The FBI Laboratory: An Investigation into Laboratory Practices and Alleged Misconduct in Explosives-Related and Other Cases (April,1997) Office of the Inspector General, Department of Justice

This investigation by the Department of Justice Office of the Inspector General (OIG) concerned allegations of wrongdoing and improper practices within certain sections of the Federal Bureau of Investigation (FBI) Laboratory. Those allegations involved some of the most significant prosecutions in the recent history of the Department of Justice, including the World Trade Center bombing, the Oklahoma City bombing, and the mail bomb assassination of U.S. Circuit Judge Robert Vance (which was referred to within the FBI as the VANPAC case). The allegations implicated fundamental aspects of law enforcement: the reliability of the procedures employed by the FBI Laboratory to analyze evidence, the integrity of the persons engaging in that analysis, and the trustworthiness of the testimony by FBI Laboratory examiners. The allegations were brought to the OIG's attention by Supervisory Special Agent Frederic Whitehurst, a Ph.D. scientist employed in the FBI Laboratory. We also investigated problems that we ourselves identified in the course of our investigation, as well as information brought to our attention by other employees in the Laboratory.

The investigation spanned more than eighteen months and addressed a very large number of allegations. Most of Whitehurst's allegations were not substantiated; some important ones were. Our investigation identified policies and practices in need of substantial change. Since the allegations involved incidents that occurred over nearly a decade, some of those policies had already been changed by the FBI or were in the process of being changed before the draft report was completed. In a number of key instances, we found problems that Whitehurst had not raised. We also saw examples of superb work and encountered Laboratory personnel dedicated to the highest traditions of forensic science. But we also found some Laboratory supervisors and examiners whose performance merits critical comment, and raises serious questions about whether they should continue in their current roles within the Laboratory. Accordingly, in addition to general recommendations we made about Laboratory practices and procedures, we recommended that certain supervisors and examiners be reassigned from their current positions.

This investigation and our findings primarily concerned three units of the FBI Laboratory -- the Explosives Unit (EU), the Materials Analysis Unit (MAU), and the Chemistry-Toxicology Unit (CTU), all of which were in the Scientific Analysis Section (SAS), one of five sections of the Laboratory. Our findings and conclusions regarding certain cases in those units should not be imputed to other cases within those units, nor to other units in the SAS or other sections of the Laboratory that we did not investigate.

The next section of this Executive Summary provides an overview of our principal findings and recommendations. The Summary then generally corresponds to the organization of the Report. Section II describes the OIG investigation (Part Two of the Report). Section III summarizes the significant cases that are treated in detail (Part Three, Sections A-G of the Report). Section IV sketches the many other matters investigated (Part Three, Sections H1-H13 of the Report). Section V describes our findings and conclusions on Whitehurst's allegations of retaliation (Part Four of the Report). Section VI describes our findings and recommendations with respect to the conduct and performance of particular individuals (Part Five of the Report). Section VII summarizes our recommendations regarding general Laboratory practices and procedures (Parts Six and Seven of the Report).

We did not substantiate the vast majority of the hundreds of allegations made by Whitehurst, including the many instances in which he alleged that Laboratory examiners had committed perjury or fabricated evidence. We found, however, significant instances of testimonial errors, substandard analytical work, and deficient practices. Those findings with respect to individual cases appear in Section III of this Executive Summary and are treated in detail in Part Three of the Report. The types of problems we found included:

• Scientifically Flawed Testimony in the Psinakis, World Trade Center, Avianca, and Trepal cases.

• Inaccurate Testimony by an EU examiner in the World Trade Center case, by a former Laboratory examiner (who is still an FBI agent) in a hearing conducted by the judicial committee of the Judicial Council of the Eleventh Circuit regarding then-Judge Alcee Hastings, and by the CTU Chief in the Trepal case.

• Testimony Beyond the Examiner's Expertise in the World Trade Center, Avianca, and Hastings cases.

• Improper Preparation of Laboratory Reports by three EU examiners who altered, omitted, or improperly supplemented some of Whitehurst's internal reports (dictations ) as they were being compiled into an official report of the Laboratory. A former EU Chief failed to substantively review all of the reports in his unit, authorized EU examiners to modify Whitehurst's dictations when incorporating them into EU reports, and fostered a permissive attitude toward changes to Whitehurst's dictations.

• Insufficient Documentation of Test Results by the examiner who had performed work on hundreds of cases, including Psinakis and the UNABOM investigation, and by the CTU Chief.

• Scientifically Flawed Reports in the VANPAC and Oklahoma City cases, and in numerous cases by the former MAU examiner who worked on Psinakis, and in a few instances by an EU examiner who altered Whitehurst's reports.

• Inadequate Record Management and Retention System by the Laboratory.

• Failures by Management to resolve serious and credible allegations of incompetence lodged against the examiner who worked on the Psinakis case; to review properly the EU report in the Oklahoma City case; to resolve scientific disagreements among Laboratory examiners in three cases, including Avianca; to establish and enforce validated procedures and protocols that might have avoided problems in examiner reports in the Psinakis and VANPAC cases; and to making a commitment to pursuing accreditation by the American Society of Crime Laboratory Directors/Laboratory Accreditation Board before 1994.

• A Flawed Staffing Structure of the Explosives Unit that should be reconfigured so that examiners possess requisite scientific qualifications.

The OIG investigation exonerated most of the persons named in Whitehurst's allegations. Regarding some personnel, however, we criticized certain practices and performances in particular cases and recommended reassignments from their current positions and other actions. Our principal recommendations included:

• Because we recommended that the EU be restructured so that its unit chief and examiners have scientific backgrounds, EU Chief J. Thomas Thurman and all non-scientist EU examiners should be reassigned outside the Laboratory when the restructuring is accomplished. In the interim, the FBI should assess whether Thurman should continue to hold a supervisory position.

