Shielded from Justice: Police Brutality and Accountability in the United States
New Data on Federal Prosecutions and Sentencing
To obtain more detailed information regarding prosecution under federal criminal civil rights statutes, Human Rights Watch worked with the Transactional Records Access Clearinghouse (TRAC), a private data collection and analysis agency that has collected Justice Department records through the Freedom of Information Act. To our knowledge, these data have not been made public prior to this report's publication.201 Tables of the TRAC analysis are in Appendices A through C, and some material appears below and in individual city chapters.
The TRAC data show exactly how rare it is for police officers to be federally prosecuted for civil rights violations, district-by-district. They also show that, despite applying the same law for the same type of offense, each district varies dramatically in the number of complaints referred for prosecution and the number pursued. Finally, the TRAC data show that the officers who are occasionally prosecuted successfully - presumably for the most serious crimes and with the strongest evidence - do not generally receive long sentences.
The Civil Rights Division does not compile information broken down by federal district, but the Justice Department's Executive Office of United States Attorneys(EOUSA) does.202 Unfortunately, the EOUSA said its data are not broken down so that cases involving law enforcement agents can be examined separately.203 The office, therefore, said it could not provide the information requested by Human Rights Watch regarding the number of referrals for federal criminal civil rights prosecutions of law enforcement agents acting under color of law for the fourteen U.S. districts examined in this study.204 The EOUSA wrote: "In response to your Freedom of Information Act and/or Privacy Act request, a search for records located in this office has revealed no record for the information you seek. We have been advised by our Case Management staff that our computer system cannot identify defendants who are law enforcement agents."205 In other words, both offices are maintaining somewhat similar, yet incomplete, data: by failing to monitor criminal prosecutions of law enforcement personnel in their respective federal districts, these offices cloud assessment of the adequacies of their own law enforcement efforts.
To add to the difficulty in analyzing the data compiled by the Justice Department, a third agency - the FBI's civil rights section, which is responsible for investigating possible federal criminal civil rights violations - compiles its own statistics that do not match those compiled by either the Civil Rights Division or the EOUSA at the Justice Department. For example, in Fiscal Year 1996, the FBI reports initiating 3,700 civil rights investigations (with approximately 70 percent, or 2,590, related to law enforcement officers).206 Yet the chart provided by the Civil Rights Division of the Justice Department states that the FBI initiated 2,619 civil rights investigations, without a breakdown provided for law enforcement. Neither the FBI nor the Civil Rights Division could provide a reason for this discrepancy. Said the FBI representative, "[Y]ou'd think we'd all be counting from the same page."207 The FBI notes that it initiates its own preliminary investigations in the vast majority of cases, meaning that cases passed from Justice Department headquarters are not necessarily acted upon because the FBI typically will have already started investigating the complaint. The FBI's local field offices receive complaints from the public, and learn about high-profile cases from local newspapers.
According to the data provided by the EOUSA and analyzed by TRAC, prosecution rates differ greatly among the ninety-four different federal districts in the nation. TRAC collected and analyzed data on prosecution rates for federal criminal civil rights violations under 18 U.S. Code, sections 241 and 242 between 1992 and 1995.208 Of the districts including the cities examined in this report, the federal district of Georgia North (including Atlanta) decided on how to proceed with 133 cases during this four-year period, and prosecuted none. Rhode Island (including Providence) decided on 164 and prosecuted three. California North (including San Francisco) decided on 342 and prosecuted two. And Louisiana East (including New Orleans) decided on 819 cases and prosecuted just nine. Prosecution efforts were much more likely in California Central (including Los Angeles), where of thirty-nine cases considered, twelve were prosecuted, and in Pennsylvania East (including Philadelphia), where of fifty cases considered, thirty were prosecuted.209 Although these numbers represent just a piece of the picture, they show wide variation district by district regarding the percentage of cases prosecuted by the relevant U.S. attorney (in consultation with the Civil Rights Division).
Except in rare instances, such as the Rodney King beating in Los Angeles, federal prosecutors do not pursue cases in which local prosecutors attempt but fail to indict or convict. In deciding whether to proceed with a case in which local prosecutors have failed to obtain a conviction, federal prosecutors consider whether the original trial was affected by prosecutorial incompetence, corruption, or jury tampering, whether they can introduce crucial evidence not allowed in state proceedings, or whether there is a compelling federal interest to prosecute. In practice, following a high-profile failure to indict or convict, federal prosecutorsgenerally report that they are "reviewing" the case, but that is often the last the public hears about federal action.
