Use of Force: ATF Policy, Training and Review Process Are Comparable to DEA's and FBI's (Chapter Report, 03/29/96, GAO/GGD-96-17). Pursuant to a congressional request, GAO reviewed the Bureau of Alcohol, Tobacco, and Firearms' (ATF) use of deadly force and dynamic entry, focusing on: (1) ATF policies on the use of deadly force; (2) how ATF conveys its policies to its agents; (3) the reasons for and the extent to which ATF uses dynamic entry and what equipment it uses; (4) ATF compliance with its procedures for investigating shooting and alleged excessive force incidents; and (5) how the Drug Enforcement Administration (DEA) and the Federal Bureau of Investigation (FBI) address similar issues. GAO found that: (1) except for a few instances, the 1988 ATF policy on deadly force was consistent with prior DEA and FBI policies and the 1995 Departments of the Treasury and Justice uniform policies which superceded their agencies' policies; (2) agents may use deadly force only when they reasonably believe that suspects pose an imminent threat of death or serious injury to themselves or other persons; (3) the three agencies' new agent training in their deadly force policies is similar and all agents are required to be retrained on a quarterly basis throughout their careers; (4) dynamic entry is used to ensure personal safety when access to premises is needed in high-risk situations or when suspects might swiftly destroy evidence; (5) the three agencies' method of dynamic entry and weaponry and equipment used was similar; (6) ATF reporting, investigating, and review procedures for shooting and excessive force incidents are consistent with recommended standards and similar to the other agencies'; (7) ATF and DEA excessive force procedures are generally comparable, but FBI procedures require that all allegations be submitted to Justice for possible criminal or civil rights violations before the allegations are self-investigated; (8) ATF generally complied with its investigative procedures during fiscal years 1990 through 1995; (9) ATF found that all intentional shootings were justified, most allegations of excessive force were unsubstantiated, and 5 agents warranted sanctioning for misconduct; and (1O) ATF is implementing lessons learned from the incidents, particularly the Waco, Texas, raid. --------------------------- Indexing Terms ----------------------------- REPORTNUM: GGD-96-17 TITLE: Use of Force: ATF Policy, Training and Review Process Are Comparable to DEA's and FBI's DATE: 03/29/96 SUBJECT: Law enforcement agencies Search and seizure Crimes or offenses Firearms Human resources training Law enforcement personnel Arrests Constitutional rights Investigations by federal agencies IDENTIFIER: Waco (TX) FBI Firearms Training System ATF Inventory Tracking and Equipment Management System ATF On-the-Job Training Program Federal Law Enforcement Training Center Criminal Investigator Training Program Federal Law Enforcement Training Center Use of Force Model M16 Automatic Weapon ****************************************************************** ** This file contains an ASCII representation of the text of a ** ** GAO report. Delineations within the text indicating chapter ** ** titles, headings, and bullets are preserved. Major ** ** divisions and subdivisions of the text, such as Chapters, ** ** Sections, and Appendixes, are identified by double and ** ** single lines. The numbers on the right end of these lines ** ** indicate the position of each of the subsections in the ** ** document outline. These numbers do NOT correspond with the ** ** page numbers of the printed product. ** ** ** ** No attempt has been made to display graphic images, although ** ** figure captions are reproduced. Tables are included, but ** ** may not resemble those in the printed version. ** ** ** ** Please see the PDF (Portable Document Format) file, when ** ** available, for a complete electronic file of the printed ** ** document's contents. ** ** ** ** A printed copy of this report may be obtained from the GAO ** ** Document Distribution Center. For further details, please ** ** send an e-mail message to: ** ** ** ** ** ** ** ** with the message 'info' in the body. ** ****************************************************************** Cover ================================================================ COVER Report to the Chairman, Subcommittee on Treasury, Postal Service, and General Government, Committee on Appropriations, House of Representatives March 1996 USE OF FORCE - ATF POLICY, TRAINING AND REVIEW PROCESS ARE COMPARABLE TO DEA'S AND FBI'S GAO/GGD-96-17 ATF Use of Force (187013) Abbreviations =============================================================== ABBREV ACC - ATF's Associate Chief Counsel ATF - Bureau of Alcohol, Tobacco and Firearms CITP - Criminal Investigator Training Program DEA - Drug Enforcement Administration ELRB - ATF's Employee Labor Relations Board FBI - Federal Bureau of Investigation FLETC - Federal Law Enforcement Training Center FTCA - Federal Tort Claims Act HEAT - High Risk Entry and Arrest Team IACP - International Association of Chiefs of Police MSPB - Merit System Protection Board OCC - ATF's Office of Chief Counsel OE - ATF's Office of Enforcement OI - ATF's Office of Inspection OJT - On-the-Job Training OPR - Office of Professional Responsibility PCIE - President's Council on Integrity and Efficiency PRB - ATF's Professional Review Board RAC - Resident Agent-in-Charge SAC - Special Agent-in-Charge SAR - Significant Activity Report SIR - Shooting Incident Report SIRB - Shooting Incident Review Board SIRT - Shooting Incident Review Team SRT - Special Response Team SWAT - Special Weapons and Tactics Letter =============================================================== LETTER B-262134 March 29, 1996 The Honorable Jim Lightfoot Chairman, Subcommittee on Treasury, Postal Service, and General Government Committee on Appropriations House of Representatives Dear Mr. Chairman: Your August 2, 1995, letter requested that we review various aspects of the Bureau of Alcohol, Tobacco and Firearms (ATF). This report responds to your questions concerning ATF's use of deadly force and dynamic entry. Specifically, the report discusses (1) ATF's policies for the use of deadly force; (2) how ATF conveys its policies to its agents; (3) the reasons for and the extent to which ATF uses dynamic entry and the equipment it uses to accomplish these entries; and (4) ATF's compliance with its procedures for investigating shooting and alleged excessive use-of-force incidents. The report also discusses how ATF addresses these issues in comparison to the Drug Enforcement Administration and the Federal Bureau of Investigation. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies to the Secretary of the Treasury, the Attorney General, the Director of ATF, the Director of the Federal Law Enforcement Training Center, the Administrator of the Drug Enforcement Administration, the Director of the Federal Bureau of Investigation, and other interested parties. Copies will be made available to others upon request. Major contributors to this report are listed in appendix VII. Please call me on (202) 512-8777 if you or your staff have any questions. Sincerely yours, Norman J. Rabkin Director, Administration of Justice Issues EXECUTIVE SUMMARY ============================================================ Chapter 0 PURPOSE ---------------------------------------------------------- Chapter 0:1 According to the International Association of Chiefs of Police, managing officers' use of force is one of the most difficult challenges facing law enforcement agencies. The ability of officers to enforce the law, protect the public, and guard their own safety is very difficult in an environment where violent crime is commonplace and firearms are frequently used for illegal purposes. Because the Bureau of Alcohol, Tobacco and Firearms (ATF) is the federal agency primarily responsible for enforcing firearms laws, its agents often face this difficulty. Over the past several years, ATF has come under public criticism and congressional scrutiny, in part, due to citizen accusations that ATF agents used excessive force in carrying out their enforcement responsibilities. In August 1995, the Chairman of the Subcommittee on Treasury, Postal Service, and General Government, House Committee on Appropriations, asked GAO to (1) identify and describe ATF's policies for the use of deadly force; (2) determine how ATF conveys its policies to its agents; (3) determine the reasons for and the extent to which ATF uses dynamic entry--a tactic used to gain rapid entry to premises when serving high-risk search and arrest warrants and which may include forced entry--and the equipment ATF uses to accomplish these entries; and (4) determine whether ATF has complied with its procedures for investigating shooting and alleged excessive use-of-force incidents. The Chairman also asked GAO to compare how ATF addresses these issues with the way the Department of Justice's Federal Bureau of Investigation (FBI) and the Drug Enforcement Administration (DEA) address them. While GAO determined whether ATF complied with its procedures for investigating shooting and alleged excessive use-of-force incidents on the basis of a review of investigative file documents, it did not evaluate the quality and adequacy of ATF's investigations. Neither did GAO verify whether all incidents had been reported nor whether all incidents that had been reported were investigated. BACKGROUND ---------------------------------------------------------- Chapter 0:2 ATF, an agency of the Department of the Treasury, is responsible for, among other things, investigating criminal violations of federal firearms, explosives, arson, alcohol, and tobacco laws. As of September 1995, ATF had 1,944 special agents; 1,777 of these agents were assigned to 24 field divisions nationwide. Treasury's use-of-force policy, revised in October 1995 and applicable to ATF, is based on the premise of timely and effective application of the appropriate level of force required to establish and maintain lawful control. Both Treasury and Justice define "deadly force" as use of any force that is likely to cause death or serious physical injury. For example, any firearm discharge directed at a suspect is considered to be deadly force. As shown in table 1, ATF, on average, conducted over 12,000 investigations and arrested about 8,000 suspects during fiscal years 1990 through 1995. In the course of these activities, ATF was involved in fewer than 10 reported shooting or alleged excessive force incidents each year. Table 1 Summary Table of ATF Investigations, Arrests, Shootings, and Alleged Excessive Force Incidents, FYs 1990- 1995 FY FY FY FY FY FY 1990 1991 1992 1993 1994 1995 ---------------------------- ----- ----- ----- ----- ----- ----- Total investigations 11,74 13,73 15,11 13,47 11,96 10,50 1 5 7 9 5 5 Suspects arrested 7,104 8,301 8,631 6,607 9,595 6,692 ATF shooting 4 9 9 2 8 7 incidents\a Alleged excessive force 3 2 5 3 8 4 incidents\b ---------------------------------------------------------------------- \a Reported intentional shootings at suspects by ATF agents. \b Reported alleged excessive force incidents involving ATF agents. Source: ATF. RESULTS IN BRIEF ---------------------------------------------------------- Chapter 0:3 In October 1995, Treasury and Justice issued uniform policies governing the use of deadly force for each of their bureaus and agencies. The revised policies permit law enforcement officers to use deadly force only when an officer has a reasonable belief that the subject of such force poses an imminent danger of death or serious physical injury to the officer or another person. GAO determined that the 1988 ATF use of deadly force policy, which was in effect before the issuance of the October 1995 Treasury policy, was, with the two distinctions discussed in the principal findings, consistent with the 1995 Treasury policy. In addition, ATF's 1988 policy was, with the three distinctions discussed in the principal findings, consistent with FBI and DEA policies that were in effect immediately before the issuance of the 1995 uniform policies. ATF conveys its deadly force policies to new agents through training. GAO's discussions with training officials, reviews of course materials, and observations showed that the types of deadly force training provided new ATF agents were consistent with the types of training provided new DEA and FBI agents. Moreover, ATF policy requires that all ATF agents be instructed on the deadly force policy at least quarterly throughout their careers. In the fall of 1995, GAO observed three ATF divisions' quarterly firearms qualifications. At each qualification, firearms instructors reminded agents of the policy. According to ATF officials and GAO analysis of selected ATF data for fiscal year 1995, dynamic entry was a principal tactical procedure employed by ATF when access to premises was required to execute search and arrest warrants in high-risk operations--those posing a threat of violence--or in operations where evidence could be easily destroyed. According to ATF officials, ATF used dynamic entry when, given the specific situation, it reduced the potential for injury to those involved. According to FBI and DEA officials, dynamic entry also was a principal tactic used to execute their high-risk warrants where entry to premises was required. The equipment used by ATF, DEA, and FBI during dynamic entries is generally comparable on the basis of GAO's observations at division offices, discussions with agency officials, review of ATF inventory listings, and analysis of selected operations reports for fiscal year 1995. ATF's procedures for reporting, investigating, and reviewing shooting incidents and alleged use of excessive force incidents are consistent with standards recommended by organizations such as the President's Council on Integrity and Efficiency. Overall, ATF's shooting incident procedures are comparable with DEA's and FBI's with two distinctions. GAO identified distinctions between ATF and these agencies in the delegation of investigative responsibilities and representation on review boards. Also, ATF's excessive force procedures are comparable to DEA's with one distinction relating to delegation. However, ATF's procedures are not comparable to FBI's mainly because of distinctions regarding the referral of excessive force allegations to Justice for possible civil rights violations. GAO's review of ATF's investigative files for its reported intentional shooting incidents and alleged use of excessive force incidents for fiscal years 1990 through 1995 showed that ATF (1) complied with its investigative procedures with an exception discussed below; (2) found that all intentional shootings were justified; (3) found most allegations of excessive use of force were unsubstantiated; and (4) sanctioned agents it determined had engaged in misconduct. PRINCIPAL FINDINGS ---------------------------------------------------------- Chapter 0:4 ATF'S DEADLY FORCE POLICIES HAVE REMAINED GENERALLY CONSISTENT IN RECENT YEARS AND ARE GENERALLY CONSISTENT WITH FBI AND DEA POLICIES -------------------------------------------------------- Chapter 0:4.1 State rules and U.S. Supreme Court guidance on the use of deadly force have been evolving for a number of years. Approaches among the states on the use of deadly force have ranged from those that place a predominant emphasis on the apprehension of a fleeing felon to those that permit such force only with respect to dangerous suspects and impose several qualifications on the use of such force. Within