national security
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : e+RDP81 M00980R000600040017-4
reasons felt that further investigations would jeopardize
sensitive information and operations. In fact no such operation
was involved but it is possible that Helms and Walters were not
sure at that time whether a CIA operation was involved. The
Watergate case, therefore, illustrates how such arrangements
could be used, especially by White House officials, to obstruct
a legitimate investigation.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21: dl RDP81 M00980R000600040017-4
VI. PAST LEGISLATIVE AND ADMINISTRATIVE PROPOSALS
IN RESPONSE TO THE "GRAY MAIL" PHENOMENON
Over the years the CIA and its predecessors have responded
with two initiatives to the problems of enforcement of the
espionage and other statutes which risk disclosures of foreign
intelligence "sources and methods". First, especially with
respect to leaks and espionage violations, military and civilian intelli-
gence agencies have called for enactment of statutes similar to
the British Official Secrets Act. Second, since 1954 the CIA has
sought special arrangements with the Department of Justice de-
signed to avoid controversies in these kinds of cases by relieving
CIA of its responsibility to report to the Department criminal
activity where further investigation might, in CIA's judgment,
jeopardize clandestine operations.
A. Legislative Initiatives: Abortive Efforts
to Enact An Official Secrets Act
Obviously, some of the problems described earlier in the
administration of espionage statutes would be resolved if the
culpability requirements were eased. It would be immensely easier
to prosecute leaks and espionage if all that had to be proven was
that the defendant had passed classified information to unauthor-
ized persons -- essentially the rule under the Official Secrets Act.*
According to Professor Benno Schmitt of Columbia Law School,
one of the nation's experts on our espionage statutes, proponents
It should be noted that the Official Secrets Act not only
applies to divulgence but also to publication of secrets, and
that its scope extends to all official government information,
not just national security secrets.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : C11RDP81 M00980R000600040017-4
of such legislation "reached back to Civil War experience, in
which the Union cause had been hindered by newspaper detailing of
military plans prior to their execution." The most famous
confrontation in the Congress over this kind of legislation was
during the Wilson administration when, according to Professor
Schmitt, the administration "proposed to censor or make
punishable after the fact (exactly which option was never made
clear), publication of defense information in violation of
Presidential regulations, without any limiting culpability
requirement." According to Schmitt:
In response to this proposal, the Congress
engaged in its most extensive debate over freedom
of speech in the press since the Alien and Sedition
Acts. The preoccupation was not an academic one.
Opponents feared that President Wilson or his
subordinates would impede, or even suppress, informed
criticism of his administration's war effort and
foreign policy under the guise of protecting military
secrets.. .The aggrandizing of presidential powers
during wartime was a recurrent fear of Republicans,
especially Senate progressives such as Borah,
LaFollette, Norris and Hiram Johnson.
The proposal was ultimately voted down and only the more modest
of the Wilson administration's espionage proposals were adopted.
That legislation serves as the framework for our present
espionage statutes.
Similar proposals were,made during the World War II period.
In 1946 the Joint Congressional Committee for Investigation of
the attack on Pearl Harbor recommended that Congress enact
legislation prohibiting the revelation of any classified informa-
tion. During the war there had also been a study jointly con-
ducted by Army and Navy Intelligence and the FBI which made
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CL f .DP81 M00980R000600040017-4
similar recommendations transmitted by the Secretary of War to
the Attorney General in June, 1946.
In 1947, the predecessor of Section 798, making it a crime
per se to reveal communications intelligence, was introduced and
in September of 1948 an omnibus bill was proposed by the Truman
administration incorporating the Section 798 language and a
number of earlier proposals for simplifying the culpability re-
quirements of the espionage statutes. During this period the CIA,
objecting to what it called a "piecemeal" approach of amending
various sections of the espionage statutes to deal with special
limited problems, suggested a redrafting of the whole espionage
statute along the lines of the British Official Secrets Act. A
few of the technical changes proposed by the Truman administration,
and the intelligence and the military departments were incorporated
into Title 18; the most significant of those was Section 798 of
Title 18. However the intelligence community and Department of
Defense were not satisfied with those amendments and in 1952
Defense Secretary Robert Lovett proposed to President Truman that
the administration still seek legislation similar to the British
Official Secrets Act. The Justice Department prepared such legis-
lation but it did not reach the floor in either House.
In 1957 the Commission on Governmental Security suggested
legislation that would make it a crime "for any person willfully
to disclose without proper authorization for any purpose: whatso-
ever, information classified, knowing such information to have been
so classified." The Commission justified its proposal in terms of
the "gray mail" problem":
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA3 DP81 M00980R000600040017-4
Since espionage cases may frequently involve
national security information of the highest
classification, the government is confronted with
a serious problem of how far such, information can
be compromised in the course of prosecution. . .A
defendant who may have met with the greatest
success in securing our most precious secrets, may
also have secured an advantage in warding off
successful prosecution.
No action was taken on the Commission's recommendation, nor
on subsequent initiatives in 1958 in the Eisenhower administration,
nor a similar initiative in 1966 by the CIA. Indeed, legisla-
tion was never seriously considered in this area until the Federal
Criminal Code Reform legislation was introduced by the Nixon
administration. That legislation contained some of the recommenda-
tions suggested by the intelligence community in the past but met
with strenuous opposition from media and civil liberties groups.
Similarly, those same groups strongly criticized legislation
drafted by the CIA and proposed by the Ford administration in
February of 1976. No action has been taken on the CIA proposal.
Typical of opposition that the Federal Criminal Code Reform
and the subsequent Ford administration proposal provoked is the
testimony of Jack Landau of the Reporters Committee for Freedom
of the Press before a Congressional subcommittee which was con-
sidering the Federal Criminal Code Reform:
It is abundantl clear that S. ll(the Code
reform proposal)- is an unwise an& unconstitutional
proposal which could be used to silence the type
of aggressive news reporting which produced articles
about the Pentagon Papers, the Mylai massacre, the
Watergate cover-up, the CIA domestic spying, the FBI
domestic spying and other government misdeeds. News
reporting which has been embarrassing to some persons
in the government and which is dependent in whole or
in part on government compiled information and reports
frequently supplied to the press by present or former
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : Gb 9 DP81 M00980R000600040017-4
government employees without government authorization.
The new espionage provisions of the Federal Criminal Code
Reform were dropped prior to its consideration by the Senate
early this year; proponents realized that any further action on
the Federal Criminal Code Reform would be indefinitely postponed
as long as there was significant controversy over its constitu-
tionality.
