For GuaranteedGrades:

Glimpses into the Gems of American Intelligence: The President’s Daily Brief and

the National Intelligence Estimate

LOCH K. JOHNSON

The two most prestigious products prepared by the US intelligence

agencies for use by decision-makers in Washington, DC, are the

President’s Daily Brief and the National Intelligence Estimate. The Brief,

an example of ‘current intelligence,’ adds value to what policy officials in

Washington can learn about world affairs from the best newspapers,

especially in the domains of foreign weaponry, activities within closed

societies, and the machinations of terrorist organizations. The National

Intelligence Estimate, an example of ‘research intelligence,’ has added

value, too, on occasion, but has often been wrong. Each of these forms of

intelligence has their critics, and the NIE in particular is frequently

considered too long a document and too diluted in content. The production

of NIEs has varied over the years since 1950, averaging twenty-three a

year with a low of five (in 1976) and a high of fifty-six (in 1992).

INTRODUCTION

Nothing lies more at the heart of intelligence in any nation than the quality of

the information prepared for policymakers by its secret agencies. The

government in Washington, DC, reportedly spends some $44 billion a year to

provide the president and other leading officials with the most accurate facts

and insights available on threats and opportunities that face the United States.1

These findings are packaged into two premier reports. One is the President’s

Daily Brief (PDB), based on current intelligence; it is designed to give policy

officials an understanding about what happened in the world during the past 24

hours, and what is likely to happen during the next 24 hours. The second is the

National Intelligence Estimate (NIE), based on research intelligence; it is

designed to provide officials with more in-depth information on some

international topic of interest, say, the future of the Chinese military

establishment. This essay explores the world of the PDB and the NIE.

Intelligence and National Security, Vol.23, No.3, June 2008, pp.333–370 ISSN 0268-4527 print 1743-9019 online DOI: 10.1080/02684520802121257 ª 2008 Taylor & Francis

THE PRESIDENT’S DAILY BRIEF

An Elite Publication

The President’s Daily Brief came to the public’s attention dramatically in

2003 during the Kean Commission inquiry into the 9/11 attacks against the

United States.2 Thomas Kean, the panel’s chairman, referred to the PDB as

the ‘Holy Grail of the nation’s secrets’.3 The Kean Commission, reluctantly

appointed by the Bush Administration under pressure from the families of the

9/11 victims, soon found itself at war with the White House and other

executive agencies over access to documents that might help it determine

why the nation failed to prevent the tragedy. The PDB stood at the center of

this war.

The Daily Brief is a highly classified intelligence report sent each morning

from the Central Intelligence Agency (CIA) to the president and a limited

number of other officials in the top reaches of America’s national security

establishment. The bipartisan Commission concluded that it was vital to

examine the intelligence on potential terrorist activities that the president had

received in the PDB prior to the Al Qaeda attacks against the Twin Towers in

New York City and the Pentagon in 2001.

In dismay over executive branch delays in providing documents related to

the inquiry, the Commission first issued subpoenas to the Department of

Defense and the Federal Aviation Administration. Then Chairman Thomas

Kean threatened to use his panel’s subpoena powers against the White House

itself, if necessary, to obtain the necessary PDBs. At first, the White House

backed away from a confrontation and agreed to allow Commission members

limited access to the document. President George W. Bush insisted, however,

that his staff should determine who among the ten Commission members

would be allowed to review the Daily Briefs that came to the White House

during the summer of 2001 and that, further, only a few commissioners would

be allowed to participate in the review.

When the Commission balked at this attempt to interfere with its

independence, the White House retreated somewhat. Its new position was

that the Commission would be allowed to choose from among its membership

who would have access to the documents; but, for security purposes, no more

than two individuals would be allowed to read the Daily Briefs. In addition,

another two commissioners would be given the right to examine sections of

the documents that the White House deemed ‘relevant’. While a majority of

the Commission members were willing to accept this compromise, a vocal

minority objected to the restrictions. The dissenters argued that each

Commission member would have to vote on the final report; therefore, to

fulfill his or her responsibilities, each commissioner should be granted the

authority to read the key PDBs.

334 INTELLIGENCE AND NATIONAL SECURITY

From the standpoint of the dissenters, the Kean Commission needed to

determine the value of the PDB. To what extent did this current intelligence

report aid the efforts of President Bush and, before him, President Bill Clinton

to protect the nation against terrorist attacks? An answer to this question

required the judgment of the full Commission, not just two members – or even

the addition of two more members allowed to look at some parts of the

document. The right of the American people to know the answer trumped the

doctrine of executive privilege. Besides, the rightful purpose for evoking

executive privilege is to prevent a chilling effect with respect to policy advice

given to the president; intelligence agencies, in contrast, provide facts and

insights into world affairs, not policy advice. Nor was the doctrine of state

secrets in play, since the members and staff of the Kean Commission had all

the necessary security clearances and a right to know.

The ultimate purpose of the Kean Commission’s work was to establish

better counterterrorist safeguards for the United States. To this end, it had to

understand what Presidents Clinton and Bush knew, when they knew it, and

what steps they took to defend the nation. Whatever weaknesses the

commissioners discovered, either at the CIA or inside the White House, could

then be addressed.

White House stonewalling on these documents inevitably raised suspicions

that administration officials were simply trying to cover up mistakes that had

occurred in the Oval Office during the months leading up to 9/11. Did the

surprise terrorist attack represent an intelligence failure, a policy failure, or

both? Hidden within the rarified pages of the PDB, the Kean Commission

might have found an answer – if it had been given a chance. Instead, the Bush

Administration steadfastly refused to let the full Commission study the

documents.

This was not the first time a commission had been refused entry into the

hidden world of the President’s Daily Brief. In 1995, Congress charged the

Aspin-Brown Commission with the duty of evaluating the state of American

intelligence in the aftermath of the Cold War.4 The Commission decided to

evaluate the quality of intelligence reports coming into the Oval Office,

especially the PDB. The CIA permitted the author, who served as special

assistant to Les Aspin, Commission chairman and former secretary of

defense, an opportunity to review a few recent PDBs at its headquarters in

Langley, Virginia. To prepare a more meaningful evaluation, the Commis-

sion requested access to copies going back several months. In response to this

request, the door to the Daily Briefs slammed shut.

The pugnacious Les Aspin may have taken up the quest for access to

additional PDBs, but he died early in the inquiry and Harold Brown, a former

secretary of defense in the Carter Administration, replaced him. The

Commission under Brown decided not to push the issue, fearing that a fight

THE PRESIDENT’S DAILY BRIEF AND THE NIE 335

with the CIA on this matter might derail the Commission’s broader goals of

intelligence review. As usual when it comes to thePDB, the CIA got its way. At

least in 2003, the Kean Commission did manage to pry out into the public

domain a revealingBrief entitled ‘Bin LadinDetermined to Strike inUS’, dated

6 August 2001.5

Inside the PDB

From among the hundreds of classified reports prepared each year, the PDB is

considered the most prestigious document provided to senior policy officials

by the nation’s so-called intelligence community (16 agencies that often more

resemble separate fiefdoms than a ‘community’). A former Director of

Central Intelligence (DCI) has referred to the PDB as ‘our most important

product’.6

The Brief is certainly the most tightly held of the many reports written by

the intelligence agencies. Few have emerged from the CIA’s vaults into the

public domain.7 The document is distributed by CIA couriers around 7:00

each morning, but only to the president and a few top cabinet officials and

presidential assistants. The number of recipients has varied from adminis-

tration to administration, rising (for example) to as many as 14 in the Clinton

Administration and as few as five in the Reagan Administration and six in

the second Bush Administration. Always on the distribution list are: the

president, the vice president, the secretaries of state and defense, and the

national security adviser. The document often sets the agenda for early

morning discussions among these individuals and their top aides. It serves as

a ‘catalyst for further action’, in the words of a staff aide on the National

Security Council (NSC).8

Since 2005, the United States now has a Director of National Intelligence

(DNI), who is no longer located at the CIA (where the DCI kept his office);

nonetheless, the President’s Daily Brief is still put together primarily by staff

at the CIA. This is true because most of the government’s intelligence

analysts are at the CIA, housed in its Directorate of Intelligence (DI).

Moreover, the printing presses and distribution system for the intelligence

community are also located on the CIA’s compound in Langley, Virginia,

near McLean. Both the Director of the Central Intelligence Agency (D/CIA)

and the DNI attend the morning oral briefing on the PDB at the White House.

A typical sequence in the preparation of a Brief begins at 8:30 on the

morning 24 hours before it will be distributed to policymakers, with meetings

at the CIA among the staff responsible for assembling the document. The

purpose is to decide what subjects will be included in the Brief, especially the

hot topics of the moment that are likely to be addressed in the New York

Times, the Washington Post, the Washington Times, and the Wall Street

Journal. As the ‘menu’ of items to include unfolds, contacts are made by this

336 INTELLIGENCE AND NATIONAL SECURITY

group with analysts throughout the intelligence community who can

contribute to the topics selected for the Brief. This call for material goes

on from 10:45 that morning until late into the afternoon and early evening.

The various offices are asked, ‘What can you provide on this subject?’ The

responses come streaming back to the CIA and the drafting of the document

begins.

Between 7:30 and 8:00 that same evening, the PDB draft is ready for

review by the CIA’s Deputy Director for Intelligence (DDI) and the Agency’s

Director (D/CIA). If they have left their offices, the draft is sent to their

homes over a secure fax machine. Any revisions they have are returned to the

PDB working group, as soon as possible, for integration into the document.

The production staff takes over late in the evening and works through the

early morning hours of the next day, printing and collating the document –

similar to newspaper crews around the nation who are also preparing their

newspapers for delivery. At 5:30 a.m., the PDB is ready for the CIA couriers,

who (armed) carry the document to those who are on the exclusive list of

subscribers. The Brief is handed to these principals between 6:00 and 9:00

a.m., depending on when they want to receive the document and the follow-

up oral briefing that accompanies it (if they want one).

The format of the Daily Brief has varied over the years, though it has

always had three core objectives: readability, logical reasoning, and faithful

adherence to the intelligence community’s sources. During the Ford

Administration it ran over 20 pages in length, on average. President Jimmy

Carter reduced it down to some 15 pages. During the Clinton Administration,

it was 9–12 pages long and printed in impressive four-color graphics (vividly

displaying, for instance, global economic trends in colored lines on a graph).

Throughout the second Bush Administration, a DCI reports that the document

was only 1–2 pages long, printed on heavy paper, and presented in a leather

binder.9

Whatever its length or format, the PDB – ‘the book’, as it is known

inside the CIA – is designed to grab the attention of busy policymakers and

inform them about events that have just transpired around the world. It

features articles expected to be of ongoing relevance to the interests of the

White House, perhaps the health of an aging and important foreign leader

(a subject of considerable discussion in the 1990s when Soviet President

Boris Yeltsin suffered a serious decline in health). The cadre of intelligence

officers at the CIA responsible for the all-night, intense production cycle

required to prepare the PDB takes pride in its work. The Brief’s spiral-

bound, glossy pages are attractive and easy to read. Unlike regular

newspapers, policymakers do not have to flip over cumbersome pages

searching for the continuation of a page A1 story on page A6; the stories in

the Brief flow continuously. Further, the PDB focuses on topics known to

THE PRESIDENT’S DAILY BRIEF AND THE NIE 337

be high on the president’s agenda, rather than the daily smorgasbord offered

by regular newspapers.

Further, the Brief attempts to integrate information clandestinely gathered

from around the world by America’s intelligence agencies into what

intelligence professionals refer to as an ‘all-source fusion’ or (in the

Pentagon) ‘jointness’. This integration of information provides policymakers

with a comprehensive view of international affairs based on human spying,

photography from surveillance satellites and reconnaissance aircraft, and

telephone taps, all woven into a underlying framework of open-source

information. The Brief allows the president and other busy national leaders

‘one-stop shopping’ for up-to-date global information.

The PDB comes with another important service unavailable to ordinary

newspaper subscribers: follow-up oral briefings tailored to answer the

specific questions of its VIP readers – 30–60 minutes of additional

information, depending on the interest and patience of the policymaker,

presented by intelligence experts on any of the articles published in the Brief.

