Investigation of the Seattle Police Department United States Department of Justice Civil Rights Division United States Attorney's Office Western District of Washington December 16, 2011 U.S. Department of Justice Via Hand Delivery and First Class U.S. Mail The Honorable Michael McGinn Mayor City of Seattle 600 4th Avenue, 7th Floor Seattle, WA 98124-4749 Re: Seattle Police Department Civil Rights Pattern or Practice Investigation Dear Mayor McGinn: This letter reports the findings of the United States Department of Justice Civil Rights Division's and United States Attorney's Office for the Western District of Washington's (collectively, "DOJ") joint investigation of the Seattle Police Department ("SPD" or "the Department"). Our investigation is brought pursuant to the Violent Crime Control and Law Enforcement Act of 1994, 42 U.S.C. ? 14141 ("Section 14141"), the Omnibus Crime Control and Safe Streets Act of 1968, 42 U.S.C. ? 3789d ("Safe Streets Act"), and Title VI of the Civil Rights Act of 1964, 42 U.S.C. ? 2000d ("Title VI"). These laws authorize DOJ to initiate a civil lawsuit to remedy patterns or practices of conduct by law enforcement agencies that deprive individuals of rights, privileges, or immunities secured by the Constitution or laws of the United States. As we stated in our notification letter of March 31, 2011, our investigation focused on whether SPD engaged in a pattern or practice of unconstitutional policing through (1) the use of excessive force; or (2) discriminatory policing. Before sending our letter, we met in February 2011 with you, dozens of community stakeholders, City leaders, and SPD personnel and union members. While opinions differed on the causes, scope, and depth of the challenges facing SPD, there was agreement on some overarching principles. SPD's success depends upon recruiting the right officers, and then providing them with strong and consistent leadership, training, and oversight. The structural deficiencies that we identify in this report are exacerbated by the growing number of less-experienced officers in SPD. At the outset of our investigation, approximately one-third of officers had three years or less experience, and another 350 officers were retirement-eligible (meaning even more new hires and potentially half of the force with little experience). Proper leadership, training (including mentoring), and oversight are critical for molding this next generation of SPD officers. Unfortunately, most interviewed - internally and externally to SPD - believed that one or more of these critical elements is deficient. There was a clear consensus that both the source of and solution to SPD's problems would turn on the issues of leadership, training, and oversight. These early insights were borne out by our investigation. The issues and deficiencies found in our investigation will only be remedied by sustained, consistent and engaged leadership, coming from the top and carried out through every level of leadership in SPD. -2- Our investigation finds a pattern or practice of constitutional violations regarding the use of force that result from structural problems, as well as serious concerns about biased policing. Resolution of our findings will require a written, court-enforceable agreement that sets forth remedial measures to be taken within a fixed period of time. A disciplined remedial structure will provide all interested parties with the greatest assurance that violations of constitutional rights are corrected and will not reoccur. Efforts by SPD to address the findings in this letter will not only ensure that SPD meets its obligations under the United States Constitution, but will also improve public confidence in the Department and enhance its ability to provide for the public safety of all Seattle residents. The City of Seattle and SPD were cooperative with our investigation, and we acknowledge the professionalism of all the City officials and counsel involved in this matter to date. In particular, we appreciate the openness and flexibility of City and SPD personnel during our two tours of SPD, as well as their diligence in providing requested information, including voluminous responsive documents, in a timely fashion. Consistent with our commitment to conduct the investigation in a transparent manner, we provided technical assistance and advice to SPD. This letter formalizes and provides greater detail regarding concerns raised with SPD. We are encouraged by the many hours SPD devoted to meeting with us and in providing information, and by the preliminary steps that SPD has already taken to address concerns raised by our investigation. This leaves us optimistic that we will continue our collaborative relationship to craft agreed-upon remedies for the full scope of issues set out in this letter. Finally, throughout our investigation we were mindful of the realities police officers face and the service they provide. For SPD those realities include the backdrop of the murders of five police officers in and around Seattle, and the attempted murder and wounding of a sixth officer. These deaths were the result of unprovoked, unexpected attacks against on-duty uniformed officers by members of the community. We do not underestimate the impact that these events have on all police, and particularly on SPD officers. Officers often place themselves in harm's way for the good of the community and we need to give them the tools they need to protect themselves and others. Our review of the Department was made in full appreciation for the fact that SPD must account for the risk of these types of events in its training, policies, and oversight. I. A. EXECUTIVE SUMMARY OF DOJ'S FINDINGS AND CONCLUSIONS Background and Scope of Review The great majority of the City's police officers are honorable law enforcement professionals who risk their physical safety and well-being for the public good. However, a pattern of excessive force exists as a result of a subset of officers who use force improperly, and is caused by a number of systemic deficiencies that exist in spite of SPD's recent reform efforts. -3- For many years, the City of Seattle periodically has faced accusations of police misconduct, including claims of excessive force and discriminatory policing techniques. Over the last decade, the City has responded to these allegations by implementing significant measures to improve police oversight, including developing and refining an elaborate police accountability system. Despite these efforts, recently there have been a number of widely publicized incidents involving use of force by the police, leading to understandable public concern. Our investigation was not prompted by any one particular incident. Nor did we focus on, or try to resolve the facts of, any of these high-profile events. Rather, we took a broader view. Specifically, with the assistance of our national policing experts, we systematically and thoroughly examined voluminous documents and records, including hundreds of hours of video footage, a variety of police reports, policy manuals, and SPD records related to its use of force and policing practices. This effort included obtaining and analyzing all use of force reports for the approximately twoyear period preceding our review. Moreover, we did not limit ourselves to a document review. We also conducted multiple site visits and interviewed hundreds of individuals, including community leaders, individuals alleging SPD officers had violated their constitutional rights, and SPD personnel, including front-line officers, their immediate supervisors, and command level staff. B. Findings Use of Force - We find that SPD engages in a pattern or practice of using unnecessary or excessive force, in violation of the Fourth Amendment to the United States Constitution and Section 14141. Deficiencies in SPD's training, policies, and oversight with regard to the use of force contribute to the constitutional violations. Officers lack adequate training or policies on when and how to report force and when and how to use many impact weapons (such as batons and flashlights). We also find that, starting from the top, SPD supervisors often fail to meet their responsibility to provide oversight of the use of force by individual officers. Command staff does not always provide supervisors with clear direction or expectations of how to supervise the use of force. Discriminatory Policing - We do not make a finding that SPD engages in a pattern or practice of discriminatory policing, but our investigation raises serious concerns on this issue. Some SPD policies and practices, particularly those related to pedestrian encounters, could result in unlawful policing. Moreover, many community members believe that SPD engages in discriminatory policing. This perception is rooted in a number of factors, including negative street encounters, recent well-publicized videos of force being used against people of color, incidents of overt discrimination, and concerns that the pattern of excessive force disproportionately affects minorities. This perception can significantly undermine the trust necessary for SPD to conduct effective policing in minority communities. The City and SPD need to thoroughly examine the issues raised, address the policies, procedures, and training that contribute to the problem, and conduct more sustained and effective community engagement. -4- 1. SPD's Use of Force We find that SPD engages in a pattern or practice of unnecessary or excessive force, in violation of the Fourth Amendment to the United States Constitution and Section 14141. We base our legal conclusion on numerous factual findings, including the following: ? When SPD officers use force, they do so in an unconstitutional manner nearly 20% of the time. This finding (as well as the factual findings identified below) is not based on citizen reports or complaints. Rather, it is based on a review of a randomized, stratified, and statistically valid sample of SPD's own internal use of force reports completed by officers and supervisors. ? SPD officers too quickly resort to the use of impact weapons, such as batons and flashlights. Indeed, we find that, when SPD officers use batons, 57% of the time it is either unnecessary or excessive. ? SPD officers escalate situations and use unnecessary or excessive force when arresting individuals for minor offenses. This trend is pronounced in encounters with persons with mental illnesses or those under the influence of alcohol or drugs. This is problematic because SPD estimates that 70% of use of force encounters involve these populations. ? Multiple SPD officers at a time use unnecessary or excessive force together against a single subject. Of the excessive use of force incidents we identified, 61% of the cases involved more than one officer. ? In any given year, a minority of officers account for a disproportionate number of use of force incidents. Over the more than two-year period reviewed, 11 officers used force 15 or more times, and 31 officers used force 10 or more times. In 2010, just 20 officers accounted for 18% of all force incidents. Yet, SPD has no effective supervisory techniques to better analyze why these officers use force more than other officers, whether their uses of force are necessary, or whether any of these officers would benefit from additional use of force training. This pattern or practice is also the product of inadequate policy, training and supervision. SPD fails: (1) to properly monitor or investigate the use of force; (2) to implement adequate policies on the proper use of various force weapons; and (3) to adequately train its officers on the use of force, particularly the appropriate use of various force weapons. The chain of command does not properly investigate, analyze, or demand accountability from its subordinate officers for their uses of force. In particular, we further find that the secondary review process is little more than a formality that provides no substantive oversight or accountability. Tellingly, of the approximately 1,230 internal use of force reports we received, covering the period between January 1, 2009 and April 4, 2011, only five were referred for "further review" at any level within SPD. Moreover, in our investigation, we found no case in which a first-line supervisor was held accountable for the inadequate investigation or review of a use of force incident. -5- We also find that SPD's vague Use of Force policy and inadequate training encourage pervasive underreporting and render the Department's statistics on its use of force incomplete. Finally, we find that SPD's Early Intervention System ("EIS") and its internal affairs department (its Office of Professional Accountability, "OPA") do not provide the intended backstop for the failures of the direct supervisory review process, for the following reasons: ? OPA disposes of nearly two-thirds of citizens' complaints by sending them to SPD's precincts, where the quality of investigations is, according to one OPA supervisor, admittedly "appalling." (We understand that OPA has suspended the assignment of investigations to the chain of command.) ? OPA's current classification and findings systems are so complex that they damage OPA's credibility and undermine public confidence in OPA. ? OPA consistently overuses and misuses the finding "Supervisory Intervention," which results in neither a true finding nor a remediation of the officer. We find that Supervisory Interventions are often improperly used to dispose of allegations as serious as excessive use of force and discriminatory policing simply to avoid the "stigma" of a formal finding. Although we find the structure of OPA is sound, and the investigations OPA itself conducts generally are thorough, these and other concerns render the system an additional deficiency contributing to the pattern or practice described above. Indeed, none of the uses of force our review finds to be excessive were referred to OPA for its review. It is to SPD's credit that it has been open to our discussions on these topics, and that it is in the process of revamping its review of officer uses of force and OPA's classification and findings systems. We hope these findings add a sense of focused urgency and purpose to SPD's efforts. Separately, we are aware of recent incidents involving the use of Oleoresin Capsicum ("OC") spray to disperse the so-called "Occupy Seattle" protesters on November 2, 2011 and November 15, 2011. Although these incidents concern us, we do not directly address them in this letter because they occurred outside of the timeframe of our review. However, we note that Seattle has previously been criticized for its response to demonstrators, including incidents related to the World Trade Organization meetings in 1999. In reviewing SPD's response to the WTO demonstrators, the Police Executive Research Forum noted: "There is a balance to be struck between, on the one hand, First Amendment rights and other civil liberties, and on the other hand, the interventions required to protect public safety and property." Police Management of Mass Demonstrations: Identifying Issues and Successful Approaches, PERF (2006) at 5. Our assessment of the constitutionality of police-citizen encounters in these protest scenarios is not limited to Seattle; we are paying close attention to police response to Occupy and other mass demonstrations across the country. As we resolve the issues addressed in this letter, we will review relevant information related to these events, including SPD's own review, and determine whether additional findings are necessary or appropriate. -6- 2. Discriminatory Policing Although we do not reach a finding of discriminatory policing, our investigation raises serious concerns about practices that could have a disparate impact on minority communities. These practices undermine SPD's ability to build trust among segments of Seattle's diverse communities. Our investigation revealed the following: ? SPD officers exhibit confusion between a casual, social contact and an investigative detention (a "Terry" stop). SPD must ensure its officers understand that, unless they have a sufficient factual basis to detain someone, a person is free to walk away from police and free to disregard a police request to come or stay. Officers should also understand that in such circumstances, the decision to "walk away" does not by itself create cause to detain. A person on the street is not always required to comply with police orders. While not conclusive, some data and citizen input suggest that inappropriate pedestrian encounters may disproportionately involve youth of color. ? Of the cases that we determined to be unnecessary or excessive uses of force, over 50% involved minorities. ? Analysis of limited data suggests that, in certain precincts, SPD officers may stop a disproportionate number of people of color where no offense or other police incident occurred. We further find that SPD's ability to maintain the trust of the community is hindered by SPD's: (1) deficient policies that address the risk of biased policing and or govern pedestrian stops; (2) inadequate supervision and training of its officers on (a) how to avoid biased policing practices, (b) how to conduct proper pedestrian stops, and (c) tactical communications skills; (3) a failure to proactively and consistently engage the community; and (4) the failure to keep meaningful data that would permit SPD to evaluate and take action to address allegations of biased policing. SPD appropriately encourages its officers to be proactive and engage with the community and people on the streets in a number of ways. Good policing requires regular and sustained interactions between police and the community. However, SPD must ensure that its officers understand the constitutional restrictions that guide pedestrian encounters. In light of the problems that we found, it is incumbent on SPD to take reasonable measures to correct these deficiencies, prevent the risk of discriminatory policing, especially in the context of pedestrian encounters. Of the deficiencies we identified, perhaps the most important is SPD's failure to collect and analyze data that could address and respond to the perception that some of its officers engage in discriminatory policing. We recognize that there are a number of issues raised when a government agency collects data relating to someone's racial, ethnic, or other minority status. However, other police departments have succeeded in developing effective and reasonable protocols for obtaining such data. -7- II. DOJ'S INVESTIGATION This investigation was conducted by the Civil Rights Division's Special Litigation Section and the United States Attorney's Office for the Western District of Washington. We engaged nationally recognized law enforcement professionals and a social scientist with expertise in biased policing. Their combined experience and knowledge have helped inform our findings. These professionals conducted an independent analysis of SPD policies, use of force and OPA reports, other data, and community sentiments toward SPD. The City and SPD have provided full and open cooperation in the investigation. They timely provided us with access to documents, information, and personnel. As part of our investigation, we requested, received, and reviewed from the City and SPD hundreds of thousands of pages of documents, including SPD's written policies and procedures; its training materials; its internal use of force reports; SPD and OPA's public reports; OPA's complaints and investigative files; and data generated from SPD and OPA databases. The data included several hundred hours of video footage and raw computerized data, both of which we were permitted to select and retrieve. We additionally obtained thousands of pages of documents from the public record and the community. We also conducted hundreds of interviews and meetings with SPD officers, supervisors, and command staff, as well as Seattle City officials, local community advocates and attorneys, and members of the Seattle community at large. Additionally, in May and September 2011, we and our police practices experts conducted two on-site tours of SPD, meeting with SPD command staff and a range of personnel over several days. We also conducted six full days of interviews with community members, and attended separate community meetings with local advocates and community members. This letter and the specific incidents set forth in it are not meant to be an exhaustive review of all documents and incidents reviewed. However, the examples that we provide throughout the letter serve to illustrate our findings and to aid in the efficient resolution of this matter. III. BACKGROUND ON THE CITY OF SEATTLE, SPD, AND OPA Seattle is the largest city in the state of Washington and the Pacific Northwest with a population, during the last census, of 608,660 people. According to 2010 census data, Seattle's racial and ethnic demographics are as follows: 69.5% are white, 13.8% are Asian, 7.9% are Black, 6.6% are Latino, and 0.8% are Native American. The Seattle Police Department is the largest department in Washington State, staffing approximately 1,300 sworn officers. John Diaz became the interim Chief of Police ("COP") on May 7, 2009, and on August 16, 2010, was sworn into the position permanently by Mayor Mike McGinn. SPD currently receives millions of dollars of federal grants from the Bureau of Justice Assistance, Community Oriented Policing Services, and the National Institute of Justice. For the first nine months of 2011, SPD reported that major crimes are down 7% compared with 2010, which was itself a historically low crime year. The City has a three-part police accountability system. The first, and the cornerstone of the system, is OPA, which sits within SPD. A civilian Director leads OPA and reports directly to the COP, who is the ultimate arbiter of discipline. The second part of the system is the OPA -8- Auditor, who does not sit within SPD, but serves as an independent civilian advisor to the City on the quality of OPA's