United States Department of State and the Broadcasting Board of Governors Review of the Department’s

出处：按学科分类—政治、法律 BERKSHIREPUBLISHINGGROUP《PatternsofGlobalTerrorism1985-2005：U.S.DepartmentofStateReportswithSupplementaryDocumentsandStatistics》第607页（39274字）

Office of Inspector General

Executive Summary

In a June 18, 2004 letter to Secretary of State Colin Powell,six members of the U.S. Senate requested that the Office of Inspector eneral (OIG) review how inaccurate and incomplete data and statements came to be included in the Patterns of Global Terrorism—2003 report. Patterns—2003, released on April 29, 2004, asserted that acts of international terrorism had declined slightly in the past several years and that 2003 had the “lowest annual total of international terrorist attacks since 1969,” suggesting that the Administration was winning the global war on terrorism.Shortly there after, an oped piece in the Washington Post alleged that there were statistical errors in the report.Subsequent articles in various periodicals claimed terroristacts had been left out, mislabeled, and counted incorrectly.On June 22, 2004, the Department of State (the Department) issued a revised version of the Patterns—2003 report that identified an increase in the number of significant terrorist events.

Before Congress mandated that the Department report on terrorism in 1987,⁽¹⁾ the Central Intelligence Agency(CIA) produced a report on terrorism, which was viewed largely as the reference on significant trends in international terrorism.With the 1987 legislation, Patterns began its transition to serving as a “report card” on how well the U.S. government was responding to the threat of terrorism.Although the Department has overall responsibility for producing the annual report on international terrorism,the Terrorist Threat Integration Center (TTIC) produces the data that form the basis of the report’s Appendix A—Chronology of Significant Terrorist Incidents. An Incident Review Panel (IRP), which includes members of the Intelligence Community, reviews a monthly listing of terrorist incidents provided by TIC and determines which events are “significant.”

OIG’s review has found that the inaccurate statements in the report were based on omissions of IRP-adjudicated decisions and apparent in consistencies in the database of terrorist events maintained by TTIC.The database formed the basis for Appendix A, as well as for the charts and graphs in the report’s Appendix G—Statistical Review, and for the statistics used in the report’s Year in Review section.The reasons for the omissions and inconsistencies in the TTIC database were:

■ Data entry lagged toward the end of 2003, and IRP decisions of adjudicated terrorist incidents between November11 and the end of December were not entered into TTIC’s database until April 2004, after the report had been printed.

■ No one questioned the omissions before the report was printed.

■ It appears that there was no consistent, replicable methodology that IRP used for selecting events to be included in Appendix A.

OIG was also asked to look at whether the process for assemblingthe report and the participants involved in the process differed significantly from previous reports.The major difference in the process between the 2003 report and previous reports was the transfer to TTIC from the CIA’s Counter terrorist Center (CTC) of maintenance of the database. The TTIC unit manager, who was formerly from CTC and who was responsible for updating the database, left TTIC in December 2003 and was not replaced. According to TTIC officials, data entry for the events adjudicated by the IRP in January 2004⁽²⁾ did not begin until April 2004, well after AppendixA was first sent to the Department in February 2004.

The office and procedures involved in the preparation of Patterns within the Department were the same in 2003 as in prior years. However, a personnel change and staffing shortageswithin the office of the Coordinator for Counterterrorism(S/CT) likely affected oversight of the report. A Foreign Service public affairs officer who previously helped produce the report left in the spring of 2003, and the position was reassignedelsewhere within S/CT.The S/CT associate coordinator,whose office was responsible for producing the reportbegan working in that office in June 2003 and had no previousexperience in preparing the Patterns report.

In summary, OIG found that the omissions and apparent inconsistencies were due to a number of factors. The shift from CTC to TTIC of responsibility for maintenance of the database of terrorist events, along with the lack of trained,long-term personnel working in that office, also probably contributedto the lack of supervision of the database. In addition,the process for assembling the report at the Department,while differing little from that of previous years, lacked suffi-cient oversight and coordination.