• CTU Chief Roger Martz should not hold a supervisory position in the Laboratory, and the FBI should assess whether he should continue to serve as a Laboratory examiner.

• EU examiner David Williams, who worked on the World Trade Center and Oklahoma City cases, should be reassigned outside the Laboratory.

• The FBI should assess what disciplinary action is now appropriate for Michael Malone, the former Laboratory examiner who testified in the Hastings hearing.

• We concluded that Frederic Whitehurst cannot effectively function within the Laboratory and suggested that the FBI consider what role, if any, he can usefully serve in other components of the FBI. In making that determination, the FBI and the Department of Justice must weigh the significant contribution he has made by raising issues that needed to be addressed within the Laboratory against (1) the harm he has caused to innocent persons by making many inflammatory but unsubstantiated allegations, and (2) the doubts that exist about whether he has the requisite common sense and judgment to serve as a forensic examiner.

To enhance the quality of the Laboratory's forensic work, we made recommendations in the following areas: (1) accreditation, (2) restructuring the EU, (3) the roles of Laboratory examiners and resolutions of disputes, (4) report preparation, (5) peer review, (6) case documentation, (7) record retention, (8) examiner training and qualification, (9) examiner testimony, (10) protocols, (11) evidence handling, and (12) the role of management. In response to a draft of this Report, the FBI accepted full responsibility for the failings we identified within the Laboratory. The FBI's response concurred with nearly all of the OIG's recommendations and stated that the Laboratory has implemented or is taking steps to implement them. The FBI's response to the draft report is contained in an Appendix, along with our reply to specific points raised in its response.

The OIG investigation essentially occurred in two phases. The first phase, lasting from 1994 to the summer of 1995, was limited in scope. As is detailed in the Report, during that period, allegations by Whitehurst were the subject of various reviews by the FBI Office of General Counsel (FBI OGC), the FBI Office of Professional Responsibility (FBI OPR), and the FBI Laboratory itself until mid-1995. The OIG's investigation in that period focused on Whitehurst's contentions that his analytical reports had been substantively altered by an EU examiner.

By the summer of 1995, after other scientists in the Laboratory confirmed certain aspects of Whitehurst's allegations, it became clear that a more global, comprehensive investigation was warranted. With the agreement of FBI Director Louis Freeh, and the full cooperation of the FBI, the OIG undertook such an investigation and retained an international panel of five scientific experts to consult with the OIG. Those experts, whose combined experience exceeds 100 years of work in forensic and national laboratories, have been integrally involved in the process of interviewing witnesses, reviewing documents, and writing this report. Four experienced prosecutors from United States Attorneys' Offices and the Criminal Division were detailed to the OIG to lead the investigation, and have provided considerable investigative expertise in this matter.

From the autumn of 1995 to the present, the OIG team has conducted hundreds of interviews, including re-interviews of key witnesses, and reviewed more than 60,000 pages of documents and transcripts. Upon completion of a draft report on January 21, 1997, the OIG solicited comments from the FBI and from prosecutors (primarily in the United States Attorneys' Offices) and other lawyers who handled the cases at issue to ensure that no factual errors were inadvertently included. The responses themselves, as well as our replies, are contained in a separate Appendix. In evaluating those responses, the OIG made some revisions to the Report. After careful consideration, in most instances we did not agree with requests to change the language in the draft report or our findings, and have explained our reasoning either in the Report itself or in the Appendix.

One general point about the responses bears highlighting in this summary. As to cases in which we criticize the work of FBI Laboratory personnel, such as in the World Trade Center and Avianca cases, the FBI and U.S. Attorneys have responded by saying, in essence, that nothing in the Report should be read as affecting the outcome of those cases. Our purpose has not been to determine whether a defendant in any given case was improperly convicted of a crime; it was to ascertain whether the performance of the Laboratory personnel met general standards of conduct for forensic scientists and complied with policies in the FBI Laboratory in effect at the time the work was performed. Our findings of deficiencies in the work performed in cases should in no circumstance be read as expressing a view as to whether that case should have reached a different outcome. That role is properly performed by the prosecutors, defense counsel, and judges who can assess the work of the FBI Laboratory in the context of all of the evidence in the case. We, therefore, concluded that it would be inappropriate for us to make any judgments as to whether our findings will or should affect a particular case.

From the time Frederic Whitehurst first joined the FBI Laboratory in 1986, he repeatedly complained about the work practices of Agent Terry Rudolph, who preceded Whitehurst as the Laboratory's senior examiner for the analysis of explosives residue. Those complaints reached an apex with work Rudolph performed in connection with the Psinakis case. After that case ended in an acquittal, the Assistant United States Attorney (AUSA) who tried the case wrote a letter to the FBI complaining that Rudolph's performance was deficient, that the judge had nearly excluded his testimony, and that the defense had seriously impeached his scientific work and conclusions relevant to that case. That letter raised serious questions about certain Laboratory practices. For example, it noted the apparent absence within the Laboratory at the time of established protocols to determine when certain tests should be performed and of peer review to confirm the sufficiency of the analysis conducted by the Laboratory examiner.

Laboratory management responded to the AUSA's letter by directing that Rudolph's case files be audited. In August 1989, an internal audit of some of Rudolph's files found numerous shortcomings and recommended that an extensive technical review be undertaken. That review was assigned to Roger Martz, the chief of the Chemistry-Toxicology Unit (CTU). Martz reviewed 95 of Rudolph's files, concluded that Rudolph's analysis supported the results, and reported finding no technical errors. Upon the completion of Martz's review, the Laboratory determined that no further action concerning Rudolph was necessary. That decision proved to be a significant error in judgment. Our investigation showed that Martz's review was seriously deficient, that he failed to engage in the type of technical review that would actually have assessed the competence and sufficiency of the work purportedly performed by Rudolph, and that Martz's written reporting led Laboratory managers to believe that there were no problems with Rudolph's work or his files.

