Michael Gennaco Stephen Connolly Phillip Hansen Teresa Magula Samara Marion Julie Ruhlin City of San José INDEPENDENT AFTER ACTION REGARDING THE EVENTS OF MAY 29 – JUNE 7, 2020 October 2021
323-821-0586 7142 Trask Avenue | Playa del Rey, CA 90293 OIRGroup.com
Table of Contents Introduction ........................................................................................................ 1 Methodology ...................................................................................................... 6 Interviews with SJPD Sources ....................................................................... 6 SJPD Documentary & Digital Evidence ......................................................... 6 Evaluation of Body Worn Camera Footage ................................................ 7 Outreach to City Council & Other City Officials .............................................. 7 Community Input ............................................................................................ 8 SJPD Preliminary After Action Report: General Impressions ......................... 10 Overview of Operational Timeline ................................................................... 16 May 29: From Peaceful to Riotous ............................................................... 16 May 30: Unrest Continues ........................................................................... 22 May 31: Tactical Planning and Curfew ........................................................ 24 June 1 to 4: Evolving Tactics & Changing Crowd ........................................ 26 June 5 - 7: Mediation Over Escalation ......................................................... 27 Timeline ........................................................................................................... 30 Assessment & Analysis ................................................................................... 39 Command and Control: Leadership Void ..................................................... 39 Operational Planning ................................................................................ 39 Span of Control: Lieutenants and Command Supervision ....................... 40 Reliance on Special Operations ............................................................... 41 Real-Time Communication: Missing the Lieutenant Link ......................... 41 Training Concerns ........................................................................................ 42 Training Prior to May 29, 2020 ................................................................. 43 Dispersal Orders and Unlawful Assemblies ................................................. 46 Deployment of Less Lethal Force: Data and Analysis ............................ 52 Force Reporting and Review Deficiencies ............................................... 54 Force Deployment: Types and Uses ........................................................ 58 Improvised Uses: Adapting and Learning ................................................ 61 High-Velocity Rounds ............................................................................... 63 A Shifting Landscape: PIW Policy Revisions and Deployment Issues .... 64
New Approaches, Different Results ......................................................... 70 Analysis of Arrests ....................................................................................... 71 Arrest Data: Overview .............................................................................. 71 Arrest Tactics: “Wait Until They Leave” .................................................... 72 Transportation and Processing ................................................................ 74 Field Jail and Releases ............................................................................ 76 Curfew as a Tactic ....................................................................................... 77 City Collaboration: EOC ............................................................................... 81 Effective Mutual Aid ..................................................................................... 83 Communication with the Public .................................................................... 86 Equity Considerations ...................................................................................... 91 Conclusion ....................................................................................................... 93 APPENDIX A: Recommendations ................................................................... 95
Page| 1
Introduction What follows is an independent assessment of the San José Police Department’s response to protest activity in the City as it unfolded last year between May 29 and June 7. It was prepared by OIR Group, a team of police practices experts based in southern California that began its work on the project earlier this year.1 The report is meant to supplement the Department’s own detailed After Action Report (“AAR”), which it produced at the end of last summer and presented to the City in September 2020. Accordingly, it discusses the AAR’s findings and recommendations from an outside perspective, appreciating the research that the AAR reflected and the self-scrutiny it produced while re-evaluating some of its particulars through a broader lens. It also extends beyond the parameters of a typical internally generated AAR in that it includes insights from other stakeholders. This Report also differs from the AAR in that more than a year has passed since the key events of summer 2020. Much has changed, in terms of both the national backdrop and the local civic landscape. In fits and starts, we have begun to emerge from the COVID-19 pandemic that helped shape last year’s anxious, restless public mood. Derek Chauvin has been convicted and sentenced for the murder that initiated an unprecedented movement across the country. San José itself has a new Chief of Police and a new City Manager. And national attention converged with local tragedy after the VTA rail yard shooting on May 26 of this year. 1 OIR Group is led by Michael Gennaco, a former federal prosecutor and a nationally recognized leader in the field of police oversight. For 20 years, OIR Group has worked in a range of jurisdictions throughout California and in several other states. It specializes in the independent review of police operations, including officer-involved shooting cases, allegations of misconduct, uses of force, and systemic matters of policy and procedure. Since last year, it has evaluated protest-related policing issues in several cities, including Santa Monica and Santa Rosa in California, Kalamazoo (MI), and Iowa City (IA). Public reports of each of these projects can be found at OIR Group’s website: www.oirgroup.com.
2 | Page
Notably, San José has moved forward with a series of initiatives that are meant as a response to the tensions, inequities, and historic legacies that George Floyd’s murder exemplified, and that fueled the subsequent demands for change. These include the multi-faceted “Police Reforms Work Plan” (that encompasses this Report along with 19 other items) and the “Reimagining Community Safety” committee that began its work in the spring of this year. We mention this larger context because of how it informs our understanding of this Report’s potential significance. For many reasons, including accountability and the need to adapt to the dynamics of contemporary protest actions, it is important to study the events of May 29 to June 7, 2020 and to learn from them. This Report seeks to do so, and it describes a number of technical, tactical, and philosophical adjustments that SJPD has made, or should make, in response to that period. But the events that fall within the scope of this project are components of a dynamic that extends well beyond it. Accordingly, we hope that our findings and recommendations will reinforce – and be reinforced by – the City’s broader commitments to change. In terms of our specific mandate, our central impressions can be summarized straightforwardly. They track the SJPD After Action Report in some – but not all – key respects. • The outrage over George Floyd’s murder on Monday, May 25, triggered waves of impassioned protest that built steadily across America as that week progressed. While SJPD officials were tracking national developments and attempting to stay apprised of relevant intelligence as it related to San José, they did not have an accurate gauge for the scope of what began in San José on Friday, May 29. • Prepared for a different nature of protest activity based on their intelligence-gathering leading up to the day and past protest activity in their City, but not expecting the size and hostility or recalcitrance of some of the crowd that actually formed, SJPD struggled to maintain order in the City over the course of several hours, beginning in the late afternoon of that Friday. Indeed, May 29 turned out to be the most intense, violent, and disruptive day of the local protests according to various metrics – including the amount of force used by SJPD officers. • On May 29 (and subsequent days of the unrest) a portion of the crowd engaged in a series of destructive, antagonistic, and assaultive
Page| 3
behaviors that had several implications. These included the galvanizing of other protesters who may themselves have been less inclined toward aggressive behavior, thereby influencing the overall dynamic. Other protesters attempted to self-police and discourage violence with mixed success. • Another effect of aggression by some of the crowd was the SJPD response it provoked, including the issuing of dispersal orders and the use of various force options to enforce those orders and respond to crowd violence. • While many if not most officers performed with restraint and determination in extremely adverse circumstances, the apparent excessive or malicious actions of a few SJPD members greatly undermined public perceptions of agency legitimacy. • SJPD’s challenges in coordinating an effective response were magnified in part by inadequate deployment and experience. The conflict’s distinctive challenges, and the shortage of coordinated, well-supervised direction, contributed to force deployments that were both high in volume and at times indiscriminate. In some cases, this resulted in serious injury to protesters and further damaged the public’s confidence. • Indeed, many of the force deployments and acts of police aggression surprised and infuriated crowd members and at times did more to provoke than to discourage further resistance. • The City’s decision to impose a curfew, beginning Sunday, May 31, and extending for four nights, assisted in SJPD’s ability to regain control and to isolate problematic individuals and groups from the larger body of peaceful protesters. At the same time, questions about its necessity and rollout were legitimate in ways that should inform future exercise of the authority. • To its credit, the Department evolved in its handling of the protest activity and adjusted to shortcomings as they emerged from day to day. Changes included a significantly enhanced supervisory presence in the field, formalized briefings and mission planning, increased deployment,
4 | Page
and a reconsideration of the “skirmish line” strategy that had engendered considerable conflict on May 29 and subsequent days. • One shift that seemed less advisable was a mid-week adjustment to the less lethal munitions policy, expanding authorization for use at a time when legitimate concerns were arising about deployments in the first days of the unrest. However, and commendably, the Department addressed this with another, more restrictive version that it introduced by mid-June of last year. • In their various forms, SJPD’s communications with the public were of mixed effectiveness. This was true in terms of both specific directives (such as the dispersal orders and curfew announcements) and the larger framing of the daily struggles the Department was experiencing in the field. • As the summer moved forward, concern about SJPD’s handling of the protests led to responsive review projects, even as this specific enforcement issue was incorporated into a larger dialogue about the future of public safety in the City. • Individual officer accountability for specific uses of force or other allegations of misconduct remains a “loose end” that will impact community confidence. Interestingly, the adversarial nature of the encounters between SJPD and the protesters had a self-perpetuating quality, particularly at the outset of what the Department refers to in its AAR as the “operational period.” There is, of course, something inherently unenviable about the dual prominence of the police as both the subject of vehement protest and the street-level regulators of it. In San José in particular, several of the Department’s representatives left us with the impression that this dynamic was very much compounded by their own disappointment and dismay. They seemed genuinely surprised that they had been grouped in with the bad actors of the policing world, despite their deliberate and successful efforts at positive community relations and accountability. And they were insistent that the violence and aggression they experienced was genuinely and uniquely dangerous. Throughout this Report, we discuss the different aspects of this perception and look at the supporting evidence provided by SJPD.
Page| 5
Conversely, the large majority of protesters who did not engage in vandalism, looting, or assaultive behavior struggled with the seeming unfairness of the widespread “crackdown” that ensued on Friday and some of the subsequent days. They saw their own refusal to comply with dispersal orders that they believed to be unjust as fitting into the best traditions of righteous protest against oppression. And they saw the heavy-handed nature of the police response as reinforcing their worst perceptions about American law enforcement. Below, we share some of the views that were presented to us by community members and other stakeholders – whom we took to be credible and understandable. There are no easy solutions to the disconnect described above. Indeed, as with many of the more challenging divisions in contemporary society, the differing perspectives have their own elements of validity as well as a common spirit of righteousness. And, as jurisdictions across the country continue to illustrate, protests aimed at police brutality are especially difficult for public safety to navigate. Law enforcement’s regard for order and its reliance on cooperation do not mix well with demonstrations that, by their very nature, are intended to challenge and disrupt – and yet are also susceptible to unacceptable excesses by some. Ideally, though, the widespread, energetic insistence on change can be a starting point rather than a temporary wrinkle in the status quo. Constructive dialogue, responsive reforms, and an ability to learn from experience can contribute to shifts that redound to the benefit of all stakeholders. To the credit of involved parties within and outside SJPD, several concrete steps have already been accomplished since last June, both in the arena of demonstration protocols and the broader aspects of police-community relations. Much of the work, though, has yet to be completed, and the fruits of structural changes have yet to be fully realized. We hope this Report will contribute to that process through both increased understanding and constructive change.
6 | Page
Methodology The City and the Independent Police Auditor framed the scope of work for this project to focus our evaluation on the specific events of May 29 to June 7, while also ensuring that a range of perspectives would be included in the development of findings and recommendations. Because the actions of the Police Department were the central focus of this assessment, SJPD was our main source of raw information. We also benefitted from the insights of others who were connected to the events of May 29 to June 7 as responders, participants, or in their capacity as City leaders and officials.
Interviews with SJPD Sources The San José Police Department cooperated fully with our requests for information of various kinds. Foremost among these was a series of interviews with Department personnel, including current leadership of the Department. In all, we met with over a dozen SJPD representatives. We reiterate our appreciation here for each of their perspectives and willingness to participate in this review.
SJPD Documentary & Digital Evidence At our request, SJPD provided an extensive amount of relevant documentary evidence related to the events of May 29 to June 7, 2020. The documents included: • Internal memoranda related to the civil unrest, including memoranda related to policy changes and requests for additional less lethal munitions • Deployment counts for all days, including mutual aid • Training Reports and all training material related to crowd management and special operations
Page| 7
• Statistics for the period in question; specifically, arrests, citations, and damages • All materials used to create the Department’s own After Action report, which included a timeline of events and clips from media footage • Materials related to intelligence gathering efforts, such as printouts of screenshots from social media sources • Operational materials, including Operation Plans, related to the events of May 29 to June 7 • All relevant Department policies regarding uses of force, tactics, and First Amendment assemblies • Use of Force Reports submitted by SJPD officers detailing their uses of force from May 29 to June 7 OIR Group team members also collected digital evidence from public sources, including both traditional and social media platforms, to better inform our review. OIR Group discovered digital evidence in personal and organizational Facebook and Instagram pages and Twitter feeds, and we reviewed streaming video footage from local media sources. OIR Group also reviewed Computer Aided Dispatch (CAD) evidence, including written logs and radio broadcast communications, from May 29 to June 7. We used this detailed evidence, plus body worn camera and media video, to construct the detailed timeline included in this Report. Evaluation of Body-Worn Camera Footage Central to our understanding of what happened were the body-worn camera (“BWC”) videos provided by SJPD. We reviewed hours of BWC footage of officers deployed to various locations of interest throughout the City, focusing on critical moments of SJPD observations and decision-making.
Outreach to City Council & Other City Officials As part of our outreach, we also invited the Mayor and all current or recent City Council members to share their experiences and perspectives with us;
8 | Page
each had helpful insight into the events of the summer and their aftermath. We also reached out to other City leadership and officials who were particularly relevant to our inquiry and we appreciated the perceptions and viewpoints each had to offer. OIR Group also reviewed City Council meeting agenda items and listened to recorded City Council sessions from mid- and late-2020 and early-2021 during which relevant items were discussed by the Council and lengthy public comment was received.
Community Input One important element of our assignment was to engage with the San José community and listen to voices from all perspectives, to gain insight into the events of May 29 to June 7 and the reaction to those events from various segments of the community. There were several different aspects to our engagement effort.2 We conducted one public, Community Listening Session open to all participants on May 6, 2021.3 A number of public speakers shared their views and observations in sessions that were live streamed on the City of San José’s YouTube channel and recorded. Some participated in our virtual polls. City personnel provided invaluable technical support and other facilitation for these efforts; we are grateful for their assistance. We also participated in direct conversations with several engaged community members who were connected to the events of this period in various ways. These external stakeholders include those with connections to different activist organizations and individuals who attended the protests, either as direct participants or onlookers. 2 Due to COVID-19 protocols, this and all other public engagement efforts were conducted virtually – either via Zoom or telephone. 3 A recording of the public Listening Session may be viewed on the City of San José’s YouTube channel at the following link: https://www.youtube.com/watch?v=g76yLEKDrzs
Page| 9
We understand that this project – focused on specific events in the summer of 2020 – is just one component of the City’s broader response to calls for police reform. There are important conversations taking place in other forums, including a Use of Force Review, a review of 21st Century policing best practices, increased involvement by the City’s Independent Police Authority in internal investigations, consideration by the City’s Charter Review Commission of fundamental changes to the way independent oversight is performed, and a community-based process to reimagine public safety that has had some initial challenges but is persevering in efforts to reestablish trust and gain input on the path forward. Through our engagement with the San José community, we gained insight into the public’s view on the police response to the demonstrations of last summer, but we also learned about SJPD history, the status of police community relations, and recent reform efforts, all of which was important to our broader understanding. We are grateful to those who shared their experiences, views, and insights, and appreciate the value of the ongoing dialogue on these important issues.
10 | Page
SJPD Preliminary After Action Report: General Impressions This project differs in significant ways from comparable ones that we undertook in the aftermath of last summer’s protests. One way is that this review is only one element in a wide-ranging response by the City – not only to the demonstrations themselves, but also to the underlying conditions and frustrations that animated the movement in the first place. Another is that, months before we became involved in the process, SJPD had itself completed a “Preliminary After Action Report” (“AAR”) that was completed by early September of 2020 and provided a detailed account of key events. This document, though expressly limited in its scope, was a timely, constructive component in the City’s desire for information and insight. We found it to be an extremely valuable resource as we began our own evaluation. It was useful as both a reference and as a starting point for productive conversations with Department members and other City officials. And we cite it repeatedly within the pages that follow. Because its footprint was an influential one, we take this opportunity to share our overall sense of its strengths and limitations. To be clear, the Department itself was overt in framing its intentions for the document. It provided an introductory memorandum that placed the AAR into a larger context of City initiatives, including this independent project and the multi-pronged “Police Reforms Work Plan.” The articulated goal was straightforward: “The Police Department’s preliminary After-Action Report will provide the City Council and community a detailed understanding of the public protests, civil unrest, and law enforcement response from May 29 – June 7, 2020, from the Police Department’s operational perspective, including a review of command, policy, training, staffing, and tactical issues that affected police actions.” [Emphasis added.]
Page| 11
The categorization of the AAR as “Preliminary” in the title is reflective of the SJPD approach. The Department acknowledged in the same introductory memo that a broader assessment was in the works and would incorporate outside experiences and vantage points; it pledged to incorporate any of these additional reviews into its own ongoing process of adjustment and improvement based on lessons learned. The strengths of the Department’s AAR certainly include its timeliness, clarity, issue-spotting, detail, and acknowledgement of issues. There is a wealth of relevant information related to policy, procedure, and equipment. It provides an excellent overview of key events from what it describes as the “operational period” of May 29 to June 7. And it features Findings and Recommendations that are well-considered. Indeed, to the extent the numerous recommendations are implemented (as many have been already), they will leave the Department better prepared to face future similar challenges. SJPD’s production of such a lengthy, informative document in less than three months is a credit to the responsible personnel and to the agency’s leadership, which was obviously attuned to public concern and therefore willing to prioritize completion. Moreover, the AAR provides important perspectives regarding the challenges faced by the SJPD over the relevant dates. The AAR accurately describes the difficulties and complexities of policing incidents of civil disorder. It asserts an appropriate regard for First Amendment expression and the importance of facilitating peaceful protest, and of the extent and severity of the unrest that complicated the landscape of the demonstrations in various ways. To its credit, the AAR concedes SJPD’s deficiencies in training and experience that undoubtedly contributed to the problems they encountered. The primary emphasis of the AAR is on the challenges and difficulties faced by SJPD. The AAR also catalogues the perceived needs of the Department relating to personnel, training, and equipment. Many, if not all,4 of these, identified needs resonated with us as legitimate challenges that merit remediation to the extent possible (given budgetary or other constraints). 4 For example, we stop short of agreeing with the AAR’s assertion that the lack of a departmental fixed-wing aircraft was a significant obstacle to effective performance.