this context, federal law enforcement agencies have adopted policies to govern their employees' use of deadly force. (See p. 28.) In October 1995, Treasury and Justice adopted use of deadly force policies for their component agencies that are uniform with the exception of certain agency mission-specific provisions covering, for example, Justice's prisoner-related responsibilities. Both policies provide that their respective officers may use deadly force only when necessary, that is, when the officer has a reasonable belief that the subject of such force poses an imminent danger of death or serious physical injury to the officer or another person. (See p. 33.) The 1988 ATF use of deadly force policy, in effect before the issuance of the October 1995 Treasury policy, was, with two distinctions, consistent with the new policy. The two distinctions were that (1) the 1995 Treasury policy referred to the use of "deadly force" while the 1988 ATF policy referred more specifically only to the use of a "firearm" and (2) the 1995 Treasury policy allows for the use of deadly force only when the law enforcement officer has a "reasonable belief" that there is an imminent danger of death or serious physical injury while the 1988 ATF policy allowed for the use of such force when the agent "perceives" an imminent threat of death or serious physical injury. (See p. 34.) In addition, the prior ATF policy was, with three distinctions, consistent with prior DEA and FBI policies. The three distinctions were that (1) ATF's policy alone provided the additional qualifying restriction that the threat of death or serious bodily harm be "imminent," (2) the ATF and DEA policies referred to the shooting of "firearms" while the FBI policy used the term "deadly force," and (3) the ATF policy used the term "perceives" while the DEA and FBI used the term "reasonably believe" and "reason to believe," respectively. (See p. 35.) ATF, DEA, AND FBI CONVEY DEADLY FORCE POLICIES TO THEIR AGENTS IN SIMILAR WAYS -------------------------------------------------------- Chapter 0:4.2 ATF conveys its deadly force policies to new agents through training. GAO's discussions with training officials, reviews of training materials and policies, and observations showed that the training provided new ATF agents to introduce them to the deadly force policies was consistent with the Treasury/ATF deadly force policies and the types of training provided were consistent with the training provided to new DEA and FBI agents. (See p. 37.) GAO's discussions and review of course materials also showed that within the first week of training, each agency provides new agent trainees with a classroom lecture and discussion describing, with examples, the agency's use-of-force and deadly force policies. Thereafter, each agency integrates use-of-force issues into other segments of the training where force could be a relevant issue, such as training in physical control techniques. Each agency employs training techniques, such as practical exercises, that use role-playing and firearms judgment exercises that require shoot or not-to-shoot decisions. Furthermore, each agency trains new agents in how to recognize the perceived level of threat they face and in how to respond to it with an appropriate level of force. (See p. 42.) Once new agent training is completed, ATF requires that the use-of-force policies are to be reiterated to agents throughout their careers at quarterly firearms qualifications and during tactical operations briefings. In the fall of 1995, GAO observed three divisions' quarterly qualifications, interviewed the divisions' Firearms Instructor Coordinators, and reviewed documentation of prior qualifications and used these observations, interviews, and documentation to confirm that the deadly force policy was reiterated during qualifications at these three divisions. Also, agents at the qualifications with whom GAO spoke confirmed that the policy was reviewed before every tactical operation. DEA and FBI officials said that deadly force policies are also to be reiterated at their quarterly firearms qualifications. (See p. 43.) ATF'S USE OF DYNAMIC ENTRIES AND RELATED EQUIPMENT IS GENERALLY COMPARABLE TO FBI'S AND DEA'S -------------------------------------------------------- Chapter 0:4.3 Dynamic entry is one of several tactical procedures used by ATF to execute search and arrest warrants. Dynamic entry, which relies on speed and surprise and may involve forced entry, is a preferred tactic during high-risk operations--those where ATF believes that suspects pose a threat of violence--or in operations where evidence can be easily destroyed. ATF statistics on suspects arrested from firearms investigations during fiscal years 1990 to 1995 showed that 46 percent of the suspects had previous felony convictions, 24 percent had a history of violence, and 18 percent were armed at the time of their arrests.\1 (See pp. 47 and 56.) All ATF case agents, including those assigned to special weapons and tactics units, known as Special Response Teams (SRT), are to be trained in the dynamic entry technique. GAO observed new agent and SRT training sessions where ATF agents made dynamic entries into buildings during practical exercises. Moreover, according to ATF officials, ATF agents primarily used the dynamic entry technique to gain entry to buildings during high-risk search and arrest warrants. (See p. 53.) From fiscal year 1993 through 1995, ATF conducted 35,949 investigations and arrested 22,894 suspects. During this same period, SRTs were deployed 523 times, and SRT members were involved in 3 intentional shooting incidents, 1 of which resulted in fatalities. GAO reviewed the available documentation for all 157 SRT deployments for fiscal year 1995 and found that the dynamic entry technique was used almost half the time and was the predominant technique used when an entry to a building was required. However, in none of the 1995 SRT dynamic entries did ATF agents fire their weapons at suspects. (See p. 54.) According to ATF, DEA, and FBI officials, the primary purpose of dynamic entry is to ensure the safety of agents, as well as suspects and other individuals, by reducing the potential for suspects to react to the notification of warrant service. However, in October 1995, ATF decided that a dynamic entry could be planned only after all other tactical options had been considered. According to officials, DEA and FBI agents and specialized teams also often use dynamic entries during high-risk search and arrest warrant operations to ensure the safety of agents and suspects. (See p. 49.) The equipment available for use by ATF agents during dynamic entries generally includes weaponry; breaching equipment, such as battering rams; and/or other tactical equipment designed for safety, such as ballistic vests, helmets, and body bunkers. Body bunkers are ballistic shields that can provide additional protection to agents as they enter and search premises. In addition to the equipment available to all ATF agents, SRTs have access to additional firearms, such as bolt-action rifles, and specialized tactical equipment, such as diversionary devices. GAO's review of SRT deployment reports for fiscal year 1995 showed that diversionary devices represented the specialized equipment most often used by SRTs. (See p. 57.) Vehicles used by SRTs generally included vans and trucks that were used for deployments and storing equipment. GAO's review of fiscal year 1995 SRT deployment reports, found that SRTs most often used aircraft to obtain aerial photography of suspects' locations. In only one fiscal year 1995 operation did an SRT use aircraft to transport agents. ATF equipment, vehicles, aircraft, and clothing used by SRT's are generally comparable to those used by DEA and FBI agents during similar operations on the basis of GAO's observations at divisions, discussions with agencies' officials, review of ATF equipment lists, and analysis of SRT deployment reports. (See pp. 60 and 63.) -------------------- \1 ATF did not compile data for suspects armed at arrest for fiscal years 1990 and 1991. ATF COMPLIED WITH ITS PROCEDURES FOR INVESTIGATING SHOOTING AND ALLEGED EXCESSIVE FORCE INCIDENTS -------------------------------------------------------- Chapter 0:4.4 ATF's procedures for reporting, investigating, and reviewing shooting and excessive force incidents were revised in October 1994. The revisions included changes to shooting incident reporting requirements and procedures, including which types of shooting incidents are and are not to be reported. ATF procedures are consistent with guidelines and/or standards recommended by the International Association of Chiefs of Police, the President's Council on Integrity and Efficiency, and the Commission on Accreditation for Law Enforcement Agencies. For example, agents are required to immediately report shooting incidents to their supervisors, incidents are to be investigated by an independent unit, and certain reports are to be reviewed by a review board on the basis of the nature and seriousness of the incident. (See p. 65.) Overall, DEA and FBI procedures for reporting, investigating, and reviewing shooting incidents are comparable to ATF's. For example, each agency requires that investigations address similar issues and that reports contain similar types of information. Distinctions in the procedures include (1) DEA and FBI delegate some investigative responsibilities to their field divisions--ATF does not delegate such responsibility and (2) DEA's and FBI's review boards include representatives from Justice--ATF's review board does not include representatives from Treasury. While ATF's excessive force procedures are comparable to DEA's with one distinction relating again to delegation, they are distinct from those employed by FBI. ATF is to investigate allegations of excessive force first and--if warranted--is to refer them to Justice for possible criminal investigation. In contrast, FBI refers all allegations of excessive force to Justice's Civil Rights Division for possible criminal investigation before investigating the allegations itself. (See p. 73.) GAO's review of documents in ATF's investigative files for reported shooting and excessive use-of-force incidents, for fiscal years 1990 through 1995, showed that ATF complied with its investigative procedures except that two investigative files did not include a required record of review\2 by the designated unit at ATF headquarters. GAO's review also showed that ATF's investigations of 38 reported shootings involving ATF agents' discharging their weapons at suspects found each to be justified and within the scope of its use-of-force policy. In addition, ATF's investigations found that 18 of 25 reported allegations of excessive force were unsubstantiated. Four investigations found evidence of some agent misconduct, such as an agent who confiscated drugs from an informant but failed to turn in or report the drugs. Two of the investigations were ongoing at the time of GAO's review, and one was closed without action because ATF determined that there was no need for adjudication. ATF agents found to have engaged in misconduct received written reprimands and suspensions. (See p. 83.) Due to time and/or methodological constraints, GAO did not evaluate the events that resulted in the incidents or the quality and adequacy of ATF's investigations. In addition, GAO did not verify whether all shooting or excessive force incidents were reported or whether all reported excessive force incidents were investigated. Accordingly, GAO's conclusions about ATF's compliance with its investigative procedures are based on GAO's review of ATF investigative file documentation required by these procedures and apply only to these files. (See pp. 83 and 90.) GAO did not review the shootings arising from ATF's operation at the Branch Davidians' compound in Waco, TX. This incident was independently investigated by Treasury, which identified shortcomings in ATF's operations and reported its findings in September 1993. In October 1995, ATF reported the actions it had taken in response to Treasury's findings. (See p. 25.) -------------------- \2 A document maintained in the investigative file that indicates who reviewed the file and, where applicable, any annotated comments resulting from the review. RECOMMENDATIONS ---------------------------------------------------------- Chapter 0:5 GAO is making no recommendations in this report. AGENCY COMMENTS ---------------------------------------------------------- Chapter 0:6 GAO requested comments on a draft of this report from the Secretary of the Treasury and the Attorney General. In separate meetings on March 1, 1996, the Senior Advisor to the Under Secretary of the Treasury for Enforcement and ATF officials provided Treasury's comments, and the Director of the Audit Liaison Office under the Assistant Attorney General for Administration provided Justice's comments. Also present at the Justice meeting were officials from the Office of the Attorney General, the Office of the Deputy Attorney General, the Criminal Division, DEA, and FBI. Overall, the Treasury and Justice officials either characterized the report as balanced, accurate, and thorough or had no comments. They provided some technical comments that GAO has incorporated in this report, where appropriate. (See p. 27.) INTRODUCTION ============================================================ Chapter 1 According to the International Association of Chiefs of Police (IACP), managing officers' use of force is one of the most difficult challenges facing law enforcement agencies. The ability of officers to enforce the law, protect the public, and guard their own safety is very difficult in an environment in which violent crime is commonplace and firearms are frequently used for illegal purposes. For the agents of the Bureau of Alcohol, Tobacco and Firearms (ATF), suspected illegal firearms activities are the leading cause for their initiating enforcement actions. Over the past several years, ATF has come under public criticism and congressional scrutiny primarily as a result of its operation at the Branch Davidians' compound in Waco, TX, and citizen accusations that ATF agents used excessive force in carrying out their enforcement responsibilities. The February 1993 operation at the Branch Davidians' compound was initiated to serve an arrest warrant on David Koresh, the Davidians' leader, and to execute a search warrant on the compound. When Koresh refused to accept the warrants, ATF tried to forcibly enter the compound using a tactic known as dynamic entry but simultaneously was met with gunfire from the Branch Davidians. In the ensuing gun battle, four ATF agents and six Branch Davidians were killed. OVERVIEW OF ATF ENFORCEMENT ACTIVITIES ---------------------------------------------------------- Chapter 1:1 ATF is a law enforcement agency within the Department of the Treasury with responsibilities directed toward reducing violent crime, collecting revenue, and protecting the public. ATF enforces the federal laws and regulations relating to alcohol, tobacco, firearms, explosives, and arson. Among its missions, ATF is to work directly and in cooperation with others to (1) suppress and prevent crime and violence through enforcement, regulation, and community outreach and (2) support and assist federal, state, local, and international law enforcement. To accomplish its criminal enforcement responsibilities, ATF has 24 field divisions, headed by special agents-in-charge (SAC), located throughout the United States. As of September 1995, ATF had a total of 1,944 special agents of