The bombshell came at the most inopportune time.

An FBI special agent was testifying in the government’s high-profile terrorism trial against Omar Abdel Rahman, the “blind sheik” suspected of plotting the first attack on the World Trade Center.

Frederic Whitehurst, a chemist and lawyer who worked in the FBI’s crime lab, testified that he was told by his superiors to ignore findings that did not support the prosecution’s theory of the bombing.

“There was a great deal of pressure put upon me to bias my interpretation,” Whitehurst said in U.S. District Court in New York in 1995.

Even before the Internet, Whitehurst’s extraordinary claim went viral. It turned out he had written or passed along scores of memos over the years warning of a lack of impartiality and scientific standards at the famed lab that did the forensic work after the World Trade Center attack and in other cases.

With the FBI under fire for its handling of the 1993 trade center attack, the Oklahoma City bombing and the O.J. Simpson murder case, officials had to act.

After the Justice Department’s inspector general began a review of Whitehurst’s claims, Attorney General Janet Reno and FBI Director Louis J. Freeh decided to launch a task force to dig through thousands of cases involving discredited agents, to ensure that “no defendant’s right to a fair trial was jeopardized,” as one FBI official promised at a congressional hearing.

The task force took nine years to complete its work and never publicly released its findings. Not the results of its case reviews of suspect lab work. Not the names of the defendants who were convicted as a result. And not the nature or scope of the forensic problems it found.

Those decisions more than a decade ago remain relevant today for hundreds of people still in the U.S. court system, because officials never notified many defendants of the forensic flaws in their cases and never expanded their review to catch similar mistakes.

A review of more than 10,000 pages of task force documents and dozens of interviews demonstrate that the panel operated in secret and with close oversight by FBI and Justice Department brass — including Reno and Freeh’s top deputy — who took steps to control the information uncovered by the group.

“It was not open,” said a person who worked closely with the task force and who spoke on the condition of anonymity because the bureau and Justice Department maintain a strong influence in forensic science. “Maybe [a coverup] wasn’t the intent, but it did seem to look that way. . . . It was too controlled by the FBI.”

The documents and interviews tell a story of how the Justice Department’s promise to protect the rights of defendants became in large part an exercise in damage control that left some prisoners locked away or in the dark for years longer than necessary. The Justice Department continues to decline to release the names of defendants in the affected cases.

A Washington Post review of the department’s actions shows an agency struggling to balance its goal of defending convictions in court with its responsibility to protect the innocent. The Justice Department’s decision to allow prosecutors to decide what to disclose to defendants was criticized at the time and allowed most of the process to remain secret. But by cloaking cases in anonymity, failing to ensure that defendants were notified of troubles with their cases and neglecting to publicly report problems or recommend solutions, the task force obscured problems from further study.

Justice Department spokeswoman Laura Sweeney said the federal review met constitutional requirements by allowing prosecutors in the affected cases to make the final decision whether to disclose potentially exculpatory information to the defendants.

“In January 1996 the Department established a Task Force to advise prosecutors of the Office of Inspector General investigation of the FBI lab,” Sweeney said in a statement. The task force worked with prosecutors and the FBI “to notify the relevant prosecutors [local, state and federal] so that they could determine what information needed to be disclosed to defense counsel.”

Scathing report

If the Justice Department was secretive, the agency’s independent inspector general was not. Michael R. Bromwich’s probe culminated in a devastating 517-page report in April 1997on misconduct at the FBI lab.

His findings stopped short of accusing agents of perjury or of fabricating results, but he concluded that FBI managers failed — in some cases for years — to respond to warnings about the scientific integrity and competence of agents.

The chief of the lab’s explosives unit, for example, “repeatedly reached conclusions that incriminated the defendants without a scientific basis” in the 1995 Oklahoma City bombing, Bromwich wrote. The head of toxicology lacked judgment and credibility and overstated results in the 1994 Simpson investigation. After the 1993 World Trade Center attack, the key FBI witness “worked backward,” tailoring his testimony to reach the result he wanted. Other agents “spruced up” notes for trial, altered reports without the author’s permission or failed to document or confirm their findings.

The investigation led to wide-ranging changes, including higher laboratory standards and requirements for examiners.

Meanwhile, the Justice Department set out to evaluate discredited agents’ work in thousands of cases that had gone to trial.

Jim Maddock, the FBI’s assistant general counsel, told reporters that the goal of the new task force was to identify any potentially exculpatory information that had arisen in any criminal case involving agents criticized in the report.

“We are undertaking that review,” Maddock said at an April 15, 1997, news conference. “And when it is done, we will give a full accounting of our findings.”

Interviews and documents show that key decisions about the task force’s work were made at the highest levels, including the decisions to exclude defense lawyers from the review and not publicly release the findings.

Task force participants said Reno signed off on the decision allowing prosecutors to decide what to disclose, because normal legal and constitutional requirements give prosecutors that discretion.

Justice Department officials also believed that the public release of the 1997 inspector general report generated enough publicity to give defense attorneys and their clients opportunities to appeal, task force participants said.

