Investigative activities should not be based on a person’s:
- (a) race, ethnicity, national origin, gender, sexual orientation, religion, or other protected characteristic, unless these characteristics are part of a sufficiently specific suspect description; or
- (b) expressed or perceived belief, absent a plausible basis to conclude that the person is advocating conduct that poses a threat to public safety.
a. Use of protected characteristics. It is imperative that governmental activities, especially those involving law enforcement, not be contaminated by any sort of animus or bias. Many agencies have adopted the phrase “bias-based policing” to describe policing activities undertaken in whole or in part on the basis of a person’s race, gender, ethnicity, or other protected characteristic such as a person’s sexual orientation, gender expression, religion, disability, homelessness, or economic status. Investigative activity on the basis of a protected characteristic typically is prohibited unless the characteristic is part of a specific suspect description, in connection with an investigation of a specific crime or violation. Absent a specific suspect description, bias-based policing occurs if a person’s status or group identity is a factor in an investigatory decision, even if other factors are present. For example, if an officer pulls a driver over because the driver is speeding and because the driver is Latinx, it still is an example of biased policing, even though there is some other legitimate basis for the stop. Officers may rely on a person’s protected status when they are looking for a suspect matching a particular description—but even in these circumstances, officers should not take action based on that characteristic alone and should have specific evidence to identify a specific suspect rather than engaging in a dragnet of a wide range of individuals.
b. Belief and expression. Law-enforcement investigations based on individuals’ expressed or perceived political, religious, or other expressions pose a distinct set of concerns. On the one hand, any investigative activity has the potential to chill protected speech or association—and can create the perception that law enforcement is targeting individuals impermissibly for expressing unpopular views. At the same time, political, religious, or other expression—unlike race, ethnicity, or religious identity—sometimes may be tied together with, or be the stated basis for, advocating violent or criminal activity, and therefore may itself provide legitimate grounds for investigative follow-up. For example, if in the course of religious expression, or for religious motives, an individual announces an intent to commit a crime or to support another individual’s criminal activity, it may be reasonable for officers to conduct a preliminary inquiry to determine whether there is in fact any reason to think that the individual poses a credible threat that should be pursued further.
A number of agencies have developed policies that attempt to balance these two concerns by making clear that the proper focus of any investigation is on the potential for unlawful activity as opposed to mere expression, by limiting the sorts of investigative techniques that officers may use before they have reasonable suspicion or probable cause to believe that an actual offense has been or is about to be committed, and by limiting what data may be retained from any preliminary inquiry that officers undertake. In New York City, for example, officers are not permitted to undertake any sort of investigative activity based “solely on activities protected by the First Amendment,” but they may initiate a preliminary inquiry when “statements advocate unlawful activity, or indicate an apparent intent to engage in unlawful conduct . . . unless it is apparent, from the circumstances or the context in which statements are made, that there is no prospect of harm.”
1. Avoiding bias. As law-enforcement leaders themselves increasingly have recognized, it is essential that policing be conducted in an impartial manner—and that individuals and communities not be singled out on the basis of race, ethnicity, gender, sexual orientation, religion, or other protected characteristic. See International Association of Chiefs of Police, Bias-Free Policing (Nov. 1, 2003) (“[B]ias-free policing is a critical cornerstone for upholding professional ethics in law enforcement, is vitally important to strengthening public trust and confidence in our actions and responsibilities, and is an essential element in maintaining community support for tolerance and understanding of our actions as we perform our responsibilities as law enforcement officials.”), https://www.theiacp.org/resources/resolution/bias-free-policing-0; Al Baker, Confronting Implicit Bias in the New York Police Department, N.Y. Times, (July 15, 2018), https://www.nytimes.com/2018/07/15/nyregion/bias-training-police.html (noting that “elected leaders and chiefs of police have increasingly focused on [ ] implicit bias” and that “[i]f officers rely on stereotypes instead of facts, routine encounters can escalate or turn deadly”). Most major departments now have policies in place to prohibit “bias-based” policing—and some states have adopted laws requiring the same. See, e.g., Los Angeles Police Dep’t Los Angeles Police Department Manual § 345 (“Policy Prohibiting Biased Policing”); Chicago Police Department, General Order G02-04 Prohibition Regarding Racial Profiling and Other Bias Based Policing (2017); Seattle Police Dep’t, Seattle Police Department Manual Policy 5.140 (2019); San Francisco Police Department, General Order: Policy Prohibiting Biased Policing (May 4, 2011), https://www.sanfranciscopolice.org/sites/default/files/2018-11/DGO5.17%20Policy%20Prohibiting%20Biased%20Policing.pdf; N.J. Stat. Ann. § 2C:30-6 (making it a crime for a public servant to “intimidate or discriminate against an individual or group of individuals because of race, color, religion, gender, handicap, sexual orientation, or ethnicity”); Ark. Stat. Ann. § 12-12-1402 (prohibiting racial profiling in policing); Kan. Stat. Ann. § 22-4609 (making it unlawful to use “racial or other biased-based policing” in policing decisions).
