August 22, 2008 – Editor’s Note: George Washington University’s National Security Archive released a report Friday on prewar Iraq intelligence which found “the U.S. intelligence community buckled sooner in 2002 than previously reported to Bush administration pressure for data justifying an invasion of Iraq.” The report, a portion of which is based on previously undisclosed government documents, was prepared by National Security Archive senior fellow John Prados.
The documents suggest that the public relations push for war came before the intelligence analysis, which then conformed to public positions taken by Pentagon and White House officials. For example, a July 2002 draft of the “White Paper” ultimately issued by the CIA in October 2002 actually pre-dated the National Intelligence Estimate that the paper purportedly summarized, but which Congress did not insist on until September 2002.
A similar comparison between a declassified draft and the final version of the British government’s “White Paper” on Iraq weapons of mass destruction adds to evidence that the two nations colluded in the effort to build public support for the invasion of Iraq. Dr. Prados concludes that the new evidence tends to support charges raised by former White House press secretary Scott McClellan and by the Senate Select Committee on Intelligence in its long-delayed June 2008 “Phase II” report on politicization of intelligence.
U.S. Intelligence and Iraq WMD
Compiled and edited by Dr. John Prados
On June 5, 2008 the Senate Select Committee on Intelligence (SSCI) released a report examining whether the public statements made by U.S. officials, including President George W. Bush, Vice President Richard Cheney, Secretary of State Colin L. Powell, and others were consonant with U.S. intelligence information. This report forms part of a second phase of the SSCI’s investigation of Iraq intelligence issues, most especially Saddam Hussein’s possible Weapons of Mass Destruction (WMD) program, originally approved by the Intelligence Committee in February 2004 but stalled by its Republican majority for several years, until the majority changed with the current 110th Congress. Committee chairman Senator John D. Rockefeller IV (D-WV) then ordered work on this inquiry resumed, and the present report is the result.
The appearance of this long-awaited SSCI “Phase II” report coincided with controversy over the revelations of former White House press secretary Scott McClellan who, in a memoir appearing almost simultaneously, argued that “in the fall of 2002, Bush and his White House were engaging in a carefully orchestrated campaign to shape and manipulate sources of public approval to our advantage.” (Note 1) A review of new evidence along with previously-available documents sheds important new light on this debate. Among the findings:
* The Phase II report on Bush administration public statements, in conjunction with the SSCI’s original July 2004 report on Iraq’s alleged Weapons of Mass Destruction, indicates that political manipulation extended beyond the intelligence itself to affect investigation of the intelligence failures on Iraq as well as the Bush administration’s use of that information.
* In conjunction with other recently declassified materials, the Phase II report shows that the Bush administration solicited intelligence then used to “substantiate” its public claims.
* A recently declassified draft of the CIA’s October 2002 white paper on Iraqi WMD programs demonstrates that that paper long pre-dated the compilation of the National Intelligence Estimate on Iraqi capabilities.
* The timing of the CIA’s draft white paper coincides with a previously available draft of the British Government’s white paper on Iraqi WMD, demonstrating that the Bush administration and the Tony Blair government began acting in concert to build support for an invasion of Iraq two to three months earlier than previously understood.
* A comparison of the CIA draft white paper with its publicly released edition shows that all the changes made were in the nature of strengthening its charges against Iraq by inserting additional alarming claims, in the manner of an advocacy, or public relations document. The draft and final papers show no evidence of intelligence analysis applied to the information contained. Similar comparison of the British white paper shows the same phenomenon at work.
* Declassified Pentagon documents demonstrate that the CIA white paper was modified in ways that conformed to the desires of the Undersecretary of Defense for Policy and his office, in much the same way that British documents indicate that country’s white paper was changed to conform to the desires of the Blair government.