Both S/CT and TTIC personnel greatly regretted the errorsin the 2003 report, and both have begun to formulate more rigorous methodologies for the compilation of data.OIG’s recommendations, along with the innovations and remedialaction already undertaken by S/CT and TTIC, should provide future Patterns reports a more useful and accurate analysis of terrorist and anti-terrorist activities.

Objectives, Scope,and Methodology

The objectives for OIG’s review were to determine: 1) how“inaccurate” statements came to be included in the Department’sPatterns of Global Terrorism—2003, 2) whether there was a significantly different process used in assembling the report,and 3) whether there were different participants involvedin the process, compared to previous years.

OIG performed fieldwork in Washington, D.C., from July to August 2004.The team conducted interviews with Departmentmanagement officials from S/CT, the Bureau of Intelligenceand Research (INR), and the Bureau of Diplomatic Security. The Team also met with the Director of TTIC and other relevant TTIC officials. TTIC also conducted an internalreview of its support to the Department’s preparation of Patterns, and OIG met with TTIC’s review team on several occasions throughout the review process. Both teams shared their findings and recommendations. OIG’s team consisted of Ambassador Fernando Rondon, team leader, and Anita Schroeder, Margaret Ann Linn and Stephanie Hwang. This evaluation was conducted in accordance with the Quality Standards for Inspections issued by the President’s Council on Integrity and Efficiency. OIG discussed its findings and recommendationswith S/CT officials at the conclusion of the review.Their comments are addressed within the report.

Background

In a June 18, 2004, letter to Secretary Powell, six members⁽³⁾of the U.S. Senate requested that OIG identify how inaccurateand incomplete data and statements came to be included in the Patterns of Global Terrorism—2003 report. The Senatorsreferred to a need for clearer accounting in the war on terrorism as a reason why Congress mandated the annual Patterns report in 1987. Hence, they said they “were very disturbed”to learn of the inaccuracies in the report and of the incomplete picture of global terrorism that it offered. Furthermore,the Senators found the Administration’s use of the reportin press conferences and press releases to be “especially troubling.” (For the full text of the Senators’ letter, see AppendixA.)

Patterns—2003, released on April 29, 2004, indicated that acts of international terrorism had declined slightly in the past several years and that 2003 had the “lowest annual total of international terrorist attacks since 1969,” which was used to bolster the assertion that the Administration was winningthe global war on terrorism. Shortly thereafter, an op-ed article in the Washington Post alleged that there were statisticalerrors in the report. Subsequent articles in various periodicalsclaimed events were left out, mislabeled, and counted incorrectly. On June 22, 2004, the Department issued a revisedPatterns—2003 report that said there was an increase in the number of significant international terrorist events over the period.

History of Patterns

of Global Terrorism

The CIA issued the U.S. government’s first annual report on international terrorism in 1976 as a research paper.The goal of that report, International and Transnational Terrorism: Diagnosisand Prognosis, was to provide a framework for understandinginternational terrorism. In 1978, the CIA began to issue the report annually and added coverage on significant trends in terrorist activities, including foreign-government support for terrorist groups and international efforts to deter terrorism.

In 1982, the role of producing a report on international terrorism shifted from CIA to the Department when the first publication under the Department’s purview, Patterns of InternationalTerrorism: 1981, was published.The first “Chronologyof Significant Terrorist Events” (Appendix A) appeared in the following year’s report.

In December 1987, the Foreign Relations Authorization Act, Fiscal Years 1988 and 1989 (P.L. 100-204, Section 104, as amended), required that the Department provide Congress a full and complete annual report on terrorism for those countries and groups meeting the criteria for internationalterrorism.The reporting requirement changed in 1996 to include the extent to which other countries cooperate with the United States in apprehending, convicting, and punishing terrorists responsible for attacking U.S. citizens or interests.A more complete history of the Patterns reports may be found in Appendix B.

As now designed, Patterns serves as a unique, comprehensiveaccounting of significant acts of global terror and of the diplomatic record of the United States and its partners in cooperativelycountering such terror. Its unclassified nature facilitateswide distribution and contributes to greater public understanding of the global war on terrorism.