12 | Page
There were, however, flip sides to the speed with which SJPD produced its report. One that we identified was an occasional problem with accuracy in data – as with the statistical summaries of both arrests and uses of force. We discuss the particulars of this issue in more detail in the relevant sections of our own Report. Importantly, and as we explain, the inconsistencies we identified did not cut in one direction or another (for example, so as to be more favorable to SJPD); moreover, SJPD personnel were candid, helpful and reflective during our review as we sought further clarification. The larger issue, and perhaps the more substantial one, is that the Department had yet to formally review or reach determinations on particular force deployments or other actions that generated public notoriety at the time of the protests.5 To the agency’s credit, it did catalogue several of these specific incidents chronologically (and even included links to relevant videos available on the internet). And it is in keeping with the standard parameters of an AAR that conducting investigations of individual officer accountability issues, such as uses of force, are not within the scope. At the same time, this left a considerable gap in the overall reckoning with the Department’s performance, and we hope that will be bridged for the public in other future contexts. We also remain unsure about the extent to which a supervisory review process, known in the Department as the “Command Review Process,” has assessed specific force deployments to determine compliance with SJPD policy (and/or identify any non-disciplinary interventions that might be needed).6 Again, while this was understandably not the focus of last 5 It should be noted that investigations that involve potential violations of policy – and possible disciplinary consequences for involved personnel – could routinely take several months to complete, thus extending that process beyond the compressed time frame in which the Department produced its AAR. Our larger concern, however, is the relevant accountability process had still not been completed when we were finalizing our own report. This is because the Department has chosen to defer its decision-making during the pendency of related civil litigation. We discuss this – and our concerns about it – below. 6 As we discuss later in this Report, the Department stated that it was unable to complete its usual Command Review Process due to the complexities of the protest context. In the future, SJPD should develop a way to complete its supervisory review process, even in the most complex situations.
Page| 13
summer’s AAR, our overall sense of the thoroughness and effectiveness of the SJPD internal assessment is missing that key component. Meanwhile, we also identified a number of significant – and largely unanswered – questions that presumably did fall within the scope of the AAR and its articulated objectives. These include the following: • The AAR concedes that there was a lack of sufficient command personnel during the first two days. Why was that? • Was there a lack of on-scene oversight or control in the deployment of impact munitions and chemical agents? • Why was there a lack of reporting with respect to use of force? • Why was there a lack of controls regarding tracking the total deployment of less lethal munitions and gas as well as the breakdown of deployment use by individual officer? We also noted that the AAR reports that there were no patrol commanders in the field for the first two days, but does not adequately explain why this was the case. Certainly, the AAR does emphasize the downsizing of the Department over the years, but the specific deployment decision not to have patrol commanders in the field initially is not explained. And it is unclear from the AAR whether there were additional supervisors available to respond but were not called in for deployment. As we discuss in the body of our Report, that staffing shortfall was as impactful as it was surprising. Additional issues about which we sought more information, and discuss below, included the following: • The AAR notes that sergeants found themselves making independent decisions that were, at times, in conflict with each other. While the issue was creditably reported, there was insufficient explanation for why there was a lack of unity of command on the first two days of the protest activity. • The AAR notes that, at times, approval was granted to line personnel to employ impact munitions and chemical agents, but there was no discussion about what guidelines, if any, were provided to those individuals regarding how to deploy them. Nor was there any
14 | Page
discussion about whether this was a sound decision and whether such approval would be granted in the future. • The AAR talked globally about the overall shortages in SJPD being chronically understaffed but did not sufficiently discuss the possible alternative strategies (we noted that SJPD both had and used alternative emergency deployment schedules such as a 12/12 schedule, but this was not explicitly discussed in the AAR). It would have been helpful to learn whether such schedules were available and if so, why they were not deployed to boost the Department’s internal response on the initial days. • The AAR notes that impact munitions were only to be used to target violent suspects. Yet there is no explanation in the document regarding how members of the media or other non-violent members of the crowd were struck by them.7 • There is a reference in the AAR between officers’ “reported injuries” and “recorded (survey) injuries”, yet the AAR does not make the breakdown and divergence clear. • While the AAR discusses the deployment of skirmish lines, it fails to sufficiently discuss the rationale for their creation. In this situation, where the main grievance of the protest activity is police brutality, the presence of such lines only creates a “target” and potential for escalation of tensions between police and protestors. • The AAR references a policy section on the use of 37mm impact munitions for crowd control that was subsequently rescinded. However, the AAR fails to sufficiently discuss whether policy contributed to the first day’s less lethal deployment decisions – many of which were controversial – or to discuss the subsequent policy changes. To reiterate, there were many commendable aspects of the Department’s AAR, and explanations for some of the missing elements that we pursue further in our own review. We recognize that our Report benefits from the passage of time, 7 The Department did indicate within the AAR that these cases are being investigated by Internal Affairs, and information about potential possible violations will eventually emerge from that process. See Footnote 5, above.
Page| 15
the opportunity to incorporate a range of perspectives, and – not least – the good work that went into the Department’s internal accounting last summer.
16 | Page
Overview of Operational Timeline The “operational period” of the unrest in San José unfolded over the course of several days, with several different facets. In this section of our own Report, we provide a general summary of the key events as a frame of reference for the later analysis of SJPD’s response.
May 29: From Peaceful to Riotous On the afternoon of May 28, the Department learned of a planned event, the “George Floyd Solidarity Action,” scheduled to happen on Friday, May 29 at 2:00 PM in front of City Hall. On the morning of May 29, patrol personnel created an Operations Plan in preparation for what they believed would be an uneventful, peaceful protest at City Hall. The plan identified two Mobile Field Force (MFF) units of 10 officers led by one sergeant to “stand-by” for the event. A Mobile Field Force unit is typically made up of specially trained, specially equipped patrol officers under the direction of a patrol supervisor and patrol command. Participation in MFF is a collateral duty, and these units are deployed on the rare occasion of large-scale civil disorder or crowd management.8 Special Operations – the Department’s division that encompasses several unique support units – also prepared: they established two “Strike Teams,” teams of four to six Special Operations officers led by a Special Operations sergeant, to stand by.9 SJPD’s mission was to allow for peaceful demonstration of First Amendment rights while also ensuring the safe movement of the crowd and reducing the 8 A concern identified by the Operations Plan was that this event might span two shifts: dayshift, who would end their workday at 4:00 PM, and swing shift, who would begin their day at 3:00 PM. To remedy this, the dayshift patrol supervisor designated two additional MFF units made up of swing shift officers. However, as we detail later, the mid-incident shift change had implications for deployment numbers. 9 Teams are typically transported via a shuttle bus and respond together. A Strike Team is typically made up of Special Operations officers under the direction of a Special Operations Tactical Commander. These teams are smaller, more mobile and deploy using patrol cars.
Page| 17
impact on residents and businesses. Command10 staff reported that City and SJPD leadership instructed deployed officers to manage the crowd’s safety and prioritize protest rights over policing minor misdemeanor acts, such as tagging or jaywalking. As planned, the event began around 2:00 PM at City Hall and, according to officers on the scene, was largely peaceful. At approximately 3:00 PM, the crowd began to move from City Hall eastbound on Santa Clara Street, walking in all lanes of traffic. A Special Operations supervisor followed the crowd in an unmarked police vehicle to provide real-time intelligence, but no additional units were deployed. Meanwhile, the day shift supervisor observed the dynamics at City Hall and perceived a concerning level of antagonism toward law enforcement. She determined that activating the Operations Plan was appropriate. At around 2:45 PM, when she observed the crowd’s movement eastbound, she began to set up a Command Post at a local parking lot and instructed the two designated day shift MFF units that were on stand-by to report to that location. Shortly after 3:00 PM, the crowd, now approximately 250 people, walked onto the 101 Highway from Santa Clara Street and obstructed traffic in both directions. The California Highway Patrol (CHP) requested that SJPD units assist with closing the freeway on- and off-ramps at Santa Clara Street. Various SJPD units, including the day shift MFF units and at least one of the Special Operations Strike Teams, responded to this location to form lines across the freeway on and off-ramps. According to several SJPD personnel and City leadership, this was a turning point on May 29. Previously tasked with simply maintaining order while keeping a low profile, SJPD was now being asked to manage the crowd at the freeway. And some of the crowd began to act in an overtly violent manner, 10 Throughout this Report, we use the term “command” or “command staff” to distinguish between Department leadership and the line-level officers who comprised the majority of SJPD responding personnel, and were following orders as provided by these higher-ranking individuals. The command level is itself comprised of multiple ranks, ranging from the Chief to sergeants. All ‘command personnel’ have supervisory authority, but answer to each other (and ultimately to the Chief) within the management hierarchy.
18 | Page
vandalizing vehicles (both those of police and un-involved highway motorists) and throwing large objects at officers. Command staff discussed various deployment strategies. SJPD leadership later explained that, because they were not equipped with appropriate tools to properly manage this aggressive crowd at the freeway and were being overrun, they decided to withdraw officers from the area and instead sought to push the crowd back toward City Hall. The downtown area, they decided, was a safer space for protest activity and was also where protesters had parked their vehicles; the thought was that returning to that venue might encourage some participants to go home. In order to accomplish this goal, SJPD issued several dispersal orders from a Long-Range Acoustic Device (LRAD), a loudspeaker mounted on a police vehicle, to encourage the crowd to move back to City Hall. The crowd began to move westbound back to City Hall via Santa Clara Street. SJPD continued to issue dispersal orders from the LRAD as officers traveled alongside the crowd. Based on observations from officers monitoring the situation in the field, the day shift supervisor realized that, with the crowd now moving back to downtown, the initial staging location in the parking lot was no longer practical for officer safety. She accordingly moved it and set up a formal Command Post at San José High School. Command staff also called for a dayshift holdover at 4:06 PM, which was a request that all day shift personnel, who were near their end of watch that day, stay on duty. Meanwhile, new protesters had converged in downtown. According to some protest participants, this second crowd was made up of people who were late to the initial protest or who had watched media footage and wanted to join. The two protest groups converged, spanning from City Hall to 9th Street along Santa Clara Street. The two MFF units and Strike Force regrouped in downtown, formed a skirmish line across Santa Clara at 9th Street and continued to issue dispersal orders. But rather than clear the crowd, the presence of the skirmish line only seemed to heighten tension. Some members of the crowd approached the line for direct confrontation with officers while others set up protective barricades to throw objects at the line, ranging from water bottles to fireworks.
Page| 19
Others stood in the roadway, peaceful but in defiance of dispersal orders. Still others reported to us that they were simply trying to express their protest rights and, unaware of the violence, were caught off guard by SJPD’s crowd management tactics. Shortly after 5:00 PM, direct confrontation between the protesters and the officers on the skirmish line escalated further. As officers began pushing the crowd with 42” riot batons, some in the crowd reacted by pushing back. One protestor shoved his cellular phone into the officers’ faces as he marched past the skirmish line. When one officer slapped the phone out of the protestor’s hand, the protestor punched the officer, causing the officer to fall unconscious on the ground. In response to this escalating violence, the Special Operations commander authorized use of less lethal projectile impact weapons (PIWs). Special Operations officers deployed 37mm projectiles at the crowd. Around this time, various individuals captured video footage of a Special Operations officer shouting at protesters using profane language and appearing excited to use force on the crowd.11 The skirmish line managed to move the crowd one block while continuing to issue repeated dispersal orders. A brief time later, another protestor approached the line yelling that the officers were “racist.” Officers deployed 40mm PIWs specifically at this protestor, who continued to approach the line of officers. Officers took the protestor behind the skirmish line and arrested him. This scene seemingly intensified the crowd’s anger as the crowd observed the arrest. Officers used various less lethal force techniques, from the 42” baton to rounds of less lethal impact projectiles, to move the crowd toward City Hall at 5th and Santa Clara, reaching this area around 6:15 PM. Some in the crowd used dumpsters as barricades, hiding behind them to throw objects at the skirmish line, or started fires in them. Officers reported that they attempted to isolate these individuals and target them with impact projectiles. In one instance, officers struck a protester in the genitals; this individual was standing 11 We discuss this incident and other officer communication issues in greater detail later in this Report.
20 | Page
between the skirmish line and a protestor that was throwing items at officers from behind one such dumpster barricade, as if to provide additional cover.12 Meanwhile, two swing shift patrol lieutenants had reported for duty to the Command Post at San José High School. Seeing that the event had escalated well beyond the original Operations Plan for a peaceful protest, they moved the Command Post to the much larger parking lot of the SAP Center. From here, these supervisors managed deployment and equipment. When swing shift officers reported for duty, they were assigned to teams of one sergeant plus 10 officers and sent to the field. The idea was to have these teams liaison with a field lieutenant and be specifically deployed to a mission.13 But even with more resources, the skirmish lines were too stretched out and could not adequately contain the crowd. At 6:21 PM, after consulting with higher level command, Command personnel called a “Code 30,” an emergency call that requested every available officer to respond to the scene. This call also activated a Mutual Aid response from neighboring law enforcement jurisdictions throughout Santa Clara County. The violence at City Hall escalated further. Some protesters looted a construction site and began to throw construction materials at the officers on the skirmish line. With the assistance of the AIR3 helicopter’s overheard surveillance, officers identified specific individuals that were throwing objects or inciting the crowd to violence. But as they later reported, SJPD did not have the resources to strategically arrest them. In response to the deteriorating situation and at the orders of an Assistant Chief, the Special Operations teams deployed OC gas at approximately 6:30 PM from handheld canisters.14 This reportedly had little effect on the crowd’s 12 We discuss this and other questionable uses of the impact projectiles later in this Report. 13 As we discuss in detail later in this Report, command in the field was significantly lacking. As a result, most officers, many of whom lacked training and experience in crowd management, were left without adequate command. 14 We provide a detailed description of each type of force used in the section “Force Deployment: Types and Uses,” below.
Page| 21
activity; worse, it blew back toward the officers on the line, not all of whom were in immediate possession of their Department-issued gas masks.15 Protesters drove a vehicle into the roadway and used it as a barricade. From behind the vehicle, various individuals threw items at the officers. Officers continued to use less lethal projectiles on the crowd, including flashbang devices. Eventually, Special Operations officers acquired OC gas rounds that could be fired at greater lengths. They used these rounds to fire OC gas beyond the front-line protesters and reach the aggressive individuals hiding behind barricades. But this OC gas, like the gas deployed earlier, was also largely ineffective at dispersing the crowd. As extensively documented in the Department’s own AAR, the protest activity and police response continued late into the evening, with various violent confrontations at and around City Hall and in Cesar Chavez Park. The Department continued to issue dispersal orders from police vehicle loudspeakers and the AIR3 helicopter throughout the evening. Around 8:00 PM, SJPD received reports of an officer-involved shooting involving a Santa Clara County Sheriff’s Deputy. SJPD took control of the investigation, which created a further strain on resources. Around this time, officers in the field reported that their inventory of less lethal munitions was running low. Some officers were tasked with finding more munitions from the Department’s training range, Central Supply, and other locations in the City, which they brought back to the Command Post and provided to officers in the field as needed. At approximately 10:00 PM, the skirmish line began moving the crowd southbound on 4th Street. In response to this push, some in the crowd once again began throwing objects at the line. According to SJPD personnel, Department leadership determined that justification existed for additional assertive measures. It instructed Special Operations use CS gas to disperse the crowd. When some in this same crowd began lighting fires, officers again deployed impact projectile rounds. 15 SJPD reports that it has subsequently addressed the issue with officer gas masks.
22 | Page
Officers were finally able to clear the downtown core just before midnight. Before leaving for the evening, Special Operations personnel attempted to order munitions to replenish the Department’s expended supply. But the contracted munitions supply company reported that their supply was also low due to extremely high demands nationwide, and they were unable to ship new munitions to San José. When they finally did locate munitions supplies from a local vendor, officers made an “exigent purchase” outside of the normal City purchasing policy. In this way, they were able to restock the supply of less lethal munitions for the coming day.
May 30: Unrest Continues On the morning of May 30, SJPD command regrouped to plan for the day ahead. Command personnel held an informal telephonic briefing where they discussed mission and tactics. Special Operations commanders created an Operation Plan.16 All swing shift officers, now working a 12-hour shift, would be diverted from regular patrol activity to support Special Operations. SJPD maintained the Command Post at the SAP Center; all swing shift officers were instructed to report to that location for duty. According to some command personnel, the basic leadership dynamics improved considerably from May 30 onward. SJPD command began hosting a general briefing at SAP Center that reflected a more organized, coordinated approach. This included briefings on recognized deficiencies that the Department sought to address, refreshers or updates on Department policy and code of conduct, and reminders to officers to report uses of force.17 After the general briefing, SJPD held a supervisor briefing to review command and control. Command reported that they were receiving hourly updates from the SJPD Intelligence Unit regarding possible protest activity that day; SJPD personnel 16 By May 30, SJPD personnel reported, the event had officially shifted from patrol leadership to Special Operations leadership, with patrol playing a support role to Special Operations. As such, the Special Operations commander was the Incident Commander and, with the support of patrol supervisors, drafted the Operations Plan on this day and going forward. 17 Other personnel reported that these briefings started on May 31, and others recall them staring on June 1.