investigations and SPD's policies. The third and final part of the system is the seven-member civilian OPA Review Board ("OPA-RB"). OPA-RB conducts community outreach regarding accountability issues and audits the operation of OPA by reviewing some of OPA's closed investigative files. IV. A. FINDINGS AND CONCLUSIONS Use of Excessive Force We find that SPD officers engage in a pattern or practice of unnecessary or excessive force in violation of the Fourth Amendment to the United States Constitution and Section 14141. The pattern is the result of inadequate policies, supervision, discipline and training. The use of excessive force in the course of an arrest, investigatory stop, or other seizure violates the Fourth Amendment. Graham v. Connor, 490 U.S. 386, 394-95 (1989). See also Young v. Cnty. of Los Angeles, -- F.3d --, 2011 WL 3771183, at *3 (9th Cir. Aug. 26, 2011). Courts analyze claims of excessive force under an objective reasonableness standard. Graham, 490 U.S. at 394. Assessing the reasonableness of an officer's use of force is a fact-dependent inquiry based on the "totality of the circumstances." Graham, 490 U.S. at 394-96. Courts employ a balancing test that weighs the gravity of the particular intrusion on Fourth Amendment interests against the government's need for the intrusion. Miller v. Clark Cnty., 340 F.3d 959, 964 (9th Cir. 2003); Young, 2011 WL 3771183, at *3. In assessing the government's interest in the use of force, courts will examine the severity of the crime at issue, whether the subject posed an immediate threat to officer or public safety, and whether the suspect was actively resisting arrest or attempting to escape. Id. Assessing the totality of the circumstances requires consideration of "whatever specific factors may be appropriate in a particular case, whether or not listed in Graham." Mattos v. Agarano, -- F.3d --, 2011 WL 4908374, at *6 (9th Cir. Oct. 17, 2011). We did not reach this conclusion lightly, and only after carefully examining, with the assistance of our experts, hundreds of SPD use of force incidents. Specifically, we examined a randomized, stratified, and statistically valid sample of SPD's own use of force reports for a twoand-a-quarter year period from January 1, 2009 to April 4, 2011 (consisting of hundreds of use of force incident reports), as well as dozens of other cases from various other sources, including OPA reports, public sources, and those obtained during interviews of community members. The findings we made from examining just SPD's own use of force statements are compelling. We find that approximately 20% of those incidents involved the unnecessary or excessive use of force. We also find that SPD officers were particularly prone to resorting to excessive force when employing batons, using unnecessary or excessive force 57% of the time. Additionally, we reviewed dozens of other cases that may have involved unconstitutional force, but that we could not conclusively categorize as such because of deficient reporting or incomplete evidence. Table 1 provides a visual illustration of the pattern of excessive force uncovered in our review of SPD use of force reports. -9- Table 1: Uses of Excessive Force by Weapon Primarily through our review of SPD's own documents, we find the following unconstitutional patterns in SPD's use of force: (1) the use of excessive force in the course of arrests for minor offenses; (2) the use of excessive force inflicted by multiple officers on one person; (3) the premature or excessive use of impact weapons, such as batons and flashlights; (4) the use of excessive force on subjects who were already restrained; and (5) the use of excessive force in response to individuals' expression of their First Amendment rights. Below we discuss each of these five observed patterns. 1. SPD Officers Use Excessive Force in Response to Minor Offenses. We find that SPD officers engage in a pattern or practice of using unnecessary or excessive force in the course of arresting individuals for minor offenses. Courts consider the severity of the underlying offense when determining whether a use of force is constitutional under the Fourth Amendment. See Graham, 490 U.S. at 396; Mattos, 2011 WL 4908374, at *6 (speeding and failing to sign traffic citation did not constitute serious offenses and weighed against the reasonableness of use of force); Winterrowd v. Nelson, 480 F.3d 1181, 1186 (9th Cir. 2007) (holding that officer's manual manipulation of plaintiff's injured shoulder during pat down was objectively unreasonable for non-threatening suspect detained during traffic stop); Blankenhorn v. City of Orange, 485 F.3d 463, 478 (9th Cir. 2007) (misdemeanor trespass insufficient to warrant gang-tackling plaintiff). -10- The use of excessive force to arrest individuals for minor offenses is especially problematic when, given the nature of the underlying offense, the use of verbal tactics might have defused the situation without necessitating the use of force. The Ninth Circuit has rejected "police tactic[s] that needlessly or unreasonably create a dangerous situation necessitating an escalation in the use of force." Deorle v. Rutherford, 272 F.3d 1272, 1282 n.20 (9th Cir. 2001) (citing Cunningham v. Gates, 229 F.3d 1271, 1291 n.23 (9th Cir. 2000)). Our investigation finds repeated uses of excessive force for charges related to minor offenses, including pedestrian interference, obstruction, open container violations, jaywalking, and shoplifting. In a number of incidents, failure to use tactics designed to de-escalate a situation, led to increased and unnecessary force. For example, in one incident, an officer viewed a man exhibiting irrational behavior. His stressed mental state was apparent. He was standing in the street yelling at traffic lights while holding a stuffed animal. He was sweating, his eyes were bulging, and he was talking incoherently. One officer ordered the man to move to the side of the road. The man did not respond and began to walk away, at which point the officer sprayed the man without warning with a powerful form of pepper spray. When the officer did catch up to the man, the officer reported that the man "balled up his fist." In response, the officer struck the man on the arm with a baton. The man then turned and ran. At this point, four officers chased down the man and administered between 14 to 18 punches for between 1530 seconds, five to seven elbow or knee strikes, and approximately three baton strikes, with one officer additionally striking the man on the thigh with his baton because he was reportedly attempting to kick the officers. Ultimately, they arrested the man on the minor charges of pedestrian interference and obstruction. In another incident, two officers used excessive force against a small woman who had just stolen a purse from a department store. When the woman tried to walk away from the officers, one officer grabbed her left wrist and the second officer grabbed her right arm. They bent her arms behind her back to try to place handcuffs on her, and the woman began to twist her body in an attempt to escape. Even though each officer had control of one of the woman's arms, one officer sprayed three to four bursts of OC spray to the woman's face and additionally delivered two to three punches to the woman's rib cage in response to the woman's twisting of her body and attempts to push herself up from the ground where she was pinned under the officer's knee. As a result of the officers' actions, the woman received a cut lip, stitches to her chin, and small abrasions to the right side of her face. These examples illustrate an unreasonable escalation of force in violation of federal law. See Headwaters Forest Def. v. Cnty. of Humboldt, 240 F.3d 1185, 1203 (9th Cir. 2000) ("[A] rational juror could conclude that the [] nonviolent misdemeanor offense of trespass did not render pepper spray necessary to effect the[] arrests.") We also find that SPD's use of such excessive force often involves people with mental illness, or people under the influence of drugs or alcohol. There is no question that such individuals can pose a significant safety risk to both the public and officers. Seattle has seen some horrific murders committed by mentally ill offenders. However, the erratic nature of these individuals makes de-escalation techniques even more important. Assessing the appropriate force in light of a subject's mental state is not just smart policing, it is required. Officers must take into account the subject's mental state in determining the reasonableness of the use of force. Hayes v. Cnty. of San Diego, 638 F.3d 688, 697 (9th Cir. 2011) (citing Deorle, 272 F.3d at -11- 1282). Instead of consistently attempting to de-escalate these encounters, SPD officers too often resort to force quickly and excessively when engaging with this population. This is especially problematic because, by SPD's own estimates, 70% of its use of force encounters involve this population. For example, in one incident, two officers went to the home of a man that they knew was experiencing a mental health crisis. The officers chose not to enlist the assistance of the Crisis Intervention Team ("CIT"), which would have had the requisite expertise for handling an individual experiencing distress. In addition, the individual's acute mental state was apparent from the outset of the contact. When the man answered the door, his eyes were bulging and he appeared disoriented. The officers explained to him that he was under arrest for an outstanding warrant, and one officer grabbed the man's left arm in an attempt to handcuff him. The man immediately pulled away and refused to cooperate. In response, one officer swept the man's left leg with his foot and "placed him on the ground" to gain leverage. The officer then got on top of the man's body. After the man attempted to get up several times, the officer deployed his Electronically Controlled Weapon ("ECW," a.k.a. TASER) once into the man's left leg, and administered two additional ECW cycles on the man's upper back. At this point, the man began to crawl down the hallway, with both officers on top of his back. One officer delivered two strikes to the man's face with his right elbow because he reportedly feared that the man was trying to grab his firearm. The second officer struck the man several times in the back and hip area with a closed fist, and kneed him in the lower back. In the course of this incident, the man began to vomit, stopped breathing, and suffered a brain injury that has since left him hospitalized. Had the CIT team been used, or had the officers not escalated the situation, this outcome could have been averted. In another example, two officers responded to a call to assist mental health professionals with the commitment of a man who refused to leave his house. After a few minutes of discussion, the officers told the man that he had no choice but to come with them, and one officer reached out to take hold of the man's sleeve. The man then grabbed the officer's wrist. In response, the officer raised his baton, and the second officer aimed his ECW at the man. This prompted the man to release the officer's wrist, but he then reportedly took a "fighting stance." This led to the application of the ECW twice by one officer, after which the man was struck with a baton by a second officer 10 to 12 times on his left leg, left side, and left arm. Officers were aware that they were approaching an individual with mental health issues, and that no urgency existed to commit him. Officers should have developed a plan to detain the subject that might have included calling in the CIT and employing some de-escalation tactics, instead of choosing a tactical course that quickly led to the use of force. See Deorle, 272 F.3d at 1282 (finding no substantial governmental interest in using force where there was no "immediate need to subdue" an individual before response group's arrival). These incidents highlight the need for SPD to establish special protocols when making contact with persons suffering from mental illness. We note that SPD has already recognized the need to improve its interactions with this population by directing additional resources to CIT. -12- 2. SPD Officers Use Excessive Force When They Apply Force in Tandem Against a Single Individual. We find that SPD engages in a pattern or practice of excessive force when multiple officers use excessive force against a single individual. When multiple officers use force against one person, it becomes more difficult for officers to reasonably defend the use of force as necessary out of concern for an immediate threat to their safety or for the safety of the public. See Graham, 490 U.S. at 394. Officers are also required to intercede when fellow officers violate people's constitutional rights. United States v. Koon, 34 F.3d 1416, 1447 n. 25 (9th Cir.1994), rev'd on other grounds, 518 U.S. 81 (1996); Cunningham v. Gates, 229 F.3d 1271, 1289 (9th Cir. 2000) (holding that officers can be held liable for failing to intercede when they have an opportunity to do so). The issue of the use of force by multiple officers is not new. In 2007, the former OPA director conducted an analysis of use of force from 2003 to 2005, and found a significant percentage of force complaints involved more than one officer. She stated, "[I]n 2005, more often than not, a force complaint involved more than one officer, and this was true about half of the time in 2003, and 43% of the time in 2004." Similarly, we found that in 61% of the cases in which we determined there was an unnecessary or excessive use of force, more than one officer was involved. One recent incident illustrates the problem. Four officers were dispatched to a house to investigate a stabbing at a party and, upon searching the house, located a man lying face down in bed with his arm under a pillow. The man was asleep after having had too much to drink. The man was 49 years old, 5'3" tall, 130 pounds, and did not speak English as a first language. After the man failed to comply with orders to show his hands, officers reached the conclusion that the man was dangerous because his arm was out of view and tried to arrest him. As officers made contact with the man, he began to kick at the officers while lying on his back. Officers cited this resistance as reasonable cause to use force. The four officers used excessive force as follows: one officer delivered five to seven punches to the man's face; a second officer struck the man three to four times in the head with a closed fist; a third officer struck the man three times with his knee to the side of his body; and a fourth officer struck the man multiple times with a flashlight to his midsection. The use of force by four officers on one unarmed man of relatively slight stature is excessive. The incoherent state of the subject is further illustrated by the fact that when officers took the man to the precinct for booking, he said that someone had beaten him up and thanked the SPD officer for saving him. Given his inebriated slumber, it is not clear from reports whether the man heard the officers. Officers clearly have good reason to require suspects to show their hands. Officer and public safety can be put at risk by hidden weapons. But force used to gain compliance must be gauged to the level of risk and the ultimate goal of determining whether weapons are present. Here, such an assessment did not occur. -13- 3. SPD Officers Too Quickly and Unreasonably Resort to the Use of Impact Weapons. We find that SPD officers engage in a pattern or practice of excessive use of force when they too quickly resort to employing impact weapons, such as batons and flashlights, in nonexigent circumstances (i.e., not merely as weapons of necessity). Before resorting to impact weapons, officers should consider what other tactics are at their disposal. Headwaters, 240 F.3d at 1204 (holding that, before deploying pepper spray, police "were required to consider 'what other tactics if any were available' to effect [the] arrest."); see also Smith v. City of Hemet, 394 F.3d 689, 701 (9th Cir. 2005) (availability of alternative methods of capturing or subduing suspect may be a factor to consider in assessing reasonableness of use of force). Table 1 in Section IV.A details the scope of the problem. Our expert consultants reviewed 50% of all uses of batons recorded in use of force reports, and of that statistically significant sample size, found that 57% of these uses were excessive. Additionally, our expert consultants found that 20% of the uses of "other" weapons, such as flashlights, were excessive. The findings relating to the use of batons is particularly troubling because they indicate that in a majority of cases in which SPD officers use batons, they use them in a way that either is excessive or unnecessary, in violation of the Fourth Amendment of the Constitution. For example, in the course of one arrest, an officer made contact with a man whom he suspected had put a bag of crack cocaine into his mouth. The officer used his baton to pry the man's mouth open so that a second officer could take out the drugs. This is an inappropriate use of a baton. Yet, despite the fact that the officer's supervisor identified the inappropriate use of the baton, neither the supervisor nor anyone in the chain of command identified this incident as an unreasonable use of force. In one of the incidents involving an individual with a mental illness, described in Section IV.A.1, an officer used his baton to strike a man 10 to 12 times on the leg, arm, and side, before resorting to any other weapon. The officer also used his baton after a second officer had already twice deployed his ECW. Under the circumstances, this use of force was excessive. 4. SPD Officers Use Excessive Force Against Individuals Who Are Already Under Physical Control. We find that SPD engages in a pattern or practice of using excessive force against individuals who are already under control. Under the "totality of circumstances" approach, it is more likely to be unreasonable to resort to force when a subject does not pose an immediate threat to the safety of the officer or the public. See Graham, 490 U.S. at 394. An officer should be extremely hesitant to use force against an arrestee who has already surrendered or who has been restrained or rendered helpless. LaLonde v. Cnty. of Riverside, 204 F.3d 947, 961 (9th Cir. 2001) ("[I]n a situation in which an arrestee surrenders and is rendered helpless, any reasonable officer would know that a continued use of the weapon or a refusal without cause to alleviate its harmful effects constitutes excessive force"); Blankenhorn, 485 F.3d at 480 (holding that an officer's punches were not reasonable where subject had stopped struggling). Our review of use -14- of force reports identified multiple instances in which force was used against people who were handcuffed, prone, and/or otherwise under physical control. Our investigation showed multiple incidents in which this rule was not observed. For example, in one incident, a handcuffed man was being seated in a patrol car when he started to kick his feet at two officers. One of the officers then punched the subject five times in the stomach and chest with a closed fist, rather than finding alternate ways to gain full compliance from this already restrained subject. See Graham, 490 U.S. at 396. While we recognize that feet can pose a safety risk, the amount of force used was not in scale with the level of risk presented. 5. SPD Officers Use Excessive Force Against Individuals Who "TalkBack." We find that SPD engages in a pattern or practice of using excessive force against individuals who express discontent with, or "talk back to," police officers. Similarly, SPD's use of force reports, and interviews with members of the community, reveal multiple incidents in which officers resort to the use of force when verbally confronted by individuals. It is both unconstitutional and unreasonable for officers to use force to prevent the exercise of free speech, even when such speech constitutes a verbal attack on the police. Hartman v. Moore, 547 U.S. 250, 251 (2006) (holding that official reprisal for protected speech offends the Constitution because it threatens to inhibit exercise of a protected right) (internal citations omitted); Winterrowd, 480 F.3d at 1185 (holding that a belligerent attitude and calling officers "cowards" and "thugs" did not justify the use of force). The following two incidents, which were reported to OPA by third parties and detailed in OPA's records, illustrate such incidents. In one incident, an officer contacted a male pedestrian at a bus stop. The pedestrian was swearing and heckling the officer. When the officer challenged the pedestrian and asked him "what the hell was wrong" and otherwise insulted him, the pedestrian asked if he was allowed to yell at the officer. The officer then shoved the pedestrian in his chest and shoulder, causing the pedestrian to lose his balance and stumble backward. While the officer did not ultimately arrest the pedestrian as a result of this incident, he did use force to retaliate against the pedestrian, even though the incident involved only verbal heckling and the pedestrian posed no threat to the officer. See Winterrowd, 480 F.3d at 1185 (stating that a belligerent attitude "poses no physical danger" and does not justify the use of force). In another incident, officers responded to a call regarding a disturbance. They arrested a young man for his refusal to comply with an officer's orders to go down to his knees. After they put the young man in handcuffs face down on the ground, he turned around, looked at one of the officers and said, "Go ahead, do what you got to do." After the man's reported refusal to comply and turn face down again, the officer punched the young man in his diaphragm and struck him with his right knee. Because he was already prone and handcuffed, and there was no documentation that turning his head posed any safety risk, this level of force was unnecessary. The use of excessive force in retaliation for verbal challenges underscores SPD's pattern of escalating minor situations unnecessarily. It also raises the question of whether the officers were -15- seeking compliance for safety reasons, or whether they just wanted submission from their exercise of force. B. Deficiencies Contributing to Pattern or Practice of Excessive Force We find that the following systemic deficiencies have led to the above-described pattern or practice of excessive use of force. 