B. Administrative Initiatives
In February of 1954 Lawrence Houston, General Counsel for
the CIA, established an arrangement with William Rogers, Deputy
Attorney General, to obviate the need to report to the Department
of Justice certain criminal activity coming to CIA's attention.
According to a memorandum by Houston to Allen Dulles, Houston
.justified this arrangement to Rogers in the following terms:
Occasionally, however, the apparent criminal
activities are involved in highly classified and
complex covert operations. Under these circumstances,
investigation by an outside agency would not hope for
success without revealing to that agency the full
scope of the covert operation involved as well as this
agency's authorities and manner of handling the
operation.
Apparently, Rogers agreed with this assessment and "saw no purpose
in referring the matter to the Department of Justice" under the cir-
cumstances. There is some uncertainty in the materials the Commit-
tee has reviewed as to whether this arrangement was ever to have
been reduced to writing or any formal understanding between CIA and
the Department of Justice.
The ambiguity of the arrangement is highlighted by an exchange
of correspondence between the CIA and the Bureau of the Budget in
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 :-QQA-RDP81 M00980R000600040017-4
August of 1954. The CIA expressed concern regarding legislation
about to be enacted which would grant the Attorney General exclu-
sive responsibility for investigating all violations of Title 18
by government officers and employees. Notwithstanding the CIA's
concerns, that legislation was eventually enacted and codified as
5 U.S.C. Sec. 311(a) (since recodified in 28 U.S.C. Sec. 535(b)(2),
see Appendix).
In November of 1958, Rogers sent a memorandum to the heads of
all departments and agencies in the Executive branch of government
emphasizing their responsibilities under the legislation. Subse-
quent Attorneys General have issued the same reminder soon after
taking office. However, for over twenty years the CIA, based on its
1954 arrangement, assumed these directives exempted reporting the
kinds of cases Houston had described to Rogers. Although there were
minor changes in the procedures described in Houston's original
memorandum -- in 1955 and again in 1964 -- the basic thrust of the
arrangement wherein CIA took primary responsibility.for balancing
the need for secrecy against the administration of justice remained
until 1975.
In January of 1975 DCI William Colby and Lawrence Silberman
Acting Attorney General, reviewed the 1954 ,arrangement. At that
time Silberman took the position that the agency should comply with
5 U.S.C. Sec. 311(a) by providing a summary "but not an investiga-
tive report as such" in essentially every case and that the. basic
security issue should be raised, but that the Attorney General,
not the CIA, would make the decision on whether or not to prosecute.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CI2DP81M00980R000600040017-4
41-
The responsibility of the CIA to report evidence of crimes by its
employees to the Attorney General was the subject of a specific
provision in Executive Order 11905 issued by President Ford (de-
signed to regulate the activities of the intelligence community)
and its successor issued by President Carter, Executive Order 12036.
The Attorney General and DCI have recently signed a memorandum
of understanding which would serve as a successor to the 1954
arrangement.* The new Executive Order and the new memorandum
of understanding between Justice and CIA retain the principle
established by acting Attorney General Silberman that the Depart-
ment of Justice has the responsibility of balancing the needs of
secrecy against the ends of justice.
Both the memorandum of understanding and the Executive Order
purport to impose a burden on the intelligence community to report
criminal acts by its own employees. With respect to non-employees,
the new Executive Order reads as follows:
...(the head of any intelligence agency must)
report to the Attorney General evidence of
possible violations by any other person of
those federal criminal laws specified in guide-
lines adopted by the Attorney General.
No such guidelines have yet been adopted and, therefore, the
reporting requirements under that provision are unclear. Further-
more, neither the memorandum of understanding nor the Executive
Order addresses the way in which the Department of Justice should
handle evidence necessary to investigate or prosecute an allega-
tion brought to its attention under these provisions. In other
words, neither the memorandum of understanding nor the Executive
* The Committee has been informed that this memorandum may be
subject to further revision.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : ~I,-_RDP81 M00980R000600040017-4
Order is intended to resolve the controversies on the use of
classified information in the prosecution, the problem to which
this report is addressed.
Certainly one of the difficulties in developing these
policies is concern that these reporting requirements might in-
directly involve the foreign intelligence agencies in domestic
law enforcement in violation of the 1947 National Security Act.
The Committee shares this concern. However., the solution to
this dilemma may be in the distinction between passively report-
ing domestic criminal activity on the one hand and actively
seeking it out (e.g., "watchlisting" domestic subversives).
The drafters of future versions of the memorandum of understanding
and guidelines implementing the Executive Order should keep this
distinction in mind and avoid an unrealistic interpretation of
the domestic law enforcement prohibition.
VII. NEW INITIATIVES
The Committee agrees with former DCI Colby's testimony before
the Subcommittee on Secrecy and Disclosure that, "We would be
irresponsible if our revision of intelligence structure did not
recognize the need to protect the necessary secrets of intelligence
better than we do today." A resolution of the dilemma presented
by this report must be.a part of the charter legislation being
considered by the Intelligence Committee.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : Clff~-RDP81 M00980R000600040017-4
To meet the problems set out in this report, the Committee
has prepared a recommended program.* This program is designed to
serve two basic ends: first, to facilitate the enforcement of
espionage statutes and thereby protect our national secrets with-
out jeopardizing constitutional principles; and second, to facili-
tate enforcement of the criminal sanctions set out in the legis-
lative charters. Without question, the movement to apply the rule
of law to intelligence through statutory charters will be severely
undermined if leakers or spies continue to go unpunished or if
violations of the charters go unenforced.
Although unanimity exists among the members of the Committee
on the scope and significance of the problem of "gray mail", there
is substantial disagreement on a remedial program. Some members
such as Senator Wallop (see separate views) describe the recommenda-
77
tion of the Committee as resulting in only marginal improvements.
Other members find his approach or any major recasting of the
espionage laws to be fraught with the practical, legal and politi-
cal problems which have thwarted efforts to remedy this problem
in the past. This is not to say that the espionage statutes
written over six decades ago should not be subject to a serious
re-examination. There is strong sentiment that the committee should
undertake such a study but the implementation of the program
recommended herein should not await the completion of that study.
The program the.,.committee does adopt, is, however supported by
those who would take an even more fundamental approach, as being
the'minimum dictated by the record disclosed in this report. In
See pp. 62 ff.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-44-
the end the committee recognizes that if this program were
adopted in toto there would still be circumstances where some
leaks would go unpunished and some prosecutions subject to "gray-
mail" but perhaps that is the price we must pay for the consti-
tutional protections of a free press and a right to a public
trial.
A. Leaks, Espionage, and Current Law
The espionage statutes clearly cover most forms of traditional
spying. Nevertheless, prosecutions under these statutes have
often failed in the face of the "gray mail" phenomenon.