At the same time, policing agencies, courts, and commentators alike have struggled at times to articulate precisely what constitutes “bias”—and the circumstances in which race or other protected characteristics may be used in police decisionmaking. Some, for example, have argued that race may legitimately be used as part of a “profile.” And courts have at times suggested that race may be used to justify a stop or arrest so long as it is not the “sole” or “primary” factor motivating the decision. See United States v. Avery, 137 F.3d 343, 355 (7th Cir. 2001); Brown v. City of Oneonta, 221 F.3d 329 (2d Cir. 1999); United States v. Travis, 62 F.3d 170 (6th Cir. 1995); Samuel R. Gross & Debra Livingston, Racial Profiling Under Attack, 102 Colum. L. Rev. 1413, 1416 (2002) (noting that some courts continue to follow the rule that a police practice is not racial profiling unless the police are relying solely on the basis of race); R. Richard Banks, Race-Based Suspect Selection and Colorblind Equal Protection Doctrine and Discourse, 48 UCLA L. Rev. 1075, 1077-1078 (2001) (“No court has treated law enforcement reliance on a race-based suspect description as a racial classification warranting strict scrutiny under the Equal Protection Clause.”).This approach, however, is inconsistent with traditional equal protection jurisprudence, which requires that any use of race in government decisionmaking be “narrowly tailored” to a compelling government interest. See Fisher v. Univ. of Texas, 570 U.S. 297, 307-308 (2013) (applying strict scrutiny to a university’s admission process that took the applicant’s race into account);Regents of Univ. of California v. Bakke, 438 U.S. 265, 299 (1978) (same); see also Evan Gerstmann & Christopher Shortell, The Many Faces of Strict Scrutiny: How the Supreme Court Changes the Rules in Race Cases, 72 U. Pitt. L. Rev. 1, 48 (2010) (“The Court’s insistence that there is no equal protection issue unless police or border profiling is based solely on race makes this area an extreme outlier in the field of equal protection.”). A profile is, almost by definition, both over- and under-inclusive, given that the vast majority of individuals of any race or generalized description are going to be innocent of any criminal activity. See Albert W. Alschuler, Racial Profiling and the Constitution, 2002 U. Chi. Legal F. 163, 215 (2002). Profiling disproportionately burdens these individuals by subjecting them to additional law-enforcement scrutiny for no other reason other than their race or ethnicity. This is unacceptable.
For this reason, many departments have revised their policies to make clear that race—as well as other protected characteristics—only may be used if they are part of a specific suspect description. See, e.g., Seattle Police Dep’t, Seattle Police Department Manual 5.140.3-POL (2019) (“Officers may take into account the discernible personal characteristics of an individual in establishing reasonable suspicion or probable cause, only when the characteristic is part of a specific suspect description based on trustworthy and relevant information that links a specific person to a particular unlawful incident.”); Los Angeles Police Dep’t, Los Angeles Police Department Manual § 345 (2020) (“Department personnel seeking one or more specific persons who have been identified or described in part by their race [or other protected characteristic] may rely in part on the specified identifier . . . only in combination with other appropriate identifying factors and may not give [it] undue weight.”), https://www.lapdonline.org/lapd_manual/new_page_1.htm. In addition to the specific characteristics listed here—including race, ethnicity, gender, sexual orientation, and religion—some agency policies list a variety of other characteristics as well, including age, disability, economic status, and veteran status. See, e.g., Seattle Police Dep’t Manual Policy 5.140 (2019) (age, disability status, economic status, familial status, gender, gender identity, homelessness, mental illness, national origin, political ideology, race, ethnicity, color, religion, sexual orientation, use of a motorcycle or motorcycle-related paraphernalia, veteran status), https://www.seattle.gov/police-manual/title-5—employee-conduct/5140—bias-free-policing; Chicago Police Department, Prohibition Regarding Racial Profiling and Other Bias Based Policing General Order G02-04 (race, ethnicity, color, national origin, ancestry, religion, disability, gender, gender identity, sexual orientation, marital status, parental status, military discharge status, financial status, lawful source of income); LAPD Manual § 345 (race, religion, color, ethnicity, national origin, age, gender, gender identity, gender expression, sexual orientation, disability); Admin. Code of the City of New York § 14-151, Bias Based Profiling (actual or perceived race, national origin, color, creed, age, alienage or citizenship status, gender, sexual orientation, disability, or housing status), https://www1.nyc.gov/site/cchr/law/biased-based-profiling.page.
Agencies also have encouraged officers to focus on behavior that is indicative of criminal activity, as opposed to an individual’s identity or self-presentation. See id. (defining bias-based policing as officer relying on personal characteristics, rather than behavior, of suspect); Seattle Police Dep’t Manual Policy 5.140 (2019) (“Law enforcement and investigative decisions must be based upon observable behavior or specific intelligence.”). Studies suggest that focusing on observable behavior can both reduce racial or other disparities, and also improve hit rates, thereby reducing unwarranted intrusions and making better use of police resources. See Sharad Goel, Justin M. Rao & Ravi Shroff, Precinct or Prejudice? Understanding Racial Disparities in New York City’s Stop-and-Frisk Policy, 10 Annals Applied Stat. 365, 381-387 (2016) (finding that a focus on observable behavior leads to higher hit rates and decreased racial disparities); Jeffrey Fagan, Terry’s Original Sin, 2016 U. Chi. Legal F. 43 (2016) (finding a deterrent effect on criminal activity when police focus on observable behaviors such as “casing,” actions indicative of engaging in a drug transaction, or actions indicative of violent crimes).
