Return to the USDOJ/OIG Home Page
Return to the Table of Contents

An Investigation of the Belated Production of Documents
in the Oklahoma City Bombing Case

March 19, 2002
Office of the Inspector General


The bombing of the Alfred P. Murrah Federal Building in Oklahoma City, Oklahoma, on April 19, 1995, was, at the time, the most significant act of terrorism that had ever taken place in the United States. Government agencies, led by the Federal Bureau of Investigation (FBI), immediately began an extensive investigation to identify and prosecute the culprits. The investigation, known as OKBOMB, was run by a Task Force that consisted primarily of FBI investigators and support personnel and Department of Justice prosecutors. Within a few months, three individuals - Timothy McVeigh, Terry Nichols, and Michael Fortier - were indicted for crimes relating to the bombing. McVeigh and Nichols were convicted after trials, and Fortier pled guilty as part of a plea agreement with the government. McVeigh, who had devised the plot to bomb the Murrah Building and had planted the bomb, was sentenced to death.

On May 8, 2001, one week before McVeigh's scheduled execution date, the Department of Justice and the FBI revealed to McVeigh's and Nichols' attorneys that over 700 investigative documents had not been disclosed to the defendants before their trials. The government acknowledged that it had violated a discovery order in the case, and the Attorney General stayed McVeigh's execution for one month in order to resolve the legal issues arising from the belated disclosure.

Following the public revelation of the problem, and after finding and releasing more than 300 additional OKBOMB documents to the defense, the FBI came under severe criticism for its handling of the OKBOMB documents. Allegations were made that FBI personnel intentionally failed to disclose exculpatory information to the defense.

When the problem of the belated documents was first disclosed, the Attorney General requested that the Office of the Inspector General (OIG) investigate the circumstances leading to the untimely production. This report details the results of the OIG's investigation.

I. The OIG Investigation, Scope of the Report, and Conclusions

To conduct the investigation the OIG assembled a team of five attorneys, two special agents, two auditors, a paralegal, and support personnel. We conducted approximately 200 interviews of current and former FBI and Department of Justice officials. OIG investigators traveled to 13 FBI field offices to conduct interviews, view the physical premises, and review the offices' processes for handling documents. These 13 offices accounted for over 50 percent of the belated documents. The OIG also surveyed the 43 other FBI field offices for explanations about how they handled OKBOMB documents and why they failed to provide the material deemed to be belated to the Task Force during the OKBOMB investigation.

In general, our investigation sought to answer the following questions:

We investigated why more than 1,000 discoverable items had not been disclosed to the defense before McVeigh's and Nichols' trials. Although the media reported that some FBI officials were blaming the FBI's computer system for causing the problem, we did not conclude that the computer system was the cause of the belated production of documents. The FBI's computer system is antiquated and in need of substantial improvement, but we found that human error, not the inadequate computer system, was the chief cause of the failure to provide the defense with these items. The failures that we observed stemmed from individual mistakes, the FBI's complex document processing systems, inconsistent interpretations of FBI policies and procedures, agents' failures to follow FBI policies, agents' lack of understanding of the unusual discovery agreement in this case, and the tremendous volume of material being processed within a short period of time.

Shortly after the documents problem became public, FBI Headquarters asserted that the primary reason the defendants did not receive materials to which they were entitled was because the FBI field offices had failed to send the material to the OKBOMB Task Force. We found in many instances that personnel in the field offices did fail to send OKBOMB material to the Task Force. We also found that in some instances these errors could have been caught and remedied if the field offices had conducted appropriate searches for OKBOMB material in their files as requested by the Task Force. Significantly, however, we also found that the Task Force lost or mishandled some of the material sent by the field offices. Therefore, both the field offices and the OKBOMB Task Force share responsibility for the incomplete document production.

The defense attorneys alleged that the government had intentionally failed to disclose information exculpatory to the defendants. We concluded that the evidence does not support such claims. As part of our investigation into this issue, the OIG examined the evidence cited by the defense as supporting its claims; however, this evidence revealed primarily unintentional errors, not any intentional effort by government employees to withhold discoverable documents from the defense. The direct and circumstantial evidence supports the conclusion that the government did not willfully withhold material known to be discoverable.