Because the Laboratory took no action against Rudolph, Whitehurst continued to complain about Rudolph's sloppy work habits, and added charges that Rudolph had perjured himself in a case, lied to an AUSA, abused annual leave, and made racist remarks. Those allegations led to an FBI OPR investigation in 1991-1992. Although we did not find evidence of a deliberate effort to dismiss or ignore Whitehurst's allegations -- as he has maintained -- we did find significant deficiencies in the OPR investigation of this matter.

The OPR investigators lacked the technical expertise to review Whitehurst's allegations concerning Rudolph's casework, so the Laboratory itself conducted yet another review of Rudolph's case files, this time in 1992. James Corby, the chief of the Materials Analysis Unit (MAU), performed that review. Corby analyzed approximately 200 cases and found significant flaws, such as Rudolph's failing to follow his own explosives residue protocol, to form conclusions with a valid scientific basis, and to conduct necessary tests. Corby recommended that Rudolph be disciplined and removed from doing any further explosives work in the Laboratory. Corby's supervisor, Kenneth Nimmich (chief of the Scientific Analysis Section (SAS) of the Laboratory), then directed that Corby, Martz, and CTU examiner Lynn Lasswell engage in a panel review of Rudolph's files to determine whether any errors needed to be brought to the attention of any prosecutor or defense attorney. Lasswell analyzed 57 of the 200 case files and found serious deficiencies. We found no evidence, however, that Martz conducted any review of Rudolph's files or otherwise assisted in this effort.

Nimmich recommended to John Hicks, the Laboratory Director, that Rudolph be severely reprimanded. Instead, Hicks decided to orally admonish Rudolph. When Hicks delivered that punishment, however, he also gave Rudolph a check for $500, which represented an incentive payment for recent work. The monetary award meant that a decidedly mixed message was sent to Rudolph, who reported to us that he was quite surprised by how leniently he had been handled.

In 1993, Corby continued to express concern over the condition of Rudolph's files and asked James Kearney (who had replaced Nimmich as the head of the SAS) to raise the issue anew with Hicks. Hicks, however, decided that the Rudolph matter had been adequately reviewed and took no action. In 1994, Whitehurst's attorney complained in a letter to the FBI about Rudolph. The FBI Office of General Counsel (FBI OGC) conducted an investigation, determining that Rudolph's files were sloppy and that his [Rudolph's] conclusions are not supported by appropriate documentation. The FBI OGC recommended a comprehensive review, a recommendation not welcomed by the Laboratory Division.

A year after that recommendation was made, in June 1995, Corby was directed to review all cases in which Rudolph had worked as an examiner. Corby completed his review before the end of that year, and found that nearly one-quarter of Rudolph's files did not meet the administrative or technical guidelines at the time the cases were worked. (Emphasis in original.) Rudolph wrote a 200-page response in which he took issue with many of Corby's conclusions. We did not attempt to replicate Corby's work, but our review convinced us that his findings were generally correct.

Although our investigation did not reveal intentional misrepresentations by Rudolph, we did find serious performance deficiencies in his work. As the foregoing discussion of management efforts reveals, it took FBI management nearly six years to perform the type of comprehensive review of Rudolph's files that should have occurred in 1989 after Rudolph's performance in the Psinakis case was so sharply criticized by the AUSA who handled that case. Former Director Hicks was especially remiss for failing to respond adequately to the mounting concerns about Rudolph's competence. CTU Chief Martz was derelict in his technical review and misleading memorandum in 1989. The 1992 review largely failed as an effort to ascertain fully the true extent of the deficiencies in Rudolph's files. Had Laboratory managers performed responsibly, the Rudolph matter might have been appropriately resolved much earlier than 1995. Instead, the Rudolph problem continued to fester.

In 1989, mail bombs killed U.S. Circuit Judge Robert Vance and a civil rights attorney. A massive investigation ensued, ultimately leading to the indictment and conviction in 1991 of Walter Leroy Moody, Jr. Whitehurst complained to the OIG that J. Thomas Thurman of the Explosives Unit (EU) and Martz of the CTU circumvented Laboratory procedures because Thurman arranged for Martz's unit to analyze material in the mail bombs even though Whitehurst's unit, the Materials Analysis Unit (MAU), was responsible for analyzing explosives residue. Whitehurst also contended that, because Martz failed to follow the protocol for residue analysis developed by the MAU, he reached a flawed opinion in concluding that the mail bombs contained a particular smokeless powder. Whitehurst further alleged that Martz and Thurman fabricated evidence, perjured themselves, and obstructed justice in the case. He also suggested that prosecutors Louis J. Freeh and Howard Shapiro, at that time the AUSAs who tried the case, may have committed misconduct by offering the testimony of Martz and Thurman.

We found no evidence to support Whitehurst's charges that Thurman and Martz perjured themselves, fabricated evidence, obstructed justice, or violated any FBI policies or procedures in the case. We did not find any evidence of prosecutorial misconduct. In our investigation of this matter, we also reviewed the analytical work of Robert Webb, an examiner in the MAU who analyzed certain tape, paint, sealant, and glue, and whose conclusions were reported in Thurman's testimony. Although Whitehurst had made no allegations against Webb, we found that Webb stated certain conclusions about his work more strongly than were warranted by the results of his examinations. We found that Webb did not fabricate evidence or intentionally bias his conclusions.

Although we did not find the kinds of misconduct alleged by Whitehurst in this matter, our investigation of this case found ways in which Laboratory practices and procedures could have been improved. Those included: (1) establishment of clear guidelines stating the respective responsibilities of different units with regard to explosives residue analysis; (2) clearer guidance as to the proper scope of the testimony by examiners other than those who conducted the underlying analytical tests; (3) an improved record retention and retrieval system; (4) written and validated protocols for standardized procedures; and (5) contemporaneous peer review to ensure that conclusions are properly supported by analysis and data.

