SECURITY CLASSIFICATION POLICY AND EXECUTIVE ORDER 12356 TWENTY-NINTH REPORT BY THE COMMITTEE ON GOVERNMENT OPERATIONS

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP83M00914R001800100054-7
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 52
• Document Creation Date: December 20, 2016
• Sequence Number: 54
• Content Type: REPORT

Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
SECURITY CLASSIFICATION POLICY AND
EXECUTIVE ORDER 12356
TWENTY-NINTH REPORT
BY THE
COMMITTEE ON GOVERNMENT
OPERATIONS
AuauST 12, 1982.-Committed to the Committee of the Whole House
on the State of the Union and ordered to be printed
U.S. GOVERNMENT PRINTING OFFICE
97-2440 WASHINGTON : 1982
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
COMMITTEE ON GOVERNMENT OPERATIONS
JACK BROOKS, Texas, Chairman
L. H. FOUNTAIN, North Carolina FRANK HORTON, New York
rANTE B. FASCELL, Florida
BENJAMIN S. ROSENTHAL, New York
DON FUQUA, Florida
JOHN CONYERS, JR., Michigan
CARDISS COLLINS, Illinois
JOHN L. BURTON, California
GLENN ENGLISH, Oklahoma
ELLIOTT H. LEVITAS, Georgia
DAVID W. EVANS, Indiana
TOBY MOFFETT, Connecticut
HENRY A. WAXMAN, California
FLOYD J. FITHIAN, Indiana
TED WEISS, New York
MIKE SYNAR, Oklahoma
EUGENE V. ATKINSON, Pennsylvania
STEPHEN L. NEAL, North Carolina
DOUG BARNARD, JR., Georgia
PETER A. PEYSER, New York
BARNEY FRANK, Massachusetts
HAROLD WASHINGTON, Illinois
TOM LANTOS, 'California
BARBARA B. KENNELLY, Connecticut
JOHN N. ERLENBORN, Illinois
CLARENCE J. BROWN, Ohio
PAUL N. MCCLOSKEY, JR., California
THOMAS N. KINDNESS, Ohio
ROBERT S. WALKER, Pennsylvania
If. CALDWELL BUTLER, Virginia
LYLE WILLIAMS, Ohio
JOEL DECKARD, Indiana
WILLIAM F. CLINGER, JR., Pennsylvania
RAYMOND J. McGRATH, New York
HAL DAUB, Nebraska
JOHN HILER, Indiana
WENDELL BAILEY, Missouri
LAWRENCE J. DENARDIS, Connecticut
JUDD GREGG, New Hampshire
MICHAEL G. OXLEY, Ohio
WILLIAM If. JONES, General Counsel
JOHN E. MooRE, Staff Administrator
JOHN M. DUNCAN, Minority Staff Director
GOVERNMENT INFORMATION AND INDIVIDUAL RIGHTS SUBCOMMITTEE
GLENN ENGLISH, Oklahoma, Chairman
TED WEISS, New York THOMAS N. KINDNESS, Ohio
HENRY A. WAXMAN, California JOHN N. ERLENBORN, Illinois
JOHN L. BURTON, California WENDELL BAILEY, Missouri
JOHN CONYERS, JR., Michigan
EX OFFICIO
FRANK HORTON, New York
ROBERT GELLMAN, Counsel
EUPHON METZGER, Clerk
JOHN J. PARISI, Minority Professional Staff
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
LETTER OF TRANSMITTAL
HOUSE OF REPRESENTATIVES,
Washington, D.C., August 12, 19892.
Hon. THOMAS P. O'NEILL, Jr.,
Speaker of the House of Representatives,
Washington, D.C.
DEAR MR. SPEAKER : By direction of the Committee on Government
Operations, I submit herewith the committee's twenty-ninth report to
the 97th Congress. The committee's report is based on a study made by
its Government Information and Individual Rights Subcommittee.
JACK BROOKS,
Chairman.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
CONTENTS
1. Introduction Page
---------------- ---------------------- 1
Il. -----------
Historicalbackground4
A--------------------. Early secrecy requirements ------------------------------ 4
B. Presidential actions6
--------------------
C. Executive Order 101047
D. Executive Order 10290
E8
. Executive Order 10501____________________________ 8
F. Executive Order 11652
9
G. Executive Order 12065________ ------ 11
III. Analysis of major features of Executive Order 12356_______________ 13
A. What is subject to classification?_________________________ 13
B. Classification authority and procedure--------------------
_ _ _ _ _ _ _ _ _ _ 20
C. Declassification ------------------ ----------------------
---_ 24
IV. Discussion, findings, and recommendations_______________________ 27
A. Adoption of Executive Order 12356_________________ ___ 27
1. Findings 34
2. Recommendations__________________ 35
B. Reagan administration justification for the changes in the
Executive order on security classification________________ 36
1. Finding----------------
2. Recommendation________________________________ 42
C. Overclassification--------------------------------------- 1. Findings---------------------------------------- 49
2. Recommendation________________________________ 49
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
Union Calendar No. 451
97TH CONGRESS i HOUSE OF REPRESENTATIVES J REPORT
2d Session j No. 97-731
SECURITY CLASSIFICATION POLICY AND
EXECUTIVE ORDER 12356
AuGuST 12, 1982.-Committed to the Committee of the Whole House
on the State of the Union and ordered to be printed
Mr. BROOKS, from the Committee on Government Operations,
submitted the following
TWENTY-NINTH REPORT
BASED ON A STUDY BY THE GOVERNMENT INFORMATION AND
INDIVIDUAL RIGHTS SUBCOMMITTEE
On August 10, 1982, the Committee on Government Operations ap-
proved and adopted a report entitled "Security Classification Policy
and Executive Order 12356." The chairman was directed to transmit
a copy to the Speaker of the House.
I. INTRODUCTION
For over 25 years, the Committee on Government Operations has
held hearings, issued reports, and conducted general oversight on in-
formation policies and practices of federal departments and agencies.,
Much of this work began with the establishment of the Special Sub-
committee on Government Information during the 84th Congress. The
Subcommittee was asked to "ascertain the trend in the availability
1 See, e.g., Committee on Government Operations, "Availability of Information from Fed-
eral Departments and Agencies," H.R. Rep. No. 2947, 84th Cong., 2d Sess. (1956) ; Com-
mittee on Government Operations, "Availability of Information from Federal Departments
and Agencies (Scientific Information and National Defense)," H.R. Rep. No. 1619, 85th
Cong., 2d Sess. (1958) ; Committee on Government Operations, "Availability of Informa-
tion from Federal Departments and Agencies (Department of Defense)," H.R. Rep. No. 1884,
85th Cong., 2d Sess. (1958) ; Committee on Government Operations, "Availability of Infor-
mation from Federal Departments and Agencies (Progress of Study, February 1957-July
1958)"; H.R. Rep. No. 2578, 85th Cong., 2d Sess. (1958) ; Committee on Government Op-
erations. "Availability of Information from Federal Departments and Agencies (Progress
of Study, August 1958-July 1959)," H.R. Rep. No. 1137, 86th Cong., 1st Sess. (1959) ;
Committee on Government Operations, "Availability of Information from Federal Depart-
ments and Agencies (The First Five Years and Progress of Study, August 1959-July 1960),"
H.R. Rep. No. 2084. 86th Cong., 2d Sess. (1960) ; Committee on Government Operations,
"Availability of Information from Federal Departments and Agencies (Progress of Study,
July-December 1960)." H.R. Rep. No. 818, 87th Cong., 1st Sess. (1961) ; Committee on Gov-
ernment Operations, "Availability of Information from Federal Departments and Agencies
Sesr (19s8 )f Study, September 1961-December 1962)," H.R. Rep. No. 918, 88th Cong., 1st
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
of government information and [to] scrutinize the information prac-
tices of executive agencies and officials in light of their propriety, fit-
ness and legality." 2 One of the products of the. Committee's efforts
was the Freedom of Information Act.3
The classification of information by federal agencies in the name of
national security has been a regular subject of review by the Govern-
ment Operations Committee. Many Committee reports discuss aspects
of classification policy and practice, and there are two earlier Commit-
tee reports that were devoted exclusively to security classification is-
sues. In 1962, the Committee issued a report 4 on the stwtus of Executive
Order 10501,5 the Executive Order on Security Classification that was
issued in 1953 by President Eisenhower and updated at various times
during the Kennedy Administration.
A second security classification report 6 was issued in 1973, approxi-
mately one year after President Nixon issued Executive Order 11652.7
The report examined the interconnection between the Executive Order
on Security Classification and the first exemption of the Freedom of
Information Act. The report also reviewed the Nixon security classi-
fication order as well as other aspects of classification policy.
In 1977, the Subcommittee on Government Information and Indi-
vidual Rights held hearings on the security classification exemption to
the Freedom of Information Act while the Carter Administration
was preparing Executive Order 12065.$ When the Carter Administra
tion circulated a draft order for comment, the Chairman of the Sub-
committee submitted comments e
A revised Executive Order on Security Classification was issued by
President Reagan in 1982, and this report reviews the procedures
under which that order was adopted as well as the changes made by
the Reagan order. Hearings on the Reagan order were held on March
10, 1982, and May 5, 1982.16
Throughout all of the oversight on information policy issues con-
ducted by the Government Operations Committee, the need for the
protection of some government information from public disclosure
has never been questioned. The Committee has only sought to keep
government secrecy to the minimum necessary for the operation of a
democratic government in the complex world. This report has been
prepared with this goal in mind.
While some of the Committee's earlier work on security classifica-
tion issues has been superseded by improvements or changes in the
security classification process, many of the Committee's earlier find-
2 Letter from Rep. William L. Damson, Chairman, House Committee on Government Op-
erations reprinted in Committee on Government Operations, "Replies From Federal Agen-
cies to Questionnaire Submitted by the Special Subcommittee on Government Information,"
84th Cong., 1st Sess., p. iii (1955) (Committee Print).
3 5 U.S.C. ? 552 (1976).
* Committee on Government Operations, "Safeguarding Official Information in the Inter-
ests of the Defense of the United States (The Status of Executive Order 10501)," H.R. Rep.
No. 2456, 87th Cong., 2d Sess. (1962).
5 See text accompanying notes 37-44.
6 Committee on Government Operations, "Executive Classification of Information-Se-
curity Classification Problems Involving Exemption (b) (1) of the Freedom of Information
Act (5 U.S.C. 552)," H.R. Rep. No. 93-221, 93d Cong., 1st Sess. (1973) [hereinafter cited
as "1973 Government Operations Security Classification Report"].
See text accompanying notes 45-33.
"Security Classification Exemption to the Freedom of Information Act," Hearing be-
fore a Subcommittee of the House Committee on Government Operations, 95th Cong., 1st
Sess. (1977).
Id. at appendix 4.
io Executive Order on Security Classification," Hearings before a Subcommittee of the
House Committee on Government Operations, 97th Cong., 2d Seas. (1982) [hereinafter
cited as "Hearings"].
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
ings, conclusions, and recommendations are still valid today. This
report does not replow old ground. Instead, it focuses on the new
Executive Order on Security Classification and the issues raised by
its adoption.
The major findings of the report center on the failure of the Reagan
Administration to fully inform the Congress and the public about the
proposals to change security classification rules or to solicit advice at
a meaningful time during the revision process. In addition, the report
concludes that the Reagan Administration failed to identify clearly
the problems with security classification rules that the new order was
intended to solve. The Administration also failed to explain in a
satisfactory manner the purpose of the changes that were made. Fi-
nally, the report finds that overclassification of information continues
to be a serious problem and that the new Executive Order offers noth-
ing that will address the overclassification problem.
The report recommends more openness when security classification
rules are adopted or changed by the executive branch. This includes
full notice to the Congress and the public, written findings in sup-
port of proposed changes, and a timely opportunity for public and
congressional comment on proposed rules. The report also recom-
mends that changes in security classification rules be made only when
concrete problems with the protection of sensitive information have
been identified.
Finally, to deal with the problem of overclassification, the report
recommends that the National Security Council, the Information Se-
curity Oversight Office, and the Office of Management and Budget
work together to develop and apply methods of limiting abuse of
classification authority.
The Executive Order on Security Classification sets the basic rules
for the classification of information by Government agencies. The
order is the central document establishing security classification policy.
This is why it has been the subject of review by the Government Op-
erations Committee for so many years.
However, Government classifiers are influenced by other things be-
sides the Executive Order. The report of the Information Security
Oversight Office 11 for the years 1980-81 notes that agency classifica-
tion activity was influenced by the change in government in Iran, the
Iranian hostage situation, events in Afghanistan, the Middle East,
and other parts of the world. World events can have a direct impact
on classification decisions 12 The overall attitude of government policy
makers toward secrecy can also have an impact as can oversight ac-
tivities of both the executive and legislative branches.
Because of all of these different influences on classification activities,
it is difficult to measure directly the effect of a new Executive Order.
In addition, classifiers appear to have significantly increased discre-
tion in making classification decisions as a result of the vague language
in the new order.
At hearings, Administration spokesmen offered narrow interpreta-
tions for much of the broad language contained in the order. Although
serious questions about the credibility of these interpretations remain,
n The status of the Information Security Oversight Office is discussed at text accom-
panying notes 201-202.
12 Information Security Oversight Office, "Annual Report to the President 1980-81"
at 12.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
it is nevertheless possible that the new classification rules will be ap-
plied narrowly and judiciously. It remains to be seen whether Govern-
ment classifiers will take their cue from the seemingly broad new
authority in the order or from the narrow interpretations offered to
this Committee by the Administration.
Given the past abuses of classification authority and the consistent
pattern of overclassification by the executive branch, the Committee
is not optimistic that classifiers will apply the new classification au-
thority with restraint.
II. HISTORICAL BACKGROUND 13
A. EARLY SECRECY REQUIREMENTS
The practice of assigning a secret status to sensitive government
records can be traced back to the early days of the United States.14
Throughout this country's history, records have occasionally been
designated confidential either with or without direct statutory author-
ization for such an action. However, formal military secrecy proce-
dures or regulations only appeared after the Civil War. The earliest
War Department security-secrecy orders are dated 1869.15
The initial Army General Order of 1869 on security-secrecy per-
tained to the physical protection of forts preventing them from being
photographed or their layout being depicted without authorization.
This limited objective passed through a series of metamorphoses and,
shortly after the United States entered World War I, had evolved
into a fully developed information security classification system. By
then, the Navy also had a directive on this matter. Adherence to these
regulations was reinforced not only by armed forces disciplinary
penalties, but also by criminal law.
ve See generally H. Relyea, "The Evolution of Government Information Security Classi-
fication Policy : A Brief Overview (1775-1973)," reprinted in "Security Classification Re-
form," Hearings on H.R. 12004 before a Subcommittee of the House Committee on Gov-
ernment Operations, 93d Cong., 2d Sess. 505-97 (1974); 1973 Government Operations
Security Classification Report, supra note 6, at 3-11.
The Committee wishes to acknowledge the assistance of Dr. Harold Relyea, Congressional
Research Service Specialist in American National Government, in the preparation of this
part of this report.
14 Secrecy issues can be traced back at least as far as the 'Continental Congress. A recent
Freedom of Information Act case discussed some early concerns : "Our nation's earliest intel-
ligence activities were carried out by the Committee of Secret Correspondence of the Con-
tinental Congress. The Continental Congress created the Committee on 29 November 1775 to
'correspond with our friends in Great Britain, Ireland and other parts of the world,' and
Congress resolved to provide for expenses incurred by the Committee in sending 'agents' for
this purpose. In carrying out these duties, the Committee placed great importance upon
secrecy. In reference to information from its agent Arthur Lee, describing French plans to
send arms and ammunition to the Continental Army, the Committee stated : 'Considering
the nature and importance of it, we agree in opinion that it is our indispensible duty to
keep it a secret, even from Congress. . We find by fatal experience, the Congress con-
sists of too many members to keep secrets.'
"The Committee exercised broad discretionary power to conduct intelligence activities in-
dependent of the Continental 'Congress and to safeguard the secrecy of matters pertaining to
its agents, though Congress asserted greater direct control following the Declaration of
Independence. It is especially remarkable that the Committee was in a position to insist
upon secrecy even against Congress, which functioned both as the legislative and the execu-
tive power at this time and exercised control over foreign affairs.