Criteria

Title 22, Section 2656f, of the U.S. Code requires that the Department, by April 30 of the following year, produce an annual report on terrorism that includes “detailed assessments”for each country in which there were significant acts of international terrorism. The law defines an international terrorist event as premeditated, politically motivated violencethat is perpetrated against noncombatant targets by subnational groups or clandestine agents to influence people and which involves citizens of two or more countries.The definitionof “significant” is left to the Secretary. In November 1996, with further modifications in later years, criteria were developed to define which international terrorist attacks were“significant.” An incident is now considered significant if there was loss of life or serious injury to people, major property damage of $10,000 or more, or the abduction or kidnapping of people.

In 1996, the congressional reporting requirements were amended to require the Department to report on the extent to which other countries cooperate with the United States in apprehending, convicting, and punishing terrorists responsiblefor attacking U.S. citizens or interests.

Organizational Responsibilities

Although the Department has the overall responsibility for producing the annual report on international terrorism, other U.S. government entities (primarily TTIC) assist in writing portions of the report and in producing the statistics used for the report’s Appendix A.

S/CT is the Department’s program office responsible for coordinating and producing Patterns. The responsibility for preparing Patterns has belonged to S/CT since the Departmentbegan issuing the report annually in 1982. S/CT gets input from various sources (overseas posts, functional and regionalbureaus) to prepare the report, drafts the Coordinator’sYear in Review section, which discusses trends and numbers of events and writes introductions to the regional overview. S/CT also takes the lead on drafting Appendix D—U.S. Programs and Policy, with the assistance of the Department’sBureau of Diplomatic Security.

The Department’s INR draft[s] individual country reports within the regional overviews. The drafts are then sent to S/CT, which sends them to the regional bureaus for review and comment.

TTIC⁽⁴⁾ reviews information on terrorist events and maintainsthis information in a database, which is inherited from the CIA’s CTC.⁽⁵⁾ The database is used to compile Appendix A—Chronology of Significant Terrorist Incidents at the end of each year. TTIC also provides data for the charts and graphs that are used in the report.

The IRP,⁽⁶⁾ which was made up of representatives from INR, CIA, Defense Intelligence Agency, and the National SecurityAgency, reviewed terrorist incidents and determined which events met the established definition of significant internationalterrorism. A representative from TTIC, a nonvotingmember of the IRP, chaired the panel. The results of the IRP’s meetings are contained in TTIC’s database.

[Eds. Note: Appendixes B and C of Patterns contained descriptionsof “designated foreign terrorist organization” and“other terrorist groups.” These descriptions are compiled in Part 2,Terrorist Groups.]

The Process

TTIC tracks worldwide acts of terror and maintains this informationin a database. On a daily basis, TTIC analysts reviewedreporting of events for possible inclusion in the database. If an analyst determined that a terrorist incident should be considered by the IRP, a description of the incident was entered into the database and summary information,such as the date of the incident, the country, type of event(e.g., kidnapping), was also entered. Monthly, TTIC sent terroristincident summaries⁽⁷⁾ to the members of the IRP for their consideration. IRP members could study and share the information with staff from their respective agencies before they met, on the first Wednesday of each month, to determine (adjudicate) which of the previous month’s events were significantinternational terrorist events.TTIC, in turn, updated the database with a record of the IRP’s decision.The Department began drafting portions of the Patterns—2003 report in November 2003, but the chronology of significantinternational terrorist events was not received by the Department until February of 2004.The draft version of Patterns,

without Appendix A, was also circulated to regional and functional bureaus for review and comment. Because of the short timeframe in which S/CT has to coordinate the report,Appendix A was never circulated for review or comment within S/CT or with other Department offices, accompanied by the draft version of the report. In March 2004, the report was forwarded for production, to meet the April 30 congressionaldeadline.

Addressing Congressional Concerns

Congress requested that OIG determine: 1) how “inaccurate”statements came to be included in the Department’s Patterns of Global Terrorism—2003 report, 2) whether there were significantlydifferent processes used in assembling the report than in previous years, and 3) whether the participants involveddiffered significantly from those who produced previousreports.The answers to these questions are summarized here briefly and are discussed in more detail in the Findings and Recommendations section of this report.