In addition to determining the reasons why the documents had not been timely disclosed to the defense, we also investigated the conduct of FBI personnel in 2001. The evidence shows that the belated documents problem was revealed to defense counsel and to the public only one week before McVeigh's scheduled execution because several FBI supervisors did not effectively address the situation when the issue was discovered in January 2001. The supervisors failed to aggressively manage the document review process and failed to set any deadlines for completing the project. The supervisors also failed to timely notify OKBOMB prosecutors, FBI supervisors, and Department of Justice officials. This belated disclosure of the problem only one week before McVeigh's execution created a firestorm of criticism, made it appear that the FBI was hiding documents until the last moment, and led to the last-minute rescheduling of the execution date.

In addition, we found that FBI Headquarters failed to effectively manage the aftermath when the problem became public on May 10, 2001. Headquarters was too quick to assign blame without sufficient facts, and it did not adequately communicate with its field managers, many of whom found out about the issue and that they were being blamed for the problem from press reports. Headquarters also issued instructions to the field that were not well considered, contributing to a sense of confusion throughout the agency.

The belated documents issue provides an opportunity to shine light on several of the FBI's long-standing problems: antiquated and inefficient computer systems; inattention to information management; and inadequate quality control systems. Although the belated documents issue was presented as a discovery problem, the FBI's troubled information management systems are likely to have a continuing negative impact on its ability to properly investigate crimes. At the end of this report, we set forth several recommendations that address these systemic weaknesses. The recommendations address, among other issues, improved planning for complex, document intensive cases; computer system enhancements; and increased automation training.

II. Background

Immediately following the bombing of the Murrah Building, the FBI and the Department of Justice established a Task Force to investigate the crime. An FBI Inspector in Charge and a Department of Justice prosecutor led the Task Force, which at its peak consisted of over 200 investigators, prosecutors, and support personnel. The Task Force initially was located in Oklahoma City but later moved to Denver, Colorado, when the defendants' trials were moved. In addition to the personnel assigned to the Task Force, thousands of other investigators from the FBI's 56 field offices, its foreign offices, and other law enforcement agencies also participated in the OKBOMB investigation.

The belated documents consist primarily of "FD-302s" and "inserts," forms used by the FBI to record investigative activity such as witness interviews. After FBI agents in the field offices memorialized their investigative activity on the appropriate form, the documents were supposed to be sent to the Task Force. Task Force personnel organized the evidentiary material, entered a brief description of the material into appropriate databases, and filed the hard copies.

In every federal criminal trial, the defendants are entitled to have access to some, but not all, of the prosecution's files. After the OKBOMB defendants were indicted, the government decided to go beyond the discovery rules routinely used in federal criminal trials and agreed to provide the defense with all the FD-302s and inserts. The agreement was not formalized in writing, but we found no dispute about this obligation.

As the OKBOMB investigation progressed in 1995 and 1996, the Task Force realized that it was not receiving all of the documents generated in the field offices. On many occasions, the Task Force sent sternly worded instructions to the field offices that all OKBOMB-related materials were to be sent to the Task Force and directed the field offices to search their offices for OKBOMB materials. At the same time, however, some field offices complained to the Task Force that they were receiving multiple requests from the Task Force for documents that the field offices had sent previously.

III. Chronology of Events Leading to the Discovery of the Belated Documents

We describe in the report the events of 2000 and 2001 that led to the discovery of the belated documents, and we also describe the process that personnel used to determine whether the documents had been disclosed previously to the defense. Essentially, the belated documents were discovered when FBI personnel began an archival process designed to ensure the long-term maintenance of the historically significant OKBOMB materials.

The OKBOMB defendants were tried in separate trials in 1997 in Denver, Colorado, after the trial judge moved the cases from Oklahoma City. Following the trials, the evidence was packed, transferred back to Oklahoma City, and stored in a large warehouse. In early 2000, personnel in the FBI's Oklahoma City Field Office became concerned that the heating and cooling capacity of the warehouse was insufficient to maintain the OKBOMB evidence, and they sought the advice of the FBI's archivist. The archivist agreed that the warehouse was not suitable for long-term storage, and he also agreed to assist in the preservation process.