Here is a rare opportunity for a president or other PDB readers to talk back to

their ‘newspaper’. The DCI often led the oral briefing while that office

existed (1947–2005). As insisted on by the second President Bush, now the

DNI or, pinch-hitting, the D/CIA, carries out this task. Under President

George W. Bush, six days a week (Sunday is normally a day off) the briefing

starts around 7:45 and lasts an hour or less. The numbers of people present

varies from day to day, with the president, the vice president, the secretaries

of state and defense, the White House chief of staff, and the national security

adviser almost always in attendance, supplemented (depending on the most

important topics in the PDB that day) by such people as the attorney general,

the secretary of the Department of Homeland Security, the FBI director, and

the adviser to the president for homeland security.10 Four days a week, the

DNI brings in one or more intelligence analyst to participate in the briefing –

‘someone who knows more than I do’, as the current DNI, Mike McConnell,

has put it.11

During a typical year of the Clinton Administration, 42 follow-up oral

briefings took place in the offices of 13 PDB recipients; and the CIA sent an

additional 426 memoranda to those recipients who requested more detailed

written responses to their queries. About 75% of these follow-ups occurred by

the next working day.12 The PDB is more than a document; it is a process,

allowing intelligence officers to interact with decision-makers and provide

useful supportive information. As a NSC staffer has noted, this interaction

keeps ‘the CIA boys hopping, but, most importantly, it lets them know what

is of interest at any given time to the President’.13

As to whether or not the President’s Daily Brief adds value over news

available through television and newspapers, presidents and some other

338 INTELLIGENCE AND NATIONAL SECURITY

subscribers in the small ‘witting circle’ of PDB readers have often

complained about the quality of the document. George W. Bush, for

instance, received the PDB during his first presidential campaign in 2000,

along with other leading candidates – a service provided by the CIA since

1952 to individuals who may soon find themselves as the nation’s chief

executive. He found them unhelpful and remarked: ‘Well, I assume I will

start seeing the good stuff when I become president’, little knowing that the

intelligence community was already giving him the best ‘stuff’ it had.14

Nevertheless, an examination of PDBs by the Aspin-Brown Commission

provided evidence that the intelligence agencies are often able to supply

information beyond what the open media has to offer – although frequently it

falls short of this goal. Among the topics examined by the Commission was

the terrorist attack of 1995 in Japan, when the group Aum Shinriko

(‘Supreme Truth’, in Japanese) released lethal sarin nerve gas into the Tokyo

subway system; and the question of whether the Chinese were selling M-11

missiles to Pakistan between 1989 and 1995. The Commission also looked

into intelligence reporting on unrest in Burundi during 1995.15

In the sarin case, the PDB reader would have learned no more than the

average newspaper subscriber about the details of the attack, but would have

found out additional details about how the Aum Shinriko cult financed its

operations and about the background of its leader – although not appreciably

more. In fact, the US intelligence community (like the Japanese authorities)

were surprised by the subway attack. The CIA and its companion agencies had

virtually no information about the sect’s past activities in Japan, Russia, or

even the United States. Only after the attack did the CIA discover that the Aum

Shinriko leader was virulently anti-American; that he advocated chemical–

biological (CB) and even nuclear war against the United States; and that he

was making progress toward acquiring CB weapons of mass destruction.

As for the M-11 missile controversy, reporting in both the PDB and public

newspapers displayed many ambiguities. In both venues one could learn a

fair amount about the alleged missiles sales; but the intelligence community

possessed photographic and eavesdropping information that moved the case

from pure speculation to reasonably strong circumstantial evidence that the

Chinese were indeed providing the Pakistanis with missile components. The

sighting of ‘cylindrical objects’ at the Sargoha Missile Complex in Pakistan

or ‘unidentified, suspicious cargo’ being unloaded in Karachi harbor, as

reported by intelligence assets, proved nothing; however, when coupled with

telephone intercepts between Pakistani and Chinese officials about M-11

contracts and photographs of M-11 TELs (transporter-erector-launchers) at

Sargodha, the president had more clues than offered by the New York Times.

In the Burundi case, the Aspin-Brown Commission asked five news services

to provide information on the current situation in the African nation and some

THE PRESIDENT’S DAILY BRIEF AND THE NIE 339

background on the existing unrest – all within 24 hours. From among the

public sources of information, only Jane’s Information Group provided data

that was unavailable in the PDB.16 Jane’s furnished the Commission with two

loose-leaf notebooks filled with order of battle statistics, detailed descriptions

of weapons systems in the Burundi inventory, and a history of the Hutu–Tutsi

conflict. In contrast, the PDB had more information about the internal political

polarization in Burundi; the prospects for a humanitarian disaster on a par with

the Rwanda genocide a year earlier; the number and location of US and

European nationals in-country; and insights into ethnic patterns, arms

acquisition, and the prospects for massive bloodshed – all based on human

sources inside the nation. Jane’s provided the most detail on Burundi weapons

systems, but US intelligence offered what policymakers desire above all else:

‘actionable intelligence’ – information they can act on. For instance,

intelligence on an arms shipment to Burundi from another nation led to

quick diplomatic pressure by the United States to halt the shipment.

Satellite, U-2, and drone photography, plus wiretaps overseas and the

occasional well-placed spy, are bound to provide the intelligence agencies

with an edge in some instances over the Times, thereby giving the Brief added

value. This was particularly evident in October of 1962 during the Cuban

missile crisis, when agent reports, followed up with confirming U-2

photographic missions, proved vital for disclosing the new Soviet threat just

90 miles from America’s shores.

During the Cold War, the intelligence agencies added great value over

newspaper reporting when it came to knowing the number and specifications

of Soviet bombers and ICBMs. As well, the intelligence community had the

critically important capacity to quickly discern whether the Soviets were

preparing for a first strike against the United States or Europe. Pearl Harbor

surprise attacks became infinitely less likely, thanks to this capability, and, as

a result, the hair-trigger nervousness on both sides of the Iron Curtain began

to relax, tamping down the danger of an accidental missile launch. More

recently, the Aspin-Brown Commission found that PDBs were especially

effective when reporting on the weapons capabilities of foreign nations and

factions, on events within closed societies, and on the activities of terrorist

organizations – topics that regular newspaper correspondents have a difficult

time covering, because of the secretive nature of these subjects and the

danger of trying to enter restricted territories to report on them.

Even with the advantage of its clandestine human and mechanical assets

around the globe, the intelligence community’s information – if better than

the nation’s newspapers on some occasions – still remains skimpy in

significant instances, such as reporting on conditions in North Korea. With

respect to broad political and economic issues, say, the prospects for further

European integration, the latest twists in German or French politics, or the

340 INTELLIGENCE AND NATIONAL SECURITY

state of the economy in China, the open media often present better insights

than the intelligence agencies. Frequently, newspapers will have correspon-

dents who have served in some countries, such as Germany or France, longer

than any CIA officer.

In a fast-moving incident of short duration (like the sarin attack in Japan),

the public media is apt to know as much – and sometimes more – about what

has happened than the intelligence community, especially if satellite

photography is irrelevant to the situation. With subjects that unfold over a

longer period and involve deception by nations (like the M-11 story), the

intelligence community has a better chance of focusing its clandestine assets

and adding value to the public reporting. Aiding the accuracy of public

reporting is the fact that regular newspaper and magazine reporters have

sources within the intelligence community. Conversely, the secret agencies

have ongoing conversations with US journalists before and after these

correspondents travel overseas – a controversial relationship for those

concerned about keeping the media free from government influence, but a

relationship that is nonetheless widespread and persistent.17

Even those in high office who have lost faith in the ability of US

intelligence reports to add value to newspapers are likely to read the PDB

anyway, if they are fortunate enough to be on the distribution list. As

Secretary of State George P. Shultz of the Reagan Administration wrote in a

memoir, ‘I had no confidence in the intelligence community . . . [but] I

continued to read The President’s Daily Brief, in part to know what was being

put before [the president].’18

NATIONAL INTELLIGENCE ESTIMATES

NIEs in the News

Like the PDB, the subject of National Intelligence Estimates also caught the

media’s attention in the aftermath of 9/11, only this time because of the

intelligence community’s faulty prediction in 2002 about the likely presence

of weapons of mass destruction (WMD) in Iraq. Caught up in the swift

moving events that followed hard upon the 9/11 attacks, notably the US

invasion of the Taliban-led Afghanistan (the chief Al Qaeda haven), Director

of Central Intelligence George Tenet never got around to ordering the

preparation of an NIE on Al Qaeda or on suspected Iraqi WMD. Nor did the

White House. Reportedly, the president’s advisers feared that a full-blown

Estimate on the WMD question would reveal ‘disagreements over details in

almost every aspect of the administration’s case against Iraq’.19

An NIE on Al Qaeda may have been handy but not immediately necessary,

since the course of action was clear in this case: the United States would

retaliate against the terrorist group that had just attacked the nation and

THE PRESIDENT’S DAILY BRIEF AND THE NIE 341

against its host, the Taliban regime in Afghanistan. Less excusable, however,

was the lack of an NIE on Iraqi WMD at the very time the United States

was engaged in an important internal debate over whether to launch a war

against Saddam Hussein. Rumors about the presence of WMD in Iraq were

rife and inflamed by comments about ‘mushroom clouds’ appearing on

American soil, expressed by the president and national security adviser

Condoleezza Rice.20 The nation began to fear that Saddam might use

unconventional weapons against the American homeland, even though no

policy official had presented any empirical evidence to support the Iraqi

WMD hypothesis.

Tenet has admitted his error in not having prepared a timely NIE on this

topic. ‘An NIE on Iraq should have been initiated earlier, but at the time I

didn’t think one was necessary’, he has written. ‘I was wrong.’21 Various

other intelligence reports had been prepared in previous months suggesting

the probability of Iraqi WMD; but a formal, community-wide NIE had not

been written on the subject for years.

Senators Richard Durban, Democrat from Illinois, and Carl Levin,

Democrat from Michigan, both members of the Senate Select Committee

on Intelligence (SSCI, one of the two intelligence oversight committees in

Congress), believed that such a study would be important in light of the

debate over war against Iraq. They insisted on having a formal written

assessment and persuaded the SSCI Chairman, Bob Graham (Democrat from

Florida), to send a letter on 10 September 2002 to DCI Tenet requesting that

an NIE on Iraq be prepared as soon as possible. The Estimate was expected to

answer such questions as: does Saddam have WMD already, and, if not, when

could he be expected to have them? Was war against Iraq necessary? What

would be the nature of the forces used against an invading US army? What

conditions would the United States face in a postwar occupation?22

Tenet replied that he would be unable to produce the kind of

comprehensive NIE on Iraq that Graham sought, because of other pressing

intelligence duties. Nevertheless, he promised to furnish, as soon as possible,

an Estimate on the subject of WMD in Iraq.23

The preparation of an NIE can take two to four weeks on a fast track, two

to six months in normal times, and seven months or even three years on a

long track.24 Historically, an NIE on average has taken 215 days to produce –

about seven months. A crisis can accelerate the process. During the Suez

Canal crisis of 1956, the intelligence community produced an NIE on Soviet

intentions within a few hours; those responsible for NIEs, however, prefer to

have at least three months to produce an Estimate.25 In the case of Iraqi

WMD, Tenet ordered a ‘crash project’ to meet the SSCI request. The 90-page

Estimate on Iraqi WMD went to the Senate about three weeks after the

request – too hastily prepared, in the view of critics. At least one study has

342 INTELLIGENCE AND NATIONAL SECURITY

called it ‘the worst body of work in [the CIA’s] long history’.26 It arrived at

SSCI’s quarters in the Dirksen Office Building in early October and Tenet

came to the building to brief SSCI members on its main points.27 In

retrospect, Senator Graham feels that the DCI seemed to skate over dissenting

views in the report that downplay the Iraqi threat; still, dissenting views were

clearly presented in the NIE and opponents of the war grasped onto them.

Senators Graham, Durbin, and Levin next sought to have the NIE

declassified for public consumption, except for portions that might disclose

sensitive sources and methods. They made the request on 2 October 2002,

and two days later Tenet delivered an unclassified version of the longer

document, this one 25 pages in length. The problem was, at least from

Graham’s point of view, that the new version ‘did not accurately represent the

classified NIE we had received just days earlier’.28 Missing was the sense

from the still-classified document that Saddam Hussein posed no immediate

danger to the United States or to his neighbors, if he were simply left alone.

In Graham’s opinion, Tenet had diluted the original document to keep in step

with the opinion of the White House that Saddam was a great danger.29

Republican Senator Chuck Hagel flatly concludes that the condensed NIE

was ‘doctored’ to suit the political needs of the Bush Administration.30

Others lawmakers may have agreed with Graham that the declassified

version of the Estimate was substantially different from the original, except

that apparently most of his colleagues never bothered to read the first,

classified report.31 Further, its ‘Key Judgments’ section was never released to

the Congress and the public until 16 July 2003 (the US/UK invasion of Iraq

began on 19 March 2003); and a more complete, but still redacted, version

never surfaced in the public domain until 1 June 2004. In a formal report

released in July 2004, Graham’s Committee (SSCI) concluded that the Key

Judgments were, for the most part, ‘either overstated, or were not supported

by the underlying intelligence reporting’.32

Only much later, in 2007, in a memoir published in the throes of the war in

Iraq, did Tenet acknowledge that ‘we should have said, in effect, that the

intelligence was not sufficient to prove beyond a reasonable doubt that

Saddam had WMD’. He now believed that ‘more accurate and nuanced

findings would have made for a more vigorous debate – and would have

served the country better’.33

In December of 2007, the declassified sections of another NIE was

splashed across the front pages of the New York Times and other newspapers.

The key judgments of this Estimate, entitled Iran: Nuclear Intentions and

Capabilities and dated November 2007, concluded that Iran had halted its

nuclear weapons program in 2003 – a dramatic reversal of an earlier NIE in

2005 that warned that Iran was seeking a nuclear weapons capability.