1. Failure to Report Use of Force. We identified multiple cases in which SPD officers failed to report the use of force at all, including incidents involving pushing and shoving. In some cases, the officers used euphemisms such as "escorts to the ground" and "guiding" suspects to the ground. Additionally, our investigation uncovered at least 17 instances in which officers were identified as using force in other officers' use of force statements, but were omitted from the summary portion of the use of force packet. This means that SPD did not track these officers' uses of force. Furthermore, half of OPA Investigation Section ("IS") investigations we reviewed that related to complaints of use of force did not have an accompanying use of force report, despite the clear application of some level of force. Officers also consistently describe their actions in use of force incidents in isolation without referencing whether other officers used force or the timing of other officers' uses of force. This hinders the ability of supervisors or OPA to determine the full scope of the use of force at any incident. We also find that when officers do report, they routinely use patterned and nondescriptive language in their use of force reports to justify their actions. For example, instead of clearly articulating the type, nature, and seriousness of resistance exhibited by the subject that preceded the use of force, officers consistently use language such as, "the subject continued to resist," or the subject "took a fighting stance" or "struggled." Additionally, we consistently saw cases in which officers justified their uses of force by reporting that an individual "refused to remove his arms from underneath his body" or "tucked his hand under his body." Obviously, as discussed above, officers have good reason to require a subject to show his hands. However, if these situations are a common cause of the need to use force, SPD should review them carefully to determine if additional training or other tactics could accomplish safe compliance. The reporting failures relating to use of force are caused, in part, by deficiencies in SPD policies relating to the reporting of use of force. Currently, SPD's policy requires that force be reported whenever an officer "uses deadly force, physical force or less lethal force as defined in Section I of th[e] policy." Department Policy & Procedures ("DP&P") 6.240.XI.A. SPD defines "physical force" as anything less than deadly or less lethal force that "causes an injury, could reasonably be expected to cause an injury, or results in complaint of injury." DP&P 6.240.I.D. The policy on its face is vague, leaves too much room for officer discretion in reporting force, and excludes the reporting of force that should be reported, as will be discussed below. These policy shortcomings are compounded by failures in training and supervision. As with other aspects of SPD officer training (discussed below), SPD supervisors and command staff are unfamiliar with the training concerning reporting obligations their officers receive at the -16- Washington State police academy (Basic Law Enforcement Academy ("BLEA")). This disconnect creates confusion as to what supervisors can and should expect from their officers when it comes time for them to report on their uses of force. Because of these problems, SPD officers have an inconsistent understanding of when force should be reported, including in high profile incidents. During our investigation, we heard from some officers that force should be reported only when an injury occurred, others believed that the policy required reporting if an arrestee complained about the force, and still other officers understood that the policy additionally required reporting when media attention was expected. The standard should be clear, and all officers should know what it is. SPD has expressed an intent to improve its use of force policies and practices, including those practices relating to the reporting of the use of force. SPD has recently implemented a 48-hour training on the use of force, and SPD policy and expectations. We encourage SPD to emphasize the importance of use of force reporting requirements in this training to address these deficiencies. We also urge the Department to continue to develop its working relationship with BLEA so that officers receive consistent training. Only when use of force policies and reporting expectations are consistent with one another can SPD accurately track its officers' uses of force. More specifically, to avoid any uncertainty, the use of force policy should require a use of force statement for any use of force above unresisted handcuffing, including the active pointing of firearms. A clear policy will also improve the reliability of SPD's internally generated statistics or reports about use of force, and militate against the "stigma" of completing a use of force report. Universal reporting standards will also provide SPD with another way of tracking officers' interactions with individuals, which can help SPD determine whether any problematic correlations exist between use of force and the race of the individual who is the subject of that use of force. 2. Failure to Develop Adequate Policies and Training Relating to Specific Force Weapons. Although SPD has a general policy regarding Use of Force (DP&P 6.240) and a specific policy regarding Firearms (DP&P 8.010-8.080), it does not have individualized policies governing the appropriate use of specific force weapons, such as the use of OC spray, batons, or the ECW. To adequately convey Department expectations regarding other uses of force, we recommend that SPD create individualized policies specific to each weapon. In particular, the policies should create clear directives about the appropriate application of these weapons, including when it is appropriate to use the weapon, how often, and the amount of force used per weapon (i.e., number of bursts of OC spray, number of cycles of an ECW, etc.). Additionally, the policy should clearly direct the preservation of evidence when these weapons are used. SPD may want to consider providing intermediary weapons to a broader swath of officers following the development of policies and training, so that officers refrain from resorting to the use of batons and the unauthorized use of flashlights so quickly. See Section IV.A.3. Impact weapons are especially dangerous, even deadly, when applied to sensitive areas of the body, such as the head. Currently, SPD only requires officers to carry a firearm, baton, and flashlight, -17- although confusion exists about this, even at the command staff level. Moreover, SPD policy is inconsistent about whether flashlights are considered an authorized less lethal use of force. Compare DP&P 6.240.I.D and 6.240.X.E. Broader use of intermediate weapons would enable officers to use less dangerous weapons when engaging a suspect. Of course, if SPD issues intermediate weapons, it should also be careful to ensure the careful reporting of their use, and effective training on using these weapons appropriately and safely. 3. Inadequate Supervision of Use of Force. Perhaps most importantly, SPD has tacitly allowed a pattern or practice of excessive use of force by failing to provide adequate supervision of force. The failure to supervise patrol officers' use of force has occurred at every level, from the first-line supervisor's (typically a sergeant) investigation and review, to the chain of command's secondary review of that investigation, to the final review by command staff. Supervisors and the entire chain of command must receive clear direction and oversight from command staff. Supervisors then must take a more active role in demanding improved reporting, more careful on-scene investigations, and more thorough secondary review. a. Inadequate First-Line Supervisory Review of Use of Force Incidents i. Insufficient On-Scene Supervisory Investigation The Use of Force policy fails to provide adequate guidance for supervisors. First, the policy does not require supervisory investigation following all uses of reportable force, and instead requires supervisors only to "respond to the scene of any use of force incident that involved three or more TASER applications and/or circumstances requiring an on-scene medical evaluation," absent extenuating circumstances. DP&P 6.240.XII.B.1. Thus, the policy allows all use of force incidents, except those narrowly defined by the policy, to occur without supervisory on-scene investigation. This means that officers have to decide when medical attention is needed. Further, this also means that supervisors often rely only on the involved officer statements in the use of force reports to determine whether a use of force was consistent with policy. Second, the policy fails to provide supervisors with guidance about how to investigate whether a use of force was unnecessary, excessive, or otherwise inconsistent with policy. The policy does not provide guidelines about how supervisors can identify discrepancies in different versions of the incident, how to compare injuries to the officer's reported force, or what to do should the supervisor have questions about the appropriateness of the use of force. Where Department policy does clearly delineate supervisory responsibilities, supervisors frequently neglect their duty to investigate the use of force. DP&P 6.240.XII.B. First, the supervisory summary routinely fails to include a detailed description of the force used by the officer and suspect, or a detailed description of all incident-related injuries, complaint of injuries, or lack of injuries. DP&P 6.240.XII.B.8.b-c. Instead, typically the supervisory summary loosely cobbles together the individual officers' reports. In some instances, the sergeant's summary simply references and directs the second-line reviewer to the officers' statements. Supervisor summaries should detail every use of force used by every officer and every injury sustained by -18- the subject, but often, much of this information is omitted. In addition, each officer at the scene should be required to describe what he or she did and saw. Given that the vast majority of police encounters involve no use of force, enhanced supervisory responsibilities should not impose undue administrative burdens. The recommendations stated above will not impose onerous additional use of force reporting requirements on officers and will improve efficiency of the review process. In addition, as issues are identified and corrective action taken, excessive use of force incidents should decline, with a commensurate decline in required reporting. Second, supervisors also consistently fail to canvass for, interview, or obtain statements from material civilian or officer witnesses. Gathering witness statements is a crucial piece of evidence collection that greatly aids a supervisor's determination of the accuracy of an officer's use of force statement and helps resolve discrepancies in accounts of the incident. The vast majority of the use of force cases that we reviewed only describe the officer's own version of events without describing other officers' actions. Interviews with the subjects of the force are also frequently inadequate, as the questioning fails to elicit what actions may have caused the officer to use force, and what amount and type of force was used by each officer. Third, supervisors often fail to take photographs of the suspect's injury "regardless of the presence or absence of visible injury," as SPD's own policies require. DP&P 6.240.XII.4-6. Finally, supervisors do not always respond to the scene of use of force incidents when onscene medical evaluations are required. DP&P 6.240.II.D. These consistent departures from SPD policy and generally accepted police practices make it difficult to adequately supervise officers. ii. Insufficient Analysis of Use of Force Incidents Effective oversight of the use of force requires adequate supervisory analysis, which we often found significantly lacking at SPD. The sergeant should piece together the sequence of events from each officer's use of force statements and other evidence the sergeant has obtained to "make sense" of what happened. Such analysis would provide a commander reviewing the sergeant's analysis a complete understanding of the incident from beginning to end, including crucially when each officer used force, why the force was necessary at each point in time, and how each injury, if any, occurred. Ultimately, this analysis should enable a sergeant to determine if the use of force was within the Department's policy. DP&P 6.240.XII.B.8. The supervisor should also make a good faith effort to resolve any discrepancies that may exist between various subject, officer, or witness statements, to determine whether an officer's use of force statement is consistent with the types of injuries sustained by the subject, and to determine whether the type of force was proportional to the resistance offered by the subject. If a supervisor is unable to assess these questions, the supervisor should conduct further investigation, or, if misconduct seems possible, refer the complaint to OPA. -19- b. Inadequate Training on How to Investigate a Use of Force Again, our investigation revealed that sergeants do not have a clear sense of their responsibilities as investigators of the use of force, as outlined above in Section IV.B.3.a, which results in unconstitutional uses of force by inadequately supervised officers. The importance of training sergeant supervisors to properly investigate use of force incidents cannot be overstated. Sergeants are first-line supervisors who respond to use of force incidents and oversee their officers' activities on a daily basis. We understand that SPD has begun to clarify expectations and improve training for sergeants, including rolling out a new sergeant's training that all SPD sergeants and the top 20 officers on the Sergeants Promotion list are required to attend by the end of 2011. However, we urge SPD to continue to develop this training by regularly meeting with other units, including the new Force Review Committee (discussed below), to determine what areas of supervisory training require the most development. SPD sergeant training should not be limited to use of force investigations. SPD sergeants should receive general training about how to conduct thorough and effective investigations, as we have also seen deficiencies in general OPA line investigations (Section IV.B.6.a.iii.(b)), as well as in investigations of biased policing allegations (Section IV.D.3.b). c. Inadequate Oversight and Review of Use of Force The Use of Force policy requires that use of force packets be forwarded through the chain of command, typically from a sergeant, to the watch Lieutenant, to the Captain of the Precinct, and to the employee's Bureau Commander, typically the Assistant Chief over Patrol Operations. DP&P 6.240.XII.B.12. We find that this process is a mere formality, almost a rubber stamp of the first-line supervisor's conclusion. We have found little evidence that management and Department executives conduct a meaningful review of the use of force. On a macro level, Appendix A shows how many times each individual SPD officer used force between January 1, 2009 and April 4, 2011. The 600+ officers are placed in random order. The chart shows that during this period, 11 officers used force 15 or more times, and 31 officers used force 10 or more times. Of all officers who used force 10 or more times, only one officer received administrative review of any kind. This statistic indicates that there is minimal supervisory oversight over officers who frequently use force. Effective supervisory techniques should focus on the relatively small number of officers who use force frequently to better understand why they use force, when they use force, and what training or other remedies, if any, are needed to minimize the use of force. The supervisory technique of collecting data and examining the activity of particularly active officers should be extended to examining the activity of officers who may be outliers in other aspects of policing, including, but not limited to pedestrian stops. Extending supervision and data collection in this way is important to reduce the number of excessive force incidents because the officers who are outliers in use of force are often the same officers who are outliers in the context of seizures such as pedestrian stops. -20- Table 2A below shows the total number of use of force events in 2010, broken down by the number of times an officer used force. It shows that the majority of the 461 officers who used force did so once or twice. A very small number of officers are using force frequently. Table 2B below shows that the 461 officers who used force represent only approximately 37% of all the sworn officers. We note that the total number of officers includes officers who are unlikely to use force due to their rank/title or assignment. Again, we make this observation to highlight the fact that the kind of enhanced supervision and oversight that SPD needs to provide officers is limited to a subset of its force. Table 2B: Majority of Sworn Officers Do Not Use Force -21- Table 2C below shows that the 20 officers who used force seven or more times participated in 18% of the use of force events for 2010. In other words, these 20 officers used proportionally 50% more force than the 24 officers who used force five to six times in a year (12% vs. 18%), and almost as much proportionally as the 105 officers who used force just twice in the year (21% vs. 18%). This happened, despite the fact that these 20 officers represent just 4.34% of the 461 officers who used force in 2010, as Table 2C below shows. Only one of these 20 officers (the same officer referenced above) received administrative review, while the remaining 19 officers did not receive any. The tables show that the potential overuse of force is not limited to one or two potentially problem officers, and provide helpful analysis that we encourage SPD supervisors to conduct. Our intent is not, however, to show that the pattern or practice of the use of excessive force is attributed only to those officers who use force more than one time in a year. An officer who uses force multiple times, or just one time in a year, may not necessarily be using excessive force. Table 2C: A Small Number of Officers Account for a Disproportionate Amount of Force Given the failures in intervention, it is not surprising that, when we met with community members, many complained that SPD does not hold its officers accountable for misconduct, and instead empowers the relatively small number of officers who do engage in misconduct. In the future, identifying the few officers who may overuse force will help prevent abuses and the appearance of wider-spread problems. We also noted that for each individual use of force report, every reviewer in the chain of command may select one of two checkboxes on a Use of Force form to indicate whether the "force appears to be within guidelines" or whether the "force incident requires further review." As noted above, many packets were plainly insufficient. Yet, of the approximately 1,230 use of force reports reviewed in the two-and-a-quarter-year time period, only five use of force packets were referred at any level for further review, which translates to approximately 0.4% of the -22- packets. No watch commander referred a use of force packet for further review in this period. Although we were told during our investigation that the chain of command may require further review informally, we have received no documentation substantiating this practice. More generally, we are aware of no case in which a first-line supervisor was held accountable for the deficient investigation or review of a use of force incident. The inadequate secondary review of use of force incidents can be attributed, in part, to the onerous responsibility left to the patrol operations bureau commander to review approximately 500 use of force reports each year. We understand that SPD has recognized the inadequacy of its current first-line investigatory and secondary review process and are pleased to learn that, in response, SPD has recently created a Force Review Committee to give commanders greater assistance in reviewing use of force reports, and in identifying problematic use of force patterns and training deficiencies. In contrast to the past practice, we understand that this Committee is demanding more investigations of use of force incidents and is routinely sending back cases to the first-line supervisors for "further review." We look forward to further discussion and refinement of the Committee's role, including the codification of its procedures. d. The Early Intervention System is Broken We find systemic deficiencies in SPD's Early Intervention System ("EIS"), which is designed to identify employees who may be experiencing "symptoms of job stress, training deficiencies and/or personal problems that may affect job performance." DP&P 3.070. An effective EIS is a powerful tool that should enable SPD to identify officers whose at-risk behaviors exceed department guidelines, even if direct supervision of use of force incidents fail or otherwise find the force incident "within policy." SPD employs indicator criteria and threshold levels to identify employees exhibiting job performance, training, or other indicators, and then attempts to intervene on the employee's behalf in a positive manner. More specifically, the EIS is triggered if an employee