which 1,777 were assigned to its field divisions. ATF's special agents are to initiate criminal investigations when notified of suspected illegal activities by such sources as informants; undercover operatives; and referrals from ATF inspectors and other federal, state, and local law enforcement agencies. Figure 1.1 is an ATF organization chart, as of January 1996, that depicts the principal units discussed in this report. Figure 1.1: ATF Organization Chart, as of January 1996 (See figure in printed edition.) Source: ATF. As table 1.1 shows, the vast majority of ATF's enforcement activities have been directed at suspects who are believed to be engaged in illegal firearms activities. Suspicion of illegal firearms activities is the principal reason that initiates ATF firearms investigations. Table 1.1 Investigations Initiated by ATF, FYs 1990-1995 FY FY FY FY FY FY Investigation type 1990 1991 1992 1993 1994 1995 ---------------------------- ----- ----- ----- ----- ----- ----- Firearms 9,753 11,74 12,59 10,48 9,227 7,606 5 8 9 Arson 668 808 1,462 1,955 1,629 1,739 Explosives 1,192 1,079 952 873 888 890 Tobacco 8 15 13 14 19 33 Alcohol 12 9 10 17 9 29 Other 108 79 82 131 193 208 ====================================================================== Total 11,74 13,73 15,11 13,47 11,96 10,50 1 5 7 9 5 5 ---------------------------------------------------------------------- Source: ATF. On the basis of its investigations, ATF apprehends individuals that it suspects of criminal violations. As can be seen from table 1.2, most of the individuals ATF arrested were suspected of violating firearms laws. Table 1.2 Suspects Arrested by ATF, FYs 1990-1995 FY FY FY FY FY FY Investigation type 1990 1991 1992 1993 1994 1995 ---------------------------- ----- ----- ----- ----- ----- ----- Firearms 6,355 7,599 7,971 6,092 8,687 6,034 Arson 276 293 286 275 477 304 Explosives 422 365 342 218 374 284 Tobacco 8 3 3 3 10 21 Alcohol 1 10 9 4 0 12 Other 42 31 20 15 47 37 ====================================================================== Total 7,104 8,301 8,631 6,607 9,595 6,692 ---------------------------------------------------------------------- Source: ATF. WHAT ARE USE OF FORCE AND DEADLY FORCE? ---------------------------------------------------------- Chapter 1:2 According to IACP, use of force has been construed to include a wide range of techniques used to compel compliance. Such techniques range from verbal persuasion--the lowest force level-- to deadly force--the most severe force level--and everything in between, including physical force, stun guns, tear gas, batons, and other nonlethal equipment. The variety of coercive options available to agents in a confrontational setting is often referred to as the "force continuum." According to Treasury policy on the use of force, the primary consideration in its use is the timely and effective application of the appropriate level of force required to establish and maintain lawful control. ATF training materials note that the use of force by law enforcement officers in the performance of their duties has been traditionally limited to four categories. Under these categories, use of force is allowed if necessary to overcome resistance to the officer's lawful commands, effect an arrest or detain a suspect, maintain custody and prevent escape, or protect the officer or other persons. Furthermore, the training materials note that in determining how much force may be or should be used, the officer should consider such factors as the nature of the offense for which the suspect is being arrested, e.g., felony or misdemeanor arrest; number of participants on each side; size, age, and condition of participants; record and/or reputation of the suspect for violence; use of alcohol or drugs by the suspect; suspect's mental or psychiatric history; presence of innocent bystanders; and availability of less violent or nonlethal weapons. In October 1995, Treasury and the Department of Justice adopted use of deadly force guidelines that are uniform with the exception of certain agency mission-specific provisions. For example, while warning shots generally are not permitted under the Treasury and Justice policies, the U.S. Secret Service may use warning shots in exercising its protective responsibilities, the U.S. Customs Service may use warning shots on the open waters, and Justice agencies may use warning shots under certain circumstances within the prison context. In addressing the subject of nondeadly force, the Treasury and Justice uniform policies recognize that if force other than deadly force reasonably appears to be sufficient to accomplish an arrest or otherwise accomplish the law enforcement purpose, deadly force is not necessary. In commenting on the policy, both Departments committed to take all reasonable steps to prevent the need to use deadly force. Both Departments define deadly force as the use of any force that is likely to cause death or serious physical injury. Therefore, any firearms discharge that is intended to disable a suspect is considered to be deadly force. However, the use of other weapons, including those considered nonlethal, could be construed as a law enforcement officer's having used deadly force, depending on the manner in which the weapon was used. For example, hitting a suspect in the head with a baton is considered to be the use of deadly force. OBJECTIVES, SCOPE, AND METHODOLOGY ---------------------------------------------------------- Chapter 1:3 In August 1995, the Chairman of the Subcommittee on Treasury, Postal Service, and General Government, House Committee on Appropriations, asked us to (1) identify and describe ATF's policies for the use of deadly force; (2) determine how ATF conveys its policies to its agents; (3) determine the reasons for and the extent to which ATF uses dynamic entry and the equipment ATF uses to accomplish these entries; and (4) determine whether ATF has complied with its procedures for investigating shooting and alleged excessive use-of-force incidents. While we determined whether ATF complied with procedures for investigating shooting and alleged excessive use-of-force incidents on the basis of a review of case file documents, we did not evaluate the quality and adequacy of ATF's investigations. In addition, except for a limited check discussed later, we did not verify whether all shooting and alleged excessive force incidents were reported or whether all reported allegations of excessive force were investigated. The Chairman also asked us to compare how ATF addresses the above issues with the way that Justice's Federal Bureau of Investigation (FBI) and the Drug Enforcement Administration (DEA) address them. In addition, we were asked to determine (1) whether ATF applies lessons learned from its reviews of shootings and allegations of excessive use of force and (2) what authority ATF has to take adverse personnel actions against agents, particularly in connection with excessive use-of-force incidents. OBJECTIVE 1 -------------------------------------------------------- Chapter 1:3.1 To identify and describe ATF's policies on the use of deadly force and compare them to DEA's and FBI's, we reviewed pertinent Treasury, Justice, ATF, DEA, and FBI policies and accompanying commentaries on the use of deadly force that were available. Also, we reviewed certain relevant U.S. Supreme Court and lower court decisions involving the use of deadly force. We also interviewed appropriate ATF, DEA, and FBI officials concerning their policies on the use of deadly force. OBJECTIVE 2 -------------------------------------------------------- Chapter 1:3.2 To determine how ATF conveys its deadly force policies to its agents, we (1) visited the Federal Law Enforcement Training Center (FLETC) and ATF's National Academy in Glynco, GA; (2) observed training facilities, equipment, and ongoing classes--including Basic Marksmanship, Judgment Pistol Shooting, Situational Response, Non-Lethal Control Techniques, and Tactical Operations Planning; and (3) spoke with officials about the overall training courses provided new agents, in general, and the use-of-force and firearms training provided, in particular. We also reviewed teaching guides and student training materials for FLETC's Criminal Investigator Training Course and ATF's New Agent Course. At the time of our visit to FLETC, ATF did not have any students enrolled in criminal investigator training for new agents, although there were new agent classes in progress. However, we observed a class of new ATF agents attending the National Academy. We also met with FBI and DEA training officials to discuss their new agent training courses and use of deadly force training; toured the training academy at Quantico, VA, and observed the training facilities and equipment; participated in a demonstration of the Firearms Training System\1 and reviewed summaries of course materials and instructor teaching guides for training that discussed use of force. We compared use-of-force course descriptions and the types of training provided new ATF agents to course materials and DEA and FBI training officials' descriptions of the types of training provided new DEA and FBI agents. We also reviewed (1) ATF's manual for its new agent On-the-Job Training Course and identified training objectives dealing with use-of-force issues and (2) course training materials provided to Special Response Teams (SRT)--ATF's version of special weapons and tactics units--on use of force and deadly force. We identified and reviewed ATF's policies requiring use-of-force discussions during quarterly firearms training and during tactical operational planning and operation briefings. On September 28, 1995, October 13, 1995, and December 11, 1995, respectively, we observed quarterly firearms training at three ATF field divisions--Baltimore; Washington, D.C.; and Los Angeles\2 --to assure ourselves that the required use-of-force discussion took place. We also interviewed the divisions' Firearms Instructor Coordinators concerning firearms training and use-of-force training and reviewed their records documenting use-of-force discussions at prior quarters' training sessions. At the firearms qualification sessions, we also judgmentally selected and spoke with some attending ATF agents to determine whether use-of-force discussions were a regular part of tactical operation planning and operation briefings. Also, in these divisions, we interviewed Assistant Special Agents-in-Charge responsible for the SRT, SRT leaders, and group supervisors to determine whether use-of-force discussions were a part of tactical operation planning and operation briefings. Finally, we compared FLETC and ATF training materials with relevant Treasury/ATF use-of-force and deadly force policies to determine if training materials complied with the policies. While our review analyzed whether the agencies' training reflected applicable use-of-force policies, we did not assess the effectiveness of the agencies' training. In addition, we did not review new agent attendance records while at FLETC or the ATF National Academy, nor did we review agent personnel training records while we were at the field divisions. -------------------- \1 Reacting to realistic video scenarios and using laser pistols, new agents employ the Firearms Training System in learning to make split-second decisions about whether to shoot a suspect during a confrontation. \2 The Los Angeles division was chosen because of its relatively large number of SRT activations and its more extensive inventory of equipment available during high-risk operations. The remaining two divisions were chosen because of their proximity to Washington, D.C., in our effort to complete our work in a timely and cost-effective manner. OBJECTIVE 3 -------------------------------------------------------- Chapter 1:3.3 To determine the reasons for and the extent to which ATF uses dynamic entry and the equipment used in such entries, we reviewed ATF policies, procedures, and documents regarding operational planning, dynamic entries, equipment, and SRT units. Due to time constraints, we did not review a sample of all ATF enforcement actions conducted by all ATF agents to determine how often dynamic entry was used. Because (1) SRTs are to be deployed to conduct ATF's higher risk search and arrest warrants and SRTs have access to all of the equipment available to ATF agents as well as additional specialized equipment and (2) ATF maintains more thorough and readily available data on SRT operations, including tactics and equipment used, than it does on its other enforcement groups, we focused our review and analysis of ATF's use of dynamic entry and related equipment to operations involving SRT deployments. However, whenever possible, we discussed the use of dynamic entries and related equipment by non-SRT agents with headquarters and division officials. We analyzed all SRT deployment activation reports for fiscal year 1995 to identify the reason for the deployment, the extent to which the deployment used the dynamic entry tactic to enter a building, whether the deployment resulted in force being used (e.g., shootings or physical force), and whether the SRT used specialized equipment. We also obtained an ATF statistical compilation of fiscal years 1993 and 1994 SRT activation data, which showed the number of activations, the reason for deployments, deployments that resulted in force being used (e.g., shootings), and whether special equipment was used. Furthermore, we reviewed all shooting incident reports for fiscal years 1990 through 1995 and determined the number of SRT incidents in which ATF agents fired their weapons at suspects. We also interviewed ATF officials in the Special Operations Division to determine ATF's practices regarding the use of dynamic entries and other tactics as well as the type of equipment used for these operations. In addition, we discussed the use of dynamic entries with ATF division officials, SRT team leaders, Tactical Operations Officers, and agents at three ATF field divisions--Washington, D.C.; Baltimore; and Los Angeles. We reviewed training materials and observed some of the training provided new agents at FLETC regarding dynamic entries and other tactics and related equipment. We also reviewed training materials used to train new SRT members during their initial 2-week training session at Fort McClellan, AL. We analyzed all SRT training reports for fiscal year 1995 to determine the type of in-service training received, equipment used during training, and sources that provided instruction. In addition, we observed the Washington Division SRT's fourth quarter 1995 training to determine what tactical training was received. We spoke with officials from ATF's Property and Fleet Management Section and Enforcement Support Branch regarding ATF policies and controls on the equipment available for high-risk operations. We also observed the equipment maintained and issued by the Enforcement Support Branch in Rockville, MD, and the SRT equipment at the three field divisions we visited. In addition, at the Washington, D.C., and Los Angeles divisions, we obtained listings of the SRT equipment and vehicles as well as certain firearms, breaching tools, and other tactical equipment available for dynamic entries. Although we observed the SRT equipment maintained in Baltimore, we did not obtain an equipment listing because at the time of our visit the SRT had been recently merged into the Washington Division's SRT. We discussed the use and sources of this equipment with the Tactical Operations Officers, the SRT team leaders, and several agents in each of the three division