Yale University Press / New Haven and London

Secret Agencies
U.S. INTELLIGENCE IN A HOSTILE WORLD

Loch K. Johnson

There are many elements that go into every important decision.

In the first place, you must try to grapple with the facts.
What is the actual situation?
Secretary of State Dean Rusk to Eric Goldman (January 12, 1964)

CONTENTS

Preface ix
List of Abbreviations and Acronyms xv
Chapter 1 / The Meanings and Methods of Intelligence 1
Chapter 2 / The Evolution of the Intelligence Missions 31
Chapter 3 / The Ethics of Covert Operations 60
Chapter 4 / Intelligence Accountability 89
Chapter 5 / The Distinctiveness of American Intelligence 119
Chapter 6 / Intelligence and Economic Security 146
Chapter 7 / An Assessment of American Intelligence 174
Appendix A / Directors of Central Intelligence 207
vii

Appendix B / Chronology of the Cuban

Missile Crisis 209
Notes 211
Index 251

viii

Inhalt

PREFACE

This book examines how, and how well, the intelligence agencies of the
United States have been used by government officials since the end of World
War Two to guard and advance the global interests of the nation. My purpose
is to help inform the American people about the hidden side of their government. For democracy relies on a knowledgeable citizenry to provide general
guidance to those few individuals who make foreign policy decisions on their
behalf.
Americas secret agencies engage in three primary missions. First and foremost, they are expected to gather and interpret information from around the
world (referred to by intelligence officers as collection and analysis). Second,
the agencies are expected to protect U.S. government secrets from espionage
by other governments (counterintelligence). Third, from time to time they have
been directed to oppose the nations adversaries through the use of aggressive clandestine operations abroad (covert action). Throughout the Cold War
(194591) the Soviet Union was the nemesis of American foreign policy and
hence the number-one target of the intelligence agencies. The containment of
Soviet-inspired communism was the preeminent objective that shaped Amerix

icas relations with the rest of the world and provided the raison dtre for the
secret agencies.
In an earlier study, A Season of Inquiry (1985), I wrote about the beginning
of a new era for American intelligence ushered in by a series of spy scandals.
In the benchmark year 1975 government investigators had accused the secret
agencies of conducting espionage against American citizens, the very people
they had been created to protect. Probes by the executive and legislative
branches chronicled a long list of Orwellian excesses: spying on civil rights activists and Vietnam War dissenters, plotting the assassination of foreign leaders, and running unsavory clandestine operations meant to undermine or destroy regimes considered anathema to the interests of the United Stateseven
democracies (Chile is only the most well-known case).
In the light of this jarring breach of trust, U.S. intelligence agencies would
no longer enjoy the same breadth of discretion in the conduct of covert operations around the globe as they had had before. Henceforth officials within the
executive branchand, in a dramatic expansion of supervision, the legislative
branch as wellwould attempt to hold the nations spymasters to a higher standard of accountability. A Season of Inquiry traced the debates about the future
of intelligence that took place during the Year of Intelligence, as some officers of the Central Intelligence Agency (CIA) remember 1975, orfor the
more embitteredthe time of the Intelligence Wars. Scandal had forced both
the president and the Congress to grapple with the dilemma of how to tighten
control over the secret agencies without stifling their initiative and morale in
the struggle against Americas external enemies. A unique experiment in intelligence accountability had begun.
My second study of intelligence, Americas Secret Power (1989), examined
the effectiveness of the new accountability during its first decade, including the
performance of neophyte House and Senate intelligence oversight committees,
the stringent approval and reporting requirements for sensitive operations, and
the new Intelligence Oversight Board (IOB) set up in the executive office of
the president. The verdict: even after ten years, the new relationships remained
rough-hewnand they had failed altogether to prevent the Iran-contra affair
of 198687. Nevertheless, the new methods of democratic control had worked
most of the time, and clearly they represented a vast improvement over the
open-ended authority granted the secret agencies throughout the earlier era of
tolerant neglect (194574).
Americas Secret Power explored a number of problems that continued to
disturb the balance between accountability and effectiveness for the bakers
dozen departments and agencies that make up the so-called intelligence comx

Preface

munity (IC). Seven major sins of intelligence emerged from the study, the
most damning of which was the failure to provide policymakers with objective
information. The book identified a variety of pathologies that weakened the
core intelligence mission of information collection and interpretation. It also
explored the elaborate relationships that had evolved since the end of World
War Two between the secret side of government and other American institutions, particularly the media and the universities.
In the same year Americas Secret Power was published, history offered up
one of its rare sea-changes in world affairs. In November 1989 the Berlin Wall
was brought down suddenly, and the Soviet Union soon came tumbling after.
In a quick succession of astounding and exhilarating events, the Cold War was
over. These events, culminating in a splintering of the Soviet empire in 1991
into its constituent republics and once-captive nations, brought to the forefront
troubling questions about U.S. intelligence capabilities. How could the secret
agencies have failed to anticipate the dissolution of Americas deadliest international rival? What would happen to the clandestine service now that the Cold
War was over?
The present book carries forward my research into the netherworld of intelligence, further unfolding topics taken up earlier and setting out in new directions as wellamong them the debate over whether the United States should
engage in a more aggressive use of economic espionage against allies and enemies alike. I consider a range of ethical questions surrounding the use of covert
operations, while continuing to follow the thread of intelligence accountability that weaved through the companion volumes. I offer an updated appraisal
at the close of a second decade in this nobleand often shakyexperiment
meant to bring some semblance of democracy into the darkest corners of American government.
I begin by examining what is meant by intelligence, why nations with
global interests consider it important to have secret agencies, and how the use
of intelligence is beset with existential vexations (chapter 1). Chapter 2 brings
a broad historical overview of Americas secret operations abroad from the
Cold War to the present. The purpose of this chapter is to indicate how the emphasis placed on the different intelligence missions by the government has fluctuated over the years. The moral implications of clandestine operations are assessed in chapter 3, where I offer a set of guidelines for a more ethical approach
to the use of secret power.
The question of intelligence accountability, a central concern for any probe
into the interstices between secrecy and democracy, is taken up in chapter 4
with a close look at how well overseers have monitored the intelligence agenPreface

xi

cies through Congressional hearings. Chapter 5 contrasts the U.S. approach to

intelligence with that of other countries.
The issue of intelligence and economic security is the focus of chapter 6.
The key question here is: Should this nations secret agencies aid the American
business community in its struggle for success in the global marketplace
against adroit foreign competitors like Japan and Germany? The book concludes in chapter 7 with an evaluation of how well Americas intelligence agencies fared during the Cold War against the USSR, a totalitarian state bristling
with nuclear weapons and endowed with powerful secret services of its own.
Have the American people been well served in their quest for peace and security in a world marred by violence, intrigue, and uncertainty? Do the billions
of taxpayer dollars spent on intelligence over the past fifty years add up to a
wise investment or a foolish waste of money?
The methodology in this and my other books has been straightforward:
study everything of a serious nature that has been written on the subjecta
steadily burgeoning literature of government documents, periodicals, and
scholarly treatisesand interview as many intelligence professionals and outside experts as possible.1 The interviews have been with men and women at all
levels of the secret agencies and with their overseers in the executive and legislative branches, as well as with a wide range of academic specialists from the
United States and abroad.
Aunifying theme binds together this corpus of research. The information provided to policymakers by the intelligence community often contributes vitally
to the making of sound decisions, giving the secret agencies a role of unquestionable importance to the nations well-being. Yet the evidence clearly reveals
that, at the same time, the intelligence agencies have the capacity not only to
safeguard democracy but to subvert it as well. Moreover, the information they
have provided to the nations leaders has at times been wrong, as a result of errors in judgment or bias in reportingor because many things about the world
are simply unknowable. Thus the intelligence agencies indeed warrant the support of Americans, but they also require a close watchfulnesseven wariness.
This book has benefited greatly from discussions with intelligence officers
and overseers, most of whom have requested anonymity for professional reasons. I thank them profusely for their patience and generosity. Some of the
thoughtful people with whom I have spoken can be openly thanked, though,
beginning with Les Aspin, the former secretary of defense and chairman of the
Commission on the Roles and Capabilities of Intelligence. He was a wonderful source of encouragement for this project; he read and commented on porxii

