"It is basically subject to perception. If the detainee dies you’re doing it wrong."

This was perhaps the most chillingly outrageous, widely quoted statement by a government official to be aired by Senate Armed Services Committee Chairman Carl Levin, D-Mich., at hearings last summer and in the committee’s December 11 report on abuse of detainees by U.S. forces.

But the quoted official, CIA lawyer Jonathan Fredman, told the committee on November 18 that he had made no such statement. In fact, Fredman added in a heretofore confidential, five-page memo, he had stressed at the 2002 meeting with interrogators at the Guantanamo Bay detention facility described in the Levin committee’s report, "Interrogation practices and legal guidance must not be based upon anyone’s subjective perception" (emphasis added) but rather upon "definitive and binding legal analysis."

The overall effect of selective reporting by many critics has been to paint honorable former and current officials as a bunch of sadistic war criminals.

Remarkably, the 18-page report issued by the committee (headed "Executive Summary") does not mention Fredman’s vehement — and, in my view, quite plausible — denial of the horrifying words attributed to him in a document of debatable reliability that the report, and Levin, have treated as established fact.

Asked to explain, a staffer said that someone had contacted the committee on Fredman’s behalf in July and that the response was "we wanted to talk to him." However, the committee heard nothing further until Fredman’s memo arrived on November 18, three days before members voted on the report. (The CIA does not usually allow employees to talk to the Levin committee.) The report then went to classification review — most of it is still classified — and only technical edits could be made, the staffer said. The staffer added that Fredman’s memo did not make entirely clear which of the statements attributed to him in the document he denied making.

But the fact remains that the committee had Fredman’s memo in hand 23 days before it issued its public "Executive Summary" without mentioning the denial. If a newspaper did that, it might be on the losing end of a libel suit.

The Levin committee’s shabby treatment of Fredman is illustrative of a pattern. To be sure, the report makes a valid case that the military’s abuse of numerous detainees between 2001 and 2004 "cannot simply be attributed to the actions of a few ‘bad apples’ acting on their own," as Bush apologists have sought to do with diminishing credibility; the abuse was, to some extent, spurred by high-level decisions.

But Levin undermines his own credibility by overstating the evidence of high-level complicity in the most shocking abuses at Iraq’s Abu Ghraib prison and in Afghanistan, and by depicting so many officials as villainous.

One wonders how many other inconvenient facts the drafters of the report chose to omit. And one begins to notice other tendentious passages. Most are accurate as far as they go. But their wording misleadingly intimates that the undoubted approval by top-level Bush administration officials of near-torture (if not torture) of a relatively small number of high-value detainees, and the undoubtedly widespread abuses of a great many more low-level suspects and wrongly detained innocents, were part of a grand and broad conspiracy to inflict sadistic torture and commit other illegal abuses.

Take the report’s conclusion that Defense Secretary Donald Rumsfeld’s December 2, 2002, authorization of aggressive interrogation techniques for use at Guantanamo, on the recommendation of then-Pentagon General Counsel William J. Haynes II, "influenced and contributed to the use of abusive techniques, including military working dogs, forced nudity, and stress positions, in Afghanistan and Iraq."

This is true to a point. And some criticism of Rumsfeld and Haynes is warranted. But the report’s language might also foster an impression, unsupported by the evidence, that Rumsfeld, Haynes, and other top officials intended to encourage the widespread, wanton abuse of prisoners that Abu Ghraib came to symbolize.

And although the report mentions waterboarding — the most notorious interrogation method used by the CIA — it glides past the fact that no military detainee was waterboarded because, on Haynes’s advice, Rumsfeld denied an October 2002 request from Guantanamo’s commander to authorize waterboarding by military interrogators.

Nor does the report mention the fact that in September 2002 and thereafter, leading congressional Democrats including Rep. Nancy Pelosi of California, now House speaker, raised no objection when briefed on waterboarding and other interrogation methods.