After the bombing of the World Trade Center on February 26, 1993, law enforcement authorities investigated and apprehended several suspects, which led to convictions in two trials: one beginning in 1993, Salameh, which dealt primarily with the bombing, and the other in 1995, Rahman, a broader case that included evidence of the bombing. Prior to the Salameh trial, Whitehurst complained within the Laboratory about the scientific work in several respects, all of which were ultimately resolved to his satisfaction prior to the first trial. In January 1996, however, Whitehurst submitted to the OIG an 80-page critique of the Salameh testimony of David Williams, an examiner in the Explosives Unit (EU). Among the many allegations framed by Whitehurst, he specifically accused Williams of misrepresenting the truth, testifying outside his area of expertise, and presenting testimony that was biased in favor of the prosecution. We concluded that Williams gave inaccurate and incomplete testimony and testified to invalid opinions that appeared tailored to the most incriminating result. We did not substantiate Whitehurst's many other allegations.

Williams testified in the Salameh trial as an explosives expert, and his testimony was potentially significant. He opined (1) that the defendants had the capacity to manufacture about 1200 pounds of the explosive urea nitrate, an explosive rarely used for criminal purposes, and (2) that the main explosive (main charge ) used in the World Trade Center bomb consisted of about the same amount (1200 pounds) of the same explosive (urea nitrate). Normally, the way a crime laboratory determines the main charge of an exploded bomb is by finding unconsumed particles or distinctive byproducts of the explosive among the debris. The search for such residues is made by a forensic chemist. The FBI chemists specializing in the examination of explosives residue, however, did not find any residue identifying the explosive at the World Trade Center. Thus, the normal way of scientifically determining the main charge was unavailable. Williams' testimony filled that scientific void.

Williams' opinions that the defendants had the capacity to manufacture about 1200 pounds of urea nitrate and that about 1200 pounds of urea nitrate was used in the bombing were deeply flawed. As explained in detail in the Report, his testimony about the defendants' capacity exceeded his expertise, was unscientific and speculative, was based on improper non-scientific grounds, and appeared to be tailored to correspond with his estimate of the amount of explosive used in the bombing. His opinions about the explosive used in the bombing were based on an invalid inference concerning the velocity of detonation (VOD) of the main charge, an incomplete statement of the VOD of urea nitrate, invalid and misleading statements about the type of explosives that could have been used, and speculation beyond his scientific expertise that appeared to be tailored to the most incriminating result.

Ultimately, Williams conceded during our investigation that he had no basis from the crime scene for determining the type of explosive used, acknowledging that based on the crime scene the main charge could have been anything. That opinion differs substantially from the opinions he rendered in the Salameh trial that narrowed the category of possible explosives and ultimately identified the main charge as urea nitrate. During the Salameh trial, Williams testified that he was a scientist ; the prosecutors referred to him as an explosive expert witness. In contrast, Williams' identification of urea nitrate was based not on science but on speculation based on evidence linking the defendants to that explosive.

Additionally, we concluded that Williams gave inaccurate testimony regarding his role -- and the formulas used -- in the FBI's manufacture of urea nitrate, and that his testimony concerning his attempt to modify one of Whitehurst's dictations was misleading.

The Report also details many other allegations made by Whitehurst, which we found to be unsubstantiated. We also concluded that the World Trade Center case exemplifies the need for persons within the EU to have scientific expertise, examiners to understand the distinctions between their role as forensic science experts and the role of a criminal investigator, clear guidelines about matters within the expertise of an EU examiner when testifying, and proper documentation of case work.

The Avianca case involved the midair explosion aboard Avianca Airlines Flight 203 shortly after its takeoff from Bogota, Colombia, on November 27, 1989. Everyone onboard, including two Americans, were killed in the crash. Agent Richard Hahn, at that time an examiner in the EU, was assigned to the team of Americans sent to Colombia to assist with the investigation. Hahn collected evidence at the crime scene, examined evidence, and prepared a final report. He also testified both in the first trial in New York, which ended in a mistrial, and the second trial, which resulted in the 1994 conviction of Dandeny Munoz-Mosquera (Munoz).

In 1990 Whitehurst conducted chemical analysis of evidence found at the scene, and his findings were part of Hahn's final report for the Laboratory. After the Munoz trials, Whitehurst alleged that Hahn fabricated evidence, committed perjury, and testified outside his area of expertise in those trials.

Whitehurst's first disagreement with Hahn's testimony concerned the type of explosive used in the blast. Hahn testified in both trials that a high velocity explosive was used in the bombing, based on his observation of indentations on the fuselage known as pitting and cratering, a phenomenon in which an explosion causes small indentations on metal surfaces. We concluded that Hahn's correlation of the pitting and cratering to a high velocity explosive within a narrow range of velocity of detonation was scientifically unsound and not justified by his experience. Moreover, in light of scientific literature Whitehurst submitted to Hahn before the second trial, Hahn erred by not inquiring about the validity of the theory upon which he based his testimony concerning pitting and cratering.

Next, Whitehurst alleged that Hahn gave inappropriate testimony regarding Whitehurst's 1990 findings of two explosives (RDX and PETN) in the evidence from the aircraft, because Hahn failed to mention the conclusions set forth in a memorandum written by Whitehurst in 1994. That memorandum, written on the same day Hahn testified in the first trial, addressed whether the FBI could scientifically disprove a story advanced by someone in Colombia (the Confessor ) who confessed to the Avianca bombing and claimed that the defendant was not involved. We found that Hahn's testimony in the first trial was unobjectionable in that respect (since he was unaware of the memorandum) but that his testimony in the second was incomplete for having failed to take into account certain aspects of the analysis advanced by Whitehurst in the memorandum. We further concluded that SAS Chief Kearney contributed to Hahn's incomplete testimony by not properly resolving the issues raised in Whitehurst's memorandum.