Preface

tions of the manuscript as I went along, and was especially helpful with chapter 7. His premature death in 1995 was a tragedy for the country and for the
many of us who valued his friendship and keen analytic mind.
Others I am pleased and able to thank openly include James A. Barry, David
D. Gries, Arthur S. Hulnick, Carol Minor, Kay Oliver, Hayden B. Peake, Donald P. Steury, and Michael A. Turner of the CIAs Center for the Study of Intelligence and its Office of Academic Affairs; Harold P. Ford, Joseph S. Nye, Jr.,
and Gregory F. Treverton, all formerly with the National Intelligence Council;
Douglas J. MacEachin, formerly deputy director of intelligence at the CIA;
George J. Tenet, a former senior intelligence official on the National Security
Council (NSC) and presently the deputy director of central intelligence; the late
James J. Angleton, chief of CIA counterintelligence; John T. Elliff, Senator
Wyche Fowler, Richard H. Giza, Thomas K. Latimer, Senator Sam Nunn, and
Paula L. Scalingi, former legislative overseers; Carol Rindskopf, former general counsel of the CIA; Frederick P. Hitz, the CIAs inspector general; Dean
Rusk, former secretary of state; former intelligence officers George Carver,
Dr. Ray S. Cline, Jack Davis, and Walter Pfortzheimer; and each of the directors of Central Intelligence from 1966 to 1995Richard Helms, James R.
Schlesinger, William E. Colby, George Bush, Adm. Stansfield Turner, William
J. Casey, William H. Webster, Robert M. Gates, and R. James Woolseywho
kindly subjected themselves to the authors questioning.
I also want to express my appreciation to several scholars, friends, private
analysts, and reporters who have allowed me to bend their ears on the topics in
this book, often guiding me in a better direction than the one I was traveling:
Christopher Andrew, Richard K. Betts, Steven Emerson, Louis Fisher, Randall
Fort, John Lewis Gaddis, Roy Godson, Allen E. Goodman, Michael Handel,
Glenn P. Hastedt, John Hollister Hedley, Karl F. Inderfurth, Rhodri JeffreysJones, Robert Jervis, Frederick M. Kaiser, Anne Karalekas, William M. Leary,
Mark M. Lowenthal, Fred F. Manget, Ernest R. May, Harvey Nelsen, Jay Peterzell, John Prados, Harry Howe Ransom (esteemed mentor), Jeffrey T. Richelson, Harry Sepp, Frank John Smist, Jr., Robert David Steele, Stafford T.
Thomas, Richard R. Valcourt, Wesley K. Wark, H. Bradford Westerfield (who
generously and with great insight read an early draft of the manuscript), and
David Wise. No doubt they will object to some of the conclusions I have
reached in these pages; but perhaps they will see their good influence here and
there, too. The annotations throughout this volume are further testimony of my
debt to the individuals mentioned here, along with a much wider group of intelligence specialists.
I would like to express my deep gratitude, as well, for the support I have rePreface

xiii

ceived from the University of Georgia. My interview trips to Washington, D.C.,

were made possible by funding from Thomas P. Lauth, the head of the Department of Political Science; Wyatt W. Anderson, dean of the College of Arts and
Sciences; and Robert L. Anderson, the associate vice president for research. I
am grateful as well to Rick Dunn and Amy Fletcher, doctoral candidates at the
university, for their research assistance; to Chuck Grench, Otto Bohlmann, Susan Laity, and Richard Miller of Yale University Press for their guidance and
encouragement; and to the following journals and publishers for permitting me
to draw on materials I have previously published: Frank Cass, Simon & Schuster, the University of Oklahoma Press, St. Martins Press, the American
Intelligence Journal, the American Journal of International Law, Foreign
Policy, the Journal of Strategic Studies, and the International Journal of Intelligence and Counterintelligence.
Above all, I want to thank my wife, Leena, and my daughter, Kristin, for the
cheerful tolerance they have displayed toward the research trips that took me
away from the hearth and the long hours spent huddled before the pale screen
of a word processor at home. Their unwavering love and devotion have sustained me through the solitude and frustrations that accompany the writing of
a book.

xiv

Preface

ABBREVIATIONS AND ACRONYMS

ABM
ACIS
AFIO
AG
AWACS
BMD
BNL
CA
CAS
CE
CHAOS
CI
CIA
CINC
CIO
CISPES
C/CATF
CMS
CNN

anti-ballistic missile
Arms Control Intelligence Staff
Association of Former Intelligence Officers
Attorney General
Airborne Warning and Control System
ballistic missile defense
Banca Nazionale del Lavoro
covert action
Covert Action Staff (CIA)
counterespionage
codename for CIA domestic spying operation
counterintelligence
Central Intelligence Agency
commander in chief
Central Imagery Office
an FBI counterintelligence program
chief/Central American Task Force
Community Management Staff
Cable News Network
xv

COINTELPRO
COMINT
COMIREX
CORONA
COS
CPSU
CRS
DA
DAS
DCIA
DCI
DDA
DDCIA
DDI
DDO
DDS&T
DEA
DECA
DEIB
DGSE
DI
DIA
DID
DINSUM
DO
DOD
DS&T
ELINT
FBI
FBIS
FOIA
GAO
GATT
GEO
GNP
GRU
HEO
HPSCI
HUMINT
IA
IC
ICBM
IG
IIM
xvi

Counterintelligence Program (FBI)

communications intelligence
Committee on Imagery Requirements and Exploitation
codename for first U.S. spy satellite system
chief of station (CIA)
Communist Party of the Soviet Union
Congressional Research Service (Library of Congress)
Directorate of Administration (CIA)
deputy assistant secretary
Director of the Central Intelligence Agency
Director of Central Intelligence
Deputy Director for Administration (CIA)
Deputy Director of the Central Intelligence Agency
Deputy Director for Intelligence (CIA)
Deputy Director for Operations (CIA)
Deputy Director for Science and Technology (CIA)
Drug Enforcement Administration
Developing Espionage and CI Awareness (FBI)
Daily Economic Intelligence Brief
French Intelligence Service
Directorate of Intelligence (CIA)
Defense Intelligence Agency
Defense Intelligence Daily
Defense Intelligence Summary
Directorate of Operations (CIA)
Department of Defense
Directorate of Science and Technology (CIA)
electronic intelligence
Federal Bureau of Investigation
Foreign Broadcast Information Service (CIA)
Freedom of Information Act
General Accounting Office (Congress)
General Agreement on Tariffs and Trade
geosynchronous orbit
Gross National Product
Soviet military intelligence
high-elliptical orbit
House Permanent Select Committee on Intelligence
human intelligence (espionage)
Intelligence Assessment
intelligence community
intercontinental ballistic missile
Inspector General
Interagency Intelligence Memorandum
Abbreviations and Acronyms

IMF
INR
INTELINK
IOB
IRS
ISC
JCS
ITT
JCS
JETRO
KGB
KH
LEO
MASINT
MIRV
MITI
MRBM
MRC
MVD
NAFTA
NATO
NEC
NEO
NFIP
NIA
NIC
NID
NIE
NIO
NISC
NOC
NPC
NPIC
NRO
NSA
NSC
NSCID
NTM
OEOB
OMB
OOTW
OPA
OPC
OPEC

International Monetary Fund

Bureau of Intelligence and Research (State Dept.)
an intelligence community computer information system
Intelligence Oversight Board (White House)
Internal Revenue Service
Intelligence and Security Committee (Britain)
Joint Chiefs of Staff
International Telephone and Telegraph Corporation
Joint Chiefs of Staff
Japans external trading organization
Soviet secret police and foreign intelligence service
Keyhole (satellite)
low-elliptical orbit
measurement and signature intelligence
multiple independent reentry vehicle
a Japanese economic planning group
medium-range ballistic missile
major regional conflict
Soviet Ministry of Internal Affairs
North American Free Trade Agreement
North Atlantic Treaty Organization
National Economic Council (White House)
noncombatant evacuation operation
National Foreign Intelligence Program
National Imagery Agency (proposed)
National Intelligence Council
National Intelligence Daily
National Intelligence Estimate
National Intelligence Officer
National Intelligence Study Center
non-official cover
nonproliferation center
National Photographic Interpretation Center
National Reconnaissance Office
National Security Agency
National Security Council (White House)
National Security Council Intelligence Directive
National Technical Means
Old Executive Office Building
Office of Management and Budget
operations other than war
Office of Public Affairs (CIA)
Office of Policy Coordination (CIA)
Organization of Petroleum Exporting Countries
Abbreviations and Acronyms

xvii

op sec
ORD
ORR
OSINT
OSS
OSWR
OTA
OTR
PAC
PDB
PDD
PFIAB
PHOINT
PLA
PLO
PM
PRC
RADINT
ROSE
SA
SAC
SAM
SHAMROCK
SIGINT
SIRC
SIS
SLBM
SMO
SNIE
SOG
SOVA
SSCI
SVRR
TECHINT
TELINT
UAV
UN
USC
USG
USIB
USTR
VC
WMD

xviii

operational security
Office of Research and Development (CIA)
Office of Research and Reports (SOVA predecessor)
open-source intelligence
Office of Strategic Services
Office of Special Weapons Research
Office of Technology Assessment (Congress)
Office of Training (CIA)
Political Action Committee
Presidents Daily Brief
Presidential Decision Directive
Presidents Foreign Intelligence Advisory Board
photographic intelligence (imagery)
Peoples Liberation Army (China)
Palestine Liberation Organization
paramilitary
Peoples Republic of China
radar intelligence
Rich Open Source EnvironmentCIA computer software
Special Activities Division, DO/CIA
Strategic Air Command
surface-to-air missile
codename for NSA domestic wiretap
signals intelligence (special intelligence)
Security Intelligence Review Committee (Canada)
Strategic Intelligence Service (Britain)
submarine-launched ballistic missile
support to military operations
Special National Intelligence Estimate
Special Operations Group (paramilitary covert action)
Office of Soviet Analysis (CIA)
Senate Select Committee on Intelligence
Russian intelligence service
technical intelligence
telemetry intelligence
unmanned aerial vehicle
United Nations
United States Code (statutory identification system)
United States Government
United States Intelligence Board
United States Trade Representative
Viet Cong (pro-Communist faction in Vietnam War)
weapons of mass destruction

Abbreviations and Acronyms

SECRET AGENCIES

CHAPTER 1

THE MEANINGS AND METHODS OF INTELLIGENCE

In a full-page magazine advertisement that offered financial counseling for

the perplexed consumer, a New York bank presented readers with a drawing of
a man in a rowboat. Blithely oaring his way along a sparkling river, he seemed
completely unaware of the gathering currents about to sweep him over a waterfall. The copy advised, Moving ahead without looking ahead could prove
to be the greatest risk of all.
As with boating in unfamiliar waters, steering a nation through the treacherous tides of history can also be a perilous enterprise. Responsible leaders in
every nation seek knowledgeand, ideally, foreknowledgeof the world
around them. For with a better understanding of global affairs, they are apt to
protect and advance more effectively the vital interests of their citizens.
THE FOUR MEANINGS OF INTELLIGENCE