OIG found that the inaccurate statements were due to omissions and apparent inconsistencies in the database of terroristevents maintained by TTIC.These include:

■ Even though the IRP had in January 2004 adjudicated terroristevents for the remainder of the previous calendar year, data entry lagged toward the end of 2003, and informationon terrorist incidents between November 11 and the end of December 2003 was not entered into the database until April 2004.

■ No one questioned the omissions before the report was printed.

■ It appears that there was no consistent, replicable methodologyused by the IRP for selecting events to be included in the chronology in Appendix A.

Given the above circumstances, Appendix A and the statisticsused as a basis for official Department briefings cannot be viewed as reliable.

OIG was also asked to look at whether the process and the participants involved in producing the report were significantlydifferent from those of previous reports. OIG found that the process remained the same, but that the major differencebetween the 2003 report and previous reports was the involvement of TTIC in the maintenance of the database used to prepare Appendix A. Prior to the formation of TTIC, responsibilityfor maintenance of the database rested with CTC. The TTIC unit manager, who was formerly from CTC and who was responsible for updating the database, left TTIC in December 2003 and was not replaced. Efforts to fill the position were unsuccessful until after the report was published.According to TTIC officials, data entry for the events adjudicated by the IRP in January 2004⁽⁸⁾ did not begin until April 2004, well after Appendix A data was first sent to the Department, in February 2004.

The offices involved in the preparation of Patterns within the Department were the same in 2003 as in prior years.However, a personnel change and staffing shortages within S/CT likely affected oversight of the report. A Foreign Service public affairs officer who previously helped produce the reportleft in the spring 2003, and the position was reassigned elsewhere within S/CT.The associate coordinator whose of-fice was responsible for producing the report began working in that office in June 2003 and had no previous experience in preparing the Patterns report.

In summary, OIG found that the omissions and apparent inconsistencies were due to a number of factors.The shift of responsibility from CTC to TTIC for maintenance of the databaseof terrorist events, along with the lack of trained longtermpersonnel working in that office, probably contributed to the lack of supervision of the database.The process for assemblingthe report at the Department, while differing little form that of previous years, lacked sufficient oversight and coordination.Recommendations to correct these problems are presented in the following section.

OIG found that both S/CT and TTIC personnel greatly regrettedthe errors in the 2003 report and have already begun formulating more rigorous methodologies for the compilationof data.The recommendations discussed in the next section,along with innovations and remedial action already undertaken by S/CT and TTIC should provide future Patterns reports a more useful and accurate analysis of terrorist and anti-terrorist activities.

Findings and

Recommendations

The Department’s Patterns of Global Terrorism—2003 report contained erroneous information largely because of gaps in data entry, inconsistently applied methodology, and lack of oversight. In this section, OIG presents its findings and recommendationsregarding these issues.

Data Inconsistencies and the

Compilation of Appendix A

The omission from the report of terrorist attacks occurring after November 11, 2003, was perhaps the most egregious error, but OIG also found that this omission highlighted other weaknesses in the compilation of Appendix A.What appearto be inconsistencies in the application of the definitions,as described in the criteria and the adjudication by the IRP, also affected the quality of the data. Finally, Appendix A was never circulated within the Department nor was it made available to the analysts responsible for writing regional overviews and country specific narratives, individuals who could have identified the omissions and inconsistencies.

Appendix A Methodology

OIG reviewed a sample of significant international terrorist events that were identified in Appendix A in both the originaland revised Patterns—2003 reports and found that the definitions and criteria for identifying and classifying internationalterrorist events appear to have been applied inconsistently.Some incidents were removed from the original version of Appendix A, but other incidents, similar to those that were deleted, were then added. For example:

■ On March 31, IBM employees in Italy found an explosive device and notified police, who described it as a dangerous,though rudimentary, bomb.This incident was deleted from the revised Appendix A.

■ On September 8, authorities in Madrid, Spain, safely defuseda parcel bomb hidden in a book that was sent to the Greek Consulate. This incident was not included in the original version of Appendix A but was included in the revisedversion.