In December 2000, the archivist sent an electronic communication to the FBI's 56 field offices authorizing them to destroy copies of OKBOMB documents that met specific guidelines the archivist provided. The field offices were to send a list of the remaining OKBOMB materials to the Oklahoma City Field Office. By late January 2001, two field offices sent their OKBOMB files, not just a list, to Oklahoma City. When the files were examined, two Oklahoma City Field Office analysts, both of whom had worked on OKBOMB and one of whom had been involved in the discovery process, immediately became concerned because they found what they believed to be original documents in the field office boxes. Because they did not believe that the field offices should have been in possession of any original OKBOMB documents, the analysts promptly advised an Oklahoma City supervisor and two FBI managers in the Dallas, Texas, Field Office, one of whom had been the OKBOMB Inspector in Charge, that some of the material might not have been disclosed during the discovery process.

On January 30, 2001, the Oklahoma City analysts sent a communication to all FBI field offices instructing the field offices not to destroy OKBOMB documents and to send OKBOMB materials to Oklahoma City. As the material came into Oklahoma City, the two analysts plus a few additional Oklahoma City personnel reviewed every document to determine whether they had been disclosed previously to the defense. As we describe in the report, the databases used by the FBI for the OKBOMB investigation made this process difficult and inexact.

Because the Oklahoma City employees learned that the field offices had not complied completely with the January directive to send all OKBOMB material, they sent yet another directive in March 2001 again requesting that all OKBOMB material be sent to Oklahoma City. During the review process, the two analysts kept their Oklahoma City supervisor and the two Dallas Field Office OKBOMB supervisors informed about the progress of their review. In March 2001, the analysts showed the Dallas supervisors a box of problem documents that they had not been able to find in the databases that listed which documents had been disclosed to the defense. Throughout March and April, the Oklahoma City personnel continued their review. By the end of April 2001, they finished their examination of all the documents sent by the field offices and concluded that over 700 documents had never been disclosed to the defense. Concerned about the approaching execution date, the Oklahoma City personnel sent the documents to Dallas so the OKBOMB Inspector in Charge could determine how to handle the issue.

On May 7 and 8, 2001, the OKBOMB Inspector in Charge notified for the first time FBI Headquarters and an OKBOMB prosecutor that documents that had not been turned over in discovery had been found in FBI files. The prosecutor disclosed 715 documents to the defendants' attorneys on May 9. Yet, the FBI continued searching for and finding additional documents in its field offices. These documents also were reviewed, and by the end of May over 300 more documents were released to the defense. In total, 1,033 documents were provided to the defense.

IV. Causes of the Belated Production of Documents

The OIG found that both the field offices and the OKBOMB Task Force were responsible for documents not being disclosed timely to the defendants. With respect to most of the documents, we were unable to determine definitively the exact cause of the problem because of FBI employees' inability to either know or recall exactly what they did with a document that they handled over six years ago. Nonetheless, we were able to determine a number of factors that likely caused the problem.

First, the FBI's system for handling documents is complex. Many different employees are involved in processing documents, including agents, supervisors, and various administrative personnel. Documents are stored in many different locations, various databases are used to track the documents, and information is placed on different types of forms, which are handled in various ways depending on the type of form.

Second, certain procedural breakdowns added to the complexity of processing the OKBOMB documents. For example, in order to get information to the Task Force as quickly as possible, agents used teletypes (a form of instant communication similar in some ways to a facsimile) to send information. Yet, FBI and OKBOMB procedures required that information be placed in a different format - an FBI FD-302 or an insert. Some of the document-handling problems arose because field offices believed information had been sent to the Task Force (because they had sent a teletype) but, because the Task Force did not disclose teletypes, the information did not ultimately get to the defense.

We also found that some offices purposefully did not send information to the Task Force because an agent or supervisor believed the information was not pertinent to the OKBOMB investigation. The Task Force became aware of this problem and sent instructions to the field that all information relating to the bombing was to be reported to the Task Force regardless of the field office's view of its significance. Nonetheless, despite these instructions, some agents still failed to send documents to the Task Force because they deemed the information to be insignificant. Although these agents erred in making these decisions, we believe they acted from a mistaken view of the discovery requirements, which were substantially broader than in the usual case, not with the intent to hide exculpatory information.

Other factors also played a role. Agents in Resident Agencies (that is, FBI satellite offices) created many of the belated documents. We found that offices and individual employees differed as to whether Resident Agency personnel or personnel in the primary FBI field office were responsible for sending material generated in the Resident Agency to the Task Force. In some cases personnel assumed that someone else had sent the document to the Task Force when in fact no one had.