Page| 23
reported to us that, from searching social media, Intelligence learned that activist groups were training and planning as well, coordinating drop-offs of persons and potential weapons for use that day. At approximately 4:00 PM, a crowd assembled at City Hall. Officers deployed a skirmish line along Santa Clara Street and 5th Street, spanning the intersection. The intention of this line, reported the Department, was to prevent the crowd from taking over roadways as they had done on May 29. While the majority of the crowd remained peaceful, AIR3 and officers on the ground first reported specific incidents of violence at approximately 5:00 PM. One individual was arrested for throwing bottles at officers; additional individuals began to throw bottles at the skirmish line. Without declaring an unlawful assembly and issuing a formal dispersal order, officers targeted those aggressive or violent individuals with 40mm PIWs. Some of the crowd left the area. By 7:00 PM, SJPD reported that the majority of the crowd at City Hall had become violent. At 7:18 PM, SJPD issued a dispersal order and Special Operations teams deployed to City Hall. At least some of this crowd moved from City Hall to other areas in downtown San José. SJPD deployed Strike Teams to combat their specific activities. By 10:00 PM, the crowd at City Hall had escalated their attacks on officers, throwing rocks and bottles at the skirmish lines. SJPD responded again with PIWs. At 10:45 PM, Special Operations deployed CS gas to disperse the crowd. And while some left, some remained and continued to throw items at officers. Around 11:45 PM, officers observed objects being thrown at officers from an apartment window overhead on Santa Clara Street. They deployed several high-velocity rounds at the exterior walls of the apartment building. By 1:00 AM, the crowd had largely dispersed.
24 | Page
May 31: Tactical Planning and Curfew Command personnel reported that by the third day of activity, SJPD’s tactical planning and communications were carefully planned instead of reactive. Their revised plan was to only respond if police presence was needed to control criminal behavior or for public safety, versus reacting to all protest activity as it unfolded. In the early morning hours, City leadership began to draft an emergency curfew order at the request of the then-Chief, who strongly believed that having a curfew would assist with restoring order to the downtown core in the late evening hours. According to several involved in the process, drafting the order was a time-consuming process that took longer than expected, resulting in limited time to communicate the order to the public and to officers prior to enforcement.18 While City leadership was drafting the curfew, protesters began to converge again around City Hall. By 3:00 PM, the crowd was estimated to be over 300 people, mostly peacefully congregating. Around 4:00 PM, the crowd grew to an estimated 500. SJPD deployed a skirmish line on Santa Clara Street; in the AAR, they described their intent in doing so observing and managing the crowd.19 Officers reported that the crowd was mostly peaceful; some members, they noted, were making eight loops around City Hall to commemorate Mr. Floyd’s murder over eight minutes. In aerial footage from AIR3, we observed that, at various points, individuals would approach the line and verbally taunt officers; at least one officer reported that, by 4:30 PM, the crowd had become “hostile,” requiring him and others to use their batons to 18 We discuss this and other curfew-related issues later in the Report. 19 This skirmish line deployment does not align with the stated mission of the day to remain out of sight until a police response was needed. Command personnel did not provide a rationale for the change in plans except to note that the crowd size was increasing.
Page| 25
push away individuals. On the other hand, others reported that the crowd was largely peaceful.20 Officers did encounter one individual whom they determined to be aggressive toward police officers at the City Hall fountain area; this individual, whom they believed to be intoxicated, picked up a glass bottle and threw it at the officers on the line. In response, several officers deployed their 40mm launchers and arrested the individual. At the same time, a group of protesters, some elderly, had converged in the City Hall courtyard area to pray and listen to speakers. Per the leader of this group, who later spoke with us, members heard the sudden use of PIWs being deployed but claimed not to have heard prior warnings. Fearing that officers were firing indiscriminately into the crowd, they dispersed from the area. Eventually, around 7:00 PM, the skirmish lines were called back to the SAP Center and changed tactics from crowd management to enforcement. SJPD deployed Strike Teams, or small teams of officers sent to specific calls for service, such as looting, vandalism, or fires throughout the downtown area. When the curfew took effect at 8:30 PM, Strike Teams began to issue dispersal orders and, later, arrest the individuals who remained in the area. Some officers deployed additional PIWs during the course of the evening, specifically during the attempted robbery of a bank. One fleeing individual reportedly ran into the roadway in front of an SJPD Motor Unit, who struck the individual with his motorcycle. By 9:45 PM, SJPD had cleared the downtown area. Officers were sent home for the day. 20 The difference in these versions is a presumably a matter of officer position and perception – a reminder of the variations within a large crowd environment that can produce sincere discrepancies in people’s experience and subsequent descriptions. This was true of the police but also, of course, of members of the public who participated.
26 | Page
June 1 to 4: Evolving Tactics & Changing Crowd SJPD leadership reported that, after three days of protests and placing officers in the field in skirmish lines with little positive impact, they decided to “draw lines in the sand” to determine what behavior would require a police response. SJPD, in collaboration with City leadership, decided that they would “let people get away with” misdemeanor activity such as small fires and vandalism.21 On June 1, SJPD observed the peaceful protest at City Hall without deploying until approximately 5:30 PM. At that time, they reported, the crowd began moving into the roadway. SJPD personnel reported that they were concerned about possible accidents, so they established a skirmish line on Santa Clara as they had done the previous day. This approach had similar results as on May 31: the crowd turned its anger onto the deployed officers and, by 6:30 PM some individuals began throwing objects and yelling at the officers. Another group, gathered at Cesar Chavez Park, began moving toward Motor Unit officers. In response, the Motor Unit officers moved to a new location, out of sight of the protest, and this crowd returned to City Hall, where they largely remained peaceful. When the curfew went into effect at 8:30 PM, SJPD began to issue dispersal orders and the majority of the crowd left the area. Some remained and threw fireworks or vandalized property. Some of these individuals were arrested. By 10:00 PM, SJPD had cleared the downtown core. The daytime events on June 2, 3, and 4 were largely peaceful and SJPD did not deploy skirmish lines during the daytime hours. SJPD personnel theorized that there were several possible reasons for this shift: first, the daytime crowd was different, and possibly more moderate, during these weekday events; second, the crowd typically marched throughout downtown instead of staying 21 One member of City leadership later shared with us the view that this approach, while helping control the crowd’s anger and reducing civil unrest, was understandably unpopular among business owners, who felt targeted and unsupported. This dynamic reflects the tension that was shared by multiple jurisdictions, where increased latitude toward lower-level disruptive behavior was helpful – but not without cost.
Page| 27
static, preventing pockets of individuals from infiltrating the crowd; third, they had changed tactics, were receiving more detailed briefings, and were better staffed and organized. The Department, they noted, had found a balance between enforcement, protecting protest rights, and avoiding approaches that would make officers the subject of the crowd’s anger. Conversely, the evening hours presented challenges due to “sideshow” events, with individuals driving recklessly, performing stunts, and “surfing” on moving vehicles. This was obviously a different manifestation of the unrest and merited a different intervention from the Department. SJPD deployed in skirmish lines and/or Strike Teams, using the 8:30 PM curfew order to disperse or arrest offenders. On the evening of June 2, an officer-involved shooting by a SJPD officer occurred. This shooting happened when officers attempted to apprehend the suspect of a hit-and-run accident that had occurred during a side show event – a significant incident but not one that was directly related to the protest movement. On June 4, after discussion with City leadership and the community, the City lifted the 8:30 PM curfew. That day’s protest activity ended at approximately 8:00 PM without intervention.
June 5 - 7: Mediation Over Escalation The return of the weekend saw the return of larger protest crowds. On Friday, June 5, an estimated 2,000 people gathered at City Hall and marched to the SJPD headquarters. Despite this large crowd and their location, SJPD continued to employ the tactics that had worked well over the previous days: they staged teams out of sight, allowing the protest to happen without intervening and monitoring activity from the field and the Command Post. The crowd eventually dispersed, with some moving to City Hall and others to Cesar Chavez Park. Around 9:20 PM, the SJPD helicopter, AIR3, observed an individual in the crowd at City Hall shining a laser pointer at the helicopter pilots. In coordination with AIR3’s direction over the radio, a team of officers deployed to arrest the individual. But the mere presence of the officers suddenly “removing” someone from the protest area angered some in the crowd, who threw objects at the arrest team. The officers quickly took the arrested individual to a police vehicle. The team then entered City Hall to remove
28 | Page
themselves from the angry crowd. Other officers deployed to assist this arrest team with the crowd’s aggression, and individuals in the crowd threw objects at them and slashed a police vehicle’s tires. Meanwhile, the Fire Department reported that individuals were jumping onto fire trucks. Others in the crowd were observed passing out wooden boards, which SJPD believed would be used as shields against PIWs. Around 10:30 PM, some of the crowd started a “sideshow” of driving activity at City Hall. SJPD issued dispersal orders from the AIR3 helicopter, which seemingly made the crowd more “rowdy” in the later descriptions of SJPD personnel. The side show continued until 11:30 PM; some individuals dispersed, but a number of others remained. SJPD formed a skirmish line at the intersection of Santa Clara Street and 4th Street with the goal of clearing the downtown core. But, as the evening went on, SJPD personnel reported that as soon as officers arrived at any location to clear the area, the crowd “turned on them,” and officers became the targets of violence. During this time, officers deployed flashbangs, devices that create a loud explosive sound and bright light, and arrested several individuals. Special Operations was prepared to deploy CS gas. Around this time, a group of individuals approached the skirmish line. An involved commanding officer reported seeing what he called a “window of opportunity.” He and another officer walked toward the group, made up of young males, and began to speak with them. The commanding officer engaged in conversation and negotiation for several minutes, even taking a knee with the protesters, as he asked them to leave the area peacefully.22 Eventually, the group agreed to disperse, and the downtown core was cleared. 22 This event was recorded and streamed on various social and traditional media outlets. According to some in the Department, this event was controversial for several reasons. First, the apparent “leader” of the group, a young Black male who led negotiations with the commanding officer, had previously attempted to “bait” officers into confrontations by calling in 9-1-1 calls of a “man with a gun” and describing himself. While this allegation bears on the advisability of engaging with the person, the commanding officer was unaware of it at the time. Second, some were upset that the commanding officer had taken a knee, a move that they
Page| 29
On June 6 and 7, protesters engaged in smaller protests and marches throughout the City. SJPD did not deploy any crowd management teams on these days. perceived indicated solidarity with the protest. This issue proved to be a fraught one across multiple jurisdictions, and we discuss it in more detail below. Nonetheless, we also note that these were extraordinary circumstances, and the value of police connecting with disaffected members of the community was extraordinarily acute. “Case by case” exercises of discretion by law enforcement personnel (as opposed to blanket approaches of one kind or another) strike us as being the best approach to the “take a knee” scenario. This specific example appears to have been a commendable effort by the involved SJPD member.
30 | Page
Timeline Part of our Scope of Work for this evaluation was to create a detailed timeline. We selected several significant events using the documentary and digital evidence sources listed in our Methodology section as well as information gathered in our numerous interviews.23 Because the Department’s AAR provided a very detailed and, in our review, accurate timeline of daily events, we chose to present a different visual. On the following pages, we present timelines by day that depict the actions of the crowd in the blue “bubbles” on the left and the actions, or reactions, of SJPD in the bubbles on the right. The intention is to show that, in the early days of protest, the actions of SJPD were largely driven by the actions of the crowd; as we discuss throughout this Report, SJPD was in a “reactive” mode during these early days. For example, on May 29 at 6:40 PM, some in the crowd created a barricade in the roadway and began throwing objects at the skirmish line; SJPD reacted with projectile impact weapons (PIWs) and flashbangs (called noise-flash devices, or “NFDs”). However, some SJPD actions were not driven by the crowd; see, for example, the May 31 choice to deploy skirmish lines at 4:00 PM, despite a largely peaceful crowd at City Hall. The deployment of these lines, reported SJPD personnel later, only seemed to incite the crowd, resulting in objects being thrown at police personnel and necessitating uses of force. And, once SJPD changed their tactics, the crowd’s actions remained, for the most part, peaceful, despite large groups of protesters; see, for example, the June 2 Timeline when protest activity occurred without police response. 23 Note that June 4 was not included in these daily Timelines because SJPD did not have any notable deployments on that date.
Page| 31
May 29 CROWD ACTIVITY
CROWD ACTIVITY
32 | Page
May 29 (cont.)
CROWD ACTIVITY SJPD ACTIVITY
Page| 33
May 30
CROWD ACTIVITY
34 | Page
May 31
CROWD ACTIVITY
Page| 35
June 1
CROWD ACTIVITY
36 | Page
June 2
CROWD ACTIVITY
Page| 37
June 3
CROWD ACTIVITY
38 | Page
June 5
CROWD ACTIVITY
Page| 39
Assessment & Analysis Command and Control: Leadership Void Law enforcement agencies speak of “Command and Control” in reference to both clear, organized decision-making and the coherent field responses that flow from it in the context of a particular operation. Many of the more significant issues we identified within the SJPD response stemmed from shortfalls in the Department’s ability to establish appropriate command and control structures in the first days of the unrest. There were several reasons for this as we discuss below. Operational Planning First among these reasons was the extent to which the events of May 29 clearly took the Department by surprise. Department personnel were aware on some level that the protests in response to George Floyd’s murder were arriving in San José in the form of an event that Friday. Intelligence concerning widespread civil disobedience in other cities was clear, but the prevailing opinion was that nothing of an extreme nature would occur in San José. Several SJPD members with whom we spoke were consistent in their recollections of the prevailing mindset, which revolved around a distinction between the egregious officer misconduct that caused Floyd’s death and the Department’s own sense of its performance and standing in the community. As a result of this underestimation, a “pro forma” operations plan for responding to the expected protest activity was completed. A patrol lieutenant was pre-designated as the incident commander – the person responsible for coordinating the Department’s resources and managing the event from a central location. However, no substantive preparations or briefings of personnel appear to have taken place prior to the event.
40 | Page
Span of Control: Lieutenants and Command Supervision Second were issues with the number of line officers per supervising officer, commonly referred to as “span of control.” We learned that, due to staffing shortages, the Department often operates with just two lieutenants on duty during the evening hours, rather than the four that are normally called for in the Department’s organizational chart. So, even before this critical incident hit, the “span of control” (number of subordinates for which a supervisor is responsible) appeared to be too large to constitute effective levels of supervision for an agency with nearly 100 officers per shift to police over a million residents. Predictably, during a rapidly developing critical incident, this span of control quickly proved inadequate to the supervisory needs of the situation. The AAR makes reference to a significant shortage of command and management level personnel on scene during the first two days of protest. This was among the more puzzling aspects of the overall SJPD response: while the intensity of May 29 was not foreseen for many valid reasons, we remain unsure as to why an “all hands on deck” mentality did not prevail by May 30. Many personnel felt that there was no unity of command or clear understanding of the mission on the first day. That sentiment was echoed by several participants during our interviews; SJPD personnel reported that the operation was disjointed and disoriented and said that it was primarily run by sergeants. Further exacerbating these “span of control” deficiencies was the fact that Mobile Field Force (MFF) platoons, or groups of officers, were mostly missing a lieutenant/platoon leader.24 In field operations like May 29 and 30, a MFF platoon is typically comprised of one lieutenant / platoon leader and four squads, each comprised of one sergeant and 8-12 officers. The 24 SJPD normally subscribes to the Mobile Field Force (MFF) concept in responding to critical incidents and requests for mutual aid. This structure – including the presence of a lieutenant – aligns with modern best practice and is reflected in SJPD training materials that we reviewed.
Page| 41
recommended span of control in the field is one lieutenant to no more than 50 personnel. But, in this incident, the lieutenant role was not filled, leaving the sergeants largely in charge of a too-large span of control, as many SJPD personnel reported, and no field command staff. Reliance on Special Operations Third, we noted that the Department has routinely relied upon their well-trained and supervised Special Operations personnel to police high-profile events like protests, which, historically in San José, were largely peaceful and uneventful. This model works on a normal basis because problems are relatively few and there are sufficient Special Operations personnel to handle the mission. On May 29, however, the scope of the operation quickly overwhelmed the limited resources of Special Operations, which was then augmented by large numbers of patrol personnel who lacked experience and training at the officer and sergeant levels and had no lieutenants specifically assigned to oversee their involvement. Special Operations officers and Command personnel conveyed their sense that their unit performed effectively and with its usual cohesion; we concur. It was understandable that they navigated their responsibilities at a different level of proficiency than the broader patrol personnel; their backgrounds as members of this group were relatively tailored to the needs of the incident. But it also showed how much room for improvement existed among the patrol officers who were thrown into a challenging environment and not provided with sufficient controlled supervision or guidance, and how reliant SJPD had become on these specialized units for effective policing of this nature. Real-Time Communication: Missing the Lieutenant Link Finally, we noted that, while there was ongoing communication between a Deputy Chief, who saw himself as the incident commander, the Special Operations Captain who was in the field, and the Special Operation lieutenants who were in close contact with their regularly assigned subordinate personnel, there was a missing communication link between these Command personnel and the aforementioned patrol MFF squad sergeants. Typically,
42 | Page
this role would be filled by the patrol lieutenant, a role which, as we already detailed, was missing in this operation. Lieutenants maintain contact with the Incident Commander. They implement the Commander’s Intent, by communicating objectives and performance expectations to their subordinate personnel. In the field, they are a visible representation of accountability to the line personnel. This was an important missing component on day one of the unrest. Without patrol lieutenants, the span of control was far too great for the Special Operations lieutenants to effectively communicate with their own personnel and take command of patrol personnel. So, patrol personnel were largely left to their own devices with no clear communication from Command. This accounts for some personnel feeling that there was a defined chain of command and clear mission objectives, while others believed that sergeants were controlling the operation and making decisions independently and, at times, in conflict with each other. RECOMMENDATION 1 SJPD should maintain a roster of lieutenants for immediate call back, on-call if necessary, to fulfill the platoon leader role in response to critical incidents.