The many official investigations and unofficial investigations carried out, plus the statements and speeches of former CIA officials defending themselves against charges of distortion, have established a few points beyond question. Most important, following Saddam Hussein’s 1998 final expulsion of UN weapons inspectors from Iraq, very little new information fell into the hands of U.S. intelligence. Notable exceptions include data from Iraqi Foreign Minister Naji Sabri, recruited as a CIA source (Note 2), and from Iraqi scientists clandestinely approached by the CIA under a covert program. (Note 3) Both these streams of information denied the existence of Iraqi WMD. On the other side were data from Iraqi exile sources that claimed all sorts of WMD and a set of fabricated documents alleging an Iraqi deal to buy uranium ore in Niger. The only concrete “find” was of a shipment of aluminum tubes being imported into Iraq, but analysts were divided over whether these tubes had anything to do with WMD at all. U.S. intelligence largely discounted the (accurate) details from Sabri and the scientists and – despite the CIA’s expressed misgivings -made use of the exile data. This thin data conditioned the intelligence analysis.
There was also a source of intelligence failure that flowed not from bad information but from analytical procedures. American intelligence knew that Saddam had worked through the 1990s to deceive UN weapons inspectors – they assumed he was hiding his WMDs rather than concealing the lack of them. On specific weapons, for example long-range Iraqi missiles, intelligence took a standard accounting approach, and since they could not account for every Iraqi missile, assumed Saddam was hiding a covert force of ballistic missiles. U.S. intelligence was coming off a record of underestimating Iraqi WMD progress in the 1980s and now overcompensated in the other direction.
The recent SSCI Phase II report concludes that Bush administration statements, while “substantiated” by the CIA reporting, went beyond that data. The Republican minority on the committee attacked that conclusion. The main defense offered – and repeated by media commentators – is that the root cause of the administration’s Iraq hysteria was intelligence failure, not intent to manipulate the American public. A typical formulation is that of columnist Fred Hiatt in the Washington Post, who argued that “the phony ‘Bush Lied’ story line distracts from the biggest prewar failure: the fact that so much of the intelligence upon which Bush and Rockefeller and everyone else relied turned out to be tragically, catastrophically wrong.” (Note 4)
But the question of the role of threat manipulation in the origins of the Iraq war is complex and goes beyond analytical failure. Its center is the degree to which the Iraq intelligence was politicized. Absent the drumbeat for war, even exaggerated estimates of Iraqi WMD prowess would have represented only a standard foreign policy problem. Bush administration intentions made a difference. Both the SSCI Phase I report and that of the Commission on the Intelligence Capabilities of the United States on Weapons of Mass Destruction (the Silberman-Robb Commission) investigation, though arguing that no politicization had occurred, also cited cases suggesting the opposite. Former national intelligence officer Paul Pillar told an audience at the Council on Foreign Relations that the Silberman-Robb finding did not surprise him for two reasons: because any intelligence analyst would be reluctant to make the damning admission that his paper had been politicized, and because “in my experience, the great majority of cases of actual politicization – successful politicization – are invariably subtle.” (Note 5)
There were several avenues by which the Bush administration made its preferences clear. Vice President Richard Cheney questioned his CIA briefers aggressively, pressing them to the wall when he saw intelligence from other agencies that portrayed a more somber picture than that in CIA’s reporting. He sent briefers back for more information, including in instances when they checked with headquarters and returned with the same word. Cheney was especially acerbic on CIA’s rejection of claims that one of the 9/11 terrorists had met with Iraqi intelligence officers in Prague. On a number of occasions, Cheney sent his chief of staff, I. Lewis Libby, to CIA headquarters to follow up on his concerns. Mr. Cheney went there himself, not just once but on almost a dozen occasions. The practice encouraged the CIA to censor itself, driven, as Pillar put it, by “the desire to avoid the unpleasantness of putting unwelcome assessments on the desks of policymakers.” (Note 6)
A second avenue to influence U.S. intelligence lay through Donald Rumsfeld’s Pentagon. There, William Luti’s Near East and South Asia unit of the Office of the Undersecretary of Defense for Policy (OUSDP) was in close touch with the Vice President’s office. Papers circulated back and forth, and both offices utilized claims from Iraqi exiles – claims that Saddam trained terrorists or possessed various WMDs – to press the intelligence agencies for similar information. Under Paul Wolfowitz, the Deputy Secretary of Defense, and the undersecretary for policy, Douglas Feith, the Pentagon formed a special group to review reports on Saddam’s links to Al Qaeda. This unit, the Policy Counterterrorism Evaluation Group (PCTEG) has been represented by Feith as merely charged with assembling a briefing on terrorism, but its real function was to bring additional pressure to bear on the CIA.