2. Avoiding reliance on expressed religious or political belief. It is a fundamental principle—enshrined in the First Amendment—that individuals may not be singled out by the government on the basis of their political or religious views. See Rosenberger v. Rector & Visitors of the Univ. of Va., 515 U.S. 819 (1995) (“When the government targets not subject matter, but particular views taken by speakers on a subject, the violation of the First Amendment is  blatant.”). Democracy requires ample space for political disagreement and dissent, which would be undermined if majorities could leverage the power of the state to stamp out minority views. See Alexander Meiklejohn, The First Amendment is an Absolute, 1961 Sup. Ct. Rev. 245, 254 (1961); Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 424 (1996).
At the same time, the words people use and the ideas they express can themselves pose a threat to public safety when used to incite violence or otherwise create a serious risk of harm. See Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (explaining that government may target speech that “is directed to inciting or producing imminent lawless action and is likely to incite or produce such action”); Schenck v. United States, 249 U.S. 47, 52 (1919) (“The most stringent protection of free speech would not protect a man in falsely shouting fire in a theatre and causing a panic.”). For this reason, political or religious expression—unlike racial, ethnic, or religious identity—sometimes may be used to justify investigation or enforcement. See, e.g., Lawrence Rosenthal, First Amendment Investigations and the Inescapable Pragmatism of the Common Law of Free Speech, 86 Ind. L.J. 1, 35 (2011) (“Politically motivated terrorism provides a particularly vivid example of the case for intelligence-gathering investigations predicated upon expression that is protected by the First Amendment.”); Michael Vitiello, The Nuremberg Files: Testing the Outer Limits of the First Amendment, 61 Ohio St. L.J. 1175 (2000) (using an anti-abortion group’s implicit threats of violence against abortion providers as an example of when political speech crosses the line and loses First Amendment protection). On the other hand, the mere fact that an individual belongs to a particular organization or espouses a specific ideology is not itself sufficient. See United States v. Robel, 389 U.S. 258, 265 (1967) (holding that a statute forbidding members of a Communist organization from working in a defense facility unconstitutional as it “establishes guilt by association alone, without any need to establish that an individual’s association poses the threat feared by the Government in proscribing it.”); Clark v. Library of Congress, 750 F.2d 89 (D.C. Cir. 1984) (applying strict scrutiny when FBI’s investigation of Library of Congress employee was based solely on that employee’s membership in socialist group).
Agencies have adopted a variety of measures to ensure that investigations steer clear of protected First Amendment activity while at the same time enabling officers to take the steps necessary to protect the public from harm. Many agencies now have specific policies for handing investigations predicated on First Amendment activities that make clear that any investigative activity—however limited—must be predicated on a plausible threat of unlawful conduct. See, e.g., New York Police Department, Patrol Guide Procedure No. 212-72 Appendix B (2018) (“When, however, statements advocate unlawful activity, or indicate an apparent intent to engage in unlawful conduct, particularly acts of violence, an investigation under these guidelines may be warranted, unless it is apparent, from the circumstances or the context in which the statements are made, that there is no prospect of harm.”); Los Angeles Police Dep’t, Los Angeles Police Department Manual § 271.45 (2019) (instructing officers that “activities that are generally protected by the First Amendment should not be reported as a [Suspicious Activity Report] unless additional facts and circumstances can be clearly articulated [to show] that the behavior observed is reasonably indicative of criminal activity”). Many agencies also require supervisor approval either to initiate an investigation, or to use more intrusive techniques, such as undercover officers or informants. See, e.g., Chicago Police Department, Investigations Directed at First Amendment-Related Information General Order G02-02-01 § VI (2012) (noting that certain methods of First Amendment information-gathering require supervisor approval as provided for in Special Order S02-02-01); Berkeley Police Dep’t, Law Enforcement Services Manual Policy 430.9 (2018) (requiring that, when practicable, officers must obtain approval from the chief of police before conducting investigations “into individuals or groups exercising Constitutionally protected First Amendment activities”). Finally, agencies have used a variety of mechanisms to track the status of all investigative activities, and to periodically assess their continued necessity. D.C. Code § 5-333.12 (requiring panel of at least three commanding officers to review continued necessity of investigations involving First Amendment activity every 90 days and mandating a yearly outside audit of compliance with rule for First Amendment investigations); San Francisco Police Dep’t, General Order: Guidelines for First Amendment Activities § 8.10.VI (mandating designation of member of police commission to ensure compliance with department policies regarding First Amendment investigations and requiring annual audit of the Department’s compliance); Santa Cruz Police Dep’t, Policy Manual 606.4 (requiring police chief’s re-approval of First Amendment investigations every 30 days).