Whitehurst's memorandum was a deeply flawed document, however, because it: (1) reached an invalid conclusion (from Whitehurst's failure properly to review his own laboratory work) about whether he could scientifically exclude the explosive the Confessor said was used; (2) misstated a conversation he had had with Hahn on a material point; (3) rendered a misleading and overstated opinion suggesting that the data was consistent with a potential defense; and (4) improperly raised questions about whether contamination may have accounted for Whitehurst's original scientific findings.

Finally, Hahn testified to a theory that a fuel-air explosion followed the initial blast and that certain of the passengers' injuries were indicative of such an explosion. That testimony was flawed and exceeded Hahn's expertise.

The Avianca case was an unfortunate instance in which communication broke down between examiners and supervisors in the Laboratory, and in which the EU examiner testified to opinions that were not justified by his experience or the applicable science or that exceeded his expertise. It was not, as Whitehurst alleges, an illustration of a Laboratory examiner committing perjury or fabricating evidence. And indeed, Whitehurst's own conduct in this matter, especially his 1994 memorandum, was seriously flawed.

To address the defense's contention that the police had planted blood at the crime scene and on socks found in the defendant's residence, the prosecutors in the O.J. Simpson case asked the FBI Laboratory to determine whether the blood preservative EDTA was present in those blood stains. CTU Chief Roger Martz and several research chemists at the FBI Forensic Science Research Unit (FSRU) at Quantico worked to develop a method for identifying EDTA in blood. After Martz testified in the Simpson trial, Whitehurst alleged that scientists at the FSRU had commented that Martz had committed perjury, misled the jury concerning the validation studies conducted by the FSRU scientists, misled the defense by stating that all digital data from the analysis of the evidence had been erased, and generally testified in an arrogant manner.

We found no basis to conclude that Martz committed perjury or any corroboration that FSRU scientists had made such allegations. Nor did we find that Martz improperly erased digital data. Martz was unfairly criticized by the defense for not conducting certain tests. We did not criticize Martz for the substance of the analytical work performed by him and the FSRU chemists, but rather for his deficient record-keeping and note-taking and for the manner in which Martz testified. That testimony ill served the FBI because it conveyed a lack of preparation, an inadequate level of training in toxicological issues, and deficient knowledge about other scientific matters that should be within the expertise of a chief of a unit handling chemical and toxicological analyses in the Laboratory.

Not long after the EU completed its report on the Oklahoma City bombing, Whitehurst wrote a 30-page letter to the OIG criticizing David Williams, the EU examiner responsible for the report. We concluded that many of the same errors committed by Williams in the World Trade Center case were repeated in the Oklahoma City case -- principally, that Williams based some of his conclusions not on a valid scientific analysis but on speculation from the evidence associated with the defendants.

Williams' September 5, 1995, report contained several serious flaws. Just as he had done in the World Trade Center case, he offered an opinion about the velocity of detonation (VOD) of the main charge that was unjustified. His statement about the VOD of an ammonium nitrate fuel oil (ANFO) explosive -- the explosive allegedly used in the bombing -- was incomplete. His categorical identification of the main charge as ANFO was inappropriate based on the scientific evidence available to him. Here, Williams did not draw a valid scientific conclusion but rather speculated from the fact that one of the defendants purchased ANFO components. His estimate of the weight of the main charge was too specific, and again was based in part on the improper, non-scientific ground of what a defendant had allegedly purchased. In other respects as well, his work was flawed and lacked a scientific foundation. The errors he made were all tilted in such a way as to incriminate the defendants. We concluded that Williams failed to present an objective, unbiased, and competent report.

Williams' supervisor, J. Thomas Thurman, did not properly review Williams' report. Thurman left too much discretion to Williams to include certain opinions, and Thurman allowed certain conclusions to stand even though he told us that he now does not agree with them and cannot justify them, and the conclusions are unsupported in the body of the report.

All cases handled by the Laboratory deserve professional, diligent treatment. Williams' and Thurman's performances in the Oklahoma City case -- a prosecution of enormous national significance -- merit special censure.

In the course of providing more than 1000 pages of written allegations to the OIG, Whitehurst has also alleged wrongdoing in a range of other cases also addressed in our Report. In none of those cases did we find Whitehurst's allegations of intentional misconduct to be borne out by facts, even when those allegations concerned Laboratory personnel who are sharply criticized in the Report. In investigating those allegations, however, we found instances in which general practices and procedures could be improved. Those more general recommendations are set forth later in this Summary.

In the following cases, our findings and conclusions are set out in detail in the report and we will not repeat the conclusions in this Summary:

• Yu Kikumura, a 1988 prosecution of a member of the Japanese Red Army terrorist faction;

• a Laboratory report analyzing two pipe bombs found in fuel storage tanks at a marine terminal in Norfolk, Virginia, in 1991;

• analytical work conducted in connection with the disappearance of a young girl named Melissa Brannen in 1989;
• testimony and analytical work in the Italian prosecution of the murderers of Paolo Borsellino, who was killed in a car bombing in Sicily in 1992; the 1994 prosecution of the person charged with the attempted murder of Miami criminal defense attorney Gino Negretti;

• work conducted by the Laboratory after James Conlon, a hydraulic crane operator, died in an explosion while working at a scrap metal yard in New Jersey in 1992;

• the analysis of smokeless powder found in a pipe bomb sent to U.S. District Judge John Shaw in 1995;

• a Laboratory report in which David Williams offered an expert opinion about the main charge in an improvised explosive device in connection with an investigation of the Ghost Shadow Gang of New York; and

• a 1994 article describing fourteen explosive devices thought to be associated with the so-called Unabomber.