‘Rarely, if ever’, concluded a New York Times correspondent reporting from

THE PRESIDENT’S DAILY BRIEF AND THE NIE 343

Washington, DC, ‘has a single intelligence report so completely, so suddenly,

and so surprisingly altered a foreign policy debate here.’34

In light of the intelligence community’s 2003 Iraqi WMD Estimate (most

of which proved wrong) and its assertion in 2005 that Iran had a robust

nuclear weapons program underway, opinion on this new assessment was met

with skepticism in some quarters – especially among hawkish anti-Iranian

members of the Bush Administration. Even within the intelligence

community, some organizations (the National Intelligence Council and the

Department of Energy) expressed only ‘moderate confidence’ in the NIE’s

conclusions, rather than the ‘high confidence’ endorsed by the other secret

agencies.35 Whether right or wrong in its latest judgment about Iranian

weapons development, the intelligence community had demonstrated again

that its Estimates could quickly rile the political waters of the nation’s

capital.36

Inside the NIE

A National Intelligence Estimate is an appraisal of a foreign country or

international situation, authorized by the DCI (or, since 2005, the DNI) and

reflecting the coordinated judgment of the entire intelligence community.

Estimates often serve as the building blocks of national security policy. They

are the outcome of an intricate gathering and evaluation of intelligence drawn

from all sources. They are not limited to the task of predicting specific events;

indeed, their primary responsibility is to assist the president and other US

leaders to protect the nation from danger by making available the best

possible deep understanding of foreign leaders, developing events, and the

military and economic capabilities of other nations. An Estimate will set

down on paper, and often rank, a range of possible outcomes related to

developments inside another nation or faction, or the likely unfolding of a

situation somewhere in the world that threatens US interests.

A CIA official has offered this straightforward definition of an NIE: ‘a

statement of what is going to happen in any country, in any area, in any given

situation, and as far as possible into the future’.37 Sherman Kent, the early

pioneer and father figure of CIA analysis, noted that Estimates consist of

three elements: knowledge, reasoning, and guesswork. This last dimension

‘may be sullied by visceral reactions’ unless the analyst is careful not to allow

personal biases to creep into his or her work.38

Many insiders and outsiders alike regard the NIE as the crown jewel of US

intelligence analysis. A respected former CIA analyst notes that an Estimate

is the ‘most authoritative analytic product prepared by the intelligence

community . . . the bringing together of every scrap of evidence, from the

most sensitively exotic to the most openly unclassified, that the U.S.

intelligence community has on the question at hand’.39 In the judgment of

344 INTELLIGENCE AND NATIONAL SECURITY

political scientist Harry Howe Ransom, the NIE is ‘the single most influential

document in national security policy making, potentially at least’.40

The first NIE in the United States appeared on 8 November 1950, on the

subject of Chinese intervention into the Korean peninsula. The Estimate

noted that there were some 30,000 to 40,000 Chinese troops in North Korea,

and that the Chinese intention was to maintain its presence there. Ever since

this inaugural NIE, the production of Estimates has often begun with a formal

request from a senior policymaker – whether the president, the national

security adviser, the secretary of state, a military commander, or a member

(or members) of Congress – for an appraisal and prognosis of events and

conditions in some part of the world. Two of the most well known NIEs, the

2003 Estimate on Iraqi WMD and the 2007 Estimate on the Iranian nuclear

program, were both requested by members of Congress. However, in an

overwhelming majority of cases – 75% in one recent year – the intelligence

community itself has generated most of the NIE proposals, although these

internal initiatives are usually based on a perception that policymakers would

like to see a more definitive analysis of some global topic. Many of those

knowledgeable about the NIE process believe that the most useful Estimates

are the ones specifically requested by policymakers, because then the

potential readers have indicated an interest in having more information and

are thus likely to pay closer attention to the final product. In a phrase, as a

former top NIE briefer to the White House has said, ‘ideally Estimates should

be consumer-driven reports’.41

The subjects for NIEs cover a wide front. Former DCI Stansfield Turner

provides these examples from the Carter Administration:

. the balance of strategic nuclear forces between the United States and the USSR;

. the conventional military balance in Europe;

. the prospects for improvement in relations between the Soviet Union and China;

. the outlook for cohesiveness within the Atlantic Alliance; and

. the significance of the third world’s international debt problems.42

A look at NIEs prepared from 1992 to 1995 offers a sense of how they have

been distributed across the regions of the world, as well as across several non-

geographic subjects.43 Regionally, Central Asia headed the list with 41

Estimates. A lingering post-Cold War interest in Russia caused that nation – the

focus of ten NIEs – to far outdistance all others in the region in terms of US

estimating. Next came the Near East with 24 Estimates and Africa with 20; then

Latin America with 16, and East Asia with 15. Substantially below these regions

came Europe, along with the subject of global weaponry and arms control (tied

THE PRESIDENT’S DAILY BRIEF AND THE NIE 345

at nine). Bringing up the rear were a range of other global issues, such as

immigration and environmental degradation (seven); international crime (one);

health (three, including one on HIV/AIDs); drugs (two); economics (one on the

effect of the General Agreement on Tariffs and Trade, or GATT); international

crime (one); space (one); human rights (one); and, soon to be Public Enemy

No. 1, terrorism (one). During the ColdWar, the array would have been roughly

in the same order, with Central Asia still dominant but with NIEs focused even

more overwhelmingly on Russia. This brief post-Cold War portrait of Estimates

discloses the intelligence community’s attempts to move, slowly but surely,

toward the reality of a more complex combination of threats and opportunities

in a no-longer bipolar world.

A panel of intelligence experts, known since 1980 as the National

Intelligence Council or NIC (and earlier as the Office of National Estimates,

ONE), examines the merits of each proposed Estimate in consultation with

analysts throughout the community, as well as with senior policy officials, to

determine the feasibility and demand for such a study. If the decision is to

move ahead, the NIC determines what segments of the community could best

contribute to the Estimate and provides these selected agencies with an outline

of the NIE’s objectives, asking them to respond with their facts and insights by

a certain deadline. This outline is known as the Terms of Reference or, in the

inevitable Washington acronym, TOR. As a NIC document explains: ‘The

TOR defines the key estimative questions, determines drafting responsibil-

ities, and sets the drafting and publication schedule.’44

In response to the TOR, data and ideas pour back to the NIC from around

the community and are shaped into a draft NIE by one or more of the senior

analysts that comprise the NIC, in continual dialogue with experts further

down the chain of analysts. Since 1973, the senior analysts on the NIC have

been called the National Intelligence Officers or NIOs. The men and women

who serve as NIOs are expected, according to Sherman Kent, to have ‘the

best in professional training, the highest intellectual integrity, and a very

large amount of worldly wisdom’.45 Usually the NIO with particular

expertise on the subject at hand will lead the drafting. The National

Intelligence Council and its NIOs have been affiliated with the DCI, until that

office was replaced in 2005 with the new DNI. When that transition occurred,

the DNI set up shop (at least temporarily) at the headquarters building of the

Defense Intelligence Agency (DIA), located at Bolling Air Force Base, across

the Potomac River from National Airport. While most components of the

NIC remain at CIA Headquarters, the Council now reports to the DNI.

The 10–16 or so NIOs (the number varies from time to time) are considered

the crème de la crème of intelligence analysts. They are drawn from

throughout the community and occasionally from academe. A recent set of

NIOs (in 1996) consisted of four career intelligence officers; five analysts

346 INTELLIGENCE AND NATIONAL SECURITY

from academe and the nation’s think tanks; three from the military; and one

from Capitol Hill. The NIC in recent years has also maintained a Rolodex

file of some 50 consultants with security clearances and based in the worlds of

academe and the think tanks; they are consulted on a case-by-case basis.

The NIOs organize conferences with outsiders as well, in an attempt to

keep themselves well informed about world affairs. Perhaps the best known

of the intelligence community’s efforts to reach outside its walls to seek

academic consultation on the substance of a NIE came in 1976, by way of an

‘A-Team, B-Team’ review of a 1975 Estimate on Soviet intentions and

capabilities.46 The NSC selected the two teams. The CIA’s own Soviet

experts comprised the A-Team; and academics comprised the B-Team, led by

Harvard University Russian historian Richard E. Pipes, known for his

strongly hawkish views on the Soviet Union. Pipes and his panel were

convinced that the CIA had gone soft; its liberal ‘civilian’ views, reinforced

by arms-control experts in the scholarly community, had led to an NIE that

downplayed the Soviet intent on world conquest. In the B-Team view, the

reality was that the Soviets were subtly seeking a first-strike, war-winning

strategy, not peaceful co-existence. Specifically, the B-Team accused the CIA

of miscalculating Soviet expenditures on weapons systems, thereby under-

estimating the formidable strength of the Red Army; the A-Team, in turn,

charged the Pipes panel with hyping the Soviet peril.

The upshot of this attempt at ‘competitive analysis’ using outsiders: the

CIA trimmed back on some of its calculations in the next annual Estimate on

the USSR, adopting figures slightly more in line with the B-Team projections.

Nevertheless, a vast gulf between the two groups continued to exist on the

subject of Soviet intentions: the more optimistic views of the A-Team set

against the pessimism of ‘hard-liners’ like Pipes. The ‘debate’ probably

damaged the reputation of the intelligence community; the door had been

opened to doubt about the wisdom of relying alone on internal judgments

made by professional intelligence officers. Yet, overall, it was healthy for the

intelligence analysts to have their views tested by external experts – although

the selection of an outside review board known for a particular ideological

stance is apt to be less useful than recruiting specialists with no axes to grind.

During the NIE drafting process, the NIO in charge will send the first draft

back to all of the intelligence agencies working on the Estimate and thus the

process of interagency editing begins, as experts from throughout the

community hammer into shape the final NIE document. A recent analyst

recalls the process in these painful words: ‘It was like defending a Ph.D.

dissertation, time after time after time.’47

The NIC makes the penultimate judgment on the appropriateness of the

findings and conclusions presented in each Estimate, then sends the document

along to the National Intelligence Board for further review. The NIB is

THE PRESIDENT’S DAILY BRIEF AND THE NIE 347

comprised of the senior representatives of the intelligence community and

chaired by the DNI, who has the last say on the Estimate before it is

distributed to the president and senior policymakers.48 Sometimes in the past

a DCI (and perhaps a DNI in the future) has so disliked an Estimate produced

by the intelligence bureaucracy that he has written one himself on the topic,

instead of sending forward the NIC version. This practice is rare, however,

and carries with it the danger of an Estimate becoming too personalized or

even politicized.49 Sometimes intelligence chiefs can be correct and the

bureaucracy wrong, as when DCI John McCone rejected the conclusion of an

NIE in 1962 that predicted the Soviets would not place missiles in Cuba.50

The best bet, though, is usually to rely on well-trained and experienced

country or subject experts. If a DNI or a D/CIA disagrees with the experts, he

or she can forward the NIO version along with a clearly marked dissent from

the director’s office.

The bulk of the NIE drafting goes to more junior analysts within the

intelligence community – specialists who study the daily cable traffic from the

country in question or are otherwise expert in the topic under consideration

(say, the efficiencies of Chinese rocket fuel, on the narrow side, or the likely

path of leadership succession in China over the next two decades, on the

broader side). According to an NIO, the National Intelligence Council is

expected to work closely with NSC members and their staff, as well as other

intelligence consumers, ‘scrubbing information honestly and adapting to the

working style of the policymakers receiving the NIE’. The NIOs are expected,

as well, to keep in touch with the various intelligence entities whose analysts

contributed to the NIE draft. The preparation of an Estimate is, according to

this same insider, ‘an art form [that] requires a corps of floating linebackers,

flexible and easily collapsible, to charge an intelligence problem’.51

This process has not been as smooth as it may sound, even beyond

the inherent difficulty of forecasting the future of the world.52 Obviously, the

tenor of the language in an NIE is all-important, especially the confidence

levels evinced by the document. The NIOs must be careful not to claim more

than the evidence can support, especially in the executive summary (called

‘Key Judgments’) found at the beginning of an Estimate. This may be the

only portion read by a harried (or lazy) policy official and it needs to convey

the shades of gray, and the caveats, that serve as an antidote to overly

assertive and simplistic conclusions. Sometimes reporting in the PDB and an

NIE are inconsistent on the same topic, in their confidence levels and use of

assertive language, thereby sowing confusion among readers.53

In an effort to improve the clarity of NIE judgments, the NIC has

developed a hierarchy of terms to express an Estimate’s level of confidence in

the likelihood of its forecasts coming true, ranging from ‘remote’ to ‘almost

certainly’ with this gradation in between: ‘very unlikely’, ‘unlikely’, ‘even

348 INTELLIGENCE AND NATIONAL SECURITY

chance’, ‘probably/likely’, and ‘very likely’.54 Even when the language in the

PDB or the NIE is guarded, irresponsible policymakers can cause a

‘cascading’ effect whereby they ‘cherry-pick’ (select) certain parts of the

report that they like, then exaggerate their import in subsequent speeches and

conversations. Regrettably, Tenet never objected publicly to the occasional

twisting of intelligence on Iraqi WMDs, with the end result that the American

people (along with their representatives in Congress) were misled into

thinking that their homeland might be in immediate jeopardy from attack by

unconventional weaponry.55

Another concern is the relationship between the NIO and the decision-

maker. On the one hand, if the NIO becomes too cozy with those in policy

positions, the danger of politicization rises as the analyst is tempted to bend

intelligence in support of policy objectives; on the other hand, if the NIO is

too detached, the NIE risks being irrelevant to the information needs of the

policymaker.