Training Concerns Whenever law enforcement actions generate controversy, a comprehensive administrative assessment of what occurred should feature the role of training as one component of the analysis. A lack of training can account in part, for elementary tactical decisions that proved to be ill-advised, such as the deployment of chemical agents upwind from officers, and before all officers on the line were equipped with gas masks. The main question – whether the actions of officers comported with Department training and expectations – leads in short order to others. If yes, then does that training need to be re-visited in light of the incident’s outcomes? If no, then what remedial measures are required to address individual or more wide-spread shortcomings?
Page| 43
As applied to the efforts of SJPD in late May and early June of 2020, these questions were overshadowed by whether useful training on the distinctive topics of large-scale crowd management had occurred in the first place. The answer, as acknowledged within the AAR and affirmed in our subsequent discussions with Department personnel, was that it had not. The lack of sufficient time and money for training on all potentially useful subjects is an oft-repeated refrain throughout policing – and one that we largely perceive as legitimate. As much as training is a key component in ensuring a well-organized, disciplined response to acts of civil disorder, it is obviously difficult to decide how to prioritize. Because serious incidents of civil disorder are infrequent and training for such events is not mandated, it is often placed on the back burner. SJPD was far from alone in its lack of relevant experience and prior preparation for the particular challenges that arose. Nevertheless, the scope of any civil disturbance implicates the importance of a coherent, disciplined response at all levels of a police organization. The events of last May and June are also a reminder of how important the effective handling of these matters is to public confidence, and will ideally incentivize SJPD to follow through on its intentions to address training needs in this arena. Here we explore the particulars as it related to San José and discuss options for the future. Training Prior to May 29, 2020 The AAR rightfully points to Department-wide training deficiencies in the realm of crowd control and civil disorder. A review of training documents provided by the Department revealed that apart from a few hours of introductory training to recruit officers in the Basic Academy, there has been little substantive in-service training in recent years focusing on response to civil disorder. The shortage of civil disorder training impacted the response of not only line officers, but also supervisors at different rank levels and the overall “command and control” with which SJPD responded. At command levels, Incident Command System (“ICS”) training is an important component contributing to clear organization and chain of command during critical incidents. Its strength lies in part on its general applicability to a range of situations, and we certainly endorse the Department’s stated emphasis on
44 | Page
ensuring that its supervisors are well-versed in its core principles. However, as useful as it proved to be once it was fully utilized in the latter days of the protest period, it is best understood as a foundational concept – and not a substitute for targeted training on law, tactics and strategies specific to civil disorder. In 2016, 16 SJPD managers (Lieutenants and Captains) attended a three-day course on Crowd Management Strategies for Command. Unfortunately, only one of those attending this course had any involvement in the events that unfolded on May 29, and by that time he had ascended to the position of Deputy Chief. Most supervisors and managers on scene had little or no training in this realm. When one lieutenant was queried as to when their last substantive civil disorder training took place, the answer was “never.” At the line level, the vast majority of officers and sergeants had not received any civil disturbance training apart from a very rudimentary introduction during the Basic Academy. Many officers had completed qualification on less-lethal weapons such as the 37 and 40mm less-lethal weapons systems, but such qualifications are normally focused on proper physical manipulation of the weapon and its use against a single, non-compliant subject in a patrol environment. Simple qualifications on how to operate the weapons do not equate to scenario-based training that is geared toward instilling competence and discipline with the weapons system and the specialty munitions that are employed during a chaotic event of civil disorder. A 37mm round that emits five rubber baton projectiles is indiscriminate by nature, and it is a far cry from a single projectile used in a patrol setting to incapacitate a target individual who poses a cognizable threat. When and how to use the weapon in a civil disorder scenario and the importance of accurately documenting its use is a function of proper training that was not presented to San José officers. It appears that civil disorder training for SJPD personnel has been sparse for a number of years, and the training which did take place was reserved for a relatively small segment of the organization. And the Mobile Field Force training materials we reviewed25 left us with the impression that the little training that did take place at the line level centered on enforcement measures and physical reaction to confrontation. The written materials convey the sense 25 We did not observe this training and do not know what messages were conveyed by instructors during its delivery; our impressions are based strictly on the written materials provided by the Department.
Page| 45
that violent confrontation is inevitable, with no mention of the need for self-discipline and the unit cohesion that is so critical during incidents of civil disorder. We recommend that MFF training specifically address these issues and that future training incorporate a message from the Chief regarding Department expectations for conduct during these events. Further, training specific to law, tactics and strategies for use in a First Amendment context should be provided and documented on a recurring basis. It should be attended by all personnel that may be tasked with responding to, supervising or commanding such events. The training should feature a scenario-based component, and it should include performance expectations from the Chief or senior command personnel. Specific training on “crowd control in a protest context” is an essential component at all levels of the organization to ensure a disciplined response to such incidents. The events of last May and June are a reminder of how important the effective handling of these matters is to public confidence and will ideally incentivize SJPD to follow through on its intentions to address training needs in this arena. RECOMMENDATION 2 SJPD should ensure that all of its lieutenants are well-versed in the fundamentals of the Incident Command System and would be capable of serving as the Incident Commander for a given operation if necessary. RECOMMENDATION 3 SJPD should revise its Mobile Field Force training to emphasize the need for self-discipline and unit cohesion and to include a message from the Chief regarding the Department’s performance expectations. RECOMMENDATION 4 SJPD should provide public updates regarding its progress in providing the targeted, topic-specific training on Mobile Field Force and other crowd management concepts that it cited in its After Action Report.
46 | Page
Dispersal Orders and Unlawful Assemblies A dispersal order is an announcement given by law enforcement to two or more people who are gathered in what has been formally characterized as an “unlawful assembly.” The point is to address disruptive or dangerous behavior by requiring crowd members to leave the area so that additional enforcement action is not needed. The predicate for a legitimate invocation of dispersal orders is the existence of an unlawful assembly in the first place. In California, it is unlawful for persons to assemble for the purpose of disturbing the public and then fail to leave after being ordered to do so by the police. The acts of the crowd must be either violent or tending to incite others to violence. The situation on May 29 supported SJPD’s determination that dispersal orders were appropriate. However, what complicated the analysis in many jurisdictions was a phenomenon that applied in San José: namely, the challenge of a dynamic in which traditional peaceful protest – perhaps even among the large majority of crowd members – was accompanied by peripheral activity that was antagonistic and dangerous. Preserving the former while safely addressing or preventing the latter proved to be very difficult at times throughout the protest period. On May 29, a large part of the crowd throughout the afternoon was peacefully demonstrating, chanting and/or holding signs in protest of police or support of the Black Lives Matter movement. These individuals were not engaged in overt acts of violence and reported that they were surprised and upset that their First Amendment right to protest was thwarted by police enforcement actions. Conversely, some crowd members that day, particularly on Santa Clara Street between City Hall and 9th Street, were reportedly engaged in violent acts such as throwing construction materials, rocks and bottles at the SJPD skirmish lines. Commanding officers reported that these incidents were growing more hostile and coordinated as the intensity and size of the crowd increased. A review of the available video footage confirmed the existence of aggressive or assaultive behavior by some at that location as the afternoon hours progressed into early evening. While some individuals in the crowd were
Page| 47
engaged in peaceful protest, they were doing so alongside aggressive rioters.26 To its credit, SJPD modified its tactics in subsequent days in order to give more latitude to the protest movement. This included not responding to protestors who were involved in minor unlawful activity and even some overtly aggressive activity to balance the right of free speech with public safety. But, to our knowledge, these tactics have not been formalized for events going forward. SJPD and the City should continue to consider what set of circumstances should be required in a public protest setting prior to the declaration of an unlawful assembly. The City should engage with the community as it develops these guidelines, through direct outreach to residents, business owners, and groups most impacted by the events of Summer 2020. The resulting policies and guiding principles should be publicly announced so that City residents and stakeholders understand the “ground rules” in advance of the next protest. RECOMMENDATION 5 SJPD and the City should engage with the community as it considers the circumstances required for a public protest to be declared an unlawful assembly. The resulting guidelines should be publicized in a way that provides City residents and stakeholders a clear understanding of the circumstances under which the SJPD will declare an unlawful assembly. As for the mechanics of the orders themselves, we found that this contributed to some of the confusion and resentment that they had engendered. On May 29 around 4:30 PM, as the crowd began to move from the 101 freeway back toward downtown San José, a Special Operations officer began to issue the first of many dispersal orders from a Long-Range Acoustic Device (LRAD) mounted on a police vehicle. The order was repeated by this officer 26 Multiple SJPD representatives also described to us their observation of crowd members who might otherwise have been more orderly being incited into disruptive behavior by the actions of others – a phenomenon that the “unlawful assembly” concept recognizes and is designed to offset.
48 | Page
and others from that point forward, at ongoing intervals late into the evening from the LRAD and the Air3 helicopter. This dispersal order as heard on various body-worn camera clips was as follows: This is [officer name], a Peace Officer of the State of California and a police officer of the City of San José. I do hereby declare this an unlawful assembly. In the name of the people of the State of California, I command you to immediately disperse. Unlike in other jurisdictions, the audibility of the announcements was not an issue in San José. The Department’s equipment was suitable to the task of providing clear instruction. Nonetheless, in terms of contents and efficacy, there were aspects to the orders that merit re-consideration by SJPD. In its After Action Report, the Department noted that the dispersal order was largely ineffective on May 29; the crowd did not disperse as instructed. The dispersal order was repeated for nearly half an hour prior to any enforcement action. One protest participant candidly shared with us that, although she clearly heard the order repeated every few minutes, she and others in the crowd ignored it because the dispersal order became “background noise.” In her perception, there was no reason to leave. We acknowledge that the “right” approach to timing, repetition, and enforcement of dispersal orders does not lend itself to exact formulas. Circumstances vary. It is important to be clear and to give the crowd a reasonable amount of time to respond, and the obvious preference is not to have to resort to force. But when the dynamics are not shifting as a result of repeated announcements, further iterations of the same warning become easy to disregard, and ironically can contribute to reactions of surprise when physical action is finally (but seemingly “suddenly”) taken. While is impossible to affirmatively know why other individuals did not disperse, we did note that SJPD’s dispersal order lacked information that may have contributed to its efficacy. In our view, the best approaches are those that are straightforward but also detailed enough to put crowd members on notice of what is happening, why it is happening, what form compliance should take, and the potential consequences of non-compliance. Accomplishing this defuses what we consider to be the fairest criticism of such actions in the
Page| 49
multiple jurisdictions we reviewed: that people were subjected to force or arrest without fair warning. Modern crowd management best practices suggest that a dispersal order should include several key informational elements: • Potential for arrest • Warning of use of less lethal force or tear gas that may result in injury • Route(s) for dispersal • Length of time to disperse This information serves to put the crowd on notice of expectations to leave and potential enforcement activity to come. A reference to consequences, including the possibility of arrest or force, is a crucial component of effectiveness and legitimacy.27 SJPD’s dispersal order issued in English did not contain this recommended language and, as a result, the crowd was largely not on notice as to what police actions they could be subjected to if they failed to disperse.28 Nor were they informed of how long they had to leave the area before their continued presence would result in the police use of force or gas. While we recognize that some of the people with whom we spoke, and who felt called to protest police abuses in the first place, were predisposed toward distrust of law enforcement, the dispersal order process nonetheless provided additional fuel for feelings of antagonism – and in avoidable ways.29 Other protest participants or onlookers reported that they did not know where to go to safely leave the area or reach their vehicles, which were often parked behind police lines. We reviewed, for example, May 29 video footage of several protesters who approached the skirmish line and asked to cross 27 Deorle v. Rutherford 272 F.3d 1272, 1284 (9th Cir. 2001). 28 The current Department dispersal order now includes the new language. 29 One of the SJPD field supervisors with whom we spoke was candid in dismissing this as a concern – he effectively said that no one could have validly claimed to be unaware of the need to leave, and argued that at some point it is incumbent on people to simply follow legal orders. But we reiterate that the merits of this argument are diminished when the announcements are both limited in scope and repeated to the point of reduced credibility.
50 | Page
behind the line to access their parked cars on Santa Clara Street only to be told to leave without further instruction. We observed others who were given conflicting departure routes, with one line of officers directing protesters to move eastbound on Santa Clara Street and other officers directing them westbound on Santa Clara Street. Providing instruction to the crowd on ways to leave the area would also advise officers who might be similarly confused on the optimal dispersal route. The After Action Report made several recommendations regarding the dispersal order and declaration of unlawful assembly as follow: • Revise the Department’s dispersal order script to include POST recommended language, including an explicit warning about force, including chemical munitions. • Pre-record dispersal orders in the three languages most likely to be encountered in San José: English, Spanish and Vietnamese. • Incorporate into written and trained protocols a recommendation that unlawful assembly orders be repeated periodically once the Department is prepared to take enforcement action to avoid prolonged repeated announcements that may cause the crowd to become complacent. • When being used in the field, unlawful assembly orders should be given repeatedly and then enforcement action taken soon thereafter so the crowd does not become emboldened or complacent. • When dispersal orders are given, they should be published on various social media platforms and include instructions that the order applies to everyone present. OIR Group agrees with these recommendations and makes the following additional ones. First, the AAR recommends that SJPD modify its dispersal order script to including an explicit warning about force and gas; we would add that it should also clearly provide a route of safe departure and warning of arrest. And the route of departure should consider traffic, parking, and officer deployment when practicable.
Page| 51
Second, in addition to modifying the dispersal order script to add the aforementioned language, the Department should also update its policy regarding dispersal orders to include this language. Finally, we recommend that the Department consider strategies to communicate with community stakeholders in advance of future events to plan and coordinate as practicable. Modern crowd management best practices suggest that law enforcement should make attempts to meet and negotiate with protesters prior to issuing a formal dispersal order. We acknowledge that most of the activity on May 29 was informal and unorganized, but at least one protest was known and organized. Further, one event organizer shared that, despite making attempts to contact SJPD over a week in advance of her May 31 protest event, no one from the Department called her back. The Department should consider policy modification and strategies to enhance opportunities for communication with community stakeholders in advance of future events. RECOMMENDATION 6 SJPD should revise the Department’s dispersal order script in the three predominant languages to include an explicit warning about arrest, force and gas should the order be defied, and clear instruction regarding the safe routes of departure. RECOMMENDATION 7 SJPD should revise the Department’s Demonstrations and Civil Disturbances policy to include an explicit warning about arrest and use of force (including chemical munitions) should the order be defied, and clear instruction regarding safe routes of departure. RECOMMENDATION 8 In crafting dispersal orders to instruct crowds about the routes of egress from a protest, SJPD policy and training should advise officers to consider conditions such as parking, traffic and officer deployment to ensure the feasibility and safety of any direction provided about dispersal routes.
52 | Page
RECOMMENDATION 9 SJPD should consider adding language to its demonstration and civil disturbances policy advising, when practicable, that the Department contact event organizers or participants with the goal of gaining voluntary dispersal prior to issuing formal dispersal orders. RECOMMENDATION 10 SJPD should revise the Department’s Demonstrations and Civil Disturbances policy to include that when dispersal orders are given, they should also be published on various social media platforms immediately.
Deployment of Less Lethal Force: Data and Analysis SJPD used various “less lethal” force tools from May 29 to June 7. Less lethal force is defined as “force that is not reasonably anticipated and intended to create a substantial likelihood of causing death or serious injury.” As noted in the Department’s Preliminary AAR, the purpose of an After Action Report is not to evaluate each unique use of force. Which is not to say that such an exercise is not important: on the contrary, an agency should have a process for the rigorous examination of each force deployment’s appropriateness. But an AAR is less about individual accountability than a broader look at overall performance across a range of categories. While our own Report similarly refrains from an evaluation of individual deployments, there is no question that SJPD’s high volume of force uses, and the distinctive munitions that comprised those uses, was a major element of the operation – and a matter of considerable public interest and concern. This is unsurprising. The use of force by police in the demonstration context exacerbated the very tensions that had given rise to the protests, and it pushed at issues of free speech vs. law enforcement’s responsibility to
Page| 53
maintain order and public safety. The controversial injuries that resulted provided further fodder for critics of the SJPD response. Accordingly, we address the following topics specifically related to force: • Inexperience with civil unrest and limited supervision in the field set the stage for officers to deploy munitions with less focus and precision than would normally be expected for any use of force. • The unique circumstances contributed to deviations from normal reporting protocols, thus limiting the full ability to assess the scope and legitimacy of individual deployment decisions. • The Department does have knowledge of individual instances in which force was arguably or allegedly out of policy – but has thus far refrained for different reasons from making findings and addressing accountability – a dynamic that is concerning. • Apart from individual accountability issues, there were broader tactics and deployment decisions that merit further attention. These include the improvised uses of OC gas and flash bangs at different points in the operational period – not all of which comported with best practices. • The Department adjusted relevant policies during the operational period in response to unfolding events. We consider the shifts that occurred to be problematic in several ways that we detail later in this section of the Report. • At the same time, a change in deployment tactics by June 1 to a more restrained, reactive model of officer engagement appeared to contribute to a significantly lower volume of force incidents over the course of the last several days of the period. We also noted, and detail later, that the nature of the crowd changed in these subsequent days. • More positively, the Department also quickly amended relevant policy soon after the operational period in order to align it more closely with the emerging community preference for more restrictive use of force options for crowd management.