Four other matters are also addressed in this section of the Report:

A recurring theme in Whitehurst's complaints and allegations to the OIG has been that the FBI retaliated against him for raising concerns about the FBI Laboratory to the FBI and others. Retaliation is a difficult issue to investigate, because it rests on the motivations of persons taking actions with respect to the complainant. Neutral explanations may sometimes mask an unstated intent to take harmful actions. Some of the allegations in lawsuits filed by Whitehurst against the FBI and the Department of Justice involve actions taken after the OIG launched this investigation. We did not attempt to assess whether recent actions taken by the FBI -- such as placing Whitehurst on administrative leave with pay after the OIG draft report was issued -- constituted acts of retaliation. Rather, our focus was on retaliatory conduct Whitehurst alleged was directed at him before November 1995. With respect to all but one of Whitehurst's contentions, we concluded that the evidence did not substantiate his allegations of retaliation because we discerned no retaliatory purpose behind the FBI's decisions that he questioned. As for the remaining contention, we were unable to complete our investigation due to Whitehurst's decision not to provide a release form that would have permitted key personnel to speak to us about medically sensitive information regarding Whitehurst.

Whitehurst claimed that he was retaliated against for accusing Terry Rudolph of misconduct in the Psinakis case. After he criticized Rudolph, Whitehurst was suspended without pay for seven days and placed on probation for six months. We did not substantiate Whitehurst's claim. FBI management had reason to criticize Whitehurst's actions in the Psinakis case because he erred in making his concerns known only to the defense attorneys, without first discussing them with the prosecutor, case agent, or his supervisors. The evidence further showed that the FBI's internal discipline unit imposed the suspension despite opposition from Laboratory managers, who recommended the least severe form of discipline possible for Whitehurst. The disparity in treatment between Whitehurst and Rudolph appeared to reflect a failure by management adequately to appreciate the seriousness of Rudolph's conduct rather than an attempt to retaliate against Whitehurst.

Whitehurst also contended that FBI OPR ignored and covered up his allegations that personnel in the Criminal Investigative Division were unlawfully using computer software and that an agent assaulted Whitehurst's wife, who also works at the FBI. Although the evidence showed that the OPR investigation was not as thorough as it should have been, we did not substantiate charges of a coverup. Indeed, Mrs. Whitehurst herself told the OPR investigator that she did not suffer any retribution or continuing harm, although she did feel threatened by the agent at the time of the incident.

Whitehurst next maintained that FBI OPR improperly initiated an investigation into his disclosure of information to the Senate Judiciary Committee. FBI OPR investigated the disclosures, which were admitted by Whitehurst, because of concerns that confidential FBI records had been disclosed to unauthorized persons. When the Judiciary Committee refused to disclose Whitehurst's letters on the ground of protecting confidentiality, FBI OPR closed its investigation and no administrative action was taken against Whitehurst. We found no retaliatory purpose in the actions taken by FBI OPR with respect to this allegation.

In addition, Whitehurst alleged that FBI OPR improperly disclosed derogatory information about him to prosecutors in the World Trade Center and O.J. Simpson cases. After reviewing the disclosures of materials made by the FBI in those cases and interviewing the relevant FBI and U.S. Attorney personnel, we concluded that the FBI did not improperly disclose derogatory information about Whitehurst in those cases, but rather attempted to provide appropriate material regarding witness credibility.

In May 1994, the FBI reassigned Whitehurst from the explosives residue program to be an analyst of paints and polymers. Whitehurst alleged that this reassignment was in retaliation for reporting misconduct in the Laboratory and especially in the Explosives Unit. The Chief of the Scientific Analysis Section, James Kearney, made the decision to transfer the explosives residue program from the Materials Analysis Unit (MAU) to the Chemistry-Toxicology Unit (CTU). He gave two reasons for that move. One was to more closely balance the responsibilities and staffing of the CTU and MAU after a reorganization. A second was to place the explosives residue analysis program under a single unit chief; before that time responsibilities had been divided between the CTU and MAU. Although there was internal opposition to the transfer in responsibilities on the ground that CTU Chief Martz lacked the expertise to supervise the program, we found no evidence of a retaliatory purpose in the transfer of the explosives residue program from the MAU to the CTU.

Similarly, Kearney explained that the reason he moved Whitehurst out of the explosives residue program was because of Whitehurst's poor working relationship with EU and other personnel. Whitehurst acknowledged tension between himself and the EU examiners. MAU Chief Corby also noted that transferring Whitehurst to the CTU with the explosives residue program would have been problematic because of friction between Whitehurst and Martz. Thus, substantial credible evidence showed that the decision to move Whitehurst out of the explosives residue program was not made for a retaliatory purpose.

We also investigated other information proffered by Whitehurst in support of his retaliation claim, but we did not find the anecdotes he supplied to be sufficient to support his claim that an atmosphere of retaliation existed in the Laboratory.

Finally, Whitehurst alleged that in 1993, the FBI ordered him to undergo psychiatric evaluation and therapy in retaliation for his raising various complaints against the FBI Laboratory. We concluded that the Laboratory personnel did not act with a retaliatory purpose in referring the matter to the FBI Health Care Program Unit (HCPU) and the FBI Employee Assistance Program (EAP). However, because Whitehurst did not provide the necessary medical release forms to allow us to interview key personnel with the HCPU, EAP, and Personnel Section, we could not reach any definite conclusions concerning the motives of any such personnel in referring Whitehurst to psychotherapy.

Because Whitehurst made allegations of misconduct against a large number of persons in a large number of cases, we detailed in a separate part of the Report our findings and conclusions about each person against whom allegations were made or when our findings led us to conclude that the conduct of a person merited critical comment. In some instances, we made recommendations that persons be transferred from the positions they held prior to completion of our draft report, they be given special supervision, and/or their Laboratory reports be reviewed because of concerns we identified in their work.