■ On February 1, a time bomb was discovered and defused in a McDonald’s restaurant in Turkey. There were no injuriesand no one claimed responsibility.This incident was deleted from the revised report.

■ On February 25, an incendiary bomb was thrown at a McDonald’s restaurant in Saudi Arabia.There were no injuriesand no one claimed responsibility.This incident was included in both reports.

TTIC explained that there are several reasons for what appearto be inconsistencies in the application of the criteria. In some cases, details were not available when the IRP first consideredan event. Over time, as more information became available, the IRP would reconsider the event, and this may have changed how the event was categorized. Since the IRP did not keep records or minutes of how determinations were made, TTIC could only speculate on why some events were included or not included. TTIC also opined that the criteria are complex and decisions can be subjective. For example,when the IRP reconsidered the February 1 event, it may have deleted it because the panel determined that there was not enough firepower involved to cause significant damage. In contrast, the February 25 event may have been included becauseincendiary bombs cause significant property damage.

There also appear to be inconsistencies in how events are counted. Multiple similar events offering more or less simultaneouslyor in succession in one area are sometimes counted as one incident and at other times are counted individually.For example:

■ On February 25, two bombs that exploded in Caracas,Venezuela, and damaged the Spanish and Colombian embassies,were listed as one incident, as were four bomb attackson March 25 in Pristina, Serbia, against police stations of the United Nations Interim Administration Mission in Kosovo.

■ On April 12, grenade attacks at two different targets inAnantnag District, Kashmir, one at a bus station and anotherat an army patrol, were listed as separate events.

■ On November 15 in Istanbul, Turkey, vehicle-bomb attacksat the Beth Israel and Neve Shalom synagogues were listed as separate events.

TTIC and S/CT explained that the IRP had analyzed these events and that the panel had a basis for determining that some events should be considered as one incident and other events should be listed as separate incidents. Unless a rationale is explained, it is difficult for the reader to understandwhy some multiple events are counted as only one incidentand why others are counted individually.

TTIC and S/CT are developing an algorithm that explains how multiple events are counted. OIG encourages S/CT to include this supporting material in the report to assist the reader in understanding the methodology. OIG also believes that, for purposes of transparency, S/CT should include an explanatory footnote for each event that may not obviously fit the criteria and methodology.

Accepted standards of data collection and analysis require that there be a reliable and complete database (the universe of events under consideration) and that classification of events from the database be performed in a consistent, replicablemanner. This means that a separate determination of events as significant or non-significant, using the same database,should yield the same results.When terrorist events do not fit established criteria or definitions, this circumstance should be flagged, allowing classification criteria to be adjustedperiodically.The definitions for classifying events from the database (the criteria for determining if an event is significant,international or domestic, etc.) should be based on hard data. If hard data are not available, then a thorough explanationof how decisions are made must be provided.

The accuracy of the report is dependent on accurate and complete data and a comprehensive adjudication process.The role of TTIC in the preparation of the terrorism database is also vital to the accuracy of the data. Although S/CT does not have a direct role in the preparation of Appendix A, it is in the Department’s best interest to ensure that a reliable methodology is used in determining the data for the Patterns report.Therefore, OIG recommends that S/CT and TTIC formalizethe roles and responsibilities of the two agencies.

The following recommendation calls on S/CT to conclude a memorandum of understanding (MOU) with TTIC. This MOU should ensure that:

■ Specific, replicable criteria for classifying terrorist events for Patterns are established and applied consistently;

■ In cases where terrorist events do not fit the established criteria, an explanation is recorded as to how the event was classified; and

■ The classification criteria are periodically adjusted to includeor exclude certain types of events, as needed.OIG also recommends that TTIC keep complete minutes of meetings and notes on how decisions were made, and that they make this information available to the Department and other users upon request, as appropriate.

OIG also found that the Patterns—2003 report containedinternal inconsistencies, particularly between the chronology and the geographic overview sections and betweenthe charts and graphs and other sections. For example,the Year in Review section refers to “190 acts of international terrorism.” However, Appendix A lists 169 “significant terroristincidents.” Country-specific portions of the geographic portions of the geographic overviews, meanwhile, often includereferences to domestic terrorist events and to internationalterrorist events.