The importance of total secrecy in intelligence matters v-as appreciated in this era at
the highest levels. In a letter of 26 J7'ly 1777 issuing orders for an intelligence mission, Gen-
eral Washington wrote to Colonel Elias Dayton : 'The necessity of procuring good intelli-
gence is apparent and need not be further urged. All that remains for me to add is that you
keep the whole matter as secret as possible. For upon secrecy, success depends in most
enterprises of the kind, and for want of it, they are generally defeated .' ' Halperin v.
Central Intelligence Agency, 629 F.2d 144, 157 (D.C. Cir. 1980) (footnotes omitted).
v Relyea, supra, note 13, at 11.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
The Espionage Act of 191716 and a predecessor statute of 191117
prohibiting the unauthorized disclosure of national defense secrets
were directed at persons engaging in spying. Neither law specifically
sanctioned the information protection practices of the War Depart-
ment or the armed forces, and the orders and directives of these en-
tities were not promulgated pursuant to these statutes.18 The markings
and controls prescribed for the use of the military were designed for
utilization in conjunction with internal communications and docu-
ments. However, Navy security regulations of 1916 stated : "Officers
resigning are warned of the provision of the national defense secrets
act.' 19 This suggested that former naval personnel could not publicly
reveal information that had been protected under Navy regulations
without subjecting themselves to possible prosecution. Such a disclos-
ure might have been pursued under the 1911 secrets law noted above,
not the Navy's directives on the matter. This situation illustrates how
armed forces regulations pertaining to the protection of information,
although not issued in accordance with a "secrets" statute, enjoyed the
color of statutory law for their enforcement.
Shortly after the American entry into World War I, Congress pro-
vided authority for the Commissioner of Patents 20 or the President 21
to keep patent applications secret if publication of the applications
might "be detrimental to the public safety or defense, or may assist
the enemy or endanger the successful prosecution of the war." No
classification marking arrangement was devised for this action. Quite
the contrary, the means provided for maintaining this secrecy was to
withhold the grant of a patent until the termination of the war. This
appears to be the first direct statutory grant of authority to the Chief
Executive to declare a type of information secret.22
Armed forces regulations governing the creation and protection of
secrets were continued and expanded after World War I. By 1936,
Army instructions on the application of secrecy markings seemed to
embrace foreign policy material and what might be called "political"
data. A "Secret" designation was to be applied to information "of
such a nature that its disclosure might endanger the national security,
or cause serious injury to the interests or prestige of the Nation, an
individual, or any government activity. or be of great advantage to
a foreign nation." Similarly, information would be "Confidential"
if "of such a nature that its disclosure, although not endangering
the national security, might be prejudicial to the interests or prestige
of the Nation, an individual, or any government activity, or be of
advantage to a foreign nation." The term "Restricted" might be used
in instances where information "is for official use only or of such a
1040 Stat. 217 ('1917) (codified at 18 U.S.C. ?? 793-4 (1976)).
17 36 Stat. 1084 (1911).
See H. Edgar and B. Schmidt, "The Espionage Statutes and Publication of Defense
information,' 73 Columbia Law Review 929-1087 (1973).
19 Changes in Navy Regulations and Naval Instructions No. 7 (Sept. 15, 1916).
20 40 Stat. 394 (1917).
2140 Stat. 422 (1917).
22 Current patent secrecy provisions may be found at 35 U.S.C. ?? 181-188 (1976) and
50 U.S.C. App. ? 10(i) (1976). For a complete discussion of invention secrecy, see House
Committee on Government Operations, "The Government's Classification of Private Ideas,"
H.R. Rep. No. 96-1540, 96th Cong., 2d Sess. 1-62 (1980).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
nature that its disclosure should be limited for reasons of administra-
tive privacy, or should be denied the general public." 23
The outstanding characteristic of these provisions is the broad dis-
cretionary authority they appeared to confer with regard to the scope
of application. Initial security-secrecy regulations were designed only
to safeguard fort and coastal defense facility information. The
extended applicability in 1936 to almost any area of governmental
activity was without any stated reason or authority.
Such regulations were promulgated without direct statutory author-
ization, other than departmental "housekeeping" laws 24 However,
armed forces directives governing information protection during
World War I and into the late 1930's often made general reference
to criminal law to give them force. For example, Army regulations
of 1937 indicated that "to reveal secret, confidential, or restricted
matter pertaining to the national defense is a violation of the Espion-
age Act."
By the time of the arrival of the New Deal, information restriction
markings and controls were commonplace in the War and Navy
Departments and spilled over into other government entities whenever
protected records were shared. Restrictions seemingly could be applied
to any type of defense or nondefense information and appeared to
carry sanctions which left few with any desire to question their
appropriateness or application. "National security," a policy term
open to broad definition, was a central concept of the controls.
B. PRESIDENTIAL ACTIONS
The President issued the first Executive Order prescribing security
classification policy and procedures in March, 1940, relying on a 1938
statute concerning the security of armed forces installations and equip-
ment and "information relative thereto." 25 The directive, Executive
Order 838126 authorized the use of control markings on "all official
military or naval books, pamphlets. documents, reports, maps, charts,
plans, designs, models, drawings, photographs, contracts, or specifica-
tions which are now marked under the authority of the Secretary of
War or the Secretary of the Navy as `secret' `confidential,' or 're-
stricted,' and all such articles or equipment which may hereafter be so
marked with the approval or at the direction of the President." The
order made no reference to penalties or the Espionage Act, paralleled
armed forces regulations for marking and handling secret records, gave
civilian employees of the government authority to classify informa-
tion, and was confined largely to traditional national defense matters.
The legislative history of the 1938 statute, upon which the President
relied to issue his directive, provides no indication that Congress an-
ticipated or expected such a security classification arrangement would
be created. Indeed, the Executive Order may have been a substitute
for statutory authority which Congress could not be expected to grant.
ss Sea 1978 Oo -ernment Oneratinns Pec -rity Classification Report, supra note 6, at 5-6.
24 When the Revised Statutes were compiled in 1874, numerous separate statutory house-
keeping laws were combined in a single provision at section 161 which was continued in
editions of the United States Code without alteration until 1958. A 1958 amendment pro-
vided that the housekeepin2, law did not authorize the withholding of information from the
public. (72 Stat. 547 (1958)). The housekeeping law 1s now located at 5 U.S.C. ? 301
(1976).
25 See 52 Stat. 3 (1938).
76 5 Fed. Reg. 1145 et seq. (Mar. 26, 1940).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
The case of the War Security Act is illustrative. Prepared at the direc-
tion of Attorney General Francis Biddle, the proposal would have
given the executive branch broad powers for maintaining internal secu-
rity, including information matters, within the United States.27 The
measure generated considerable public controversy and was sharply
debated in the House during a time when the outcome of the war was
uncertain. However, it was not considered by the Senate.21, Neverthe-
less, a few years later, Congress did authorize secrecy controls for
atomic energy information 29 and intelligence sources and methods.30
During World War II, a patchwork of security-secrecy authorities
were relied upon to protect sensitive information. These included the
President's order, armed forces directives, special agency regulations,
and other ad hoc arrangements. In September 1942, the Office of War
Information, under authority provided by Executive Order 9103 con-
cerning the control of Federal statistical information and Executive
Order 9182 creating the agency, issued a government-wide regulation
on creating and administering classified material S.311
Elsewhere, personnel who would have access to any official secrets of
the Manhattan Project, which was under the supervision of the Army
Corps of Engineers, were subject to a background investigation to
establish their loyalty, integrity, and discretion. Approved individuals
were informed of the penalties for disclosing classified information
improperly and "then required to read and sign either the Espionage
Act or a special secrecy agreement." 32
C. ExFcuTIVE ORDER 10104
Relying on the 1938 statute concerning the security of armed forces
installations and equipment, President Truman issued a new security
classification directive, superseding Executive Order 8381, in Febru-
ary of 1950.33 The order was entitled "Definitions of Vital Military and
Naval Installations and Equipment." The only change in policy made
by Executive Order 10104 was the addition of a fourth designation,
"Top Secret," which brought American information security categories
into alignment with those used by our allies.
As with most earlier directives, regulations, and orders dealing with
use of classification markings, Executive Order 10104 continued to be
directed at the protection of military secrets. Only rarely had classifi-
cation authority extended to nonmilitary agencies or to information
related to foreign policy or diplomatic relations. One exception was the
protection provided in the Espionage Act for cryptographic systems,
communications intelligence information, and similar matters.34
At the time Executive Order 10104 was issued, plans were under way
within the executive branch to overhaul the classification program
completely. The National Security Council had begun pursuing the
27 See 88 Cong. Rec. 8311 (Oct. 17, 1942) ; 89 id. 2393-94 (Mar. 23, 1943).
28 See 89 Id. 2390-2408 (Mar. 23, 1943) ; 2780-99 (Mar. 31, 1943) ; 2877, 2878-95
(Apr. 2. 1943). For press comments, see id. A1312, A1688 (Appendix).
' 60 Stat. 755, 766 (1946) (codified as amended at 42 U.S.C. $? 2014(y), 2162 (1976)).
For a discussion of the history of atomic energy restricted data, see House Committee on
Government Operations, "The Government's Classification of Private Ideas," supra note
22, at 120-72.
80 63 Stat. 208, 211 (1949) (codified at 50 U.S.C. ? 403g (1976) ).
31 1973 Government Operations Security Classification Report, supra note 6, at 7-8.
82 A. Brown and C. MacDonald, "The Secret History of the Atomic Bomb" 201 (1977).
$315 Fed. Reg. 597 et seq. (Feb. 1, 1950).
84 See 1973 Government Operations Security Classification Report, supra note 6, at
8-9. Cf. 50 U.S.C. ? 403g (1976) (protecting intelligence sources and methods) which
was enacted in 1949.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
matter of a new policy in 1948. Actual drafting of a new directive
establishing minimum standards for the handling and processing o
classified records within the government was done by the Interdepart-
mental Committee on Internal Security.35 This effort resulted in a new
Executive Order.
D. EXECUTIVE ORDER 10290
In issuing Executive Order 10290 in September, 1951, President Tru-
man indicated he was relying upon "the authority vested in me by the
Constitution and statutes, and as President of the United States." 36
The order was entitled "Prescribing Regulations Establishing Mini-
mum Standards for the Classification, Transmission, and Handling, by
Departments and Agencies of the Executive Branch, of Official Infor-
ination Which Requires Safeguarding in the Interest of the Security
of the United States."
Four levels of ciassification were prescribed, but their use was no
longer confined to traditional national defense matters. This policy
shift was evident in two ways. First, the classification of information
in the interest of "national security" was now sanctioned under a presi-
dential order. Second, classification authority was extended to all agen-
cies and to all information "the safeguarding of which is necessary in
the interest of national security."
The order also resorted to citing portions of the Espionage Act to
reinforce its requirements. Classified materials furnished to authorized
persons "in or out. of Federal service, other than the Executive
Branch," were to bear a warning statement concerning their improper
disclosure and giving reference to two sections of espionage law.
Criticism of Executive Order 10290 prompted President Eisenhower
to seek a review of the Truman order. The Attorney General conducted
the review and recommended that a new order be issued. Executive
Order 10501 was signed in November, 1953.'' The new order was en-
titled "Safeguarding Official Information in the Interests of the De-
fense of the United States." 38
According to a Justice Department spokesman, the order differed
from its predecessor in the following major respects :
(1) It withdrew authority to classify information from 28
agencies of the Government; (2) in 17 other agencies it lim-
ited authority to classify to the head of the agency, without
power to delegate; (3) it sharply limited the authority to
classify only if required in the interest of the national defense
of the United States; (4) it completely eliminated one of the
most controversial categories of classified information, that
is, "restricted"; (5) it explicitly defined, which had not been
done before, the three remaining categories of classified infor-
mation; "Top Secret," "Secret," and "Confidential," in order
to prevent indiscriminate use of the power to classify when
35 "Report of the Commission on Government Security" 155 (1957).
E? 16 Fed. Reg. 9795 et seq. (Sept. 27. 1951).
18 Fed. Reg. 7049 et seq. (Nov. 9, 1953).
See generally House Committee on Government Operations, "Safeguarding Official In-
formation in the Interests of the Defense of the United States (The Status of Executive
Order 10501)," H.R. Rep. No. 2456, 87th Cong., 2d Sess. (1962).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
specific interests of the national defense did not so require;
(6) it included provisions for review of classified material,
for the purpose of removing the classification or downgrad-
ing the classification when tile interests of national defense
no longer required the original classification; and (7) it made
more definite and certain the procedures for handling classi-
fied information, so that employees would be more alert to
the dangers of unauthorized disclosure.39
The new order significantly improved the classification process by
limiting classification authority and by defining more precisely the
limits and purposes of classification. The elimination of the "re-
stricted" category, which had applied to training manuals and other
documents of lesser sensitivity, resulted in the removal of much in-
formation from the classification system.40
Nevertheless, Executive Order 10501 continued some past practices.
A 1973 report of the House Government Operations Committee ob-
served that President Eisenhower, like his predecessor, apparently
promulgated the directive "relying primarily on implied constitutional
powers of his office and statutes claimed to afford a basis on which to
justify the issuance of the Executive order." 41
In the ten years subsequent to the order's adoption, amendments
were made by Presidents Eisenhower and Kennedy that further nar-
rowed classification authority and established procedures for declas-
sifying and downgrading documents.41 However, no sanctions were
provided for overclassification, and congressional recommendations
regarding this omission were ignored .43 Also, although a prestigious
national study commission indicated the tripartite classification cate-
gories were overly.broad and urged, for reasons of efficiency and
economy, that the "confidential" level of classification be abolished,
this proposal was rejected.44
F. ExEcuTIVE ORDER 11652
In 1971, President Nixon directed that a review be made of the
adequacy of classification and declassification arrangements as well
as all aspects of information security within the executive branch. An
interagency committee was formed to pursue this task. The panel de-
veloped a draft revision of Executive Order 10501 during the summer
and autumn. This proposal was then circulated to selected departments
and agencies for comments during January of 1972. After adjust-
ments were made, it was promulgated in March as Executive Order
11652.45
as "Commission on Government Security," Hearings before the Subcommittee on Re-
organization of the S nate ( o m'ttee on Go,-crn._en, Operations, 84th Cong., 1st Sess.
30 (1955) (testimony of Assistant Attorney General William F. Tompkins).
40 1973 Government Operations Security Classification Report, supra note 6, at 11.
'i Id.
42 See House Committee on Government Operations, "Safeguarding Official Information
in the Interests of the Defense of the United States (The Status of Executive Order
10501)," supra note 38, at 11-13.
as Id. at 25, 26.
" See "Report of the Commission on Government Security" 174-76 (1957).
"37 Fed. Reg. 5200 (Mar. 10. 1972). For a detailed section-by-section analysis of E.O.
11652, see "U.S. Government Information Policies and Practices-Security Classification
Problems Invo'ving Subsection (b) (1) of the Freedom of Information Act (Part 7)," Hear-
ings be"ore a Subcommittee of the House Committee on Government Operations, 92nd
Cong., 2d Sess. 2849-2883 (1972).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
The most significant features of the Nixon order, as characterized by
the President at the time of its issuance, were :
The rules for classifying documents are more restrictive.
The number of departments and people who can originally
classify information has been substantially reduced.
Timetables ranging from 6 to 10 years have been set for
the automatic declassification of documents. Exceptions will
be allowed only for such information as falls within four
specifically defined categories.
Any document exempted from automatic declassification
will be subject to mandatory review after a 10-year period.
Thus, for the first time, a private citizen is given a clear right
to have national security information reviewed on the basis of
specified criteria to determine if continued classification is
warranted so long as the document can be adequately identi-
fied and obtained by the Government with a reasonable
amount of effort.
If information is still classified 30 years after origination.
it will then be automatically declassified unless the head of
the originating department determines in writing that its
continued protection is still necessary and he sets a time for
declassification.
Sanctions may be imposed upon those who abuse the
system.
And a continuing monitoring process will be set up under
the National Security Council and an interagency classifica-
tion review committee, whose chairman is to be appointed by
the President.46
The new order represented an improvement over the previous classi-
fication rules. Classification authority was further limited '47 and clas-
sification categories were more specifically defined.48 Other changes
slightly speeded up the process for declassifying and downgrading
documents.49 In addition, the order established the Interagency Clas-
sification Review Committee as an oversight body.50
However, there were also a number of serious problems with the
Nixon order. A critical evaluation of the new order by the House Com-
mittee on Government Operations in May, 1973, noted major defects,
including :
(a) the change of basic terminology of the order's applica-
tion-from "national defense or foreign policy" to "national
defense or foreign relations", referred to in the new orders as
"national security" ;
(b) the lack of sufficiently strong penalties for overclassifi-
cation;
(c) the lack of assurance to guarantee Congress the full au-
thority to properly exercise its oversight and investigative re-
sponsibilities regarding the operation of the new Executive
Order;
4s Statement of President Nixon (Mar. 8, 1972), reprinted in 1973 Government Opera-
tions Security Classification Report, supra note 6, at 56.