Leaks differ qualitatively from espionage. A leaker normally
discloses classified information not to a foreign agent but to
a journalist. In fact, this type of security leak has become part
of a flourishing informal and quasi-legal system. For example,
senior officials often disclose classified information as a means
of explaining their positions to the public, while dissenters leak
in order to expose improprieties and shoddy thinking.
There are two major drawbacks to the sub rosa practice of
providing selected intelligence information to the news media and
other sources. First, the public does not necessarily receive a
balanced view from the leaked information because the process is
informal. Second, and more importantly, information whose secrecy
is vital to our national security is sometimes disclosed.
Under current law, it is not at all clear whether most leaks
of information to the media are criminal. To the legal neophyte
in this field, it appears that Title 18, ?793(d) and (e) do address
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
the problem. ?793(d) and (e), in similar language, make
criminal the behavior of any person who, having lawful or
unauthorized,
access to . . . information relating to the national
defense which information the possessor has reason
to believe could be used to the injury of the
United States . . . willfully communicates . . . the
same to any person not entitled to receive it.
Superficially then, this statute seems to punish leaks
to journalists as well as spying. However, these statutes
are not normally used in cases unless transmittal of information
to foreign agents is involved. Whether they could be used
in cases where information is passed to a journalist is unclear
from a careful reading of the legislative record. This very
lack of clarity and Congressional reticence to clarify the statutes,
discussed earlier, has in fact discouraged leak prosecution under
these sections which in turn has precluded the growth of case law
to define the statute's meaning. Application of the statute to
publication of national defense information by a newspaper raises
serious First Amendment problems. Two distinguished commentators
have suggested that after taking into account constitutional
concerns, there is "little worth preserving in these two
remarkably confusing provisions."* As noted earlier, this
Committee has no desire to decide in this report whether these
statutes perform a necessary task or whether they do it adequately.
However, it does believe the administration should itself decide
under what, if any, circumstances it will seek to apply
Harold Edgar, Benno Schmitt. "The Espionage Statutes
Pub l1lmicedFoiER&e#eel$QO4VOfflZrmgWi-00?81 RIt OMDImD ~0A17P view 930.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-46-
criminal sanctions to leaks of national security information.
Administrations from the time of World War I have put
forward proposals that would resolve the ambiguities of the law
regarding leaks by making the disclosure of government secrets
a crime even without requiring proof of damage to the national
security. In fact, practically all of these attempts have
floundered in part because Congress has refused to make leaks
explicitly criminal which do not damage the national
security.
Although the mere classification of a document may not in
itself warrant criminal penalties for its disclosure, certain
narrow classes of information are in fact so sensitive that a
statute should protect them against any disclosure. Thus,
communications intelligence is protected against disclosure
even without proof of harm or communication to a foreign agent.
Former Director Colby testified in favor of a proposal that
would impose such strict liability penalities upon the
unauthorized disclosure by government employees of sensitive
sources and techniques of intelligence collection. To an extent
the Committee anticipated Colby's recommendation in a provision
of its proposed legislative charter (S. 2525, Sec. 431(a)). This
section penalizes the disclosure of the identity of a CIA employee
serving under cover in a manner which jeopardizes the safety
of that employee. The Committee believes that such a statute
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
would cover the type of unauthorized disclosure recently made
by former CIA employee Philip Agee. Colby, however, suggests
that the sanction be expanded to cover CIA sources as well as
employees and circumstances where political or economic reprisals
could be expected. Although Colby urges protection for
intelligence "techniques," the Committee is extremely hesitant
in going beyond the strict liability coverage already accorded
communications intelligence. Colby himself warned the Committee
of the great difficulties inherent in developing a workable
definition of "technique." Added to the difficulty of legally defining
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : A-RDP81 M00980R000600040017-4
"technique" are the difficulties of proving that any given
disclosure revealed it.
As this report clearly establishes, great difficulties
exist in enforcing current espionage statutes
due to the "gray mail" phenomenon and any new statutes
would face the same problem. Even under the "strict liability"
of Section 798, the law is unclear as to whether the prosecution
would have to establish that the classification of the
material is substantively valid. If so, the government
would face the prospect that much sensitive information
would have to be revealed in the course of litigating that
question.
The modest expansion of the espionage statutes to cover
disclosure of agents under cover warrants serious consideration,
despite the risk of "gray mail." However, the Committee is
not prepared to recommend a major restructuring of those
statutes to encompass all leaks. First, in light of the
experience under Section 798 of Title 18, most members of
the Committee have serious doubts as to whether even a radi-
cal restructuring of the rest of the espionage law along the
lines of the British Official Secrets Act could have an
appreciable impact on leaks. Second, the Committee is unan-
imous in the view that countless practical, legal and political
differences lie in the path of such an undertaking.
What other than criminal sanctions will diminish the
frequency and gravity of leaks? Any comprehensive law
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/24494A-RDP81M00980R000600040017-4
against leaks cannot be effective so long as it is impossible
to distinguish between a criminal act and a widely accepted
governmental practice. Past Executive Orders on classifica-
tion have failed to protect the most important national
security information by providing for the classification of
much information that ought to be made public. Recently,
President Carter promulgated a new order dealing with secrecy
and classification. This new Order is an improvement over
past practices, but if it is not strictly construed and
vigorously enforced, it will foster disrespect for the
whole classification system. In the words of Justice Stewart
in the Pentagon Papers case: "When everything is secret,
nothing is secret." Perhaps the mechanisms contained in the
new executive order will avoid overbroad classification and
will allow for declassifying intelligence necessary to informed
public debate and thus minimize the incentive behind unauth-
orized disclosure of information.*
Yet, given the ingrained nature of the leaks system and
the fact that leaks often result from bureaucratic infighting,
some unauthorized disclosure is bound to continue. To deal
with leaks administrative sanctions are better suited in most
cases than criminal ones because they are more enforceable.
* Of course, such a declassification system must be impartial.
Otherwise, the public will be faced with a biased view and
officials disagreeing with this view would have added incentive
to leak.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-50-
No risk of "gray mail" would exist, because proceedings
could be secret. Due process rights--these might include
a right to present evidence, to be represented by counsel, to
challenge accusations, and to appeal to the courts--must, of
course, be preserved. At the same time, administrative
sanctions would be less onerous. Dismissal or loss of
security clearance are at times more appropriate sanctions
for leaking than criminal prosecutions.