In general, OIG found that the commingling of discussionsof domestic and international terrorism and of significantand non-significant terrorism led to confusion and difficulty in interpreting the numbers cited. OIG suggests that S/CT include in the report’s introduction an explanation of the report’s methodology for classifying significant internationalterrorist events, preventing readers from becoming confusedabout which types of events are being discussed. OIG notes that previous early editions of Patterns included such a reference. S/CT has agreed and intends to include a section on methodology in future reports.That section would define international and domestic incidents and significant and non-significant incidents and state the procedures used to make these determinations. References to major incidents in future reports will make clear whether the incident is domesticor international.

Frequency and Review of Appendix A

In the past, the annual list of significant international terroristevents was forwarded to S/CT in February of the subsequentyear. However, S/CT analysts who prepare the regional overviews did not have access to Appendix A. Moreover, the chronology has not traditionally been vetted throughout the Department by such bureaus as INR,⁽⁹⁾ the Bureau of ConsularAffairs, and the Bureau of Diplomatic Security, all of which have independent sources of information on terrorist events. S/CT said there was insufficient time in the publishingschedule for it to circulate Appendix A to other Departmentoffices for comment. It also stated that the chronology represented the vote of the IRP, and S/CT did not question the decisions made by the IRP. However, had Appendix A been circulated to other analysts at S/CT or elsewhere in the Department, these analysts would likely have identified the omissions and inconsistencies contained in Patterns—2003.

Analyzing patterns of terrorism requires a reliable, periodicchronology of significant events that can be compared and examined in context, along with information from other sources.The chronology is more likely to be complete and accurateif it is vetted by a number of offices dealing in antiterroristactivities and terrorism reporting.The final arbiter of the chronology must, however, be TTIC.

TTIC agrees that the chronology can be produced more frequently and has suggested that it will, in the future, producea quarterly chronology.

Oversight and Staffing

OIG found that the process for assembling the Patterns—2003 report lacked oversight and adequate review. Statistics presented in the Year in Review section were not examined for appropriateness before being finalized. An experienced analyst might have questioned the inconsistencies and data abnormalities in the chronology while preparing the Year in Review figures; however, an individual experienced in data analysis and reporting did not draft that portion of the report.Data collection and analysis requires that analysts verify the suitability and completeness of the data and double-check their calculations against historical patterns and other existinginformation.

An S/CT Civil service public affairs specialist, who had handled production of the report since 1986, directed general coordination of the report. S/CT previously had a second publicaffairs position, a Foreign Service officer who helped producePatterns.When this individual left, in the spring 2003,the position was transferred to other functions within S/CT.

Management of the preparation of a report as important as Patterns should include oversight of the entire publication by a knowledgeable person who has responsibility for the accuracyof the data analysis presented, the consistency of data presented in different sections of the report, and the cohesivenessof the entire report. Given the importance of the global war on terrorism to U.S. national interests and foreign policy,Patterns should have the support and staff needed to produce a world-class product.

Attribution

When the errors in the data were first brought to the public’s attention the Department was presumed to be entirely responsible.With responsibility for maintaining the database comes acknowledgement of authorship and accountability for the charts, graphs, and lists of events generated from that information.This attribution is not acknowledged in the Patternsreport, but OIG believes that it is an important factor in ensuring that the information presented in the report is accurate,transparent, and verifiable.

Progress to Date

As it conducted its review, OIG found that S/CT and TTIC had already begun to review the steps involved in the preparationof Patterns and to discuss changes. S/CT is proposing to include more information in the report on the membership of the IRP and on how possible incidents are identified,processed, and adjudicated. Future reports will also discuss international versus domestic terrorism and significant versus non-significant events. In addition, Patterns reports will clarifythe adjudication of incidents in Kashmir, Israel, and Chechnya, since all of these incidents involve cross-border support for terrorist groups.

S/CT has informed OIG that it has prepared a draft MOU with TTIC outlining the responsibilities of each office or agency. S/CT has also reviewed the clearance and approval procedures for Patterns and is improving the process. Finally,S/CT plans to add two positions having direct responsibility for Patterns, a Public Diplomacy unit chief and a Public Diplomacy analytic officer.