participates in seven uses of force, or receives three OPA complaints, in a period of six months. When an officer is identified by the EIS, the department intervenes by conducting a review of the officer's pattern of behavior that triggered the system. The EIS's failures are best exemplified by examining EIS interventions made in 2010 on behalf of two employees, Officer A and Officer B. In 2010, Officer A received three separate EIS interventions, all of which related to five OPA complaints (four of which related to use of force). This officer was also involved in 14 uses of force in a two-year time period between October 2008 and October 2010. See Appendix B for a timeline of the interventions and the precipitating incidents. Officer A's pattern of force very clearly demonstrates several flaws of the EIS system. First, the EIS thresholds are far too high and interventions on officers' behavior far too late. SPD should not allow an officer to take part in as many as seven uses of force before the department intervenes to further examine the behavior, or as many as 14 uses of force in one year. SPD should: (1) adjust EIS thresholds to identify at least 3-5% of the line officer population; (2) revise the aggregate indicator to include uses of force (DP&P 3.070.II.I); -23- (3) create single-event thresholds for events so critical that they require immediate department intervention; (4) implement rolling thresholds, thereby ensuring that an officer who has received an intervention for use of force is not permitted to engage in seven additional uses of force before again triggering EIS; and (5) include a threshold specifically related to biased policing complaints. Second, the interventions that follow an EIS trigger happen far too long after the triggering incident, which diminishes the effectiveness of the intervention and the ability to remedy an officer's behavior. For example, between September 2009 and March 2010, Officer A participated in seven uses of force, averaging approximately one use of force per month. Even though the seventh use of force incident occurred in March 2010, the intervention did not occur until almost seven months later in October 2010. In the interim period, Officer A participated in another six uses of force, almost enough to trigger the EIS again. Third, the EIS review by the supervisor is superficial at best, as evidenced in the EIS reviews of Officer B. See Appendix C. In a 14-month period, Officer B participated in 20 uses of force and was the subject of four OPA complaints regarding the use of force. In his four EIS interventions, the reviewing supervisor noted that Officer B applied many best practices and acted with good reason, and that his uses of force were in accordance with Department policy. Meanwhile, our investigation revealed several noteworthy patterns that Officer B's supervisor should have identified. In Officer B's 20 uses of force, we observed that he used a flashlight and/or baton in 60% of those incidents, some of which we found to constitute excessive force. In eight of those incidents, Officer B used force with the same partner officer. A supervisor should have discussed these patterns with Officer B as part of the EIS review and come up with appropriate resolutions or training opportunities specifically targeted to these particular issues. In addition, it is worth noting that, in some instances, the supervising Captain's recommendations were never completed. Finally, several other deficiencies with EIS exist. SPD fails to track officers over time to see if interventions have successfully curbed the behavior that initially triggered the EIS; SPD allows the sergeant who initially signed off on the use of force subject to the EIS to review the EIS; officer participation in the EIS process is voluntary; and officer performance evaluations frequently fail to reference EIS interventions. 4. Inadequate Verbal De-Escalation Training. In addition to the shortcomings in training relating to use of force weapons (such as batons), use of force reporting, and sergeant training discussed above, we also find deficiencies in training relating to verbal de-escalation techniques. The incidents discussed in our findings, in Section IV.A, illustrate that force could have been avoided in many cases if the officers had better strategies for using verbal commands before resorting to the use of force. SPD's most recent Special Report on the Use of Force (2006-2009) emphasized the "command and control" culture at SPD. It states: "To put it bluntly, officers are not trained to fight fair. Instead officers are trained to take appropriate action to bring a situation under control as quickly as possible in order to minimize the risk of harm to everyone. There is no matching of -24- action/reaction, and no requirement to try varying levels of force." In other words, officers are trained how to win conflict, but not how to avoid it. In response, OPA auditors have repeatedly recognized, since as early as 2004, the necessity of implementing training that assists officers in learning how to de-escalate situations to avoid "the escalation of minor street confrontations into situations involving forceful arrests" and to make "better early tactical decisions" to avoid forceful arrests. Many community members we spoke to also emphasized that they believe SPD officers should be doing much more to de-escalate confrontations. We understand that SPD has committed to develop the LEED (Listen and Explain with Equity and Dignity) training, which will focus on respect, listening skills, and the use of verbal tactics as an alternative to the use of force. This is a positive step forward, and particularly important in terms of developing SPD officers' skills on communicating with the diverse communities and populations they encounter on a daily basis. We also encourage SPD to expand its training of officers, in conjunction with its CIT unit, on how to handle encounters with people who have mental illness or are under the influence of alcohol or drugs. We urge SPD and the Training Unit to maintain its sense of urgency with respect to its priorities and implementation plans, and to make improvements swiftly. 5. Garrity Protections. A significant systemic contributor to our finding of a pattern or practice of unconstitutional uses of force by SPD officers are SPD's policies relating to Garrity protections. They compromise use of force reporting, investigations, and supervision, and make it difficult to conduct effective OPA investigations. Given the serious nature of our concerns, we provided specific guidance regarding SPD's Garrity policies and practices in a separate letter, dated November 23, 2011. See Appendix E. We understand that SPD has charged its newly-instituted Professional Standards Section with reviewing our recommendations, and look forward to working closely with them on reform. 6. OPA Has Not Provided the Necessary Accountability. As with EIS, we find that OPA does not provide the intended backstop for the failures of the direct supervisory review process. We are particularly concerned with: (1) the quality of the investigations performed by the precincts; (2) the consistent overuse and misuse of the finding, "Supervisory Intervention," and (3) OPA's current classification and findings systems, which we find damage OPA's transparency and accessibility to the public. Attached as Appendix D is a complete discussion of our findings regarding OPA. The Appendix reviews each stage of the OPA complaint process. -25- C. Discriminatory Policing Although we do not make a finding on the issue, our investigation found troubling practices that could have a disproportionate impact on minority communities. At a minimum, there is a strong perception among segments of Seattle's diverse communities that SPD officers engage in discriminatory policing practices against racial and ethnic minorities, in violation of the Fourteenth Amendment, the Safe Streets Act, and Title VI. The Equal Protection Clause of the Fourteenth Amendment prohibits selective or discriminatory enforcement of the law. Whren v. United States, 517 U.S. 806, 813 (1996). The Equal Protection Clause is violated when a government official administers a facially neutral law in a way that is motivated by a discriminatory purpose and results in a discriminatory effect. See Washington v. Davis, 426 U.S. 229-40 (1976). Evidence of discriminatory effect may include evidence of similarly situated individuals who were not subjected to the enforcement actions, statistical evidence, or both. United States v. Armstrong, 517 U.S. 456, 467 (1996). Discrimination - whether overt or implicit - results in an equal protection violation. GonzalezRivera v. I.N.S., 22 F.3d 1441, 1450 (9th Cir. 1994) (recognizing the unconscious effects of impermissible biases on law enforcement decision-making). In addition to prohibiting the intentional discrimination described above, Title VI's implementing regulations and the Safe Streets Act also proscribe recipients of federal funding, such as SPD, from engaging in law enforcement activities that exert a discriminatory effect on the basis of race, color, or national origin. 42 U.S.C. ? 2000d; 42 U.S.C. ? 3789d(c)(1); see also Alexander v. Sandoval, 532 U.S. 275, 281-82 (2001). Over the years, SPD has grappled with complaints regarding its treatment of racial and ethnic minorities. The OPA Auditor's 2009 Report about SPD's Relationship with Diverse Communities found that 43% of residents and 56% of Blacks believe that racial profiling by the police is a problem in Seattle. The report also found that a third of pedestrians felt that the police did not treat them professionally, and 40% felt the police did not have a valid reason to stop them. This is an area where community perceptions can have significantly detrimental consequences on a police department's ability to perform its mission, and it is an area where SPD faces real challenges. SPD has implemented measures in response to allegations of discriminatory policing. For example, the Department began collecting traffic stop data in response to allegations of racial profiling in traffic stops; OPA has conducted studies, reviews, and audits to identify problems and offer recommendations and solutions; and SPD has expanded its community outreach programs to include cultural competency seminars and meetings with community members and youth. We reviewed these steps and took them into account during our investigation. Despite SPD's reform efforts, our investigation showed SPD has more work to do, and there are some indicators of potentially biased policing that SPD needs to address. A critical area of concern is SPD's policies and practices regarding pedestrian stops. SPD's policy does not clearly distinguish a casual, social contact from an investigatory stop. Many in the community perceive that pedestrian stops are over-used and target minorities. This was evident -26- both in the reports we heard in community meetings and in our review of various SPD documents and reports. This perception has been hardened by recent publicized videos of force being used against people of color, incidents of overt discrimination and, to a limited degree, the collateral consequences of the pattern or practice of excessive force discussed above. Our conclusions were drawn based on several factors, including the lack of complete data, structural shortcomings, accounts shared by community members, and implicit bias. Our use of force analysis and two categories of data we did review raise a red flag about the potential of biased policing that SPD should fully investigate. Our investigation identified three systemic deficiencies that contribute to the problem that SPD should correct: (a) the failure to adequately collect data necessary to assess allegations of discriminatory policing; (b) the failure to develop adequate policies and procedures; and (c) the failure to develop appropriate training curricula that properly addresses the potential for implicit bias. Solid social science establishes that certain police officers will unconsciously engage in prohibited biased policing without proper accountability systems in place to prevent this. Although SPD has taken some steps to address concerns of bias, systemic failures in policies, training, and supervision may explain why indicators of discriminatory policing still exist. 1. Unlawful Pedestrian Encounters. Our investigation raises serious concerns about SPD's practices related to pedestrian stops. SPD appropriately encourages its officers to be proactive and engage with the community and people on the streets in a number of ways. Good policing requires regular and sustained interactions between police and the community. However, SPD must ensure that its officers understand the constitutional restrictions that guide pedestrian encounters. Officers must have a sufficient factual basis to detain or investigate someone, or a person is free to walk away from police and free to disregard a police request to come or stay. In these circumstances, a person's decision to "walk away" does not by itself create cause to detain. SPD's policy and practices blur the line between a social contact or casual encounter, and a temporary investigatory detention pursuant to Terry v. Ohio, 392 U.S. 1, 30 (1968). Consequently, officers lack adequate guidance as to when someone must heed an "order" to stop or stay. This means that Seattle residents may be detained for an investigatory or Terry stop by officers who are entitled to no more than a casual social contact with that individual. SPD needs to implement better policies, training, and supervision to ensure officers constitutionally detain someone in a pedestrian encounter. Federal and state laws do not allow officers to detain an individual for anything less than reasonable suspicion. However, it is well established that the "Fourth Amendment is not implicated when law enforcement officers merely approach an individual in public and ask him if he is willing to answer questions." United States v. Washington, 490 F.3d 765, 770 (9th Cir. 2007), citing Muehler v. Mena, 544 U.S. 93, 101 (2005). In State v. Harrington, the Washington Supreme Court has also stated that the term "'social contact' does not suggest an investigative component." 222 P.3d 92, 96 (2009). -27- SPD's policy manual states that a social contact is "voluntary or 'consensual,'" which means that the individual "is under no obligation to answer any questions and is free to leave at any point." DP&P 6.220.II.A.2. Even though an officer may engage in the encounter for the purpose of asking questions and gathering information, an officer does not need to have reasonable suspicion or probable cause to initiate the contact. DP&P 6.220II.A-A.1. As the OPA Auditor noted in a 2008 annual report, during a social contact, an "individual is free to walk away, refuse identification or even take off running." We are concerned that these inappropriate practices may particularly affect racial and ethnic minorities, which would constitute a violation of the Fourteenth Amendment's Equal Protection Clause. That is certainly the perception of a significant segment of communities of color. During our community interviews, individuals reported several instances in which SPD officers reportedly interacted with individuals in ways that, if accurate, went well beyond social contact stops. In these reported instances, officers stopped individuals without reasonable suspicion or cause, detained them, and indicated to them that they were not free to leave. Many of the incidents reported involved officers stopping youths of color. These reports often involve people who do not understand they are free to walk away, and are afraid to disobey officers' unlawful commands to stay and answer questions. We understand that the reported events represent only one side of the story; however, the stories are numerous enough that they clearly contribute to the negative perception of SPD within segments of the community. Since social encounters should be very brief encounters that are not necessarily recorded, it is of the utmost importance that SPD officers properly conduct these stops so as not to violate law or generally accepted police practices. The limited data that SPD was able to provide us with raises concerns that, in some precincts, there are disproportionate stops of non-whites, even when no offense or other police incident has occurred. 2. Incidents of Overt Discrimination Contribute to the Problem. Perceptions of biased policing are hardened by recent videos of police employing force against minorities, and are also caused by incidents in which SPD officers have used racially insensitive or racially inflammatory language toward, and against, racial and ethnic minorities. During our investigation, a number of individuals reported incidents in which racial epithets were used or minorities were singled out for harsh treatment. We also reviewed the video of the notorious incidents involving an officer's threat to "beat the f'ing Mexican piss" out of a suspect. It is troubling that the use of this racial epithet failed to provoke any of the surrounding officers to react, suggesting a department culture that tolerates this kind of abuse. Of greatest concern, neither of the two supervisors present admonished the officer at the scene. Nor did anyone report the incident to OPA until a third-party video of the incident was posted publicly. The number of people present, the failure to correct the officer, and the failure to immediately report the conduct all could be seen as a reflection of a hardened culture of accepting racially charged language. -28- In a similar incident reported to us by the community, an SPD officer, who was in the process of detaining a Native American suspect, referred to him as a "f'ing Indian." In another incident, a Black male bystander at an accident asked the SPD officer on site if he could help because he worked at a hospital. The officer replied, "[w]hat are you the janitor?" A white bystander watching the encounter reported the incident as a biased policing complaint to OPA because she believed the officer's statement contained some racially insensitive undertones. Despite agreeing to the complainant's version of the facts, the officer was not formally disciplined for lack of professionalism, and only received counseling from his commander. As acknowledged by SPD, incidents involving the use of overt discriminatory language cause great concern and undermine public trust. 3. Additional Indicators of Potential Biased Policing. Again, perceptions persist that SPD exhibits bias both in its street encounters, and in its use of force. As discussed above, similar issues have been addressed over the years by various oversight entities, including a review of "contempt of cop" cases, and a report of SPD Relations with Diverse Communities. However, our findings and analysis relating to the use of excessive force also may implicate the issue of biased policing. Of the cases our investigation determined involved excessive or unreasonable uses of force, more than half involved non-white subjects. While this single statistic does not establish the existence of bias it should be a source for concern given how disproportionate it is to the size of Seattle's minority population. Moreover, we note that previous reports have touched upon (but did not resolve) the issue of disparate impacts of SPD policing on communities of color. For example, OPA's review of use of force complaints from 2003 to 2005 revealed that "a high proportion of complaints about force [were] made by citizens of color." And in 2008, following a report in a local newspaper examining the racial disparities in SPD's contempt of cop charges, the OPA Auditor wrote a report examining patterns related to obstruction arrests. The OPA report did not resolve the issue of biased policing. However, the report did show that among the 76 "obstruction only" charges reviewed, 51% involved Black individuals. While the Auditor noted that this data was consistent with rates of arrests for violent offenses, this disparity, and those discussed above, warrant further investigation. Finally, while SPD does not collect or maintain adequate data relating to issues of biased policing, it does collect some data relating both to its street stops and through its Computer Aided Dispatch ("CAD") system. This data too is incomplete and not sufficient to support any conclusion one way or the other with regard to biased policing. However, as discussed below, the data does raise red flags regarding whether minorities are subjected to a disproportionate amount of policing. Taken together, sufficient indicators exist to prompt SPD to conduct a more rigorous analysis of these issues. -29- D. Deficiencies Contributing to Potential Discriminatory Police Practices 1. Inadequate Biased Policing and Pedestrian Stop Policies. SPD has two primary policies that contribute to concerns regarding discriminatory policing: (1) Unbiased Policing, SPD Policy 5.140, and (2) Social Contacts, Terry Stops & Arrests, SPD Policy 6.220. a. The Unbiased Policing Policy The current policy states that "race or ethnicity shall not be motivating factors in making law enforcement decisions...." DP&P 5.140.I.C.2. The policy lacks precise language regarding the definition of biased policing, and fails to clearly identify prohibited acts covered by the policy that officers might not think of as biased policing, including issues related to gender, religion, and sexual orientation. We recommend that the Department revise its policy to provide more guidance to SPD officers about how to avoid biased policing practices. Supervisors are essential to ensure their officers do not engage in discriminatory policing. We, therefore, also urge SPD to revise the policy to amplify the importance of supervisory responsibility and to formalize the mechanism by which supervisors confirm that their officers are both familiar with the policy and operating in compliance with it. Currently, the