For more than fifty years, the problems endemic to municipal policing in the United States—brutality, racial discrimination, corruption, and opacity—have remained remarkably constant. This has occurred notwithstanding the advent of modern constitutional criminal procedure and countless judicial opinions applying it to the police. The municipal police can evade criminal procedure’s legality-based paradigm through formal and informal means. That paradigm presupposes that the police’s primary role is fighting crime, the zealous pursuit of which leads them to violate civil rights. The history and sociology of American policing, however, suggest that courts and law scholars have misconceived the municipal police. They are not, in the main, fighters of crime. They are guarantors of a social order that benefits dominant groups. This Essay advances an original descriptive account of the municipal police: They are “street sovereigns” whose power derives from law but cannot be contained by it. Police have the power to derogate from law as necessity requires, and it is the police themselves who usually have final say as to what constitutes a necessity. Police use of force and plainclothes tactics illustrate the descriptive theory. The theory suggests that law cannot restrain police in the way that American courts and commentators typically think it can. But law nonetheless remains important as an enabler of popular and institutional resistance to police abuses.


    1. Legality’s Three Core Assumptions
      1. There Is a Crime-Control “System”
      2. Courts Make Crime-Control Tradeoffs
      3. Police as “Street Magistrates”
    2. A Short History of Municipal Policing from the Nineteenth Century Until the Present
      1.  “Crime Control,” After the Fact
      2. The Dangerous Classes as Continuing Preoccupation
    3. History and Law, Disjointed
    1. Excessive Force
    2. The Sociology of Police Violence
    3. The Thin Blue Paradox
      1. Sovereignty and the State of Exception
      2. Street Sovereigns
    4. Plainclothes Policing and Legality’s Subversion
      1. Plainclothes Policing
      2. Entrapment and Sovereign Prerogative



Nothing works. History warrants cynicism among those who study the police. 1 See, e.g., Joshua Chanin, Evaluating Section 14141: An Empirical Review of Pattern or Practice Police Misconduct Reform, 14 Ohio St. J. Crim. L. 67, 111 (2016) (concluding that federal oversight under 42 U.S.C. § 14141 has little lasting effect past the point that oversight ends); Cynthia Lee, Reforming the Law on Police Use of Deadly Force: De-Escalation, Preseizure Conduct, and Imperfect Self-Defense, 2018 U. Ill. L. Rev. 629, 632 (noting that despite the public attention, shootings keep occurring); Udi Ofer, Getting It Right: Building Effective Civilian Review Boards to Oversee Police, 46 Seton Hall L. Rev. 1033, 1039 (2016) (noting the pervasive failure of civilian review boards). Five years after the Ferguson protests, the number and demographic profile of officer-involved shooting deaths have remained roughly constant. 2 See John Sullivan, Zane Anthony, Julie Tate & Jennifer Jenkins, Nationwide, Police Shot and Killed Nearly 1,000 People in 2017, Wash. Post (Jan. 6, 2018), https://www.washingtonpost.com/investigations/nationwide-police-shot-and-killed-nearly-1000-people-in-2017/2018/01/04/4eed5f34-e4e9-11e7-ab50-621fe0588340_story.html (on file with the Columbia Law Review) (“Experts said they are uncertain why the annual total shows little fluctuation—the number for 2017 is almost identical to the 995 killed by police in 2015.”). This occurred despite vigorous advocacy for more stringent rules constraining police violence and harsher repercussions when the rules are violated. 3 See Lee, supra note 1, at 632 n.3 (listing the “plethora” of law review articles advancing such proposals in recent years). And brutality is just one of many abuses endemic to municipal policing in the United States. Discriminatory enforcement against minorities, 4 See, e.g., Paul Butler, Chokehold 1–7 (2017) (cataloging different ways that police are harsh on black men); Jeffrey Fagan, Anthony A. Braga, Rod K. Brunson & April Pattavina, Stops and Stares: Street Stops, Surveillance, and Race in the New Policing, 43 Fordham Urb. L.J. 539, 599, 601 (2016) (showing racial disparity in the Boston Police Department’s use of stop and frisk and in surveillance of blacks and Hispanics); Rob Voigt, Nicholas P. Camp, Vinodkumar Prabhakaran, William L. Hamilton, Rebecca C. Hetey, Camilla M. Griffiths, David Jurgens, Dan Jurafsky & Jennifer L. Eberhardt, Language from Police Body Camera Footage Shows