TTIC, meanwhile, has also conducted an internal review of the Patterns process. One of its key findings was the need to restructure and expand the scope of responsibilities of the incident adjudicative body that replaces the IRP. The new panel will include a more diverse representation from across the counterterrorism community and will include TTIC assigneesfrom each of the original agencies whose members originally comprised the IRP, as well as TTIC assignees form the Department of Homeland Security and Federal Bureau of Investigation. TTIC is also requesting that a representative from S/CT, as a non-voting member, attend each session to provide insight into incidents that will be included in Patterns.TTIC is providing guidelines to address complex issues such as multiple events, disputed borders, timing of incidents,and the maintenance of record of the rationale behind these decisions, should questions arise in the future. TTIC agrees that there is a need to distribute the list of significant internationalterrorist incidents more frequently.

OIG believes that all parties involved in the preparation of Patterns are taking effective steps to ensure that future reportscontain the most complete and accurate depiction possibleof international terrorism.

Appendix A: Request

Letter from Congress

The letter below was sent to Secretary of State Colin I. Powell on 18 June by Senators Tom Daschle, Harry Reid, Joseph R.Biden Jr., Patrick Leahy, Carl Levin, John D. Rockefeller IV.

Dear Mr. Secretary:

We are writing to express out deep concern with the latestedition of the Department of State’s Patterns of Global Terrorism2003 report.

Defeating global terrorism is a goal we all share.Yet, over the course of the last several months, Defense Secretary Rumsfeld has repeatedly indicated that the Administration lacks the “metrics” to determine whether the United States is winning the war on international terrorism. Given that this effortis likely to take considerable time and resources, we agree with Secretary Rumsfeld that success requires that the Americanpeople be provided a clearer accounting from their leadersof how we are fighting and faring in the war on terrorism.

The desire for clearer accounting of our efforts against terrorismwas part of the rationale that led Congress in 1987 to pass legislation mandating an annual Patterns of Global Terrorismreport. Congress believed this report could help ensure that our government’s decisions to defeat terrorism are informedby the latest and most accurate data.

Because defeating terrorism will require accurate data, we were very disturbed to learn that the State Department’s Patternsof Global Terrorism annual report for 2003 presented an inacurrate and incomplete picture. As you know, the Patterns of Global Terrorism 2003 report indicated a decline in global terrorism when, in fact, the number of terrorist incidents increased—to the highest level in more than 20 years. Especiallytroubling is the fact that senior Administration officials used this erroneous data in a series of press conferences and press releases to claim that the United States is winning the war on international terrorism.

We are pleased that you committed to address this problemand urge you to aggressively follow through on these pledges.We submit a series of additional recommended steps for you to consider while you are editing the earlier report.

1. immediately remove from the State Department’s website the inaccurate and incomplete version of the report as well as any public testimony, speeches or transcripts based on that report;

2. upon completion of the revised report, forward the new report to any Congressional Committee that heard testimonybased on the faulty report;

3. in order to clear up any misimpression that the earlier reportmay have left with the American public or with Congress,particularly as Congress begins deliberation on FY2005 appropriations measures, hold a public press conference to describe the corrections made to the originalreport and the implications of the new conclusions for our government’s success in the war on terror;

4. initiate an investigation by the State Department’s InspectorGeneral into how the inaccurate statements found their way into the latest version of this report and whether the process for assembling this report and the participants involved differed significantly from previous years.

Prevailing in the war on terrorism will require the longtermcooperation of the American people and the world. Our government’s credibility is essential to these efforts. By releasingan inaccurate report, the Administration has undermined out nation’s credibility at a critical moment in the war on terrorism.We believe the best thing you can do to restore some of this lost credibility is to pursue the initiatives described above.

Thank you for your consideration.

Sincerely,

Tom Daschle Harry Reid

Joseph R. Biden Jr. Patrick Leahy

Carl Levin John D. Rockefeller IV

1 Report contains caveat that study does not represent a CIA position, but that judgments are those of the author.

2 Ibid.

Report Number SIO-S-04-18, September 2004