A prudent awareness of the dangers and opportunities that confront a nation

can be achieved only through painstaking collection of information about key
events, circumstances, and personalities worldwide. This gathering of infor1

mation, followed by its careful sifting, lies at the heart of intelligence as that
term is applied to affairs of state.
More formally, professional intelligence officers define strategic intelligence as the knowledge and foreknowledge of the world around usthe prelude to Presidential decision and action.1 At this global level the objective is
to acquire an understanding of the potential risks and gains confronting the
nation from all compass points. At the more restricted level of tactical intelligence the focus turns to an assessment of likely outcomes in specific battlefields or theaters of warwhat military commanders refer to as situation
awareness.
From this point of view (and it is by far the most common usage) intelligence
is information, a tangible product collected and interpreted in order to achieve
a sharper image of political and military conditions worldwide. A typical intelligence question at the strategic level would be, If a coup toppled the Russian president, who would be among the field of leading contenders to replace
him, and what political and military views do they have? Or at the tactical
level, one can imagine General H. Norman Schwarzkopf demanding during the
Persian Gulf War in 1991, I want the precise location of Iraqs Republican
Guardand I want it now! To prevail in battle, a nation must have data on the
enemys terrain, roads, airfields, ports, waterways, and bridges. Can that
bridge support a tank? Is the runway long enough for a C-47? Is the beach
firm enough to support an amphibious landing? Is aviation fuel available on
the island? Even the types of local parasites cannot be overlooked if troops are
to be properly inoculated against infectious diseases.
What makes intelligence different from other forms of information are the
strands of secret material woven into it. As Abram N. Shulsky emphasizes, intelligence often entails information some other party is trying to deny:2 agent
dossiers locked in Kremlin safes; telephone conversations between Beijing
commanders and artillery units of the Peoples Liberation Army (PLA) on maneuvers near Changchun; the flight plans of cocaine-filled Caravelle jets from
Colombia headed for landing strips in Mexico along the Texas border.
Still, much of the information gathered and analyzed by American intelligence agencies is drawn from open sources in the public domain, such as Iranian television broadcasts, Japanese economic reports, or editorials in Rossiiskaya Gazeta and the hundreds of other new Russian newspapers. Allen
Dulles, the chief of intelligence from 1953 to 1961, testified before the Senate
Armed Services Committee on April 25, 1947, that about 80 percent of intelligence analysis is based on the public recordalthough CIA old-timers hasten

Meanings and Methods

to add that he was including in this figure information gathered by diplomats

and military attaches.
Whatever the precise mix of covert and overt information in intelligence reporting during the Cold War, both are necessary ingredients for good analysis.
The overt information provides a context for the coverta way of putting the
clandestine nuggets into perspective. Yet classified studies (some by reputable outside scholars on contract) that have looked at the added value of
clandestine reporting conclude that policymakers really do gain information
from the secret agencies beyond what can be found in the New York Times, the
Economist, or Foreign Policy.3
Nonetheless, many policymakers prefer the public literature, because it is
written in a felicitous style and, since it is unclassified, can be talked about
openly. Few, though, are prepared to relinquish their access to the Presidents
Daily Brief or PDB (if they are lucky enough to be one of the thirteen policy
elites to receive it), the National Intelligence Daily (NID), the Defense Intelligence Digest (DID), or the many other publications prepared by the intelligence agencies.
Policymakers understand that intelligence sources offer unique access to
data on terrorist activities or enemy weapons systems, for instance, via worldwide coverage by agents in almost every capital and via surveillance satellites.
Most important, decisionmakers know they can talk back to these newspapers, asking intelligence officers to follow up with tailored oral briefings or
written reports. In a word, intelligence is responsive to their needs.
During the Cold War much of the information sought by policymakers was
secret (denied) and had to be acquired through clandestine means. Espionage
thus became a defining feature of intelligence-as-information. Even if the bulk
of what was reported by intelligence officers came from open sources, it
reached far beyond the policymakers usual brief sampling of the daily Washington newspapers and the New York Times.
Since the end of the Cold War the intelligence agencies have tended to concentrate on the secret pieces of the global puzzle. Sensitive to the charge (however wrong) that it adds little to what the newspapers report, the intelligence
community has made a concerted effort to demonstrate the value added from
its clandestine tradecraft. The overt/covert mix also depends on the subject.
With respect to terrorism, counternarcotics, and proliferationor hard targets like North Korea or Iranthe overwhelming percentage (75 to 90) of all
the material in intelligence reports is likely to come from clandestine sources.
In contrast, political and economic subjects are often well reported in the pub-

Meanings and Methods

lic media, and the secret agencies turn to these sources too for a reliable context in which to place their covert findings (anywhere from 10 to 40 percent of
the total).
One intelligence analyst has observed that roughly 60 percent of the sources
used by his technical branch of the CIA are open, including scientific journals,
computer databases, newspaper articles, and reports from the CIAs Foreign
Broadcast Information Service (FBIS), which translates thousands of foreign
periodicals and newspapers into English. Another 25 percent is based on insider information, that is, hard-to-find gray literature (such as technical-conference proceedings), diplomatic reporting, contract studies, and surveys financed by the intelligence agencies. Only 15 percent of its information comes
from mechanical and human espionagethough, it should be kept in mind, this
information often proves the most valuable.4
From another vantage point intelligence may be considered a process: a series of interactive steps formally referred to as the intelligence cycle.5 At the
beginning of the cycle officials plan what information to target around the
world; then they order the information to be collected and organizedor
processed in the narrower sense of that wordfor close study (analysis) by
experts.
Once the expert analysts have assessed the information, it is disseminated
in the last step of the cycle to top policy officers in the executive branch and selected members of Congress with foreign policy responsibilities. An illustration of this usage of the word intelligence might be, Analysts in the Directorate
of Intelligence (DI), the CIAs analytic shop, play a vital intelligence role as
they attempt to interpret the goals and modus operandi of Islamic radicals.
From a third perspective intelligence may be thought of as a set of missions
carried out by the secret agencies. The first is collection and analysis, a shorthand phrase for the full intelligence cycle;6 second, counterintelligence, the
thwarting of secret activities directed against the United States by foreign entities (usually hostile intelligence services);7 and third, covert action, the secret
intervention into the affairs of other states8sometimes called special activities or, for the benefit of the occasional Latin scholar who might come across
the Special Activities Division (SA) crest at CIA Headquarters, Actiones
Praecipuae. An example of this usage might be, What mix of secret intelligence operationscollection-and-analysis, counterintelligence, and covert
actionmight be most effective to prevent North Korea from developing an
arsenal of nuclear weapons?
Finally, the term intelligence is used from time to time to denote the structures or organizations that carry out these core missions. Intelligence in this in4

Meanings and Methods

stance, refers to the actual network of officials and agencies involved in the
gathering, processing, interpreting, and disseminating of information, as well
as those who plan and implement counterintelligence (CI) and covert action
(CA). Using this sense of the word the president might remark, Make sure intelligence is present at the Tuesday meeting of the National Security Council.
Or a battalion commander might say, Get intelligence on the line; I need the
exact coordinates of Serbian artillery near Bihac.
The establishment of intelligence as an organization in the United States has
a long history, beginning with George Washingtonone of the few presidents
with a deep and abiding interest in the subject.9 As general during the Revolutionary War he had his own secret code number (711) and made use of an effective network of spies led by Paul Revere and including Nathan Hale.
Intelligence organizations have played a role in each of Americas military
conflicts since the Revolutionary War.10 General Ethan Allen Hitchcock
formed a highly successful spy ring in the U.S. Army during the 1840s that
helped lead to victory in the war with Mexico. Allan Pinkerton assembled a talented team of spies for the Union Army during the Civil War, and Rose ONeil
Greenhow (Rebel Rose), a resourceful agent for the South, contributed to the
Confederate success at the first Battle of Bull Run. The outbreak of war in Europe in 1914 stirred some modest efforts in Washington to create a more sophisticated secret service for the nation, but only with the onset of World War
Two did this objective receive the full attention of President Franklin D. Roosevelt. In June 1942 he ordered the formation of a new intelligence agency,
called the Office of Strategic Services (OSS), which vigorously pursued each
of the intelligence missions against the Axis powers.11
Still, as the former secretary of state Dean Rusk remembers, the U.S. intelligence services during World War Two remained bare-boned. When I was assigned to G-2 [Army Intelligence] in 1941, well over a year after the war had
started in Europe, he once told a Senate subcommittee, I was asked to take
charge of a new section that had been organized to cover everything from
Afghanistan right through southern Asia, southeast Asia, Australia, and the Pacific. . . . Because we had no intelligence organization that had been giving attention to that area up to that time, the materials available to me when I reported
for duty consisted of a tourist handbook on India and Ceylon, a 1924 military
attaches report from London on the Indian Army, and a drawer full of clippings
from the New York Times that had been gathered since World War One. That
was literally the resources of G-2 on that vast part of the world a year after the
war in Europe had started.12
At the end of the war President Harry S Truman turned toward the task of
Meanings and Methods