Further, the CIA has been criticized for taking over the drafting of Estimates

and, as a result, alienating other agencies and undermining the important goal

of the all-source intelligence fusion. Especially tricky has been the question of

how to represent dissenting views in an Estimate. To the extent possible, the

intelligence community attempts to resolve its disagreements before an NIE is

presented to top policymakers – although therein lies the danger of reports that

are diluted by a search for the lowest common denominator. The intelligence

agencies sometimes have quite different perspectives on a world situation.

Military intelligence agencies, for example, are notorious for a ‘worst-case’

approach to estimating. This is a result, critics contend, of pressures on

analysts applied by the Department of Defense and the military-industrial

complex to justify larger military budgets or new weapons systems by scaring

the American people and members of Congress with testimony about dire

threats from abroad.56 Conversely, military intelligence officials often seem to

consider the CIA and INR as naı̈ve and too ‘civilian’ to understand the true

nature of foreign military threats.57

A joke about the different cultures in the intelligence community arises

from how supposedly the intelligence agencies perceived Soviet intentions to

attack the United States during the Cold War, with the military weighing in

with the most frightening (‘worst case’) forecasts:

US Air Force: ‘The Russians are here!’

Defense Intelligence Agency: ‘The Russians are not here yet, but they

are coming.’

CIA: ‘The Russians are trying, but they won’t make it.’

Intelligence and Research (INR, in the Department of State): ‘The

Russians? They aren’t even trying.’58

THE PRESIDENT’S DAILY BRIEF AND THE NIE 349

The clash of differing views among intelligence agencies – known as

‘competitive intelligence’ – can be healthy, if driven by an objective analysis

of facts rather than policy bias. Debate among analysts can provide

policymakers with a wide range of views, instead of just the lowest-

common-denominator consensus characteristic of British intelligence reports.

Sometimes NIEs are guilty of homogenization, offering up views of a tapioca

consistency that rob policy officials of the nuances they need to understand.

The CIA has enjoyed a special advantage over the years in its capacity to

eschew policy pressures. It is an independent agency, outside the framework

of a policy department – the only US intelligence agency that can make this

claim – and, as a result, it has a chance of escaping immediate in-house

policy pressures from cabinet secretaries.

Sometimes agency dissents have been relegated to obscure footnotes, if

included at all in an NIE. This is unfortunate; failing to flag uncertainties or

disagreements is a cardinal error of estimating. More recently, however, NIC

chairs have been careful to make sure that analytic dissents are stated at some

length in the text of the NIE itself, not hidden in a footnote – if only to avoid the

resentment of dissenting agencies that may claim their findings have been

shunted aside. Dissenting agencies have insisted that their contrary opinions be

highlighted boldly in the text, often in a boxed form obvious to every reader.

An additional challenge has been to ensure that NIOs keep good liaison

relationships with consumers, and among themselves. ‘The difficulty lies not

only in predicting the future, in a world of many variables, incomplete data,

and intentional deception’, writes a seasoned intelligence officer, ‘but in

convincing policy makers that the prediction is valid.’59 Experience has

shown that unless a policymaker knows and feels comfortable with an NIO or

other intelligence briefer, he or she is less likely to pay much attention to the

proffered product.60 An NIO responsible for global environmental issues

during the Clinton Administration had never met the NSC staff person with

these same responsibilities, even though both had been in their positions for

over a year!61 This is no way to engender rapport, so necessary to the

intelligence process (although leading as well down the pathway to

politicization, unless the analyst is always on guard).

These personal relationships are critical. A top intelligence manager has

observed that when he guided the National Intelligence Council, he came to

believe that NIEs ‘were not our real product; rather, our real product was

National Intelligence Officers – not paper but people, experts, in a position to

attend meetings and offer judgments’.62 Of course, the judgments presented

by NIOs are rooted in the hard research and thought that they and lower

analysts have put into the NIEs as they are being drafted. As for maintaining

good contacts among NIOs themselves, Paul Wolfowitz, a member of the

Aspin-Brown Commission and later deputy secretary of defense in the second

350 INTELLIGENCE AND NATIONAL SECURITY

Bush Administration, has remarked that ‘the most important activity in which

[NIOs] can engage is internal debate over their judgments’.63 Yet sometimes

this internal debate never takes place.

Despite all of these hazards, NIEs do get written in an attempt to report on,

and often rank, possible outcomes in the world that might endanger the

United States. The idea is to assess trends (say, what is the ongoing situation

among extreme Islamists in Pakistan?), and then – to the extent possible for

mere mortals – to anticipate the course of history (the jihadists will storm the

government’s main buildings in Islamabad next Wednesday at 9:00 a.m.).

Sherman Kent has commented on the goal of in-depth analysis: ‘The guts of

the matter is the synthesizing of the pieces and setting them forth in some

meaningful pattern which everyone hopes is a close approximation of the

truth.’64 The idea is to help protect America’s security by making available to

Washington officials the best possible understanding of foreign capabilities,

leaders, and developing events.

In an NIE, the emphasis is placed on the most likely outcome; but some

Estimates also offer possible alternative scenarios. Still, the end result

remains something of a best guess resulting from consultation among the top

analysts in the intelligence community and others they may talk to outside the

intelligence establishment. As Kent once put it, ‘Estimating is what you do

when you do not know.’ In the process, he continues, one enters ‘into the

world of speculating’.65 However shrewd the forecasts may be in an NIE,

they still remain hunches – better than blind luck to be sure, but nonetheless a

far cry from certainty.66

Sometimes NIEs have been as accurate as a Swiss watch in predicting the

outcome of a world event or situation; on other occasions they have been

wide of the mark. Examples of successful predictions include: broadly, the

likely conduct of the Soviet Union in world affairs (the USSR would try to

expand, but not to incur the risk of general war67); the likely conduct of

Communist and Nationalist China; the Soviet Sputnik in 1957; the Sino-

Soviet split of 1962; the Chinese A-bomb test in 1964; the development of

new Soviet weapon systems throughout the Cold War;68 developments in the

Vietnam War (1966–75); the Arab–Israeli War of 1967; the India–Pakistan

War of 1971; the Turkish invasion of Cyprus in 1974; the Chinese invasion of

Vietnam in 1978; the mass exodus from Cuba in 1978; the Soviet invasion of

Afghanistan in 1979; the sharp deterioration of the Soviet economy just

before the end of the Cold War (1984–89); the investment strategies of the

Organization of Petroleum Exporting Countries (OPEC) consistently over the

years; and the rise and fall of various political leaders around the world,

including the breakup of Yugoslavia in the 1990s.

As a generalization, most of the intelligence community’s mistakes during

the Cold War were about what the Soviets intended, not what weapons

THE PRESIDENT’S DAILY BRIEF AND THE NIE 351

systems they had.69 The ability to track Soviet weaponry was vital during

the Cold War, and knowledge of Russian weaponry remains vital today.

America’s arms negotiations with the Russians (and others) depends on the

ability of the intelligence agencies to detect, through a range of sources and

methods, any significant violation of weapons agreements – a process known

as verification.

Examples on the debit side of analysis include: the failure to predict the

outbreak of the Korean War in 1950, or the placement of Soviet offensive

missiles in Cuba in 1962; the reporting of a (non-existent) bomber and missile

gap between the Soviet Union and the United States in the 1950s and early

1960s; underestimating during the Vietnam war the support flowing to the

Viet Cong through Cambodia; underestimating the pace of the Soviet strategic

weapons program; the failure to track where the Pakistani physicist A.Q. Kahn

had sent nuclear materials in the developing world; and the inability to

forecast the Soviet invasions into Hungary in 1956 and Czechoslovakia in

1968, the Arab–Israeli war in 1973, the fall of the Shah of Iran in 1979, or the

collapse of the Soviet empire in 1989–91. More recently, the most notorious

failures have been the inability to warn precisely when Al Qaeda would attack

the United States in 2001 (although there were several non-specific warnings

provided by the intelligence agencies from 1995 through the summer of 2001)

and the faulty prediction that Saddam Hussein possessed WMD.

One can only conclude that NIEs, just like PDBs, have been uneven in

their ability to provide the president and other officials with accurate

predictions about history’s probable trajectory. Especially difficult are the

long-range prognostications found in NIEs, since the skill of human beings in

forecasting diminishes with the distance one attempts to peer into the future.

‘The CIA Directorate of Science and Technology has not yet developed a

crystal ball’, a Senate intelligence overseer once observed. He continued:

Predicting the future must remain probabilistic. Though the CIA did

give an exact warning of the date when Turkey would invade Cyprus [in

1974], such precision will be rare. Simply too many unpredictable

factors enter into most situations. The intrinsic element of caprice in the

affairs of men and nations is the hair shirt of the intelligence estimator.70

Reflecting back on his experiences with intelligence while serving as

secretary of state for Presidents John F. Kennedy and Lyndon B. Johnson,

Dean Rusk agreed with this viewpoint. In his words: ‘Providence has not

provided human beings with the capacity to pierce the fog of the future.’

Rusk added that NIEs ought to begin with the statement, ‘‘‘Damned if I

know, but if you want our best guess, well, here it is’’, to alert the policy

officer that there cannot be certainty in the matters being discussed.’71

352 INTELLIGENCE AND NATIONAL SECURITY

Intelligence scholar Richard Betts has emphasized this point in his writings.

When it comes to intelligence predictions, he notes, ‘some incidence of

failure [is] inevitable’. Consequently, he urges a higher ‘tolerance for

disaster’.72

It is plain enough why policymakers need to have a realistic understanding

about the limited capacities of intelligence. As a leading former CIA analyst

has stated: ‘there is seldom, if ever, enough intelligence present to make

absolute predictions or warnings’.73 Information is usually scarce or

ambiguous, and the situation in question may be fluid and changing.

Intelligence scholar Arthur S. Hulnick advises: ‘Policy makers may have to

accept the fact that all intelligence estimators can really hope to do is to give

them guidelines or scenarios to support policy discussion, and not the

predictions they so badly want and expect from intelligence.’74

This realistic sense of intelligence limitations is unhappy news for

presidents and cabinet secretaries who seek clear-cut answers, not hunches

and hypotheses; but such is the existential reality of intelligence. Yet it bears

repeating that having intelligence agencies collect information worldwide

and try their best to make sense of the findings is far better than operating in

an information vacuum, just as one is ill-advised to cross a busy street

blindfolded. As analyst Harold Ford has correctly observed, ‘There is no

substitute for the depth, imaginativeness, and ‘‘feel’’ that experienced, first-

rate analysts and estimators can bring to the often semi-unknowable questions

handed them.’75

Moreover, there is the possibility that intelligence estimating will further

improve in the coming years, depending on whether the United States is able

to advance its technical and human abilities to improve its gathering of

information overseas, as well as its ability to sift through the information

quickly and effectively, separating important ‘signals’ from all the other

‘noise’. It depends, too, on whether Americans are able to expand their

knowledge of foreign languages, cultures, and histories; and whether the

government can attract the best young minds in the nation to take up the

challenge of preparing NIEs for Washington decision-makers. Vital, in

addition, will be a renewed commitment among intelligence professionals to

resist political pressures from policymakers to twist intelligence in a manner

that suits policy preferences at the expense of the truth – the soul-destroying

politicization of intelligence.

Even if NIEs are less than perfect instruments for predicting future events,

they at least have the virtue of marshaling together in one place a reliable set

of facts about a situation abroad of interest to the United States. This frees up

the nation’s decision-makers to focus attention on sorting out disagreements

they may have over which policy options to choose. As William Odom has

written, ‘The estimate process has the healthy effect of making analysts

THE PRESIDENT’S DAILY BRIEF AND THE NIE 353

communicate and share evidence. If the NIEs performed no other service,

they would still be entirely worth the effort.’76 Almost 40 years ago, Kent

noted, too, that ‘the intelligence estimate will have made its contribution in

the way it promoted a more thorough and enlightened debate’.77

CURRENT VERSUS RESEARCH INTELLIGENCE

An important discussion in the world of intelligence surrounds the question of

how many resources should be sunk into the quick production of current

intelligence, which can be highly perishable, at the expense of preparing

more deeply considered strategic products of research intelligence.78 What is

the proper balance between short reports on world affairs, like the PDB, and

lengthier treatments, like the NIE?

The starting point in this debate should be the realization that almost all

policymakers prefer to receive current intelligence reports over Estimates.