CTU Chief Roger Martz lacks the judgment and credibility to perform in a supervisory role within the Laboratory. If Martz continues to work as an examiner, we suggest that he be supervised by a scientist qualified to review his work substantively and that he be counseled on the appropriate manner for testifying about forensic work. We further recommended that another qualified examiner review any analytical work by Martz that is to be used as a basis for future testimony.

EU Chief J. Thomas Thurman deserves special censure for his inadequate supervisory review of Williams' report in the Oklahoma City bombing case. Because we concluded that all examiners in the EU, including the Chief, should have a scientific background, we recommended that he be reassigned outside the Laboratory when that restructuring occurs.

EU examiner David Williams should be reassigned outside the Laboratory. Although we did not find that Williams had perjured himself in the World Trade Center case, his work in that case and in the Oklahoma City investigation demonstrate that he lacks the objectivity, judgment, and scientific knowledge that should be possessed by a Laboratory examiner.

EU examiner Wallace Higgins should be reassigned outside the FBI Laboratory when the restructuring of the EU occurs. In the interim, while Higgins remains in the EU, the SAS Chief should counsel Higgins on the proper preparation of reports and monitor his work. A qualified explosives examiner also should review any reports prepared by Higgins.

Richard Hahn no longer works in the Laboratory. If in the future he is called upon to testify about his work as an examiner, we recommended that he be specially counseled about the importance of not testifying on matters beyond his expertise and that his testimony should be reviewed by qualified examiners to ensure that it is appropriately limited.

Michael Malone no longer works in the Laboratory, having been transferred from the Hairs and Fibers Unit in 1994. We concluded that Malone testified falsely and outside his expertise in the Hastings matter. We recommended that the FBI assess what discipline is appropriate and monitor future expert testimony to assure that it is accurate and limited to matters within his knowledge and competence.

Robert Webb also has been transferred out of the Laboratory. We found that Webb's report in the VANPAC case stated conclusions more strongly than were justified by the results of his examinations and the background data. We recommended that another qualified examiner review Webb's analytical work in the event it is to be used as the basis for future testimony.

J. Christopher Ronay was the EU Chief from 1987 through October 1994, when many of the problems raised by Whitehurst first surfaced. We found that he exhibited poor judgment as a manager in approving EU reports. Because he is retired, we did not recommend any action concerning Ronay.

Terry Rudolph is now retired from the FBI. Although we were told that he worked as a consultant for a period of time after his retirement, we recommended that he not be employed in any capacity by the FBI in the future. We further recommended that a notation referring the findings of this Report be placed in each of his case files.

With respect to managers in the FBI Laboratory, we found important instances of deficiencies and failures to handle situations in an expeditious, thorough, and effective manner. A significant example of that finding occurred at the very outset of Whitehurst's criticisms and the weak response of Laboratory management to AUSA Burch's letter to the Laboratory Director regarding deficiencies in Rudolph's performance in Psinakis in 1989. More recent examples involved significant problems in explosives-related cases. Management lapses included failures to supervise appropriately the drafting of Laboratory reports in the EU, to evaluate the competence of examiners, and to establish a climate in which meaningful peer reviews and the professional resolution of scientific disagreements were the norm. The Report singles out for criticism Charles Calfee, Kenneth Nimmich, James Kearney, and John Hicks, all of whom are now retired from the FBI. We did not, however, substantiate criticisms of Alan Robillard, who transferred out of the Laboratory in 1994.

Our investigation exonerated a number of persons against whom allegations of misconduct were made. Those persons included: Roger Asbury, Edward Bender, Louis J. Freeh, Donald Haldiman, Ronald Kelly, Lynn Lasswell, Richard Laycock, Thomas Mohnal, Bruce McCord, Mark Olson, and Howard Shapiro. Furthermore, we did not substantiate Whitehurst's allegations against Alan Jordan, and although we did not substantiate allegations against Robert Heckman in the Borsellino matter, we did find reason to criticize Heckman for his work in the Conlon case.

Finally, the Report discusses Frederic Whitehurst, the complex person whose expression of concern about problems in the Laboratory sparked this investigation. He is an experienced scientist who identified significant problems in certain cases and in certain practices within the Laboratory. He also accused many of his colleagues of perjury, fabrication of evidence, and conspiracy. Those allegations were not supported by the facts uncovered in the investigation. Any decisions about Whitehurst must involve a careful weighing of the substantial contribution he made in bringing to light issues in the Laboratory that needed to be addressed against the considerable harm he has caused to the reputations of innocent persons and the fact that his frequently overstated and incendiary way of criticizing Laboratory personnel will make it extremely difficult if not impossible for him to work effectively within the Laboratory. Our own view is that Whitehurst lacks the judgment and common sense necessary for a forensic examiner, notwithstanding his own stated commitment to objective and valid scientific analysis.

Although we made recommendations with respect to individuals, we perceived our principal mission to be to make systemic recommendations on Laboratory practices and procedures, the full implementation of which would help the FBI Laboratory avoid in the future the problems we encountered in the matters we investigated. The recommendations as to individuals are, however, a necessary concomitant to achieving the type of organizational and cultural changes that should be undertaken by the FBI. The FBI has recognized in the immediate past that some aspects of its policies and procedures demand change, and upper management has taken steps to put new policies into effect. It is not clear from those policy changes that Laboratory top management has acknowledged that appropriate assessments of personnel are also required. Steps must be taken to provide personnel with the appropriate training, background, and commitment to quality that is required in a first-class forensic laboratory.

In its response to our draft report, the FBI concurred with nearly all of the OIG's systemic recommendations, even though it frequently disagreed with how we applied those general principles in assessing individual performances. Our emphasis in the previous section on individuals, therefore, should also be read in light of the importance

of investing personnel in the Laboratory with the appropriate skills and motivations to change old practices, as well as of underscoring the need for personal accountability as those changes are made. Thus, although virtually all of the following general recommendations are recognized within the FBI as appropriate and have been accepted as valid, the best proof of acceptance will not be in the articulation of new practices, but in their complete implementation in the coming years.