54 | Page
Force Reporting and Review Deficiencies A repeated theme across jurisdictions nationwide after the summer 2020 civil unrest was the deficiency in tracking, reporting, and accurately counting uses of force, and SJPD was no exception. This was a function of several factors, but two seem to be most significant. The first was a “worst of both worlds” phenomenon in which individual officers were not only using much more force than usual, but doing so in a context with far less time for reporting and reflection than the usual shift – or work week – would offer. The second factor that created a challenge even for well-intentioned agencies is that the subjects of force were, for the most part, not in custody and therefore not identifiable or a resource for follow-up investigation. This too was a difference from the usual scenario, in which force is used against people who ultimately become arrestees. SJPD seems to have suffered from these dynamics in its own reporting and subsequent ability to review. As noted in the AAR, the Department lacked a clear and effective mechanism to track officers’ uses of force on all levels from May 29 to June 7. Officers were directed to report their uses of force as soon as practicable after each deployment, but there does not appear to have been emphasis at the outset of the operation (and subsequently) that expectations of the Department were to be followed. The majority of officers did so by submitting a General Offense Supplemental Incident Report. The quality and detail in these reports varied. For example, some officers detailed the exact count and circumstances of their force, including a description of the actions that precipitated use of force and, when applicable, of the subject. Others, however, simply noted (for example) that they deployed “multiple” uses of force at “aggressive subjects” who were “throwing rocks and bottles.” While some of this imprecision is understandable for the above-stated reasons, it is nonetheless regrettable. It does little to reassure members of the public who questioned the legitimacy of individual force uses or the overall deployment strategies of SJPD. And it merited subsequent attention from the agency in terms of creative thinking about how to improve the documentation in situations such as this, which are admittedly challenging.
Page| 55
An effective review of these reports by first level supervisors would have differentiated between the reports that were sufficiently detailed – articulating the threats presented and explaining the involved officers’ decision-making – versus those that lacked the detail to meet Department expectations when force was used. Sergeants are expected to return for clarification and detail those reports that are insufficient. Yet there is no evidence that robust review occurred here. Going forward, the Department should reinforce expectations for first level supervisors on what information is expected in a force report. One alternative – apart from a clear emphasis on Department expectations in briefings and as needed throughout deployment – is the use of body-worn camera recordings to aid in the process. We have seen examples in other jurisdictions of officers narrating their own actions, and the reasons for them, in real time as part of the recordings – a practice which obviously aids in later recollection and accuracy. RECOMMENDATION 11 SJPD should ensure that briefings in advance of crowd control situations include reinforcement of the Department’s expectations regarding the reporting of uses of force. RECOMMENDATION 12 SJPD should ensure that supervisors responsible for reviewing use of force reports ensure the reports are sufficiently detailed before approval. RECOMMENDATION 13 SJPD should consider practices such as body-camera narration that might enhance officers’ ability to subsequently capture their observations and decision-making when reporting on dynamic, extended deployments. The AAR team used the Incident Reports that it did have to create an estimated force count and tracked these counts in an Excel workbook by officer, date, force type and count. As part of this evaluation, we conducted an audit of SJPD’s work. Using the Department’s own tracking workbook, OIR Group selected a random sample of officers across dates. We then
56 | Page
crosschecked the Department’s documentation against the officer’s Incident Report. Using this methodology, OIR Group found the Department’s own internal force count in their workbook, overall, to be an accurate reflection of officers’ reporting. However, this accuracy did not seemingly translate to the counts that were reported in the AAR: we noted that the total counts in the workbook did not always match the counts reported in the AAR’s Force Tables. SJPD personnel worked diligently with OIR Group to identify and explain these inaccuracies during our review process. After engaging with them, OIR Group has no reason to believe that these errors were intentional. The After Action team was fully transparent in providing its working papers for our review, and the Report Tables both over and underreported counts – thus undermining the idea that there was an attempt to mislead in service of a desired impression. And we reiterate that the Department’s production of the AAR in a compressed time frame was both impressive and a possible contributing factor to issues like this. As always, though, accuracy in public reporting is a value worth prioritizing. Our audit also discovered: • Special Operations personnel were, in general, more specific in their use of force reports than their counterparts in patrol. As we note elsewhere in this report, Special Operations personnel have more training and experience around force deployment in specialized events, especially the use of less-lethal tools, and are perhaps more accustomed to counting rounds and reporting deployment. Even then, however, we noted that some Special Operations personnel did not keep accurate count of their uses of force. • Uses of the 42” riot baton were particularly difficult to quantify. Perhaps because of this, the After Action team counted the number of officers who reported using force with their baton versus the actual counts of baton pushes or strikes. However, in reviewing body-worn camera footage, we observed several questionable uses of the 42” riot baton. These ranged from ineffective holding and pushing techniques to strikes on uncooperative, though non-aggressive, subjects. As we discuss in the Use of Force policy section and elsewhere in this Report, this may indicate the need for additional training and internal review.
Page| 57
The Department also seemingly lacked any method to count their munitions inventory from the start. To our knowledge, there were no counts of how many munitions the Department started with on May 29. And, when SJPD “ran out” of munitions in the field on May 29, officers scrambled to find munitions elsewhere in the City, such as from the Department’s firing range. But the range training “check-out” logs that were provided for our review had no records for May 29 to June 7. Finally, when new munitions were finally acquired, they came from a source outside the normal Department purchasing process due to a nationwide shortage. This unusual process also may have contributed to the inability to track inventory. Further, in our interviews, some SJPD personnel commented that officers may have carried extra rounds beyond what was issued to them, and that this is a normal practice, especially for Special Operations personnel. While there is no inherent problem with officers’ carrying extra rounds, it does contribute to the Department’s inability to control and track the uses of these munitions. Accordingly, for sake of consistency and tracking, SJPD should consider what an optimal number of rounds are and ensure that its personnel be advised regarding Department expectations. The Department made two specific recommendations related to tracking uses of force. The AAR recommended that: • [SJDP] establish a system to accurately record and document the deployment of less lethal weapons, to include the date, time, circumstances and number of munitions. • Supervisors and commanders should ensure the accurate documentation of all events, facts and uses of force as soon as practicable after the event. OIR Group concurs with these two recommendations and offers the following additions. RECOMMENDATION 14 SJPD should create a Department-wide inventory system specifically for less-lethal munitions so that the Department can track inventory and less lethal usage at any point in time.
58 | Page
RECOMMENDATION 15 SJPD should evaluate its practices and protocols governing the number and type of munitions issued to officers, including Special Operations personnel, and reinforce with officers the importance of uniformity and consistency in the number of rounds carried. We also take this opportunity to note again the unresolved nature of several force instances and related allegations of misconduct. The Department explained to us that it has chosen to defer its administrative investigations during the pendency of related civil litigation. Lawsuits about the conduct in question do constitute an exception to the normal state statutory requirement that any disciplinary consequences for officers must be imposed within one year of the date that the agency is aware of the potential policy violations. However, while the statute allows for the underlying investigation to be put on hold, it does not require it. We recognize the potential implications for liability when agencies take the formal step of finding fault with their personnel. For that reason, some jurisdictions do choose to wait. But in our view, the importance of timely investigations and resolutions and meaningful remediation also has a significance that outweighs concerns about civil litigation. It speaks to public confidence as well as the agency’s commitment to accountability and adaptation. Delaying accountability and transparency until the civil litigation has taken its course creates its own costs well beyond the concerns about the lawsuits. We’re disappointed by the idea that the Department’s full reckoning with these issues is still months if not years away. And waiting for the litigation also has consequences for complainants (who will need to wait an extraordinary time for their matter to be resolved) as well as involved officers (who will also need to wait to learn the outcome of the investigation). It seems like a missed opportunity to make needed and timely interventions in the most effective ways. Force Deployment: Types and Uses Before evaluating our concerns with uses of force, we present an overview of the force options used by SJPD over the days of summer unrest as a frame of
Page| 59
reference. Any counts provided herein are estimates from our evaluation of force reports and other Department materials. Type of Force à Projectile Impact Weapons (PIW) Projectile Impact Weapons, or “PIWs,” are less lethal impact munitions (sometimes referred to as “bean bags” and “rubber bullets”). They are typically most effective when deployed in encounters with a single individual. The San José community was particularly concerned with SJPD’s use of PIWs on the evening of May 29, when SJPD patrol and Special Operations officers deployed an estimated 550 PIWs, an amount that reportedly depleted their entire inventory.30 The first use of a PIW was reported at approximately 5:30 PM in the vicinity of Santa Clara Street and 7th Street and occurred throughout the evening. As we detail later, the use of PIWs in this context requires additional review and consideration. SJPD also deployed PIWs in specific instances on May 30 and 31 and June 2 and 5. Over the operational period, SJPD deployed the following types of PIWs: • Patrol and Special Operations officers used 40mm launchers to deploy 40mm foam baton rounds, which were meant to be used against specific subjects who were aggressive or assaultive. Department policy regarding these 40mm rounds changed mid-operation, as discussed below. • 40mm Impact Oleoresin Capsicum (OC) rounds were also used to disperse OC gas; these were fired at specific subjects or at targets around subjects (e.g., at a wall behind a group of individuals to disperse the OC gas over individuals). • Special Operations officers used 37mm launchers to deploy foam baton rounds, which were meant to be “skipped” off the ground in front of individuals to disperse a crowd. Per Department policy, use of these rounds required declaration of an unlawful assembly, reasonable time for the crowd to disperse, and sufficient distance between the officer 30 As previously noted, an exact count could not be ascertained due to deficiencies in counting/reporting.
60 | Page
and the crowd to allow the energy of the round to sufficiently dissipate to prevent any substantial risk of injury to any person. • Special Operations used a modified shotgun to deploy “stun bags,” a less-lethal munition designed to target a specific subject. Department policy regarding these types of rounds also changed mid-operation, as discussed below. Type of Force à Tear Gas Officers used two types of chemical munitions on May 29, 30, and 31: • Oleoresin Capsicum (OC) gas, commonly referred to as “OC gas” (or, when deployed from a handheld canister, “pepper spray”). OC gas is an inflammatory agent derived from the oil of hot pepper plants, which causes heat, redness, and swelling to the skin and irritation to the nose and eyes. According to Department records, OC gas was deployed via an OC Blast Stinger grenade and the 40mm OC Projectile Impact round (defined above). • Ortho-Chlorobenzalmalononitrite (CS) gas, or what most people refer to when they say “tear gas.” CS gas is an irritant, which causes intense stinging to the eyes and respiratory system. According to SJPD policy, the deployment of CS gas is held to a higher standard; only specially trained MERGE (a unit within Special Operations) officers are allowed to deploy CS gas after authorization from the Chief level. SJPD command personnel reported that they did not intend to use OC gas as a dispersal mechanism or to blanket the entire crowd. The OC gas, like the PIWs discussed above, was intended to strategically target and disperse “aggressive” protestors. We also noted that, in contrast to other jurisdictions that used either type of tear gas first, before other munitions, on May 29, SJPD officers did not deploy any gas until approximately 6:30 PM, nearly one hour after using PIWs. Type of Force à Flashbangs Noise Flash Devices, more commonly known as “flashbangs,” create a loud explosive sound and bright light that is meant to shock, disorient, or otherwise
Page| 61
distract a subject. Typically, these devices are used in tactical situations, such as a barricaded suspect, to distract or disorient a suspect. Special Operations officers deployed Noise Flash Devices (NFDs) at two points during the unrest. As we discuss in the following section, these uses of NFDs were largely “improvised” and should be evaluated by the Department, both for past accountability and future uses in crowd management contexts. First, during the course of crowd management on May 29, Special Operations officers deployed approximately 13 NFDs between approximately 6:45 and 10:00 PM at various locations on Santa Clara Street and 4th Street. Officers reported that at least two NFDs were deployed at or near moving vehicles to get the vehicles to move or stop driving erratically.31 The other NFDs were deployed approximately 3-5 yards in front of the skirmish line to disperse the crowd. Second, on June 5, Special Operations officers deployed two NFDs at approximately 11:30 PM in front City Hall (4th and Santa Clara Street). These officers reported that members of a “hostile” crowd were throwing frozen water bottles and paint bottles at the skirmish line. They used two NFDs in rapid sequence to push the crowd away from the skirmish line and create a safe distance between the officers and the crowd. Improvised Uses: Adapting and Learning As we have noted throughout this Report, the circumstances of last summer’s civil unrest were new and unexpected for many jurisdictions. While some officers were trained in each force option detailed above, many were not trained to use them for crowd management of this nature. As a result, force deployment was sometimes “improvised” to meet the needs of the moment. SJPD was no exception. Personnel acknowledged that their force deployment happened on a steep “learning curve,” especially on May 29, the first day of major unrest. 31 This use seems especially concerning. The intended “messaging” of a flashbang seems difficult for a driver to discern and comply with; moreover, its disorienting effect could prove particularly problematic to an individual operating a car adjacent to a crowd of protesters.
62 | Page
SJPD used some force tools, most notably OC gas and flashbangs, in an “improvised” way that was not initially effective or not consistent with intended use. For example, on May 29, Special Operations personnel first deployed OC gas via hand-tossed canisters. Personnel reported that this was ineffective for several reasons: protestors threw the gas canisters back at officers on the skirmish line, the wind shifted the gas toward officers, and Motor Units behind the line did not have gas masks. Accordingly, per SJPD’s own report, officers adapted. They acquired 40mm less lethal launchers, which they used to deploy OC gas via rounds fired from a distance. This allowed them to reach individuals who were hidden behind crowds or objects. While this was reportedly successful and avoided the pitfalls of the canister deployments, we noted that others in the crowd, who were not directly aggressive or combative, nonetheless experienced the negative effects of the gas. This latter dynamic is one of the reasons that tear gas use, of any type, proved controversial in a number of jurisdictions: along with the negative associations it carries of notorious, heavy-handed police “crackdowns” from the past, it fails to distinguish between aggressors and mere bystanders sincerely engaged in protest. Similarly, SJPD used flashbangs (not typically a tool for crowd management) in an improvised way, because the devices seemingly met the needs of the immediate moment. On May 29 specifically, the officers were concerned that their inventory of less-lethal munitions was running low and would soon run out. They had flashbangs in their “toolbox” and decided to use them; they admitted to us that they made this choice out of necessity, not because the flashbang was a tool designed for this job. Later that evening (and later that week), they observed that the loud noise of the OC Blast Stinger Grenade had seemingly been an effective deterrent to the crowd. They decided to again use the flashbangs because they produced an even louder explosive sound. This improvisation, they reported, successfully ended some “violent” encounters. While SJPD reported successful outcomes in these instances, we urge the Department to evaluate these and other “improvised” uses of force tools. Indeed, we question the overall safety of flashbangs as a tactical choice in a crowd-intensive context, given the variables of unpredictable movement and the reality that these devices initially combust at extremely high temperatures
Page| 63
(up to 2600 degrees centigrade), which can lead to injury or unintended explosions. SJPD should specifically examine the uses of flashbangs in a crowd management context and assess the advisability of this deployment under the circumstances presented. RECOMMENDATION 16 SJPD should examine the “improvised” uses of less lethal force tools through the lenses of accountability, advisability, and remediation, and should train and provide written guidance to officers on the appropriate uses of these tools. RECOMMENDATION 17 SJPD should specifically examine the use of flashbangs on May 29 and June 5 and develop protocols for limiting future deployment in a crowd control context. High-Velocity Rounds In another example of unique – and arguable – deployment strategies that emerged in these days, the Department acknowledged that it used approximately two 40mm “high-velocity” foam baton rounds in response to objects being allegedly thrown at officers from upper-level windows on Santa Clara Street. These rounds use the same foam baton projectile as the standard issue 40mm foam baton round, though the high-velocity round has a greater maximum effective range. The Department reported that these high-velocity rounds were effective in ending the overhead assault on officers. According to one commanding officer, the purpose of these rounds was as a “direct deterrent,” to blow out the apartment’s windows or strike external walls and thereby scare the
64 | Page
individuals throwing objects into stopping their attacks. SJPD did not intend to hit individuals with these rounds, nor were any injuries reported.32 Perhaps in a specific tactical operation with robust intelligence and focus, these rounds would be the most effective choice. But considering the potential for injury that use of these rounds during crowd management operations might engender, with chaotic and ever-changing circumstances, high numbers of bystanders, little to no intelligence about the apartments’ occupants (e.g., innocent occupants, children) and limited resources, their use should be carefully reviewed. RECOMMENDATION 18 SJPD should examine the use of high-velocity rounds through the lenses of accountability, advisability, and remediation, and consider if these rounds should be deployed during crowd management operations, and if so, under what conditions. A Shifting Landscape: PIW Policy Revisions and Deployment Issues While the above less lethal tools may have been technically “in policy” the use of PIWs from 37 and 40mm launchers generated a significant amount of controversy –issues of policy compliance were particularly confusing. Interestingly, and not coincidentally – relevant policy shifted repeatedly during the course of a month. These shifts occurred before, during, and after the operational period, in reflection of the intense experiences and reactions that were generated by those weeks. Step One: A Coincidental Change on May 22 On May 22, 2020, three days before the George Floyd murder occurred in Minneapolis, the Department happened to revise its Projectile Impact Weapon 32 Special Operations officers reported that they attempted to make contact with individuals in the apartments where they fired rounds to check on the occupants’ well-being, address injuries as needed, and potentially arrest identified suspects. The relevant people were not cooperative, though, and a decision was made not to forcibly enter the apartments to avoid additional confrontation.