modernizing the governments intelligence organization. The attack by the

Japanese air force at Pearl Harbor on December 7, 1941, had caught the U.S.
Navy by surprise and caused extensive destruction to the Pacific fleet. This day
of infamy, in President Roosevelts phrase, is still considered the most disastrous intelligence failure in American history.
Until the attack the U.S. military was unaware that the Japanese possessed
a new type of aerial torpedo that could navigate the relatively shallow waters
of Pearl Harbor. Nor did government officials have reliable information about
the likely targets of a Japanese air attack; conventional wisdom at the time
pointed to the Philippines as the probable site. Moreover, the fragments of information obtained by U.S. military intelligence that did point to Hawaii were
never adequately analyzed and coordinated within the government; the president and other high officials were never given access, for example, to decoded
intercepts of Japanese military communications that indicated that Pearl Harbor could be in jeopardy.13
With the establishment of the CIA by way of the National Security Act of
1947, President Truman hoped to improve the capabilities of the United States
to anticipate security dangers. His objective was to upgrade the collection,
analysis, andespeciallythe interagency coordination and dissemination of
information useful to policymakers as they dealt with world affairs. Above all,
the goal was to have no more Pearl Harbors. At the time Truman gave little
thought to counterintelligence or covert action; indeed, mention of these missions was omitted altogether from the National Security Act, although they
would soon take on a life of their own as U.S.-Soviet hostilities deepened.
The Cold War sired and nourished strapping espionage bureaucracies in
both the United States and the Soviet Union. Today, Americas spy empire
the intelligence communityconsists of thirteen major and several minor secret agencies. According to various newspaper accounts, the IC employs over
150,000 people and, in recent times, has spent some $2830 billion a year.14
Beneath the president and the National Security Council (NSC) in the intelligence chain-of-command stands the director of Central Intelligence or
DCI. This chief intelligence officer is in chargetitularly at leastof the entire secret government. (Appendix A provides a list of the seventeen men who
have served in this position since 1947.) The DCI simultaneously heads the
Agency, as the CIA is called by insiders, and in this capacity is referred to as
the DCIA (director of the CIA).15
The CIA is the best known of the secret agencies. Its headquarters are in the
Washington, D.C., suburb of Langley, Virginia, in a campus-like setting along
the banks of the Potomac Riverknown sarcastically by some intelligence of6

Meanings and Methods

ficers outside the CIA as Langley Farms.16 The DCI has his main office on
its seventh floor, but he also occupies a suite on the third floor of the Old Executive Office Building, or OEOB, next to the White House. The CIA is mainly
responsible for the analysis of strategic information and has also been granted
control over the planning and conduct of covert action. (Counterintelligence is
a responsibility shared by all the intelligence agencies, in coordination with a
newand still inchoateNational Counterintelligence Center, established in
1994 in the wake of the Aldrich Ames spy scandal.) The CIAs organizational
chart (as of 1995) is presented in figure 1.1.
The CIAs major companion agencies include the National Security Agency
(NSA), located at Fort Meade, Maryland, responsible for codebreaking and
electronic eavesdropping; the National Reconnaissance Office (NRO), with
quarters in newly constructed buildings near Dulles Airport in the Virginia
countryside and chartered to coordinate the development and management of
surveillance satellites; the Central Imagery Office (CIO), in the Department of
Defense (DOD), which supervises the photographic side of foreign surveillance; the Defense Intelligence Agency (DIA), also in the DOD and in charge
of military intelligence analysis; and the four military intelligence services,
each gathering tactical intelligence from all corners of the globe. Each of these
entities is under the command of the secretary of defense (as well as the DCI
a sure prescription for blurred lines of authority) and as a result are considered
the nations military intelligence agencies.17
On the civilian side of intelligence stand (along with the CIA): the Bureau
of Intelligence and Research (INR), at the Department of State; the FBIs intelligence units, housed within the Department of Justice; the Department of
the Treasury, home of the Secret Service and the Internal Revenue Service, both
of which have an intelligence component; and the Department of Energys intelligence corps, which (among other duties) tracks the flow of fissionable materials around the globe.18 Together, these military and civilian agencies comprise the largest organization for the production of information in the history
of civilization (see figure 1.2).19

AN ENCOMPASSING VIEW OF INTELLIGENCE

Regardless of how the term is usedas product, process, mission, or organizationintelligence is widely considered Americas first line of defense.20
The assumption behind this perspective is that sound choices for U.S. foreign
policy depend on decisionmakers having the most accurate, complete, and
timely information possible about the capabilities and intentions of other naMeanings and Methods

Figure 1.1 The Office of the DCI and the Central Intelligence Agency

Figure 1.2 The United States Intelligence Community

tions or factions. This is not an easy assignment on a vast planet where nations
keep their political ambitions closely veiled and hide their development of new
weapons inside heavily guarded buildings and even, as in North Korea, in deep
underground caverns.
At bottom the intelligence community, with its intricate worldwide network
of mechanical and human spies, has but one overmastering objective: to safeguard the United States and its international interests. This can mean anything
from promoting democracy to ensuring access to foreign oil and preventing internal subversionan important mission of the domestically oriented intelligence agencies, like the FBI. To achieve these goals, it is first necessary to acquire and understand information about the potential threats and opportunities;
consequently, reliable facts and analysis are seen by many scholars and government practitioners as the sine qua non of effective decisionmaking. Every
morning I start my day with an intelligence report, President Clinton has remarked. The intelligence I receive informs just about every foreign policy decision we make.21
A former secretary of state has suggested why decisionmakers often display
a healthy appetite for information of all kinds, including intelligence: The
ghost that haunts the policy officer or haunts the man who makes the final decision is the question as to whether, in fact, he has in his mind all of the important elements that ought to bear upon his decision, or whether there is a missing piece that he is not aware of that could have a decisive effect if it became
known.22
The situation in the Persian Gulf in August 1990 provides an illustration of
how vital intelligence can be to policy officers. No question pressed more heavily on those in the White House and the Pentagon during that month than the
exact size and strength of the Iraqi military units that were headed south to invade Kuwait. An effective American response would have to rely substantially
on accurate intelligence about the troop and weapons strength of the Iraqi
forces. Drawing on a combination of intelligence sources (including the orderof-battle expertise of an Iraqi military defector), the CIA and the DIA quickly
provided answers.
In this instance of competitive analysis the two agencies disagreed dramatically on the potency of the Iraqi military. It took another two months of
data scrubbing before it became clear that the DIA figures had been based on
outdated information from the Iran-Iraq war and consequently were inflated.
Yet even in those frustrating instances where the secret agencies disagree, the
debate that ensues gives a more reliable result than if leaders were able to turn
to only one agency for an answer. Out of this particular interagency disagreeMeanings and Methods

11

ment came a useful cross-checking of sources and methodologies which eventually produced a highly reliable order-of-battle assessment. Regrettably, a
president will not always have the luxury of waiting so long before dispatching troops into battle. Nor will the United States always possess the resources
even in more robust economic timesto provide intelligence support for every
possible military contingency the country may face overseas. Even the idea
(endorsed by the bottom-up review conducted by Secretary of Defense Les Aspin in 1994) of fighting simultaneously two so-called major regional conflicts
(MRCs)say, in North Korea and Iraqwould stretch American intelligence
support and warfighting capabilities to the limit.
Defense Secretary William J. Perry, Aspins successor, questioned the feasibility of the 2MRC concept in public hearings. Its an entirely implausible
scenario that wed fight two wars at once, he conceded before the Senate Defense Appropriations Subcommittee in 1994.23 Yet demands for intelligence
support for military operations (referred to as SMO in the Pentagon) extend
even beyond the prospect of two major wars. Intelligence support is needed
for small-scale interventions (like Haiti and Somalia) as well as Operations
Other Than War (OOTW in Pentagonese), which include dispensing military
and humanitarian aid, staging counternarcotics operations, noncombatant
evacuations (NEOs), and United Nations (UN) peacekeeping operations, as
well as counterterrorism operations, interdicting weapons of mass destruction
(WMDs), and assisting foreign forces.
Each of these activities stands to benefit from good intelligence support; and
this is only the Pentagons list. The civilians in the government who deal with
foreign affairs have their own intelligence requirements, too, from information
on trade matters to support at international environmental conferencesall
tugging at the same finite resources. The tension between uniforms and suits
tactical intelligence for the military field commanders and strategic intelligence for the president and the rest of the civilian part of the governmentlies
at the heart of the current debate over future directions for American intelligence.
The president, as the commander in chief and the highest civilian officer in
the government, is caught in this cross fire between contending intelligence requirements. Added to the complexity of the rival claims on the intelligence dollar is the fact that most of the time, happily, the United States is at peace. Yet
when war comes, the nation must be ready. In the first instance, the president
can tilt toward the civilian side of intelligence, using the assets of the intelligence community to gather and analyze information that may head off a war.
In the second instance, however, he must tilt toward success on the battlefield,
12

Meanings and Methods

with the fewest American casualties possible. (Zero-body-bag wars is the

quixotic goal of some military planners in whose heads dance visions of remote-control, penny-arcade weapons.) These are quite different postures (despite some overlap); as a result, sorting out the nations future intelligence
needs is hampered by turf battles within the bureaucracy.
Given these multiple dimensions of intelligence, how shall it be defined? If
one prefers a narrow dictionary definition, the idea of intelligence as product
secret informationis apt to be most satisfying. As we have seen, however,
this perspective leaves aside a good many activities carried out by the secret
agencies. For that reason, in this book I prefer a broader perspective. Regardless of ones favorite definition, the most important point is to have an understanding of what duties the secret agencies actually perform. In this spirit, one
can say that intelligence has to do with a cluster of government agencies that
conduct secret activities, including counterintelligence, covert action, and,
foremost, the collection and analysis of information (from a mixture of open
and covert sources) for the illumination of foreign policy deliberations.