B. Facilitating Enforcement of Existing
Statutes and the Charters
The review of the cases described earlier and the
hearings of the Secrecy and Disclosure Subcommittee have
led the Committee to recommend a program of both administra-
tive and legislative action designed to facilitate enforcement
of the espionage statutes. In essence, on the administrative
side, the Committee recommends a streamlining of decision-
making within the Executive branch on cases where leaks or
espionage occur and the use of administrative
sanctions in less serious breaches of security or other vio-
lations of the law. On the legislative side, the Committee
recommends some new judicial procedures intended to strengthen
the hand of the judge and encourage accommodation between the
defendant and the prosecutor concerning the use of classified
information in litigation -- to seek solutions which encourage
proceeding with prosecution rather than dropping the case
out of fear of disclosure of sensitive information.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : 5IA_RDP81 M00980R000600040017-4
(1) Administrative Recommendations
At the heart of its administrative recommendations (see
pages 62-64)
is the Committee's concern that there is no
effective administrative system currently operating in the
Executive branch for investigating and penalizing unauth-
orized disclosures and the crimes of bribery, perjury and
others described in Part V. Leakers occasionally are pena-
lized on an ad hoc basis.* Violations of the Executive
Order on classification, and even espionage, are not subject
to formal administrative sanction.
In the case of leak investigations the FBI takes the
position that it should not investigate a leak unless there
is clear evidence of a crime. The Committee also believes
that the FBI should not conduct investigations of citizens
for leaks without their consent except in cases involving
a nexus with criminal activity.
But where there is such a nexus, even where prosecution
of the crime is impossible because of the risk of further
disclosures, the FBI should investigate when the leak endangers
sensitive int-elligence sources or methods and is reasonably
believed to violate the criminal statutes of the United States.
* E.g., Donald Stewart, formerly the chief leak investigator
for the Department of Defense, supplied examples of cases during
his tenure when high-ranking military officials received a "slap
on the wrist" for. what appeared to be serious compromises.
Mr. Stewart's prepared statement appears as part of the Sub-
committee's public hearing record.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-52-
The persons investigated should be officials, employees,
or contractors of the executive, legislative, or judicial branch having
access to the information leaked; the investigation and
any intrusive investigative techniques should be authorized
in writing by the Attorney General;* and the investigation
should terminate within 90 days, unless such authorization
is renewed. The Attorney General should submit information
concerning the leak to the head of the employing agency,
or to the President, for appropriate administrative action.
These standards do not go as far as the recommendations
of the Rockefeller Commission (on alleged CIA abuses),
which proposed FBI investigations without evidence of a
crime or the Attorney General's approval. Nevertheless,
they break sharply with current Justice Department policy
foreclosing FBI investigations of damaging criminal leaks
where administrative action, rather than prosecution, is
the intended result.
The Justice Department is properly concerned that such
cases waste time and money because they often turn. out to
be leaks -either formally or informally sanctioned by appropriate
authorities. Nevertheless, where such a leak endangers sensi-
tive sources or methods and violates the criminal statutes
investigation is appropriate.
The Director of Central Intelligence has extraordinary
powers under the 1947 National Security Act, and he and the
Court orders would be required for electronic surveillance
or searches and seizures; such techniques would rarely be
appropriate in most "leak" cases.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIRDP81 M00980R000600040017-4
director of the National Security Agency would have similar
authority under the proposed legislative charters, to dis-
miss their employees. These charters should also recognize
the authority of the directors of CIA and NSA as well as
that of the heads of other agencies to take disciplinary
action against employees who leak classified information.
With that authority should come the implied responsibility
of the agencies to investigate employees' past activities
which would warrant action.
The leak cases reviewed indicate that these initial
investigations are often not conducted because no one official
at the intelligence community level has the authority to
require individual agencies to pursue particular leads in
an investigation. Some intelligence community body should
be required to ensure that individual agencies investigate
activity by intelligence agents, employees or informants
which violates security or charter prohibitions. However,
this investigative responsibility should not be delegated
to the FBI until there is evidence of criminal violations.
As stated, the advantage of administrative sanctions
over criminal prosecution is that procedures under the former
do not require extensive public disclosure of classified
information. Therefore, both the staff of the Committee
and representatives of the Executive branch should explore
what possibilities exist for formalizing and upgrading
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : Cl 42DP81 M00980R000600040017-4
administrative review and investigation procedures for
violations of security and other unlawful acts by intelli-
gence officials. For example, a possible alternative is
an administrative review procedure for employees similar
to courts martial in the military. Officials of the agency
would hear complaints of violations, especially in circum-
stances where the decision has been made to forego criminal
proceedings for national security reasons. These administra-
tive review procedures could be applied to former employees
who violate charter prohibitions, assuming that a deferred
compensation pension plan could be conditioned upon continued
compliance with security and charter requirements. Former
employees who violate prohibitions could be made subject
to loss of pension rights through the administrative pro-
cedure, if it were made clear in the pension agreement that
payments were contingent on such compliance. A decision
to take away pension rights would presumably be reviewed
by the courts to ensure that no former employee's rights
were violated.
Another major goal of the Committee recommendations for
administrative action is to improve accountability in Executive
branch decisionmaking concerning cases involving national
secrets. The Committee agrees with the testimony of Philip
Lacovara before the Secrecy and Disclosure Subcommittee:
I have the sense that the government may be
aborting cases prematurely or unnecessarily because
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 Lq ;RDP81 M00980R000600040017-4
of a failure to press the alternatives to their
fullest, as we did, for example, in the Special
Prosecutor's Office in the Ellsberg break-in
prosecution, where defense efforts to use "national
security threats" to stymie the case were beaten in
the courts.
During the course of the hearings the Subcommittee
members and witnesses agreed on a number of fundamental
points about decisionmaking in these cases. There is
little controversy that the ultimate decision on whether
to proceed on these types of cases must be centralized
within the Attorney General's office. Nevertheless, the
DCI should have authority, through the "sources and methods"
provision of the National Security Act, to make his views
known on whether to halt prosecution of a criminal case.
The Deputy Attorney General and the DCI in testimony
before the Subcommittee agreed that it was up to the Attorney
General, with disputes settled by the President, to decide
whether or not the jeopardy to national secrets in pursuit of
a prosecution outweighs the ends of justice.
If the intelligence community disagrees with an Attorney
General's decision, the DCI or any other agency head should
have the right to appeal to the President. The decision to
drop a national security case should be made in writing by
a high-level official within the Department of Justice, an
Assistant Attorney General or a Deputy Assistant Attorney
General. Included in that written decision should be a
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 :cARDP81M00980R000600040017-4
detailed explanation of the. information which would have
been revealed in the course of trial, why the information
would be revealed, and what damage the disclosure of the
information would have to the national security. The mere
fact that a written record must be made will discourage
thoughtlessly dropping a potential prosecution.