offices we visited. We also obtained and reviewed comprehensive listings of rifles and tactical carbines, SRT and armored vehicles, and aircraft in inventory throughout ATF from ATF's Inventory Tracking and Equipment Management System. In addition, we visited the FBI and DEA Washington field divisions to compare ATF's use of dynamic entries and equipment to other federal law enforcement agencies. At each division, we interviewed division officials, including entry team leaders, to determine their use of dynamic entries and other tactics and observed the equipment used by the FBI Special Weapons and Tactics (SWAT) and DEA entry teams during high-risk operations. On the basis of the standardization of training provided for high-risk warrant service, both FBI and DEA officials opined that their division's use of dynamic entry and related equipment generally was representative of other field divisions in their respective agencies. We also reviewed the literature available from IACP and other law enforcement experts regarding the use of equipment, dynamic entries, and other tactics by law enforcement agencies. OBJECTIVE 4 -------------------------------------------------------- Chapter 1:3.4 To determine whether ATF complied with its procedures for investigating shooting and use-of-force incidents, we obtained and reviewed the following information: (1) procedures for reporting, investigating, and reviewing shooting and misconduct incidents; (2) policies on administering adverse personnel actions against agents found to have violated use-of-force policies; (3) policies on protecting complainants from retaliation; (4) policies on ensuring that lessons learned from investigations are transmitted to agents; and (5) investigative guidelines and/or standards recommended by IACP, the President's Council on Integrity and Efficiency (PCIE), and the Commission on Accreditation for Law Enforcement Agencies. We obtained similar information, where applicable, from DEA and FBI. We also identified and reviewed legislation, regulations, and court cases related to the use of force by law enforcement agencies. We identified and reviewed files related to the investigation of reported shooting and use of excessive force incidents during fiscal years 1990 through 1995. For shooting incidents, we identified and reviewed 38 of 39 incidents where ATF agents intentionally discharged their weapons at suspects.\3 For use of excessive force incidents, we identified and reviewed 92 investigations in three categories of alleged agent misconduct: (1) misconduct during the execution of a search warrant, (2) violation of a person's civil rights, and (3) assault by an agent on a person. Because ATF does not maintain a separate category for use of excessive force, we judgmentally selected these categories following consultations with ATF officials and a review of misconduct incident categories. The selection was based on the likelihood that these categories would include most, if not all, incidents of alleged use of excessive force. Of the 92 investigations, we found that 25 involved allegations of the physical abuse of persons and/or property. To place the shooting and use of excessive force incidents in perspective, we obtained statistics related to ATF enforcement actions, such as arrests and SRT deployments. At the request of the Subcommittee, we also obtained shooting incident and enforcement action data from DEA and FBI. However, it should be emphasized that these data were not comparable to ATF's, given the agencies' differences in missions, personnel levels, and some data definitions. These data are presented in appendix V. As agreed with the Subcommittee, we did not verify the accuracy of ATF's, DEA's, or FBI's statistical data because of time limitations. To determine ATF's compliance with its investigative procedures, we reviewed ATF's investigative files for all 38 intentional shooting incidents that were reported to and investigated by ATF from fiscal years 1990 through 1995 as well as for the 25 alleged excessive force incidents we selected. We based our compliance determination on whether the information in the files indicated that the investigative procedures had been followed. We looked for the required information on (1) the incident, such as whether it resulted in injuries or the type of law enforcement activity that resulted in the incident, and (2) the investigation, such as who conducted the investigation, who reviewed it, the types of information the investigation obtained and analyzed, and the outcome of the investigation. Where documentation was not initially found, we obtained documents and/or explanations from ATF officials and considered them in our determination. Due to time and methodological constraints, we did not evaluate the quality and adequacy of the shooting and use of excessive force investigations or the validity of their conclusions. We also did not evaluate the circumstances, such as law enforcement actions, that resulted in the shooting or use of excessive force incidents. In addition, we did not verify the accuracy of the information in ATF's files. Finally, we did not verify whether all shooting and alleged excessive force incidents were reported or whether all reported allegations of excessive force were investigated. We did, however, do a limited check related to this matter by searching a computerized news database and contacting two organizations with possible knowledge of some incidents. The results of this limited check are discussed in chapter 5. We discussed issues related to our review, including the use of excessive force, with officials from (1) ATF's Office of Inspection, Office of Chief Counsel, and Office of Enforcement; (2) DEA's Office of Inspections; (3) the FBI's Office of Inspection; and (4) organizations that monitor law enforcement practices. -------------------- \3 We did not review the Waco incident. No ATF file existed because Treasury investigated the incident. ADDITIONAL ISSUES -------------------------------------------------------- Chapter 1:3.5 To determine whether, and how, ATF applies lessons learned from its investigations, we (1) identified and reviewed the relevant sections in ATF's investigative procedures; (2) obtained from ATF and reviewed examples of lessons learned being implemented; (3) reviewed ATF's October 1995 report on the actions taken in response to the lessons learned from the Waco operation; and (4) discussed related issues with cognizant ATF officials, including the Associate Director for Enforcement. To determine ATF's authority for administering adverse actions against its personnel--including managers who perform poorly--we (1) obtained and reviewed the relevant ATF adverse action orders, (2) identified examples of personnel actions from our review of ATF's investigative files, and (3) discussed adverse action issues with staff from ATF's Employee Labor Relations Branch (ELRB) and the chairman of the unit charged with reviewing incidents that may result in adverse action being taken against ATF personnel. We also obtained relevant documentation from DEA and FBI and compared it with ATF's to identify any similarities and differences. Our review was made between August 1995 and January 1996 in accordance with generally accepted government auditing standards. We provided drafts of this report to the Secretary of Treasury and the Attorney General for comment. Responsible Treasury and Justice officials provided oral comments at separate meetings on March 1, 1996. AGENCY COMMENTS ---------------------------------------------------------- Chapter 1:4 At the March 1, 1996, meetings, the Senior Advisor to the Under Secretary of the Treasury for Enforcement and ATF officials provided Treasury's comments, and the Director of the Audit Liaison Office under the Assistant Attorney General for Administration provided Justice's comments. Also present at the Justice meeting were officials from the Office of the Attorney General, the Office of the Deputy Attorney General, the Criminal Division, DEA, and FBI. The Treasury and Justice officials either characterized the report as balanced, accurate, and thorough or had no comments. They provided some technical comments that we have incorporated in this report, where appropriate. TREASURY AND JUSTICE USE OF DEADLY FORCE POLICIES HAVE BEEN ADOPTED TO REFLECT SUPREME COURT GUIDANCE AND ARE GENERALLY CONSISTENT ============================================================ Chapter 2 State rules and Supreme Court guidance on the use of deadly force have been evolving for a number of years. Approaches among the states on the use of deadly force have ranged from those that place an emphasis on the apprehension of a fleeing felon to those that permit such force regarding dangerous suspects but with certain qualifications. Within this context, federal law enforcement agencies have, over the years, adopted policies to govern their employees' use of deadly force. In October 1995, Treasury and Justice adopted uniform policies on the use of deadly force. These uniform policies, like those they replaced, were adopted to reflect applicable Supreme Court guidance. Treasury's and Justice's commentaries,\1 in general, explain that their policies were formulated to be more restrictive on the law enforcement officer than constitutional or other legal limits. ATF's 1988 use of deadly force policy, which was in effect before the issuance of the 1995 Treasury policy, was, with two distinctions as discussed in this chapter, consistent with the new Treasury policy. In addition, the 1988 ATF policy, was, with three distinctions as discussed in this chapter, consistent with prior DEA and FBI policies. -------------------- \1 Treasury and Justice issued written comments discussing and elaborating upon their 1995 use of deadly force policies. STATE RULES AND SUPREME COURT GUIDANCE ON THE USE OF DEADLY FORCE ---------------------------------------------------------- Chapter 2:1 STATE RULES HAVE VARIED -------------------------------------------------------- Chapter 2:1.1 By 1985, the rules in the states governing the use of deadly force by law enforcement officers varied.\2 These rules can generally be grouped into three categories: (1) the common-law rule, (2) a modified common-law approach, and (3) the Model Penal Code approach. Many states followed something similar to the English common-law rule on the use of deadly force, which existed at the time of this country's founding. Generally, deadly force could be used by a law enforcement officer if necessary to arrest a felony suspect. Because the type of felony involved is not taken into account, this rule is generally referred to as the "fleeing felon" rule. An officer could use deadly force when he reasonably believed that he was justified in arresting an individual for a felony as long as the officer also reasonably believed that such force was necessary to protect himself or prevent escape. To a great extent, the rationale behind the fleeing felon rule was based on the fact that common-law felonies\3 were punishable by death, and the use of deadly force was seen as merely accelerating the penal process, albeit without providing a trial. For example, the 1982 Tennessee statute, which was found unconstitutional in a landmark Supreme Court decision discussed later, was based on the fleeing felon rule. The decision provided, in part, that "[i]f, after notice of the intention to arrest the defendant, he either flees or forcibly resists, the officer may use all the necessary means to effect the arrest."\4 Some states used a modified common-law rule by specifying the felonies for which deadly force may be used to arrest or by stating that only "forcible felonies"--also called dangerous felonies--justify the use of deadly force. An Illinois statute, for example, listed the felonies that may trigger the use of deadly force: "'Forcible felony' means treason, murder, voluntary manslaughter, aggravated criminal sexual assault, criminal sexual assault, robbery, burglary, arson, kidnapping, aggravated battery and any other felony which involves the use or threat of physical force or violence against any individual."\5 Legislatures in other states abandoned the common-law rule for some form of the Model Penal Code approach, which imposes several qualifications on the use of deadly force. The Model Penal Code, as formulated by the American Law Institute\6 in 1962, generally permits the use of deadly force only when the crime for which the arrest is made involves conduct including use or threatened use of deadly force or when there is a substantial risk that the person to be arrested will cause death or serious bodily harm if his apprehension is delayed. More specifically, under the Model Penal Code approach, the use of deadly force is not justified unless (1) the arrest is for a felony, (2) the actor effecting the arrest is a peace officer or is assisting a peace officer, (3) the actor believes such force creates no substantial risk of injury to innocent persons, and (4) the actor believes that the felony included the use or threatened use of deadly force or there is a substantial risk that the suspect will cause death or serious bodily harm if apprehension is delayed.\7 The Supreme Court noted, in 1985, that while there was not a constant or overwhelming trend away from the common-law rule, a long-term movement has been away from the emphasis that deadly force may be used against any fleeing felon.\8 -------------------- \2 The issue of the use of deadly force was described by a 1975 decision of the U.S. Court of Appeals for the Second Circuit as an area characterized by "shifting sands" and "obscured pathways." Jones v. Marshall, 528 F. 2d 132, 141 (2d Cir. 1975). \3 Common-law felonies were murder, rape, manslaughter, robbery, sodomy, mayhem, burglary, arson, and larceny. \4 Tenn. Code. Ann. section 40-7-108 (1982). \5 Criminal Code of 1961 section 2-8, Ill. Ann. Stat. ch. 38, section 2-8. \6 The American Law Institute was founded in 1923 and is composed of judges, law professors, and practitioners. Its objective is to encourage the fair administration of justice throughout the nation by advancing the uniformity of law, whenever practicable. This goal is accomplished by the work of the Institute in proposing Model Acts--statutory provisions that legislatures may enact in whole, in part, or not at all. \7 American Law Institute, Model Penal Code, section 3.07(2)(b). \8 Tennessee v. Garner, 471 U.S. 1, 18 (1985). TWO SUPREME COURT CASES HAVE PROVIDED SOME GUIDANCE ON THE USE OF DEADLY FORCE -------------------------------------------------------- Chapter 2:1.2 In the 1985 Tennessee v. Garner decision\9 and the 1989 Graham v. Connor decision,\10 the Supreme Court addressed the issue of when police may reasonably use deadly force and provided some clarification as to how courts should examine allegations that law enforcement officers have used excessive force. In Garner, a police officer shot and killed Edward Garner to prevent his escape from the scene of a burglary, even though Garner did not appear to be armed. Garner, after being told to halt, tried to climb over a fence at night in the backyard of a house he was suspected of burglarizing. With the aid of a flashlight, the officer was able to see Garner's face and hands. Even though the officer saw no sign of a weapon, he shot Garner in order to prevent his escape. The officer argued that his actions were reasonable under a Tennessee statue that provided that a law enforcement officer could use any means necessary to make an arrest. The Supreme Court, basing its determination on a Fourth Amendment\11 balancing test, struck down Tennessee's statute to the extent that it authorized the use of deadly force against all fleeing suspected felons, including nondangerous suspects. The Court noted that a seizure occurs whenever an officer restrains the freedom of a person to walk away.