THE METHODS OF INTELLIGENCE

Human beings have always needed information to secure their livelihood

and their safetythe location of the best fishing stream, the site where firewood
might be gathered, when deer herds were likely to appear. During the Cold War
the presence of nuclear warheads and rapid-delivery systems held out for
Americansand perhaps for all humankindthe prospect of sudden extinction. This ominous condition made accurate information about the intentions
and capabilities of the well-armed adversary, the USSR, more vital than ever.
In this current information age we are constantly bombarded by facts,
opinions, speculation, rumor, and gossip from every direction. Television carries into our homes each night unsettling images of squalor and death from
around the world (not to mention our own backyard). Computers draw us into
an interactive milieu where e-mail gives, and expects in return, ever more rapid
exchanges of information. The cellular telephone assures that a flow of information will follow us everywhere: into the car, the mall, the meetingplace.
What effect has this rising tide of informationand its secret undercurrents we
call intelligencehad on decisions made in the high councils of government?
Foreign policy decisions are preceded in most cases by the gathering and interpretation of information by government officials about the costs and benefits that may accrue to their nation from various options. In prehistoric times,
people were touched by only small eddies of data about the world around them:
Meanings and Methods

13

hints of changing weather in the cloud formations, the scent of game, the sound
of a twig snapping at night that warned of an intruder. In our own time, American leaders stand in the middle of a deep and rushing stream of information
from across the globefrom newspapers, computers, radio, telephone, and especially television. As Ronald Steel observed in 1995, We would probably not
be involved in any of these areas [Somalia, Rwanda, the Balkans] were it not
for the power of television to bring the most horrifying images into the American living room.24
The form of some information that comes to the president and other top officials has changed little from the early days of the republic: whispers from the
First Lady, ruminations over drinks in the Georgetown parlors, the counsel of
confidants offered in the privacy of the Oval Office. Yet consider these dramatic
changes: thousands of high-resolution satellite photographs arrive each day in
the offices of intelligence analysts; data in the form of signals intelligence (SIGINT) pour into the receiving antennae at the NSA; live, ghastly pictures of the
carnage in Rwanda and Bosnia fill the television screens in the White House
and most every other house; a deluge of citizen opinion jams Internet terminals
throughout the government, including the warrens of the Old Executive Office
Building, where NSC staffers prepare their influential option papers on foreign
affairs. The advance of technology has produced a downpour of information
that falls relentlessly on intelligence officers and policymakers alike.

Information Collection
Sophisticated spy machines, designed for the purposes of broader and faster
information collection, have exercised a fascination on those in public office.
Over the years since 1947 the managers of the secret agencies have successfully promoted a steadily rising investment for technical intelligence, or
TECHINT.
By definition TECHINT refers chiefly to IMINT and SIGINT. IMINT is the
acronym for imagery intelligence, also called photographic intelligence
(PHOTINT), electro-optical intelligence, or, in plain English, photography.
SIGINT, also known as special intelligence, encompasses the interception
and analysis of communications intelligence (COMINT)say, two drug dealers talking to one another via cellular telephones in Colombiaand electronic
intelligence (ELINT), such as the electronic signals associated with radar
jamming.25
Foreign radios, satellites, cellular telephones, and land-line and fiber-optic
communications all are inviting targets for SIGINT collectors hoping to learn
the intentions of adversaries. Electronic eavesdropping can be the key to avert14

Meanings and Methods

ing war. For example, it could tip off the attack plans of a belligerent nation that
might be countered by stepped-up diplomacy or a show of military strength. It
may also save the lives of individual Americans abroad. Recently a U.S. ambassador was forced to plan an evacuation because of a civil war that was
spreading through the country in which he was stationed. A SIGINT intercept
disclosed that a team of assassins had learned of the proposed evacuation route
and intended to slay the ambassador, his wife, and children. Warned of the trap,
the ambassador and his family took a different route to the airport and escaped.
Another of the technical ints is MASINT, which stands for measurement
and signature intelligence. MASINT exploits the physical properties of foreign
targets (an enemy missile, for example) through the use of special technical
sensors. These properties might include energy emitted from a nuclear warhead, mechanical noises, or telemetry intelligence (TELINT), the collection of
data emitted by weapons as they are being flight-tested, which reveals their
specifications.
Prior to the advent of the U-2 spy airplane in the 1950s, the most important
TECHINT efforts against the Soviet Union came from radar sites in Turkey and
Iran (collecting RADINT, or radar intelligence, a form of MASINT); from EC135 and RC-135 aircraft lumbering along the perimeter of the USSR; and from
camera-laden, unmanned balloons drifting across Soviet airspace. Some of the
balloons made it to Japan and the Pacific, but most crashed somewhere in the
vast Soviet territory.
The U-2 is an imagery collector and the most outstanding of the early
TECHINT innovations.26 Developed in an accelerated program to obtain reliable data on the extent of the feared bomber gap, this sleek spyplanethe
so-called Black Lady of Espionagemade its debut with a flight over the
Soviet Union on July 4, 1956. A series of twenty-nine additional U-2 flights
deep into the USSR during the late 1950s and early 1960 (brought to a halt for
six months beginning on May 1, 1960, when the Soviets shot down over
Sverdlovsk a U-2 piloted by Gary Francis Powers) provided IMINT impressive enough to persuade American leaders that the Soviets had far fewer longrange bombers than initially feared. The Bison and Bear aircraft simply were
nowhere to be found in the anticipated numbers on Soviet airfields.
Evidence regarding the next alarma missile gap, stemming from concern over a possible acceleration of the Soviet ICBM programremained inconclusive.27 Following the U-2 shootdown in 1960, President Dwight D.
Eisenhower had promised his Kremlin counterpart, Nikita Khrushchev, that he
would curb further U-2 flights over Soviet territory, so the answer to the missile debate would require a different approach: satellite photography from the
Meanings and Methods

15

more secure confines of space. With a new sense of urgency, the government
rushed forward with its nascent satellite program.
After a frustrating concatenation of technical disasters, in 1960 the United
States at last placed a reliable surveillance satellite in space (Project
CORONA). The first CORONA image, taken on August 18, 1960, was disappointingly fuzzy but clear enough nonetheless to discern a Soviet airfield at
Mys Shmidta. Unfortunately, most of the satellite photos taken in 1960 were
dark and difficult to read; during 1961, however, the spy cameras improved
greatly, and their pictures of military installations in the USSR did indeed disclose the existence of a missile gapbut one that favored the United States.
By the 1970s America had launched several types of satellites into the heavenssome as big as a Mack truck. A few relied on electro-optical technology,
others on infrared sensors and radar. Some circled the planet in a low elliptical orbit (LEO), others in a high elliptical orbit (HEO), and a few remained in
a stationary posture over a single nation or region (achieved by orbiting in
synchrony with the earths own spin velocity, called geosynchronous orbit
or GEO). The perigees and apogees ranged from less than one hundred to
more than twenty-four thousand miles in space. Together, the constellation
of satellites (platforms) offered an exciting new TECHINT blend of collection cameras and sensors that allowed several perspectives of the same
target.
Harold Brown, the secretary of defense during the Carter presidency, has
commented on the value of this intelligence synergism:
Our national technical means [NTM, the accepted euphemism at the time for satellites and other TECHINT machines] enable us to assemble a detailed picture of Soviet forces, including the characteristics of individual systems, by using information from a variety of sources. . . . We regularly monitor key areas of the Soviet
ICBM test ranges. We monitor missile test firings with a wide variety of sensors:
cameras taking pictures of launch impact areas; infrared detectors measuring heat
from the engine; radars tracking ICBMs in flight; and radios receiving Soviet
telemetry signals. . . . The use of multiple sources complicates any effort to disguise or conceal a violation.28

The technological advances were fairly steady and remarkable from 1956
to the 1980s, though always punctuated by setbacks. By 1963 the Keyhole
or KH cameras (a generic term for spaceborne image collectors, just as Talent refers to cameras aboard aircraft like the U-2) could peer from remote
space into newly dug Soviet missile silos. In the 1970s the Rhyolite generation of satellites tracked missile telemetry with ever greater accuracy and,

16

Meanings and Methods

joined by its cohorts Chalet and Jumpseat, achieved major breakthroughs

in COMINT. The infrared and radar satellites of the late 1960s and the 1970s
were especially important innovations because, unlike electro-optical photography, they are able to penetrate through cloud cover and the darkness of night
(relying on star glow alone to provide the necessary definition). The KH-11 imagery satellite launched in 1976 presented as a gift to incoming president
Jimmy Carter one of the greatest advances of all: real-time imagery of the
USSR and other foreign targets. The main points of friction now were the processing and interpretation of the images, not their delivery to earth.
Into the 1980s the TECHINT wizards in the intelligence community and
their colleagues in the private sector spun out more devices for watching American adversaries more closely. The speed with which data were moved from
satellite platform to earth-bound photointerpreters accelerated, new cameras
provided wider swaths of coverage, and engineers produced an expanded range
of camera angles for greater comprehension of such matters as a missiles dimensions. Further, the lifespan of the satellites rose from a few days to months,
then years; and the number of ground stations increased to process more rapidly
the stream of data from space. Failed launches that so plagued the early days
of the spy satellite program became a rarity.29
Spy satellites have their limits, of course. Despite their sophisticated phototechnology, they do not have x-ray vision and cannot see through roofs.
Moreover, nations like Russia and China have learned how to track their orbits.
Foreign regimes often halt their use of sensitive communications and telemetry testing and hide their weapons as the birds pass overhead. The North Koreans solved this problem by locating their most sensitive weapons facilities
underground. Yet the reconnaissance satellites have contributed in a major way
to making the world more transparent and therefore safer from the dangerous
hysteria that has frequently arisen over the possible machinations of unseen enemies.
The recruitment of human spies who can steal secrets from vaults or overhear important conversations among foreign adversaries is still a high priority
for Americas secret agencies. During the Cold War, however, spending on
TECHINT far outdistanced spending on old-fashioned espionage (known as
human intelligence or HUMINT).30 A strong proclivity exists among those
who make budget decisions for national security to focus on warheads, throw
weights, missile velocities, fuel range, and the specifications of spy satellites
things measurable.
Briefings to legislators who hold the intelligence purse strings are in-