47 E.O. 11652 at ? 2.
48 Id. at 1.
'? Id. at 5.
60 Id. at 4 7(A).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
(d) the legitimation of dozens of access or control mark-
ings that apply to classified or unclassified data; and
(e), loopholes in the mandatory review provisions affect-
ing the declassification of exempt classified information.-51
The Government Operations Committee concluded its assessment of
security classification policy and practice by recommending strongly
"that legislation providing for a statutory security classification sys-
tem ... be considered and enacted by the Congress." 52 Subsequently,
reports from two other congressional panels addressed the need for
legislative action on classification standards.53
G. ExEcuzzvE ORDER 12065
President Carter established a special task force to conduct a
comprehensive review of the information security program on
June 1, 1977.54 The memorandum which set up this group directed
that a new Executive Order be developed which would simplify the
system and provide improved protection for essential national security
information.
The Carter Administration Executive Order on Security Classifica-
tion (Executive Order 12065) was issued on June 28, 1978, and became
effective on December 1, 197855 It replaced Executive Order 11652.
The Carter order continued the pattern of the-previous security clas-
sification orders in further restricting classification authority.
The Carter order was the first to list specific categories of informa-
tion subject to classification. Previous orders had only established
standards for classification in terms of the damage that would result
from disclosure. The Nixon order had included some examples of the
types of information that would be subject to classification, but the
examples were not comprehensive. The Carter order required that in-
formation fall in one of seven categories in order to be considered for
classification.
The classification categories covered information concerning: (a)
military plans, weapons, or operations; (b) foreign government infor-
mation; (c) intelligence activities, sources or methods; (d) foreign
relations or foreign activities of the United States; (e) scientific, tech-
nological, or economic matters relating to the national security; (f)
United States Government programs for safeguarding nuclear mate-
rials or facilities; or (g) other categories of information which are
related to national security and which require protection against un-
authorized disclosure as determined by the President, by a person
designated by the President, or by an agency head.b6
611973 Government Operations Security Classification Report, supra note 6, at 102.
12 Id. at 104.
51 House Committee on Standards of Official Conduct, "Report on Investigation Pursuant
to H. Res. 1042 Concerning Unauthorized Publication of the Report of the Select Com-
mittee on Intelligence,' H.H. Rep. No. 94-1754, 94th Cong., 2d Sess. 43-4 (1976) (recom-
mending that the House leadership initiate research and study leading to establishment of
a classification and declassification system for congressional use) ; Staff of the Subcom-
mittee on Administrative Practice and Procedure, Senate Committee on the Judiciary,
"Agency Implementation of the 1974 Amendments to the Freedom of Information Act,"
95th Cong., 2d Sess. 36 (Committee Print) (1980) (recommending that classification
standards be legislated).
"Letter from Richard M. Neustadt, reprinted in Hearings, appendix 5.
53 43 Fed. Reg. 28949 (July 2, 1978), reprinted in Security Classification Exemption to
the Freedom of Information Act," Hearings before a Subcommittee of the House Committee
on Government Operations, 95th Conk, 1st Sess. 104-117 (1977) [hereinafter cited as
"1977 Security Classification Hearings"].
N E.O. 12065 at 11-301.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
While the inclusion of classification categories represented a step
toward the goal of specificity, the categories in the Carter order-par-
ticularly the last one-have been criticized as being too broad. 57
In addition to providing new classification categories, the Carter
order required that information falling within one of the categories
could be classified only if its release could reasonably be expected to
cause identifiable damage to the national security. The previous order
did not require that the damage be identifiable.58
The Carter order was also the first to require explicitly a balancing
of the public's interest in access to government information against
the need to protect certain information for reasons of national secu-
rity. If, in some cases, the public interest in disclosure outweighed the
need to protect information, then the information was to be declassi-
fied.59 This balancing was to take place, however, only when a request
for declassification was made.
The order provided for the first time that declassification should be
given emphasis equal to that afforded to classification.- Documents
were to be declassified as early as national security permitted'61 and a
justification was required for the extension of classification beyond six
years.62 Declassification rules were also strengthened.63
However, the order maintained some practices that were the subject
of criticism. In issuing it, the President continued to rely upon very
general and implied powers : "the authority vested in me as President
by the Constitution and laws of the United States of America." 84 The
order also continued the practice of referring to "national security"
considerations, a term not totally consistent with the language of the
security classification exemption of the Freedom of Information Act.65
The Carter order created the Information Security Oversight
Office 66 as a successor to the Interagency Classification Review Com-
mittee that had been established under the Nixon order. 67 The new
oversight body, which was made a unit of the General Services Admin-
istration, received some critical attention. An analysis of the order by
the staff of the Government Information and Individual Rights Sub-
committee included the following comment : "Given GSA's lack of
political or economic leverage over most agencies with classification
authority, placing the Oversight Office within GSA does not seem to
portend particularly vigorous enforcement of the order." 68
Further, the activities of the oversight office were limited by the new
order which provided that, when "inspection would pose an exceptional
national security risk," the director of the Information Security Over-
sight Office may be denied the opportunity to scrutinize some informa-
ar See text accompanying notes 89-114.
58 E.O. 12065 at ? 1-104. See also text accompanying notes 74-88.
6' E.O. 12065 at ? 3-303. See also text accompanying notes 168-185.
e 0E.0. 12065 at ? 3-301.
61 Id.
62 Id. at 11-402.
63 Id. at 3.
a Id. at preamble.
es The title of the order is "National Security Information".
66 Id. at ? 5-102.
sTE.O. 11652 at ? 7(A).
68 Staff of the Subcommittee on Government Information and Individual Rights, "Se-
curity Classification : The Experience Under Executive Order 11652 From 1973 to 1976",
reprinted in 1977 Security Classification Hearings, supra note 55, at 95, 101 (1979).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
tion security systems, thus weakening the role of this administrative
accountability monitor.69
The Carter order was criticized for eliminating the General Declassi-
fication Schedule contained in the previous order in a way that might
permit documents to remain classified for longer periods, as well as for
failure to provide more accountability in the classification process.70
Executive Order 12065 was also criticized for allowing the creation
of "special access programs" to control access, distribution, and protec-
tion of "particularly sensitive information." 71 Relying upon this au-
thority, the Carter Administration experimented with a new category
of highly sensitive intelligence information during 1980. Designated
"Royal," materials of this type reportedly were available to only eight
Members of Congress who could examine them only within the secure
facilities of the House and Senate intelligence committees.72 Informa-
tion Security Oversight Office Director Steven Garfinkel told the Sub-
committee that, to the best of his knowledge, the "Royal" special access
system was never employed, but other special access programs are in
existence.73
III. ANALYSIS OF MAJOR FEATURES OF EXECUTIVE
ORDER 12356
A. WHAT IS SUBJECT TO CLASSIFICATION?
Classification Levels.-Both the Carter and Reagan Executive Or-
ders establish three levels of classification : "Top Secret", "Secret", and
"Confidential 11.74 The Carter order provided that if there is reasonable
doubt about which level is appropriate or if information should be
classified at all, the less restrictive level should be used or the informa-
tion should not be classified.75 This continued a policy embodied in the
implementing directive to the Nixon order.76 The Reagan order omits
this rule. It provides instead that if there is reasonable doubt about
the need to classify information, the information shall be safeguarded
as if it were classified pending a determination within 30 days by an
original classification authority. If there is doubt about the appropriate
level of classification, the information shall be safeguarded at the
higher level pending a determination by an original classification au-
thority.77 However, the Reagan order provides no guidance for resolv-
ing doubts when that final determination is made.
The standards for "Top Secret" and "Secret" are the same in both
orders,78 but there is a difference in the standard for "Confidential"
?'E.O. 12065 at ? 5-202(h). See generally, General Accounting Office. "Improved Execu-
tive Branch Oversight Needed for the Government's National Security Information Classi-
fication Program" (LCD-78-125) (Mar. 9, 1979) ; General Accounting Office, "Oversight
of the Government's Security Classification Program-Some Improvement Still Needed"
(LCD-81-13) (Dec. 16, 1980).
70 Staff of the Subcommittee on Government Information and Individual Rights, supra
note 68, at 97-102.
73 E.O. 12065 at ? 4-201.
79 See J. Taylor, "New System Tightly Curbs Access to Intelligence", Washington Post,
September 7, 1980, at A7.
73 Hearings at 194.
74 E.O. 12065 at ?? 1-102 to 1-104; E.O. 12356 at ?? 1.1(a) (1)-(3).
75 E.O. 12065 at ? 1-101.
7637 Fed. Reg. 10053 et seq. (May 19, 1971).
77E.O. 12356 at ? 1.1(c).
78 E.O. 12065 at ?? 1-102. -103. E.O. 12356 at ?? 1.1(a) (1) and (2).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
information. The Carter order provided that "Confidential" shall be
used for information if its unauthorized disclosure "reasonably could
be expected to cause identifiable damage to the national security." 79
The Reagan order deletes the word "identifiable." 80 This is a return to
the standard in the Nixon order.81
The word "identifiable" conveyed the idea that there had to be some
specific type of harm that would result from an unauthorized disclo-
sure. It discouraged reliance on generalized or hypothetical fears
about the consequences of disclosure as a basis for classification. It
was an attempt to combat the tendency of bureaucrats to classify too
much information.
The stated reason for the change in the Reagan order is "to avoid
litigation problems that may arise if a quantum standard is applied"
to the word "identifiable." 82 At the May 5 hearing, Steven Garfinkel,
Director of the Information Security Oversight Office (ISOO),83 said
that he would be "very surprised if there were any practical differ-
ences" between the Carter and Reagan orders on this point 84 He ex-
plained the reasons for the change :
We didn't remove the word "identifiable" because we did
not want the classifiers to avoid the identification of damage.85
Denying that the change would have the effect of permitting classi-
fiers to avoid the identification of damage, Garfinkel elaborated:
As we review classification decisions, we are still going to
require classifiers to justify their assessments about damage
and to identify the damage. It was the use of the word in a
legal context that created the problems, not the fact that we
wanted our classifiers to do other than identify and be con-
scious of potential damage that would result from unau-
thorized disclosure of particular information.86
Later in the same hearing, Garfinkel once again insisted that the
differences between the Carter and Reagan orders were minimal :
In other words, the requirement in Executive Order 12065
for identifiable damage, the concept of conscious thought, is
no different than will exist now. The classifier will still be in a
position where he or she must be able to explain the rationale
behind the classification, the reason for the classification .117
Asked to give some examples of information that qualifies for
classification under the Reagan order, but would not qualify under
the Carter order, Garfinkel responded that "I doubt I could come up
with any examples at all." 88
*5 E.O. 12065 at ? 1-104 (emphasis supplied).
80E.O. 12356 at ??1.1(a) (3).
81E.O. 11652 at ? 1(C).
81 Information Security Oversight Office, "Comparison of Executive Order 12065 and the
(February 4, 1982] Draft Order" 2, reprinted in Hearings, appendix 2.
b' Rep. Glenn English, Chairman of the Subcommittee on Government Information and
Individual Rights, invited National Security Adviser William P. Clark to appear at the
flay 5, 1982, hearing. Clark declined to appear but designated Mr. Garfinkel to "repre-
sent the Administration" at the hearing. Letter from William P. Clark to Glenn English
(April 26, 1982) (available in files of Subcommittee on Government Information and
Individual Rights).
8' Hearings at 154.
85 Id. at 150.
85 Id.
?+ Td. at 162.
81 Id. at 150.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
It is not apparent that the deletion of the word "identifiable" will or
could have the litigation effect that was intended. If the requirements
on classifiers to justify classification decisions will be the same under
both orders, then any judicial review of the propriety of a particular
classification will also be the same. The government's obligation to
justify a classification may, in fact, be no different under the two
orders.
Classification Categories.-The Carter order was the first Executive
Order to identify specific categories of information that were subject
to classification.89 Previous orders had only established standards for
classification in terms of the damage that would result from disclosure.
The Reagan order adds three entirely new categories and modifies one
of the existing categories.
Conflicting reasons were given for the addition of the new categories.
According to an ISOO comparison of the Carter order and the Febru-
ary 4. 1982, draft,. the additional categories were "based on litigation
experience under the Freedom of Information Act." 90 However, when
Mr. Garfinkel was asked if the categories were included because of the
FOIA, he said "no." 91 Further, no known FOIA cases have turned on
the question of whether or not information fit into one of the enumer-
ated categories.92
The first new category in the Reagan order covers information
concerning :
vulnerabilities or capabilities of systems, installations, proj-
ects, or plans relating to the national security; 93
Garfinkel explained that this category includes "such information as
information relating to the protection of the President, information
relating to the protection of our embassies and information relating to
civil preparedness." 94 No specific explanation was offered to justify
the breadth of the language of the category, although Garfinkel did
say that the category was broad just like categories under the Carter
order.95
The classification categories in the Carter order were not models of
specificity. For example, the Carter order included a category covering
"scientific, technological, or economic matters relating to the national
security." 90 The Reagan order continues this category 87 in addition
m E.O. 12065 at 11-201 (a) (5).
w Information Security Oversight Office, "Comparison of Executive Order 12065 and
the [February 4, 1982] Draft Order" 5, reprinted in Hearings, appendix 2.
91 Hearings at 196.
92 Hearings at 75 (Statement of Allan Adler and Morton Halperin, Center for National
Security Studies) ; Hearings at 196 (Statement of Steven Garfinkel, Director, Informa-
tion Oversight Security Oversight Office). After the hearing, Garfinkel indicated that the
only FOIA case in which the result might have been different under the new Executive
Order Is Taylor v. Department of the Army, Civil Action No. 81-2353 (D.D.C. Nov. 20,
1981), appeal docketed, No. 81-2280 (D.C. r. Dec. 4, 1981). The court's decision turned
on an Army regulation that provided that the requested information was unclassified.
However, the court went on -to say that the information would have failed to qualify
for classification anyway because the government was unable to show that disclosure
of the information would cause "identifiable damage" to the national security.
At the May 5 hearing, Deputy Attorney General Richard Willard said of the Taylor
decision: "One of the bases for the Court's decision was that the readiness statistics
didn't really concern military plans or operations." Hearings at 196. This is not correct.
The judge never determined that the requested information failed to fall within any of
the categories. See Letter from Steven Garfinkel to Glenn English (June 22, 1982) (avail-
able in files of Subcommittee on Government Information and Individual Rights).
E.O. 12356 at 11.3 (a) (2).
w Hearings at 196-97.
Id. at 197.
< E.O. 12065 at 11-301 (e
.
n E.O. 12856 at 11.3 (a) ().
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
to the new "vulnerabilities or capabilities" category. The degree of
cverlap in these two categories is uncertain, although it is safe to say
that it is Substantia188
Concern was expressed that the first new category will permit the
classification of information concerning data processing, telecommuni-
cations, and other technologies which have only a tangential or specula-
tive relationship to national security.99
Another witness expressed the fear that the new category would
be used to protect basic information about problems with new weapons
systems.100 The explanation of the intended scope of the new category
provided by Mr. Garfinkel suggests that these concerns may be mis-
placed.101 However, it remains to be seen whether or not classifiers
will exercise the restraint that Garfinkel indicated is expected.
The second new category added in the Reagan order covers "cryp-
tology." 102 The need for this new category is uncertain. Cryptological
information is already classifiable under the Carter order. The word
"cryptology", as added by the Reagan order, is not qualified or
defined?0S
An explanation for this new category was provided by Garfinkel
after the hearings in response to a written question from Subcommit-
tee Chairman Glenn English :
The appearance of "cryptology" as a classifiable category
is not new. The classifiability of this type of information was
clearly recognized in previous Executive orders, including
E.O. 10501 and E.O. 11652. The drafters of E.O. 12065
omitted specific reference to "cryptology" because they be-
lieved that it was adequately covered by the other classifica-
tion categories. However, experience indicated that while the
other categories make clear the classifiability of signals intel-
ligence, similar protection for communications security in-
formation, the other element of cryptologic information, was
not as apparent. The decision was made to clarify this issue by
including "cryptology" as a classification category under
E.O. 12356.