Page| 65
and Chemical Agents policies in several notable ways. This was in reaction to new equipment that had become available to the agency, and for which it needed to establish parameters for permitted use. One significant change was that 40mm OC rounds were now authorized for crowd control, whereas they had previously been prohibited for that purpose. At the same time, the new policy prohibited Department personnel from using chemical devices (including delivery munitions or devices) without first receiving training from a Department-approved Chemical Agents and/or Less Lethal Impact Munitions instructor. It also restricted the deployment of CS gas to only members of the Department’s MERGE Unit, required prior approval from a MERGE commander, and called for emergency medical personnel to be staged on-scene prior to deployment except in exigent circumstances. Importantly, the policy also prohibited the use for crowd control of those 40mm Projectile Impact Weapons that do not contain chemical agents. And it maintained its previous prohibition against stun bag shotguns use for crowd control purposes. Simply put, the new policy provided officers new latitude to use “launchable” OC in order to control crowd movement (without other justification or provocation in terms of aggressive or assaultive behavior), while reinforcing prior limitations on the use of other launchable less lethal munitions for that purpose. Despite this explicit policy prohibiting use of 40mm PIWs and stun bag shotgun rounds, SJPD estimated that, during the first three days of the protests, the Department deployed almost 400 40mm PIW rounds (foam baton rounds) and at least 21 stun bag shotgun deployments on May 29 and May 30.33 And here is where the evaluation becomes complicated: while the use of these PIWs was prohibited for general use in crowd control, pre-existing policy allowed officers to use them against specific individuals who posed a threat of serious injury to the officers or others. 33 This was in addition to multiple deployments of the 37mm less lethal munitions that were previously authorized for crowd control as “skip rounds,” to be fired at the ground in front of aggressive individuals.
66 | Page
When asked, SJPD Command personnel took the position that all of these rounds had only been used against aggressive individuals whose behavior fell within the parameters of the policy. But the sheer number of individual deployments, the multiple assertions to the contrary by subjects of and witnesses to those deployments, the unique circumstances, and the imprecision or incompleteness of documentation all combine to raise questions about that claim of policy compliance. It is true that some officers’ Use of Force Reports articulated the aggressive/assaultive behavior. For example, some force reports stated that the intended target was stepping out from the crowd to throw a projectile at the skirmish line and detailed the suspect’s appearance and actions. In those specific cases where the suspect was directly aggressive and the officer articulated as much, the use of PIWs may have been aligned with policy. But, in other cases, these less lethal tools were used against suspects whose activity was not apparently assaultive. For example, in reviewing body-worn camera footage and reading officers’ General Order reports, we noted the following questionable PIW deployments, among others, that certainly merited scrutiny – and perhaps remediation – beyond anything we are aware of the Department’s having done: • PIWs used on passively resisting protestors standing with locked arms on Santa Clara Street in front of the skirmish line. These protestors were refusing to move but there was no observable evidence of assaultive behavior. • PIWs fired at one individual who refused to move “out of the way” of the skirmish line when these officers were attempting to target a group of aggressive individuals who were hiding behind a large dumpster.34 • PIW fired at a protestor who was walking by the line pouring what appears to be alcohol out of a beer can but is doing so in a passive manner. 34 This individual suffered the loss of a testicle as a result of the PIW strike.
Page| 67
• Deployment of PIWs on individuals who were fleeing the scene of an attempted bank robbery / looting on May 31 in an attempt to stop and arrest them. And we noted that these PIW deployments, in addition to all others, occurred in the context of a larger crowd management situation, which raises questions of inadvertent harm to unintended targets. Notably, the majority of officers who reported using PIWs reported that they could not confirm if the subject had been hit, or if she or he had been injured. To our knowledge, no use of impact munitions was connected with an arrest. In its AAR, the Department acknowledged the community’s concerns about the Department’s use of PIWs. It also explained that Internal Affairs would investigate the specific allegations that officers violated prescribed policy in their specific uses of PIW, including 40mm launchers and stun bag shotguns. To our knowledge, as of the date of publication, these cases are still under review and awaiting the outcome of civil litigation. Step Two: Mid-Operation Policy Revision on June 1 Three days into the operational period, on June 1, Department leadership adjusted the relevant policies – not to reinforce or expand restrictions, but instead to increase the latitude with which officers could use different less lethal force options. Instead of being a reaction to the negative reaction that force uses had already engendered, the shifts were meant to provide officers more force options in order to respond to an environment that the Department had internally and externally described as highly chaotic and dangerous. The June 1 changes reiterated that, like the previous policy, 37mm PIW and 40mm OC rounds were the only pre-authorized crowd dispersal tools. However, the policy change added stun bag shotguns and 40mm PIW foam as a crowd control option, if these were authorized by a member of Command Staff. These were the very munitions that had resulted in injuries and engendered much concern in the first days of unrest. These rounds could be used as “skip rounds” in crowd control, stated the new policy, “in exceptional circumstances,” though they had previously been prohibited from that specific use. Moreover, the June 1 revisions did not define “exceptional circumstances.”
68 | Page
The policy change also permitted the use of 37mm rounds without a lawful dispersal order if authorized by the Chief, Assistant Chief or Special Operations.35 In interviews, Command Staff reported that they communicated the mid-week policy changes during daily briefings held at the SAP Center.36 While their impression was that this approach was somewhat effective in providing officers with a high-level orientation, there was no time to provide the sort of “hands-on” training or reinforcement that such a switch would normally have warranted. Accordingly, officers with already limited training on these tools in a civil disorder context were deployed with a brand-new set of “rules” for their use and little, if any, supplemental guidance. Beyond the challenges that the changes posed to officers are questions of why these specific changes were made. To us, expanding the policy regarding use of PIWs in the face of (at the time, alleged) injuries and public concern was a confusing change that seemingly would increase the use of rounds that had proven problematic in the initial days of the unrest. In the course of our interviews, some personnel theorized as to the reasons. For example, two officers suggested that, because the authorized 37mm skip rounds were not effective (and primarily served to antagonize the crowd members, rather than deter their aggression), adding the 40mm and stun bag rounds to the authorized “toolkit” was actually a way to improve the effectiveness of PIWs and the overall ability to meet the challenges that the demonstrations had thus far posed. Still, the disconnect between much of the public sentiment and this element of the Department’s response was noteworthy. RECOMMENDATION 19 SJPD should consider options for improving the dissemination and clarification of mid-operation policy changes, so that affected personnel in the field are suitably equipped to adapt as intended. 35 The previous policy required a dispersal order. 36 It would have been helpful if the subject matter of these briefings – conveying new policy directives – had been documented.
Page| 69
Step Three: A Further Reconsideration on June 16 As described in the AAR, the then-Chief considered the experiences of the operational period and the public concerns that arose and moved swiftly to revise the PIW policy again. In the new iteration, which took effect on June 16, the Department prohibited the use of all PIWs against persons for the purposes of crowd control and crowd dispersals. It continued to authorize the use of PIWs against individuals who are actively attacking an officer or another person or who is armed and posed a threat to officers or other persons. And it discourages officers from using a PIW if the crowd is too dense for “accurate projectile placement.”37 The then-Chief explained the new policy to City Council, stating that the older policy versions were intended to disperse individuals engaged in unlawful assemblies without causing significant injuries. In skipping the rounds off of the ground, he stated, officers could disperse a crowd without purposefully striking individuals. But, he admitted, “[SJPD] did not have real-world experience using this technique,” and the technique “had likely caused injuries in the crowd.” This prompt adjustment by the Department is notable and creditable. We consider the new version to be appropriately responsive to the will of the community and current thinking about the balance between restraint toward demonstrations and police ability to protect themselves and others. However, there are still individuals who remain skeptical about whether less-lethal munitions can be effectively used in a crowd control situation, even under the limiting conditions set out by policy. Accordingly, SJPD should evaluate future uses of less lethal munitions in a crowd control context to assess whether unintended consequences of their deployment continue to occur. 37 SJPD’s Revised Duty Manuals states, “Such use shall be in compliance with DM section L 2629 – USE OF PROJECTILE IMPACT WEAPONS. When aiming a PIW at a violent individual during crowd control situations, officers are reminded of their responsibility for accurate round placement and their duty to avoid striking unintended subjects. In such circumstances, officers shall consider alternate solutions if the crowd density creates an unnecessary risk of striking individuals against whom the use of an PIW is not intended.”
70 | Page
RECOMMENDATION 20 SJPD should continue to evaluate use of less lethal munitions in a crowd control context in future incidents, and if the more restrictive policy still results in injuries to non-assaultive individuals, should reconsider whether to ban their use in a crowd control context altogether. New Approaches, Different Results One surprising finding was that no officers reported using force on June 1, 3 or 4. As previously noted, SJPD leadership determined that the Department would take a reactive stance starting on June 1 and onward, responding only if significant acts of violence occurred. SJPD did not deploy officers to protest activity on June 3 or 4, so we would not expect to see any force on those days. But SJPD did deploy officers on June 1: Department personnel formed at least two skirmish lines (at City Hall and outside of Cesar Chavez Park) and affected arrests later that evening. There were also reports of the crowd becoming aggressive at various points in the day. But no force was reported – and this seemingly corresponded with the reality in the field. OIR Group reviewed footage from AIR3 and did not observe any obvious physical engagements with the crowd. Interviews with Command staff also confirmed these observations. That no force was used on these days, particularly on June 1, seems to reflect the positive impact of the Department’s newly adopted, “observe and react” tactical stance, coupled with increased briefings and command and control. While other variables – such as crowd behavior – presumably played a role as well, this shift is significant and to SJPD’s credit.
Page| 71
Analysis of Arrests Arrest Data: Overview Between May 29 and June 5, SJPD made 155 arrests.38 The vast majority of these were for misdemeanor violations for which arrestees were released after issuance of a criminal citation with a later court date. As was true for other areas of data collection for reasons we discussed in detail earlier in this Report, various reports of the number and types of arrests for this time period were difficult to reconcile. The number of arrests detailed in the Department’s After Action Report were not supported by data we later received from the Department in response to our document request. There is no clear way to account for the differences, beyond the Department’s acknowledgement that the numbers reported in the AAR were hastily put together and do not reflect later, more accurate accounting. The differences do not favor the Department one way or the other, and we did not find the misreporting to be calculated or in any way intended to mislead. The difficulty with arrest data is partially attributed to initial confusion and miscommunication between arresting officers and those transporting arrestees to the Command Post for processing. On May 29, many arrestees arrived at the processing center without detailed information regarding their identity, reason for arrest, or identification of the arresting officer. This both delayed processing and muddled it. But even when these communications issues were remedied on subsequent days by regular reminders from command staff about the importance of conveying complete and accurate information, the data challenges persisted because technological issues at the Command Post and the general confusion and chaos that persisted meant the information was 38 The Department’s AAR reported 176 arrests made across this time period. But this number is at odds with the arrest log the Department produced to us in response to our data request. The total number is not the only discrepancy between the two documents; the daily arrest totals and counts of specific types of citations issued were also divergent in ways that we could not reconcile. We discussed this discrepancy with SJPD personnel who prepared the AAR and concluded that the numbers contained in the arrest log were likely more reliable.
72 | Page
not being electronically linked to reports in the Department’s CAD as they would be under more normal circumstances. Demonstration-related Arrests: May 29 – June 5 Misdemeanors Felonies Unknown Classification Total arrests May 29 – – 32 32 May 30 46 5 0 51 May 31 22 8 1 31 June 1 20 1 0 21 June 2 16 1 0 17 June 3 0 0 0 0 June 4 0 0 0 0 June 5 2 1 0 3 On Saturday, May 30, all but one of the 46 misdemeanor citations was for unlawful assembly (the one exception was an arrest for disorderly conduct). After imposition of the curfew on May 31, citations for violating the curfew order replaced those for unlawful assembly, with some additional arrests for resisting or obstructing the police. In all, given the size and scale of the demonstrations, and relative to protest-related arrests we saw in other cities during the same time period, the number of arrests over the span of days was relatively low. Arrest Tactics: “Wait Until They Leave” The arrest numbers were driven, in part, by staffing limitations and strategic decision making. Even though officers reported seeing numerous crimes committed by individuals within the crowds, they were generally unable to effectuate arrests. In a typical crowd control scenario, officers try to control violent or criminal elements within an otherwise peaceful crowd by sending a team of officers into the crowd to surround or encircle an offender. Often,
Page| 73
again in a typical scenario, those peacefully demonstrating cooperate with these efforts, with the understanding that violent protesters create larger problems for the entire assembly. But the protests of May and June 2020 – in San José and across the country – featured distinct crowd dynamics and created challenges outside the norm. Here, SJPD decided early on that it would be unwise to attempt to pull those involved in suspected criminal activity out of the crowd, given the volatile nature of the protests and the levels of hostility being broadly displayed toward officers. Special Operations commanders articulated the concern that officers attempting to effectuate arrests within the larger crowd would face substantial risk of assault and, more broadly, were likely to further inflame or incite the crowd. In addition, as we have discussed elsewhere throughout this report, commanders believed they did not have sufficient resources to staff teams of officers capable of moving through the crowd to make arrests. The crowd control strategy, therefore, did not emphasize arresting people, except to the extent that aggressive individuals could be identified and apprehended after they separated themselves from the crowd. For example, we observed one instance where individuals were exiting and entering their vehicle as they threw objects at the skirmish line on Santa Clara Street at City Hall. A description of the vehicle and its occupants was communicated by AIR3 via the radio and AIR3 tracked this vehicle as it moved through the downtown core. We do not know if the occupants were eventually apprehended but, according to SJPD personnel, the goal was to wait for this vehicle to leave the downtown core, pull it over, and then cite or arrest the occupants (rather than attempt to arrest the offenders in the midst of crowd control operations). However, unless the individual(s) could be identified and tracked after they left the downtown core (as in the above example), SJPD’s tactic was to simply allow some low level criminal behavior to go unchecked, to the extent that it was not causing danger to others. As we observed in other jurisdictions across the country, the Department did not have the resources to both effectively manage the crowd and police the more minor criminal elements. On subsequent days, when their tactical approach transitioned away from resource-intensive skirmish lines to more strategic monitoring and Strike Teams, SJPD personnel did engage in more targeted arrests. This approach
74 | Page
still had mixed results. For example, as we detail in the Operation Overview section above, on Friday, June 5, AIR3 observed an individual pointing a laser at it and communicated his exact location and description to the Command Post. A team of officers deployed to arrest this individual; as we described above, the operation was successful but also incited the previously peaceful crowd, and the arrest team was “trapped” in City Hall until they could be rescued by another team of officers.39 These decisions are, of course, easier to assess after the fact than they are to get right in the moment. Every approach comes with its assorted pros and cons, as well as its critics. In our interviews with Department leaders, they acknowledged the challenges of these different tactical approaches and concerns raised by the various options, and understand the potential for learning from these decision points. Transportation and Processing Those that were arrested were detained in the field until they could be transported to the field jail (located at the SAP Center, but at the opposite end of the parking lot from the Command Post) for processing. SJPD had two vans (referred to as “wagons”) equipped for transporting arrestees, but only one was operational at the time. At times, the single van was not available to respond quickly, and patrol cars were deployed to transport arrestees. This resulted in occasional delays, with restrained arrestees left sitting on curbs awaiting transport and requiring officers to remain present standing guard, unable to move on and respond to additional service needs. The Department has since purchased two new transportation vans, which should ameliorate this issue in the future. From the outset of the protests on May 29, the Santa Clara County Sheriff’s Office provided staffing, equipment, and transport buses to assist with processing a potentially large number of arrestees.40 There was a clear 39 We note this specific incident in contrast to an earlier incident of an individual pointing a laser at AIR3 on Friday, May 29 at approximately 9:30 PM. That individual was identified but SJPD did not pursue or arrest him/her, to our knowledge. 40 The buses were used to hold arrestees pending transport to jail but were not nimble enough to be deployed to pick up individuals arrested in the field.
Page| 75
division of labor once arrestees arrived at the field jail, with Sheriff’s personnel in charge of booking and processing, while SJPD maintained personnel available to write citations for the vast majority of arrestees whom were released rather than transported to jail. Transporting, booking, and safely and humanely detaining a large number of individuals at one time is challenging for any law enforcement agency. SJPD’s ability to call on the Sheriff’s Office and work cooperatively with that agency was an advantage not every police agency enjoyed as they grappled with problems posed by large demonstrations last summer. Because SJPD staff was already stretched thin, it depended on Sheriff’s Office assistance in these critical areas of operation. Nonetheless, SJPD acknowledges it could have done better for its part. As we noted above, problems arose on May 29 because officers in the field did not provide appropriate paperwork to the transport officers before placing arrestees in vans. This delayed processing at the field jail and created confusion and inaccuracies. This issue was largely resolved on May 30 and subsequent days by greater supervisory attention, and the Department’s After Action Report recommended development of guidelines and protocols that provide clear direction to arresting, transporting, and booking officers as well as field supervisors. We agree with these recommendations, and further recommend that these issues be covered in a more comprehensive policy and training governing mass arrests and bookings. A new policy should establish a clearly articulated plan for how to handle the various issues related to mass arrests, including the minutiae of things like which forms officers should use to document arrests, and include detailed expectations around staffing and division of responsibilities.