\12 The constitutionality of a seizure is determined by balancing the nature and quality of the intrusion on the individual's Fourth Amendment interests against the importance of the governmental interests alleged to justify the intrusion. The Court then reasoned that an apprehension using deadly force is also a seizure subject to the reasonableness requirement of the Fourth Amendment. In using this balancing test, the Court was not persuaded that shooting nondangerous fleeing suspects is so vital as to outweigh the suspect's interest in his own life. Thus, the Court found that the use of deadly force to prevent the escape of all felony suspects, whatever the circumstances, is constitutionally unreasonable. Hence, the Court found Tennessee's statute unconstitutional to the extent that it authorized the use of deadly force against nondangerous fleeing suspects. However, the Court also found that it is not constitutionally unreasonable to prevent escape by using deadly force in certain limited circumstances: "Where the officer has probable cause to believe that the suspect poses a threat of serious physical harm, either to the officer or to others, it is not constitutionally unreasonable to prevent escape by using deadly force. Thus, if the suspect threatens the officer with a weapon or there is probable cause to believe that he has committed a crime involving the infliction or threatened infliction of serious physical harm, deadly force may be used if necessary to prevent escape, and if, where feasible, some warning has been given." In the 1989 Graham decision,\13 the Supreme Court provided some clarification as to how courts should examine allegations that law enforcement officers have used excessive force. In Graham, a diabetic felt the onset of an insulin reaction and drove with a friend to a convenience store to purchase orange juice. Upon entering the store and seeing the number of people ahead of him at the checkout line, Graham hurried out of the store to go to a friend's house instead. A police officer became suspicious, followed Graham's car, and made an investigative stop. Backup police officers arrived, handcuffed Graham, and ignored Graham's attempts to explain and treat his diabetic condition. Graham sustained various physical injuries during the incident, was thrown headfirst into the police car, and the officers refused to let him have some orange juice as a remedy for his condition. Graham was later released when the officers learned that nothing had happened at the store. Graham brought an action against the officers involved in the incident alleging that the officers had used excessive force in making the investigatory stop. The District Court applied a four-factor test\14 and ruled in the officers' favor. The Court of Appeals for the Fourth Circuit, without attempting to identify the specific constitutional provision under which Graham's claim arose, endorsed the four-factor test applied by the District Court and affirmed the District Court decision. However, the Supreme Court ruled that the lower courts had applied the incorrect legal standard and remanded the case in order for the Court of Appeals to consider the claim under the Fourth Amendment's reasonableness standard. In doing so, the Court declared that it was making "explicit what was implicit in the Garner analysis--that all claims alleging that law enforcement officers have used excessive force--deadly or not--in the course of an arrest, investigatory stop, or other seizure of a free citizen should be analyzed under the Fourth Amendment's 'reasonableness' standard . . . ." The Court explained that determining whether the force used to effect a particular seizure is "reasonable" under the Fourth Amendment requires a careful balancing of "the nature and quality of the intrusion on the individual's Fourth Amendment interests against the countervailing governmental interests at stake." While recognizing that the test of reasonableness under the Fourth Amendment is not capable of precise definition or mechanical application, the Court explained that its proper application requires careful attention to the facts and circumstances of each particular case, such as the severity of the crime at issue, whether the suspect poses an immediate threat to the safety of the officers or others, and whether the suspect is actively resisting or attempting to evade arrest by flight. Among other things, the Court noted that the reasonableness of a particular use of force must be judged from the perspective of a reasonable officer on the scene, rather than with the 20-20 vision of hindsight. The Court further noted that the calculus of reasonableness must allow for the fact that police officers are often forced to make split-second judgments--in circumstances that are tense, uncertain, and rapidly evolving--about the amount of force that may be necessary in a particular situation. -------------------- \9 Garner, 471 U.S. 1 (1985). \10 Graham v. Connor, 490 U.S. 386 (1989). \11 The Fourth Amendment contains a prohibition against unreasonable seizures of the person. The Supreme Court has stated that the Fourth Amendment's proper function is to constrain, not against all intrusions as such, but against intrusions that are not justified in the circumstances, or that are made in an improper manner. Schmerber v. California, 384 U.S. 757, 768 (1966). \12 Garner, 471 U.S. at 7 (1985). It is well settled that police officers may use some degree of force in effectuating a lawful arrest. Graham, 490 U.S. at 396 (1989). \13 Graham, 490 U.S. 386 (1989). \14 The District Court considered the following four factors: (1) the need for the application of force, (2) the relationship between that need and the amount of force that was used, (3) the extent of the injury inflicted, and (4) whether the force was applied in a good faith effort to maintain and restore discipline or maliciously and sadistically for the very purpose of causing harm. TREASURY AND JUSTICE HAVE ADOPTED UNIFORM POLICIES ON THE USE OF DEADLY FORCE ---------------------------------------------------------- Chapter 2:2 In October 1995, Treasury and Justice adopted use of deadly force policies to standardize the various policies their component agencies had adopted over the years. The policies are uniform with the exception of certain agency mission-specific provisions covering, for example, Justice's prisoner-related responsibilities. Justice's Resolution 14, which created the Justice uniform policy, notes that in view of Supreme Court decisions addressing constitutional restrictions on the use of deadly force,\15 Justice's investigative agencies have, over the years, adopted policies to govern their employees' use of deadly force, albeit in a manner that was not standardized. Both Justice and Treasury note in their commentaries that the policies are intended to maintain uniformity among their various respective departmental components and to achieve uniform standards and training with respect to the use of deadly force. While components may develop and conduct their own training on the use of deadly force, the commentaries state that the new uniform policies govern the use of deadly force under all circumstances. The Justice and Treasury uniform policies provide that their respective officers may use deadly force only when necessary, that is, when the officer has a reasonable belief that the subject of such force poses an imminent danger of death or serious physical injury to the officer or another person. The Treasury and Justice commentaries, in general, explain that their policies were formulated to be more restrictive on the law enforcement officer than constitutional or other legal limits. Following are Treasury's and Justice's 1995 policies: Treasury: "Treasury Law Enforcement Officers may use deadly force only when necessary, that is, when the officer has a reasonable belief that the subject of such force poses an imminent danger of death or serious physical injury to the officer or to another person."\16 Justice: "Law enforcement officers and correctional officers of the Department of Justice may use deadly force only when necessary, that is, when the officer has a reasonable belief that the subject of such force poses an imminent danger of death or serious physical injury to the officer or to another person."\17 Accompanying Treasury and Justice commentary provide that "probable cause," "reason to believe," or a "reasonable belief," for purposes of their policies, mean facts and circumstances, including reasonable inferences, known to the officer at the time of the use of deadly force, that would cause a reasonable officer to conclude that the point at issue is probably true. The commentaries also recognize that the reasonableness of a belief or decision must be viewed from the perspective of the officer on the scene, who may often be forced to make split-second decisions in circumstances that are tense, unpredictable, and rapidly evolving. Justice and Treasury commentaries also state that as used in their respective policies, "imminent" has a broader meaning than "immediate" or "instantaneous." The commentaries further state that the concept of "imminent" should be understood to be elastic, that is, involving a period of time dependent on the circumstances, rather than the fixed point of time implicit in the concept of "immediate" or "instantaneous." Thus, a subject may pose an imminent danger even if he or she is not at that very moment pointing a weapon at the officer if, for example, he or she has a weapon within reach or is running for cover carrying a weapon or running to a place where the officer has reason to believe a weapon is available. In addition, the policies provide that if force other than deadly force appears to be sufficient to accomplish an arrest or otherwise accomplish the law enforcement purpose, deadly force is not necessary. The commentaries further provide that if force less than deadly force could reasonably be expected to accomplish the same end, such as the arrest of a dangerous fleeing subject, without unreasonably increasing the danger to the officer or others, then it must be used. -------------------- \15 The Treasury and Justice commentaries explain that in developing the policies, it became apparent that decisional law provides only limited guidance regarding the use of deadly force. \16 The Department of the Treasury, Uniform Policy on the Use of Force, Treasury Order No. 105-12, section (4)(a) (1995). \17 U.S. Department of Justice, Uniform Policy on the Use of Deadly Force, section I (1995). ATF'S 1988 POLICY ON THE USE OF DEADLY FORCE WAS GENERALLY CONSISTENT WITH THE 1995 TREASURY POLICY ---------------------------------------------------------- Chapter 2:3 The 1988 ATF use of deadly force policy, which was in effect before the issuance of the October 1995 Treasury uniform policy, was, with two distinctions, consistent with the October 1995 Treasury policy. ATF's 1988 use of deadly force policy stated that "A firearm may be discharged when the special agent believes that there is no other means of control and perceives an imminent threat of death or serious bodily injury to himself/herself or other innocent persons."\18 The 1988 ATF and 1995 Treasury policies are consistent in that both policies generally authorize the use of such force only when the law enforcement officer reasonably believes or perceives that there is an imminent threat or danger of death or serious physical injury to the officer or another person. Moreover, both the 1988 ATF and 1995 Treasury policies limit the degree of force authorized to that which is needed to accomplish the law enforcement purpose. More specifically, the 1988 ATF policy provided that the degree of force authorized was limited to that which was necessary to establish lawful order and control in a timely manner, and the 1995 Treasury policy provides that if force other than deadly force appears to be sufficient to accomplish an arrest or otherwise accomplish the law enforcement purpose, deadly force is not necessary. One distinction between the policies is that the 1995 Treasury policy refers to the use of "deadly force" while the 1988 ATF policy referred more specifically only to the use of a "firearm." With respect to the 1988 ATF policy, an ATF official noted that until 1995, firearms were the only equipment issued to ATF agents that could inflict deadly force. A second distinction is that while the 1995 Treasury policy allows for the use of deadly force only when the law enforcement officer has a "reasonable belief" that there is an imminent threat of death or serious physical injury, the 1988 ATF policy allowed for the use of such force when the special agent "perceives" an imminent threat of death or serious physical injury. An ATF official noted that, under the 1988 policy, the special agent's perception of an imminent threat would have been within the context of additional policy language which provided that "the authority to bear firearms carries with it an obligation to exercise discipline, restraint, and good judgement." -------------------- \18 ATF Order No. 3000.8, ch. A, section (3)(a) (1988). ATF'S 1988 POLICY ON THE USE OF DEADLY FORCE WAS GENERALLY CONSISTENT WITH PRIOR DEA AND FBI POLICIES ---------------------------------------------------------- Chapter 2:4 The 1988 ATF use of deadly force policy was, with three distinctions, consistent with DEA and FBI policies in effect prior to the issuance of the 1995 uniform policies. Following are DEA's and FBI's prior policies: DEA: "Agents are not to shoot any person except in self-defense, when they reasonably believe they or another person are in danger of death or grievous bodily harm."\19 FBI: "Agents are not to use deadly force against any person except as necessary in self-defense or the defense of another, when they have reason to believe they or another are in danger of death or grievous bodily harm."