This addition should not result in an increase in the amount
of information classified under the new Order since it covers
the same type of information that is currently being classified
.under E.O. 12065.204
The third new category added in the Reagan order covers "con-
fidential sources." 105 No cogent explanation of the purpose of this
new category was provided.106 "Confidential source" is defined to mean :
Any individual or organization that has provided, or that
may reasonably be expected to provide, information to the
88 Hearings at 197 (Statement of Steven Garfinkel, Director, Information Security
Oversight Office).
w Id. at 76 (Statement of Allan Adler and Morton Halperin, Center for National Se-
curity Studies).
100 Id. at 98 (Statement of Bob Schaefer, Society of Professional Journalists).
101 See also text accompanying notes 139, 145.
102 E.O: 12356 at 11.3(a) (8).
103 For a discussion of cryptography issues, see House Committee on Government Oper-
ations, "The Government's Classification of Private Ideas," supra note 22, at 62-118.
104 Letter from Steven Garfinkel to Glenn English (June 22, 1982) (available in files of
Subcommittee on Government Information and Individual Rights).
405 E.O. 12356 at 11.3 (a) (9).
la See text accompanying notes 265-274.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
United States on matters pertaining to the national security
with the expectation, expressed or implied, that the informa-
tion or relationship, or both, be held in confidence.107
In commenting on the February 4, 1982, draft, the Chairman of
the House Select Committee on Intelligence called this definition
"overly broad" and questioned whether the disclosure of the name
of a person who is only a potential source could damage national
security.b08
In response to questioning at the May 5 hearing about the confiden-
tial source category, Mr. Garfinkel indicated that anyone who would
be in a position to provide information could be a confidential source
within the meaning of the order. However, he also pointed out that
there are other requirements in the order that would have to be satis-
fied before information would be subject to classification.l? The clas-
sification of information about potential, but not actual, sources should
be an extraordinary event.110
In addition to the new categories, the Reagan order modifies an
existing category to read: "intelligence activities (including special
activities), or intelligence sources or methods." 111 The words in paren-
theses are new. Mr. Garfinkel indicated that "special activities" will
be given the same meaning as in Executive Order 12333 on intelligence
activities.112 The explanation is that the change was made to ensure
that special intelligence activities receive protection." 118
Further, all three new categories have to be interpreted in light of
Garfinkel's assurances that the universe of information that can be
classified under both orders is "essentially the same." 114 This supports
the narrowest possible interpretation of the new categories and of
other changes in the Reagan order.
Mandatory Classi fccation.-The classification categories describe
the information that can qualify for classification. The standards
included in the classification levels (the degree of harm to national
security) determine when information that is subject to classification
should be classified.
Both orders provide guidance to classifiers on how to apply the
classification rules. The major difference between the Reagan and
Carter orders in this area is that the Reagan order could, be inter-
preted to require that anything meeting the criteria for classification
must be classified.115 This has been referred to as a "mandatory classi-
fication" rule. The Carter order merely permitted the classification of
information that met the minimum criteria.116 It clearly allowed clas-
sifiers the discretion not to classify information.
107 E.O. 12356 at 6.1(f).
IN Letter from Edward P. Boland to William P. Clark, National Security Adviser
(Mar. 9, 1982), reprinted in Hearings, appendix 4.
108 Hearings at 204.
nc The Freedom of Information Act allows the protection of information provided by
confidential sources in law enforcement investigations through Exemption 7(D), 5 U.S.C.
? 552(b) (7) (D) (1976). Although the Reagan Administration has proposed amendments
to the Freedom of Information Act, none of the amendments indicates any need for pro-
tection of an expanded category of confidential sources. See H.R. 4805, 97th Cong.
Ill E. 0. 12356 at ? 1.3(a) (4).
11246 Fed. Reg. 59941 et seq. (Dec. 8, 1981).
113 Hearings at 202-03. See text accompanying notes 261-264.
114 Id. at 184.
1'$E.O. 12365 at ? 1.3(b).
no E.O. 12065 at ? 1-302.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
One witness expressed concern that, if taken literally, no official
short of the President would have any authority under the Reagan
order to release classified information no matter how important it
might be to public debate or understanding of major policy issues.117
Garfinkel denied that there would be any practical differences
between the two orders on this point. The change in language, was
"a conscious effort to change the tone of the order, to make the order
a little more positive about the requirements of security." 118 .
Garfinkel also stated that individuals who have the authority to
classify information have the authority to declassify it if the public
interest requires disclosure, and that the authority to classify inher-
ently includes the authority to disclose information if it is in the public
interest.119 If such authority is inherent in the classification process,
then the only difference between the two orders in this area is that the
Carter order is explicit whereas the Reagan order is not.'20
In addition, the Reagan order includes a presumption that unauthor-
ized disclosure of foreign government information, the identity of a
confidential foreign source, and intelligence sources and methods could
reasonably be expected to cause damage to the national security.121 A
similar presumption in the Carter order only includes foreign govern-
ment information and the identity of confidential foreign sources. 112
No justification was presented for the expansion of a presumption of
classifiability to sources and methods, and there is reason to question
the need for the expansion. As one witness pointed out, the CIA has
voluntarily disclosed information about sources and methods in the
past.123 The large number of known past disclosures or methods casts
doubt on the reliability or utility of the presumption in this area. The
Chairman of the House Select Committee on Intelligence expressed
similar doubts : "It is difficult to accept as the basis for a presumption
that all sources and methods would damage the national security if
disclosed ..." 124
The limited use of such presumptions for narrow categories of in-
formation that tend to be uniformly sensitive may be unobjectionable
as a guide for classifiers. However, the expansion of presumptions to
areas where significant quantities of data do not meet the minimum
damage standard of the Executive Order threatens the advantages of
the use of presumptions, permits overclassification of information, and
invites close judicial scrutiny if exclusively relied on in FOIA litiga-
tion.125
Limitations on Classification.-Both orders contain similar provi-
sions prohibiting the use of classification to conceal violations of law,
inefficiency, administrative error; to prevent embarrassment; or to re-
strain competition. 126
"7 Hearing at 71 (Statement of Allan Adler and Morton Halperin, Center for National
Security Studies).
lax Hearings at 154.
33' Id. at 205-06.
lm See text accompanying notes 180-185.
E.O. 12356 at ? 1.3(c).
'22 E.O. 12065 at 11-303.
123 Hearings at 77-78 (Statement of Allan Adler and Morton Halperin, Center for Na-
tional Security Studies).
124 Letter from Edward P. Boland to William P. Clark, National Security Adviser (Mar. 9,
1982), reprinted in Hearings, appendix 4.
= Id
320 E.O. 12065 at ? 1-601; E.O. 12356 at ?1.6(a).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
19
The Carter order included a number of additional specific limitations
on classification that are not repeated in the Reagan order. These limi-
tations prohibit the classification of :
a product of non-governmental research and development
that does not incorporate or reveal classified information to
which the producer or developer was given prior access .. .
until and unless the government acquires a proprietary inter-
est in the product; 127
references to classified documents that do not disclose clas-
sified information.128
The February 4 draft proposed to eliminate another of the Carter
limitations, but the limitation was reinstated in the final order. This
limitation prevented classification of':
basic scientific research information not clearly related to the
national security.129
The explanation for the original deletion of all three limitations
was that they are "self-evident," 130 and that the information covered
by the limitations is not subject to classification under existing tests.131
The limitation for basic scientific research was restored in the final
order because "its absence had caused such a concern among the
scientific community . . ." 132
The importance of the limitations in the Carter order was that they
allayed fears about the scope of the government's classification au-
thority for private technology and basic scientific research. The limita-
tions made it clear that both categories of information were not sub-
j ect to classification. 33
The proposed deletion of these two limitations-especially the
limitation against classifying private technology-gave these ques-
tions renewed importance. Witnesses at the March 10 hearing ex-
pressed concern about the significance of the deletions. At the center
of the controversy was the definition of "information" in the Reagan
order. Information is defined to mean :
Any information or material, regardless of its physical
form or characteristics, that is owned by, produced by, ppro-
duced for, or is under the control of the United States Gov-
ernment.134
The Carter order contains substantially the same definition,135 but
the presence of the limitations made the ambiguities of the definition
less important.
Professor Mary Cheh of George Washington University's National
Law Center found the February 4 draft "equivocal with respect to
whether or not it applies to privately-generated information." 186 She
227 E.O. 12065 at 11-603.
lze Id. at ? 1-604.
'" Id. at ? 1-602; E.O. 12356 at ? 1.6(b).
'' Information Security Oversight Office, "Comparison of Executive Order 12065 and
the [February 4, 1982] Draft Order" 8, reprinted in Hearings, appendix 2.
"a Hearings at 172 (Statement of Steven Garfinkel, Director, Information Security Over-
sight Office).
in Id. See text accompanying notes 252-260.
'n See Hearings at 21-22 (Statement of Professor Mary Cheh).
2"E.O. 12356 at ? 6.1(b).
'ffi E.O. 12065 at ? 6-102.
1w Hearing at 8.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
suggested that "you could take the same definition [of "information"]
in the new order, given the new classification categories, and potentially
extend it to privately developed information." 137 She also questioned
whether the concept of "control" in the definition of "information"
would extend to information submitted when applying for a patent or
a license to export information., -18
Garfinkel denied that the deletion of the limitation on classification
of private technology increased the scope of classifiable information 139
He also clarified when data qualified as "information" within the
meaning of the order. On the question of when information is "pro-
duced for" the Government, Garfinkel explained that information
produced for the general market does not qualify.140
With respect to the meaning of the word "control" in the definition
of information, Garfinkel said that "the idea there is that the Govern-
ment is in a position to require its production, to retain the informa-
tion or to have the information destroyed." 141 He also explained that
"control means that some statute or legal instrument or some sort of
legal commitment between the holder of that information and the
Government enables the Government to prevent its dissemination." 142
Information submitted in connection with patent applications is not
controlled by the Government, according to Garfinkel, because the
submitter is not bound by the Government's control over the informa-
tion and because there is no agreement between the patent seeker and
the Government concerning control.143 Where information is submitted
to the Government under export reporting or licensing laws, that in-
formation is not controlled by the Government because the submitter
retains a copy and has entered into no agreement with the Government
to control dissemination. 144
As with the interpretation of the classification categories and other
features of the Reagan order, the deletion of the limitations must be
interpreted in light of Garfinkel's assurances that the universe of in-
formation classifiable under both orders is essentially the same.146
B. CLAssIFIcATIox AtTnoRrrY AND PROCEDURE
Original Classification Authority.-The Carter order included a list
of agencies whose officials are given original classification authority
for the Top Secret, Secret, and Confidential levels.146 The total num-
ber of officials with original classification authority under the Carter
order was 7,229. This was a significant reduction from the Nixon
order (17,626 classifiers), which was in turn a significant reduction
from the Eisenhower order (59,316 classifiers) .147 Garfinkel predicted
2 ";Id. at 17.
' Id.
In Id. at 171.
'' IS. at 175.
lu Id.
lu Id. at 17T.
'As Id. at 176. On the issue of invention secrecy laws, see House Committee on Govern-
ment Operations, "The Governments Classification of Private Ideas," supra note 22, at
1-6.2.
1 Hearings at 177.
14 See text companying note 114.
11? E.O. 12065 at 91-201.
141 Information Security Oversight Office, "Comparison of Major Features of 'Executive
Orders Governing The Information Security Program," reprinted in Hearings, appendix
2. See also Hearings 164-65 (Statement of Steven Garfinkel, Director, Information Se-
curity Oversight Office).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
that there would be no significant increase or decrease under the
Reagan order.148
Unlike the Carter order,149 the Reagan order did not include a list
of the agencies with classification authority.150 The list is contained in
a separate Presidential order that appeared in the Federal Register a
month after the order was signed.151 The stated explanation for this
change is because it is more practical from an administrative stand-
point. 152
The most notable change in the list of agencies with classification
authority is the addition of the Environmental Protection Agency as
an agency with authority to classify information as "Confidential." 153
Delegation of Classification Authority.-Both the Carter and Rea-
gan orders contain rules for the delegation of original classification
authority. Separate rules are stated for the delegation of each classi-
fication level. As an example of how the delegation rules work, the
rules for delegation of "Top Secret" will be explained. For both the
Carter and Reagan orders, the rules for "Top Secret" are similar to
those for the other levels.
The Carter order provided that "Top Secret" classification authority
may be delegated only to "principal subordinate officials who have a
frequent need to exercise such authority as determined by the President
or by agency heads" who have original "Top Secret" classification au-
thority.154 The Carter order specifically prohibited the redelegation
of any delegated original classification authority.155
The Reagan order permits delegation of original "Top Secret" clas-
sification authority by the President; by an agency head or official who
has been granted original "Top Secret' classification authority under
the Presidential order, or by the senior agency official with responsi-
bility for the agency information security program, provided that the
senior agency official has been delegated original "Top Secret" classi-
fication authority by the agency head' .156
Reclassification.-The Carter order provided that documents that
have been declassified and publicly released may not be reclassified.157
The Reagan order permits reclassification of information that has
been declassified and released if it is determined in writing that (1) the
information requires protection in the interest of national security and
(2) the information "may reasonably be recovered." 158 This is one of
the few changes in the Reagan order that is specifically identified as a
change in policy.159
In addition, the Reagan order makes it easier to classify or reclassify
a document after an agency has received a request for it under the
FOIA, Privacy Act, or under the Executive Order itself. The Carter
order prohibited such classifications unless authorized by the agency
in Hearings at 165.
140 E.O. 12065 at ? 1-2.
160 E.O. 12065 at 1.2:
'' 47 Fed. Reg. 20105-6 (May 11, 1982), reprinted in Heart appendix-1.
Information Security Oversight Office, 'Comparison of Executive Order 12065 and
the [February 4, 1982) Draft Order 3, reprinted in Hearings, appendix 2.
12 Hearings at 153 (Statement of Steven Garfinkel, Director, Information Security Over-
sight ce).
''E.O.Offi 12065 at j 1-204(a).
166 Id. at 1204 (d).
100E.O. 12356 at 51.2(d) (2). The senior agency official is designated under ? 5.3(a)(1).
167 E.O. 12065 at ; 1-607.
150E.O. 12356 at 11.6 (b).
10 Hearings at 182 (Statement of Richard Willard, Deputy Assistant Attorney. General);
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
22.
head or deputy.180 The Reagan order permits lower level officials to
classify documents after a request for the documents is received."'
Mr. Garfinkel explained the origin of the reclassification authority :
During the course of our consideration of the new Execu-
tive Order, it came to our attention that there were several
instances in a couple of agencies where information had been
erroneously declassified and released and was in the hands of
one person. This person was quite willing to have the informa-
tion retrieved by the originating agency and reclassified, but
the originating agency felt by the terms of the language of the
existing order, it could not do so.
So it was viewed by us in our drafting procedure that there
needed to-be some.degree of flexibility on this issue, that we
could not always say that that information could not be re-
classified.162
The reclassification provision generated as much controversy as any
other feature of the.new order. Allan Adler and Morton Halperin of
the Center for National Security Studies found it-
particularly troubling because it is not limited to cases where
sensitive information has been declassified by some gross error
on the part of processing officials and would appear to permit
the Government to compel the return of such information by
lawsuit, if necessary.
Consider, for example, the situation where an FOIA re-
quester could be asked to return a previously-released docu-
ment on the basis of the Government's claim that the informa-
tion has been classified and cannot be disseminated without
official authorization and clearance. If the requester refuses
to voluntarily give up the document, is a lawsuit one means by
which the document "may reasonably be recovered"? Would
the requester, who sought no access to classified information
in the first place, now find himself in possession (i.e., knowl-
edge) of information that he cannot further disseminate with-
out penalty? 188
There have been a number of recent instances where, notwithstand-
ing the prohibition in the Carter order against reclassification, the
Government has threatened or attempted to reclassify information
that was either declassified or otherwise in the public domain. In one
incident, a researcher who obtained declassified documents from the
National Archives was asked to return them to the Archives to be re-
produced. The documents were then reviewed for possible reclassifica-
tion, although in the end the documents were returned to the research-
er in their entirety 184
In a second case reported in the press, the Justice Department de-
manded the return of documents from an individual who had received
them from the Department in response to a Freedom of Information
Act request. The Department contended that the information had been
100 E.O. 12065 at ? 1-606.
a01E.O. 12356 at 11.6 (c).
=Hearings at 179.