76 | Page
RECOMMENDATION 21 The Department should develop a policy on mass arrest and booking procedures that establishes a clearly articulated plan for handling the range of necessary tasks and includes detailed information about necessary forms, staffing assignments and division of responsibilities. Field Jail and Releases Department leaders decided to locate a field jail to handle the processing of arrestees at the SAP Center, at the opposite end of the parking lot which was serving as the Command Post. While this setting had some advantages – it was open and relatively easy to secure, and its proximity to the Command Post simplified some communications and control issues – it also had some significant downsides. Specifically, there were no bathrooms or running water, and it was too close to the hub of protest activity to allow for direct releases. The lack of facilities was a source of a significant number of complaints. Restrooms in the SAP Center were available for law enforcement use, but security concerns dictate that a holding facility for arrestees have separate bathrooms. Pandemic-related closures further exacerbated these conditions. The Sheriff’s Office brought in portable toilets, but not until May 31. The desire and need to expeditiously release people from the jail created another set of problems. With the exception of a handful of individuals arrested on felony charges, everyone arrested was ultimately “cited out” on the night of their arrest, meaning they were given a citation and admonition to appear in court on a later date. But the process of writing citations was complicated and delayed at times by the lack of reliable documentation in the field. Another issue that complicated the release process – and led to a large number of complaints we heard throughout our review – was transportation. Because the field jail was located adjacent to the Command Post and near the site of the major protest activity, the Sheriff’s Office decided to transport people away from the field jail, generally well away from downtown San José and toward the County’s northern and eastern borders, prior to releasing them.
Page| 77
This resulted in some delays, but created more significant concerns about the location. Sheriff’s personnel reportedly chose the release spots based on access to bus transportation, but we repeatedly heard about arrestees being dropped off “in the middle of nowhere.” Because these were issues within the Sheriff’s control, they are outside the scope of our review and beyond the span of SJPD authority. Nonetheless, because they were SJPD arrests, the Department bore the brunt of the criticism about citations and releases. In interviews with City leadership, we learned that members of City Council raised this issue with the Sheriff’s Office in a letter outlining their concerns with the transportation and release process but, to date, had not received a response. Because the Sheriff’s operations have such an impact on San José residents and SJPD arrests, City and Department leaders should continue efforts to dialogue with the Sheriff’s Office about protocols for handling similar situations in the future, with the goal of establishing guidelines and expectations for the timing and location of releases. RECOMMENDATION 22 City and Department leadership should engage in conversations with the Santa Clara County Sheriff’s Office concerning protocols for handling the release of individuals arrested and cited during protests or other mass gatherings, with the goal of establishing guidelines and expectations for the timing and location of releases. Curfew as a Tactic On Sunday, May 31, the City Manager (acting in his role as Director of Emergency Services) proclaimed a local emergency related to the civil unrest as a predicate to imposing a city-wide curfew, beginning at 8:30 PM through 5:00 AM the following morning. These curfew hours continued for the next four nights, ending on the morning of June 4. City leaders had considered the idea of a curfew as early as Friday night, struggling to strike the right balance between protecting protesters’ First Amendment rights and the desire to protect property and restore order. After the first two nights of protests devolved into some level of destructiveness and vandalism, and in response to
78 | Page
intelligence that he had received regarding planned looting, the Chief requested that the City impose a curfew beginning Sunday night, to give the Department a tool to limit activity and thereby maintain control. The Emergency Proclamation and curfew order was written sometime beginning Sunday morning, and announced by the Mayor, City Manager, and Chief at a press conference around 5:00 that afternoon. The City also publicized the curfew via social media platforms. However, the general perception was that there was not a significant amount of “lead time” to ensure clear and effective dissemination of the curfew order’s parameters before it went into effect. We discuss these communications issues in more detail below; it is clear, though, that confusion and uncertainty about the order (particularly on the first night of the curfew) fueled protesters’ frustration about the police response in general and, more specifically, about curfew enforcement efforts. City leaders understood the purpose of the curfew order was to provide a sort of “cooling off” period – to allow protests to continue during the day but to let police clear the streets after dark, in a way that was not dependent on law enforcement’s judgment call about when a particular assembly had become “unlawful” and subject to dispersal orders. By providing clarity that any gathering that continued past 8:30 PM was illegal, the City gave SJPD the means to cite and arrest anyone who remained in the area, with some exceptions. The Department shared that this was effective in dispersing most peaceful protesters, leaving behind those intent on perpetuating the lawlessness of the civil unrest. Simplifying SJPD’s ability to detain these individuals limited vandalism and other impacts on downtown businesses. On Tuesday, June 2, following a lengthy public discussion, City Council voted to end the nightly curfew as of Thursday morning at 5:00 AM. Over the four days the curfew order was in effect, SJPD arrested or cited 54 individuals for violating the curfew; 18 on May 31, 19 on June 1, and 17 on June 2. No arrests were made on June 3. Eighteen of these 54 individuals were arrested on additional charges (most often resisting or obstructing an officer, or failure to disperse).41 These arrests accounted for nearly 80% 41 As we discuss in greater detail in the Arrest section of this Report, these numbers differ from those presented by SJPD in its After Action Report.
Page| 79
of the total arrests made those three days. The City Attorney later decided not to prosecute these cited curfew violations. The City was not the first jurisdiction in the State or region to impose a curfew. Indeed, similar orders were used across the country during that week in ways that have become part of a larger public discussion about balancing free speech and assembly rights with protecting property interests. The police argue that curfew orders were – and are – effective and necessary tools for clearing the streets by making large-scale mass arrests during periods of civil disturbance. It gives officers a “short-cut” to detain and arrest suspected or potential vandals or looters without having to gather and document evidence of those specific crimes. Instead, mere presence on the street past the appointed hour is sufficient cause. But San José’s decision to shut down its streets was not universally seen as an effective “cooling off;” rather, it further inflamed existing anti-police sentiment among those who saw the curfew order as nothing more than a tool to suppress protesters’ criticism of police. The ACLU spoke out against curfews throughout the Bay Area, and has filed a lawsuit in Southern California challenging the constitutionality of curfew orders in that part of the state, seeking decisions that may have statewide impact. Criticism of curfews throughout California and nationwide has included concerns about selective enforcement in communities of color and the burden they impose on people who work irregular hours, most often in service industry jobs where workers are disproportionately people of color.42 Enforcement of curfews leads to increased police interactions with individuals and potential confrontations that can be seen as exacerbating the dynamics that gave rise to the protest movement in the first place. And while 42 In communications with us, SJPD was clear about the conscious efforts it made to avoid the excesses and collateral harms that curfews can create.
Curfew Violations: May 31 – June 2 Curfew citations (SJMC 8.08.260) Total arrests May 31 18 31 June 1 19 21 June 2 17 17
80 | Page
the curfew provides an effective enforcement tool for keeping streets clear, it creates the potential for overreach, disparate impact, and largely unfettered discretion to police on who to arrest and when. Balancing these concerns with the need to give police tools to prevent looting and violence is a delicate task, and one that requires trust and confidence in the police to wield a rather blunt instrument (a curfew order) judiciously. We recommend the City engage with its community on these critical issues, to develop some expectations and guidelines on whether and how to use curfews in future circumstances. Beyond the fundamental concerns about the impact of curfews on individuals’ constitutional rights, some raised concerns about the way in which the City communicated the imposition and parameters of the curfew order. Announcement of the curfew came just three and a half hours before it was to go into effect and was provided only in English.43 The short notice left residents uninvolved in the protests but wishing to comply with the order little time to plan their evenings or seek clarity on questions like exceptions for travel to and from work.44 It also gave SJPD little time to brief its officers about its enforcement expectations, or the limits and exemptions of the curfew order. While recognizing that establishing a curfew in response to an emergency is almost definitionally challenging from a “lead time” and notification perspective, we encourage the City and SJPD to consider ways to improve upon its dissemination of this critical information. 43 We also note that, in our discussions with San José elected officials about period of protest, there were expressions of concern that they themselves had not been looped into the decision-making or notification process. While the City Manager’s authority to act unilaterally in this regard – at least in the short run – exists for good reason, the effort to involve government officials in the process to the extent possible is also a priority worth remembering. 44 The Department, in collaboration with other City Departments, coordinated messaging through portable digital SMS signs that were placed at all major arteries into San Jose. The curfew and hours were listed. This was obviously a helpful step for which the City deserves credit.
Page| 81
RECOMMENDATION 23 The City should engage with its community in developing guidelines on whether and how curfews should be utilized, particularly when related to activity protected by First Amendment protections. RECOMMENDATION 24 Guidelines should include clear expectations about how any future curfew orders will be communicated to the public, including assuring that they are communicated at least in the three predominant languages in San José. City Collaboration: EOC San José has experienced major events, both emergent (such as the COVID-19 pandemic outbreak, major flood in 2017, and ongoing wildfires) and pre-planned (such as the annual Cinco de Mayo and Mardi Gras celebrations), that required all City agencies – from Fire to Public Works – to work as a collaborative team. That collaboration, City officials informed us, typically occurs in the City’s robust Emergency Operations Center (EOC). The San José EOC’s purpose is to coordinate peripheral support for large-scale events to allow the lead agency to focus on field and operation management. A liaison from the lead agency works closely with the EOC to manage this collaborative effort. For example, in recent wildfires, the EOC worked with the Fire Department to provide City resources, such as coordination of evacuation orders and public-facing communication, so that the Fire Chief and his personnel could focus on fighting fires in the field. A similar protocol was followed with the Public Health Department to manage the COVID-19 pandemic. And while both emergency and pre-planned events have typically resulted in successful outcomes for the City, especially in the recent past, City partners respectfully expressed a common frustration regarding working with SJPD:
82 | Page
namely, that the Department is inclined much more toward independence and autonomy than a prioritization of inter-department communication and teamwork. This dynamic is, perhaps, a reality of police culture that is far from unique to San José. The very nature of police work has a tendency to promote this: police are used to responding spontaneously to unpredictable circumstances, and in many cases, going it alone. Flexibility and adaptability are valued, and, accordingly, reliance on other City partners can be relatively limited.45 Until recent months, in fact, SJPD did not have a seat (literally or figuratively) in the City’s Emergency Operations Center; SJPD had its own, state-of-the-art center for incident command and planning in the same building as the EOC and, as such, sometimes shared information. The City acknowledged that, in summer of 2020, the EOC was largely focused on a public health crisis, something that did not require a strong police relationship. But when the City unrest erupted on May 29 and police responded, the EOC, which had usually played a major role in supporting operations for other agencies, was out of the proverbial “loop.” And SJPD did not request its involvement for logistics such as setting up barricades or traffic control. Furthermore, use of the EOC to coordinate the public-facing communication through the City’s “Emergency” Public Information Officer may well have eliminated some of the contention and pushback that SJPD’s announcements produced in an adversarial environment. Again, though, this step was not taken. We address the concern with public communication later in this Report. Eventually, effective City-wide collaboration did occur. At some point, for example, SJPD asked the EOC to coordinate the “Beautify San José” team who provided clean-up services for graffiti and other vandalism. But there was, and perhaps remains, room for improvement. There was contention about the creation of the City-wide curfew on May 31.46 The City Manager’s 45 Some within SJPD hypothesized that the Department was also reticent to be perceived as being part of “liberal City politics” due to political pressures from the officer’s employee association and uniformed personnel. 46 Even after collaborating with the EOC and other City agencies to create the curfew order, the then-Chief was quoted as stating that the City had not briefed him about the curfew order. This was a source of some consternation among involved City officials with whom we spoke.
Page| 83
Office had the lead responsibility for coordinating the drafting in conjunction with City Attorney personnel; one involved person described the process as “siloed” in a way that worked to the disadvantage of an effective roll-out. Encouragingly, City personnel reported to us that coordination between the City and SJPD since the summer unrest has improved, citing the creation of an Incident Management Team to complement the police team. Building on this momentum would be advisable. In our view, the integration of police officials with other stakeholders and responsible parties is at the heart of a functional model for local governments. The varied components of an effective response to large-scale events – including episodes of civil unrest – are representative of this dynamic. Accordingly, we encourage San José to develop a coordinated plan for responding to future civil unrest events, with designated liaisons and protocols for ensuring the awareness and contributions of all potentially helpful City entities. RECOMMENDATION 25 SJPD and City leadership (specifically, the Deputy Managers assigned to manage the Emergency Operations Center) should continue to collaborate, especially around intelligence-sharing. RECOMMENDATION 26 City leadership in the City Manager’s Office should consider collaborating with SJPD and other relevant agencies to create City-wide plans for managing civil unrest, especially spontaneous occurrences. Effective Mutual Aid The Department’s After Action Report outlined the successful use of mutual aid during the early days of civil unrest. Unlike some other jurisdictions nationwide that faced issues with mutual aid ranging from deployment to radio compatibility and equipment, SJPD successfully utilized over 400 officers from other agencies plus support from the Sheriff’s Office, from May 29 to 31. The AAR noted that having a dedicated mutual aid liaison at the Command Post would have helped coordinate more effective on-going deployment of
84 | Page
mutual aid resources. OIR Group concurs. As we note throughout this Report, more command resources could have alleviated this and other issues at the Command Post. The AAR did not identify one concern that SJPD personnel shared with OIR Group regarding mutual aid: mutual aid teams consisted largely of sergeants plus officers, with no command-level personnel. When these teams were sent into downtown or other areas of the City, SJPD’s already limited command capabilities in the field were further strained. SJPD may have benefitted from requesting additional Lieutenant and higher-ranking personnel as part of their mutual aid response. RECOMMENDATION 27 When requesting mutual aid, SJPD should specify the need for command-level personnel to be included in the request. Additionally, and as we mentioned above in our initial section about the strengths and limitations of the Department’s AAR, SJPD drew attention to the fact that the county-wide mutual aid agreement among law enforcement agencies dated back to 2009. We acknowledge and concur with the resulting recommendation: that the Police Chiefs’ Association of Santa Clara County convene to determine whether updates are necessary or appropriate. But we reiterate that surrounding agencies appear to have provided considerable assistance in spite of the dated nature of any written protocols. Moreover (and unlike in other jurisdictions), there did not appear to be problematic issues in terms of officers from elsewhere proving difficult to coordinate or align with Department policy and objectives. That being said, law enforcement in Santa Clara County should endeavor to align policies regarding use of force in a crowd control context. The rules of engagement should not differ, for example, depending on whether one skirmish line in San José was composed of SJPD officers while another was comprised of another agency with different use of force rules. Finally, we noted that, with respect to tracking and documentation, SJPD did not collect detailed reports, body worn camera footage, and other documentation relating to responding agencies. As a result, this material was not available for the AAR or other internal review. Any updated Mutual Aid
Page| 85
Agreement should set out expectations that such information will be collected by responding agencies and provided to the requesting agency for review. RECOMMENDATION 28 As the largest law enforcement agency in the County, SJPD should work with other agencies to develop uniformity among polices for use of force in a crowd control context. RECOMMENDATION 29 When updating the mutual aid agreement with local partners, SJDP should develop a process for information-sharing that, at a minimum, requires assisting agencies to document and share information regarding incident reports, arrest reports, and uses of force. One noteworthy mutual aid resource that the City did not seek to utilize was the National Guard. In our conversations with elected officials, we learned that this was a conscious choice, based on an impression that, for all of their attributes in providing assistance after natural disasters, their law enforcement training was perhaps not commensurate with the needs of the moment. This approach also meant that San José avoided the negative reaction that was produced in other jurisdictions by the Guard’s obviously militarized presence. As with so many aspects of the protest movement, deployment of the National Guard engendered differing – and sometimes clashing – perspectives in the many jurisdictions that experienced unrest. The Guard was used in two other cities whose responses we evaluated formally in public reports. Controversy followed in both instances – but in one, a significant source of criticism was that they hadn’t been called sooner. While the “right” answer can be in the eye of the beholder as well as a matter of individual circumstances, San José appears to have been intentional in its decision-making in this regard. And the outcomes of the operational period – particularly after the distinctive challenges of May 29 – validate the choice that was made.
86 | Page
Communication with the Public A police agency’s communication with the public both during and after a large-scale incident is of critical importance in a few ways. One necessity is for the basic sharing of information about what is happening, why it is happening, and what the public needs to know in order to stay safe and remain aware of relevant developments. Another is the projection of the law enforcement perspective in a way that reflects understanding of public concerns, accountability for shortcomings, and a commitment to the community’s best interest. Accomplishing these goals effectively can be challenging. Here, with the dynamic conditions in the City and the underlying tensions that animated and fueled the protests, the challenges were magnified. We acknowledge that, under the prevailing circumstances, SJPD would have been hard pressed to take any approach to communication that decisively resolved the unrest and avoided a rancorous response. As one City official noted in speaking with us, this was a week in which the Department was inevitably going to be on the receiving end of negativity. But even so, the Department’s After Action Report noted limitations and missteps that offer learning opportunities for the future, and we amplify and supplement those here. The category of “Communication” is, of course, a broad one. It covers everything from the verbal interactions of officers clashing with protesters on the skirmish line to the dispersal orders issued over loudspeakers in the field to the more formal and controlled announcements and updates provided to the press and public as the week progressed. The AAR also featured a highly publicized example of an officer “taking a knee” with protesters in an effort to defuse disruptive activity through a show of solidarity. The issue of when and whether officers should engage in this specific and symbolically notable form of outreach is a complex one that we discuss below. This too is an element of communication. To highlight a few of these issues here, we begin by again mentioning the dispersal orders as an example of the Department’s mixed effectiveness in communicating. Those issues are covered above, as we commend and add to the relevant recommendations emerging from the Department’s AAR.