\20 The prior ATF policy was consistent with prior DEA and FBI policies in that they generally authorized the use of deadly force only when the agents reasonably believed or perceived that there was a threat or danger of death or serious bodily harm to the agent or another person. One distinction among the three policies was that the ATF policy alone provided the additional qualifying restriction that such threat be "imminent." In addition, the aforementioned "firearm/deadly force" and "reasonably believes/perceives" distinctions also existed among the prior policies. More specifically, (1) the ATF and DEA policies referred to the shooting of a "firearm" while the FBI policy used the term "deadly force" and (2) the ATF policy used the term "perceives" while the DEA and FBI used the terms "reasonably believe" and "reason to believe," respectively. The policies described in this chapter contain additional guidance regarding specific situations, such as fleeing persons, escaping prisoners, verbal warnings, warning shots, and shooting at or from vehicles. This additional information can be found in appendix I. -------------------- \19 Drug Enforcement Administration, Policy on the Use of Deadly Force, section 6122.13 (A). \20 Manual of Investigative Operations and Guidelines, Part II - Section 12. FBI materials assert that this policy is more restrictive than the constitutional standard announced in Garner. CONCLUSIONS ---------------------------------------------------------- Chapter 2:5 As Supreme Court guidance and state rules on the use of deadly force have evolved over the years, so have the policies of federal agencies. In addition, the ability of officers to enforce the law, protect the public, and guard their own safety is a very difficult task. Officers are often forced to make split-second judgments in circumstances that have been described as tense, uncertain, and rapidly evolving. Recently, Treasury and Justice adopted uniform policies to standardize the policies of their component agencies. Since 1988, ATF has maintained a policy that was, with three distinctions as discussed in this chapter, consistent with prior FBI and DEA policies. ATF's 1988 policy was also, with two distinctions as discussed in this chapter, consistent with Treasury's current uniform policy. ATF, DEA, AND FBI CONVEY DEADLY FORCE POLICIES TO THEIR AGENTS IN SIMILAR WAYS ============================================================ Chapter 3 Use of deadly force training provided new ATF agents at FLETC\1 and the ATF National Academy reflected the ATF/Treasury deadly force policy. This policy is also to be reiterated to new agents during their probationary period when they receive on-the-job training (OJT). Furthermore, the types of deadly force training new ATF agents received at FLETC and the ATF National Academy were consistent with the types of training provided new FBI and DEA agents. Moreover, ATF policy requires that ATF agents be reminded of the deadly force policy at least quarterly throughout their careers. -------------------- \1 FLETC, which was established in 1970, is a Treasury bureau but has an interagency Board of Directors. FLETC serves as a law enforcement training organization for more than 70 federal agencies, as well as offering training programs for state and local enforcement personnel. INITIAL TRAINING IS DESIGNED TO ESTABLISH THE FOUNDATION FOR NEW ATF AGENTS' UNDERSTANDING OF USE-OF-FORCE POLICIES ---------------------------------------------------------- Chapter 3:1 The first year with ATF, new agents are to receive about 17 weeks of formal training--about 8 weeks in general criminal investigator skills and techniques at FLETC and 9 weeks in ATF-specific training at the National Academy. In addition, agents also participate in ATF's OJT for new agents. TRAINING AT FLETC -------------------------------------------------------- Chapter 3:1.1 FLETC requires that all students who attend the basic criminal investigator course be trained in Treasury's/FLETC's Use-of-Force Policy and Firearms Policy, including deadly force. Because Treasury's use-of-force and deadly force policies are applicable to all of its bureaus, the use-of-force policies taught at FLETC are generally consistent with ATF's policies. Once trained on the policies and tested on their knowledge of them, students are required to demonstrate their knowledge and apply the policies where applicable throughout their training. All new ATF agents are required to attend the FLETC's Criminal Investigator Training Program (CITP).\2 This program, which in fiscal year 1996 is to be expanded to approximately 9 weeks from slightly over 8 weeks in fiscal year 1995, provides basic training in a broad range of skills that criminal investigators require. Among the more than 70 course topics presented are interviewing, case management, surveillance, undercover operations, crime scene investigation, fingerprints, constitutional law, court testimony, and search and seizure. All Treasury bureaus' and many non-Treasury agencies are to send their new agents to FLETC for basic criminal investigator training. CITP training consists of three methods of presentation-- classroom/lecture; laboratory, where students practice skills under an instructor's guidance; and practical exercises, where students participate in a related law enforcement scenario and demonstrate law enforcement skills. Students are to receive over 175 hours of training in the classroom/lecture, 117 hours of laboratory work, and 39 hours of practical exercises. They are to be graded in both lecture material and practical exercises. During CITP, students are to be given 5 written examinations on which they must score at least 70 percent and satisfactorily complete all required tasks during the practical exercises. -------------------- \2 Exceptions are made for new agents who have transferred from other law enforcement agencies and have completed equivalent training. USE-OF-FORCE TRAINING ------------------------------------------------------ Chapter 3:1.1.1 According to FLETC officials, in about 1990, the FLETC Use-of-Force Oversight Committee developed a use-of-force continuum model, consistent with Treasury policies, which has been used to train all students. Furthermore, the Committee recommended and FLETC agreed to integrate use-of-force issues, where applicable, into all FLETC courses. As a result, FLETC provides a 2-hour course on Firearms Policy during the first week of training that presents, among other things, the basic concepts in the use of force, including deadly force, and introduces students to FLETC's Use-of-Force Model (discussed below). Among the performance objectives of this course are that the student is to be able to (1) identify basic principles governing the use of force, (2) identify and apply the appropriate force, and (3) identify and apply the firearms policy and guidelines to hypothetical situations and practical exercises that are given throughout the training program. THE FLETC USE-OF-FORCE MODEL ------------------------------------------------------ Chapter 3:1.1.2 The FLETC Use-of-Force Model is composed of five color-coded levels of force designed to correspond to officers' perceptions of the level of threat with which they are confronted and describes the progression or de-escalation of force on the basis of the demonstrated level of compliance or resistance from a subject. Students are shown a video illustrating a situation that poses various levels of threat and emphasizes how threats in real-life situations can escalate and de-escalate from one level to another. Table 3.1 shows the levels of threat and the corresponding force represented in the Use-of-Force Model. Table 3.1 Levels of Threat and Corresponding Force Level of threat Corresponding force ---------------------------------- ---------------------------------- (1)Compliant (blue level) Communication, such as verbal commands (2)Passive resistance (green Low-level physical tactics, such level) as grabbing a suspect's arm (3)Active resistance (yellow Use of come-along holds, pressure level) points, and chemical sprays (4)Assaultive with the potential Defensive tactics, such as for bodily harm (orange level) striking maneuvers with the hands or a baton (5)Assaultive with the potential Deadly force for serious bodily harm or death (red level) ---------------------------------------------------------------------- Source: FLETC. In addition to its presentation in training courses, FLETC officials said that the Use-of-Force Model is prominently displayed in all classrooms and throughout FLETC hallways. We observed that the Use-of-Force Model was displayed in the FLETC classrooms we visited as well as in hallways, firing ranges, and the cafeteria (see fig. 3.1). Figure 3.1: The FLETC Use-of-Force Model (See figure in printed edition.) Source: FLETC. OTHER CITP COURSES THAT ADDRESS USE OF FORCE AND DEADLY FORCE ------------------------------------------------------ Chapter 3:1.1.3 Our review of FLETC CITP course materials identified seven other courses (besides the Firearms Policy course) in which use-of-force topics were presented in varying amounts. These courses were (1) Detention and Arrest, (2) Execution of a Search Warrant, (3) Judgment Pistol Shooting, (4) Situational Response, (5) Introduction to Physical Techniques, (6) Non-lethal Control Techniques, and (7) The Removal and Positioning for Transportation of Reluctant Suspects. These courses contain components whose objectives are to train students in identifying threats and use-of-force concepts, applying the proper use of force to the threat, and honing judgmental skills in applying the various levels of force, including deadly force. For example, in the Detention and Arrest course (10 hours), terms used in the use-of-force policy are to be defined, Supreme Court rulings on deadly force are to be discussed, and the FLETC Use-of-Force Model is to be presented. One of the objectives of the course is to train students to recognize the degree of force, which may include deadly force, that may be used to effect an arrest, according to Treasury/FLETC Firearms Policy. In Judgment Pistol Shooting (3 hours), students are to use a weapon that has been altered to shoot a laser beam. They are confronted by realistic video scenarios on a giant screen that require them to use proper judgment in making shoot or not-to-shoot decisions. Students are required to identify the elements of jeopardy with which they are confronted and to provide a rational explanation in each instance of questionable judgment. Failure to properly respond to the video could result in the video perpetrator's "killing" the agent or another person. To successfully pass the course, students must score 100 percent in judgment and 70 percent in shooting accuracy. In Situational Response (2 hours), students are to be given a scripted scenario in which they are placed into a situation and have to react to the threat posed by an instructor/role-player. Both students and role-players are to wear protective clothing and have weapons that fire paint bullets (commonly referred to as simunitions), which mark the target they hit. Students have to react to shoot and not-to-shoot situations and demonstrate the application of the deadly force policy. In Non-Lethal Control Techniques (30 hours), students are to be provided the basic skills required to control and arrest a compliant and a noncompliant suspect without using deadly force. As part of the training, students are required to assess the threat and resistance level of the suspect and respond with the correct level of force and control as required by the FLETC Use-of-Force Model. TRAINING AT ATF'S NATIONAL ACADEMY -------------------------------------------------------- Chapter 3:1.2 ATF students who successfully complete CITP are also required to attend New Agent Training at ATF's National Academy, which is located on the FLETC campus. New Agent Training is 9 weeks and focuses on the laws, policies, procedures, and specialized investigative techniques that are specific to ATF and designed to orientate new agents to their roles as special agents. Our review of course materials identified three courses that address ATF's use of deadly force policy:\3 (1) Firearms Usage Policy, (2) Situational Response for ATF New Agent Training, and (3) Tactical Operations Planning. A portion of the Firearms Usage Policy course is to be devoted to reiterating ATF's use-of-force and deadly force policies and the authority and limitations that agents bear in exercising the use of force. In Situational Response for ATF New Agent Training (4 hours), students are to work in two- or more person teams with simunitions and participate in 7 realistic scenarios. These scenarios are designed from real-life experiences of ATF agents to challenge the students' ability to make proper decisions regarding the use of force, tactics, the use of cover, and, if appropriate, marksmanship. In this course, emphasis is placed on resolving the exercises with the use of surprise, speed, and, if necessary, violence without the use of deadly force. Among the objectives of the course are to allow the students to (1) demonstrate the ability to make proper deadly force decisions, (2) utilize the principles of tactics to gain control of situations and to avoid shootouts, and (3) demonstrate the ability to articulate a rational explanation of shooting decisions. In Tactical Operations Planning (2 hours), students are to compile the necessary intelligence to develop and execute tactical plans relating to search warrants, high-risk search warrants, arrests, and undercover operations. As part of the training in operational briefings, students are to be taught that the Treasury/ATF Firearms Policy is required to be presented before the execution of tactical plans. -------------------- \3 Use-of-force policies, including deadly force, are contained in ATF's Firearms Policy, ATF Order 3000.8. ATF ON-THE-JOB TRAINING -------------------------------------------------------- Chapter 3:1.3 During the first year with ATF, new agents are supposed to complete phase 1 of the training program. For phase 1, new agents, in addition to attending CITP and New Agent Training, are to continue their training at their post of duty under the guidance of an experienced agent who has been designated as an OJT instructor. The purpose of OJT is to acquaint the new agent with applicable policies and procedures and to expose the trainee to various investigative activities. During this period, the trainee is expected to display the appropriate skill, knowledge, and judgment that is needed during their involvement in situations related to the training objective, such as during an arrest situation. For example, during OJT on arrest procedures, students are expected to understand the limitations on the use of force and the restrictions and limitations on using firearms when making an arrest. The student is also to understand and demonstrate proper custody and control of subjects. TRAINING PROVIDED NEW ATF AGENTS IS CONSISTENT WITH THE TYPES OF TRAINING DEA AND FBI PROVIDED TO THEIR NEW AGENTS ---------------------------------------------------------- Chapter 3:2 The types of training provided to new ATF agents in fiscal year 1995 to introduce them to and train them in the use-of-force and deadly force policies was consistent with the types of training provided to DEA and FBI new agents. Each agency provided new agent trainees with an initial 2-hour classroom lecture\4 and discussion describing with examples the agency's use-of-force/deadly force policy within the first week of the training. (App. II provides FLETC and FBI excerpts from nine training scenarios in which the use of force was used and the rationale for whether the force used was appropriate.) Thereafter, each agency employed a building-block approach that integrates the use-of-force/deadly force issues into other segments of the training in which the use of force could be a relevant issue, such as physical control techniques, arrests, and on search and seizure training. Each of the agencies employed training techniques, such as practical exercises using role-playing, simunitions exercises, and firearms judgment exercises that use realistic video scenarios requiring shoot or not-to-shoot decisions. Furthermore, each agency trained its new agents in recognizing the perceived level of threat they face and in responding to it with an appropriate level of force. FBI training officials stated that DEA and FBI have similar use-of-force training programs for new agents. One official noted that although the exact language of their policies might be somewhat different (the Treasury and Justice policies applicable to DEA and FBI were revised in October 1995 and are now generally uniform for DEA and FBI), DEA and FBI interpret and apply their use-of-force policies almost identically in training programs. This official said that he instructs on legal issues for the use-of-force policy course for FBI's new agents and has