103 Hearings at 79.
101 This incident is discussed in Hearings at 180-81.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
released "in error" and threatened legal action if the information were
published.165
Another similar instance involved the seizure of documents from
American journalists returning from abroad. The documents were
part of a 13 volume archives of secret United States documents on
American relations with Iran purportedly taken from the U.S. em-
bassy in Teheran when it was seized in 1979. The documents were
reportedly for sale in the bazaars of Tehran, and highlights were
later published in the Washington Post.166
Some surprising recovery authority was claimed by Government
representatives at the May 5 hearing. Representative Weiss asked a
series of questions about what constitutes reasonable recovery action
within the meaning of the draft order:
Mr. WEISS. Is deception a reasonable course of action?
Mr. GARFINKEL. I think ordinarily it is not reasonable.
Mr. WEISS. Is force a reasonable course of action?
Mr. GARFINKEL. Physical force?
Mr. WEISS. You go up to somebody and say, "OK, buddy,
hand it back or else."
Mr. GARFINKEL. No. Ordinarily, that would not be reason-
able.
Mr. WEISS. What do you mean by "ordinarily?"
Mr. GARFINKEL. I would say under all but the most unbe-
lievable circumstances. I am not in a position to even describe
them.
Mr. KINDNESS. Suppose the guy had just stolen it out of the
file. He tried to make away with it. What about that?
Mr. WEISS. This is where the agency declassified it and the
Government now changes its mind. Can the Government
agent walk up with a gun and say, "Hand it back ?"
Mr. GARFINKEL. I don't want to be on the record to say that
could never happen. What I am saying is that I cannot con-
ceive of a situation when that would happen.
Mr. WEISS. Is an illegal entry, a reasonable course of action
by way of recovery?
Mr. GARFINKEL. My answer would be the same.
Mr. WEISS. Would you like to comment on this?
Mr. WILLARD. My answer would be, no. An illegal entry
would not be reasonable in my opinion. As to force, unauthor-
ized force would not be reasonable. Obviously at some point,
if people resist a court order, United States Marshals are
authorized to enforce the court order. If that requires force,
it might occur, but certainly not ad hoc or the kind of force
that you describe. That would not be reasonable.
Mr. WEISS. What about deception?
Mr. WILLARD. It would depend on the circumstances and
who was involved. Obviously, in the case you describe [in-
volving the researcher who was misled into returning docu-
ments to the Archives], the Government determined that it
was not reasonable to engage in misleading the researcher,
and the documents were returned. 167
See New York Times, March 14, 1982.
Washington Post, January 31, 1982.
Hearings at 181.
Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
24
Thus, while one government spokesman refused to rule out either
physical force or illegal entries as a reasonable means to recover and
reclassify documents, a second spokesman determined that this type
of activity would be unreasonable. Neither would rule out the use of
deception as a recovery technique.
The reclassification of information by the Government is not unrea-
sonable if the information meets the standards for continued classifica-
tion and if the recovery can be accomplished through voluntary co-
operation. However, it is far from certain that a change in an Execu-
tive Order can give the Government new authority to recover declassi-
fied documents. Nor is it apparent that the President of the United
States could or would determine that deception, force, illegal entries,
or similar unsavory techniques are reasonable actions to be undertaken
by the Government.
C. DECLASSIFICATION
Duration of Classification.-The Carter order required that each
classified document include a date or event for declassification or re-
view.168 Any document classified for more than six years must contain
additional information, including the name of the classifier and the
reasons why longer classification is required.169
The basic declassification requirement of the Reagan order is simply
that documents shall continue to be classified as long as required by na-
tional security considerations. 1170 The Carter order contained similar
language 171 and enforced it with the six year declassification period.
The Reagan order omits all references to the six year period or other
periods of time for declassification. It permits a document to be marked
with a date or event for declassification 172 or with a notation "originat-
ing agency's determination required." 17s
At the March 10 hearing, Dr. Anna Nelson, who represented the
American Historical Association, the Organization of American His-
torians, and the Society for Historians of-American Foreign Relations
was particularly troubled by this change :
I would like to point out to the committee that section 1.4
(a), the section that eliminates the current time limitation on
the classification of most information, is the most disturbing
section in this entire draft. This section represents a signifi-
cant change in the classification policies of the last three dec-
ades. It eliminates the concept that at some date, all classified
information should be reviewed for declassification and re-
verts back to policies in effect before President Eisenhower's
Executive Order first addressed the massive accumulation of
classified information. 174
Dr. Nelson went on to explain why orderly declassification is impor-
tant to the process of historical research :
To understand the relationships between the individuals,
institutions and ideas which encompass political, diplomatic
or military history, the historian needs not one document or
lea E.O. 12065 at 11-401.
N? Id. at 1-502.
170 E.O. 12356 at 11.4 (a).
171 E.O. 12065 at 13-301.
1' E.O. 12356 at 1.4(a).
lea Id, at 11.5(a)(3).
14' Hearings at 110-11.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
even 10 documents, but an array of information from a broad
spectrum of sources. Therefore, the ideal declassification proc-
ess for historical research is the very kind of orderly declas-
sification which this draft Executive Order will now elimi-
nate.175
It is not clear if the Carter order's six-year default declassification
date for most documents has been successful. It has not been in effect
for six years, and the Reagan order will eliminate it before it is tested.
Mr. Garfinkel indicated that many documents that were supposed
to be declassified after six years have been marked "review before de-
classification." This is a term that is not recognized under the Carter
order. The widespread use of this marking suggests that agencies are
reluctant to allow the automatic declassification of documents after six
years.176 According to Garfinkel, anybody marking documents "review
before declassification" had to be "well aware that there was going to
be no such review at the six-year mark." 177
Declassification Policy.-The Carter order states that declassifica-
tion shall be given emphasis comparable to that accorded classifica-
tion.""" There is no similar policy statement in the Reagan order. Gar-
finkel told the Subcommittee that the statement was dropped "simply
as an economy measure." 179
The Carter order also included a provision generally referred to as
"the balancing test" :
It is presumed that information which continues to meet the
classification requirements in Section 1-3 requires continued
protection. In some cases, however, the need to protect such
information may be outweighed by the public interest in dis-
closure of the information, and in these cases the information
should be declassified. 180
If questions arise about the disclosure of information on grounds of
the public interest, the agency head, a senior agency official, or an offi-
cial with "Top Secret" classification authority will make the final
determination.181
The balancing test is dropped in the Reagan order. Deputy Assistant
Attorney General Richard Willard explained why :
The balancing provision produced unsatisfactory results
and is properly eliminated from the new executive order.
While the provision was never intended to introduce a man-
datory requirement for proper classification, plaintiffs have
been able to argue, not without some success, that it was so
intended, and that the courts could properly review and even
overrule the agenev's decision whether to balance. There has
been a corresponding erosion in the certainty and the ap-
pearance of certainty in classification decisions. The agencies
have had to edce closer to disclosure of classified information
to support their decisions on the public record, and classified
affidavits have necessarily grown in complexity and detail in
176 Id. at 111.
176Id. at 218.
177 id.
178 E.O. 12065 at 13-801.
179 Hearings at 216.
180 E.O. 12065 at 13-303.
1e1 Id.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
26
order to overcome arguments of opposing counsel and satisfy
inquiries of reviewing courts. Moreover, the balancing provi-
sion adds another necessarily judgmental requirement courts
nevertheless view as subject to judicial review. It further com-
plicates the ability of the government to defend classification
decisions and the ability of the agency to get on with its pri-
mary job of protecting national security rather than defend-
ing itself in court.182
Willard's concern over the degree of interference with an agency's
ability to carry out its work is notable. Steven Garfinkel said that only
an infinitesimal amount of classified data is ever the subject of FOIA
litigation.183 It is not apparent how this limited amount of litigation
could be the cause of interference of the magnitude suggested in Wil-
lard's statement.
In response to questioning at the May 5 hearing, Willard continually
returned to the point that the public interest is not always served by
the disclosure of information.184 The truth of this statement has never
been in dispute, and the Carter order specifically recognized that the
public interest will outweigh the need for protection only "in some
cases."
The ultimate significance of the deletion of the balancing test is
uncertain, since Garfinkel stated that the authority to classify "in-
herently includes the authority to make the determination that the
public interest requires disclosure." 185 Misapplication of this inherent
authority may be the subject of judicial review even if the order does
not contain an explicit reference.
Systematic Review for Declassification.-The Carter order insti-
tuted a new procedure for the systematic review for declassification of
most classified documents as they become 20 years old. The National
Archives and Records Service (NARS) was given the primary re-
sponsibility for the review of these documents.186 The Carter proce-
dure replaced general declassification schedules established under the
Nixon order.
The Reagan order includes a much less specific procedure for the
systematic review of documents for declassification. NARS would
still have the primary function of reviewing documents, but no period
is specified for the conduct of declassification reviews.187
This part of the Carter order has been controversial. In October,
1980, the General Accounting Office issued a report recommending
that the program of systematic review be eliminated and that all de-
classification be done only when specifically requested by the public.188
192 Hearings at 141
lea Id. at 210. The information Security Oversight Office estimates that between 800,000
and 1 million documents are classified each year. In addition, another 16 million documents
are derivatively classified. There are annroxlmately 691 million pages of classified docu-
ments at the National Arcbieves, and the member of classified documents throughout the
government is much larger. Sce Letter from Steven Garfinkel to Glenn English (June 22.
1982) (available in files of the Subcommittee on Government Information and Individual
Rights).
184 Hearings at 160, 161.
185M. at 205-06.
195 E. 0. 12065 at J 3-4.
199 E.O. 12356 at 1 3.3.
199 General Accounting Office. "Systematic Review for Peclassi$cation of National Se-
curity Information-Do Benefits Exceed Costs" (LCD-81-3) (Oct. 15, 1980).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
GAO felt that the review was not cost effective and that NARS did not
have the resources to conduct the mandated review.1119
Steven Garfinkel described the systematic review under Executive
Order 12356 as a compromise between the recommendation made by
GAO and the Carter procedure. 190 He also indicated that although no
period is set in the order, a thirty year systematic review period would
be set administratively-",'
Requests for Declassification.-The Carter order included a man-
datory review procedure for the handling of requests from the public
or from agencies for the declassification of information. If, after a
request and review, it is determined that information no longer re-
quires protection under the Executive Order, the information is to be
released.192 Although the procedure was described as "slow and often
frustrating",193 a significant proportion of requests for declassification
have resulted in the release of information.
The Reagan order continues the mandatory review program with
some significant changes. First, it limits requests to citizens, perma-
nent resident aliens, federal agencies, and state and local govern-
nlents.194 This change is similar to proposed Reagan Administration
FOIA amendments.195
The Reagan order requires that the request for declassification de-
scribe "the document or material containing the information with
sufficient specificity to enable the agency to locate it with a reasonable
amount of effort.196 Although similar to the Carter requirement that
the request "reasonably describe the information," 197 there is a differ-
ence. The Reagan order adds the requirement that a requester identify
the document to be reviewed for declassification. How this will be in-
terpreted remains to be seen.
IV. DISCUSSION. FINDINGS. AND RECOMMENDATIONS
A. ADOPTION OF EXECUTIVE ORDER 12356
The process used by the Reagan Administration for consideration
and adoption of the new Executive Order on Security Classification
has been a significant source of controversy. When he signed Executive
Order 12356, President Reagan said that it "reflects a coordinated ef-
fort involving officials of the executive branch, Members of Congress,
and representatives of concerned private organizations." 198 However,
the record demonstrates that, with very few exceptions, the entire
revision effort was carried on in isolation within the executive branch.
'"See "Oversight of the National Archives and Records Service," Hearings before a
Subcommittee of the House Committee on Government Operations, 97th Cong., 2d Seas.
(Mar. 1982) (to be printed). See also Information Security Oversight Office, "Annual
Report to the President 1980-1981" 12.
190 Hearings at 188-89.
191 Id. See also $ 2001.31(c) (iii) of the Imnlementine Directive for E.O. 12356, 47 Fed.
Reg. 27836 et seq. (June 25, 1982) (to be codified at 32 C.F.R. ? 2001.31 (c) (iii) ).
197 E.O. 12065 at ? 3-501.
153 Hearings at 119 (Statement of Dr. Anna Nelson, American Historical Association,Or-
ganization of American Historians, and the Society for Historians of American Foreign
Relations).
194 F..O. 12356 at 13.4 (a) (1).
195 See H.R. 4805 and S. 1751, 97th Cong. See also S. 1730, 97th Cong.
19^E.O. 12356 at ? 3.4(a) (2).
197 E.O. 12065 at $ 3-501.
IN 18 Weekly Comp. Pres. Doc. 431 (Apr. 2, 1982), reprinted in Hearings, appendix 1.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
This is in sharp contrast to the process for adoption of the Carter Ex-
ecutive Order on Security Classification, a process that included public
notice, comment, and participation.
The effort to revise Executive Order 12065 began in February, 1981.
Presidential Counselor Edwin Meese asked agencies of the intelligence
communities for prospective amendments ? to Executive Orders 12036
and 12065 "that would enhance the United States' intelligence capa-
bilities." 199 An intelligence community task force chaired by the Cen-
tral Intelligence Agency was established in response and eventually
determined to draft a complete rewrite of Executive Order 12065. A
draft was submitted on August 28, 1981, to Richard Allen, Assistant
to the President for National Security Affairs. On September 2, 1981,
Allen submitted this draft to the Information Security Oversight
Office (ISOO) with instructions to coordinate the development of a re-
placement for Executive Order 12065 with other executive branch
agencies.200 ISOO was created under Executive Order 12065 as a suc-
cessor to the Interagency Classification Review Committee.201 Admin-
istratively, ISOO is part of the General Services Administration, but
it takes its policy direction from the National Security Council.202
The draft prepared by the intelligence community task force "served
as framework for ISOO in its development of a draft Order to be
circulated for further agency comment." 203 An ISOO draft was cir-
culated to agencies on October 16, 1981, and agencies were given 30
days to comment. A revised draft was submitted to the Acting As-
sistant to the President for National Security Affairs on December 4.
1981. This draft was forwarded to the Office of the Vice President and
ten of the major classifying agencies for comment on December 23,
1981. The next draft, dated February 2, 1982, was the one that was cir-
culated on Capitol Hill.204
Although the revision effort was not classified, no formal public an-
nouncement was ever made about it,205 and the first notification to the
Government Operations Committee came on February 4, 1982, when
ISOO Director Steven Garfinkel sent a copy of the February 4 draft
to the General Counsel of the Government Operations Committee.
Copies of the draft order were provided at the same time to several
other congressional committees. The routing slip that accompanied the
draft sent to the Government Operations Committee stated that the
proposed draft was provided for the Committee's review and com-
ment. The only restriction noted on the routing slip was the following :
"The materials are being provided with the understanding that access
to them will be limited to Committees members and necessary staff
personnel."
Although the routing slip did not indicate that there was a deadline
for congressional comments, Committee staff was advised by Mr. Gar-
finkel that the deadline was February 22, 1982. The House was in
.recess for a majority of the period allowed for congressional comments.
122 Letter from Steven Garfinkel to Glenn English (June 22. 1982) (available in files
of Subcommittee on Government Information and Individual Rights).
200 Id.
201 E.O. 12065 at $ 5-102.
202 Hearings at 195 (Statement of Steven Garfinkel, Director, Security Oversight Office)
E.O. 12065 at $ 5-101 : E.(1. 12.156 at ? 5.1(a).
203 Letter from Steven Garfinkel to Glenn English (June 22. 1982) (available in files of
Subcommittee on Government Information and Individual Rights).
204 Id.
N5 Id.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
On February 10, 1982, Rep. Glenn English, Chairman of the Sub-
comittee on Government Information and Individual Rights, sent a
letter to William Clark (who had succeeded Richard Allen as National
Security Adviser) asking that the deadline for comments on the draft
be put off "in order to allow an adequate opportunity for review by
the Congress." 206 The letter was cosigned by Rep. Jonathan Bingham,
Chairman of the Subcommittee on International Economic Policy
and Trade of the Committee on Foreign Affairs; Rep. Don Edwards,
Chairman of the Subcommittee on Civil and Constitutional Rights of
the Committee on the Judiciary ; Rep. George Brown, Chairman of the
Subcommittee on Department Operations, Research, and Foreign
Agriculture of the Committee on Agriculture; Rep. Doug Walgren,
Chairman of the Subcommittee on Science, Research, and Technology
of the Committee on Science and Technology; Rep. Peter Rodino,
Chairman of the Committee on the Judiciary; Rep. Edward Boland,
Chairman of the Select Committee on Intelligence; and Rep. Robert
Kastenmeier. Chairman of the Subcommittee on Courts, Civil Liber-
ties, and the Administration of Justice of the Committee on the
Judiciary.