policy states that "supervisors shall ensure that all personnel in their command are familiar with the content of this policy and are operating in compliance with it." DP&P 5.140.I.E. We believe that the policy should also contain clearly delineated procedures for a supervisor to follow when investigating a potential biased policing incident, including on-scene investigation and incident analysis. This dovetails with our recommendations regarding additional training on biased policing for OPA detective sergeants. b. Social Contacts, Terry Stops & Arrests Additionally, as we discussed in Section IV.C.1, SPD's policy on Social Contacts, Terry Stops & Arrests blurs the lines between a social contact and a Terry stop, and thus encourages officers to conduct stops that do not comport with federal and Washington law or nationally accepted police practices. We recommend amending the policy to provide clearer guidance regarding what constitutes a social contact. The policy currently defines a social contact to be "for the purpose of asking questions and gathering information." A social contact should be a purely voluntary encounter. The language regarding "asking questions and gathering information" blurs the distinction between a social contact and an investigative Terry stop, when officers do "stop a person for questioning." DP&P 6.220.II.A and 6.220.III.A.1. -30- Consistent training and supervisory review on these issues is also critical. Our investigation revealed that the language of the policy, in combination with the training provided at the Washington state BLEA (discussed below), contributes to officers' improper stops of pedestrians. A policy that clarifies when officers can lawfully detain individuals should have the effect of decreasing the number of incidents that might lead to uses of force, especially in situations when officers may incorrectly believe that they have the right to detain individuals, and may resort to the use of force when those individuals assert their right to leave. Lastly, we also recommend that SPD review its policy to be consistent with State constitutional requirements regarding pretextual stops. See, e.g., State v. Ladson, 138 Wash. 2d 343, 979 P.2d 833 (1999) ("Article 1 sec. 7 [of the Washington State] Constitution forbids the use of a pretext as a justification for warrantless searches and seizures because our constitution requires we look beyond the formal justification for the stop to the actual one. In the case of pretext, the actual reason for the stop is inherently unreasonable; otherwise the use of pretext would be unnecessary.") Terminology, policies, and training should be in accordance with federal and state constitutional law. 2. Inadequate Data Collection. SPD does not collect adequate data to self-assess whether biased policing is occurring. The collection and analysis of data, particularly in pedestrian encounters, will assist the Department in identifying trends, patterns, and practices that are or could be perceived as biased policing. This in turn will allow the Department to address the community's concerns in a proactive manner. Accordingly, to address concerns regarding unlawful pedestrian encounters, especially as they affect communities of color, we recommend that SPD concentrate on data collection efforts to better track officer activity related to pedestrian stops. We note that the OPA Auditor has also previously recommended that SPD collect data on pedestrian stops as a means of addressing biased policing concerns. Data collection is just one piece necessary to address discriminatory policing practices. Other necessary remedies include heightened supervision, revised policies and procedures, and training that address the root causes of discriminatory practices, including implicit bias. a. Demographic Data Collection Standing alone, disparities in stop and arrest data are insufficient to show discriminatory policing. Still, it can be one indicator as to whether a Department needs to look further to determine if the data can be explained or if it is a reflection of discriminatory policing. SPD does not track demographic data by City precinct or district. This failure prevents SPD from tracking the information it collects through its various databases. Without the ability to identify the racial and ethnic demographics of the community they serve, SPD cannot compare how stop and arrest data relate to the demographics. Without this, SPD lacks an important tool to identify potential biased policing trends. -31- b. Street Check Database The Street Check database is the primary SPD database that collects information related to the types of pedestrian encounters that are most susceptible to concerns regarding biased policing. Officers enter "street checks" to create a record for encounters with individuals even if no offense or other police incident has occurred. The Street Check database "provide[s] other officers and detectives with information concerning suspicious activity," observations an officer makes that may lead to suspicious activity, and situations when a pedestrian is arrested but ultimately released in the field. DP&P 6.220 and 6.010. We believe that the Street Check database is limited in its effectiveness for the following reasons. First, SPD does not have an official policy or standard operating procedure that clearly defines the purpose of this database, or describes when officers should use the database. We believe that developing clear guidance will lead to more accurate and complete data collection. Second, as a result of the lack of clear guidance, officers do not consistently enter encounters into the Street Check database, and sometimes enter these encounters incorrectly into the CAD database. Confusion still exists regarding when officers should use the database, which means that the data only provide a partial picture of SPD officer activity. We understand that SPD is already in the process of updating its training manual and database to ensure uniform reporting to proper databases. Although this is a good start, we recommend that SPD also conduct periodic audits to make sure that officers are accurately entering and reporting their activity to the proper database. Third, officers are not required to input demographic data such as race, ethnicity, or gender into the Street Check database. As a result, SPD's ability to analyze the impact of its activities on various subpopulations is limited. We recommend that SPD begin collecting demographic data for all Street Check encounters. In sum, improved data collection would assist SPD in identifying outlier officers and trends relating to discriminatory policing, and provide the community with helpful information. Data collection allows a police department to respond to community concerns of discriminatory policing. On the flip side, the failure to collect this data does not absolve a department of its duty to ensure racially neutral treatment and policing that meets constitutional standards. 3. Inadequate Supervision. Our investigation indicates that SPD fails to provide adequate supervision to assess biased policing concerns by (1) failing to conduct data analysis regarding its officers' activity and (2) failing to conduct thorough investigations of biased policing allegations. a. Data Analysis Data analysis will provide SPD command staff and supervisors with a potentially powerful supervisory tool to monitor officer activity and pedestrian encounters, some of which may relate to biased policing concerns. This section summarizes the results of our data analysis -32- that identifies outlier precincts, officers, and activity. Although we do have concerns that SPD's data are not complete, our observations are based on a thorough and complete examination of the available data. In combination with implementing controls to ensure that officers consistently report data related to pedestrian encounters, we also urge SPD to conduct similar data analysis on a regular basis as a means of preventing or identifying possible biased policing trends. iii. Street Check Database When an officer makes either a social or Terry stop, the officer is directed to enter information regarding the stop and individual into a Street Check database. Obviously, these data are limited in nature, not complete, and not sufficient to support any conclusion. In particular, the data do not show the reasons or context of such stops. Accordingly, we do not suggest that this data standing alone can prove or disprove the existence of biased policing. However, the data can provide useful information on trends, including where and when stops are happening, who is being stopped, and which officers are using stops most frequently. SPD command staff and supervisors should more carefully collect and examine the data to identify any issues or trend that may relate to biased policing. For example, we made the following observations based on our analysis of the Street Check database. First, it suggests that SPD officers may conduct a disproportionate number of street checks on people of color in certain precincts when compared to population percentages. For example, in the East precinct, non-whites make up only 33% of the population, however, they made up 64% of the street checks. Similarly, in the West precinct, non-whites only make up 26% of the population, but made up 47% of the street checks. In the Southwest precinct, nonwhites make up 32% of the population, but made up 49% of the street checks. Again, the data are limited in nature, and other factors may explain or contribute to these numbers. Still, the disproportionate ratios should be reviewed. Second, we recommend that SPD focus particularly on street checks that result from officer discretion. When street checks are conducted in response to suspicion (as opposed to responses to complaints, event/incidents, and other), such encounters are considered discretionary officer-initiated activity. According to our analysis, the three districts with the greatest proportion of officer-initiated suspicion street checks are all located in the West precinct. This is also one of the areas of Seattle with the greatest racial and ethnic disparities in the overall number of street checks, as compared to a representative population. For example, Officer B, whose use of force activity was discussed in Section IV.B.3.d, also had high activity levels that could have biased policing implications. Not only was Officer B in the top 25% of officers with the most CAD activity in his respective district, Officer B's CAD activity indicated that 23% of his stops, the highest percentage of all of his stops, were stops of suspicious people, which tends to be a highly discretionary activity. This data does not alone establish proper or improper practices, but it should have prompted the Department to analyze his activity more carefully. -33- Third, we observed some disparities in the street checks data regarding stops of youth. In two different districts in the South precinct, there was a lower mean age of individuals stopped as compared to other districts. Since the South precinct is comprised of 71% people of color, this is an area where SPD should review to determine if this reflects an issue related to pedestrian stops of youth of color or whether it is anomalous. iv. CAD Database ("Computer-Aided Dispatch") The CAD database captures officer activity that originates from emergency 911 calls made by Seattle community members to police dispatchers, and from officers making calls to dispatchers to announce the activities in which they are engaged. As with the Street Check database, we recommend that SPD conduct an in-depth analysis of officer activity, both by precinct and district, in order to identify and address any trends in officer activity deserving of further review. Although the scope of the CAD data collected by SPD appears reasonable, the Department must do a better job of using these data as a supervisory tool for analyzing officer activity. For example, through our examination of 2010 CAD data, we discovered that a very small proportion of officers take the majority of calls for service. In fact, just 5% of responding officers account for 20% of all CAD recorded activity. This may reflect appropriate proactive policing. However, if any of these officers were to engage in inappropriate or overaggressive policing practices, they would have a disproportionately negative effect on the community in which they work. Moreover, SPD should ensure that the workforce demands are not shouldered by too few officers. A stressed officer with an excessive workload itself can lead to problems. In short, CAD data can be a tool for identifying outlier officers, whether overactive or underactive, and some policing trends in precincts and districts where they work. This could provide supervisors with the ability to focus on a smaller subset of officers whose activity may warrant additional attention. Additionally, analyzing the data would enable supervisors to better incorporate real life data into Department training. b. Investigations We are concerned about the thoroughness of SPD's investigations of biased policing allegations. OPA at times improperly classifies cases involving allegations of discriminatory policing. Some of the cases OPA classifies as involving something other than biased policing should in fact be classified as biased policing cases. Further, too many of the cases that have been designated as biased policing are classified as less serious. Because of these classification failures, OPA often fails to investigate allegations involving bias, make formal findings on those cases, or (where a case is sustained) issue discipline. See Appendix D. Instead, complaints may get referred to the individual's supervisor for investigation. We believe OPA should investigate all such allegations itself. In the event OPA sends any such allegations to the precincts, we urge SPD to develop more stringent supervisory responsibilities for supervisors investigating biased policing complaints. -34- We additionally recommend the following improvements for discriminatory policing investigations by OPA: (1) improve efforts to identify the subject officer; (2) interview all witnesses; and (3) clearly establish that officer testimony will not automatically be granted greater credibility. Further discussion of SPD investigations can be found in Appendix D. 4. Insufficient Training a. Unbiased Policing Training SPD uses two principal unbiased policing training modules: Perspectives in Profiling, and Race, the Power of Illusion. The Unbiased Policing policy also requires that the "Department's ongoing training curriculum [] include regular biased policing updates." DP&P 5.140.I.H. The modules narrowly address biased policing, the effects of institutional discrimination, and constitutional requirements. SPD's current training fails to adequately address some of the underlying causes of racially biased policing, namely, that biased policing is not primarily about the ill-intentioned officer but rather the officer who engages in discriminatory practices subconsciously. A wellmeaning officer can violate the Equal Protection Clause of the United States Constitution by engaging in racially biased policing based on implicit biases that impact that officer's behavior or perceptions. Gonzalez-Rivera, 22 F.3d at 1450. Understanding this phenomenon is the first step toward safe and effective policing. Several well-respected organizations have analyzed how to avoid discriminatory policing involving implicit as well as explicit discriminatory biases, including DOJ's Community Oriented Policing Services (COPS) Office, the Community Policing Consortium, the National Organization of Black Law Enforcement (NOBLE), the International Association of Chiefs of Police (IACP), the Police Executive Research Forum (PERF), and the Commission on Accreditation for Law Enforcement Agencies (CALEA). SPD has a relationship with CALEA, and has undergone its routine certification audit. It could also benefit from the solid social science that has come to the fore in recent years relating to implicit bias in policing and the training practices that have been developed to address the problem. We encourage SPD to work with Seattle's diverse communities to address cultural competency skill building. We also urge SPD to emphasize the importance of unbiased police training at all levels of SPD command staff and supervisors. Lastly, we also encourage SPD to conduct informal training and roll-calls as part of SPD's "regular biased policing updates." Id. b. Social Contact and Terry Stop Training Based on our review of State Academy training materials related to pedestrian stops, it appears the officers are being trained incorrectly. We believe that the training materials' incorrect definition of a "social contact" encourages officers to respond only to suspicions lacking in specific and articulable facts, and has the predictable effect of inviting officers to act on their conscious and subconscious biases and prejudices. Moreover, it has the unfortunate consequence of blurring the line between Terry stops and social contacts. Any attempt to investigate criminal activity should be based on real indicators as opposed to vague and nonspecific hunches. We urge SPD to devise supplemental training regarding social contacts and -35- Terry stops, and to make this training a part of the mandatory annual Street Skills training, as the OPA Director advised long ago. c. Communication Skills Training SPD recognizes the need to revise its training related to communication skills. OPA also has recognized that "disrespectful officer conduct cuts against the public trust necessary for effective policing." In an attempt to address this issue, SPD is partnering with the King County Sheriff's Office and the Criminal Justice Training Commission to develop training for officers focused on promoting dignified and respectful treatment of citizens. This program stresses four basic principles - Listen and Explain with Equity and Dignity - and encourages professional interactions, and emphasizes use of verbal tactics as an alternative to use of force, where practical and without compromising officer or public safety. The OPA Director is coordinating with the Training Unit in the development of this curriculum and its completion is anticipated in 2012. In the interim, we understand that supervisors and command staff have received in-service training on these core principles. 5. Community Engagement. Community trust is critical to effective policing. A series of high-profile incidents - both related to use of force and discriminatory policing - has eroded the relationship between SPD and the community it serves. Many community members we met with appreciated the steps taken by Chief Diaz and his command staff to address their concerns, but they felt that more sustainable programs were needed. Some residents expressed concern that while the Department's outreach following a highly publicized incident may be strong and targeted, the outreach itself is limited and short-lived. This is particularly noteworthy when SPD does not operate in as transparent a manner as possible. Any notable decrease in SPD's community outreach after the proverbial "storm has passed" does not effectively resolve the community's concerns and may exacerbate the tension. SPD's community outreach initiatives should be proactive and sustained rather than a reaction to a particular crisis. As the Department conducts outreach to inform the community of its efforts and initiatives, it should remember that an equally important purpose of these outreach efforts is to listen carefully to the community's input, and to establish a constructive two-way dialogue. V. RECOMMENDED REMEDIAL MEASURES During the course of our investigation, we provided technical assistance and advice to SPD on the above findings. As we have noted throughout this letter, we are encouraged by the steps that SPD has taken in response to that technical assistance and commend it for beginning corrective actions even before our findings were concluded. These actions by SPD include: ? SPD has recently created a Force Review Committee to give the commander greater assistance in reviewing use of force reports, and identifying problematic use of force patterns and training deficiencies. We look forward to further discussion and refinement of the Committee's role, including the codification of its procedures. -36- ? SPD has recently developed a Professional Standards Section, which will be responsible for conducting research on standards and best practices, internal audits and inspections, and managing strategic initiatives. This Section will also specifically address the recommendations laid out in the DOJ's November 23, 2011, technical assistance letter regarding Garrity protections. ? SPD has directed additional resources to CIT. ? SPD has recently implemented a 48-hour training on the use of force, and SPD policy and expectations. ? SPD has begun to clarify expectations and improve training for sergeants, including rolling out a new sergeant's training that all SPD sergeants and the top 20 officers on the Sergeants Promotion list will attend in 2011. ? SPD has committed to develop the LEED (Listen and Explain with Equity and Dignity) training, which will focus on respect, listening skills, and the use of verbal tactics as an alternative to the use of force. ? SPD has instituted a zero-tolerance policy for racially inflammatory language. ? A Working Group comprised of the OPA Director, OPA Auditor, and representatives of OPA-RB has convened to simplify the classification and findings systems, to provide a clearer delineation of the new classification and findings categories, and to improve turnaround time to its target time of 120 days. We eagerly anticipate the results of the Working Group. ? OPA recently has suspended referring complaints for investigation to the chain of command until sergeants receive additional training on how to conduct such