A version of this same problem - incorrect assumptions about other employees' actions - likely caused documents relating to confidential informants, about 10 percent of the belated documents, to not be disclosed. Documents relating to confidential informants are kept separately from other field office files, different personnel are involved in handling the documents, and the database used to track confidential informant documents is particularly difficult to use effectively.

We found, however, that many of these problems might have been caught if the field offices had responded properly to the Task Force's document requests. In 1995, 1996, and 1997, the Task Force repeatedly requested that the field offices send all OKBOMB materials to it. Indeed, a sternly worded communication from the FBI Director was sent to the heads of all the field offices in November 1996 directing them to ensure that all investigative activity had been documented properly and forwarded to the Task Force. The directive even listed various locations in the field offices to be searched for OKBOMB material. The evidence showed that many field offices did not follow these instructions, however. In 2001 original documents were found in many of the same locations that the field offices had been directed to search.

Although the field offices failed to send all OKBOMB materials to the Task Force, the Task Force also shares responsibility for documents not being disclosed to the defense. We found some of the belated documents in the Task Force files misfiled in subfiles that were not used to compile discovery. In other instances, although we could not find the actual document, the evidence strongly indicated that the field office had sent the document to the Task Force. In those cases, the evidence suggests that Task Force personnel misfiled, mislaid, or lost the document.

The massive quantity of documents coming into the Task Force placed enormous stress on a document-handling process that utilized hundreds of employees (many of whom were on temporary duty), involved multiple databases, and required documents to travel to an assortment of teams for processing. Under these circumstances, it was inevitable that some documents would be mishandled. The problem was compounded, however, by several factors. First, the Task Force did not have a routine policy of checking to ensure that items a field office said were being sent actually arrived at the Task Force. Also, documents easily could be accidentally lost or placed in the wrong filing cabinet drawer at the Task Force, and the error would not be noticed. Task Force supervisors did not recognize the deficiencies in their document processing, and they gave little consideration to whether any measures should be taken to plug the gaps.

In addition, the FBI did not have an effective automated quality control system to help the Task Force track documents when they were generated. Although the FBI assigns a serial number to every document, in 1995 each field office assigned its own set of serial numbers to the documents stored in its files, resulting in duplicate serial numbers. In order to generate a unique number for each document, the OKBOMB Task Force gave an OKBOMB serial number to each document sent by the field once the Task Force had received the document. Consequently, the Task Force could only track documents once they had been received and serialized at the Task Force. The Task Force developed several manual quality control projects to help it identify missing documents, but these were limited, prone to human error, and unable to ensure that all discoverable documents were received from the field offices.

We carefully examined the allegation that the government intentionally withheld exculpatory information from the defense. We concluded that the evidence did not support a finding that government personnel deliberately withheld exculpatory information from the defense. For the reasons stated previously, the evidence showed that mistakes and inadequate document handling systems were the chief cause of the failure to properly disclose documents. In addition, the vast majority of the documents contained information of no significance or information that had been disclosed to the defense in other forms. In a few instances, we did find that agents had made the decision not to send certain items to the OKBOMB Task Force. As previously stated, however, we determined that in those cases the agents mistakenly believed that documents they judged to be non-pertinent to the OKBOMB investigation did not need to be sent to the Task Force. We did not find that the agents intended to hide evidence from the defense.

As part of our investigation regarding intent, we investigated eight belated documents that McVeigh's attorneys asserted were particularly significant and were evidence of the government's intentional misconduct. We interviewed the agents who drafted the documents, their supervisors, and in some cases the administrative personnel who processed the documents. As a result of this review, we did not find evidence showing intentional misconduct. Rather, as with the other belated documents, the evidence indicated breakdowns in the document-handling process due to human error.

V. Analysis of FBI Officials' Actions in 2001

As part of the OIG investigation, we examined and analyzed the actions of FBI officials in reacting to and managing the belated documents problem after it first arose. Three FBI managers were aware of the potential problem as early as January 2001 - Danny Defenbaugh, the Inspector in Charge of the OKBOMB investigation who later became the Special Agent in Charge of the Dallas, Texas, Field Office; OKBOMB Supervisory Special Agent Mark White, who became a Supervisory Special Agent in the Dallas Field Office; and William Teater, a Supervisory Special Agent in the Oklahoma City Field Office.