National Security Adviser Clark responded to the letter by tele-
phone on February 22,1982, telling Rep. English that more time would
be allowed for congressional review, but setting no new deadline. The
next day, Rep. English announced that hearings would be held on the
proposed draft by the Subcommittee on Government Information and
Individual Rights on March 10 and 11, 1982.207 Late in the day on
February 2.3, aletter . dated February 22 from Mr. Clark was hand
delivered to Rep. English. The letter suggested that any comments
should be referred to Edwin Meese, Counsellor to the President, and
to Steven Garfinkel, Information Security Oversight Office, prior to
March 5.2'8
Letters dated February 25, 1982, were sent to National Security
Adviser Clark and Attorney General William French Smith inviting
them to appear or to send representatives to the hearings scheduled on
March 11.209
Assistant Attorney General Robert A. McConnell responded on
behalf of the Attorney General by letter dated March 4, 1982.210 The
Attorney General declined the invitation because it would be "inap-
propriate" to send a witness since the process of revising the Executive
Order had not been completed and the matter had not yet been pre-
sented to the President for decision.
Michael O. Wheeler. Staff Secretary of the National Security
Council, responded on behalf of William Clark in a letter dated
March 8 that was not received until March 9. The invitation to testify
was declined :
Copies of the revised Order were provided to selected Con-
gressional committees on the explicit understanding that
Administration witnesses would not appear at hearings while
the internal deliberative process was underway. Since that
2"" The letter is reprinted in Hearings, appendix 3.
2*7 728 Cong.Rec. H465 (daily ed. Feb. 23, 1982).
f0B This letter is reprinted in Hearings, appendix 3.
za These letters are reprinted in Hearings, appendix 3.
au This letter is reprinted in Hearings, appendix 3.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
30
process continues, we must respectfully decline your invita-
tion to appear on March 11.211
The "explicit understanding" referred to by Wheeler in this letter
was not mentioned in the routing slip attached to the draft order
provided to this Committee. During the hearing on March 10, Chair-
man English stated : "If any such condition was imposed, it is news
to me. No one on this committee had ever heard of such an under-
standing until it was mentioned in the letter." 212
All three Members who attended the Subcommittee hearings on
March 10 expressed frustration that no Administration witness was
willing to appear to explain the purpose of the changes in the draft
order. At the suggestion of Rep. Ted Weiss, a letter was sent on
March 11 asking the President to designate a spokesman to appear
before the Subcommittee prior to the issuance of the revised Executive
Order.213
On April 2,1982, President Reagan signed the new Executive Order
on Security Classification (E.O. 12356).214 The effective date of the
new order is August 1. 1982.215
It was not until two weeks after the order was signed by the
President and one month after the March 11 letter was sent to the
President, that a response to that letter was received. On April 15,
1982, Kenneth M. Duberstein, Assistant to the President, wrote to
Chairman English :
This is just to assure you that your .March 11 letter, regard-
ing the Executive Order on Security Classification, was
brought to the President's immediate attention. I have also
taken the liberty of sharing a copy with the national security
staff so that they may provide you with a full response.
Please be assured that we appreciated receiving your views,
and those of Congressman Kindness and Congressman Weiss,
on this very important issue.21e
No response from the national security staff was ever received.
The Administration showed even less interest in public comments on
the draft order than it showed in congressional comments. It was never
announced publicly that a revision of the security classification rules
was being considered, and none of the drafts that circulated within
the executive branch was ever released for public review or comment.
However, unofficial copies of the last several drafts were in circulation
among some interested groups outside the executive branch. Since no
drafts were ever formally released by the Administration, some or all
of the reactions from interested groups must have ,been based on the
unofficial copies.
At the May 5 hearing, Garfinkel was asked about contacts by his office
with interested private organizations. Mr. Garfinkel listed seven groups
that were "consulted": American Bar Association Task Force of Gov-
ernment Information and National Security; American Council on
Education; American Historical Association; Association of Former
Intelligence Officers; Center for National Security Studies; Institute
2u This letter is reprinted in Hearings, appendix 3.
2" Hearings at 61.
This letter is reprinted in Hearings, appendix 3.
21447 Fed. Reg. 14874 et seq. (Apr. 6, 1982), reprinted in Hearings, appendix 1.
E.O. 12356 at 5 6.2(d).
21 This letter is reprinted in Hearings, appendix 3.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
of Electrical and Electronics Engineers; and International Classifica-
tion Management Society.217
Garfinkel indicated that no consultations with private organizations
took place before the February 4 draft had already been prepared.
Also, consultations were mostly limited to those groups which had
previously contacted ISOO and expressed interest.218 Other groups
that might have had an interest, but that did not contact ISOO, were
not consulted.219
The Reagan Administration apparently felt that this comment
process for the draft Executive Order was extraordinarily open and
that a broad degree of outside input was solicited. Steven Garfinkel
wrote to Subcommittee Chairman Glenn English :
By soliciting comment from interested congressional com-
mittees and non-government interest groups representing a
spectrum of opinion, the Administration took extraordinary
steps to seek a wide range of comment in an exercise that is
ordinarily closed to anyone outside the executive branch. The
scope of comments received and considered certainly reflects
the broad degree of input solicited from outside the executive
branch.220
The approach taken by the Reagan Administration to the solicita-
tion and consideration of public and congressional comments is in sharp
contrast to that of the Carter Administration when it drafted and is-
sued Executive Order 12065. The Carter Administration revision effort
was characterized by a significant degree of openness and public
participation.
Shortly after taking office, President Carter directed the Domestic
Policy Staff and National Security Council staff to study the security
classification process and make recommendations for change. An inter-
agency task force was established to draft the new order.221
Richard M. Neustadt. Associate Director of the White House Domes-
tic Policy Staff, was co-chairman of the Carter task force. It was recog-
nized at the start that the limited perspectives of a task force composed
solely of civil servants prevented a fully balanced presentation of secu-
rity classification issues:
The problem, however, is that a group of this kind has a
limited perspective. The people who represented the agencies
generally argued for making the classification system highly
inclusive, against any mandate to declassify documents, and
against any oversight of agency classification procedures.
These viewpoints were the natural perspectives of those whose
careers are tied to classification bureaucracies. For example,
mandatory declassification requires review of the affected
documents. which imposes a burden on those people. Over-
sight threatens the autonomy of the classification offices within
each agency.
211 Hearings at 191.
"Id. See also letter from Steven Garfinkel to Glenn English (Jane 22. 1982) (avail-
able in files of Subcommittee on Government Information and Individual Rights.
=19 See, e.g.. letter from Bruce W. Sanford. Counsel. entiety of Professional Journalists,
to Glenn English (Anr. 5, 1982) renrinted in Hearings at 221.
229 Letter from Steven Garfinkel to Glenn English (June 2?. 1982) (available in files
of Subcommittee on Government Information and Individual Rights).
P21 Letter from Richard M. Neustadt, reprinted in Hearings, appendix 5.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
This perspective, however, could not produce a fully bal-
anced view of the security classification system or an executive
order that served (President Carter's] objectives. There was
no way to bring balance to the process from within the Gov-
ernment because there are no institutional advocates for re-
form of a classification process within the agencies.222
Since the Carter White House staff was aware that it did not have
the time or expertise to ensure that all alternatives were considered, a
decision was made to open the drafting process to congressional and
public comment. This was felt to be the "only way to bring a full range
of expert viewpoints into the process and thereby give the decision-
makers a full range of options." 223
Objections were raised within the Carter Administration to the pro-
posal to seek public comments, but the objections were overruled on the
grounds that the quality of the work would be improved :
The decision to circulate drafts in spite of that objection was
based on the judgment that the quality of the work was certain
to be improved through a comment procedure that would give
us additional ideas and viewpoints. Moreover, we felt that a
President should be informed about knowing the problems of
a proposed course of action-both substantive and political-
before he takes that action, not afterwards.224
There was a tremendous amount of public and congressional interest
in the security classification issue, and the Carter Administration re-
ceived over 500 comments on the draft that was circulated .225 Neustadt,
had no doubts that the public comment process was helpful in the prep-
arations of the final draft of what became Executive Order 12065:
I can tell you unequivocally that this public comment proc-
ess improved both the range of options we were able to present
to the President and his senior advisors for their final deci-
sions and the quality of the final order .226
The time that it took 'for compiling and reviewing the public com-
ments was about three weeks. Neustadt did not find this to be a signifi-
cant delaying factor in a revision process that took a year.221 The total
time spent on the revision effort by the Reagan Administration was also
about a year.221
There was some recognition by Reagan officials of the limited per-
spectives and interests of agency classification officials. Presidential
counselor Edwin Meese reportedly said in a mid-March speech before
the National Newspaper Association that the controversy over the
executive order actually was the fault of an overzealous bureaucracy
trying to have its own way. Meese said "that early on, as they always
do, the bureaucracy tested us and they tried to expand classification.
And so I think you '11 find that that is being corrected in the current
drafts of the classification executive order that is now being studied
222 Id.
229 Id.
224 Id.
225 Information Security Oversight Office, "Annual Report to the President" 2 (1979).
Letter from Richard M. Neustadt, reprinted In Hearings. appendix 5.
22 Letter from Richard M. Neustadt, reprinted in Hearings, appendix 5.
21t Id.
zze The revision process started in February. 1981, and the final order was signed on
April 2, 1982. See text accompanying notes 75-94.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
by us."'229 It has never been clear which drafts Meese was referring
to in his statement.
Despite this statement by Meese, the Reagan Administration made
minimal efforts to broaden its perspective on classification issues by
seeking outside advice. Nevertheless, Garfinkel claimed that those few
unsolicited comments that were received were in fact relied upon as a
basis for change. He indicated that at least one provision was included
in the final order because its omission in the February 4, 1982, draft
caused significant concern in the scientific community.230 Citation of
public concern as a explanation for a change in the Executive Order
is recognition on the part of the Reagan Administration that public
comments can be helpful. Nevertheless, no organized effort was made
to seek public comments.
It is impossible to assess how many other interested parties might
have expressed sufficient concern to warrant other changes had public
comments been solicited. Garfinkel indicated that he did eventually
meet with representatives of seven interested organizations. Since the
Carter Administration received over 500 comments on its draft order,
it is reasonable to conclude that there were individuals or organiza-
tions who might have been willing to offer additional points of view.
The consequences of the Reagan Administration's failure to seek
counsel from outside its own ranks were recognized shortly after the
order was signed by the President. Senator David Durenberger, in
introducing a bill to overturn portions of Executive Order 12356, made
the following comments about the role played by Administration
bureaucrats in the drafting of the new order:
This is an order that only a bureaucrat could write. It was
drafted by security bureaucrats, who think only of how to
keep everything secret, and legal bureaucrats who think only
of how to get away with filing fewer affidavits.
Nobody gave much thought to the public's right to know
what big government is doing. Nobody worried about main-
taining public support for the governmental security system,
which is essential if we are ever to stem the flow of leaks that
is the real security problem. Nobody thought to buttress
public support for judges who have almost always followed
the Government to withhold information that it says should
be kept secret.
Ed Meese, the Counselor to the President, was right when
he described the draft executive order as "the fault of an
overzealous bureaucracy trying to have its own way." Alas,
the bureaucracy got its way. It got an order that will increase
secrecy far beyond what experience has taught us is needed
to protect the security of our country, even though that is not
the administration's intent.231
Another contrast between Carter and Reagan Administration ap-
proaches comes from the response of the two Administrations to con-
gressional requests for testimony. The Carter Administration sent a
Justice Department official to a hearing on the security classification
220 Washington Post, March 15, 1982 at All,
230 Hearings at 172. See also text accompanying notes 255-260.
233128 Cong. Rec. 54211 (Apr. 28, 1982) (statement upon introduction of S. 2452, a bill
to provide certain standards for the application of the Freedom of Information Act
exemption for classified information).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
exemption of the Freedom of Information Act in which that official
testified about the proposed revision to the Executive Order on Secur-
ity Classification. 232 The hearing took place while the order was being
revised. However, the Reagan Administration refused a similar re-
cluest for testimony during the period when it was considering revi-
sions.233
At the hearings on May 5, Deputy Assistant Attorney General
Richard Willard said that such an appearance would have been inap-
propriate.234 Exactly what dangers would have resulted were not ex-
plained. There is no evidence that the Carter revision process was dis-
turbed or delayed by the appearance of an Administration witness at
a congressional hearing. 235
Each branch of Government has its own constitutional responsibil-
ities, but nothing in the Constitution precludes cooperation, open dis-
cussion, and public consultation. Indeed, public hearings with partic-
ipation by executive branch officials are a standard feature of the leg-
islative process. It would certainly be unusual for a congressional com-
mittee to refuse to accept testimony from an executive agency on the
grounds that participation by such an agency in the legislative process
would be "inappropriate."
Further, like many pieces of legislation, the Executive Order on
Security Classification establishes guidelines of general applicability
throughout the Government. The rules have an effect on the Congress,
Government contractors, scientists, researchers, the press, and others.
In addition, the order has a direct and immediate impact on the
availability of information under the Freedom of Information Act.
Any information that is properly classified under Executive Order is
exempt from disclosure under FOIA.236 Therefore, an expansion of
the security classification rules means that agencies will be able to
withhold more information from the public under FOIA.
There is no reason why an appearance at a congressional hearing
prior to issuance of an executive order would restrict the President's
ability to act or would create a dangerous precedent. Such an appear-
ance would not necessarily bind the president or future presidents.
This is amply demonstrated by the fact that the Ragan Administra-
tion had no difficulty ignoring the "precedent" established by the Car-
ter Administration when it permitted testimony on a draft security
classification Executive Order.
1. Findings
The Committee finds that the security classification rules estab-
lished under Executive Order are guidelines of general applicability
that affect not only the executive branch, but the Congress, Govern-
ment contractors, Government grantees, and others as well.
The Committee finds that executive branch bureaucrats who partici-
pate in the drafting of security classification rules have very narrow
interests that almost uniformly favor increased secrecy.
232 See "Security Classification Exemption to the Freedom of Information Act" : Hearings
before a Subcommittee of the House Committee on Government O^erations, 95th Cong., 1st
Sess. (1977) (testimony of Assistant Attorney General John Harmon).
233 See text accomranving notes 209-216.
234 Hearings at 157. On the same day that Willard made this statement, the Subcommit-
tee on Civil Service of the House Committee on Post Office and Civil Service held a hearing
on a proposed Executive Order at which renresentatives of two cabinet department testi-
fled. The proposed order would have suspended certain labor negotiation rights of overseas
Defense Department civilian employees.
aia Letter from Richard M. Neustadt, reprinted in Hearings, appendix 5.
5 U.S.C. ? 552(b) (1) (1976).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
The Committee finds that consultation by the executive branch with
the Congress and the public during the decisionmaking process for
security classification rules will help to define and clarify secrecy
issues, identify problems with the operations of the classification sys-
tem, and improve the quality of the final product.
The Committee finds that failure to fully inform the Congress and
the public about proposals to change security classification rules in-
creases mistrust for necessary government secrecy rules.
The Committee finds that the Reagan Administration made only
minimal efforts to seek advice from congressional committees prior to
issuance of the new security classification rules. Although the revision
effort had been ongoing for over a year, selected congressional com-
mittees were formally notified only toward the end of the process and
were originally allotted less than three weeks for comment. The com-
ment period was extended upon request, but the short additional period
of time allowed was insufficient to enable committees to solicit public
comments or to conduct public hearings. Also, the Administration re-
fused to explain on the record what problems the draft order was de-
signed to solve or what the proposed changes were intended to
accomplish.
The Committee finds that the Reagan Administration made no
effort to inform the public of its plans to revise the security classifica-
tion rules or to solicit public comments at a meaningful time during
the revision process.
The Committee finds that virtually all of the controversial provi-
sions of Executive Order 12356 were already in place before a draft
was released for review by anyone out side the executive branch. In
essence, the consultation period was so limited and so late that most of
the decisionmaking had already been completed.
2. Recommendations
The Committee recommends that any future plans to revise security
classification rules, whether by executive order or other executive ac-
tion, be announced publicly.
The Committee recommends that any proposals to change security
classification rules he circulated publicly fora period of at least sixty
days and that public comments be accepted and considered by the
President.