Page| 87
A second example of a communication concern related to the curfew that was imposed on May 31 and lasted for the next four nights.47 In this case, the City Manager’s Office, the Mayor’s Office, and the Chief of Police aligned on the need to introduce the curfew in response to the level of unrest that had already occurred, and out of concern for planned looting activities about which they had received intelligence. San José was neither the first nor the only Bay Area jurisdiction to take such a step. But it was extraordinary nonetheless, and there were divisive aspects to it. Some of these related to the fact that it was happening at all, as a significantly escalated show of government authority at a time when public dissent was itself quite pervasive. But there were other critiques relating to the effectiveness with which the orders were communicated to the public. This was particularly true on the first day, when things moved quickly in terms of internal authorization, drafting, and sharing of the plan and its particulars. The curfew was announced in an afternoon press conference held by the Mayor, City Manager, and Chief. But there was not a significant amount of “lead time” to disseminate the information through a range of mechanisms and to ensure that parameters were clear before it took effect. It fell to SJPD to enforce the curfew, and it made relevant announcements by helicopter and loudspeaker. Nonetheless, there was confusion and uncertainty among some members of the public; we heard from multiple representatives of community groups who professed that they and their counterparts were unaware of the new restrictions. This obviously compounded the tension and frustration that police responses were generating throughout that period. The first night of a curfew presents inherent difficulties, since by its nature it is an unusual, far-reaching step that occurs in response to unplanned circumstances. Still, we talked to City officials who were involved in the preparation of the order who felt that better internal interaction among departments was possible, and that more could have been done to make the public aware and to make enforcement more graduated and less confrontational. 47 We discuss the broader dynamics of the curfew above.
88 | Page
RECOMMENDATION 30 The City should review its internal protocols for deciding upon, communicating, and effectuating curfew parameters, including the coordinated interplay among departments, review of timing and enforcement strategies, and use of the City’s existing public messaging systems. Additionally, press conferences held on May 31 and June 4 created a narrative of officers “at war” with their City residents; messages of apology or fault were lost amid what one commenter called “sob stories” from SJPD.48 The narratives of violence and of a city “under siege” shared during these press conferences did not match the experiences of at least some protesters and residents – and seemed abrasively out of step with the prevailing concerns about policing in general and SJPD’s “crowd control” efforts in particular. The tone and points of emphasis were off-putting to many, at a time when public confidence was already strained. While the intended purpose was communication and transparency, the press conferences were received by parts of the public as spin that defended questionable uses of force and sought to mitigate the controversial behavior of one Special Operations officer in particular.49 A five-minute video shown during the June 4 press conference showcased violence and vandalism among the protesters; again, though, this selective, edited presentation could be and was interpreted as manipulation rather than eye-opening corroboration.50 It is unclear why a Department-designated Public/Press Information Officer (PIO) did not lead the public communication from May 29 to June 7. During our interviews, OIR Group learned that Department leadership chose a field tactical commander, who was also a former Department PIO, for the press conference because this commander was both prepared for the role and the 48 The May 31 press conference may be viewed at: https://www.youtube.com/watch?v=BXTynYisTMs 49 To his credit, the then-Chief later acknowledged that his initial remarks about this officer (which mixed pledges of accountability with testimonials to the officer’s career and contributions to the agency) had failed to strike the appropriate tone. 50 The June 4 press conference may be viewed at: https://www.youtube.com/watch?v=ulacZshfwW0
Page| 89
most knowledgeable about tactics and operations. While this specific tactical commander obviously brought relevant credentials to the task, his own immersion in the front lines of the protest response may have been an impediment to his striking the most effective tone in this context. The Department acknowledged this deficiency in its After Action Report, making the following recommendations regarding public communication: • During large-scale incidents, a Press Information Officer (PIO) should be assigned to the command post for impromptu press interviews and messaging. • Press Information Officers should leverage social media to communicate important messages to the public during events. OIR Group concurs with these recommendations. Lastly, we return to the “micro” level of communication – that between individual officers and members of the public in the context of enforcement activity. It is striking that so much of the general public’s negative perception of SJPD and its handling of this incident arose from the widely circulated video clips that featured one officer’s alarming verbal belligerence and aggression on May 29. The reputational harm this caused is interesting for multiple reasons. One point is surely and understandably a source of frustration for law enforcement professionals everywhere: that egregious lapses by a small number of officers can so easily end up reflecting poorly on all. In this case, the officer’s behavior exacerbated the underlying narrative of abusive police actions, and overshadowed the control, restraint, and effectiveness of many of his peers during this same period. Another takeaway is that the ability to absorb constitutionally protected hostility as expressed by the public is a skill that modern policing requires to an unprecedented extent. We make this observation in full recognition that saying this is easier than accomplishing it in the field – indeed, the body-worn camera footage shows moments of antagonism toward police that would strain anyone’s equanimity. Accordingly, we encourage the Department to reinforce skill-building that assists officers in coping with these scenarios. We also noted the issue of “taking a knee” as it played out in San José (as well as numerous other jurisdictions). This symbolic gesture has emerged as
90 | Page
a powerful one, but it is not without complication as it unfolds in a dynamic environment. We have heard assertions of officer safety concerns, for example (though these can be overblown, and should not be the justification for a blanket policy). And there are situations in which benign requests for a show of understanding are mingled uncomfortably with the “testing” of law enforcement representatives or a more aggressive posture of demand. Within police agencies, there is concern that the political nature of the gesture makes it inappropriate regardless of one’s sympathy with the underlying sentiment, and that “ad hoc” decisions by some officer to take a knee create pressures or tensions for those who, for whatever reason, do not wish to do so. Moreover, the “results” of these actions, in terms of the positive influence on crowd members or effectiveness in de-escalation, are mixed. We respect the decisions of those officers who choose to symbolically demonstrate their respect for the protest movement, whether it emerges from genuine solidarity, sensitivity to community dynamics, or even just a willingness to make a concession in order to help defuse a tense situation. At the same time, we recognize that the lack of clear guidance or thoughtful training for officers in how to respond is a gap deserving of attention. Accordingly, we encourage SJPD to give more thought to a coordinated response to this or similar issues in the future. RECOMMENDATION 31 SJPD should explore training options and “in field” strategies to enhance officers’ ability to communicate effectively with the public in the context of protest or demonstration activity, and to avoid lapsing into unprofessional commentary or behavior. RECOMMENDATION 32 SJPD should focus additional attention on developing its response to protest-related interactions with the public in the context of “taking a knee” or similar shows of community solidarity, and should provide its officers with the tools to communicate effectively with protesters over the issue.
Page| 91
Equity Considerations One component of our scope of work for this project was to examine how different racial or ethnic groups were affected by the decisions or actions taken by SJPD and City leadership from May 29 to June 7. Sometimes referred to as an “equity lens analysis,” this type of evaluation is intended to provide a more inclusive perspective of events by considering the underlying assumptions and impacts of decisions and actions. Traditionally, an equity lens evaluation includes detailed analysis of large enough data sets to produce statistically significant results.51 But, as we discuss earlier in this Report, and as we discovered in our evaluations of other jurisdictions, the limited reporting and accurate data collection during the period of unrest did not generate a sufficiently robust data set. For example, uses of force, one area that might garner an equity lens evaluation (e.g., “did officers use more force on people of color?”) rarely had an identifiable individual upon whom force was used.52 Because of this, our analysis for San José goes well beyond the numbers. Throughout this Report, we evaluate how various aspects of the response may have impacted the San José community; for example, we discuss implications of the City-wide curfew and the undue burden that it may have created for people of color who might commute to jobs at irregular hours. And, we have always found that a forward-looking approach to equity has the most impact on an organization. We make recommendations throughout this Report for organizational or policy changes that are deliberately inclusive. And we are aware that the City is actively engaged in various projects meant to reimagine and reframe policing with an eye toward equity and inclusion. 51 “Statistically significant” means that the outcome of something was caused by something other than random chance. 52 As we previously noted, SJPD had challenges in their use of force counts, and rarely identified unique “victims” of force. As such, with respect to use of force data, and, in fact, the nature of this incident (for example, deployment of tear gas of either kind impacts indiscriminately and without identifiable, unique victims), did not allow us to quantify identifiers, such as age, sex, or race of the victims of force.
92 | Page
These reviews are intended for two fundamental goals: as ways to uphold agency standards through public accountability, and as vehicles to enhance agency operations through identification of issues and concerns that evidence room for potential improvement.
Page| 93
Conclusion The aftermath of critical incidents involving law enforcement often unfolds in phases. There is an initial wave of public attention and concern that can, among other things, serve as an impetus for accountability and broader reform. That dynamic certainly applies to San José. Understandably, though, many of the processes that are needed to accomplish desired changes – such as thorough investigations into officer performance, or thoughtful reconsiderations of specific tactics, or the development of innovative new approaches to public safety, or the pursuit of more equity in the justice system – require longer term commitments. We know from our own experience that these later, more deliberate phases of response can often produce value that justifies the time they take. But the danger of lost momentum is real. When the focused energy of an initial movement fades or is re-directed toward new problems, the likelihood of constructive reform can be diminished. And when the complexities of a given situation emerge with more clarity, the need for effective dialogue and collaboration increases. To the credit of San José, the City and its residents have taken concrete steps to preserve and channel the positive energy that animated so many of the manifestations of the Black Lives Matter movement last summer. While some of those initiatives have moved forward by fits and starts, the City is nonetheless committed to hearing from its communities on issues of police reform. Much has been achieved, and there are ongoing projects to address not only new ideas about policing, but also larger social needs. A reckoning with the events of the protests themselves – and the SJPD efforts to respond to very unfamiliar challenges – is one component of that commitment. While more than a year has passed, there are specific questions from those 10 days that remain unresolved (including matters of civil litigation and individual officer conduct). As we say throughout the Report, we encourage the Department to give its internal review mechanisms all the attention that is warranted in relation to those issues.
94 | Page
Meanwhile, the evolution of SJPD policy, procedure, decision-making, tactics, and community relations in light of “lessons learned” should be a continuous process – both in relation to last year’s unrest and for all other aspects of its operations. The Department’s own Preliminary After Action Report, produced with impressive speed last summer and containing a significant amount of important information, featured a number of recommendations with which we concur, and which illustrate the value of rigorous self-scrutiny. Based on our review, our impressions of those days overlap in many – but not all – respects with those reflected in the Department’s AAR. We agree that the circumstances were extremely challenging, and that the level of active unrest and problematic behavior exceeded what we saw in other jurisdictions we reviewed – even as the breadth of sincere peaceful protest was also greater. We found several attributes in the performance of individual officers and in the agency’s willingness to shift strategies under pressure and “on the fly.” And we were struck by the cooperation, candor, and insight that was consistent in our discussions with a number of involved members of the Department. Their collective thoughtfulness was encouraging. At the same time, we identity places where we find greater room for SJPD to take “ownership” of some of the deficiencies that it acknowledges, particularly as they related to the initial days of the response. If inexperience and staffing issues were at play (and they were), so too were curious shortcomings in the amount and effectiveness of field supervision. And if the adversity – including physical danger – that confronted the officers in those days merits attention, so too do the bewilderment and resentment that police actions repeatedly produced among swaths of peaceful crowd members, and the injuries that resulted from specific force deployments. We mean for this Report to provide a useful independent perspective that will help shape SJPD’s approaches in the future. And, as we indicate above – and as the Department commendably articulated in its AAR – there are elements of our recommendations that will ideally be considered in conjunction with other City officials and stakeholders. We look forward to observing those ongoing discussions and their outcomes.
Page| 95
APPENDIX A: Recommendations 1: SJPD should maintain a roster of lieutenants for immediate call back, on-call if necessary, to fulfill the platoon leader role in response to critical incidents. 2: SJPD should ensure that all of its lieutenants are well-versed in the fundamentals of the Incident Command System and would be capable of serving as the Incident Commander for a given operation if necessary. 3: SJPD should revise its Mobile Field Force training to emphasize the need for self-discipline and unit cohesion and to include a message from the Chief regarding the Department’s performance expectations. 4: SJPD should provide public updates regarding its progress in providing the targeted, topic-specific training on Mobile Field Force and other crowd management concepts that it cited in its After Action Report. 5: SJPD and the City should engage with the community as it considers the circumstances required for a public protest to be declared an unlawful assembly. The resulting guidelines should be publicized in a way that provides City residents and stakeholders a clear understanding of the circumstances under which the SJPD will declare an unlawful assembly. 6: SJPD should revise the Department’s dispersal order script in the three predominant languages to include an explicit warning about arrest, force and gas should the order be defied, and clear instruction regarding the safe routes of departure. 7: SJPD should revise the Department’s Demonstrations and Civil Disturbances policy to include an explicit warning about arrest and use of force (including chemical munitions) should the order be defied, and clear instruction regarding safe routes of departure.
96 | Page
8: In crafting dispersal orders to instruct crowds about the routes of egress from a protest, SJPD policy and training should advise officers to consider conditions such as parking, traffic and officer deployment to ensure the feasibility and safety of any direction provided about dispersal routes. 9: SJPD should consider adding language to its demonstration and civil disturbances policy advising, when practicable, that the Department contact event organizers or participants with the goal of gaining voluntary dispersal prior to issuing formal dispersal orders. 10: SJPD should revise the Department’s Demonstrations and Civil Disturbances policy to include that when dispersal orders are given, they should also be published on various social media platforms immediately. 11: SJPD should ensure that briefings in advance of crowd control situations include reinforcement of the Department’s expectations regarding the reporting of uses of force. 12: SJPD should ensure that supervisors responsible for reviewing use of force reports ensure the reports are sufficiently detailed before approval. 13: SJPD should consider practices such as body-camera narration that might enhance officers’ ability to subsequently capture their observations and decision-making when reporting on dynamic, extended deployments. 14: SJPD should create a Department-wide inventory system specifically for less-lethal munitions so that the Department can track inventory and less lethal usage at any point in time. 15: SJPD should evaluate its practices and protocols governing the number and type of munitions issued to officers, including Special Operations personnel, and reinforce with officers the importance of uniformity and consistency in the number of rounds carried. 16: SJPD should examine the “improvised” uses of less lethal force tools through the lenses of accountability, advisability, and remediation, and should train and provide written guidance to officers on the appropriate uses of these tools.
Page| 97
17: SJPD should specifically examine the use of flashbangs on May 29 and June 5 and develop protocols for limiting future deployment in a crowd control context. 18: SJPD should examine the use of high-velocity rounds through the lenses of accountability, advisability, and remediation, and consider if these rounds should be deployed during crowd management operations, and if so, under what conditions. 19: SJPD should consider options for improving the dissemination and clarification of mid-operation policy changes, so that affected personnel in the field are suitably equipped to adapt as intended. 20: SJPD should continue to evaluate use of less lethal munitions in a crowd control context in future incidents, and if the more restrictive policy still results in injuries to non-assaultive individuals, should reconsider whether to ban their use in a crowd control context altogether. 21: The Department should develop a policy on mass arrest and booking procedures that establishes a clearly articulated plan for handling the range of necessary tasks and includes detailed information about necessary forms, staffing assignments and division of responsibilities. 22: City and Department leadership should engage in conversations with the Santa Clara County Sheriff’s Office concerning protocols for handling the release of individuals arrested and cited during protests or other mass gatherings, with the goal of establishing guidelines and expectations for the timing and location of releases. 23: The City should engage with its community in developing guidelines on whether and how curfews should be utilized, particularly when related to activity protected by First Amendment protections. 24: Guidelines should include clear expectations about how any future curfew orders will be communicated to the public, including assuring that they are communicated at least in the three predominant languages in San José. 25: SJPD and City leadership (specifically, the Deputy Managers assigned to manage the Emergency Operations Center) should continue to collaborate, especially around intelligence-sharing.
98 | Page
26: City leadership in the City Manager’s Office should consider collaborating with SJPD and other relevant agencies to create City-wide plans for managing civil unrest, especially spontaneous occurrences. 27: When requesting mutual aid, SJPD should specify the need for command-level personnel to be included in the request. 28: As the largest law enforcement agency in the County, SJPD should work with other agencies to develop uniformity among polices for use of force in a crowd control context. 29: When updating the mutual aid agreement with local partners, SJDP should develop a process for information-sharing that, at a minimum, requires assisting agencies to document and share information regarding incident reports, arrest reports, and uses of force. 30: The City should review its internal protocols for deciding upon, communicating, and effectuating curfew parameters, including the coordinated interplay among departments, review of timing and enforcement strategies, and use of the City’s existing public messaging systems. 31: SJPD should explore training options and “in field” strategies to enhance officers’ ability to communicate effectively with the public in the context of protest or demonstration activity, and to avoid lapsing into unprofessional commentary or behavior. 32: SJPD should focus additional attention on developing its response to protest-related interactions with the public in the context of “taking a knee” or similar shows of community solidarity, and should provide its officers with the tools to communicate effectively with protesters over the issue.