investigations. ? SPD has made renewed efforts in community outreach. In addition to formalizing the above improvements, some additional remedies are needed to address the deficiencies identified in this letter. Below we review some of the important remedial measures SPD should take in order to avoid engaging in a pattern or practice of unconstitutional conduct by its officers. While we understand that many of the issues identified in this letter may implicate some aspects of the SPD Collective Bargaining Agreement, Seattle's policing community has an established history of coming together to address shortcomings. We have every reason to believe it will do so now. The measures are based on professional and generally accepted practices in the field of local law enforcement. They are not meant to be exhaustive, but they include this letter's key recommendations. -37- A. 1. USE OF FORCE SPD should revise its Use of Force policy to clarify that officers must report any use of force above un-resisted handcuffing, including the active pointing of firearms. Even in cases of un-resisted handcuffing, force should be reported if the subject complains of injury or excessive force. SPD should develop and implement a use of force policy that includes a specific force policy for each and every weapon available to SPD officers, including the following: OC spray, flashlights, ECW, batons, and the active pointing of a firearm. These policies should clearly define and describe when, how, and how much force is appropriate, and when such force will be considered deadly force. SPD should also consider providing intermediate weapons to more of its officers, but only following the development of policies and training. SPD should develop and implement protocols with the CIT unit on how to handle interactions with individuals with mental health issues and individuals who appear to be under the influence of drugs and or alcohol. The protocols should include when CIT should be consulted, and how situations involving impaired subjects should be addressed when CIT cannot respond. SPD should ensure that supervisors perform the following actions in response to any use of force incident: (a) ensure that a medical unit report to the scene of every use of force resulting in injury, actual or complained; (b) conduct a thorough analysis of the incident based on all obtainable physical evidence, adequately descriptive use of force reports, witness statements, and independent investigation; (c) resolve any discrepancies in use of force reports or witness accounts and explain and document all injuries; and (d) complete a summary analysis regarding the reasonableness, proportionality, and legality of the force used. If the supervisor cannot resolve any factual discrepancies, determine the source of any injury, or determine the lawfulness of a use of force, the supervisor should refer the matter immediately and directly to his/her supervisor and to OPA. Every level of supervision thereafter should be held accountable for the quality of the first-line supervisor's force investigation. SPD should use multidisciplinary roll-out teams, including OPA, the Training Unit, and a prosecutor to investigate incidents involving serious uses of force. Serious uses of force can be defined either by the injury sustained by a subject, or as any use of deadly force or unjustified force, regardless of the severity of the injury. Trained investigators should roll-out to officer involved shootings and other serious incidents to make the determination about whether to consult a prosecutor. Following SPD Use of Force Policy 6.240.XII.A.7, SPD should require officers to submit use of force statements before they go off-duty, including following officer-involved shootings. Exceptions may be made if an officer is physically or mentally incapacitated. 2. 3. 4. 5. 6. -38- 7. Supervisors should review and take appropriate disciplinary or non-disciplinary corrective action, when warranted, in situations in which he or she becomes aware of potential misconduct or criminal behavior by an officer. Supervisors should conduct focused reviews of any officers involved in a disproportionately high number of use of force incidents for any training or discipline issues. When the supervisor becomes aware of potential misconduct or criminal behavior by an officer, the supervisor should take appropriate disciplinary or non-disciplinary corrective action SPD should track the prosecutorial disposition of all arrests as one tool to identify possible trends in abuse of law enforcement discretion. EIS reviews should be conducted by a higher level supervisor, such as a captain. The supervisor should conduct a timely EIS review that identifies patterns in officer behavior and specific training deficiencies. Officers in the EIS program should have a monitoring supervisor designated to ensure that all training and other Department-initiated remedial steps are taken and documented, and that the officer's performance is closely monitored and evaluated. At the end of the monitoring period, the precinct captain should either certify that the officer has successfully completed the EIS program or initiate another plan for improvement. EIS should be expanded to track supervisor and precinct activity. Participation in EIS must be mandatory. The following changes to EIS thresholds should be made: (a) adjust EIS thresholds to identify at least 3-5% of the line officer population; (b) revise the aggregate indicator to include uses of force (DP&P 3.070.II.I); (c) create single-event thresholds for events so critical that they require immediate department intervention; (d) implement rolling thresholds, such that after an officer who has received an intervention for use of force should not, for example, be permitted to engage in seven additional uses of force before again triggering EIS; and (e) include a threshold specifically related to biased policing complaints. B. OFFICE OF PROFESSIONAL ACCOUNTABLILTY (See Appendix D) 8. 9. 10. 11. 12. 1. An independent expert should conduct a bi-yearly randomized, stratified audit of OPA intake. Intake sergeants and lieutenants should receive formal orientation, training, and written protocols that conform to best national practices regarding (a) intake, (b) complaint classifications, (c) when an investigation is necessary, and (d) how to avoid conflicts of interest in investigations. 2. -39- 3. OPA should develop protocols with its officers, precincts, and any other outside agency that receive complaints, so that all police misconduct complaints are timely received by OPA for its review and investigation. OPA or the City should conduct anonymous Integrity Checks. That is, OPA (or third party auditor) should retain a qualified individual to file dummy complaints with OPA (and with the precinct or City) and report on how the complaint is directed and handled. OPA should provide final approval of any action or inaction taken in Preliminary Investigative Review and Supervisory Referral matters, and the outcome should be reported back to the precinct. SPD should expand upon existing investigation techniques and continue to train all of its investigators on: what factors to consider when evaluating complainant and witness credibility; examination and interrogation of accused officers and other witnesses; identifying misconduct, even if it is not specifically named in the complaint; and using the preponderance of the evidence standard as the appropriate burden of proof. An SPD officer's statement should not be given any preference over a civilian's statement, nor should SPD completely disregard a witness's statement merely because the witness has some connection to the complainant. OPA investigators should make efforts to resolve material inconsistencies between witness statements. Supervisory Interventions should only be allowed once SPD has issued a definitive and true finding. OPA should simplify its classifications, and have two findings, for purposes of reporting to the complainant ("Sustained" or "Not Sustained"), for each allegation against each officer. OPA may explain to the Complainant, named officer, and public what remedial steps will be taken in Sustained complaints (i.e., "Formal Discipline" or "Training") or why a complaint was Not Sustained, whether because the conduct was "Exonerated/Lawful," "Unfounded/Officer Not Involved," "Inconclusive," or "Administrative/Procedural" reasons. SPD, in consultation with OPA and appropriate experts, should consider issuing a discipline matrix that guides the Director and the COP when making disciplinary decisions. Each precinct should have an Integrity Control Officer ("ICO"). The ICO, who would report directly to the OPA Director, would review the quality of Line Investigation investigations and the disposition of PIRs and SRs (or its equivalents). The ICO would also review the completeness of use of force files. SPD and OPA should ensure that all in car video recordings are made available to supervisors for review. 4. 5. 6. 7. 8. 9. 10. 11. -40- C. 1. DISCRIMINATORY POLICING Each precinct should regularly collect the demographic data for their area, including districts. The Training Unit and Special Operations Bureau should update SPD databases and training materials to ensure that all officers report pedestrian stop-related activity to the proper database. SPD should conduct regular audits of SPD databases, including the CAD and Street Check Databases, to ensure compliance with the reporting requirements of each database. SPD command staff, precinct commanders, and supervisors should conduct regular analysis of officer activity data produced by the Street Check and CAD databases. SPD should revise the Social Contact, Terry Stop, & Arrest Policy to ensure that definitions of social contacts and Terry stops are in accordance with individuals' constitutional rights. SPD should revise the Unbiased Policing Policy to outline what behavior(s) are impermissible and to identify supervisory responsibility in investigations of biased policing incidents. SPD should increase training and understanding regarding (a) investigatory encounters and temporary investigatory detentions pursuant to Terry v. Ohio, 392 U.S. 1, 30 (1968), (b) pretextual encounters, (c) discriminatory policing investigations, and (d) communication skills and cultural competency. Supervisors should, in conjunction with the SPD Training Unit and OPA, develop a roll call training curriculum that reviews issues related to discriminatory policing, including activity identified within the Department. VI. CONCLUSION 2. 3. 4. 5. 6. 7. 8. We have attempted to conduct this investigation as transparently as possible. On more than one occasion, we discussed with SPD some of the problematic trends our investigation has shown and the remedial steps we were contemplating. We again commend SPD for the amount of time it has devoted to meeting with us, and the seriousness with which it has taken our requests and technical assistance. We now look forward to engaging the City in a more in-depth dialogue about how to remedy the deficiencies we have identified in the context of structured formal discussions. We are confident we can maintain the momentum of the investigation and engage in an aggressive timeline to resolve the issues we have raised. Ultimately, a resolution will be formalized by written, court-enforceable agreement that sets forth remedial measures to be taken within a fixed period of time. Such a disciplined remedial structure would provide all interested parties with the greatest assurance that violations of constitutional rights are corrected and will not reoccur. -41- Thank you again for your ongoing cooperation in this matter. Because of that cooperation, and the productive working relationship we have created, we are optimistic that we can resolve these issues. We will contact you soon to discuss the issues referenced in this letter, and to set a date and time to meet in person to discuss a remedial framework. Please note that this letter is a public document and will be posted to the Civil Rights Division's website. If you have any questions, please feel free to contact us. Sincerely, Thomas E. Perez Assistant Attorney General Civil Rights Division cc: John Diaz Chief of Police Peter Holmes Seattle City Attorney APPENDIX A: JANUARY 2009 - APRIL 2011 USES OF FORCE BY INDIVIDUAL OFFICERS Officer who did not receive Disciplinary and/or Supervisory Interventions for Use of Force 25 Officer who received Disciplinary and/or Supervisory Interventions for Use of Force 20 Number of Uses of Force 15 10 5 0 Use of Force by 659 Officers January 1, 2009 - April 4, 2011 APPENDIX B: OFFICER 2010 IN TERVEN TION OFFICER "A" Oct 2, 2008 OPA Complaint Unnecessary Force, Exonerated May 12, 2009 OPA Complaint Review: Unnecessary Force, Supervisory Intervention Aug 2, 2009 Performance Evaluation 10/01/2008 06/25/2009 "Exceeds job requirements." Dec 2, 2009 Use of Force Review: Force within guidelines Jan 18, 2010 Officer "A" - First EIS Recommendation: Attend training on Use of Force Issues (as previously recommended), period of observation by first line supervisor May 18, 2010 Use of Force Review: Force within guidelines Sep 16, 2010 Use of Force Review: Force within guidelines Jan 1, 2011 Performance Evaluation 12/29/2009 11/15/2010 "Fully competent." Oct 8, 2008 OPA Complaint Unnecessary Force, Exonerated May 29, 2010 Use of Force Review: Force within guidelines Oct 2, 2010 Officer "A" - Second EIS Recommendation: No further action required. Identified troubling pattern with use of force, provided additional training, and chain of command will continue to monitor. Feb 18, 2011 Officer "A" - Third EIS Recommendatio n: Unknown Sep 16, 2009 Use of Force Review: Force within guidelines Dec 3, 2009 Use of Force Review: Force within guidelines Jan 20, 2010 Use of Force Review: Force within guidelines Jun 6, 2010 Use of Force Review: Force within guidelines Sep 26, 2009 Use of Force Review: Force within guidelines Mar 17, 2010 Use of Force Review: Force within guidelines Jun 17, 2010 Training: Advanced Training Unit, as a result of 09-IIS-0193 and 09INTRV-0006 Jun 30, 2010 Use of Force Review: Force within guidelines Dec 5, 2009 Training re Use of Force Policy Oct 9, 2010 Use of Force Review: Force within guidelines Dec 26, 2009 Use of Force Force within guidelines LEGEND o o o o Oct 2008 First EIS Second EIS Third EIS Performance Evaluations Jul 19, 2010 Use of Force Review: Force within guidelines Jan 2009 Apr 2009 Jul 2009 Oct 2009 Jan 2010 Apr 2010 Jul 2010 Oct 2010 Jan 2011 APPENDIX C: OFFICER 2010 IN TERVEN TION Officer "B" Aug 19, 2009 Use of Force Flashlight Review: Force within guidelines Oct 26, 2009 Use of Force Flashlight with Officer "C" Review: Force within guidelines Jan 6, 2010 Use of Force Baton with Officer "C" Review: Force within guidelines Feb 28, 2010 OPA Complaint Unnecessary use of force, Exonerated Apr 15, 2010 Use of Force Flashlight Review: Force within guidelines May 29, 2010 Use of Force with Officer "C" Review: Force within guidelines Sep 22, 2010 Officer "B" - Second EIS Missing one Use of Force report. EIS file still open. No recommendation yet. Sep 24, 2010 Mar 4, 2011 Offficer "B" - Third and Fourth EIS Recommendations: nor further action necessary; SGT to continue to monitor and mentor. Jun 2, 2010 Feb 28, 2010 Aug 30, 2009 Use of Force Review: Force within guidelines Use of Force Review: Force within guidelines Apr 20, 2010 Use of Force Baton Review: Force within guidelines Jun 26, 2010 OPA Complaint Unnecessary use of force, Exonerated Jun 26, 2010 Use of Force with Officer "C" Review: Force within guidelines Use of Force Review: Force within guidelines Use of Force Review: Force within guidelines Nov 1, 2009 Jan 11, 2010 Use of Force with Officer "C" Review: Force within guidelines Use of Force Baton Review: Force within guidelines Sep 25, 2010 OPA Complaint Unnecessary use of force, Exonerated with Officer "C" (not part of EIS) Mar 18, 2010 Use of Force Flashlight Review: Force within guidelines Sep 25, 2010 Use of Force Baton with Officer "C" Review: Force within guidelines Nov 1, 2009 OPA Complaint Unnecessary use of force, Exonerated (involved Officer "C") Apr 30, 2010 Use of Force Baton with Officer "C" Review: Force within guidelines Jul 27, 2010 Use of Force Baton Review: Force within guidelines Sep 29, 2010 Performance Evaluation 10/16/2009 - 08/30/2010 "Exceeds job requirements." LEGEND Aug 6, 2010 Use of Force Baton; flashlight with Officer "C" Review: Force within guidelines Oct 12, 2010 Officer "B" - First EIS Recommendations: Review sufficient; no other action necessary. Mar 30, 2010 Use of Force Review: Force within guidelines Nov 4, 2009 Use of Force Flashlight Review: Force within guidelines ? o o o First EIS Second EIS Third EIS Performance Evaluations Aug 2009 Sep 2009 Oct 2009 Nov 2009 Dec 2009 Jan 2010 Feb 2010 Mar 2010 Apr 2010 May 2010 Jun 2010 Jul 2010 Aug 2010 Sep 2010 Oct 2010 Nov 2010 Dec 2010 Jan 2011 Feb 2011 Mar 2011 APPENDIX D: OFFICE OF PROFESSIONAL ACCOUNTABILITY APPENDIX D: OPA Has Not Provided the Necessary Accountability 1. Additional Background on Seattle's Police Accountability System As previously stated, a civilian director leads OPA and its five to seven detective Sergeants, two Lieutenants, and Captain. The Director, who is nominated by the Mayor and confirmed by the Seattle City Council, serves as a member of the COP's command staff and reports directly to the COP. The OPA Director may serve up to three (3) three-year terms. In addition to policy review, the OPA Director oversees the receipt, classification, and investigation of complaints of misconduct and related issues concerning police practices. When OPA sustains a complaint, OPA recommends disciplinary action to the COP. The COP, however, is the ultimate arbiter of discipline. Conversely, the OPA Auditor, who also is nominated by the Mayor and confirmed by the City Council, does not report to the COP and does not sit within SPD. Instead, the Auditor serves as an independent civilian advisor to the City as a whole. More specifically, the Auditor reviews OPA's classification and completed investigations of (still) open complaints, as well as SPD's policies and procedures; contemporaneously reviews all OPA investigations completed in a given week and has the discretion to order additional investigation; and issues written reports to the public at large on a routine basis. 2. A Summary of Our Findings on the Accountability System Although we believe that the structure of OPA is sound, and the investigations OPA itself conducts are thorough, we find that OPA fails to provide adequate oversight to prevent a pattern or practice of excessive force. We are particularly concerned with: (1) the quality of the investigations performed by the precincts; (2) the consistent overuse and misuse of the finding: "Supervisory Intervention"; and (3) OPA's current classification and findings systems, which we find damage OPA's transparency and accessibility to the public. However, as detailed below, we have identified problems with each stage of OPA complaint process: a. Intake Although OPA tracks every citizen contact with OPA, it does not capture, for tracking or statistical purposes, complaints made directly to individual officers, to the precincts, or to the City itself (i.e., through the Mayor's complaint line). We have discovered written complaints by citizens regarding significant issues (i.e., use of force) directed to such locations, that were not referred to OPA for appropriate action. If OPA does not capture all such complaints, they likely will not be investigated. Accordingly, OPA should develop a protocol with its officers, precincts, and any other outside agency that receives complaints, so that all police misconduct complaints are timely received by OPA for its review and possible investigation. Second, the current intake audit process requires the Director, Auditor, and perhaps an OPA Lieutenant alone to review literally hundreds of contacts with OPA per quarter. For instance, in the most recent report, covering two quarters, the Director and Auditor reviewed 681 complaints or inquiries. We believe that it asks too much of the Director and Auditor to 1 provide a thoroughly documented, detached, reliable, and systematic audit of the intake of all citizen complaints. Third, not every complaint made by telephone or in person is audio-recorded. This is not a minor concern because nearly 38% of all OPA complaints over the past two years have come via telephone. OPA should, by default, audio-record all complaints, allowing a more thorough audit of intake. Fourth, OPA's Intake Sergeants lack substantive, written guidance, and formal training, on how to assess the nature of the claims citizens present. Intake Sergeants are often the newest members of OPA and, as such, they require formal orientation, training, and written protocols on the intake process. Without further guidance, Intake Sergeants may misunderstand what citizens are actually complaining about and what evidence they should be eliciting. Finally, and most importantly, during our investigation, we were told that referrals from supervisors about misconduct by their supervisees are admittedly "rare to non-existent." We understand that no supervisor has ever been disciplined for failure to report misconduct to OPA. As discussed in more detail above, SPD needs to clarify and strengthen the obligations of supervisors (and all employees) to report misconduct or potential misconduct to OPA, and not merely undefined "serious" misconduct. b. Classification In public documents, OPA rightly asserts that "[t]ransparency and accountability to the public are the key ingredients for an effective program." We find that OPA's current classification (a "triage") system is not transparent or accessible because, by OPA's own admission, it is far too complicated. Currently, OPA classifies complaints into four categories: Investigation Section ("OPA-IS") Investigations; Line Investigations ("LIs"), i.e., those conducted by the