Indeed, consumers (policymakers) recently rated NIEs eighth among products

forwarded to them by the intelligence community.79 Mark M. Lowenthal

writes that in the past several years the intelligence community has ‘put its

greatest emphasis on shorter, more current products’, a response to ‘a fairly

consistent decline in policymaker interest in intelligence community products

as they get longer and more removed from more current issues’.80 Former CIA

analyst Harold Ford observed a similar phenomenon a generation earlier in the

early 1990s: ‘The great majority of policymakers have to concern themselves

with fairly immediate, pressing problems. More distant and more uncertain

[issues] have a lesser constituency and fewer advocates.’81 Or as Richard K.

Betts succinctly states: ‘Immediate problems drive out distant ones.’82

Among the hottest items of current intelligence desired by policymakers

are foreign leadership profiles – facts and insights (sometimes salacious)

about the public and private lives of the men and women whom US

policymakers will be talking with at conferences, summits, and other

international meetings. Without these profiles, provided in reports the size of

baseball cards for easy reading, second-echelon political officials in other

lands would remain strangers to US negotiators. In one recent year, almost

16,000 of these profiles made their way from the CIA to policy offices around

the District of Columbia.83

The advent of cable news and the Internet have affected the status of NIEs.

‘Long-term research and in-depth analysis suffered as CIA managers and

analysts became fixated on the race to get late-breaking tidbits of intelligence

into the President’s Daily Brief’, notes New York Times reporter James

Risen.84 Added to this PDB advantage over Estimates, NIEs have ‘on

occasion been wrong, or in more cases late, or in still more cases, too cloudy

to be of much use’.85

354 INTELLIGENCE AND NATIONAL SECURITY

The upshot is that about 80–90% of the analytic resources of the

intelligence community are presently dedicated to clarifying for policy-

makers what happened today and yesterday, and what is likely to happen

tomorrow.86 Estimates have receded on the agenda of priorities for the US

intelligence agencies. Former senior intelligence official Mark Lowenthal

notes that the intelligence community has ‘gotten out of the knowledge-

building business. Now it is: current, current, current’.87 Another leading CIA

analyst ruefully agrees: ‘Life in the Directorate of Intelligence [DI, the home

of the CIA’s analysts] is no longer contemplative.’88 Happily so, critics of

National Intelligence Estimates would add. They point to a series of

shortcomings in the Estimates process. First, NIEs are too long. ‘If the

intelligence product is not two pages or less’, an assistant secretary of defense

emphasized to the Aspin-Brown Commission, ‘it is unlikely to be read. I have

only about five minutes [a day] that I can devote to reading intelligence.’89

One recent NIC chair, Richard N. Cooper, decided to make the rounds with

policy officials in 1994 to see what they thought of recent NIEs they had

received. They looked at him blankly; they simply had not read any of them.

At the same time, though, they expressed great enthusiasm for the NIOs they

knew and appreciated the occasional oral briefings from them, which were

often based on fresh NIEs.90

Tongue in cheek, but nevertheless stressing a serious point, an experienced

former intelligence officer notes that ‘some policymakers don’t read, some

won’t read, and some can’t read’.91 A former DCI conceded in 1994 that

‘policymakers basically don’t care about National Estimates’.92 In a nutshell,

NIEs may continue to be written, but they may end up just sitting on a shelf

somewhere, untouched by decision-makers.

Others, though, reject this thesis. They point, for instance, to busy policy

officials who have also been avid readers of Estimates, such as secretaries of

defense Les Aspin, James R. Schlesinger, Harold Brown, and Robert M.

Gates (himself a former DCI) – all of whom devoured NIEs as if they were

reading pulp fiction. Moreover, even if government principals never get

around to reading an entire NIE (perhaps only the Key Judgments section),

their aides – the deputy assistant secretaries (DASs) of Washington – usually

will. The information they absorb is then recycled to their seniors in oral

briefs and conversations, as well as in memoranda and reports they write to

their bosses. And even if no policy official reads the NIEs, they would still

serve a purpose: the process educates NIOs and other analysts, who then

become better equipped to provide oral briefs to key officials.93 Further, as

Richard Cooper states, ‘NIEs focus the intelligence agencies; they allow

quality control for NIOs over other analysts throughout the community.

Moreover, they provide valuable grist for the mill at the level of the deputy

assistant secretaries and office directors.’94 Finally, as one cabinet aide

THE PRESIDENT’S DAILY BRIEF AND THE NIE 355

reminds us: ‘No one reads an encyclopedia from A to Z, but it is still helpful

to have encyclopedias.’95

An additional criticism of NIEs is that they are too democratic:

‘estimating by plebiscite’, scoffs Lowenthal. ‘Since when is the FBI view

on Dafur just as good as INR’s?’ he asks, yet both are able to weigh in with

their views in an Estimate on this subject.96 Or when it comes to WMD in

Iraq, one could pose the question: why should the DIA’s view be

showcased as prominently as the judgment of intelligence officers in the

Energy Department, whose primary responsibility is to monitor nuclear fuel

proliferation around the world (and who, in 2002, dissented against the

prediction that Iraq possessed WMD)? Critics suggest, too, that NIEs are

intellectual buffets, so broadly worded that one can find whatever one likes

within them. They are, Lowenthal concludes, ‘intellectual and moral dead-

ends’.97 Former DCI Turner also worried about the ‘limited’ value of NIEs

that require ‘so many compromises’.98

NIEs raise concerns, too, about the extent to which they have become

political footballs from time to time. Since 9/11, some lawmakers and

officials in the executive branch seem to go after the blood of any NIO with

whom they disagree. The unfairness of this is patent, given the difficulty of

predicting events. A senior officer gives this illustration of how politically

dicey the art of estimating can be: ‘I was supposed to tell a lawmaker, who

didn’t know if he was going to be re-elected or not, who is going to win the

Israeli elections!’99 Another, more famous example comes from the Soviet

invasion of Czechoslovakia in 1968. Throughout the summer of that year, the

Politburo in Moscow remained deeply divided over whether to use force to

tame a rebellion in Prague. Historians now know that the Soviets fluctuated

back and forth on what to do, all the way until the eleventh hour when the

decision was finally made on 20 August to launch an invasion. The CIA had

reported on the fact that Soviet troops had been mobilized, but the Agency

was unable to say whether an invasion would actually occur. Not even the

Politburo knew until the last moment. So precise predictions remain an

elusive ideal; nonetheless, the intelligence agencies can at least alert the

White House to a developing situation, as the CIA did in this instance.

THE FREQUENCY OF NIEs

From the first DCI in 1946 until the last in 2005 – a 60-year period – the

intelligence community produced 1307 National Intelligence Estimates,

averaging 23 a year. It is worth examining the data on the frequency of these

NIEs during the administration of each Director of Central Intelligence.100

Have Estimates, in fact, gone into decline, pushed aside by the PDB and other

forms of current intelligence? The trend line in Figure 1 displays a more

356 INTELLIGENCE AND NATIONAL SECURITY

complex reality, with the numbers of NIEs fluctuating over the years and

throughout the tenures of DCIs.101

The frequency of NIE production in any given year will be a reflection of

the priorities of a DCI or DNI, the interest an administration has in receiving

Estimates, and changing world circumstances that may require the

preparation of new NIEs. In times of war, for example, policymakers are

likely to be especially focused on current intelligence for battlefield

exigencies, with NIEs pushed to a back burner.102

While a pale imitation of a community-wide NIE was published as early as

1946, it was written by just a single author inside the CIA and is not counted

here.103 The production of true, community-wide NIEs began in 1950, under

the supervision of DCI Rear Admiral Roscoe H. Hillenkoetter – nine

Estimates in total for that year. General Walter Bedell Smith, one of the most

renowned of the US intelligence chiefs because of his outstanding

management skills, replaced Admiral Hillenkoetter in October 1950. Under

Smith, the number of NIEs leapt upward, averaging 28 for 1951 and 1952.

This increase reflected his determination to build up the nation’s neophyte

intelligence capabilities. Among Smith’s first decisions as DCI was to set up

an Office of National Estimates, whose sole job was to prepare Estimates. He

made it clear that the ONE would report directly to him, signaling that

estimating would be the DCI’s special providence.104

With the advent of Allen W. Dulles’s term as DCI in 1953, the production

of NIEs rose further, to a new peak of 45 in 1955 – a figure that would remain

unsurpassed until 1992 and is still the second highest number. In his first two

FIGURE 1

NUMBER OF NATIONAL INTELLIGENCE ESTIMATES BY YEAR, 1950–2005

DATA SOURCE: CENTRAL INTELLIGENCE AGENCY, 2006. THE AVERAGE NUMBER OF NIEs PER

YEAR IS TWENTY-THREE

THE PRESIDENT’S DAILY BRIEF AND THE NIE 357

years, Dulles maintained NIE production at the same average set by Smith:

28. The levels soon shot up, though, throughout the rest of his tenure (1955–

61), averaging 34 a year.

This climb is remarkable, given Dulles’s reportedly limited interest in

analysis.105 As one of his successors, Richard Helms, recalls: ‘Dulles tended

to take the [CIA’s Directorate of Intelligence] for granted. He appreciated the

value and quality of the product, but rarely addressed himself to the DI or any

of its problems.’106 Journalist Tim Weiner recounts how the intelligence

director appeared more interested in the latest baseball scores than the latest

analyses of world affairs.107 Dulles was devoted to the operations side of

intelligence: recruiting agents, planting propaganda, secretly trying to change

the course of history. He had enjoyed a life of derring-do in the Office of

Strategic Services during World War II, running spies against the Nazis.

Nevertheless, his attention to NIE production was sufficient to continue their

upward trend; indeed, the Dulles years saw more Estimates produced – 306 –

than during any other DCI’s tenure, by far (see Table 1).108

John A. McCone followed Dulles as DCI. Like Bedell Smith, McCone was

popular inside the intelligence community and similarly admired for his

managerial skills. His dedication to the production of NIEs matched Smith’s,

at an average of 28, though fell short of Dulles’s overall average of 34. It is

claimed that McCone gave ‘new energy to the Estimates’,109 but in fact his

tenure displayed an initial decline in NIE production. The figure of 25

Estimates for 1962 was the lowest number since the first year of estimating in

1950. In 1963 and 1964, however, McCone’s number of Estimates rose to 29

in 1963 and to 30 in 1964. Perhaps the infamous NIE of 1962, mistakenly

predicting that the Soviet Union would not introduce nuclear missiles into

Cuba, contributed to McCone’s closer attention to the Estimates process.110

(McCone himself, recall, had personally cautioned policymakers that the NIE

of 1962 was likely wrong, and that the Soviets could be expected to support

their Latin American satellite with missile emplacements.)

This upward trend continued to 36 Estimates in 1965 under McCone’s

successor, Vice Admiral William F. Raborn, Jr. – two higher than Dulles’

average.111 The next DCI, Richard Helms, boosted this level further, to 38

NIEs in his first year in office (1966); but then the number of NIEs went into

their sharpest decline to that date, as Helms averaged 31 Estimates a year for

his tenure as DCI. This includes a decline to 24 in 1970 and 27 in each of the

years 1971 and 1972.

In part, this downward trend reflected Helms’ greater interest in current

intelligence. The old newspaper man in him (his first career had been in

journalism) conceded that: ‘the pressure of meeting urgent deadlines and

responding immediately to developing crises always sparked a pleasant

mnemonic echo’.112 Moreover, President Lyndon B. Johnson, who failed to

358 INTELLIGENCE AND NATIONAL SECURITY

mention the subject of NIEs even once in his autobiography, The Vantage

Point, had little interest in lengthy intelligence reports. Chester Cooper, a

CIA liaison to the White House during the Johnson years, recalls that at the

time ‘in-depth analyses were far from best sellers’.113 Johnson was, though,

an avid reader of current intelligence related to the war in Vietnam – until the

rising crescendo of negative reports became too much for the president to

bear.114

During the tenure of Helms’s successors, James R. Schlesinger (who served

only five months as DCI in 1973, although longer than any of the other DCIs

that year) and then William E. Colby, the numbers continued to fall. The

record shows 19 NIEs in 1973, nine in 1974, and ten in 1975. Colby served at a

time of great crisis for the intelligence community, during the ‘intelligence

wars’ of 1974–76 when the secret agencies were under investigation for

domestic spying.115 This DCI had matters to attend to other than NIE

production: the survival of the CIA itself seemed to be on the line.116 Colby’s

TABLE 1

FREQUENCY OF NATIONAL INTELLIGENCE ESTIMATES BY DCI

ADMINISTRATION, 1946–2005

DCI Period of Tenure* Number of NIEs Average Yearly NIE Per DCI

Souers 1946 0 0.0 lowest Vanderberg 1946–1947 0 0.0 lowest Hillenkoetter 1947–1950 50 3.0 very low Smith 1951–1952 55 27.5 high Dulles 1953–1961 306 34.0 very high McCone 1962–1964 84 28.0 high Raborn 1965 36 36.0 very high Helms 1966–1972 216 31.0 high Schlesinger 1973 19 19.0 low Colby 1974–75 19 9.5 very low Bush 1976 5 5.0 very low Turner 1977–80 35 9.0 very low Casey 1981–1986 111 18.5 low Webster 1987–1991 78 15.5 low Gates 1992 56 56.0 highest Woolsey 1993–1994 76 38.0 very high Deutch 1995–1996 29 14.5 low Tenet 1995–2004 160 20.0 moderate Goss** 2005 13 13.0 low

*These years are not the precise full dates of service for each Director of Central Intelligence, but rather the years in which a DCI served a preponderant number of days in a particular year and was thus accorded the year’s full total of NIEs. Dividing the NIEs into fractions according to the number of days a DCI served each year (not shown here) did not alter the basic analysis presented here and seemed even more arbitrary.