The Committee recommends that any proposals to revise security
classification rules be provided to the Congress with sufficient time to
permit interested congressional committees to consider the proposals,
to hold hearings, and to prepare comments.
The Committee recommends that the National 'Security Adviser or
other official charged with policy responsibility for presidential se-
curity classification rules provide written findings detailing the prob-
lems that any proposed security classification rules changes are in-
tended to solve, and written explanations to the Congress and to the
public of the purpose and scope of any proposed changes. If the find-
ings or explanations are classified, then a classified version should be
provided to congressional committees and an unclassified version
should be made public.
Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
B. REAGAN ADMINISTRATION JUSTIFICATION FOR THE CHANGES IN THE
EXECUTIVE ORDER ON SECURITY CLASSIFICATION
On its face, the Reagan Executive Order on Security Classification
significantly changes or modifies many of the basic classification poli-
cies established by President Carter and previous presidents. Public
witnesses who testified before the Committee pointed out that the
order represented a major change in a thirty year trend toward nar-
rowing and refining classification guidelines and procedures.237
However, statements by Reagan Administration officials about Exec-
utive Order 12356 do not indicate that the changes made by the new
order are as sweeping as others have contended. In fact, there are im-
portant conflicts between the language of the order and the explana-
tions offered by Administration officials. The gap between the order
and the explanations is wide, and it casts serious doubts on the credi-
bility of Reagan Administration spokesmen who have attempted to
minimize the substantive differences between Executive Order 12356
and previous orders.
For example, the Reagan order adds three completed new categories
of information that are subject to classification and modifies one of the
existing categories. In addition, several prohibitions against the clas-
sification of categories of information that appeared in the Carter
order are eliminated. The Reagan order also drops provisions requir-
ing the balancing of the public interest in disclosure against the
national security need for secrecy when deciding whether classified
information should be declassified.
Reagan Administration officials who testified before the Subcom-
mittee denied that these changes would have a significant effect on the
amount of information that will be subject to classification. Steven
Garfinkel of the Information Security Oversight Office stated that the
universe of information that could be classified was essentially the same
under both the Carter and Reagan orders.238
These assurances that the intent and scope of the new order are
narrow are difficult to reconcile with the sweeping language of the
order. All of the changes mentioned above appear to expand the
authority of Government officials to classify information. It is, of
course, possible that the new authority will be used sparingly and
that, as asserted by Administration spokesmen, no additional informa-
tion will be classified as a result of the new order. However, it is well
established that existing classification authority has been regularly
abused in the past.239 Also, the order establishes no new controls that
would be necessary to prevent abuse of classification authority in the
future. Thus, it is reasonable to conclude that, whether or not it was
intended, the new order will lead to the classification of additional
information.
Why were the changes made? During the course of the May 5 hear-
ing before the Subcommittee on Government and Individual Rights,
many different reasons were cited. Some of the reasons were general,
and some were specific justifications of specific provisions. There was
ear Hearings at 50 (statement of Morton Halperin, Director, Center for National Security
Studies) ; Hearings at 87 (statement of Charles Rowe, American Newspaper Publishers
Association) ; Hearings at 97 (statement of Bob Schi'ler, Society of Professional Jour-
nalists).
231 Hearings at 184. See also text accompanying notes 82-84, 114, and 145.
239 See teat accompanying notes 277-283.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
In response to a followup question after the hearing, Garfinkel
confirmed that while other types of information may fall within this
category, "no other general subject areas were involved in formulating
the new `vulnerabilities and capabilities' category." 248
This new classification category was the source of a great deal of
concern on the part of some observers because the scope of the new
category appears to be overly broad. Allan Adler and Morton Halperin
of the Center for National Security Studies testified that the new
category-
would also appear to permit classification of information
concerning data processing, telecommunications and other
"systems, installations, projects or plans" in which the pri-
mary current applications of technological research and de-
velopment have only a tangential or speculative relation-
ship to national security.249
CBS News correspondent Bob Schieffer, representing the Society of
Professional Journalists, expressed similar fears that a vast quantity
of information could be classified under the new category :
I fear that anything open to criticism-any vulnerability-
in a proposed new plane or tank or bullet would be shielded
from scrutiny under the new Order, and that our ability to
learn how well our money is being spent to arm America
would virtually cease.250
If the Administration intended to provide protection only for the
three narrow categories of data that were described by Mr. Garfinkel,
then the new order should have identified the categories much more
specifically. The broad language used in the order increased the con-
cerns of those outside the Administration that sweeping new clas-
sification authority was being claimed. The problem was exacerbated
by the failure of any responsible Administration official to provide,
at an early stage of the revision process, a complete public explana-
tion of the purpose of the change. Also, the discrepancy between the
actual language of the order and the eventual explanation did little
to increase the credibility of Administration spokesmen.
However, even if a narrow explanation had been provided in a
timely fashion, the broad language of the provision still raises valid
questions. It is reasonable to assume that the Administration has
draftsmen capable of writing a more specific provision. Thus, one
might conclude that that the language was intentionally broad and
that the explanation provided to the Committee does not fully reflect
the purpose of the changes.
The language used in the new classification category may continue
to present problems in the future. Classifiers may read the new clas-
sification category and be encouraged by its breadth, as well as by
other changes in the order, to classify documents that do not require
protection. This may increase overclassification 251 and the attendant
expense of protecting, maintaining, and storing the classified infor-
mation.
249 Letter from Steven Garfinkel to Glenn English (June 22. 1982) (available in files of
Subcommittee on Government Information and Individual Rights).
249 Hearings at 76.
250 Hearings at 98.
ss1 See generally text accompanying notes 275-291.
Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
The credibility of the Administration on classification issues was
also diminished by the use of inconsistent justifications for some of
the changes in the order. For example, some provisions were deleted
from the Carter order on grounds of economy or redundancy. But
this principle was not applied consistently. Some admittedly redun-
dant provisions were not removed and other provisions were added
even though there was little question that the existing language was
adequate.
The clearest instance of the inconsistent application of the redun-
dancy principle comes from several provisions of the Carter order
that limited classification authority.252 The February 4, 1982 draft
order proposed to remove all three limitations because they were "self-
evident." 253 In response to a question, Garfinkel indicated that the
provisions were redundant and that they had been removed from
early drafts because of "a conscious effort to keep the order as short
as possible and as simple as possible." 254
The final order retained one of the limitations, 255 the one prohibit-
ing the classification of basic scientific research not clearly related to
the national security, and left out the limitations prohibiting classifi-
cation of private technology and references to classified documents.
The explanation for the difference in treatment was "the provision
regarding basic scientific research raised such concern within the
scientific community that we thought it was better to be redundant." 256
Garfinkel stated that there wasn't much concern expressed about the
other two limitations.
This explanation.confirms the inconsistent application of the redun-
dancy principle but suggests that it was sacrificed in this instance, to a
higher principle. This would be admirable if it could be accepted as
offered. But it is not apparent that the expression of concern from the
scientific community was limited to the basic scientific research limita-
tion. For example, a February 26. 1982, letter from Frank Press. Presi-
dent of the National Academy of Sciences to National Security Adviser
William Clark complained about the deletion of all three limitations.257
Similarly, witnesses at the March 10 hearing (who were not scientists)
who complained about the deletion of limitations did not limit their
comments to the scientific research limitation.253
Further, it is difficult to understand why the Administration relied
on expressions of concern by outsiders as a reason for change. The draft
order was never circulated publicly by the Administration. No attempt
was made to collect comments from a range of interested parties.259
Other concerns of the few groups that were able to obtain information
about the revision process were ignored. What it was about this single
comment from this single interest group that made it more compelling
252 E.O. 12065 at 11-6.
=3 Information Security Oversight Office, "Comparison of Executive Order 12065 and
the [February 4. 1982] Draft Order" 3, reprinted in Hearings, appendix 2.
254 Hearings at 172.
sss E.O. 12356 at ? 1.6(b).
256 Hearings at 172.
257 A very similar letter from Frank Press to Glenn English is reprinted in Hearings, ap-
pendix 6. A copy of the letter from Frank Press to William Clark is available in the files
of the Subcommittee on Government Information and Individual Rights.
zsa See, e.g., Hearings at 81-82 (Statement of Allan Adler and Morton Halperin, Center
for National Security Studies).
29 See text accompanying notes 217-221.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
than all of the other comments from this and from other interest groups
is a mystery.260
At the same time that some classification limitations were removed
as redundant, other apparently redundant language was added else-
where. The classification category covering intelligence activities was
modified to include a parenthetical reference to "special activities." 261
Yet neither Garfinkel nor Willard expressed any serious doubt that
special intelligence activities were classifiable under the Carter order.
The only explanation for the addition was because the issue was ques-
tioned once by a person seeking information from the CIA.262
Subcommittee Chairman English, noting that neither Garfinkel nor
Willard had any problem with the classification of special intelligence
activities, pressed to learn "where did the ambiguity come from ?" The
response from: Garfinkel was "I do not know what gave rise to the
particular concerns of the Agency." When asked if he had pursued the
issue with the CIA, Garfinkel responded : "It didn't seem to be. such a
big deal that it was necessary to pursue any further." 263
The savings resulting from the deletion of a few words from the
order were admittedly insignificant.264 But these savings were appar-
ently enough to justify the elimination of words in those instances
where the words imposed restrictions on classification authority.
Where words increased the ability of the government to classify data,
however, the cost of those extra words was truly insignificant, and
the words were added.
A final example of the difficulties arising from the explanations pro-
vided comes from the new classification category covering "confiden-
tial sources." 265 In this case, the explanation offered to the Committee
was devoid of any apparent logic.
Under the new order, "confidential source" is defined to include :
any individual or organization that has provided, or that
may reasonably be expected to provide, information to the
United States on matters pertaining to the national security
with the expectation, expressed or implied, that the informa-
tion or relationship, or both, be held in confidence.266
In explaining the new classification category, Mr. Garfinkel stated
that it would not give the government authority to classify informa-
tion that was not classifiable under the Carter order. The new category
was added because "it was expressed to me and others that some
266 One part of the February 4, 1982, draft that was criticized and was changed in the
final order involved the portion marking requirements. The February 4 draft proposed to
delete the Carter order's strict rules for portion marking. The Carter order required that
each portion of a classified document be marked so that classified portions could be dis-
tinguished from unclassified portions. E.O. 12065 at ? 1-504. The draft Reagan order
weakened the portion marking rules by requiring marking only when a classified document
was transmitted outside the originating agency. See ? 1.5(b) of the February 4. 1982. draft,
reprinted in Hearings. appendix 2. The proposed change to the portion marking rules was
criticized by Allan Adler and Morton Halperin of the Center for National Security Studies,
and by Bob Schaeffer, Society of Professional Journalists. See Hearings at 78 and 97-98.
No detailed explanation was provided for the delegation or restoration of the portion
marking rule. The restoration of the Carter order portion marking rule was one of the
few substantive differences between the February 4 draft and the final order as signed
by the President.
X63 D.O. 12356 at ? 1.3(a) (4). See text accompanying notes 1.11-113.
262 Hearings at 203 (Statement of Steven Garfinkel, Director. Information Security
Oversight Office).
266 Id.
2" Id. at 216 (Statement of Steven Garfinkel, Director, Information Security Oversight
Office).
266 E.O. 12356 at Q 1.3(a) (9). See also text accompanying notes 105-110.
266 E.O. 12356 at ? 6.1(f).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
agencies have severe problems with the idea of identifying some of its
sources as intelligence sources. It creates problems in their relation-
ships with these sources." 261
This explanation is not believable. How would a source know how its
identity was classified by an agency? When information is classified,
the applicable classification category is not identified on the document.
The document is simply marked as containing classified information
of a particular level. There is no practical way for a source to know on
what basis its identity was kept secret.200 Further, it is not apparent
why a source would care what bureaucratic reasoning was employed
as a justification for that secrecy as long as secrecy was assured.
Further, the possibility that the identity of a source might be
requested under the FOIA fails to provide an adequate explanation.
First, the identity of sources is protected by statute,209 and identifying
information is not available under the FOIA. Second, even if the
question of disclosure were seriously contested in an FOIA case, the
identity of the source would not be revealed during the course of the
case. 'Thus, there is no way that the classification category could be
publicly linked with the identity of the source. Lastly, in the extraor-
dinarily unlikely event that a source were identified, surely it would
be the fact of identification rather than the category of classification
that would be most disconcerting to all parties involved.
Finally, the vagueness of the Carter order classification categories
would have permitted sources to be classified without treating them
as intelligence sources. The final classification category in the Carter
order allowed the classification of "other categories of information
which are related to national security 11.270 If the classification of some
sources as intelligence sources were a significant problem, then the
classification could have been easily justified under this category on
other grounds. However, it appears that the problem was not sig-
nificant enough to warrant this simple step.271
In an attempt to clarify the new category, Subcommittee Chairman
Glenn English wrote to Steven Garfinkel after the hearing asking him
to explain how sources know on what basis their identity is classified.
Garfinkel responded :
The Department of State has reported to ISOO the follow-
ing scenario in which a source may determine the manner in
which his or her identity is being protected. There may be a
press briefing in which the Department of State releases cer-
tain information and indicates that the information was re-
ceived from a "confidential" source, rather than an "intel-
ligence" source. The source can recognize himself or herself
from the information released and reported in the press. Us-
ually, the source does not want to be identified, even indirectly,
as an "intelligence" source, a label that may cause uneasiness
or worse in some parts of the world. Even though the likeli-
'8' Hearings at 204.
268 If the confidential source requested his own records under the Freedom of Informa-
tion Act, the request could be denied on the grounds that the records were classified. See
5 U.S.C. ? 552(b) (1) (1976).
289 See 50 U.S.C. ? 403g (1976) ; 5 U.S.C. ? 552(b) (7) (D) (1976).
279 D.O. 12065 at ? 1-301(g) .
2U The authority to establish additional categories was only used once. See Hearings at
205 (Statement of Steven Garfinkel, Director, Information Security Oversight Office).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
hood of his or her name surfacing is remote, it is still deemed
preferable to be identified as a "confidential" source.272
This explanation sheds no more light on the subject than the ex-
planations at the hearings. What the State Department discloses at
press briefings is totally within the control of the State Department.
It may use any labels it wishes to categorize sources referred to at those
briefings. This additional explanation is no more credible than the
earlier explanations.
In the same letter, Garfinkel was asked what type of problems were
reported to ISOO with respect to the classification of sources as intel-
ligence sources. The answer :
The Department of State has also reported that it is con-
cerned lest a court consider that sources of, for example,
clearly political or economic information are not, in fact,
intelligence sources within the generally accepted meaning of
the term, and that they cannot therefore, be validly protected
under the category "intelligence sources and methods." Spe-
cific provision is required to protect these non-intelligence
sources of information. 273
Garfinkel cited no actual instances of problems that have arisen in?
this area. No cases were cited for the proposition that. sources of polit-
ical or economic data are not "intelligence sources" within the gen-
erally accepted meaning of the term. In fact, no court cases were
cited in connection with the alleged problems with the protection of
confidential sources. The "problems" that Garfinkel referred to appear
to be hypothetical at best and non-existent at worst.
It is more likely that the addition of "confidential sources" as a
classification category was done in order to allow a definition to be
included. The new definition covers not only actual sources of informa-
tion but "potential" sources as well. Although Administration officials
denied that the new definition would permit the classification of any
information that was not classifiable under the Carter order,274 it is
not certain that potential sources could have been classified under the
Carter order. No information was provided explaining why authority
to classify "potential sources" was needed.
1. Finding
The Committee finds that the Reagan Administration failed to
provide the Congress and the public with a full and complete explana-
tion of the changes that were made in the Executive Order on Security
Classification. Many of the explanations offered after the order was
signed by the President were inadequate, inconsistent, incomplete, or
not credible. It remains uncertain why many of the changes were
made, and there is substantial doubt that the changes could be justified.
f. Recommendation
The Committee recommends that changes in the Executive Order
on Security Classification that increase classification authority be
made only when a concrete problem with the protection of sensitive
information has been clearly identified. Amendments to the text of
272 The letter is dated June 22, 1982, and is available in the files of the Subcommittee on
Government Information and Individual Rights.
ara Id.
~' Hearings at 184 (Statement of Steven Garfinkel, Director, Information Security Over-
sight Office).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
the order should be precisely and narrowly drafted to address the
problems that have been identified. In making changes, the President
should give careful consideration to the interpretation likely to be
given to new provisions by those with classification authority who
have a known predilection to overclassify information and to abuse
existing classification authority.