A final area appropriate for administrative action
pertains to the requirement that intelligence agencies
report to the Department of Justice evidence of criminal
activity by employees. As noted in Part VI of this report,
the administration is currently at work attempting to im-
plement provisions of the new Executive Order and has
recently updated the so-called Silberman-Colby understanding
as to the requirements of the intelligence community to
report crimes of its employees to the Department of Justice.
If there is no mechanism through which the Department
of Justice is so notified, the law enforcement process is
likely to break down. The guidelines should be promptly
issued and the Attorney General and the DCI should quickly
determine whether any further revisions are necessary in
the memorandum of understanding on reporting crimes by
employees. It is equally important that either the prospective
guidelines or an expanded memorandum of understanding address
not only criminal activities of intelligence agents, employees
or assets, but also criminal activity known by the intelligence
community which does not involve its employees or assets.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CI -RDP81 M00980R000600040017-4
Such an understanding must consider the protection of
(2) Legislative Initiatives
The purpose of the legislative suggestions set out in
Part VIII is to provide alternatives which will allow pro-
secutors to avoid what one witness described as the "disclose
or dismiss dilemma." Because of ambiguities in existing
judicial procedures or because of a general reluctance on
the part of the intelligence community and the Department
of Justice to take the chance of pursuing these cases, the
administration must decide whether to disclose intelligence
information or to dismiss a criminal case or not pursue an
investigation at the outset.
However, the dilemma posed by the introduction of sensi-
tive intelligence information into criminal cases, especially
at the behest of the defendant, can frequently be avoided
because the information is requested for an irrelevant matter.
For example, Lacovara described to the Subcommittee the
following sequence in the prosecution of the Watergate burg-
lars for the break-in of Dr. Ellsberg's phychiatrist:
After the indictment was returned, the defendants
did in fact demand the production of highly classified
files, including nuclear missile targeting plans. The
defendants were seeking to utilize discovery to obtain
national security information in order to support the
purported defense that they believed the break-in was
justified by national security concerns. The special
prosecutor argued, however, and both District Judge
Gesell and the U.S. Court of Appeals for the District
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-58-
of Columbia Circuit agreed, that the information
sought was irrelevant because "good faith" motiva-
tion was not a valid defense against the crime
charged, conspiracy to violate Fourth Amendment
rights. Thus the difficulty of choosing between
forfeiting an important criminal prosecution or
disclosing information potentially damaging to our
national security was avoided.
In many other cases it is possible that if the prosecutor
had forced the court to carefully examine the relevancy of the
intelligence information to a purported defense or motion, the
judge may well have been forced even under the present standards
of relevancy to decline the request for the information. How-
ever, administration witnesses were reluctant to rely on the
relevancy standard. They argued that what one judge found
relevant another judge would find irrelevant and that many
judges grant the discovery motion first before deciding
whether or not the intelligence information will be used in
the case. Furthermore, defense counsel routinely make sequen-
tial discovery motions which harrass the prosecution and thus
tie up the prosecutors in negotiations with the CIA over
sensitive documents.
Mr. Lacovara suggests that Congress enact an omnibus
pretrial proceeding for use in all cases where classified
exhibits or testimony would be required. The defendant would
be required to put the prosecutor and the court on notice in
advance of trial of all motions he would make requiring
discovery of sensitive classified intelligence information
when he might have reasonably known of the need for discovery
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : e00RDP81 M00980R000600040017-4
prior to trial.* He would have to argue successfully the
relevancy of each motion before the court in order to secure
discovery of the documents or testimony. For the purposes
of argument, the court could assume that the documents
existed without actually providing the defendant the docu-
ments and could decide in advance whether the defense would
be permitted or the motion granted as a matter of law. This
process would be intended to "weed out" irrelevant defenses
and thus simplify prosecution of the case. If at some later
time a new matter arose requiring a special motion or defense
which in turn required the disclosure of secrets, the court
could still entertain an appropriate discovery motion and
both the government and the defendant would be entitled to
an interlocutory appeal.
If such a special omnibus procedure is adopted, the
Committee recognizes that there will be cases where the "weed-
ing out" process will actually arrive at motions and defense
arguments that are relevant and do require the use of intelli-
gence information. At that point the judge must decide two
basic questions: (1) Is the information in question truly
national security information, the disclosure of which would
damage the national security? (2) What action should he
11 Of course, if the prosecution is to go forward, the
government must turn over all materials relevant to the
defense notwithstanding the fact that some of them may be
classified. See, supra, p. 21.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 - UA-RDP81M00980R000600040017-4
take against the prosecution if it withholds the documents
or testimony (e.g., dismissal of the case)? Of course, the
Government always has the option of dismissing a prosecution
if the court's decision on these matters would require what
it believes to be excessive disclosure.*
In 1974 the Supreme Court proposed the Federal Rules
of Evidence. These Rules of Evidence were extremely contro-
versial in the Congress because they contained a provision,
Section 509, that defined a "secret of state" privilege.
An invocation of the privilege by the government would prompt
an in camera adversary proceeding in which the parties
would litigate whether the information in question was in
fact "a secret of state."
Section 509 was rejected by the Congress as it reviewed
the rules proposed by the Supreme Court. However, several
witnesses agreed that perhaps Section 509 might serve as the
basis for an in camera adversary proceeding that would resolve
the use of intelligence information in the course of a trial
after the "weeding out" process described above. Furthermore,
several adjustments to the Section might be made to respond
to criticism which led to congressional rejection in 1974.
For example, the new state secret privilege might more narrowly
define the types of information to which the government could
invoke the privilege. It might give a greater role to the
The Government does not undertake prosecution on a whim.
In deciding to drop an indictment the Attorney General must
weigh the expenditures of time and money in investigation
and prosecution, as well as fairness to the defendant who
must live with the stigma of an unchallengeable indictment.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-61-
court in reviewing the claim of privilege, including authority
to go beyond and behind the classification to determine the
actual damage to the Rational security if the information
were disclosed. It might guarantee the presence of the
defendant and his counsel in the in camera procedure, sub-
jecting both to contempt of court and possible espionage
prosecution if they disclose the results of the procedure.
The primary purpose of such a procedure would be to set
standards to place the prosecution and the government on notice
in advance on what types of information could be subject to
privilege and to give the judge primary responsibility for
administering the privilege. Lacovara in his testimony
emphasized the importance of providing judges with some guidance
as to what action should be taken if they find the privilege
is legitimately invoked. Lacovara suggests a "sliding scale"
of sanctions available to the judge so that "the remedy avail-
able to the defendant would vary depending upon the circumstances
of the case." Lacovara goes on to further describe his pro-
posal as follows:
At one end of the scale, for example, if the
defendant's possible use of the information is
totally speculative, the case simply would continue
without disclosure. At the other end of the scale,
where the information is central to the question
of guilt or innocence and where no other alternative
to public disclosure is possible, dismissal may be
necessary. In between, procedures such as instructing
the jury to assume that the missing information would
have proven a given proposition may be possible.