The report dismisses with scorn the Bush team’s view that terrorists were unprotected by the Geneva Conventions, while ignoring the fact that this view had deep historical roots and was defended by highly respected scholars.(The Supreme Court rejected this position in 2006.)

The overall effect of such selective reporting by many critics — including human-rights activists, authors, and commentators far less judicious than Levin — has been to paint honorable former and current officials as a bunch of sadistic war criminals. All of this feeds the deeply destructive clamor for "bringing criminal charges" — in the reckless words of a New York Times editorial — against Rumsfeld, Haynes, and perhaps others.

To the contrary, these were public servants seeking to protect the country from terrifying threats, who sometimes went to extremes that seem unwise to people (including me) who have the luxury of judging from the sidelines, with benefit of hindsight.

A closer look at the Levin committee’s treatment of Jonathan Fredman provides a window into the kind of unfairness to which many officials have been subjected.

Fredman’s supposed statements came during a discussion of aggressive interrogation techniques including sleep deprivation, death threats, and waterboarding at an October 2, 2002, Guantanamo staff meeting. During a brief visit to the prison camp, Fredman, then chief counsel of the CIA’s Counterterrorist Center, had been asked to advise about legal restraints on tough interrogations.

Levin’s much-publicized quotation, at a June 17 hearing, of the statement that "if the detainee dies you’re doing it wrong" came months before Fredman’s denial. But how hard did Levin and his staff try to check the accuracy of the allegation before the chairman condemned him?

Fredman’s denial was known to Levin’s staff, and presumably the chairman himself, when they issued the committee’s December 11 report. And there had been reason all along to wonder about the reliability of the document on which the panel relied, headed "Counter Resistance Strategy Meeting Minutes." One of the statements that the "minutes" attributed to Fredman, for example, was absurd on its face and obviously garbled: "In Turkey, they say that interrogation at all, or anything you do to that results in the subject betraying his comrades is torture."

But instead of acknowledging the shakiness of the "minutes" — and despite the fact that the first page states in bold type that "all questions and comments have been paraphrased" — the Levin committee’s report put Fredman’s supposed statements in quotation marks.

"I did not say the obscene things that were falsely attributed to me [by Levin] at the Senate hearing," Fredman told the committee in his November 17 memo. To the contrary, "I … emphasized that all interrogation practices and legal guidance must not be based upon anyone’s subjective perception; rather they must be based upon definitive and binding legal analysis from the Department of Justice; that DOD must ensure that its treatment of detainees is fully lawful and authorized by the military chain of command; … and that comprehensive investigations must be conducted should a detainee pass away."

(The memo was shown to me by a source outside the executive branch. When I called Fredman, he said that under CIA rules, he is not at liberty to comment.)

This is not to suggest that Fredman denies making all of the controversial statements attributed to him in the committee’s report. The report (and the "minutes") may well be accurate in stating that he had described the vaguely written criminal law against torture as banning only physical pain so severe as to cause permanent damage to major organs or body parts and mental pain so severe as to lead to permanent, profound damage to the senses or personality.

These statements — which might reflect badly on Fredman had he come up with them on his own — in fact came almost verbatim from the Justice Department’s advice to the CIA in an August 1, 2002, memo. That memo was widely attacked as extreme and shoddily reasoned (by me, among others) after it leaked in June 2004. But until it was repudiated by Justice in 2004, the memo was binding on CIA lawyers such as Fredman.

As for Fredman’s supposed statement that "if the detainee dies you’re doing it wrong," maybe Levin or his staff did not believe Fredman’s denial or decided that his exact words in 2002 were less relevant than his memo’s admission that he had raised the possibility of a detainee’s dying during interrogation.

That would be a fair point. But it was far from fair to attribute a horrifying statement to a career CIA professional without mentioning his denial that he had ever made it.

This article appeared in the Saturday, January 10, 2009 edition of National Journal.