precincts; Preliminary Investigation Reports ("PIRs"); and "Supervisory Referrals" ("SRs"). We understand that a Working Group comprised of the Director, Auditor, and representatives of OPA-RB has convened since at least the beginning of this year to simplify the classification system and provide a clearer delineation of the new categories. We eagerly anticipate the results of the Working Group and hope this finding adds to the Working Group's sense of urgency. Nonetheless, we have found several other deficiencies in the current classification system that the Working Group should seek to avoid repeating. First, in OPA's 2010 Statistics Report, the Director notes that over 60% of the complaints OPA received were sent to the officer's supervisors at the precinct level for handling as PIRs or SRs - neither of which can result in formal findings or any discipline. As a result of this classification, OPA does not investigate or play an active role in such complaints. None of this on its own is objectionable, but OPA must still monitor how all complaints are handled at the 2 precincts. Based on our review of PIRs and SRs, we find that OPA does not sufficiently monitor or review the disposition of PIRs and SRs. This failure of oversight is significant because DOJ additionally finds that the line supervisors in the precincts, for the most part, fail to adequately review, document, or remediate PIRs and SRs, whether in the form of one-time counseling, more thorough training, or more recurrent mentoring. OPA's failure to document its remediation is particularly concerning. The files, at most, contain a one paragraph memorandum to the file that simply states that the supervisor (often the sergeant) spoke to the named officer. No further documentation of followup is included to show whether the one-time counseling session was effective. OPA is the ultimate repository of all citizens' complaints. Thus, OPA must provide final approval of any action or inaction on those investigations, including on those redirected to the precincts. Second, the Working Group must provide adequate guidance to the Intake Sergeants and Lieutenants to ensure a consistent application of the new classifications. As with Intake, the Intake Sergeants and Lieutenants must have clear guidance as to (a) when it is appropriate to classify a certain type of complaint as a PIR or SR; (b) when an investigation of some kind is necessary; and (c) how to avoid any conflicts of interest in investigations, perceived or otherwise. Third, use of force and discriminatory policing complaints have been categorized as PIRs and SRs. This means that, contrary to public reports, some allegations of use of force were referred to the officer's supervisor, but not investigated, or subject to findings or discipline. Similarly, in 2010, only 2 of the 17 allegations of discriminatory policing were investigated by OPA itself; only one was sent for investigation by the officer's supervisor. The remaining allegations were classified as PIRs or SRs. This should not happen. An OPA detective Sergeant should always investigate allegations as serious as excessive use of force and discriminatory policing. If the allegations are indeed unfounded or insignificant, an OPA detective Sergeant should have the skill and discretion to appropriately review and so conclude, and then document that conclusion. c. Investigation Complaints that OPA determines require investigation are classified either as (a) OPA-IS investigations that are conducted directly by OPA detective Sergeant Investigators and overseen by OPA Lieutenants, or (b) LI Investigations that are conducted by the officer's chain of command. In 2010, the line of command handled 13% of all complaints sent for investigation. (1) OPA-IS Investigations DOJ finds that OPA-IS investigations are generally thorough, well-organized, welldocumented, and thoughtful. However, in our review, we have discovered the following recurrent problems with OPA-IS investigations: an overabundance of leading questions, particularly favorable leading questions of the named officer; gaps in evidence collection; a misunderstanding of the nature of (and an apparent underlying lack of training regarding) unbiased policing investigations; and a tendency to give the named officer's testimony 3 disproportionate weight based on credibility over that of the complainant and, even, over other third-party witnesses. Additionally, we have reviewed files in which a labor representative, who is rightly present during the interview of the named officer, does not remain silent, but appears to be coaching responses to the investigating Sergeant's questions. We understand that the Director is working on a Training and Operations Manual for OPA that will codify OPA's policies, best practices, and its training requirements. This is overdue and, again, we hope that these findings provide a sense of urgency to complete this vital resource. (2) LI Investigations In unfortunate contrast to OPA-IS investigations, DOJ finds that the quality of the LI investigations is, by OPA's own admission, "appalling." Among other problems, OPA fails to probe the substance of the complaint, collect evidence, interview witnesses (even named witnesses or other officers on scene), avoid leading questions, and perform a thorough analysis of and justification for the proposed disposition. We understand that, because of such failures, OPA recently suspended referring complaints for investigation to the chain of command until sergeants receive additional training on how to conduct such investigations. The importance of this systemic failure should not be underestimated. When combined with the precincts' deficient treatment of PIRs and SRs described above, this means that (for example in 2010) nearly two-thirds (63%) of citizens' complaints are at great risk of being mishandled, unmonitored, or investigated inadequately. Finally, as OPA itself recognizes, these investigations, whether OPA-IS or LI, take a very long time. In 2010, the average investigation took 177 days, 3 days short of the 180 days permitted by the governing labor agreement, and just shy of six months. OPA recognizes this deficiency and is exploring in its Working Group how to improve turnaround time to its target time of 120 days, which would be nearly 33% quicker than the current average. In our consultant's experience, police departments of comparable size complete most internal affairs investigations within 90 days. d. Findings All investigations end with a finding. Currently, OPA uses eight types of findings: Sustained, Supervisory Intervention, Exonerated, Not Sustained, Unfounded, Administratively Unfounded, Administratively Exonerated, and Administratively Inactivated. DOJ has three concerns with OPA's practices at this stage of its investigations. First, and most importantly, OPA has overused and misused the quasi-finding of Supervisory Intervention ("SI"). According to public documents, a "finding" of SI means, "while there may have been a violation of policy, it was not a willful violation and/or the violation did not amount to misconduct. The employee's chain of command is to provide 4 appropriate training, counseling and/or to review for deficient policies or inadequate training." Between 2008 and 2010, significant percentages (between 12% and 19%) of OPA investigations were closed as SIs. As OPA personnel have admitted, an SI is not really a finding (i.e., a "ruling" or conclusion on the merits) and, under the labor contract, does not subject an officer to formal discipline. Furthermore, we learned that OPA uses SIs to avoid the "stigma" of finding misconduct and discipline in favor of (the laudable goal of) rehabilitating the named officer. However, in its review of OPA files, DOJ finds that later confirmation of rehabilitation has been weak, at best. An SI usually is followed by a one-time "counseling" session. Thereafter, neither OPA nor the employee's supervisor vigorously tracks an officer's improvement. Additionally, the quasi-finding of SI is not placed in an employee's personnel file and, historically, the underlying OPA file has not been subject to public disclosure, both in contrast to Sustained complaints. Both facts further undermine accountability and the expressed goal of rehabilitation. Complaints of significant importance, including allegations related to Use of Force and Discriminatory Policing, were often closed as SIs. Indeed, more often than not, between 2009 and 2011, when a complaint related to a Use of Force appeared factually founded, an SI was nonetheless the "finding." This should not happen. Second, as with its current classification system, DOJ finds that OPA's current findings system is not transparent or accessible because, by OPA's own admission, it confuses rather than illuminates an investigation's final conclusion. The demarcation line between the various findings is hazy at best, and not well understood by officers or the public. DOJ finds that this system discourages citizens from making complaints and, thus, lessens police accountability in Seattle. OPA has recognized the many shortcomings of this system and the aforementioned Working Group is seeking ways to simplify the findings system and clarify each finding's definition, so that it is more easily understood by the public. Again, we look forward to the results of the Working Group and hope this finding adds to the Working Group's sense of urgency. Third, and more generally, OPA's communications with the Complainant and the named Officers should be clearer, more regular, and as transparent as allowable under state and federal disclosure law. Currently, OPA sends one letter upon receipt and one letter at disposition. Unless the Complaint is sustained, there are no further communications or releases of documents. This practice does not encourage citizen participation. e. Director's Role and Certification When an OPA detective Sergeant and his supervising Lieutenant complete an investigation, they produce and send a Proposed Disposition Memo ("PDM") to the named officer's commander (i.e., Captain of the employee's Precinct) for comment. The PDM includes the Sergeant's and Lieutenant's analysis of the facts and law, and provides their proposed finding(s). Following the commander's review, the Director reviews the PDM and the 5 commander's comments and, if it is a finding other than Sustained, the Director certifies the completion of the investigation. OPA then transmits the outcome of the complaint in a final letter to the employee and the complainant. Again, generally, that letter is the only substantive information, and the end of all communications from OPA to the citizen. We have several concerns with this process. First, we understand that, in addition to comments from the employee's commander, undocumented communications occur between the Director and the command staff prior to the Director's certification. We believe that the Director and command staff should not communicate prior to the certification. The Director's decision whether to certify an investigation should be free from any influence, perceived or otherwise, of the command staff. Technical or tactical questions should be directed to the detective Sergeants and Lieutenants, who also should have the requisite expertise in this area. Second, we similarly question why the commander's concurrence is sought prior to the Director's certification. The Director should have no contact with those outside of OPA prior to certification. If the commander does not concur with the finding, that fact can be brought to the COP's attention following certification. The back and forth between the commander and the Director tolerated under the current system undermines the Director's independence and risks compromising the outcome of the investigation. Third, OPA personnel have acknowledged that it is "not an unusual" occurrence that a PDM is overturned following the Director's discussions with the commander. It is our position that there should not only be a very high bar to overturning a PDM, but the Director should also provide ample documentation validating this decision. The City should consider whether OPA- RB or the Auditor should have a formal auditing role in evaluating the findings, if not the discipline, in such situations. f. Discipline When a complaint is Sustained, the Director recommends discipline to the COP. An intermediary disciplinary meeting then is held with the Deputy Chief of Police ("D/C"), but without the COP. The D/C then proposes discipline to the COP, at which time the COP makes a preliminary disciplinary decision. The employee is then afforded an opportunity to present his or her position directly to the COP (referred to as a Loudermill hearing). The Chief then makes a final disciplinary decision, which, of course, can be appealed or grieved. We have two concerns with this process. First, this process, which already follows an investigation that on average lasts six months, further delays resolution to the dissatisfaction of the officer and the complainant. To cut down on the delay, SPD should consider dispensing with the intermediary meeting between D/C and OPA, so that OPA's recommendation can proceed straight to the COP. Additionally, we understand that parallel administrative and criminal investigations are permitted under the current labor union contract. Consistent with our recommendations in our prior Garrity letter, we would encourage exploring this possibility in certain cases. 6 Second, there are wide-spread concerns that discipline is inconsistently applied. As previous Directors have recommended, SPD, in consultation with OPA, should consider the development of a discipline matrix that guides the Director and the COP when making disciplinary decisions. Of course, special circumstances may arise that necessitate deviation from the matrix. However disciplinary guidelines that are applicable to most situations will provide for a greater degree of uniformity, fairness, and accountability throughout the Department. 3. Additional Concerns about the Accountability System a. Structural Concerns Regarding OPA The independence of OPA's civilian Director is the lynchpin of the public's trust in the system. This independence must be both actual and perceived. Unfortunately, concerns about the independence of the OPA Director have arisen in our investigation from a broad range of persons and communities. An OPA Director must never become or be widely seen as an advocate of SPD, but must remain objective and welcoming of any criticism of SPD practices from persons of good will. This is more easily said than done, particularly as the Director reports to the COP, sits horizontally as an equal to the command staff, and leads a team of sworn SPD personnel. It is a tough job. To further promote the Director's independence, we recommend that the City consider a shorter number of terms (from three to two) for the Director and set a term that ends a short time after the election of the Mayor, who appoints the Director and should be responsible for the Director's actions and responsive to the community's perceptions. We also recommend that a springing mandatory review occur at the end of the Director's term or upon the election of a new Mayor. This review should expressly require community input. The City should also consider enumerating the situations in which the Director should report an event directly to the Mayor's office, as opposed to reporting to the COP or to the public at large about policy matters. These reporting requirements could include, for example, reporting to the Mayor those communications with command staff prior to certification, or cases in which the COP does not follow the Director's recommendation. Additionally, there currently is no operational Captain of OPA-IS, as there traditionally has been. The lack of an operational Captain may undermine the stature of OPA in the eyes of SPD personnel. Presently, a Captain embedded within OPA leads an "Ethics and OPA-IS" hybrid. This position does not have a clearly defined role at the moment. Furthermore, although the quality of OPA-IS investigations is high and the personnel strong, we recommend institutionalizing the strength of this important department. OPA duty must be mandatory for any aspiring Sergeant. There should be minimum scores and competency for OPA duty. As in other departments, OPA must be drawn, and be seen to draw from the best of the best investigators. Increased pay or a bonus for OPA investigators should also be considered. 7 Finally, consistent with SPD Manual and City policy generally, SPD must have and vigorously enforce a zero-tolerance policy regarding any interference with OPA-IS or LI investigations. b. In car video Over ten years ago, in July 2011, SPD began a pilot program that equipped patrol cars with video cameras. Unfortunately, the policies governing the use of video remain unclear in several key respects and the training for the proper use of video is inconsistent. Additionally, several technical issues with synching and preservation remain unresolved. More importantly, it is our position that, contrary to SPD's apparent current policy, supervisors need access to videos to adequately supervise and train their officers and thus that SPD must preserve patrol videos for supervisory and administrative review. We understand that the Director, Auditor, and COP have formed an Audit Committee that will issue a report on these failings (as well as developments in other data tracking) soon. We look forward to receipt of this report. c. OPA-RB OPA-RB faces the following three difficulties in performing its tasks. First, a fine line exists between educating the community about the accountability system and appearing to advocate for it. Likewise, a tension exists between those outreach efforts and making recommendations to improve the system. Second, OPA-RB has reported encountering reluctance from other branches of the oversight system to review cases and engage in outreach. Finally, unpaid volunteers do all of the work of OPA-RB. If the City Council finds this arm of police accountability to be valuable (and we believe it could play a valuable oversight role), then the City needs to staff the board with paid employees and needs to provide it enumerated authority sufficient to insulate it from these difficulties. 8 APPENDIX E: GARRI LETTER U.S. Department of Justice Civil Rights Division Special Litigation Section - PHB 950 Pennsylvania Ave, NW Washington DC 20530 NOV 23 2011 The Honorable Michael McGinn Mayor City of Seattle 600 4th Avenue, 7th Floor Seattle, WA 98124-4749 Re: United States' Investigation ofthe Seattle Police Department - Garrity Protections Dear Mayor McGinn: On March 31, 2011, the Civil Rights Division and the United States Attorney's Office for the Western District of Washington initiated an investigation of the Seattle Police Department ("SPD"), pursuant to the Violent Crime Control and Law Enforcement Act of 1994,42 U.S.C. ? 14141, as well as the anti-discrimination provisions ofthe Omnibus Crime Control and Safe Streets Act of 1968,42 U.S.C. ? 3789d, and Title VI of the Civil Rights Act of 1964,42 U.S.C. ? 