Certainly the Department of Justice should press
for some intermediate treatment like that before
deciding that the case must be abandoned.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-62-
VIII. RECOMMENDATIONS
The recommendations which follow were formulated by the
Secrecy and Disclosure Subcommittee and are endorsed by the full
Committee. They will serve as an agenda for the Committee as it
proceeds with consideration of legislative charters. The Committee
will be developing specific legislative proposals to implement
these recommendations for inclusion in the charters to be
discussed in the course. of its ongoing hearings. It is the
Committee's hope that the Executive branch will work with the
Committee on these matters and, in particular, on its recom-
mendations for administrative action.
I. At this time Congress should focus primarily upon
developing statutory and administrative procedures
which would facilitate enforcement of the espionage
law and other statutes subject to the "gray mail"
phenomenon. The Committee is not prepared at this
time to recommend a general recasting of the federal
espionage statutes along the lines of the British
Official Secrets Act. However, limited further
protection of intelligence sources, especially the
identities of agents and employees under cover,
appears to be necessary.
II. The Executive branch should interpret the new
Executive Order on security classification with an
emphasis on decreasing the amount of unnecessary
secrecy. The intelligence community, the Intelligence
Oversight Board, and the intelligence committees of
the Congress should declassify as many as possible of
their reports and studies on matters of public concern
to discourage the "leaking" of versions which have not
been sanitized to protect "sources and methods" in-
formation. These reports and studies must be declas-
sified in a disinterested manner, so that the public
receives the true view of a given situation.
III. Administrative procedures for disciplining employees
responsible for violations of security or other laws
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-63-
should be developed. At the same time the intelligence
community should centralize responsibility, perhaps
in the Intelligence Oversight. Board, for investigations
of breaches of security and all violations which do not
constitute crimes. The purpose of these procedures
would be to permit sanctions against employees through
internal agency procedures in which it is easier to
cope with classified documents or testimony than in
traditional public criminal trials. In many leak cases
administrative sanctions may be more appropriate than
a criminal conviction. Of course, these administrative
proceedings would grant due process rights to the'.emp:loyee.
Some consideration should also be given to applying
these administrative review procedures to former
employees through withdrawal of pension rights for former
employees who violate security.*
IV. The FBI should continue to have exclusive responsibility
for investigating criminal violations involving the
intelligence community. In leak cases the FBI should
initiate investigation when:
(1) the leak endangers sensitive intelligence
sources or methods and is reasonably believed
to violate the criminal statutes of the
United States;
(2) the persons investigated are officials, employ-
es, or contractors having access to the
information leaked;
(3) the investigation and any intrusive in-
vestigative techniques are authorized in
writing by the Attorney General;**
(4) the investigation terminates within 90 days,
unless such authoriztion is renewed; and
(5) the Attorney General submits information
concerning the leak to the head of the
employing agency, or to the President, for
appropriate administrative action.
V. The Attorney General should issue guideline`szunder the
authority of Executive Order 12036 on the responsibility
of the intelligence community to report crimes to the
Department of Justice. The guidelines should cover
reporting of all activity in violation of U.S. laws
coming to the attention of the intelligence community,
but must consider protection of sensitive sources and
methods.
For discussion of the Committee's rationale for recommen-
dations III, IV, V and VI, see pp. 51-57, supra.
Court orders would be required for electronic surveillance
or searcrWr?QOI'A51tiIR4~1o998Db0~e4'(IEf1ja0fi4
appropria in most eak cases.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
VI. The Attorney General should issue regulations that
are binding upon all departments of the government
which set out the procedures whereby agencies of the
intelligence community are to provide necessary
information to attorneys of the Department of Justice
to proceed with a criminal investigation or prosecu-
tion. The regulations should also set out how the
decision is to be made not to proceed in national
security cases and who is authorized to make such a
decision. These regulations should require that any
such decision be made in.writing, and the decision
paper should include the precise intelligence infor-
mation which would have been disclosed in the course
of the trial, why the official believes it would have
been disclosed, and the damage the information would
have to the national security if the case proceeds.
The decision paper should be available to the intel-
ligence oversight committees of the Congress and such
cases should be reported to the committee annually
or as required.
VII. Congress should consider the enactment of a special
omnibus pre-trial proceeding to be used in cases where
national secrets are likely to arise in the course of
a criminal prosecution. The omnibus procedure would
require the defendant to put the prosecution and the
court on notice of all motions or defenses or
arguments he intended to make which would require the
discovery and disclosure of intelligence information
or the use of intelligence community witnesses. The
judge would be required to rule in advance of the
trial on the admissibility of the intelligence in-
formation and on the scope of witnesses' testimony
as well as the general relevancy of the motion or
defense prior to granting discovery of any intel-
ligence information to the defendant. On the other
hand, the defendant would be permitted a discovery
motion during the course of trial if the
prosecution presents a matter not originally suggested
by indictment or for which the defendant could not
fairly have been expected to be on notice at the time
of the omnibus procedure.*
For a discussion o the Committee's rationale for
recommendations VII and VIII,-see pp. 57-61, supra.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
VIII. The Congress should reconsider the secret of state
privilege proposed by the Supreme Court in 1974.
That privilege needs to be considerably revised
along the lines described above but at a minimum
should provide for an in camera adversary pro-
cedure on the privilege,define the scope of the
privilege, the standards for its invocation, provide
increased judicial authority for its procedural
administration, and provide a sliding scale of
sanctions available to the judge in the case where
the privilege is successfully invoked.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
, Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Additional Views of Adlai E. Stevenson
October 6, 1978
The Report on Secrecy is a positive step, as far as it
goes, but in my view it unnecessarily stops short of tackling
the need for new laws to cover espionage and other unauthorized
disclosures.
The Report emphasizes improvement of procedures to enforce
existing laws -- which could aid enforcement of new laws as
well. However, I believe the Senate Select Committee on Intel-
ligence should recommend criminal laws that will address the
threat to national security from breaches of security.
The laws affecting espionage are sparse. Those which
treat leaks are almost non-existent. Grave damage can be done
to the national security through purposeful leaks which may
violate no statute.