Not all the manipulation was visible. Behind the scenes at the State Department, Undersecretary of State John R. Bolton, also closely allied with the Office of the Vice President, pressured both the State Department and the CIA to fire individuals who refused to clear text in his speeches leveling the most extreme charges against other countries. Although Bolton’s actions did not concern Iraq directly, they came to a high point during the summer of 2002 – the exact moment when Iraq intelligence issues were on the front burner – and they aimed at offices which played a central role in producing Iraq intelligence. These included the Bureau of Intelligence and Research at State plus the National Intelligence Council (NIC) and the Weapons Intelligence, Proliferation and Arms Control (WINPAC) center at CIA. Analysts working on Iraq intelligence could not be blamed for concluding that their own careers might be in jeopardy if they supplied answers other than what the Bush administration wanted to hear.
Under the circumstances, it is difficult to avoid the impression that the CIA and other intelligence agencies defended themselves against the dangers of attack from the Bush administration through a process of self-censorship. That is the very essence of politicization in intelligence. And the degree to which public statements on Iraq by Cheney, Bush, and others were “substantiated” by the existing intelligence must be viewed through that prism.
We shall offer only a few examples here. First is the case of the CIA white paper, “Iraq’s Weapons of Mass Destruction Programs.” That document is dated October 2002 [Document 1] and was issued on October 4. It has been represented as a distillation of the National Intelligence Estimate (NIE) on Iraq published two days earlier, with the most sensitive, secret information stripped out. Posted here today is the major portion of the text of the same paper in draft [Document 2], as it existed in July 2002. This document demonstrates that the white paper existed long before the NIE was even requested by Congress. In fact the illustrations in the July version are the same as those in the final report. A close comparison of the text shows, further, that much of the argumentation is identical, and that the differences between the two are strictly in the nature of separating text to insert more charges or to sharpen them. The entire product has the character of rhetoric. Little of the text shows the kind of approach characteristic of intelligence analysis. The fact that this document was in preparation at the CIA in July indicates that the Bush administration was actively engaged in a process of building support for war months ahead of the time it has previously been understood to have done so. In fact evidence exists that the CIA white paper was commissioned as early as May 2002. (Note 7)
This point is made even sharper by recently declassified Department of Defense documents, including a memorandum from the OUSDP that details the kinds of information seen as desirable to obtain from intelligence in order to strengthen the case for war against Iraq [Document 3]. The timing of this document suggests that this text was a response to the draft CIA white paper, created at a point when Pentagon critics of CIA reporting were actively pressing their case against the agency’s refusal to accept arguments that Saddam Hussein was allied with Al Qaeda. Changes in the CIA white paper between its July draft and the final document track closely with the OUSDP comments. The net impression is that the CIA white paper was rewritten to conform to administration preferences. If so, U.S. intelligence a priori made itself a tool of a political effort, vitiating the intelligence function and confirming the presence of a politicized process. The specific analytic failures on Iraq intelligence become much less significant in such a climate, especially in that they all yielded intelligence predictions of exactly the kind the Bush administration wanted to hear.