We concluded that Defenbaugh and White did not effectively manage the review process of the OKBOMB documents. The Oklahoma City personnel who were reviewing the documents kept White, and through him Defenbaugh, informed that they were unable to find evidence establishing that many of the documents they were examining had been disclosed before the defendants' trials. Both White and Defenbaugh traveled to Oklahoma City in March 2001 and examined some of the documents that had been set aside as "problems." Yet, despite McVeigh's fast approaching execution date, neither White nor Defenbaugh set any deadlines or timetables by which the review process was to be concluded, nor did they involve themselves in the process of getting the documents from the field offices to Oklahoma City for review. And they failed to notify the prosecutors or anyone at FBI Headquarters about the potential problem until May 7, approximately one week before the scheduled execution date. We believe their failure to take timely action to resolve, or report, the problem of the belated documents was a significant neglect of their duties, and we recommend that the FBI consider discipline for these failures.

We also concluded that Oklahoma City Supervisory Special Agent Teater did not adequately supervise the document review project. Although Teater kept apprised of the status of the project and periodically asked the Oklahoma City employees whether they needed more resources, he did not consult with either Defenbaugh or White to ensure that the review process was completed expeditiously.

A closer question was the inaction of Oklahoma City Special Agent in Charge Richard Marquise. In March and April 2001, Teater sent Marquise e-mails describing the activities of his squad. The e-mails reference the OKBOMB review project, although not in a particularly detailed or descriptive fashion. We concluded that under the unusual circumstances of this case, Marquise should have questioned Teater further regarding the review project and the meaning of his e-mails.

We also concluded that the reactions of FBI officials at Headquarters following the public disclosure of the belated documents problem contributed to confusion within the agency. Early statements about the cause of the problem incorrectly placed blame for the problem solely on the field offices. Communication with the field offices was deficient and led to some field office Special Agents in Charge learning of the problem from the media. In addition, Headquarters officials gave instructions to the field without a complete understanding of the nature of the problem, thereby resulting in field offices having to complete multiple and duplicative time-consuming searches.

VI. Inadequate Responses by Field Offices to Requests for Information in 2000-2001

During our investigation, we reviewed hundreds of communications between and among FBI Headquarters, the OKBOMB Task Force, and the field offices. We found that a significant number of the communications contained inaccurate information, many of the responses were untimely, and in some cases offices did not respond at all to urgent requests for information. Although these failures did not cause the belated documents problem, they raise serious questions regarding the FBI's attention to detail, managerial accountability, and the reliability of information sent by field offices to Headquarters and to other field offices.

In some instances, we found that FBI personnel responded to the Oklahoma City Field Office's requests for OKBOMB documents in January and March 2001 by making inaccurate entries in the FBI's central database indicating that they had completed the task requested by Oklahoma City. In fact, our investigation showed that the personnel had taken no steps to search for, examine, or send the OKBOMB documents as requested by Oklahoma City. We also found that some field offices reported to Oklahoma City in January 2001 that they had no OKBOMB documents, only to send boxes of documents to Oklahoma City in May 2001.

We also found that some field offices appeared to have "lost" the FBI archivist's December 2000 request for OKBOMB documents and never took any action on it. Additionally, although the January and March 2001 requests from Oklahoma City for OKBOMB documents were marked as requiring "immediate" action, in many instances the field offices took weeks or months to respond.

VII. Destruction of Documents

As described earlier, through an electronic communication dated December 20, 2000, the FBI archivist authorized FBI field offices to destroy copies of OKBOMB documents that remained in their files, if the field office followed guidelines set out in the communication. Our investigation found that 13 field offices destroyed some portion of their OKBOMB files following the archivist's authorization, and two offices had destroyed files even before the archival process had begun. Of the 13, only one reported following the archivist's guidelines regarding destruction.

We attempted to determine whether these field offices had destroyed any FD-302s or inserts - in other words, the type of documents that were covered under the discovery agreement. Nine field offices either acknowledged destroying such documents or could not rule out the possibility that they had been destroyed.

These nine offices insisted, however, that they destroyed only copies of materials that had been sent to the OKBOMB Task Force. While probably true with regard to most of the destroyed documents, it is difficult to verify that the destroyed documents had been sent to the Task Force previously. Furthermore, even if the documents had been sent to the Task Force before the defendants' trials, the information might not have been disclosed to the defense. If the field offices had sent their files to Oklahoma City for analysis rather than destroying them, some of the documents may not have been found in the Oklahoma City databases that listed what documents had been disclosed to the defense. In other words, without the actual documents to compare with the FBI databases, it is difficult to determine whether all the destroyed documents had been disclosed to the defense.