New laws should not relate only to information derived
from intelligence sources and methods; sensitive information
derives from many sources. A State Department or White House
official could improperly release information just as damaging
as information which happens to be labeled "intelligence." The
law should be drawn accordingly.
The Report's brief mention of the court martial system
deserves fuller consideration to see if it could legally
serve as a model for civilian officers who handle government
secrets. The uniform code of military justice provides a
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
system whereby errant members can be disciplined without
breaching their constitutional rights. Any such arrangements
could, of course, provide for appeal procedures and congres-
sional oversight. It might also be possible to require officers
entrusted with government secrets to enter contractual arrange-
ments by which they would agree to submit to special
disciplinary procedures should they violate their contractual
and legal obligations.
I urge the Committee to address the inadequacy of
existing laws. It is not enough to support the improved
enforcement of laws which do not exist.
67
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
THE SEPARATE VIEWS OF SENATOR MALCOLM WALLOP
The Committee's report amply documents the quandary
variously known as "disclose or dismiss" or as "Grey
Mail." Because prosecution requires disclosure of information
likely to compound damage to the nation, leakers and spies
have been allowed to go unpunished. The sensitive substance
of the documents stolen or leaked must be discussed in open
court because the court cannot assume that the documents were
properly classified in the first place. Therefore, the
courts have refused to enforce Sec. 793, Title 18 U.S.C.,
intended to punish anyone for revealing classified information,
without, first determining what the proper classification of the
information should have been. Courts cannot avoid this function,
because the classifying authorities cannot be deemed infallible
or disinterested. Only an.impartial determination of the proper
classification can form the basis for judicial punishment of
leakers and spies.
The Committee's analysis of the possible ways out of the
quandary is circumscribed by three very firm facts. (1) The
Constitution requires open trials. (2) Prosecution of spies
or leakers often requires evidence the disclosure of which in
open court would do more harm than the prosecutor's success
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
6 1
would do good. (3) Civil libertarians, not unjustly, are afraid
of making mere disclosure of classified information a strict --
liability crime because information is often classified improperly.
Thus, it is not surprising the report can point the way to
only marginal improvements in our ability to enforce laws
safeguarding secrets. These improvements would be effected by
conferring greater powers on judges to exclude certain evidence
from espionage trials, and greater reliance on administrative
sanctions to curb leaking by the government's present or former
employees. Yet the former may well affect the fairness of trials,
while the latter would surely provide for the non-judicial
execution of penalties weightier than those meted out in most
judicial proceedings. These may be excessive prices for such
modest improvements.
A somewhat different line of analysis, however, can lead
us to a solution at once much more efficacious against "disclose
or dismiss," at least as respectful of civil liberties, and
patently fairer than the solution advanced by the report. In
brief: One need not alter the Sixth Amendment's guarantee of
public trials, and one need not disclose classified information at
public trials if the only question to be decided at such trials
is whether the accused did or did not disclose classified
information to unauthorized persons. But the question "did he
unlawfully disclose" is logically independent of the one
regarding the effect of the disclosure. Penalties need be
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
imposed only if it is determined that the disclosure caused
or could cause harm to the United States. The question of
harm done -- as distinguished from the questions of guilt and
innocence and of motive -- could be tried in camera by a judge,
with or without a cleared jury, and with cleared attorneys,
without violating the letter or the spirit of the Sixth
Amendment.
-- The two questions "did he do it" and "what harm did
it do" are logically separate. Heretofore our judicial system
has mixed them. Unless a statute is enacted to provide for their
separate resolution, we will find no solution to the quandary
"disclose or dismiss."
Did he do it?
It is useful to start from the fact that officials of the
executive branch who classify information often do it erroneously
and sometimes maliciously. Nevertheless, information does exist
the disclosure of which would harm the country. Moreover, the
people who work with it, no matter how imperfect their minds and
motives, cannot help but be charged with the task of deciding
which information deserves special protection and which does not.
Their decision should not be final, but neither should the law
regard it as merely one opinion among others. Were the law to
give its judgment no special weight, the executive branch could
70
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
'1/
not lawfully keep reporters from TASS out of the Pentagon's
war room.
Of course the executive branch's judgments on classification
must be open to challenge in court. But, until such challenges are
upheld, the executive must have the right to operate on a day-to-day
basis as if its judgments were correct. That requires, at a
minimum, that the judicial process be allowed to determine
whether a given individual did or did not handle classified
information in a manner deemed unlawful by the executive branch.
In recent years, the judiciary has not been able to make
such determinations, because it has mixed the question "did he
do it" with the question of whether the information involved
was properly classified. The latter question is essential,
but it is separate.
What harm did it do?
Civil libertarians are correct in stating
that information is often improperly and sometimes maliciously
classified, and that those who bring it into the public domain
deserve praise. But no one disputes that some unauthorized
disclosures are harmful. No one should object to determining
whether and to what extent any particular disclosure was harmful.
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
1
We should object to accepting uncritically the intelligence
agencies' own assessment of the harm done. Such assessments
are the bases of the prosecution's case, and the chief targets
for the defense. It follows therefore that an impartial court
must decide between adversary presentations on the question of
harm done.
But it does not follow that courts must decide the question
of harm done in public. This question can and should be
answered without any reference to the identity of the person(s)
suspected of disclosing the classified information, or to their
motives. The resolution of this question cannot in any way be
considered the trial of a person. Therefore, the Sixth Amend-
ment's guarantee of a public trial, which refers to trials of
persons does not apply to this question.
There are several ways for the judicial system to decide
such questions in camera. Grand juries routinely and secretly
decide questions of fact, often examining evidence and arguments
from varied sources. Perhaps major unauthorized disclosures of
classified information could be brought before grand juries, which
would issue their findings concerning the harm to be expected
from the disclosure without even knowing the identity of any
suspect(s). This would not constitute a secret trial because although
the grand jury's decision would ultimately affect anyone found guilty
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
of disclosing the information in question, it would affect
any such person equally. Its evaluation would prescind
entirely from the accused's identity. The grand jury's
assessment of harm done would be immune to the defense's
challenges in an open trial for leaking or espionage. Indeed
the question of harm done would be immaterial. Then, assuming
such a trial resulted in conviction, the judge would turn to
the grand jury's report to decide whether the defendant deserved
a token sentence and contratulations for having served his
country, or severe punishment for having endangered his fellow
citizens, or any sentence in between.