This impression is strengthened, and suspicions of collusion broadened, when the record of the British government’s white paper on Iraqi WMD is laid side by side with that of the CIA. In the course of British official investigations of the antecedents to the war, and the death of physicist David Kelly, a draft of the British white paper was released that is dated June 20 [Document 4]. As in the American case, the Joint Intelligence Committee, which originated this document and plays a role similar to that of the National Intelligence Council in the U.S., modified its draft to issue a final version on September 24, 2002, that was even more somber [Document 5]. There is a considerable record on the Blair government’s efforts to shape the content of the British white paper in directions not supported by the intelligence.
The second example concerns the U.S. government’s use of information drawn from Iraqi exile sources, principally those of the organization known as the Iraqi National Congress (INC) [Document 6]. This anti-Saddam group has had a long and stormy history with the CIA, which actually severed relations with it, an action the Clinton administration’s NSC Deputies Committee approved in December 1996. The agency was later forced to resume ties, and even to fund the group, as a result of the Iraq Liberation Act, which Congress passed in 1998. Proponents of that legislation included many individuals who became senior officials of the Bush administration. The State Department took up funding of the INC. Both State and CIA questioned the value of the intelligence it provided, and State in turn sought to end the relationship. In 2002, the Defense Intelligence Agency (DIA) took over responsibility for the Iraqi exiles. During this period the INC opened channels to the Office of the Vice President as well as Pentagon units responsible to Douglas Feith. In his own account of this period, Feith takes pains to defend the exile group and its leaders. (Note 8)
In the summer of 2002, the intelligence community compiled a detailed assessment of the material provided by the INC on several subjects and found it to have little current intelligence value, with sourcing and attribution impossible to verify. (Note 9) Despite this, and in spite of the fact that the INC went beyond providing intelligence to using the defectors it brought to the attention of the U.S. government as part of an anti-Saddam publicity campaign, the SSCI report on the group concludes that “false information from the Iraqi National Congress (INC)-affiliated sources was used to support key Intelligence Community Assessments on Iraq and was widely distributed in intelligence products prior to the war” (pp. 113-122). Intelligence agencies also avoided identifying these sources as INC-related in their reporting. Among the defectors was the notorious source “Curveball,” whose false allegations concerning Iraqi mobile biological weapons factories underlay some of the most alarming Bush administration charges against Baghdad.
This begs the question why, given distrust of the INC’s information at both the CIA and State Department, and an awareness of these doubts even within the DIA, the data was used at all, much less relied upon. Part of the answer no doubt has to do with the desperation of U.S. intelligence to obtain any information from inside Iraq – in itself a reflection of an intelligence failure. But the other part of the answer most likely flows directly from the prodding of the intelligence community by high levels at the Pentagon and White House for reactions to the defector information. This point stands out in stark relief when contrasted with the fact that the alternate stream of Iraqi insider information – from high-level agent sources and Iraqi scientists – seems to have had no discernable role in U.S. intelligence reporting. That is very arguably politicization.
Our third example has to do with the charges that Saddam sought to buy uranium ore from the African country of Niger. As widely reported, this affair involved fabricated documents, a Bush administration effort to discredit the U.S. envoy sent to check on the report by outing his wife, a CIA undercover officer; and ultimately, the criminal trial of Vice President Cheney’s top national security aide; but those matters are not of concern now. What is disturbing here, in the context of politicization of the intelligence, is the specific treatment the CIA gave to the information it developed. The record is established by the SSCI Phase I report, the Silberman-Robb report, and the proceedings of the trial of “Scooter” Libby, Mr. Cheney’s national security assistant. (Note 10)
On February 13, 2002 Vice President Cheney asked his CIA briefer about reports that Iraq was procuring uranium in Niger [Document 7]. Cheney represented the information as having come from the DIA, which indeed had issued an “executive highlight” on February 12. If this was in actuality what Cheney saw, the DIA was basing its account on information provided by Italian military intelligence, already aware of the fabricated Nigerien documents that later became the heart of this affair. The CIA had reported the same information a week earlier. The briefer promised to check, and the CIA’s WINPAC center prepared a note which observed that the foreign information on which the claim was based was only single-source and lacked crucial detail [Document 8]. The agency subsequently set up a trip to Niger by retired Ambassador Joseph V. Wilson IV, who returned with the conclusion that there was no substance to these claims. Wilson arrived in Niger on February 26 and returned on March 4. Just as Wilson came home, Vice President Cheney renewed his inquiry into the Niger allegation, and WINPAC responded by noting that the foreign intelligence service had no new information, that the Nigerien government insisted it was making all efforts to ensure that its uranium was used only for peaceful purposes, and that CIA was about to debrief “a source who may have information related to the alleged sale.”