In the vast majority of cases, the Civil Rights Division "declines" prosecution for a variety of reasons. According to TRAC analysis of fiscal years 1994 and 1995, the most common reasons for declining prosecution were: weak or insufficient admissible evidence (this was the most common reason for both years); lack of evidence of criminal intent; declined per instructions from the Justice Department; staleness; prosecution by other authorities anticipated; statute of limitations; and lack of investigative or prosecutorial resources.210
Even in the rare cases in which police officers and others are convicted on federal criminal civil rights charges, they spend little or no time incarcerated.211 In 1994 and 1995, twenty-five defendants, out of ninety-six convicted, were sentenced to three months or less in prison (including serving no time at all). Forty-eight, or half, were sentenced to twelve months or less. These data are not linked to specific cases, making it difficult to assess whether the sentences were adequate, yet given the serious nature of the civil rights crimes prosecuted by the Justice Department, the sentences do appear to be lenient.212
The Civil Rights Division of the Justice Department provides these statistics on its activities.
In fiscal year 1997, the Civil Rights Division received a total of 10,891 complaints, with thirty-one grand juries and magistrates to consider law enforcement officers leading to twenty-five indictments and informations, involving sixty-seven law enforcement agents; nine were convicted, nineteen entered guilty pleas, and four were acquitted. It appeared that more law enforcement defendants were prosecuted per case in the same number of total cases, on average, than in previous years.
These data are of interest for several reasons. First, they show that, despite rhetoric to the contrary, the Clinton administration has neither dedicated significantly greater resources nor had much more success than previous administrations in, prosecuting law enforcement officers for civil rights violations. Between fiscal years 1993 and 1996 (four years), there were forty-four law enforcement officers convicted after trial on civil rights charges, and eighty-eight plea bargains, for a totalof 132 convictions.214 Between fiscal years 1989 and 1992 (four years primarily under President Bush's administration), there were fifty-seven convictions after trial of law enforcement officers, and sixty-eight plea bargains, for a total of 125 convictions. There does not appear to be a great difference in the outcome in these cases from one administration to the next.
Second, these data show a dramatic rise in the number of citizen complaints received by the Justice Department between fiscal years 1995 and 1996, from 8,864 to 11,721. The 1996 figure is the highest number at least since 1981 (the first year for which data were made publicly available). Unfortunately, the Civil Rights Division's data do not distinguish between types of civil rights complaints, meaning that no explanation or analysis of the 25 percent increase in complaints is possible. If police abuse complaints make up a large percentage of this jump, it may indicate an increase in abusiveness, an increase in community awareness about the Justice Department's role in handling these complaints leading to more complaints being filed, or an increase due to the large influx of new officers on the streets.
Third, the data show law enforcement officers make up almost all of the acquittals in cases prosecuted by the Civil Rights Division, yet constitute only half of the indictments in civil rights cases. These data demonstrate juries' general unwillingness to hold police officers responsible for criminal acts.
201 None of the information compiled by TRAC and cited below and in the appendices was easily available to the public. TRAC tirelessly pursues Freedom of Information Act requests and received these data in a form that required special data entry and analytical skills to be understood by the lay reader. A summary of these data is provided below, with fuller charts describing the data can be found in Appendices A through C.
202 The Civil Rights Division does compile summaries of cases it has prosecuted in a given fiscal year, and that information contains federal district information, but no tally is kept and made public, and its summaries do not include the number of referrals by district.
203 When Human Rights Watch submitted basic questions to the fourteen U.S. Attorney's offices relevant to this study, the two offices that responded directed us to the Executive Office for U.S. Attorneys at the Justice Department.
208 Those prosecuted under this statute include all types of law enforcement officers (city, state or federal law enforcement officers, sheriffs' deputies, correctional officers), magistrates and judges.
212 Under 18 U.S.C. §§241 and 242, ten years in prison and a fine are the maximum penalties when bodily injury is inflicted. When death results, the convicted officer (or other individual who commits an offense "under color of any law") shall be subject to imprisonment for any term of years or for life. It appears no life sentences were given during 1994 and 1995.
213 LE refers to Law enforcement officers. Without LE information for the number of complaints or investigations, LE breakdown on grand jury presentations is useless. Indictments are brought by grand juries, while "informations" come from magistrates.
© June 1998
Human Rights Watch