**DCI Goss is given full credit for the thirteen NIEs of 2005. The new Director of National Intelligence came into office in April of that year, but took several months to settle in; therefore, the DNI was unlikely to be much involved in the Goss ‘‘spill over’’ Estimates for 2005.

THE PRESIDENT’S DAILY BRIEF AND THE NIE 359

concentration on placing the intelligence agencies back on an even keel after

the scandals preoccupied his successor as well: DCI (and later President)

George H.W. Bush, whose production of NIEs set a record low at five in 1976.

Next came Admiral Stansfield Turner of the Carter Administration (1977–80)

who reversed the estimating decline, but still recorded a relatively low level of

production: nine, seven, seven, and twelve in each of his four years as DCI, for

an average of nine – even less than the besieged Colby.

When the Reagan Administration came to power in 1981, NIE production

dipped for a year and then took on new life. Under DCI William J. Casey, the

numbers jumped from nine in his first year to 28 in 1984. They dipped again

in the next two years and plummeted as the administration found itself

wrapped around the axle of the Iran-Contra affair. The average during

Casey’s tenure of six years was 18.5 NIEs, below the overall annual average

of 23 during the full span from 1950 to 2005. William H. Webster replaced

the deceased DCI Casey in 1987, just as the Iran-Contra scandal was really

heating up. While the affair was under investigation, NIEs dropped to seven

in 1987 and 15 in 1988. Clearly, DCIs find it difficult to focus on research

analysis while embroiled in intelligence scandals. Webster slowly turned

back to the job of more energetic estimating after the scandal cleared,

recording 16 NIEs in 1989, 12 in 1990, and finally ending his tenure with 28

1991 – his only year above the overall average.

When George H.W. Bush won the presidency in 1990, he soon replaced

Webster with the first (and only) professional intelligence analyst ever to fill

the position of DCI: Robert M. Gates, formerly a Deputy Director for

Intelligence in the CIA. The brief Golden Age of NIEs was about to begin. In

1992, Estimates production under Gates hit a still-standing record of 56. If

one wants to emphasize analysis, drawing the nation’s intelligence chief from

the ranks of analysts appears to be a good approach.

Gates’ successor, R. James Woolsey (the first of three DCIs for President

Bill Clinton), kept the NIE production levels high, registering 42 in 1993 (the

third-highest rate ever, after Gates in 1992 and Dulles in 1955), followed by a

strong 34 in 1994. When DCI John M. Deutch came into the DCI office in

1995, the numbers went into decline again: to 15 in 1995 and ten in 1996.

Deutch’s deputy, George J. Tenet, took over the DCI reins in 1997 and

held them for seven years (a period of service second only to Dulles).

Throughout this period, the number of NIEs again fell, congruent with the

thesis that current intelligence in recent years has been driving out research

intelligence. The average number of annual NIEs during the Tenet years was

20 – three below the overall annual average since 1950. Perhaps most

significantly, 1997 was the last time the community wrote an Estimate on

terrorism all the way up to the 9/11 attacks (although there were some other

analytic reports with warnings about Al Qaeda’s activities).117

360 INTELLIGENCE AND NATIONAL SECURITY

Porter Goss, a Republican former member of Congress from Florida (the

first and only lawmaker to head the intelligence community), assumed the

office of the DCI in July of 2004. By April of 2005, this office had been

abolished and replaced by the DNI office, headed by former ambassador John

D. Negroponte. The production of Estimates during this time of transition

(amidst considerable confusion over who would lead the NIC and the NIOs)

remained low: 11 in 2004 and 13 in 2005 – even though Goss vowed upon

heading up the CIA that he was going to concentrate on long-range strategic

intelligence rather than producing spot news.118 The top intelligence official

in the Department of State has noted recently that ‘we haven’t done strategic

intelligence [NIEs] for so long that most of our analysts don’t know how to

do it anymore’.119

It appears as though the intelligence scandals drove attention away from

NIE production during the middle years of this history (Colby and Bush;

Casey and Webster). So have the presence of DCIs with limited interest in

research intelligence (Turner, Deutch, Goss); periods of transition and

institutional turbulence in the intelligence community (Hillenkoetter,

Negroponte); and the growing policy focus on current intelligence in recent

years (Tenet, Goss, Negroponte). The danger of global terrorism, underscored

for the United States by the attacks against its embassies in Kenya and

Tanzania in 1998, against the U.S. Cole in the Gulf of Aden, and the 9/11

attacks in 2001, as well as by a steady drumbeat of threats from Al Qaeda

thereafter, has contributed substantially to the hunger for immediate

intelligence that might warn of new attacks. The war in Iraq has further

riveted the attention of policymakers on current intelligence.

One thing is certain: the NIE process is now in one of its periodic troughs.

This is a result not so much of scandal (although the intelligence agencies

have been under seemingly constant investigation since 9/11 about various

analytic inadequacies and the use of warrantless wiretaps) but rather a new,

terrorist-driven desire for current information about Al Qaeda, the Taliban,

Iraqi insurgents, and related perils facing the United States. The irony is

that Washington desperately needs at this time a deeper understanding

of places like Iraq, Afghanistan, and Pakistan, and a more thorough

knowledge of Islamic extremists and their motivations – precisely the stuff of

NIEs.

The argument, though, between advocates of current or research

intelligence, or between PDBs and NIEs, poses a false dichotomy. In reality,

the nation needs both. Clearly, the conduct of warfare in Iraq and Afghanistan

begs for a rich flow of tactical, current intelligence to guide soldiers and their

commanders, including the commander-in-chief. Other topics, though, beg

for a more steeped analyses, such as on the future of the Saudi royal family.

Policymakers and their aides, usually starved for information that will help

THE PRESIDENT’S DAILY BRIEF AND THE NIE 361

them deal with problems in their in-box, will continue to want both PDBs and

NIEs.

The frequency of Estimates will no doubt ebb and flow in the future. The

present ebbing is a result, in part, of NIOs finding themselves swamped by

pressure to provide oral briefings to policymakers on current issues. Still, the

intelligence community steadily produces many other forms of research

intelligence, whether specialized reports, memoranda, or oral briefings.

According to former CIA senior analyst John L. Helgerson, ‘a bunch of

research intelligence is done, not necessarily estimative – just everything we

know about subject X; then someone says, ‘‘It’s about time we do a formal

Estimate’’’.120

The PDB and the NIE, then, are just two among thousands of other reports

written by the intelligence community, both current and research in nature.

Among them are Intelligence Memorandum (IM), pithy analyses sent to

specific policymakers; intelligence dailies; weekly reports; information

cables; working group reports; various research papers less detailed than

NIEs (in the ‘middle range’ of analysis, between current intelligence and

Estimates); Special Intelligence Reports and Intelligence Reports, of which

there are thousands each year focused on information requests from

individuals or smaller groups of policymakers; Net Assessments, which

examine foreign in comparison with US military capabilities; ‘baseball card’

foreign leadership profiles, which rose in number to account for some 60% of

all CIA Intelligence Directorate products in a recent year;121 personal letters

from the DCI or DNI to the president and other officials – a favorite of

George Tenet;122 assessment sections of covert action findings; and daily or

on-call oral briefings. (In 2005, Lowenthal referred to oral briefings as

‘probably the most important analytical service provided by the intelligence

community’, because of the interaction it permits between consumers and

producers.123) There are many more analysts in the intelligence community

than during the Cold War, particularly after the hiring rush that followed 9/11

and the erroneous WMD predictions, and that has led to an increase in the

production volume of both current and research intelligence.

CONCLUSION

Although this essay has focused on the two premier intelligence products in

the United States, the President’s Intelligence Brief and the National

Intelligence Estimate, it is important to remember that there are many other

reports that flow from the producers to the consumers of intelligence. To be

useful to policymakers, each of these products – whether the exalted PDBs

and NIEs or some more obscure report – is supposed to exhibit high

professional standards in the form of solid facts and thoughtful analysis.

362 INTELLIGENCE AND NATIONAL SECURITY

The key attributes of intelligence reporting include: accuracy (CIA stands for

‘Can’t Find Anything’, quipped comedian Jay Leno when in the 1990s the

intelligence community sent a NATO bomber pilot coordinates for an arms

depot in Serbia that turned out to be the Chinese Embassy); timeliness

(history runs on nimble feet); relevance (for the White House, no

dissertations on local elections in Monrovia, thank you); readability (no

econometrics); conciseness (presidents are busy); all-source fusion (tapping

into the knowledge of all 16 agencies); and a degree of specificity that can

allow policymakers the ability to take action in the nation’s defense (vague

‘amber alerts’, out; precise Al Qaeda attack plans, in).

This list sums to a high standard. Often the intelligence community will

fall short, for reasons of inadequate intelligence collection; the slow

translation of foreign-language and coded materials, along with other data

processing problems; flawed analyses, since the brains of human beings are

imperfect; or the misuse of reports by policymakers, from their disregard of

disagreeable information to cherry-picking and politicization. Despite these

challenges, the intelligence community must continue to reach for as near

perfect intelligence reporting as possible; and policymakers must constantly

resist the temptation to twist these reports for political purposes. To the extent

that both succeed in achieving these goals, intelligence reports across the

board will rise in stature and will be closely read by the nation’s leaders.

Regarding the value of PDBs and NIEs, the verdict is clear: they

contribute. Improvements are necessary, though.124 For example, PDBs

should be more nuanced and free of exaggeration. Further, NIE production

levels should rise; and the document should be shorter in length, say, 30

rather than 100 pages long (although the text could provide references to

computer links for additional information sought by the more avid reader).

Estimates should also be completed in six months at the most, or faster in

times of emergency. Those D/CIAs and DNIs who allow the number of

Estimates to fall below the annual historical average of 23 are not paying

adequate attention to a core intelligence mission. It would be sensible, too,

for the DNI to move back to CIA Headquarters, where the NIC and the

Directorate of Intelligence are located – the primary reservoirs of analysis in

the federal government. To avoid the impression that the CIA is the only

show in town, however, the DNI needs to bring along to CIA Headquarters a

good number of analysts and managers for the NIC (including NIOs) whose

primary base is located in one of the other intelligence agencies; NIEs must

be truly a community-wide product and this kind of amalgamation at DNI/

CIA Headquarters would make that clear.

Will PDBs and NIEs have a readership among key officials in Washington,

DC – the sine qua non of intelligence relevance? If presidents read them, so

will others. The bottom line: elect as America’s chief executive men or

THE PRESIDENT’S DAILY BRIEF AND THE NIE 363

women who are likely to take the nation’s top intelligence findings seriously,

and who will read them (and pick cabinet officers who will read them) as

though the quality of their decisions depended on this practice – as it

well may.

NOTES

The author would like to express his deep appreciation to Leena S. Johnson for her editorial assistance and to Kristin E. Johnson for her help with Figure 1 in the text.

1 This figure has been widely cited in the media, for instance: Mark Mazzetti, ‘Spymaster Tells Secret Size of Spy Force’, New York Times, 21 April 2006, p.A18.

2 A former governor of New Jersey, Republican Thomas H. Kane, led the Kean Commission, known more formally as The National Commission on Terrorist Attacks upon the United States. The title of its final report is The 9/11 Commission Report (New York: Norton 2004). For a history of the Commission, see Philip Shenon, The Commission (New York: Twelve 2008).

3 Quoted by Linton Weeks, ‘An Indelible Day’, Washington Post, 16 June 2004, p.C1. 4 See Loch K. Johnson, ‘The Aspin-Brown Intelligence Inquiry: Behind the Closed Doors of a Blue Ribbon Commission’, Studies in Intelligence 48 (Winter 2004) pp.1–20.

5 See the 9/11 Commission Report, p.261. 6 George Tenet, with Bill Harlow, At the Center of the Storm: My Years at the CIA (New York: HarperCollins 2007) p.30. In a court deposition, the CIA argued recently that the PDB ‘is the quintessential pre-decisional, deliberative document’ (cited in Steven Aftergood, Secrecy News, Federation of American Scientists Project on Government Secrecy, 30 November 2007, p.2).

7 For examples from the Johnson Administration, as well as the Brief on Bin Laden acquired by the Kean Commission, see Loch K. Johnson (ed.) Strategic Intelligence, Vol. 2: The Intelligence Cycle (Westport, CT: Praeger 2007) pp.296–307.

8 Author’s interview, 19 November 1984, Washington, DC, quoted in Loch K. Johnson, America’s Secret Power: The CIA in a Democratic Society (New York: Oxford University Press 1989) p.90. For a look at the PDB process as it operated in 1970, see Andrew W. Marshall, Memorandum to Henry A. Kissinger, 18 March 1970, Nixon Presidential Library, 13 MB PDF file, cited in Aftergood, Secrecy News.