Ambassador Wilson was in fact debriefed by two CIA officers on March 5. The way this was handled is what raises questions. Wilson’s data was recorded by the officers and written up by a reports officer who, according to the SSCI, “added additional relevant information from his notes.” The declassified text of this March 8, 2002, report [Document 9] shows that CIA Headquarters added the comment that the officials who provided information to Wilson “may have intended to influence as well as inform.” The ambassador himself was described as “a contact with excellent access who does not have an established reporting record.” However, Wilson had in fact carried out a mission on behalf of CIA previously, and he had been the senior U.S. envoy in Baghdad (the deputy chief of mission) before the first Gulf War. Therefore, Wilson did have an established reporting record. The comment regarding the Nigerien officials was gratuitous. The combination of these remarks cast doubt within the U.S. government on the information.
The report on Wilson’s information was then circulated in routine channels but never given to the Vice President. Director George Tenet’s comment: “This unremarkable report was disseminated, but because it produced no solid answers, there wasn’t any urgency to brief its results to senior officials such as the vice president.” (Note 11) But a look at the trip report we post here shows Wilson’s information was in fact quite solid. It simply does not say the uranium charge was real. Tenet has a secondary defense that the report was completed just after Vice President Cheney left on a trip to drum up support for war with Iraq, and that when he returned other matters seemed more pressing. Yet Cheney had renewed his inquiry into the Niger claim and surely its refutation had an impact on the arguments he had just made to encourage support for an American military option.These points drive the conclusion that the CIA was loathe to confront Mr. Cheney with a direct refutation of the Niger uranium claim. This too smacks of politicization.
Wilson’s was only one of a number of streams of reporting that undermined the Niger story, including an investigation by French intelligence and inquiries from the current U.S. ambassador and a senior U.S. military officer. Likely based on these materials and on the embassy cables reporting on Wilson from Niger, State Department intelligence filed a report doubting the claims of a Nigerien sale to Iraq [Document 10], and filed a dissent when the claim was included in the 2002 National Intelligence Estimate. But the developments of early 2002 became only the beginning of a highly ambivalent treatment of the uranium claim. On the one hand, the CIA intervened to keep this material out of the major speech President Bush gave in Cincinnati in October 2002, and also objected when British intelligence included it in their own white paper about the Iraqi threat. On the other hand, senior CIA officials mentioned the uranium claim in congressional testimony at the same time, permitted it to be included in a December 2002 “fact sheet” on Iraq, and mounted only tepid opposition to inclusion of the charge in President Bush’s 2003 State of the Union Address, where it would become notorious as the “16 Words.”
The SSCI later investigated the Iraq intelligence in detail, reporting on it in 2004. This was followed by the Silberman-Robb commission account. The SSCI Phase II report on the use of that intelligence [Document 11] examines Bush administration public statements regarding Iraqi weapons of mass destruction plus certain other topics related to war in Iraq against the inventory of intelligence reports circulated within the U.S. government. The idea was to determine whether administration claims were supported by the available intelligence. The “public statements” were winnowed down to a few, essentially the speech by Vice President Cheney in Nashville on August 26, 2002, those by President Bush to the United Nations General Assembly, in Cincinnati, and before the U.S. Congress at the 2003 State of the Union address (September 12 and October 7, 2002, and January 29, 2003), and the presentation to the United Nations Security Council by Secretary of State Powell (February 5, 2003). The subjects covered include nuclear weapons, biological weapons, chemical weapons, weapons of mass destruction generally, delivery systems, the Saddam Hussein regime’s alleged links to terrorists, Iraqi regime intent, and predictions for post-war Iraq.