One could object to the above procedure by maintaining that
the interest of anyone accused of unlawful disclosure in the
assessment of damage done is so great that no such assessment
can be considered legally authoritative unless the defendant
has had the opportunity to take part in the selection of the
jury and in the arguments before it. A corollary of this
objection is that courts may not give opinions outside the
framework of "cases and controversies." These objections could
be met by having the question of harm done decided in camera
by the same jury which had tried the case of unlawful disclosure
in open court, and by allowing the accused and his attorneys to
compel and present whatever evidence they and the court deemed
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-7-
relevant. Of course all parties to the proceedings would be
sworn to secrecy and liable to severe penalties for violating
it. Such a proceeding would not constitute a secret trial
because, although the convict would have an interest in its
outcome, the jury's decision on the harm done by the disclosure
would depend not at all on what the convict had done, or his motives,
or on his person. It would depend only on the qualities
inherent in the information disclosed. These qualities would
be on trial, not the person.
The Question of Harm
The question before us is how to punish those who harm
their fellow Americans by unauthorized disclosures, and how to
do so without infringing on the Constitution or civl liberties.
The distinctions between leaks and espionage, between good and
bad motives, between the release of substantive information and
information regarding sources and methods are all of secondary
importance. In fact, when disclosure cases from each of these
different categories have come to trial, the proceedings have
turned on one paramount question: "What harm did this do?" This
is true even of cases under Section 798, U. S. Code (the statute
protecting communications intelligence). And it seems reasonable
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
ii
that this be so. Of course, the centrality of this question is
the source of the legal quandary "disclose or dismiss."
Lately, attempts have been made to sidestep the question of
harm done in order to make punishment of disclosures easier
without recourse to an "official secrets act." The essence of
these attempts has been to establish categories of disclosure
which ipso facto result in harm to the United States. The release
of substantive information may or may not do harm, while presumably
the disclosure of intelligence sources and methods, as well as
of intelligence operatives, is always harmful. Thus the proposals
to make the disclosure of CIA employees, or of sources and methods,
into strict liability crimes. But they will not work. No doubt
disclosure of such information is harmful and should be punished.
But why give these categories special attention and a higher
likelihood of punishment and not to others (e.g., the location of
SSBNs) the disclosure of which would be even more harmful? The
only answer, that it is legally easier to do so, is unsatisfactory.
The ruling criterion for punishment is, and must remain, harm.
Administrative Sanctions
Another attempt to handle the problem without recourse to
strict liability for unlawful disclosure is reliance on adminis-
trative sanctions by agencies against offending employees or
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
against the pensions of former employees. But such sanctions are
indefensible except as punishment for exposing the country to
harm. Should any employee of the United States be punished
administratively except for having done harm to his country?
We rightly answer "no" because agencies' rules are not made for
the agencies' heads' convenience, but for the good of the
country.
Leaking by senior officials has become a part of our
government's folkways -- a dangerous and unfair part. Senior
officials can now punish junior ones very seriously by withdrawing
their security clearances ostensibly for unauthorized disclosure
of classified information, while they themselves disclose it
without authorization but with impunity. Selective leaks of
intelligence by senior officials is the most common and most
dangerous means by which the CIA interferes in American politics.
Today administrative sanctions are liable to the charge of
arbitrariness. However, as we impose legal penalties upon
leakers, we must not hinder the President, and senior officials
designated by him, from wielding classified information, selectively
and covertly as a weapon in the nation's arsenal. Ultimately,
nothing; can prevent a President from wielding such information
to his partisian advantage except a public opinion that is well-
informed and fairminded. But the law can control employees
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
-10-
of the United States, regardless of rank, who are not
following the Pr'esident's, direct orders. If such an
employee does harm to his country by leaking, even for
the best motives, why shouldn't he be punished by the law?
To put this from another perspective: Why should the legal
system countenance the imposition of very heavy penalties
(dismissal and loss of clearance for a career intelligence man
are worse than jail) except for harm done to the United States?
Does not the establishment of a non-judicial system for meting
out such punishment mean a possible detour into corporativism --
a destination far more fearsome than an official secrets act?
Conclusion
In short, the Government may keep information out of the
public domain only if its possession by our enemies would harm
the United States. Now and again, Government officials will
err, sometimes maliciously, in classifying information. But
they must have the right to classify, and the fact of unlawful
disclosure must be legally ascertainable. The key question of
harm done -- on which punishment depends -- is both separate and
not subject to the requirements for an open trial. That is because
a trial of the question of harm done does not determine the guilt
or innocence of any person, but rather the impersonal effect of an
11
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
it
action. Since the finding concerning this effect would bear
upon the sentence imposed upon a person, that person ought to
have the privilege of being present, of having counsel, of
compelling witnesses, etc. But nothing in our Constitution,
laws, or indeed in common sense, argues that these proceedings
ought to be open. Parties to such proceedings must be held to
secrecy, even as grand juries are.
It is immaterial whether the determination of harm done
takes place before or after the trial of the person accused of
unlawful disclosure. It is essential that this determination
be made in secret -- so that it may be done with the
benefit of all relevant information and without danger to the
country,. It is equally important that the determination be made
judicially -- that is, by an impartial judge and jury, with
adversary counsel -- for the sake of accuracy and of legal validity.
Soon the country will witness the attempt of our legal
system to try Mr. Kampiles for allegedly selling the technical
manual of the KH-11. Did he do it? That will be easy enough
to prove or disprove. But his attorney will ask, legitimately,
just how much harm the disclosure did. Any accounting of harm
must weigh our knowledge of any countermeasures the Soviets may
have taken since the disclosure, against the information on the
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
KH-11's targets provided by other sources. That kind of
evaluation -- most relevant to the question of harm done --
simply cannot be discussed in public. So, if we hamstring
our legal system by imprudently mixing two questions which
are logically separate ("Did he do it?" and "What harm did
it do?"). We will, irresponsibly, have to conclude we cannot
try Mr. Kampiles. And if we can't try him, whom can we try?
The country could not accept imprudent adherence to
unsound doctrine as an excuse for such irresponsibility.
This Committee should consider legislation to:
(1) establish stricter guidelines for classifying
information as important to the nation's security.
(2) establish procedures for releasing classified
information to the public.
(3) make the unauthorized disclosure of classified
information a strict-liability crime, to be tried in open
court with full constitutional guarantees.
(4) make punishment for the crime of unauthorized
disclosure vary between purely nominal (e.g., one dollar
fine) and heavy penalties, depending on the disclosed
information's importance to the nation and to the harm
to be reasonably expected from the disclosure.
101
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4
(5) vest the right to determine harm done by the
disclosure in the trial jury, meeting in camera, subject
to security clearance and bound to secrecy. The attorneys
would be similarly cleared and bound.
(6) ensure that the two questions be handled
separately in the appeals. process and that the question
of harm done continue to be decided in camera.
DU
Approved For Release 2004/05/21 : CIA-RDP81 M00980R000600040017-4