taught DEA's new agents as well and that he had also provided training assistance at FLETC. -------------------- \4 According to FBI officials, for fiscal year 1996, legal instruction on the deadly force policy was being expanded to 4 hours during the first week of new agent training. AGENTS ARE TO BE KEPT AWARE OF USE-OF-FORCE POLICIES AT TACTICAL OPERATIONS BRIEFINGS AND AT QUARTERLY FIREARMS QUALIFICATIONS ---------------------------------------------------------- Chapter 3:3 Even after the training of new agents is completed and agents are provided full special agent responsibilities, they are to be frequently exposed to the use-of-force and deadly force policies. ATF policy requires that these policies be reiterated during the planning process for tactical operations and quarterly during firearms requalification training. TACTICAL OPERATIONS PLANS AND BRIEFINGS -------------------------------------------------------- Chapter 3:3.1 For over a decade, ATF policy has required that for every search warrant obtained, a plan is to be developed to execute the warrant and that all persons participating in the warrant are to be briefed on the plan. Moreover, the policy requires that every person participating in the plan, especially those who are not Treasury enforcement officers, are to be advised of Treasury's policy on the use of firearms. A January 27, 1995, ATF policy brief stated that due to the increase in violence encountered by agents during the execution of search warrants, arrest warrants, and undercover operations, special agents planning to execute such operations are required to prepare an operational plan. The ATF guidance for operational plans stated that "the use of a well written operational plan, in concert with a thorough briefing, substantially enhances the safety of the special agents, public, and suspects." ATF policy requires that all enforcement officers involved in the operation be provided a copy of the operational plan. Among the issues to be discussed at the operational plan briefing is ATF's firearms policy on the use of deadly force. The plan, which is to be prepared on a standardized form, contains a block that is to be checked when the policy is discussed. ATF agents with whom we spoke at the Washington, Los Angeles, and Baltimore divisions stated that the firearms policy on use of deadly force was reiterated before all operations. QUARTERLY FIREARMS TRAINING -------------------------------------------------------- Chapter 3:3.2 ATF's firearms policy requires all special agents to qualify in marksmanship with their primary duty firearm each quarter. Agents who fail to meet minimum qualification requirements are not to be certified to use that weapon until they requalify. Because the requirement applies to the primary duty weapon, each agent,\5 even those in supervisory positions at headquarters, must attend and qualify each quarter. Furthermore, the ATF policy requires that as part of each firearms training session, no less than 1 hour of instruction is to be provided on ATF firearms/ammunition standards and procedures and the use-of-force policy. Each Firearms Instructor Coordinator is required to document the training provided and certify that the firearms policy instruction was provided. We observed one quarterly firearms training session at each of the Washington, Los Angeles, and Baltimore divisions. At each session, the divisions' Firearms Instructor Coordinator read the use-of-force and deadly force policies to the agents. At the fourth quarter of the Washington Division's fiscal year 1995 qualification, the coordinator elaborated on various points in the policies, such as the prohibitions against firing at moving vehicles and firing warning shots. At the Los Angeles Division's first quarter of fiscal year 1996 qualification, the coordinator reviewed the revised October 1995 Treasury use of deadly force policy and confirmed that all agents had received copies of the new policy. He also gave the agents a quiz that included questions on the policy, among other topics. At the Washington, Los Angeles, and Baltimore divisions, we discussed with the Firearms Instructor Coordinators the training they provided and also reviewed their records, including quarterly firearms training documentation, to determine if the use-of-force policies were discussed at all quarterly firearms training in fiscal year 1995. Our review of ATF agent firearms qualification records for the Washington Division showed that at each quarterly session the division's coordinator had certified on the records to reviewing ATF's firearms policy. In Los Angeles, the division's former Firearms Instructor Coordinator said that he reviewed the policy at each quarterly qualification session during fiscal year 1995 and had every agent sign their qualification records to attest that they understood the policy. However, Los Angeles' new coordinator for 1996, said that instead of having agents' attest that they understood the policy, he would meet the policy requirement by certifying on the agents' qualification record that the use-of-force policy had been reviewed. At the Baltimore Division, the Firearms Instructor Coordinator had instituted, on his own initiative, a new computerized firearms qualification record form for each division agent. The computerized document did not show whether the firearms policy had been discussed. When asked about whether the policy had been discussed at each session, the coordinator said that he had discussed it and that, henceforth, he intended to certify to doing so in his written records. Furthermore, documentation at each of these offices showed that the firearms policy had been discussed at qualification sessions going back to at least the early 1990s. DEA and FBI officials confirmed that deadly force policies are to be reiterated at their quarterly firearms qualifications. -------------------- \5 Exceptions to the qualification requirement can be permitted by the SAC due to medical or leave reasons or where operational duties do not permit (e.g., agents who are working in certain undercover capacities or agents who are required to be in court when the qualifications take place). DISSEMINATING THE REVISED TREASURY USE-OF-FORCE POLICY ---------------------------------------------------------- Chapter 3:4 In October 1995, shortly after Treasury revised its use-of-force policy to make it uniform among its components and with the Justice policy, ATF sent the revised policy to all of its field divisions. In his cover letter transmitting the revised policy, ATF's Associate Director for Enforcement pointed out that the policy sets forth uniform standards for the use of deadly force and provides broad guidelines for all Treasury enforcement agencies. Moreover, he emphasized that the uniform policy was effective immediately and that it was the responsibility of each supervisor to ensure that all special agents under their supervision receive a copy of the policy. The letter also stipulated that the policy should be addressed at the next quarterly firearms qualification. Agents we spoke with in the Washington, Los Angeles, and Baltimore divisions all confirmed that supervising agents discussed the revised Treasury use-of-force policy with agents under their supervision. And, as noted above, we observed the Los Angeles Division's quarterly firearms qualification in which the Firearms Instructor Coordinator discussed the new policy. CONCLUSIONS ---------------------------------------------------------- Chapter 3:5 ATF conveys its deadly force policies to new agents through training. Use-of-force and deadly force training provided new ATF agents reflected Treasury/ATF policies. The types of training new ATF agents receive were consistent with the training provided DEA and FBI new agents. Furthermore, ATF policy requires that the use-of-force and deadly force policies be reiterated to agents throughout their careers during quarterly firearms qualifications and tactical operations briefings. According to DEA and FBI officials, their use-of-force policies are also to be reiterated during firearms qualification. ATF, DEA, AND FBI OFTEN USE DYNAMIC ENTRIES AND COMPARABLE EQUIPMENT TO EXECUTE HIGH-RISK WARRANTS ============================================================ Chapter 4 Dynamic entry has been a principal tactical procedure ATF has used to gain entry to premises when executing search and arrest warrants in high-risk operations. ATF believes dynamic entry is a useful tactic that can reduce the potential for injury to both agents and suspects in particular situations. However, on the basis of Treasury's report on the Waco operation\1 and views of tactical operations experts and ATF's own personnel, ATF decided in October 1995 that dynamic entry would only be planned after all other options have been considered and began to adjust its training accordingly. Similarly, according to DEA and FBI Washington Division officials, their agencies use dynamic entry when necessary to execute high-risk warrants and believe the tactic promotes safety. ATF, DEA, and FBI use generally comparable weaponry and equipment to effect dynamic entries. The exceptions are noted in this chapter. In addition, all three agencies have aircraft that can be used for intelligence and surveillance operations, such as obtaining aerial photography, and their specialized teams generally have similar vehicles, such as sports utility vehicles, from which they can deploy and in which they store equipment. The clothing and additional gear worn by agents of all three agencies when executing warrants are designed to promote agent safety. -------------------- \1 Report of the Department of the Treasury on the Bureau of Alcohol, Tobacco, and Firearms Investigation of Vernon Wayne Howell also known as David Koresh, September 1993. DYNAMIC ENTRY IS A PRINCIPAL TACTIC USED BY ATF DURING HIGH-RISK SEARCH AND ARREST WARRANT OPERATIONS ---------------------------------------------------------- Chapter 4:1 Dynamic entry is one of several tactical procedures used by ATF to gain entry to premises to execute search and arrest warrants. Dynamic entry, which may involve a forced entry, relies on speed and surprise and often is used during high-risk operations, such as ones where suspects pose a threat of violence, or where evidence can be easily destroyed. Both ATF's case agents and SRTs are to be trained in the dynamic entry technique. However, ATF did not compile any statistics regarding the number of times various tactics, such as a dynamic entry, were used during enforcement operations, according to ATF officials. Due to time constraints, we did not review a sample of all ATF enforcement operations to determine how often various tactics were used. However, we discussed the use of dynamic entries with ATF headquarters and division officials who all agreed that dynamic entry was the principal tactic used by ATF agents during high-risk search and arrest warrant operations. Furthermore, as agreed with the Subcommittee, since SRTs are to be deployed to conduct ATF's higher risk search and arrest warrants and have access to all of the equipment available to ATF agents as well as additional specialized equipment, we primarily focused our review of ATF's use of dynamic entry and related equipment on operations involving SRT deployments. Our review of SRT deployments for fiscal year 1995 found that the dynamic entry technique was used almost half of the time and was the predominant technique used when an entry to a premise was required. Moreover, during the period we reviewed, when the dynamic entry technique was used, no SRT member fired a weapon at a suspect. EXECUTING HIGH-RISK SEARCH AND ARREST WARRANTS WAS THE PRIMARY PURPOSE OF SRT DEPLOYMENTS -------------------------------------------------------- Chapter 4:1.1 ATF has established an SRT in each of its 24 criminal enforcement field divisions\2 to conduct high-risk operations. These situations include high-risk arrest, search, and undercover operations. SRT membership is voluntary and a part-time duty and ranges from 11 to 20 ATF agents depending on the location of the team. ATF defines high-risk situations, in which activation of the SRT should be considered, as those in which an increased propensity for violence exists based on the nature of the subject, the monetary value of the transaction, or the underlying circumstances of the situation. Some of the factors to be used to determine whether the SRT should be deployed include the suspect's criminal history and propensity for armed violence, the weapons expected at the location, and the fortification of the buildings involved. SRTs are to be deployed at the discretion of the SAC of the division to ensure the safety of ATF agents, other law enforcement officers, and the public during high-risk operations. As seen in table 4.1, SRTs were most often deployed to execute search and/or arrest warrants. Table 4.1 SRT Deployments, FYs 1993-1995 Percen Percen Percen Reason for deployment Number t Number t Number t ---------------------- ------ ------ ------ ------ ------ ------ Search warrant 150 69 97 66 68 43 Arrest warrant 21 10 13 9 23 15 Both search and arrest 25 12 22 15 38 24 warrants (at same time) Other (undercover 22 10 16 11 28 18 operations, etc.) ====================================================================== Total 218 100\a 48 1 00\a 1 57 100\a 1 ---------------------------------------------------------------------- \a Does not add to 100 percent due to rounding. Source: ATF for fiscal years 1993-1994. GAO analysis of SRT After Action Reports for fiscal year 1995. -------------------- \2 In fiscal year 1996, ATF plans to restructure the SRTs to more effectively use its resources from 24 divisional teams to 5 regional teams consisting of about 40 agents each. Each regional team is expected to consist of several squads of about seven agents located geographically throughout its region. The teams are expected to obtain more multifunctional training, in areas such as surveillance, as well as continue to receive tactical training (e.g., entry techniques). DYNAMIC ENTRIES AND OTHER TACTICS USED TO PROMOTE SAFETY -------------------------------------------------------- Chapter 4:1.2 According to ATF, DEA, and FBI officials as well as training literature prepared by IACP and local law enforcement agencies, dynamic entry is a principal tactic available to law enforcement agencies for use during high-risk enforcement operations. According to these officials, the primary purpose for using dynamic entry is to ensure the safety of law enforcement personnel as well as suspects and other individuals during a high-risk operation. In addition, dynamic entries may be a preferred tactic when the possibility exists that evidence may be destroyed. Furthermore, these officials as well as other law enforcement officials agree that two characteristics of dynamic entry are speed and surprise. Various law enforcement articles as well as ATF, DEA, and FBI officials assert that law enforcement studies show reaction time to be slower than action time--it takes longer for individuals to respond to a threat than to make a threat. Thus, through the use of dynamic entries in certain high-risk situations, law enforcement agents hope to act so quickly that the suspects do not have time to respond or, at a minimum, give agents the advantage by forcing suspects to react to agent actions rather than the reverse. While dynamic entries may require a forcible entry, such as breaking down a door, dynamic entries can also be accomplished through open or unlocked doors. Dynamic entry is only one of several tactical