Meanings and Methods

17

eluctably accompanied by state-of-the-art visual aids: flashy four-color slides

(grabbie graphics, a CIA specialty), videotapes, and CD-ROMs. They portray satellites outfitted with all the latest bells and whistles, and cladlike the
Great Gatsbys famous motorcarin shiny metal and glass that mirror a dozen
suns as they rotate the earth.
Unlike the traditional human spy (whose identity is a tightly held secret
no pictures allowed), the spy satellite has a tangible presence. Not only can
the DCI show it off with slides during closed-door hearings, he can also pass
around the photographs it has produced: startlingly detailed displays of the enemys missile sites and tank deployments; infrared tracings of hot radioactive material flowing through the pipelines of a weapons factory deep within
the territory of a nation whose leaders claim that the facility is merely a pharmaceutical laboratory; radar impressions, taken at night or through cloud
cover, of fighter aircraft bearing missiles on a remote runway. Satellite cameras neither lie nor defect to the enemy, while their human counterparts (recruited by trolling bars in foreign capitals) have been guilty on both counts.
Technical intelligence is, in a word, trusted by collectors, analysts, and policy
officers alike.
One result of this growing reliance on TECHINT has been the acquisition
of more and more information collected at ever faster rates. And the intelligence
agencies have worked to improve the mobility of the collection platforms and
achieve greater flexibility in reorienting their instrumentation toward fresh targets at a moments notice. The aspiration is to create a surge capacity that will
allow the quick shifting of platforms toward whatever newly threatening targets may suddenly ariseSomalia today, Suriname tomorrow.
Once information is captured by an intelligence platform, the ability to send
the data hurtling back to Washington for processing has also been tremendously
accelerated. Film from the early CORONA satellites had to be catapulted from
space back toward earth, then plucked out of the ether by ponderous C-119 and
C-130 aircraftwhich sometimes failed to snare the precious eighty-fourpound capsules as they descended by parachute toward the Pacific Ocean.31
The data were flown home while fidgeting photointerpreters awaited the next
batch of black-and-white images. Now, as a result of modern digital communications, the trip from satellite to Stateside takes only moments.
Recent technological advances have improved overt information collection
too. Intelligence officers are turning increasingly toward new computer-based
information search tools (like Lexis-Nexis) and the daily reporting of information from around the world by private companies (like Oxford Analytica),

18

Meanings and Methods

along with the burgeoning use of the Internet, facsimile machines, and
e-mail. Academe, business, the media, and government are busy harnessing
these powerful tools of information management.32 At the CIA, an impressive
system called ROSE (Rich Open Source Environment) allows agency analysts
to tap into more than two thousand full-text on-line journals, from the African
Economic Digest to the Yale Law Review.
Recently a program called INTELINK, based on Internet technology, has
been introduced as a means of spinning the governments secret agencies into
at least a limited web of classified-information exchanges, to be supplemented
eventually with access to the ROSE materials. After a number of false starts,
the infrastructure for modern computer information management is growing
steadily and drawing the analytic side of the secret agencies closer together than
ever before. The CIA now has secure e-mail facilities to maintain contact with
its stations around the world; and fax intelligence, sent over secure lines, has
become a favorite means by which intelligence officers communicate with policymakers.
In spite of efforts by the intelligence agencies to keep up with technological
advances in communications, close observers suggest that in some respects
they have fallen behind the private business sectorand even some college
dormitoriesin desktop information management. Inside the State Department, for instance, the INRs e-mail system is self-contained (for security purposes). This prevents intelligence officers from sending classified e-mail to the
diplomats they are supposed to supportnot to mention adding to INRs sense
of isolation in the building. Policy officers in the OEOB, an antiquated (if
charming) structure, are similarly without secure e-mail connections to the intelligence agencies; NSC staffers must hike over to the Situation Room in the
basement of the White House to read classified cable traffic. Impressive recent
progress aside, the ICs communications infrastructure still has a long way to
go before analysts are connected to each other, to collectors, to open-source
data banks, and to the policy community in a sophisticated network of work
stations.
While technology has undoubtedly made the task of information collection
more efficient, human beings continue to play a vital role. The case officer engaged in HUMINT overseas must carry out the sensitive agent-recruitment operations abroad and attempt to calculate the intentions of foreign leaders.33 For
as Ephraim Kam has emphasized, an adversarys most important secrets often exist in the mind of one man alone . . . or else they are shared by only a few
top officials.34 This kind of information is accessible, if at all, only to an in-

Meanings and Methods

19

telligence officer with ties to someone inside the closed councils of the target
government.
No matter how good our technology, well always rely on human intelligence to tell us what an adversary has in mind, President Clinton has acknowledged. Well always need gifted, motivated case officers at the heart of
the clandestine service. Well always need good analysts to make a clean and
clear picture out of the fragments of what our spies and satellites put on the
table.35 In the early days of tracking the Soviet target, when TECHINT was
still in its infancy, HUMINT sourceseven though good ones were rare
sometimes proved of great value. Colonel G. A. Tokaty-Tokaev, for example,
defected to the United States in 1948 with useful information on the state of the
Soviet ICBM program; and Colonel Oleg Penkovskys espionage on behalf
of the United States and Great Britain during the 1960s was an even greater
windfall.
During the Carter administration the nation was reminded again of the importance of HUMINT when Iranian student militants took American diplomats
hostage inside the U.S. embassy in Tehran. In planning a rescue operation,
satellites could provide excellent eagle-eye pictures of Tehran but could not
see inside the embassy or find precisely where the hostages were being kept.
We had a zillion shots of the roof of the embassy and they were magnified a
hundred times, remembers one of the rescue planners. We could tell you
about the tiles; we could tell you about the grass and how many cars were parked
there. Anything you wanted to know about the external aspects of the embassy
we could tell you in infinite detail. We couldnt tell you shit about what was going on inside that building.36
The question of intelligence targeting further illustrates the cardinal role of
the human being in matters of intelligence gathering. The most important targets for the intelligence community are those nations or factions that present a
danger, or potential crisis, for the United States (so-called Tier 0 nations in current jargon). Yet while North Korea, Iraq, Iran, and other rogue states are easy
enough to place into this category, will U.S. leaders have the sagacity to anticipate what other targets should be at the top of the list in the immediatelet
alone the long termfuture?
When I became Secretary of Defense [in 1993], I served several months
without ever giving Rwanda a thought, recalled Les Aspin. Then, for several
weeks, thats all I thought about. After that, it fell abruptly off the screen again
and I never again thought about Rwanda.37 Knowing where to position the nations high-tech intelligence platforms is not a simple task, since countries have
an annoying habit of leaping suddenly from Tier 4 (the outer fringes of the tar20

Meanings and Methods

geting list) to Tier 0Grenada, Panama, Kuwait, Yugoslavia, and Somalia,

among other recent shooting stars or flavors of the month, as analysts call
them.

Information Processing
The next step in the intelligence cycle is called processing, which involves
the refinement of freshly gathered, rawinformation into a form that is more
easily studied by intelligence analysts. Coded data are decrypted, foreign
languages translated, and the focus of photographic material sharpened to provide maximum resolution of the imagery. Advances in technology have made
a major contribution here too. State-of-the-art computer methods make foreign
diplomatic codes more vulnerable to unraveling by cryptographers at the NSA
and help sort out the elaborate calculations involved in converting radar images into digital data.
Here again technology rubs up against the human dimension of intelligence.
The surveillance satellitesoften described as gold-plated vacuum cleaners
in the skyyield far more data than the government has the resources to
process. The information coming down from these [satellites] is just going to
choke you, laments the physicist Jerry Nelson. You cant buy big enough
computers to process it. You cant buy enough programmers to write the codes
or to look at the results to interpret them. At some point you just get saturated.38 Near the end of the Cold War the NSAreportedly processed only about
20 percent of the SIGINT it collected; recently another NSA official estimated
that the figure has dropped to about 1 percentalthough new techniques have
improved (though by no means perfected) the NSAs ability to focus on the
most important 1 percent.39 Little wonder that a recent NSA director, Vice Admiral J. M. (Mike) McConnell, was often heard declaiming, I have three
major problems: processing, processing, and processing.40
Another processing headache is language translation. The shortage of qualified linguists available to the secret agencies remains a serious deficiency, particularly with respect to the more exotic languages. Moreover, the technology
to machine-read and translate texts reliably and quickly from foreign languages
into English will not reach high levels of proficiency for decadesalthough it
is reasonably good now for some limited tasks where the language is precise,
such as translating Russian scientific texts.