We also discuss in the report the failure of the FBI and the Department of Justice to bring the fact of the document destruction to the attention of the trial court and the defense attorneys in May and June 2001 when McVeigh's stay of execution motion was being litigated. We attributed this failure to a breakdown in communication within the FBI and between FBI officials and prosecutors, the fact that senior FBI officials did not understand the role that copies played in the belated documents problem, and the intense focus on finding the remaining OKBOMB documents in the field offices.

VIII. Recommendations

Although our investigation revealed numerous problems with the FBI's handling of the documents in this case, we believe the failings also need to be placed in context. The OKBOMB Task Force and the field offices were dealing with what, at that time, was the largest criminal investigation ever undertaken by a United States law enforcement agency. The FBI processed millions of documents and items of physical evidence, conducted thousands of interviews, and managed an investigation that involved thousands of investigators and support personnel from the FBI and other agencies. The belated documents problem should not diminish their efforts. Rather, the problems encountered in this case should be used to help ensure that the FBI's record management system is substantially improved so that information can be obtained, stored, and retrieved efficiently and effectively in the future.

To aid in that effort, our report makes a series of systemic recommendations, most of which relate to FBI computer systems and document management. The FBI initiated an automation system in 1995 that could address, in large part, the problem that we saw in OKBOMB - the inability of the case investigators to know what documents have been created by other investigators. The Automated Case Support (ACS) system utilizes "universal serialization" for document management. In any given case, all documents are given serial numbers that follow sequentially regardless of the field office that created the document. Accordingly, case investigators can easily tell if they are missing the hard copy of a particular document because there will be a gap in the sequence.

We found, however, that ACS is so difficult to use that many agents and supervisors have abandoned the effort. As a result, the FBI has both a paper and an electronic information management system in place, neither of which is both reliable and effective.

The FBI is trying to develop upgraded information technology systems as part of a project it calls Trilogy. We did not investigate Trilogy as part of this review and therefore cannot state whether ultimately it will solve the FBI's substantial information management problems. But the success of any system depends on the FBI's full commitment to its use. All FBI employees must be fully trained on the system, and the efficient use of automation must become part of the basic job requirements for all employees, not only administrative support personnel but also agents and managers.

We recommend that the FBI simplify its record keeping process as part of the development of the new information management system. The FBI uses a multitude of forms, and various record keeping procedures are used depending on the form. We believe that some of the forms could be eliminated. In addition, sending documents through an office electronically might help to reduce the number of opportunities for documents to go astray simply by reducing the number of employees and steps involved in processing the documents.

Over the years, the FBI has developed various independent databases to perform tasks that ACS does not perform or does not perform well. The OKBOMB investigators used at least three major information management systems, none of which were linked to each other. We believe that any new automation system must be flexible and expandable so that independent systems can be eliminated.

We make other specific recommendations regarding document tracking and lead management and tracking. We also recommend substantially enhanced computer training and suggest that the FBI should consider making computer usage part of the core skills needed to graduate from the new agents training academy.

As part of our investigation we also interviewed prosecutors and senior investigators who participated in other major FBI cases. We found that there is no formal process in place for learning from these prior experiences. We recommend that the FBI and the Department of Justice initiate a post-case review process and develop case management protocols for large investigations like OKBOMB. Substantial time and effort can be saved if the case investigators and prosecutors do not have to reinvent administrative and substantive solutions to problems and issues that commonly arise during major cases.

We believe that implementation of these recommendations will help address the significant problems with the FBI's computer systems and document handling. These problems are not new. The FBI has known about many of them for some time, either because the OIG has discussed them in other reports or because the FBI has found them through their own reviews. But the FBI has not done enough to address these problems.

As the tragic attacks of September 11 revealed, the FBI continues to be faced with cases of the scale and dimensions of OKBOMB, and the lessons learned from OKBOMB continue to be important. To adequately fulfill its responsibilities in major cases, as well as in smaller ones, the FBI must significantly improve its document handling and information technology. This requires a sustained commitment of resources and effort, but we believe the FBI must make this commitment if it is to avoid the serious problems that occurred in the OKBOMB case.