Our first recommendation was one already accepted by the FBI -- that the Laboratory should pursue accreditation by the American Society of Crime Laboratory Directors/Laboratory Accreditation Board (ASCLD/LAB). In 1994, Director Freeh announced that the Laboratory would pursue accreditation at the earliest possible time, and the FBI's response to the OIG draft report acknowledged that the Laboratory could and should have sought ASCLD/LAB accreditation a decade ago. We commend the FBI for now making such accreditation a top priority for the Laboratory. The criteria imposed in the accreditation process should promote valuable and productive interchanges with other laboratories to change some of the insular and parochial views we encountered in the EU, CTU, and MAU, as well as to assist the Laboratory in modernizing policies and practices. Accreditation is not a panacea, nor is the absence of accreditation an indictment of all tests performed in the Laboratory. But the process of undergoing accreditation should enhance quality performance.

Second, we recommended that the Explosives Unit be restructured and its mission clarified. One existing problem in the EU is that its personnel are not forensic scientists. We recommended (and the FBI agreed) that examiners in the EU have scientific background in pertinent disciplines such as chemistry, metallurgy, or engineering, as well as technical training in the assembly, deactivation, and use of explosive devices. Although EU examiners should be available to consult at crime scenes, primary responsibility for conducting investigations and directing crime scene management functions should rest with components of the FBI outside the Scientific Analysis Section. (The recommendation concerning the proper role of EU examiners at the crime scene was the only recommendation discussed in this Section with which the FBI disagreed.)

Third, the Laboratory should abolish its current distinction between principal and auxiliary examiners, in which the auxiliary examiners' reports are combined into

a single report by a principal examiner. In cases in which more than one examiner is called upon to evaluate evidence, we suggested that a coordinating examiner assume the role of ensuring that the correct units of the Laboratory have been enlisted to work on the case and that the reports generated by those units are accurately included in the final set of reports. Although we were told that an unwritten policy (prior to a formal written directive in September 1994) had long been that auxiliary examiner reports were incorporated verbatim, we found numerous instances in which that policy was not followed.

Fourth, we recommended that, instead of one report emanating from the Laboratory with analytical results reflected in the body of that report without attribution to individual examiners, each examiner who performs work should prepare and sign a separate report, even if such individual reports are ultimately collected together as the unified report of the Laboratory as a whole.

Fifth, analytical reports should also be substantively reviewed by the unit chief or another examiner (if the unit chief lacks the requisite expertise or has performed the analysis) before they are released in final reports. Forensic science is sufficiently complex that such substantive review need not always follow hierarchical lines within the Laboratory management structure. A junior examiner who is qualified in the area should be capable of substantively reviewing a unit chief's analysis. Our central point is that peer review by qualified personnel is an essential aspect of a high-performing forensic science laboratory. The Rudolph matter, certain conclusions in the Oklahoma City report, and other cases demonstrate the importance of vigorous, substantive peer review.

Sixth, reports must be supported by adequate case files. The Rudolph files and some of Martz's work underscore the importance of case files containing all of the documentation necessary for another appropriately qualified examiner to be able to understand and replicate the examiner's data and analysis. We encountered the problem of incomplete or missing documentation in many case files. Accreditation will require the Laboratory to maintain a rigorous system of case filing, which has not existed in the past.

Seventh, not only must the files contain all relevant documentation of results, but the records themselves must be maintained so as to facilitate ready retrieval. We suggested that the Laboratory keep its own files rather than integrating Laboratory files with the Bureau's general case filing system.

Eighth, we recommended that the Scientific Analysis Section of the Laboratory Division develop and implement a coordinated training program for examiners. Training has been conducted at the unit level, and has developed in an ad hoc manner. As suggested in the ASCLD/LAB accreditation process, a unified curriculum for common issues and moot courts for testimony would be helpful. At the unit level, managers should clearly articulate training criteria and document completion of curricula.

Ninth, the FBI should develop a uniform program for training examiners with respect to court testimony and monitoring such testimony. We found the problem of examiners testifying to matters beyond their expertise or in ways that were unprofessional in Hahn's testimony in the Avianca case, Williams' testimony in the World Trade Center case, and Martz' testimony in Trepal and Simpson.

Our tenth and eleventh recommendations addressed the development of written protocols generally for the scientific procedures utilized. For the analysis performed in the FBI Laboratory to have wide-ranging credibility in courts and in the forensic science community, examiners must strictly adhere to established protocols for the analysis of evidence or document the reasons for departing from them. The same is true for the handling of evidence and the adoption of measures to prevent and detect contamination.

Finally, the role of management is critical to achieving the types of reforms needed in the Laboratory. As we have noted, before and during our investigation Laboratory managers have begun the process of implementing many of the recommendations we noted above, as the process of preparing for accreditation continues. Those reforms must be substantive and should be structured to address the fundamental issues raised in our Report.

The FBI's cooperation with the OIG investigation and acceptance of our systemic recommendations should be lauded. The process of managing necessary changes will be challenging in an environment in which scientific knowledge is expanding and forensic science is increasingly under scrutiny. We welcome the FBI's suggestion of our continued involvement in oversight to assist in ensuring that needed reforms are fully implemented. We will seek to perform that function in a manner consistent with the Laboratory's expeditious efforts to obtain ASCLD/LAB accreditation and its ongoing development of first-class examiners and standards. Although we have rejected the most inflammatory allegations made by Whitehurst, the FBI Laboratory must fully acknowledge past problems that have been identified as it continues its pursuit of excellence in forensic science.

Michael R. Bromwich

Inspector General

#####

The FBI Laboratory: An Investigation into Laboratory Practices and Alleged Misconduct in Explosives-Related and Other Cases (April,1997) Office of the Inspector General, Department of Justice