techniques ATF agents can use to execute search and arrest warrants or conduct other enforcement operations. Additional techniques include "stealth" or "static" entries, which involve slow, methodical entry and movement in a premise during which each area or room is cleared of danger before proceeding (e.g., when a suspect opens the door in response to the knock and announcement, agents may arrest or detain the suspect and then slowly clear the remainder of the premise of danger before proceeding with a search); containment call-outs, which are situations where agents surround a location and, from covered positions, contact the suspect and order the person to exit the premise; ruses, such as when agents create a ploy to draw the person out of their premise before making an entry or arresting them; and arresting or detaining the suspect away from the location (e.g., vehicle stops) before making an entry or, in the case of an arrest, to avoid having to make an entry. Whether an entry is required after using certain tactics, such as containment call-outs or ruses, would depend on the purpose of the operation. If a search warrant needs to be executed, even after arresting or detaining the suspect outside of the premise, an entry may still need to be made. According to ATF agents and our review of SRT deployment reports for fiscal year 1995, agents often conducted a stealth or static entry, rather than a dynamic entry, after detaining or arresting the primary suspect. According to ATF, DEA, and FBI officials, flexibility in tactical operations is important. Thus, agents may use a combination of these tactics or change tactics during an operation, as necessary. For example, ATF training materials and officials stressed that even after a decision is made to use a dynamic entry, a situation can emerge in the middle of an operation that dictates a change in tactics. Accordingly, during one SRT deployment in fiscal year 1995, agents planned to conduct a dynamic entry to execute arrest and search warrants. However, after the SRT's arrival at the primary suspect's home, the suspect was located in the backyard and detained before the SRT entered the premise. The agents then used a stealth entry to execute the search warrant. (App. III provides detailed examples of actual SRT operations in which these various tactics were used.) According to ATF, DEA, and FBI officials, the decision regarding whether to use dynamic entry or another technique is dictated by the unique circumstances presented in each operation, with safety as the primary objective. ATF, DEA, and FBI officials agreed that factors, such as the suspect's criminal history and violent tendencies, the location of the premise, and the amount of fortification expected, are to be considered when determining whether dynamic entry or another tactic should be used. In 1994, ATF developed and began requiring an Operational Risk Assessment form to be completed by agents when planning an operation.\3 This document is designed to identify critical elements that can affect high-risk tactical operations. The assessment is divided into four categories, including the type of enforcement activity, the suspect's criminal history, the weapons possessed by the suspect, and the suspect's location. According to ATF agents, factors, such as the suspect's violent tendencies, the location of the premise, and the amount of fortification expected, also are considered. The factors developed in the assessment are assigned point values and are totaled to determine the amount of risk believed to be present in the operation. Depending on the amount of risk present, a decision is made whether to use the SRT to accomplish the operation. Above a certain point total, deployment of the SRT is highly recommended. Below that point total, deployment is to be considered but is optional. The information gathered for the risk assessment represents critical intelligence information needed by ATF agents to develop an operational plan (discussed below). According to one SRT leader, one specific factor, if present, would not dictate the use of a particular tactic. Instead, he said agents consider the totality of the factors in a situation when developing the operational plan. For example, because a person is expected to be armed and has a history of violence, does not result in the SRT employing the same tactic in each case. Other factors also present are considered in conjunction with what is known about the suspect--possibly leading agents to chose different tactics in each case. Thus, according to the SRT leader, agents must consider all factors and determine the most appropriate tactic for each operation. For example, agents have to consider whether a prolonged containment call-out could result in needed evidence being destroyed; whether the surrounding neighborhood would have to be evacuated; or, if the location is in certain high-crime neighborhoods, whether a call-out would create a more dangerous situation (e.g., sniping or a riot). In January 1995, ATF began requiring agents to complete a standard written planning document for enforcement operations called an operational plan. According to ATF documents, ATF instituted this change due to the increase in violence ATF agents encountered during the execution of search and arrest warrants and undercover operations. An operational plan is required before executing any search or arrest warrants or conducting certain undercover operations.\4 The plan is to specify the tactics and personnel to be used during the operation. According to ATF policy, the agent(s) responsible for the planning portion of the operation prepares the plan, and the group supervisor or Resident Agent-in-Charge (RAC) reviews and approves it. According to division officials, in operations involving the SRT, the SRT team leader and/or other SRT members are to develop the plan, which is then to be reviewed and approved by the assistant SAC responsible for supervising the SRT.\5 Copies of the approved plans are to be sent to the SAC of the division. Since the decision to use a dynamic entry or other technique requires consideration of the unique factors present in each situation, ATF has not had any specific policies regarding the use of dynamic entry. However, as a result of Treasury's review of the Waco operation and the views of tactical operations experts and ATF's own personnel, ATF decided in October 1995 to implement lessons learned from Waco. One change ATF decided to make was that dynamic entry would only be planned after all other options had been considered. Given the choice, it was decided that the first tactical option to be considered during operational planning would be a ruse--luring the suspect out. It was believed that luring the suspect out would reduce the risks to the public and agents and ensure a safe, peaceful resolution to the situation. Regardless of whether dynamic entry or another technique is used, ATF agents are required to follow Treasury's use-of-force policy as discussed in chapter 2. Also, ATF agents generally are required by law to knock and announce their identity and purpose before executing search and arrest warrants. Courts, nevertheless, have permitted unannounced entries in certain exigent circumstances. Several agents stated that knocking and announcing also helps to protect their safety--through identifying themselves and their purpose they sometimes can prevent a situation in which the suspect might react without knowing that they are law enforcement officers. -------------------- \3 Some divisions were using similar risk assessment forms before being directed to do so by the Tactical Response Branch. \4 Previously, ATF agents were required to complete a written plan for search warrant operations. However, there was no standard planning requirement throughout ATF for all enforcement operations, such as arrest warrants and undercover operations. Also, previous ATF guidance permitted plans to be in outline form. \5 The only exception to mandatory completion of the operational plan is on SRT activations in which the SRT leader chooses to complete a five-paragraph order. According to ATF officials, a five-paragraph order is a type of operational planning document that is commonly used by special weapons and tactics teams and contains all of the information required in ATF's operational plan. ATF AGENTS AND SRTS ARE TO BE TRAINED IN DYNAMIC ENTRY AND OTHER TACTICS -------------------------------------------------------- Chapter 4:1.3 ATF training is to emphasize that no one tactic is an absolute and that a number of factors, beyond the agents' control, will influence whether a dynamic entry or other technique is best. During New Agent Training, ATF agents are to be taught to differentiate between situations that require a dynamic or stealth/static entry. Agents are to be taught to consider various factors--such as the characteristics of the suspect, location, and weapons expected--when determining the best tactics to employ during an operation. Agents are to participate in simulation exercises during which they are required to determine and use the appropriate tactics to gain control of situations. ATF course materials emphasize resolving these exercises by using tactics involving speed, surprise, and, if necessary, violence without employing deadly force. During our visit to the ATF National Academy, we were able to confirm through observation that a class of new agents were trained on dynamic entries through practical exercises. Agents assigned to SRTs are to receive additional training in tactics, including dynamic entries, during their 2 weeks of SRT basic training at Fort McClellan. Among the more than 20 course topics to be presented are tactical shooting, hostage situations, vehicle assaults, felony vehicle stops, and tactics. Over half of the instruction time allotted during this training is to cover tactics that include team entry and movement techniques during high-risk operations. However, according to an October 1995 ATF report, the basic SRT course curriculum has been revised to include instruction on some of the techniques needed to conduct containment call-out operations. This change was initiated by ATF, on the basis of lessons learned from Waco, to address situations in which there was no evidence in the suspect's premise that could be easily destroyed. The new basic SRT training also is to emphasize that dynamic entries are to be planned only after all other tactical options have been considered. SRTs also are to continue to train on these skills at regularly scheduled in-service sessions. SRTs are required to receive a minimum of 8 hours training each month, or 24 hours each quarter, in addition to the quarterly firearms training to be received by all ATF agents. This continuing training must be provided by qualified sources, which includes other law enforcement agencies. We reviewed SRT quarterly in-service training records for fiscal year 1995 and determined that all 24 SRTs conducted training on entry techniques, such as dynamic and/or stealth/static entries, including practical exercises frequently using simunitions or live fire rounds. However, we did not confirm whether all SRT members attended all training sessions. Our review also showed that most of the training was conducted by ATF instructors. The most common non-ATF source of training was state and local law enforcement agencies. For example, we observed the Washington Division's SRT training for the fourth quarter of 1995. Two days of this training was conducted by representatives of the Los Angeles County Sheriff's Department's Special Enforcement Branch--the SWAT team--and primarily consisted of instruction and practical exercises on stealth entry techniques. In 10 instances, Department of Defense or National Guard units provided some training to SRTs but generally not regarding entry tactics. ATF forward observers\6 received most of the training provided by Department of Defense or National Guard units, covering subjects such as winter survival and surveillance techniques. Furthermore, our review showed that some SRTs' quarterly training exceeded the minimum hourly requirements, such as training conducted by the Los Angeles and Washington divisions' SRTs. -------------------- \6 ATF forward observers support SRT operations by providing up-to-date intelligence and by ensuring the safety of participating enforcement personnel. Their initial training at Fort McClellan emphasizes weapons selection, marksmanship, and observation techniques. ATF MOST OFTEN USED DYNAMIC ENTRY TACTIC IN HIGH-RISK SEARCH AND ARREST WARRANT SERVICES -------------------------------------------------------- Chapter 4:1.4 According to ATF division and headquarters officials, dynamic entry has been the primary technique used by both SRTs and non-SRT agents to gain entry to premises when executing high-risk search and arrest warrants. We reviewed reports of each of the total 157 SRT deployments during fiscal year 1995. During these deployments, about 185 suspects were arrested. As seen in figure 4.1, SRTs used the dynamic entry technique almost half the time during these deployments, and it was the predominant tactic used when an entry to a premise was made. However, none of the 77 deployments in 1995 in which SRTs used dynamic entries resulted in ATF agents' firing their weapons at suspects, according to the deployment reports. During only one SRT deployment, involving a "buy/bust" undercover operation in an open area where no entry was required, did ATF agents fire their weapons at suspects. During this incident, three suspects were wounded. Figure 4.1: Primary Entry Tactic Used During SRT Deployments in FY 1995 (See figure in printed edition.) Note: Does not add to 100 percent due to rounding. \a Includes situations such as undercover operations in open areas, vehicle stops, and arrests following containment call-outs or ruses in which search warrants were not executed. \b Includes situations such as entering a premise to execute a search warrant after detaining or arresting a suspect via a vehicle stop, containment call-out, or ruse. Source: GAO analysis of fiscal year 1995 SRT After Action Reports. Furthermore, from fiscal years 1993 through 1995, ATF conducted 35,949 investigations, arrested 22,894 suspects, and deployed SRTs 523 times. During this same period, as seen in table 4.2, SRT members were involved in 3 intentional\7 shooting incidents, one of which resulted in fatalities. Table 4.2 SRT Intentional Shooting Incidents, FYs 1993-1995 FY FY FY 1993 1994 1995 ---------------------------------------------- ------ ------ ------ SRT intentional shooting incidents 1\a 1 1 ---------------------------------------------------------------------- \a The Waco operation. Source: GAO analysis of ATF Shooting Incident Reports, fiscal years 1993 through 1995. According to several ATF agents, one reason they used dynamic entries so frequently was the type of suspects they generally encountered. According to these agents, with most of their enforcement investigations involving firearms, ATF suspects frequently had previous convictions, were known to have committed or are suspected of past violent acts, and were believed to be armed. According to ATF reports of its firearms enforcement investigations from fiscal years 1990 to 1995, 46 percent of the suspects ATF arrested had previous felony convictions, 24 percent had a history of violence, and 18 percent were armed at the time of their arrests.\8 According to one SRT leader, ATF encounters basically three types of suspects, those (1) who do not attempt, or even consider, doing any