9 These figures come from the author’s interviews with CIA personnel involved with the PDB production in each of these administrations; for the description of the Bush PDB, see Tenet, At the Center of the Storm, p.31.

10 Lawrence Wright, ‘The Spymaster’, The New Yorker, 21 January 2008, pp.47–8. 11 Author’s notes on remarks by DNI McConnell, Conference on ‘Intelligence Strategies: New

Challenges and Opportunities’, National Defense Intelligence College, Washington, DC, 26 September 2007.

12 Author’s interviews with Clinton Administration officials throughout 1992–97, Washington, DC.

13 Johnson, America’s Secret Power, p.90. 14 Quoted by Richard L. Russell, ‘Low-Pressure System’, The American Interest 2 (Summer

2007) p.123. On the practice of providing the PDB and oral briefings to presidential candidates, see John L. Helgerson, Getting to Know the President: CIA Briefings of Presidential Candidates, 1952–1992 (Washington, DC: Center for the Study of Intelligence, Central Intelligence Agency, no date, but circulated in the late 1990s).

15 These observations are based on the author’s interviews with intelligence officers during 1998, Washington, DC.

16 The other sources consulted by the Aspin-Brown Commission were the Institute for Scientific Information Research Services Group, Oxford Analytica, East View Publications, Inc., and Lexus-Nexus. This is not to say that these groups would not have performed well

364 INTELLIGENCE AND NATIONAL SECURITY

on other occasions; but at least in this instance, in 1995 over a one-day period, they did not compete well with the PDB (and the Institute missed the deadline by four days).

17 See Johnson, America’s Secret Power, ch. 9. 18 George P. Shultz, Turmoil and Triumph: My Years as Secretary of State (New York:

Scribner’s 1993) p.864. 19 Barton Gellman and Walter Pincus, ‘Depiction of Threat Outgrew Supporting Evidence’,

Washington Post, 10 August 2003, p.A1, quoting a senior intelligence official. 20 See, for instance, ibid.; and ABC Evening News, 7 October 2002. 21 Tenet, At the Center of the Storm, pp.321–2. 22 Ibid., pp.322–3. 23 See Senator Bob Graham, with Jeff Nussbaum, Intelligence Matters: The CIA, the FBI,

Saudi Arabia, and the Failure of America’s War on Terror (New York: Random House 2004) p.180.

24 The NIE on ‘The Terrorist Threat to the United States Homeland’, released into the public domain by the National Intelligence Director in July of 2007, reportedly took three years to prepare; Wright, ‘The Spymaster’, p.55.

25 Roberts Report, in Johnson, Strategic Intelligence, Vol.2, p.320. 26 Tim Weiner, Legacy of Ashes: The History of the CIA (New York: Doubleday 2007) p.487. 27 The NIE was entitled Iraq’s Continuing Program for Weapons of Mass Destruction (NIE

2002–16HC). On the timing, see Senator Graham, Intelligence Matters, pp.179–80. 28 Graham, Intelligence Matters, p.187. 29 Ibid., pp.185–9. 30 Interviewed by Wil S. Hylton, ‘The Angry One’, Gentleman’s Quarterly (January 2007). 31 Dana Priest, ‘Congressional Oversight of Intelligence Criticized’,Washington Post, 27 April

2004, p.A1, reported that (outside of members of the Senate and House Intelligence Committees) ‘no more than six senators and a handful of House members’ read more than the five-page ‘Key Judgments’ of the full NIE; see also Elizabeth Kolbert, ‘The Lady Vanishes’, New Yorker, 11 and 18 June 2007, p.133, who reports on research indicating that not even presidential candidate Senator Hillary Clinton, Democrat of New York, had bothered to read the classified Estimate. Senator Clinton refutes the charge that she was uninformed about the NIE, arguing that she had been briefed orally on the Estimate by several members of the Administration. According to one report, not even the president read the 2002 NIE (Michael Isikoff and David Corn, Hubris: The Inside Story of Spin, Scandal, and the Selling of the Iraq War (New York: Crown 2006) p.116, citing White House sources).

32 Report on the U.S. Intelligence Community’s Prewar Intelligence Assessments on Iraq, Senate Select Committee on Intelligence (Washington, DC: US Government Printing Office, 7 July 2004) p.14.

33 Tenet, At the Center of the Storm, p.338. 34 Steven Lee Myers, ‘An Assessment Jars a Foreign Policy Debate’, New York Times, 4

December 2007, p.A1. The Times printed the NIE’s unclassified ‘Key Judgments’ (p.A10). 35 David E. Sanger and Steven Lee Myers, ‘U.S. Says Details in Military Notes Led to Iran

Shift’, New York Times, 6 December 2007, p.A12. 36 See Glenn Hastedt, ‘Washington Politics, Intelligence, and the Struggle against Global

Terrorism’, in Loch K. Johnson (ed.) Strategic Intelligence, Vol. 4: Counterintelligence and Counterterrorism (Westport, CT: Praeger 2007) pp.99–126.

37 Lyman B. Kirkpatrick, ‘United States Intelligence’, Military Review 41 (May 1961) pp.18– 22, quote at p.20 (based on a speech presented in Detroit, Michigan, in November 1960), cited by Harry Howe Ransom, The Intelligence Establishment (Cambridge, MA: Harvard University Press 1970) p.147. For examples of NIEs that have been declassified, see Johnson, Strategic Intelligence, Vol.2, pp.241–95.

38 Letter from Sherman Kent of Allen Dulles (dated 3 May 1963), Sterling Library Collection, Yale University, Series 1, Box 18, Folder 390, accessed by the author in 2005.

39 Harold P. Ford, Estimative Intelligence, Intelligence Profession Series, Number Ten (Washington, DC: The Association of Former Intelligence Officers 1993); see also his Estimative Intelligence: The Purposes and Problems of National Intelligence Estimating

THE PRESIDENT’S DAILY BRIEF AND THE NIE 365

(New York: University Press of America 1993). Declassified portions of an NIE on Iran released in 2007 refer to NIEs as ‘the Intelligence Community’s (IC) most authoritative written judgements on national security issues and designed to help US civilian and military leaders develop policies to protect U.S. national security interests’ (National Intelligence Council, National Intelligence Estimate, Iran: Nuclear Intentions and Capabilities (November 2007), CNN News, 3 December 2007).

40 Ransom, The Intelligence Establishment, p.147. For an example of the occasionally strong influence an NIE can have on policy, see former DCI Stansfield Turner’s discussion of how an Estimate in 1954, which concluded that the US use of an A-bomb against Vietnam that year would ignite World War III, contributed to President Dwight David Eisenhower’s decision against the bombing (Stansfield Turner, Burn Before Reading: Presidents, CIA Directors, and Secret Intelligence (New York: Hyperion 2005) p.79). Robert Gates, another former DCI, believes that only ‘one or two Estimates really made a major impact on policy’ between the years 1968 and 1983. His favorite example is an NIE from 1978 on Soviet global intentions. Warnings in the Estimate about a vigorous Soviet ideological offensive in the Third World convinced the Carter Administration, Gates argues, to counter this initiative with a more aggressive program of its own in the developing nations (author’s interview with Gates, Washington, DC, 28 March 1994). Michael Herman is correct, though, in his observation that it is as unrealistic to think that NIEs will lead to immediate policy actions as it is to think that ‘daily newspapers expect to change the world with every issue’ (Intelligence Power in Peace and War (Cambridge: Cambridge University Press 1999) p.143).

41 Author’s interview with Charles Peter, Washington, DC, 26 June 1995. 42 Stansfield Turner, Secrecy and Democracy: The CIA in Transition (Boston: Houghton

Mifflin 1985) p.243. 43 Based on the author’s interviews with intelligence analysts, NIOs, and managers

periodically during these years, in Washington, DC. 44 NIC, Iran: Nuclear Intentions and Capabilities. 45 Sherman Kent, Intelligence for American World Policy (Princeton, NJ: Princeton University

Press 1942) pp.64–5. 46 See Rhodri Jeffreys-Jones, The CIA and American Democracy (New Haven, CT: Yale

University Press 1989) pp.212–13; Richard Pipes, ‘Team B: The Reality Behind the Myth’, Commentary 82 (October 1986) pp.24–40; John Ranelagh, The Agency: The Rise and Decline of the CIA (New York: Simon & Schuster 1987) pp.622–4; and Weiner, Legacy of Ashes, pp.351–52.

47 Chester Cooper, retired CIA analyst, interviewed by Ron Nessen, ‘Intelligence Failure: From Pearl Harbor to 9/11 and Iraq’ (television transcript), America Abroad Media (July 2004) p.11.

48 For an account of this process, see Report on the U.S. Intelligence Community’s Prewar Intelligence Assessments on Iraq (the Roberts Report, after the chair of the Senate Select Committee on Intelligence, Pat Roberts, Republican of Kansas), US Senate, 108th Cong., 2d Sess. , 7 July 2004, excerpted in Johnson, Strategic Intelligence, Vol. 2, pp.318–20.

49 For an example of a DCI-penned NIE, see the discussion of DCI Stansfield Turner in Douglas F. Garthoff, Directors of Central Intelligence as Leaders of the U.S. Intelligence Community, 1946–2005, Center for the Study of Intelligence (Washington, DC: Central Intelligence Agency 2005) p.153n.; Gerald K. Haines and Robert E. Leggett (eds.)Watching the Bear: Essays on CIA’s Analyis of the Soviet Union (Washington, DC: Central Intelligence Agency 2003) pp.169–70; and Ranelagh, The Agency, p.662. Stansfield Turner has stated that, from time to time, he elected to ‘inject my views into [NIEs] in order to give them some real substance’ rather than the watered down comprise documents he thought they sometimes were (Turner, Burn Before Reading, p.96). Turner was careful to make sure that ‘the President clearly knew that the opinion expressed in the NIE was that of his intelligence adviser, and me alone’ (p.184). Both Garthoff and Turner note that DCI William J. Casey of the Reagan Administration often appended his own personal notes to NIEs, registering his dissent (pp.153n and 199, respectively).

50 For an account by Sherman Kent, see his ‘A Crucial Estimate Relived’, Studies in Intelligence 27 (Winter 1985) pp.1–18 (originally published under this same title in Studies

366 INTELLIGENCE AND NATIONAL SECURITY

in Intelligence 36 (Spring 1964) pp.111–19). A history of the CIA refers to this Estimate as ‘a high-water mark of misjudgment for forty years, until the CIA assayed the state of Iraq’s arsenal [in 2002]’: Weiner, Legacy of Ashes, p.195. See also Michael Douglas Smith, ‘Revisiting Sherman Kent’s Defense of SNIE 85–3–62’, Studies in Intelligence 51 (September 2007) pp.29–32.

51 Author’s interview, 19 November 1984, in Johnson, America’s Secret Power, p.3. 52 See, for instance, Stephen J. Flanagan, ‘The Coordination of National Intelligence’, in

Duncan L. Clarke (ed.) Public Policy and Political Institutions. United States Defense and Foreign Policy – Coordination and Integration (Greenwich, CN: JAI 1985) pp.157–96.

53 See Tenet, At the Center of the Storm, p.370. 54 NIC, Iran: Nuclear Intentions and Capabilities. 55 For examples of this twisting, such as the Administration’s distortions about alleged Iraqi

attempts to acquire yellow-cake uranium from Niger, see accounts throughout Tenet, At the Center of the Storm, and a summary in Loch K. Johnson, ‘A Framework for Strengthening U.S. Intelligence’, Yale Journal of International Affairs 1 (Winter/Spring 2006) pp.166–31, reprinted in Loch K. Johnson and James M. Wirtz (eds.) Intelligence and National Security: The Secret World of Spies (New York: Oxford University Press 2007) pp.497–505.

56 See, for example, Frank Church, Congressional Record, 11 November 1975, p.S35787; Richard Helms, with William Hood, A Look Over My Shoulder: A Life in the Central Intelligence Agency (New York: Random House 2003) p.235; Turner, op.cit., p.245.

57 See, for instance, the debate over Soviet military intentions during the Reagan years, described in Church, ibid.; Robert M. Gates, From the Shadows (New York: Simon & Schuster 1996) pp.264–6; and Daniel O. Graham, ‘Estimating the Threat: A Soldier’s Job’, Army Magazine (April 1973) pp.14–18.

58 Richard K. Betts has commented on the repercussion of worst-case analysis: ‘The norm of assuming the worst poses high financial costs and potential risks increasing political friction that could make crises escalate unnecessarily’ (Enemies of Intelligence: Knowledge and Power in American National Security (New York: Columbia 2007) p.36).

59 Arthur Hulnick, book review, Conflict Quarterly 14 (Winter 1994) p.73. 60 See, for instance, Ford, Estimative Intelligence, pp.41–2, 48–9; Chester L. Cooper, ‘The CIA

and Decision-Making’, Foreign Affairs 50 (January 1972) pp.223–36. 61 Author’s interviews, 28 August 1997, Washington, DC. 62 Gregory F. Treverton, RAND Corporation and former vice chair of the National Intelligence

Council, ‘Intelligence for an Era of Terror’, memorandum sent to the author, 23 March 2007.