“Our job was to do the scientific reviews and then to send the results to the prosecutors, and they were responsible for determining whether they were going to disclose or not,” Lucy L. Thomson, the chief of the task force, said in an interview. “That was just the way Janet Reno decided to do it.”

Reno is physically ailing and was unable to comment for this article.

Her deputy attorney general until April 1997, Jamie Gorelick, said Reno “was very, very interested in assuring that we weren’t keeping in prison people who deserved to have their convictions reviewed.”

“I am sure she tried as hard as she could to keep the pressure on the bureau and on the criminal division,” Gorelick said.

Delays, omissions

Documents show that the FBI and Justice Department set strict rules about what information would be disclosed as they prepared to battle defendants who challenged convictions.

The department planned to “monitor all decisions” by federal prosecutors over whether to disclose information, the head of the criminal division, John C. Keeney, wrote in a memo to all U.S. attorneys on Jan. 4, 1996. The division stood ready, if necessary, to “evaluate the allegations and, if appropriate rebut them,” he wrote.

In addition, the Justice Department and the FBI negotiated over the limit and scope of the task force review, the documents show.

For example, in a June 1997 memo, Keeney told federal prosecutors that the criminal division and the FBI would “arrange for an independent, complete review of the Laboratory’s findings and any related testimony” in all convictions in which they found there was a “reasonable probability” that work by discredited agents had affected the conviction or sentence.

But two months later, the senior attorney in charge of the task force told Keeney’s deputy that the FBI indicated that it planned to require “a cursory paper review” only and generally did not plan to reexamine evidence.

That attorney in charge, Thomson, told Deputy Assistant Attorney General Kevin V. DiGregory in an Aug. 19, 1997, memo that the FBI also wanted to keep the focus off the most vulnerable cases by not conducting reviews if a case was still in litigation or on appeal — even though the panel’s work would have been most relevant to a judge at those times.

There were other hitches. One year later, in August 1998, Thomson complained to DiGregory that “no scientists have been retained to date” by the FBI to conduct reviews of cases in which defendants may have been wrongly convicted.

Reviews were “needed as soon as possible in order to avoid possibly undercutting prosecutors’ arguments . . . and to ensure that defendants will not exhaust opportunities to file post-conviction relief motions,” Thomson said.

As it turned out, reviews would continue for six years, leaving defendants in jail after having been convicted in cases with faulty forensics.

Keeney died last year after retiring in 2010 as the longest-serving federal prosecutor in U.S. history. DiGregory did not return messages left at his home and passed through an associate.

Thomson, now a privacy expert, said that the reviews were not cursory and that she did not know whether any defendants had lost opportunities to appeal their convictions.

Reduced paper trail

As the cases became known to state and local prosecutors, many moved swiftly and made full disclosures. Others stymied the effort, whether intentionally or not.

Because of the sheer passage of time, files, trial transcripts or other records often were lost or destroyed. Personnel turnover in prosecutors’ offices often left behind no living memory of cases. Many state and local prosecutors worked in small offices with enormous active caseloads and had little stake in the Justice Department process.

As a result, reviewers dropped plans to require that state and local prosecutors sign statements when they determined a discredited agent’s work was pivotal to a case, or to explain in writing if they determined it was not, records show.

That reduced the paper trail. As long as the task force got the information, a participant said, it did not matter whether it was written down.

The task force did order reviews for multiple cases in which prosecutors refused to cooperate. For example, Tampa prosecutor Harry Lee Coe III, now deceased, told the department that his lawyers were too overworked to review questioned death penalty cases, documents show.

In South Carolina, the task force completed a scientific review in late 2002 in the case of Roy David Brooks, who had been convicted of murder. But the review came after the state had destroyed records. And the destruction of records came days after the task force wrote to prosecutors for the third time in four years seeking such records.

Even when cases were disclosed to defense counsel, it was not clear what was disclosed.

In some cases, one-sentence notifications were sent to defendants, many of whom were indigent, still in prison or without attorneys.

“Please find enclosed a copy of the Attachment to Independent Case Review Report for CDRU#6480 Case File #95-253567, which we received, from the U.S. Department of Justice,” stated the entirety of a letter from prosecutors in Tampa to one defendant in April 2001. That letter came 18 years after the offense.

The attached three-page report did not contain the defendant’s name — only strings of four- and eight-digit FBI and Justice Department code numbers. It had nothing to indicate that it involved the particular defendant’s case or the meaning of bland statements of scientific results.

In other cases, records indicate that prosecutors told defendants or their attorneys early on about the inspector general’s report but never mentioned that the task force found more-specific problems.

The task force gradually wound down when Thomson and DiGregory departed. A new administration arrived months before the Sept. 11, 2001, terrorist attacks, which transformed priorities. In 2002, Michael Chertoff, then assistant attorney general for the criminal division, narrowed the review to speed its completion, dropping unspecified “small cases.”

Through a spokesman, Chertoff declined to comment.

In addition, the criminal division stopped asking prosecutors to notify it if they turned over review results to defense attorneys.

Research director Madonna Lebling and researcher Aaron Carter contributed to this report.