C. O VERCLASSIFICATION
There is no doubt whatsoever that classification authority is used
to protect information that does not require protection in the inter-
est of national defense or foreign policy. This has been a consistent
finding of presidents, congressional committees, commissions, and
other observers.275 Presidents Eisenhower, Nixon, and Carter each re-
vised the rules for security classification in order to restrict and con-
trol the use of classification authority.276
Recent studies by the General Accounting Office provide additional
evidence that overclassification continues to be a significant problem.
A 1979 GAO report on classification problems at the Department of
Defense, the agency that classifies more information than any other
government agency, found that about 24 percent of the classified docu-
ments reviewed contained one or more examples of overclassification.
Both Executive orders [E.O. 11652 and E.U. 12065] and
the various implementing instructions describe the types of
information that should be classified for national security
reasons-to preclude overclassification and underclassifica-
tion and to make information about the Government avail-
able to the maximum extent possible. Nevertheless, we identi-
fied examples of improperly classified information at each
DOD installation and office visited. Of the 556 documents
reviewed, 133, or about 24 percent, contained one or more
examples of improper classification. None of these cases in-
volved classified information which was incorrectly treated
as unclassified.
The following are some of the more significant problems
noted :
Information not related to national security was classified.
References to classified documents were classified.
The same information was classified inconsistently.
Information that lost some of its sensitivity was not down-
graded.
When there was doubt about the level of classification, a
higher classification level was assigned.277 [Italic supplied.]
A later GAO report on the classification practices of Defense De-
partment contractors with derivative classification authority found
similar problems of overclassification. There are over ten thousand
contractor facilities which have been cleared to handle classified infor-
21'' See, e.g., the discussion of previous studies of security classification systems in 1973
Government Operations Security Classification Report, supra note 6 at 15-27. See also
Committee on Government Operations, "Availability of Information from Federal Depart-
ments and Agencies (Department of Defense)," H.R. Rep. No. 1884, 85th Cong., 2d Sess
(1958).
271 See generally text accompanying notes 37-73.
277 General Accounting Office, "=Continuing Problems in DOD's Classification of National
Security Information" 12 (LCD-80-16) (October 26, 1979).
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
mation, and the GAO estimated that contractors were holding approxi-
mately 16 million classified documents. 2711
The GAO auditors found that over 50 percent of the documents they
reviewed were overclassified :
We reviewed 235 classified documents to determine if vari-
ous portions of the documents were classified correctly. We
discussed the documents with the contractor employees who
had derivatively classified the information and verified the
basis for classification-classification guides or documents al-
ready classified. Of the 235 documents, 119, or about 51 per-
cent, contained one or more examples of improper classifica-
tion. Of the 119 documents, 109 had portions that were over-
classified, and 10 had portions that were both underclassified
and overclassified 279
The problem of overclassification was recognized early in the Reagan
Administration by top officials. In an interview with Meg Greenfield,
editorial page editor of the Washington Post, Presidential Counselor
Edwin Meese III responded to a question about overclassification :
Question. But haven't you found since you have been here
that there is too much classification?
Answer. Oh, yes, I think there is way too much classifica-
tion, and I think that's one of the things that has given rise
to the relative freedom with which some of the reporters use
this material, and I think that's one of the problems of gov-
ernment-that is, the overclassification of documents. You
really should only classify something if its revelation would
actually harm the national . security. Not that it would em-
barrass someone or make someone look bad.28?
At the May 5 hearing before the Subcommittee on Government In-
formation and Individual Rights, Steven Garfinkel, the Director of
the Information Security Oversight Office was asked if he agreed with
Mr. Meese's statement that "there is way too much classification."
Garfinkel responded :
I would 'agree. There are many documents that were orig-
inally classified that didn't merit that classification. Now, as
to oversight experience, on our reviews, about 5 percent of the
documents that we review clearly don't merit classification
and we request that the agency review it for purposes of
declassifying it.281
Later in the hearing, Garfinkel elaborated :
You could say any overclassification is way too much, but
I think that a 5-percent figure, or perhaps even a 10-percent
figure, is not to be unexpected. Bureaucrats, like most people,
are naturally very cautious. If the situation ordinarily arises
where they have information that they believe is sensitive and
they have to make a decision about it, whether it has been en-
der the present order or not, generally the people are classify-
s78 General Accounting Office. "DOD Should Give Better Guidance and Training to Con-
tractors Who Classify National Security Information" 2 (PLRD-81-3) (March 23, 1981).
171, Id. at 14.
=? Washington Post, July 7, 1981.
nearings at 146.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
ing this information. So there is a tendency to classify the
information when you are in doubt. Sometimes that results
in overclassification.282
Garfinkel's estimate of five or ten percent government-wide is con-
siderably less than the amount of overclassification found by GAO at
the Department of Defense. The Department of Defense originally
classifies more documents than any other government agency.283
The fact that overclassification exists is undisputed. However, the
serious consequences of overclassifiction are only rarely considered. Not
only does the overclassification of information interfere with the pub-
lic right to know about the operations of government, but it interferes
with the security that classification system is designed to protect. In
the Pentagon Papers case, Justice Stewart made the point this way :
For when everything is classified, then nothing is classified,
and the system becomes one to be disregarded by the cynical
or the careless, and to be manipulated by those intent on self-
protection or self-promotion. I should suppose, in short, that
the hallmark of truly effective internal security system would
be the maximum possible disclosure, recognizing that secrecy
can best be preserved only when credibility is truly main-
tained.284
In its report on classification problems at the Department of Defense,
the General Accounting Office pointed out that the consequences of
overclassification include reduced public confidence, weakened protec-
tion of truly sensitive information, and increased cost :
Even though the Executive order [E.O. 12065] and imple-
menting instructions clearly describe the types of information
that should be classified and even specify other types that
should not be classified, our review indicated that a sizable
percentage of the information classified was not classified
properly. Improper classification causes less information to be
made available to the public, reduces public confidence in the
system, weakens protection for truly sensitive information,
and increases administrative costs.285
At the March 10, 1982, hearing, Glenn English, Chairman of the
Government Information and Individual Rights Subcommitttee, com-
pared classification and money :
[C]lassification is like money. It cheapens as you expand it.
The more money you print, the cheaper it is. The more you
print, the more inflation you get. The more classification you
use, the cheaper the classification mark itself becomes.286
President Carter personally noticed that overclassification was a se-
rious problem. In a letter to the Committee, former White House
28 Hearing at 151-52.
sea General Accounting Office, supra note 277, at 2. See also General Accounting Office,
"Im
proved Executive Branch Oversight Needed For the Government's National Security
Information Classification Program", supra note 69 at 6-7; Information Security Over-
sight Office. "Annual Report to the President (Fiscal Year 1979)" 36.
2% New York Times Co. V. United States, 403 U.S. 713, 729 (1971) (concurring opinion).
z_ General Accounting Office, supra note 277, at 17.
m Hearings at 15.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
staffer Richard M. Neustadt told of Carter's experiences and provided
an example of the consequences of classification "grade escalation" :
[President Carter] recalled that during his period of mili-
tary service the classification "Confidential" was reserved for
documents of extraordinary sensitivity. Since then, overuse
has downgraded these classification levels, and routine rub-
berstamping has undermined respect for the whole system.
Everyone who participated in the process of writing E.O.
12065 agreed that vast amounts of routine paperwork con-
taining no real national security information had been classi-
fied at the "Confidential" level. As a result, genuine national
security material required a higher classification level. This
problem of "grade escalation" had reached the point that in
the White House it was widely assumed that any document
classified below the "Top Secret" level was not worth reading.
This escalation forced the creation of extra classification
levels above "Top Secret," such as the eleborate "codeword"
classification schemes administered by the intelligence com-
munity.287
The Reagan Administration does not dispute the negative conse-
ouences of overclassification. Steven Garfinkel of the Information
~ecurity Oversight Office said that the "most critical problem con-
cerning overclassification is its impact on public disclosure." 288 He
went on to described other results :
There are other reasons that ISOO does not condone over-
classification, and gives priority to preventing it through its
oversight activities. Overclassification results in far greater
costs to the government and taxpayer to create, store, trans-
mit, maintain and destroy information. It also lessens intra-
agency or inter-agency accessibility, which may impact ad-
versely on the decision-making process. (These problems
are also a product of overgrading, but to a lesser extent.)
Finally, and most importantly save for public access, a pat-
tern of overclassification tends to subvert the entire informa-
tion security system, jeopardizing the respect necessary to
protect information that truly warrants classification 289
Overclassification also interferes with the functioning of the Con-
gress. Rep. Lee Hamilton, Chairman of the Subcommittee on
Europe and the Middle East of the House Foreign Affairs Committee
and a member of the Permanent Select Committee on Intelligence,
described this aspect of overclassification in a statement on the House
floor made shortly after President Reagan signed Executive Order
12356:
Overclassification has more of an effect on the proper func-
tioning of Congress than one might imagine. In principle,
Members of Congress are supposed to have access to classi-
fied material, and constitutionally they are allowed to discuss
sensitive material in public if it pertains to valid legislative
duties. In practice, however, overclassification makes quite
284 Letter from Richard M. Neustadt, reprinted in Hearings, appendix 5.
288 Letter from Steven Garfinkel to Glenn English (June 22. 1982) (available in files of
Subcommittee on Government Information and Individual Rights).
asa Id.
Approved For Release 2007/06/14: CIA-RDP83M00914R001800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
a difference. Many legislators hesitate to discuss secret docu-
ments even if they think the classification is unjustified.
Given the tight schedules on Capitol Hill, most lawmakers
are unwilling to spend time getting such documents declassi-
fied. Also, the level of classified information presented in
closed briefings varies with the Members of Congres who
attend, and Members who have direct access to most classified
material, such as those who serve on the Intelligence Com-
mittees, are restricted by internal rules in what they may dis-
cuss with their colleagues. Overclassification breeds mistrust
between Congress and the administration, friction grows
when legislators can find out more from the newspaper than
they can from the administration. Overclassification also in-
creases the likelihood that claims of executive privilege will
lead to a constitutional clash.
As a member of the House Intelligence Commitee, I appre-
ciate the fact that access to certain information must be
restricted. But as a member of the House Foreign Affairs
Committee, I know how difficult it is to formulate and explain
foreign policy when the administration acts on undisclosed
information, when the access to sources is overly restricted,
or when the release of information is unnecessarily delayed.
My impression is that Congress' current access is not what it
should be. The President's proposal will impair it further.290
GAO has identified some of the causes of overclassification. They in-
clude lack of knowledge of classification requirements and a lack of
incentive against excessive classification :
We believe that the examples described in this chapter dem-
onstrate that a serious problem exists within DOD in that in-
dividuals who originally or derivatively classify information
either are not fully knowledgeable of the requirements of the
order and implementing instruction or prefer to follow a
course of action that would result in a lesser penalty to them if
they incorrectly classify information 291
There is little reason to think that Executive Order 12356 will reduce
the unnecessary classification of government information. All of the
major changes made by the new order loosen the restrictions on classi-
fication. These changes will have the effect of-
Increasing the amount of information subject to classification
by the addition of broad, new classification categories;
Weakening the minimum standard for determining whether
information qualifies for classification by dropping the require-
ment that damage to the national security be "identifiable";
Dropping the balancing test that required classifiers to con-
sider the public interest in disclosure against the need to protect
information ;
Removing limitations on abuse of classification authority that
appeared in the previous order; and
Permitting the reclassification of information that was de-
classified and publicly released.
210128 Cong. Rec. E1564-65 (Apr. 6, 1982).
'2 General Accounting Office, supra note 277, at 17.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
In fact, there is virtually nothing new in Executive Order 12356 to
inhibit the overclassification of information. This is not surprising.
Although Administration officials readily acknowledged that over-
classification of information was a problem under previous classifica-
tion rules, the solution of this problem was never identified as a major,
or even minor, goal of the new order. No Administration spokesman
claimed that the order was designed to restrict classification authority
for the purpose of preventing overclassification.
Perhaps the best that can be said for Executive Order 12356 on the
issue of overclassification is that as more and more documents qualify
for classification, there will necessarily be fewer documents that can
be improperly classified. The Reagan Administration has changed the
rules in a way that will define away the problem. This is akin to re-
sponding to a prison escape by expanding the boundaries of the prison
so that it can be claimed that the escaped prisoners are no longer tech-
nically outside of prison facilities.
There is an apparent need for some new administrative control to
prevent the overclassification of information in the name of "national
security." Previous Executive Orders were not successful in limiting
overclassification, and it is unlikely that Executive Order 12356 will
do anything to improve the situation.
Great progress has been made in recent years in the management,
control, and use of information. Some of the progress has been the
result of new technology,292 and some has been the result of better
understanding of the importance of information as a resource.293 A
number of recent federal laws have addressed the maintenance and
use of information by the federal government'294 the credit reporting
industry,295 and schools and universities.29e States have also been ac-
tive in passing laws affecting information.291
The new understanding of information and information policy may
provide some useful ideas for controlling classified information and
for preventing overclassification. One of the most interesting recent
federal laws that may provide a model for a classification control sys-
tem is the Paperwork Reduction Act of 1980.298 The objectives of that
Act include reducing the cost of collecting, managing, and disseminat-
ing information by federal agencies; ensuring that agencies collect
only as much information as they need and can use effectively, and
reducing the information burden imposed on the public by the gov-
ernment.299 The Office of Management and Budget, which overseas the
operation of the Act, claims that the information collection budgeting
system has allowed a 17 percent reduction in federal paperwork
burden during fiscal years 1981-1982.3""
97 See, e.g., "The Information Science and Technology Act," Hearings on H.R. 3137
before the Subcommittee on Science, Research and Technology of the House Committee on
Science and Technology, 97th Cong., 1st Sess. (1981) ; Office of Technology Assessment,
"Computer-Based National Information Systems" (1981).
See, e.g., R. V. Head. "Federal Information Systems Management : Issues and New
Directions" (1982) (Brookings Institution) ; National Telecommunications and Informa-
tion, U.S. Department of Commerce, "Issues in Information Policy" (1981).
20+ See, e.g., Privacy Act of 1974, Pub. Law No. 93-579, 88 Stat. 1896 et seq. (1974)
(codified in part at 5 U.S.C. ? 552a (1976)).
sec See, e.g., The Fair Credit Reporing Act, Pub. Law No. 91-508, 84 Stat. 1127 et seq.
(codified at 15 U.S.C. ? 1681 et seq. (1976)).
zsa Family Educational Rights and Privacy Act, Pub. Law No. 93-568, 88 Stat. 1858
(1974) (codified at 20 U.S.C. 1232g (1976) ).
aer See, e.g., R. E. Smith, "Compilation of State and Federal Privacy Laws" (1981).
208 Public Law 96-511, 94 Stat. 2812, 44 U.S.C. ? 3501 et seq. (Supp. IV 1980).
20 See Office of Management and Budget, "Managing Federal Information Resources,
First Annual Report Under the Paperwork Reduction Act of 1980" 1 (April 1, 1982).
"Id. at 11.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
 Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7
Information that is classified in the interests of national defense or
foreign policy differs in many respects from information maintained
in the private sector or maintained elsewhere in the federal govern-
ment. Nevertheless, the principles of information management that
have been recognized in recent years may still be applicable to the
control of classified information and the prevention of overclassifica-
tion. Given the high costs of overclassification, further study is needed.
1. Findings
The Committee finds that abuse of classification authority and over-
classification of government information continues to be a serious
problem. Overclassification results in unnecessary restrictions on the
public availability of information, a reduction in public confidence in
the classification system, a weakening of the protection for informa-
tion that is truly sensitive, and an increase in the cost of government.
The Committee finds that Executive Order 12356 offers nothing that
will address the problems of overclassification. Unless new action is
taken to control overclassification, the new order is likely to make mat-
ters worse because it gives classifiers vaguer guidelines, fewer restric-
tions, and unnecessary additional classification authority.
2. Reconwnendation
The Committee recommends that the National Security Council,
the Information Security Oversight Office, and the Office of Informa-
tion and Regulatory Affairs of the Office of Management and Budget
work together to develop and apply new methods of preventing over-
classification and of limiting abuse of the classification authority con-
tained in Executive Order 12356. These agencies should report to the
Congress on their efforts no later than September 30, 1983.
Approved For Release 2007/06/14: CIA-RDP83M00914RO01800100054-7