2000d. At the beginning of our investigation, we committed to providing SPD with real time technical assistance to enhance SPD practices and procedures, and to ensure compliance with constitutional rights. During our meetings with ChiefDiaz and the SPD command staff in May 2011 and September 2011, we advised that, if appropriate, we would provide in writing specific recommendations prior to completion of our investigation. In this letter, we convey our recommendations regarding SPD's practices relating to an officer's protections against self- incrimination pursuant to Garrity v. New Jersey, 385 U.S. 493 (1967). Garrity provides important and fundamental protections for police officers, but its protections are limited. Our investigation has shown that SPD attempts to apply Garrity to all use of force and police involved shooting incidents. SPD's inappropriate blanket invocation of Garrity may result in the exclusion of important evidence from an investigation. Moreover, SPD's failure to shield criminal investigators from Garrity materials could taint and render unusable other critical evidence. These practices compromise both SPD's ability to supervise officers' use of force, and its ability to fully and efficiently conduct criminal and administrative investigations. Put simply: This practice makes it too difficult to quickly exonerate officers who have followed policy and to properly discipline officers who have not. Further, these practices compromise the ability of prosecutors or other outside agencies to adequately assess incidents and to hold officers acconntable for their actions. The net effect of these consequences is diminished public trust in SPD. ------ - - - - - _._---_._---------------------- -2- This risk created by current SPD Garrity-related policies and practice is illustrated by a recent Seattle City Attorney's prosecution of an officer for fourth degree assault. It is our understanding that the defendant officer has filed a motion to dismiss the charge, claiming that prosecutors improperly used his "compelled and involuntary use of force statement" in deciding to bring the charges. While we take no position on the ultimate merits of the motions or this particular case, SPD's policies and practices on Garrity have exposed it to these claims, which have the potential to weaken both external public trust, and internal trust among the rank and file, who deserve a clear understanding of how their statements may be used. Given the serious nature of our concerns, we are providing this technical assistance in a stand-alone letter that highlights our recommendations. The recommendations detailed below were developed in close consultation with our police practices consultants, who include a former chief of police and deputy sheriff, and follow the productive dialogue we have had with SPD to date. We strongly urge SPD to consider these recommendations immediately in revising its policies, procedures, practices, and training associated with Garrity admonitions. This letter is limited to providing guidance regarding the issues associated with improper administration of Garrity warnings, and does not address any other areas of our investigation. I. Garrity v. New Jersey In Garrity v. New Jersey, 385 U.S. 493 (1967), the Supreme Court held that an incriminating statement made by a police officer is inadmissible against the officer in a criminal trial if the officer made the statement under the threat that the officer would lose his or her job if the officer invoked the right to remain silent. The Court concluded that, under those narrow circumstances, the statement would be considered coerced because the officer was denied any meaningful opportunity to assert his Fifth Amendment rights. Id. at 499-500. Garrity is premised on the fact that it is coercive for the government to put an officer "between a rock and a whirlpool," Garrity, 385 U.S. at 498, by forcing the officer to choose whether to incriminate himself or to lose his job for invoking the Fifth Amendment. United States v. Cook, 526 F. Supp. 2d 1, 6-7 (D.D.C. 2007). In order for an officer to establish that a statement is Garrity-compelled, he must demonstrate that he wanted to invoke the Fifth Amendment, but was prevented from doing so by an express or implied threat that he would be fired if he remained silent. United States v. Trevino, No. 05-51309, 215 Fed. App'x. 319, 321-22 (5th Cir. 2007). To make this showing, he must establish both that he subjectively believed that he would be fired for refusing to talk, and that this belief was objectively reasonable. Cook, 526 F. Supp. 2d at 7-8; United States v. Vangates, 287 F.3d 1315, 1322 (11th Cir. 2002). Garrity is intended to apply narrowly to situations where the officer is required to give a statement or face termination, and the officer reasonably believes that the statement could be self-incriminating. It is not meant to apply to officers' routine documentation of their activities, including, for example, the completion of incident and use of force reports, or to discussing the same with department officials. Cook, 526 F. Supp. 2d at 8 ("Garrity does not stand for the proposition that a statement made in a standard report is coerced whenever an officer faces both the remote possibility of criminal prosecution if he files the report and the arguably even more -3- speculative possibility of termination if he declines to do so."); United States v. Camacho, 739 F. Supp. 1504, 1516 (S.D. Fla. 1990) (declining to find that "the mere existence of a departmental policy of disciplining those officers who refuse to give statements always operates as a matter of law to render officer statements involuntary"); United States v. Tsou, 1993 WL 14872, at *4-5 (5th Cir. Jan. 18, 1993) (unpublished) (holding that an FBI agent's statement was not compelled, despite an FBI policy requiring agents to cooperate with any administrative investigation). II. Recommended Garrity Reforms A. Standard Use of Force Statements Are Not Compelled Statements Under Garrity. We recommend that SPD amend its Use of Force policy to clarify that standard use of force statements are not compelled under Garrity. Currently, SPD's Use of Force policy requires that each use of force statement be preceded by the following language: "This is a true and involuntary statement given by me in compliance with Section 6.240 of the Seattle Police Department Manual." The policy clarifies that "[n]o other language will be acceptable." SPD Department Policy and Procedure ("DP&P") 6.240.XII.A.4. Mandating the inclusion of this language in all use of force statements has resulted in all such statements being treated by some as potentially Garrity-compelled. Some SPD officers have stated that they subjectively believe that use of force statements are considered Garrity-compelled, and the SPD Office of Professional Accountability ("OPA") also mistakenly considers this language to be a grant of Garrity protection. This shared understanding by various audiences could undermine a valid assertion that these routine use of force statements are not Garrity-compelled, and could severely limit the ability of prosecuting agencies to review and possibly use the statements. Attempting to provide blanket Garrity protection for every use of force statement is bad policy and goes beyond what is required by law or necessary to protect officers' Fifth Amendment rights. Use of force is sometimes unavoidable in police work, but is a serious event. Routine police practice requires departments to accurately record the circumstances surrounding uses of force. Use of force statements are an invaluable training and officer-safety tool, and they are critical for maintaining accountability and managing risk. A use of force statement should be a factual recitation of events that constitutes a necessary and routine part of an officer's job duty. SPD officers file approximately 500 use of force reports each year as an obligatory part of their professional responsibilities. Of these 500 matters, we are aware of a very small number that resulted in a criminal referral, much less prosecution. Thus, in any given case, the completion of a report carries only a "remote possibility" of criminal prosecution or administrative investigation. See Watson v. Cnty. of Riverside, 976 F. Supp. 951, 955 (C.D. Cal. 1997) (finding that officer's report regarding an arrest and use of force was "a requirement of [officer's] job and did not constitute a compelled self-incrimination"); United States v. Hill, No. 1:09-CR-199TWT, 2010 WL 234798, at *9 (N.D. Ga. Jan. 13, 2010) (finding that a Sheriff's Department policy stating that adverse action may be taken against employees for refusal to cooperate in investigations was insufficient to invoke Garrity, as that interpretation would mean that "every public employee statement is subject to Garrity immunity"); Devine v. Goodstein, 680 F.2d 243, 247 (D.C. Cir. 1982) (holding that a public employee could not have reasonably believed that preparation of a routine report would be used in a criminal investigation). -4- SPD's attempt to treat all use of force statements as Garrity protected does not conform with best practices and is unlike its treatment of other SPD reports. For example, SPD does not require Garrity language to precede SPD incident reports, log sheets or booking records, even though those reports could expose officers to the same hypothetical jeopardy. Treating use of force statements differently is unnecessary. Moreover, the presumption implicit in SPD policy and practice, i.e. that every use of force implicates criminal conduct by an officer, is unfounded and undermines public trust. It also compromises the ability to hold officers accountable in those infrequent instances where there is officer misconduct. In those rare circumstances where an officer reasonably believes that a truthful statement will be self-incriminating, the officer can affirmatively exercise his or her Fifth Amendment rights and refuse to make a statement. However, the Fifth Amendment privilege should only apply where "the claimant is confronted by substantial and 'real,' and not merely trifling or imaginary, hazards of incrimination." Marchetti v. United States, 390 U.S. 39, 53 (1968); see also Hiibel v. Sixth Judicial District Court of Nevada, Humboldt Cnty. 542 U.S. 177, 190 (2004) (defendant may invoke the Fifth Amendment privilege where there is "reasonable cause to apprehend danger from a direct answer"). Once an officer has invoked the right to refuse to make a statement, SPD, after consulting with a prosecutor as discussed below, can determine if the officer's statement should be compelled. B. Officer-Involved Shooting Statements Should Not Be Treated As Compelled Under Garrity. We recommend that SPD revise DP&P 8.060 regarding Officer Discharges of Firearm so that SPD does not treat all statements provided during officer-involved shooting ("OIS") investigations as Garrity-compelled. Currently, when a shooting results in death or injury, SPD's policies direct that SPD compel the following statements from the involved officer: (1) a public safety statement at the scene of the shooting, where a supervisor orders an involved officer to answer standardized questions related to the firearms discharge, such as suspects' descriptions, whether evidence requires protection, and whether there are any known witnesses; (2) a verbal narrative of what occurred with the Homicide/Assault Unit either on scene or at the office; (3) a scene walkthrough; and (4) a written use of force statement to be submitted within three days. DP&P 8.060.III.B.6.a-d; 8.060.I.D. This "Garrity advisement" "orders" the Involved Officer to answer questions and states that a "refus[al] to answer questions relating to the performance of ... official duties ... could result in dismissal." For the same reasons stated in the preceding section, SPD should not attempt to confer automatic Garrity protections on the verbal and written statements provided following an OIS. As these are some of the most serious uses of force, SPD should also shield criminal investigators from any Garrity statements to avoid any potential issue of taint. As explained above, an officer may invoke the Fifth Amendment privilege if he or she believes a truthful answer will be self-incriminating. Absent affirmative invocation, SPD should not universally attempt to render these statements compelled under Garrity. -5- C. SPD Should Provide All Officers with the Opportunity to Give Voluntary Statements Following Use of Force Incidents. One significant concern with SPD's policy is that the policy makes it more difficult for an officer to provide a voluntary statement. While SPD must abide by an officer's Fifth Amendment right against self-incrimination, SPD's application of Garrity attempts to overly expand this protection. In practice, this may undermine the enforcement of other constitutional rights by compromising prosecutors' ability to investigate and prosecute potential criminal misconduct by officers. Attempting to automatically "Garritize" all officer statements risks generating protracted litigation that could result in inconsistent treatment of statements and create uncertainty as to the exact scope of the officer's protections. Use of force is a necessary reality in law enforcement. However, it is critical that the community has confidence that force will be used appropriately, and that individuals will be held accountable when it is not. As many agencies and officers recognize, it is almost always in an officer's best interest to provide his or her own statement regarding the force that was used. Even in the minority of cases that may result in a referral to a prosecutor, a review of an officer's statement usually provides a more complete view of an incident that quickly ends an investigation. In those relatively rare circumstances where an officer might have engaged in criminal misconduct, it is a disservice to the Department, those officers who follow the law, and the community to unnecessarily create artificial obstacles to holding that officer accountable. SPD's current policies and practices do just that. SPD's automatic wholesale Garrity invocations could result in the exclusion of important evidence from an investigation, and SPD's failure to shield criminal investigators from Garrity materials could result in unnecessary litigation over the effects of any taint. While taint seldom constitutes the prohibited use contemplated by Garrity, careless practices can result in confusion, unnecessary litigation, and the unnecessary expenditure of resources. In short, these policies could prevent prosecutors from using statements that would make an officer's culpability more clear, which in turn has the effect of undermining the community's ability to hold officers accountable when they commit criminal misconduct. In the San Diego Police Department's ("SDPD") peer review of the August 30, 2010, SPD shooting of the late John T. Williams, SDPD provided a description of its Garrity procedures, which appear consistent with generally accepted practices. The SDPD Homicide Unit "does not compel statements from shooting officers, since they would be inadmissible in a District Attorney's review." (emphasis in original).1 Instead, when a shooting incident occurs, an officer's attorney is sent to the scene, and SDPD "ask[s] the attorney if the officer is willing to participate in a voluntary walk-through of the incident scene and provide a voluntary statement about the incident" (emphasis in original). If the officer agrees, the officer participates in both with an attorney present. Id. If the attorney or officer does not agree, the investigation continues without the officer's participation, even though, as SDPD recognizes, "the most important part of Letter from San Diego Police Department Lieutenant Kevin Rooney to Chief of Police John Diaz, Seattle Police Department, re August 30, 2010 Shooting Involving Officer Ian Birk #7505, Jan. 20, 2011, available at http://www.seattle.gov/police/OPA/Docs/Birk/SanDiego_Report.pdf. 1 -6- an officer-involved shooting investigation is the shooting officer's statement." Id. In the peer review, SDPD Lieutenant Kevin Rooney observes that "absence of a legally-admissible officer statement" in the SPD shooting of the late John T. Williams made it "impossible to know what Officer Birk saw, how he felt, and why he chose to use deadly force" because there was "no first person account of what happened." Id. Adopting a similar type of protocol to SDPD would give an SPD officer the choice of providing a voluntary statement, as well as facilitate a full and fair investigation. D. SPD Should Consult with the Prosecuting Agency Before Administering Garrity Admonitions Following Officer Involvement in Serious Use of Force Incidents. As noted above, it is a best practice in law enforcement to not compel statements pursuant to Garrity, at least until the completion of the criminal investigation. If the officer invokes his or her Fifth Amendment right against self-incrimination, the criminal investigation proceeds without the officer's statement. Where a statement from the officer is compelled, usually after closure of the criminal investigation and as part of the agency's internal administrative investigation, a generally accepted practice is to consult with a prosecutor before compelling the statement and thereby granting Garrity protections. In contrast, SPD's policies automatically attempt to grant Garrity protection, depriving SPD of the opportunity to consult with prosecutors based on the circumstances of each incident. The protections of the Fifth Amendment should involve incident-specific consultation with a prosecutor in advance of granting Fifth Amendment privileges. SPD should develop a formal process for conferring, from the outset, with prosecutors in every serious use of force or potentially criminal matter to determine whether a criminal investigation or prosecution is warranted. To ensure that this process is not itself an obstacle, serious uses of force can be defined by the level of injury sustained by a subject, or any use of deadly force, regardless of the severity of the injury. Trained investigators should roll-out to OIS and other serious uses of force to make the determination about whether to consult a prosecutor. Supervisors should similarly be trained to make determinations on scene about when to refer use of force incidents to OPA, which can then decide whether to refer the incident to a criminal prosecutor. After consultation with prosecutors, SPD may learn that a criminal investigation is not warranted or viable and decide to compel the statement of the officer. If so, SPD should compel a statement only after taking appropriate steps (outlined below) to ensure the separation of the criminal and OPA administrative investigations in order to maintain the integrity of the criminal investigation, should it later become apparent that a criminal prosecution is appropriate. E. SPD Should Carefully Administer Bifurcated Criminal and Administrative Investigations. There is no question that SPD has its own need to determine what happened in an officerinvolved shooting or use of force. Thorough, fair, and timely administrative investigations are one of the most critical responsibilities of a police department. Effective investigations of -7- shooting or use of force incidents can quickly identify whether there are any policy, training, officer-safety, or accountability concerns. Currently, SPD tolls its administrative OPA investigations pending the completion of a criminal investigation, which creates several problems. First, this can result in a lengthy time delay, which in turn can compromise an investigation as witness memories fade and facts become more difficult to gather. Second, a delay in corrective action diminishes the meaningfulness of the intervention or disciplinary action on the individual officer and the department as a whole. Third, unnecessary delays in completing an administrative investigation undermines the public's confidence that SPD takes police misconduct allegations seriously, and creates additional stress for the employee who awaits the final determination of the investigation. To avoid the problems inherent in tolling an administrative investigation, we recommend that SPD develop procedures for conducting effective parallel criminal and administrative investigations that do not compromise the integrity of either investigation. To do so, SPD should not compel a statement from the subject officer until it has been determined that a criminal investigation is not viable or has been completed. This determination should be made in consultation with the prosecuting agency and the OPA Director. With the exception of compelling the officer's statement, the administrative investigation may proceed even while the criminal investigation is ongoing. SPD must also develop and implement processes to ensure the separation of the criminal and administrative investigations in the event that a statement is Garrity-compelled before completion of the criminal investigation. This process will ensure that any information learned from a Garrity-compelled interview statement does not accidentally taint other parts of the criminal investigation. III. Recommended Remedial Measures We recommend that SPD revise the following policies pursuant to the recommendations contained in this letter. These measures are not meant to be exhaustive, but highlight key recommendations that will remedy the deficiencies described in this letter. SPD should: 1. Revise SPD "Use of Force" Policy, "DP&P" 6.240.XII.A.4. SPD should make clear that all officer statements in use of force reports are part of each officer's routine job duties and are not compelled statements under Garrity. SPD DP&P 6.240.XII.A.4 should make clear that all use of force statements should be true and voluntary. Add a provision to the SPD manual that officer's statements and reports are voluntary and provided as part of an officer's routine job duty. The only officer statements to be considered compelled under Garrity are those given following an explicit written Garrity advisement Revise SPD "Officer Discharge of Firearm" Policy, DP&P 8.060: a. Revise DP&P 8.060.I.D and 8.060.III.B.2 to eliminate any reference to the public safety statement being compelled or "Garrity protected." 2. 3. -8- b. Revise DP&P 8.060.III.B.6 to eliminate the reference to the Garrity advisement so that the policy reads: "In summary, when a shooting results in death or injury, the Involved Officers may expect the following requests for information." Where an officer believes that providing a verbal or written statement will be self-incriminating, the officer should not be compelled to provide a statement without a prior consultation with the prosecuting attorney's office and the OPA Director subsequent to the completion of any criminal investigation.2 4. Ensure that criminal investigators and prosecuting attorneys are appropriately shielded from any Garrity compelled statement. Develop a policy regarding the referral of potentially criminal conduct to the prosecuting agency by an OPA investigator or SPD supervisor who may determine whether there is sufficient reason to indicate that an officer or employee has engaged in criminal conduct in the course of complaint intake or investigation. Revise SPD "Public and Internal Complaint Process" Policy, DP&P 11.001.V.J., to state that OPA administrative investigations will begin immediately, irrespective of the initiation of criminal proceedings, provided that the OPA administrative investigation does not interfere with a criminal investigation. The subject officer shall not be compelled to provide a statement to OPA where there is a potential or ongoing criminal investigation or prosecution of an officer, until the remainder of the OPA investigation has been completed, and only after consultation with the prosecuting agency and the OPA Director. * * * * 5. 6. If an officer consults with counsel on whether to invoke his constitutional rights, it is important that counsel be independent and not be counsel for SPD, the City of Seattle or a prosecuting agency. 2 . . -9- We hope that thi s letter provides helpful useful technical assistance, and remain open to continuing a dialogue about how best to implement these policy changes. We greatly appreciate the cooperati ve and productive rel ati onship that we have had with the City of Seattle thus far, and look forward to working with you as the in vestigation proceeds. Please note that this lctter is a public document and will be posted on the Civil Rights Division's website. If you have any questions, please feel free to contact us. n 4~ Jenny A. Ourkan United States Attorney United States Attorney's Office Western District of Washington cc: Chief John Oiaz Seattle Police Department Peter Holmes Seattle Ci ty Attorney . .