Information Analysis
Technology has also aided the third crucial step in the intelligence cycle:
analysis. At this stage the experts assess what the unevaluated intelligence acMeanings and Methods

21

tually means for the security of the United States. The objective is to produce
fully interpreted intelligence based on a blend of covert collection products
from all the secret agencies (all-source intelligence) and open-source materials. The output of intelligence materials has been prodigious. In 1994, for example, the DI alone produced over thirty-five thousand intelligence reports of
one kind or another, from oral briefings to encyclopedic studies.41
The written form of finished intelligence may be either an intelligence report orthe crown jewel of community-wide analysisa full-blown National
Intelligence Estimate (NIE). In both cases the focus may be on a single foreign
country or a specific topic (say, Iraqi oil production).42 In contrast, the intelligence product may also consist of short, up-to-the-minute reports known as
current intelligence. These can take the form of special intelligence reports
(crisp, highly focused papers no longer than three pages), intelligence memoranda (five-to-seven pages), or, in sharply abbreviated form (in-briefs), one
to several paragraphs in the prestigious PDB or one of several other intelligence
newspapers.
According to a recent unclassified CIA document, hundreds of reports derived from SIGINT, imagery, and human sources are sent to consumers [policy officers] and other producers [fellow analysts] each day.43 Interviews with
intelligence managers conducted in 1994 indicate that a majority of the papers
written by the DI are foreign leadership analyses, chiefly personality profiles
of political and military elites.
For decisionmakers, the favorite product from among this extensive menu
is no doubt current intelligence. Research reports [like the lengthy NIEs] work
their way from the in-box to the burn bag unread, concludes an INR analyst
ruefully. Why? Because consumers dont have time to read them, the analyst
continues. The demands today are for the quick report and the quick answer
bumper sticker or time-bite intelligence.44 This same analyst reports that
at INR the number of extensive research papers has plummeted over the past
decade from 250300 to just fifteen a year.
Some policy officers prefer reports that are briefer still: the raw intelligence alone. I would ask for some of the raw data which was behind the reports, Dean Rusk once recalled, so I could make my own check.45 At the
NSC staff level a former senior aide has said, When I wanted intelligence, I
went straight to the Sit [Situation] Room and read the raw cable traffic coming
in from overseas.46
Other policymakers prefer not to read any intelligence whatsoever, raw or
evaluated; they rely instead on spoken communication. Commenting on the
widespread use of oral intelligence briefings, Allen E. Goodman of George22

Meanings and Methods

town University wryly remarks that among policymakers, some dont read,
some wont read, and some cant read.47 About one-third of the products
created by DI analysts are oral briefings48mainly presented to policy officers
in the executive branch but increasingly to members of Congress as well. Now
and then the briefings are delivered on the run down the corridors of power, as
VIPs rush to the next meeting, or in the back seats of limousines on the way to
Washington National Airport.
The oral briefing, despite its obvious shortcomings, plays a vital part in the
intelligence cycle. Estimation is more an oral than a written process, a chairman of the National Intelligence Council (NIC) has explained. It starts with
oral contacts between NIOs [National Intelligence Officers, senior analysts in
the intelligence community assigned to the NIC] and policy makers, to find out
whats on the policy makers mind. Then it can take various written forms: an
NIE, a two-page update on an earlier NIE, a short NIC memo of two or three
pages. And it ends in an oral process, with the NIO briefing the policy maker
on the key conclusions, because theyre probably not going to have read the
written report.49
Intelligence managers value the oral briefing highlyunlike many analysts,
who prefer the opportunity to work on carefully nuanced written papers that
display their expertise and allow them more room to hedge. The situation we
find the best, declares a former CIA manager, is . . . when one of our substantive officers sees the president every day for a period, however brief, to get
the intelligence [to the decisionmaker] and receive his reaction to it, including
tasking for the next day.50 This way the intelligence manager knows for certain that the product has reached the intended consumer instead of the circular
file, and he or she can learn immediately what information the policymaker
ideally, the presidentwants next.
Gerald R. Ford and, even more so, George Bush accepted this approach, for
the most part. Some presidents, though, have refused oral briefings, preferring
short written summations. Richard Nixon cut off DCI Richard Helms from the
Oval Office after the director had enjoyed good access during the Johnson presidency; Helms remembers Nixon as the ultimate loner.51 Ronald Reagan, a
former screen star, showed an enthusiasm for intelligence presented on videotape. Whether current intelligence, raw intelligence, oral briefings, or intelligence movies, the declining emphasis on in-depth research holds a danger
for the future. The intellectual resources stored by the secret agencies may simply dry up. Long-term research is putting money into the bank, says former
DCI Robert M. Gates; current analysis is taking money out of the bank.52
By all accounts the secret agencies provide some of the best forums in the
Meanings and Methods

23

government for the analysis of international events. According to one experienced government official, Intelligence analystsessentially DI analysts
do 90 per cent of the analysis of the USG [United States Government] on foreign affairs.53
Further, regardless of all the help that machines have provided in manipulating data and crafting eye-catching graphic displays, the analytic process remains vitally dependent on the experience and intellectual abilities of the men
and women preparing the written reports and delivering the oral briefings. Yet,
does the analyst have the requisite skills to make accurate forecasts? Are the
right experts available to give a full and timely response to the policymakers
request for an assessment of some foreign event? How deep-keeled is the analysts knowledge of the country, or the circumstance, he or she is attempting to
evaluate? Too few analysts have spent adequate recent time in the countries
they are expected to understand. How many intelligence officers preparing reports for the NSC have lived in Somalia or Rwanda, Haiti or Iraq?
Moreover, the analytic process is replete with disputes over which of several competitive interpretations of the facts ought to be forwarded to the next
level of the bureaucracy before going on to the White House. In the formal estimating process by which NIEs are produced, analysts have an opportunity (if
their managers see fit) to register their dissent in the form of a footnote or, during the Clinton administration, in the text itself. Technology plays a role here
too, as Lawrence Freedman shows. As a profession, intelligence analysts are
dedicated empiricists with a shared respect for certain types of hardevidence,
sufficient to force them to acknowledge it even if it contradicts strongly-held
beliefs, he writes. Such evidence is that which comes from technical collection programs, such as radar and satellites. Other evidence will have varying
degrees of softness and its reliability may be disputed. . . . The more estimators have to guess, speculate, infer, induce and conjecture in order to reach a
conclusion, the greater the possibility of open disagreement.54
Most troubling is when the DCI or another manager decides to bury the work
of an analyst because he finds his own interpretation of events more compelling, or because he hopes to curry favor with the White House by providing
intelligence to please. At times the DCI has been an ideologue who wants the
intelligence community to shape its interpretations to match his own worldview. Robert Gates has testified that as deputy DCI he watched his boss,
William J. Casey, on issue after issue sit in meetings and present intelligence
framed in terms of the policy he wanted pursued.55
For the most part, though, DCIslike the analysts below them in the intelligence hierarchyhave exercised a professionalism that wards off tempta24

Meanings and Methods

tions to distort intelligence. Know the truth and the truth shall make you free
is the CIAs motto, and it is taken seriously by virtually all of the men and
women who enter the analytic side of the profession. Thus, the recommendation of a well-regarded former DDCI is valid most of the time: You have to
have faith that the CIAs professionals are strong enough to make straight
calls.56

Information Dissemination
Technology has had a major effect as well on the last phase of the intelligence cycle: the dissemination of information to the policy officerthe consumer of intelligence. Stewart A. Baker, a former intelligence official, is not
alone in his conclusion that from Pearl Harbor on, the intelligence failures that
hurt the worst have not been those of collection but rather those of dissemination.57
To start with a positive case, Operation Desert Storm in 1991 provides a
vivid example of swift and reliable intelligence support to the consumer. American surveillance satellites sensed the Iraqi anti-aircraft radar the moment it was
activated and relayed that information rapidly to waiting fighter pilots and
cruise-missile commanders. The word soon spread in Baghdad that it was suicidal to flip the on switch inside a radar facility, as moments later the person
at the switch would be annihilated by American F-117 aircraft or self-propelled
Tomahawk cruise missiles.
The dissemination architecture for intelligence during the Persian Gulf
War was by no means flawless, however. In the field the military had fourteen
different kinds of receiving devices for incoming intelligence, only two of
which were compatible.58 This lack of battlefield connectivity no doubt contributed to the frustrations later vented by General Schwarzkopf, who was unquestionably correct in this postmortem: We just dont have an immediately
responsive [imagery] intelligence capability that will give the theater commander near-real-time information that he personally needs to make a decision.59
In the aftermath of the Gulf War, General James R. Clapper, Jr., the talented
DIA director, concentrated his attention on making improvements in the dissemination of battlefield intelligence. His objective was the prompt delivery
to all combat commanders, regardless of echelon, of the pictures, not reports
they tell us are essential to accomplishing their mission.60 High-tech planners
in the intelligence community foresee a time in the near future when all satellite and aircraft IMINT and SIGINT will be downlinked to vans in the backlines of the battlefield, where the processing and dissemination of data will be
Meanings and Methods

25

carried out close to the soldiersnot back in Washington. In General Clappers

vision, the ultimate ideal is to have a constant Gods-eye view of the battlefield. Anywhere, anytime, all the time.61 One must wonder, however, about
the practicalitynot to mention the expenseof staring down on Earth as if
one were God.
Whatever its shortcomings, the flow of information from sensor-to-soldier
during Operation Desert Storm set a new benchmark for intelligence achievement in support of the fighting men and women. Indeed, the dissemination of
information to distant battlefields has proven easier in some respects than
across the few miles that separate the intelligence agencies from the White
House and the National Security Council.