63 Remarks to the author, 9 November 1995, Washington, DC. 64 Letter from Sherman Kent to Frank Wisner (dated 18 November 1963), Sterling Library

Collection, Yale University, Series I, Box 18, Folder 390. 65 Sherman Kent, ‘Estimates and Influence’, Foreign Service Journal (April 1969) p.17. 66 President George W. Bush said in a press conference on 21 September 2004 that the CIA

was ‘just guessing’ when it came to analyzing foreign affairs. The president had in mind specifically the pessimistic predictions coming from the Agency about the US occupation of Iraq, but he seemed to imply that intelligence generally was merely guesswork. This was an overstatement by the president that did not help morale in the intelligence community; nevertheless, strategic intelligence does have an element of guessing involved, although normally by men and women who have closely studied the subject in question and are better informed than most policymakers.

67 See Percy Cradock, Know Your Enemy (London: John Murray 2002). 68 Senator Frank Church (Democrat, Idaho), known as a harsh critic of the CIA, noted in 1975

that ‘In the last twenty-five years, no important new Soviet weapons system, from the H- bomb to the most recent missiles has appeared which had not been heralded in advance by NIEs’ (Church, Congressional Record, p.S35787). An example is the Soviet Polaris-type ‘Y’ missile and the submarines on which it was carried, both well anticipated before the first boats were launched. See also Lawrence Freedman, U.S. Intelligence and the Soviet Strategic Threat (Princeton, NJ: Princeton University Press 1986); and John Prados,

THE PRESIDENT’S DAILY BRIEF AND THE NIE 367

The Soviet Estimate: U.S. Intelligence Analysis of Russian Military Strength (New York: Dial 1982).

69 See also United States General Accounting Office (GAO), Foreign Missile Threats: Analytic Soundness of Certain National Intelligence Estimates, Report to the Chairman, Committee on National Security, House of Representatives (August 1966).

70 Church, Congressional Record, p.S35786. 71 Remark to the author, 21 February 1988, Athens, Georgia. Critics of the viewpoint have a

lively rejoinder at hand in Twain’s maxim: ‘There are many scapegoats for our blunders, but the most popular one is Providence’ (Mark Twain, Collected Tales, Sketches, Speeches, and Essays, 1891–1910 (New York: Library of America 1992) p.946).

72 Richard K. Betts, ‘Analysis, War and Decision: Why Intelligence Failures Are Inevitable’, World Politics 31 (October 1978) p.78. See, also Betts, Enemies of Intelligence.

73 Ford, Estimative Intelligence, p.7. 74 Hulnick, book review, p.74. 75 Ibid., p. 49. 76 William E. Odom, Fixing Intelligence for a More Secure America, 2nd ed. (New Haven,

CT: Yale University Press 2004) p.81. 77 Kent, ‘Estimates and Influence’, p. 45. 78 See, for example, Russell, ‘Low-Pressure System’, p.122. 79 Author’s notes on remarks by Mark M. Lowenthal (a former NIC vice chair), speaker,

Canadian Association for Security and Intelligence Studies (CASIS) Conference, Ottawa, 27–28 October 2006.

80 Mark M. Lowenthal, ‘Intelligence Analysis: Management and Transformation Issues’, in Jennifer E. Sims and Burton Gerber (eds.) Transforming U.S. Intelligence (Washington, DC: Georgetown University Press 2005) pp.220–38, quote at p.227.

81 Ford, Estimative Intelligence, p.38. 82 Betts, Enemies of Intelligence, p.70. 83 Author’s interview with a senior DI official, 28 August 1997, Washington, DC. 84 James Risen, State of War: The Secret History of the CIA and the Bush Administration (New

York: Free Press 2006) p.6. 85 Ford, Estimative Intelligence, p.38. 86 The 80% figure comes from the author’s interview with a senior CIA manager in the

Agency’s Intelligence Director in 1997 (28 August 1997, Washington, DC); Risen, State of War, cites former CIA analyst Carl Ford to the effect that ‘today, about 90 percent of analysts do nothing but current reporting’ (p.7). Amy Zegart underscores how performance evaluations in the CIA have been based on the number of reports produced by analysts. ‘For career-minded analysts, the message was clear’, she writes, ‘stay away from strategic assignments’ (Spying Blind: The CIA, the FBI, and the Origins of 9/11 (Princeton, NJ: Princeton University Press 2007) p.92).

87 Lowenthal, CASIS Conference. 88 Author’s interview with a former Vice Chairman for Estimates, 28 August 1997,

Washington, DC. 89 Joseph F. Nye., Jr., testimony before the Aspin-Brown Commission, open hearing, 19

January 1995, Washington, DC. 90 Author’s interview, 9 November 1995, Washington, DC. 91 Author’s notes of Allan E. Goodman’s remarks, Conference on US Intelligence, CIA,

Langley, Virginia, 1 October 1993. 92 Author’s interview with Robert M. Gates, Washington, DC, 28 March 1994. Gates went on

to say, though, that the NIE content filters back to policymakers through oral briefings and memos from their staff.

93 Betts notes: ‘Most intelligence products, even NIEs, are never read by high-level policymakers. At best, they are used by second-level staffers as background material for briefing their seniors’ (Betts, Enemies of Intelligence, p.40).

94 Cooper, ‘The CIA and Decision-Making’. 95 Author’s interview, under conditions of anonymity, 22 August 1997, Washington, DC. 96 Lowenthal, CASIS Conference.

368 INTELLIGENCE AND NATIONAL SECURITY

97 Ibid. 98 Turner, Burn Before Reading, p.243. 99 William Nolte, remarks, CASIS Conference. 100 The author is grateful to the CIA for providing these statistics, which come independently

from two different sources within the Agency. Of course, a much more important question than the annual production rates (the frequency) of NIEs is their quality. Unfortunately, this question cannot be systematically addressed here, since most NIEs remain classified. In 2007, Director of National Intelligence Michael McConnell vowed to keep them classified, arguing that ‘the integrity of the NIE process could be harmed by expectations that all or portions of the NIE are likely to be declassified’ (‘Guidance on Declassification of National Intelligence Estimates Key Judgments’, Memo, DNI to the Intelligence Community Workforce, 24 October 2007). For examples of some Estimates that have been made available to the public, see Johnson (ed.) Strategic Intelligence, Vol.2, Appendix F, pp.241– 95. For attempts at an evaluation of some NIEs that have become a part of the public record, see the remarks of Senator Frank Church on the Senate floor (Congressional Record, 11 November 1975, p.S35787), recounted in Loch K. Johnson, Season of Inquiry: The Senate Intelligence Investigation (Lexington: University Press of Kentucky 1985) pp.119–20; Freedman, U.S. Intelligence and the Soviet Strategic Threat; Ford, Estimative Intelligence; Loch K. Johnson, Secret Agencies (New Haven: Yale University Press 1996) pp.197–201; Prados, The Soviet Estimate; and Weiner, Legacy of Ashes. For an insightful and eloquent appraisal by a leading British intelligence official, see Percy Cradock, Know Your Enemy: How the Joint Intelligence Committee Saw the World (London: John Murray 2002) especially pp.271–80. As these and other studies indicate, NIEs are certainly fallible; there have been many miscalculations over the years, such as the misreading of the imminence of the Arab–Israeli war in 1973. (For a testy exchange between DCI Tenet and the Kean 9/11 Commission about the failure of an NIE on Somalia in 1993 to warn the White House of early Al Qaeda involvement in operations against US soldiers abroad, see the Commission’s public hearing on 14 April 2004, cited in Amy B. Zegart, ‘‘‘CNN with Secrets’’: 9/11, the CIA, and the Organizational Roots of Failure’, International Journal of Intelligence and Counterintelligence 20 (Spring 2007) pp.18–49, at note 107.) Yet, at the same time, Estimates have often proved to be accurate and valuable, as with the CIA’s reliable monitoring of military weapons systems in the Soviet Union and its tracking of petrodollars worldwide during the Cold War. As Michael Herman has put it in classic British understatement, ‘When the full history of the Cold War is written, the record of CIA’s Directorate of Intelligence in assessing the USSR may well emerge with some credit’ (Intelligence Power in Peace and War, p.369). See also Les Aspin, ‘Debate Over U.S. Strategic Forecasts: A Mixed Record’, Strategic Review 8 (1980) pp.29–43, 57–9; and Les Aspin, ‘Misreading Intelligence’, Foreign Policy 43 (Summer 1981) pp.166–72. On the early days of estimating, see Douglas Keane and Michael Warner (eds.) ‘The Intelligence Community, 1950–1955’, Foreign Relations of the United States, 1950–1955 (Washington, DC: Government Printing Office 2007); David C. Humphrey, ‘Organizational Management of U.S. Foreign Policy, 1969–1972’, Foreign Relations of the United States, 1969–1976, Vol. II (Washington, DC: Government Printing Office 2006) documents 179–290; and Donald P. Steury (ed.) Intentions and Capabilities: Estimates on Soviet Strategic Forces, 1950–1983, CIA History Staff, Center for the Study of Intelligence, Central Intelligence Agency (1996).

101 The data came to the author by year, not by DCI tenure; the decision rule here was to award all of the Estimates in a given year to the Intelligence Director who served the most time in that particular year. The figures presented here, then, should be considered accurate yearly portrayals of NIE production, but the discussion of NIEs by DCI is more approximate. The author also looked at the data by dividing NIEs according to the faction of time a DCI served in a given year; this approach did nothing to alter the conclusions offered in the text. For the precise dates of service for each of the DCIs, see David S. Robarge, ‘Directors of Central Intelligence, 1946–2005’, Studies in Intelligence 49 (2005) p.2.

102 A point stressed by Lowenthal, ‘Intelligence Analysis’, p.227. 103 See Garthoff, Directors of Central Intelligence, p.19.

THE PRESIDENT’S DAILY BRIEF AND THE NIE 369

104 See Ranelagh, The Agency, p. 192. 105 See Richard Kovar, ‘Mr. Current Intelligence: An Interview with Richard Lehman’, Studies

in Intelligence 43 (1999–2000) p.27; Ranelagh, The Agency, p.242; and Turner, Burn Before Reading, pp.78, 88.

106 Helms, A Look Over My Shoulder, p.236. 107 Weiner, Legacy of Ashes, p.122. 108 This is in part a function of the fact that Dulles was in the DCI office longer than any other

Director; but his average per year was still the third highest, outpaced only during the brief tenures of DCI Robert M. Gates in 1992 and R. James Woolsey in 1993–94 (see Table 1).

109 Ranelagh, The Agency, p.414. 110 Kent, ‘A Crucial Estimate’. 111 In this examination of NIE production, it should be understood that any particular DCI may

not deserve the credit apportioned here; in the case of Raborn, for instance, some historians of the CIA have concluded that Richard Helms, the deputy to Adm. Raborn, actually ran the Agency during the Admiral’s tenure (Weiner, Legacy of Ashes, p.252). Nevertheless, just as with presidents, DCIs reap the praise and the blame for what happens during their tenures.

112 Helms, A Look Over My Shoulder, p.237. 113 Cooper, ‘The CIA and Decision-Making’, p.227. 114 Thomas L. Hughes, ‘The Power to Speak and the Power to Listen: Reflections in

Bureaucratic Politics and a Recommendation on Information Flows’, in Thomas Franck and Edward Weisband (eds.) Secrecy and Foreign Policy (New York: Oxford 1974) pp.13–41.

115 For an account, see Loch K. Johnson, A Season of Inquiry: The Senate Intelligence Investigation (Lexington: Kentucky University Press 1986).

116 See William E. Colby with Peter Forbath, Honorable Men: My Life in the CIA (New York: Simon & Schuster 1978).

117 The 9/11 Commission Report, pp.341–3. 118 Weiner, Legacy of Ashes, p.499. 119 Carl W. Ford, Jr., interviewed by Weiner, Legacy of Ashes, p.500. Elsewhere, Carl Ford has

observed: ‘If I had to point to one specific problem that explains why we are doing such a bad job on intelligence, it is this almost single-minded focus on current reporting’ (Risen, State of War, p.7).

120 Remarks to the author, ‘Accountability of Intelligence and Security Agencies and Human Rights’, International Symposium, sponsored by the Dutch Review Committee on the Intelligence and Security Services and the Faculty of Law, Radboud University, Nijmegen, The Hague, Holland, 8 June 2007.

121 Author’s telephone interview with a senior CIA analyst, 20 February 2006. 122 See Tenet, At the Center of the Storm, p.122. 123 Lowenthal, ‘Intelligence Analysis’, p.227. 124 For additional thoughts on this subject, see Johnson, Strategic Intelligence, Vol.2;

Lowenthal, ‘Intelligence Analysis’; and Richard L. Russell, Sharpening Strategic Intelligence (New York: Cambridge University Press 2007).

370 INTELLIGENCE AND NATIONAL SECURITY