In most of these cases the SSCI study found administration claims “substantiated” by the available intelligence but portraying the data as more certain than it was, thus going beyond the intelligence, while failing to convey disagreements among intelligence experts. The Committee found claims regarding Saddam’s intentions were contradicted by the intelligence (p. 82) and those about a rosy post-Saddam future as not reflecting intelligence concerns (p. 88). In the case of Bush administration claims about links between Saddam and terrorists the report reached several conclusions, judging that the intelligence substantiated general claims of Iraqi knowledge of and support for terrorist activities, but that claims of an Iraqi-terrorist alliance or of Iraqi training of terrorists were not backed up by the intelligence reporting (p. 71-2). In general Bush administration claims asserted greater certainty than existed in CIA reports.
This analysis was assailed by Republican members even before the SSCI report appeared. In minority statements attached to the eventual primary document (pp. 100-170) they detailed their objections. The minority charges that the investigation improperly confined itself to comparisons with finished intelligence products rather than the wider range of material actually available to top officials, and that it did not make similar assessments of the statements made by Democratic Party politicians, including Senator Rockefeller himself. Republican members and staff were not permitted to be involved in the drafting work on the report and the numerous amendments they offered were rejected.
The question of whether the “Iraqi threat” resulted from manipulation, as Scott McClellan and the SSCI majority suggest, or simple intelligence failure, as in the view of the Committee minority, is a key issue for all concerned. A real intelligence failure did occur. This is plain from the Intelligence Committee’s 2004 “Phase I” report as well as that of the Silberman-Robb Commission. (Note 12) The present author argued as much even before those studies appeared. (Note 13) The CIA director of that time, George Tenet, concedes, “In many ways, we were prisoners of our own history.” (Note 14) Retired CIA analyst Melvin A. Goodman, observing this analytical effort from the outside, concludes, “The U.S. rush to war against Iraq marked the worst intelligence scandal in the history of the United States.” (Note 15) But intelligence failure was abetted and magnified by the Bush administration’s drive to use charges about alleged Iraqi WMDs as justification for war.
Ascertaining the truth in this matter does not seem to have been as important as seeming to do so, at least for the Senate Intelligence Committee. A review of the minority statements contained in the new SSCI report in comparison to similar ones in the Phase I report – by the then-Democratic minority [Document 12] – reveals identical complaints regarding the conduct of the investigation. The present Republican minority’s charge that the report errs on politicization because both the Phase I SSCI and Silberman-Robb Commission concluded there was no evidence of this flies in the face of the strong assertions by the Democratic minority during Phase I that allegations of this type had not been taken seriously. Current Republican charges that the report erred by failing to check the public statements of Democrats against the intelligence are a red herring: they effectively rely upon the Bush administration’s success at hoodwinking political opponents and then take those opponents’ statements as authoritative evidence, an example of reverse logic. The SSCI staff rules which the Republican minority now says were used to shut it out of the investigation are the same ones a Republican majority previously relied upon to limit Democrats’ influence on the scope and content of the inquiry.
The preparation of white papers on both the United States and British sides also needs to be taken into account. That Bush and Blair each turned to their intelligence agencies for the papers is significant – they were evoking the imprimatur of secret intelligence to justify policy preferences. Both papers had the function of justification, not analysis, and neither government waited until it had compiled all the evidence before demanding these products. Neither government asked for intelligence estimates, fashioned in secret, in order to inform policy on Iraq. Instead, both Bush and Blair did want their intelligence agencies to carry out avowed political agendas. And the timing of the white paper drafts – now established as being in the summer of 2002, before there ever was a UN debate or a Security Council resolution – clearly indicates their true function. The accumulating weight of evidence currently supports the interpretation Scott McClellan gives, not that supplied by apologists for the Iraq war.