INFORMATION AND THE POINT OF DECISION

At some point a decision must be made. Until then, technology contributes

mightily to the production of the richest stream of information, laced with secrets, ever enjoyed by a nations leaders. At the moment of decision, however,
statecraft becomes paramount, and all the sophisticated technology of a modern superpower is to little avail.
As officials prepare to deliberate on foreign policy, often they are too busy
to absorb new information (let alone deep analysis); or their ideological lenses
may distort the information that does reach them. Sometimes the problem is
mutual ignorance: the intelligence officer is unsure what the decisionmaker really wants, and the decisionmaker is unaware of what the intelligence officer
has to offer. As a former government official recalls, when he was on the NSC
staff in 198990, he did not read a single [National Intelligence] Estimate. Not
one. He explains why: DI analysts did not have the foggiest notion of what I
did, and I did not have a clue as to what they could or should do.62 Only years
later, as a participant in arms control negotiations (a CIA forte), did he discover
how a close working relationship with intelligence officers could prove beneficial.
Among the hazards found at the intersection between information dissemination and decision is the trap of intelligence to pleasethe politicization
or cooking of intelligence, in which the facts are slanted to suit the political needs of the current administration. As DCI, Richard Helms reportedly
changed an estimate on Soviet military intentions at the urging of a Nixon administration official. He is said to have gone along with the Pentagons position on Soviet first-strike preparations, despite contrary views among analysts
within the CIA, because an assistant to [Secretary of Defense Melvin] Laird
26

Meanings and Methods

informed Helms that the [views of the CIAs analysts] contradicted the public
position of the Secretary.63
As a result of intimidation, good information sometimes never even makes
it to the table where decisions are made in Washington. Nothing permeates the
Cabinet Room more strongly than the smell of hierarchy, Peter Wyden remarks in his study of why DI analysts capitulated to the views of more senior
government officials during deliberations over the proposed Bay of Pigs operation in 1961.64 Policymakers in the Kennedy administration and their allies in
the CIAs Operations Directorate (some of whom enjoyed the advantage of a
Georgetown bon vivant relationship with the president) were so intent on toppling Castro that DI analysts convinced themselves that any discouraging prognosticationsand they had more than a fewwould not only have been fatuous but would also have been sharply resented and would have threatened their
careers.
According to an expert on organizational behavior, this tendency to get
along with others and go along with the system is preferred [in all government
bureaucracies].65 Steve Chan has discerned this conformist instinct inside the
secret agencies. Like other bureaucrats, intelligence analysts have to conform
to the regimes basic views about the nature and morality of international relations if they wish to be treated as responsible and serious, he writes.
Therefore, they refrain from asking the really tough but crucial questions
such as [during the Cold War] the aggressiveness of the Soviet Union, the
morality of the Vietnam War, and the validity of the domino theory.66
The attempt to ensure that policy officers appreciate and understand information provided to them by the intelligence agencies, without misperceiving
or otherwise distorting its meaning, presents another challenge. At times those
in power will embrace intelligence only if it conveniently corresponds to their
existing beliefs and ideologies, rejecting the rest. They quickly learn, observes
a former INR director, that intelligence can be used the way a drunk uses a
lamppost . . . for support rather than illumination.67
The Eisenhower administration reportedly discouraged any assessments
from the intelligence community as to Soviet policy motivation that departed
from the implicit stereotypical cold war consensusespecially the hardline
stance advocated by Secretary of State John Foster Dulles.68 Former DDI
Dr. Ray S. Cline has chronicled the unwillingness of the Johnson and Nixon
administrations to accept the CIAs discouraging reports on the likelihood of
an American victory in the Vietnam War.69
The rejection of objective intelligence became particularly controversial
during the Reagan administration. The White House is said to have dismissed
Meanings and Methods

27

the conclusions of intelligence analysts who called into question the administrations views: that Syria was merely a puppet of the Soviet Union, or that
Nicaragua aggressively exported arms to Marxist guerrillas throughout Central America; that a Soviet oil pipeline to Western Europe would significantly
increase the vulnerability of U.S. allies to Soviet pressure; that the shooting
down of a South Korean passenger airline in 1983 was an intentional murder
of civilian passengers rather than a mistake made by a Soviet fighter pilot
who thought it was a spyplane; and that the assassination plot against Pope John
Paul II in 1984 had been concocted in Moscow.70
The danger of distortion by policymakers is thought to be greatest with political intelligence. On technical mattersmilitary weapons and other difficult scientific or economic subjectsthe policymaker is more inclined to accept the judgment of intelligence experts. Hardware [weapons] estimates . . .
have traditionally been first in acceptance and impact, reports an intelligence
official.71
Wishful thinking is another form of self-delusion that can cause a policy officer to ignore or distort intelligence. A senior CIA officer likes to tell of the man
who bought an expensive new barometer. He took it home only to discover the
needle was stuck on Hurricane, yet there had not been a hurricane for years in
his part of the country, and it was perfectly sunny outside. He shook the barometer gingerly and tapped on the facing. No movement. The man sat down at his
desk and wrote a scathing letter of rebuke to the manufacturer. Then he left home
on a trip. When he returned, the barometer was gone. So was his house.
Ego defense further complicates the use of intelligence. James Thomsons
reflections on decisionmaking during the Vietnam War emphasize the central
fact of human ego investment. Men who have participated in a decision develop
a stake in that decision. As they participate in further, related decisions, their
stake increases.72 Fresh intelligence assessments that call into question their
basic views are unlikely to be well received by individuals in leadership roles
especially when they may have already sent thousands of soldiers to an early
grave to implement their policies. Yaacov Vertzbergers analysis of Indias failure to anticipate a 1962 Chinese invasion concludes similarly: The need to
prove methodically, all through the period in question, that the policy pursued
had been the right one, and that the level of aspirations had been realized, made
it necessary [for Indian policymakers] to ignore any information that contradicted this.73
Even if no distortion of information occurs, have a nations leaders sufficient time to evaluate carefully the implications of the reports placed before
them by the intelligence agencies? A profile of Secretary of Defense Caspar W.
28

Meanings and Methods

Weinberger, who served in the Reagan administration, reported him

swamped, overwhelmed, left with not enough time to think forward.74
Another study of the highest decision echelons in America during the Vietnam
War found widespread executive fatigue, which had a deadening effect on
freshness of thought, imagination, a sense of possibility and perspective. . . .
The tired policy maker becomes a prisoner of his own narrowed view of the
world and his own cliched rhetoric.75 Not exactly a hospitable environment
for the absorption of fresh intelligence insights.
Times winged chariot pulls leaders toward brief forms of current intelligence, as seen in a description of the intelligence cycle offered by a former head
of the NIC. [The analyst must] mine the great lode of outside material, compress it, add the clandestine nuggets, and put it in a form that is usable to policy makers. If you cant get it to them in three pages or three minutes, theyre
not going to get it.76
Perhaps nothing so underscores the importance of the human dimension in
the making of foreign policy decisions as the fragile relationship between the
producer and the consumer of intelligence. Dialogue, rapport, trusthere are
the girders that attempt to bridge the gap between the technology-driven intelligence cycle and the deeply human point of decision. Ambassador Robert D.
Blackwill advocates this widely endorsed prescription: The key [to the success of intelligence] is getting close enough to the individual policy maker to
find out what he needs.77
No doubt many a fine analytic report has died in the in-box simply because
the requisite bonds of trust had never been established between the worlds of
the intelligence officer and the policymaker. A balance between the two can be
hard to achieve, though, because in establishing rapport the intelligence officer must at the same time avoid the trap of intelligence to pleasethe politicization of intelligence, the unforgivable sin.
Every nationlarge or small, rich or poorfaces these intelligence/decision traps. What can be done to avoid them? The answer has roots in ancient
philosophy: select leaders (and intelligence officers) imbued with wisdom and
a love of truththe human virtues, which continue to lag far behind our technological achievements.
The nations secret agencies are but one source of information competing
for the ear of the policy officer.78 Friends and confidants, television news, radio talk shows, influential newspapers, lobbying groups, opinion polls, public
and private pronouncements of foreign leaders, even at times astrologersthis
information stream that feeds into the government is wide and deep.
Meanings and Methods

29

Intelligence from the secret agencies can be a dominant current in this

stream, notably on matters where they enjoy special access to covert information and can proffer a unique, synergistic mix of SIGINT, IMINT, MASINT,
and HUMINT. With respect to weapons proliferation, terrorism, or events inside closed regimes, the clandestine services often have more reliable intelligence (based on covert sources inside an adversarys government) than the media or academe. On other occasions the reverse may be true. Determining the
situation in Rwanda [in 1994] was best ascertained from the people on the
scene, writes a former NRO director. Analyzing its significance and its relevance in that part of the world was best accomplished by scholars and others
dedicated to understanding that society and that area, not members of the current intelligence community, which was developed to address quite different
cultures.79
The secret agencies are likely to be considered by some policymakers a national asset of the highest order, but most think of them simply as one of many
tributaries feeding the information streamsometimes helpful, sometimes
not. And for a fewusually those who have never taken the time to discover
the value of intelligencethe secret government will be discounted altogether,
as if its bed had run dry, leaving nothing to offer that could not be found in the
nations best newspapers.

30

Meanings and Methods