Racial Disparities in Officer Respect, 144 Proc. Nat’l Acad. Sci. 6521, 6521 (2017), https://www.pnas.org/content/pnas/114/25/6521.full.pdf [https://perma.cc/27VC-L7HD] (concluding that officers are less respectful to black drivers than white drivers during traffic stops). corruption, 5 See, e.g., Al Baker & William K. Rashbaum, 4 New York Police Leaders Reassigned Amid Corruption Inquiry, N.Y. Times (Apr. 7, 2016), https://www.nytimes.com/2016/04/08/nyregion/4-new-york-officers-are-put-on-desk-duty-amid-us-corruption-inquiry.html (on file with the Columbia Law Review) (reporting that four high-ranking New York Police Department (NYPD) officers were reassigned to desk jobs as a response to a federal corruption investigation); Justin Fenton & Tim Prudente, Commissioner Davis Says Plainclothes Policing in Baltimore Is Over, Balt. Sun (Mar. 8, 2017), https://www.baltimoresun.com/news/crime/bs-md-ci-plainclothes-policing-ends-20170308-story.html [https://perma.cc/T4RN-BQZ3] (describing how “seven members of the department’s Gun Trace Task Force were indicted on federal racketeering charges”); Jason Meisner, Jeremy Gorner & David Heinzmann, Chicago Cops Stripped of Powers as FBI Probes Ripoffs of Drug Dealers, Sources Say, Chi. Trib. (Feb. 1, 2018), https://www.chicagotribune.com/news/breaking/ct-met-chicago-cops-stripped-fbi-sting-20180131-story.html [https://perma.cc/EJ5D-9UGY] (reporting that police officers stole from drug dealers). and the lack of political transparency and accountability 6 See Barry Friedman & Maria Ponomarenko, Democratic Policing, 90 N.Y.U. L. Rev. 1827, 1843–49 (2015) (noting that police agencies are not subject to the same rules of transparency as other executive agencies); Erik Luna, Transparent Policing, 85 Iowa L. Rev. 1107, 1117 (2000) (criticizing police departments’ “undemocratic opaqueness”). are examples of other serious criticisms recently leveled against police departments. Today’s ills echo those of the 1960s, when municipal police were also the object of sustained public criticism. 7 See, e.g., Nat’l Advisory Comm’n on Civil Disorders, Report of the National Advisory Commission on Civil Disorders 302–05 (1968) [hereinafter Kerner Report] (describing problems of brutality and racial discrimination); see also, e.g., Kenneth Culp Davis, Police Discretion 29 (1975) (noting the invisibility of police discretion in the official record); Patrick V. Murphy & Thomas Plate, Commissioner: A View from the Top of American Law Enforcement 139, 146–70 (1977) (describing endemic corruption in the NYPD in the 1960s and 1970s). Since then, there have been many state and federal reforms, including the advent of modern constitutional criminal procedure. 8 See Michael J. Klarman, Rethinking the Civil Rights and Civil Liberties Revolutions, 82 Va. L. Rev. 1, 63–64 (1996) (providing a historical account of the “criminal procedure revolution” of the 1960s). But the reforms have not dramatically changed municipal policing. And today’s police reform proposals—more restrictive rules governing the exercise of police discretion, 9 See, e.g., President’s Task Force on 21st Century Policing, Final Report of the President’s Task Force on 21st Century Policing 20–21 (2015), https://cops.usdoj.gov/pdf/taskforce/taskforce_finalreport.pdf [https://perma.cc/ZJN2-4FNE] [hereinafter 21st Century Policing] (recommending that law enforcement agencies create “comprehensive policies on the use of force”); David Alan Sklansky, Democracy and the Police 33 (2008); see also Lee, supra note 1, at 664 (proposing model legislation on police use of deadly force). better officer training, 10 See 21st Century Policing, supra note 9, at 51–59 (recommending a variety of police trainings). more diverse officers, 11 See id. at 16 (“Law enforcement agencies should strive to create a workforce that contains a broad range of diversity including race, gender, language, life experience, and cultural background to improve understanding and effectiveness in dealing with all communities.”). greater responsiveness to civilian complaints, 12 See id. at 22 (recommending the establishment of a “Serious Incident Review Board”). and more robust community engagement and communication 13 See id. at 2–3 (recommending the adoption of strategic community engagement efforts). —echo reforms proposed more than fifty years ago. 14 See Kerner Report, supra note 7, at 311–20 (recommending the creation of a specialized agency to handle citizen complaints, more stringent guidelines governing citizen contacts, increased minority recruitment, and improved community relations efforts); see also The President’s Comm’n on Law Enf’t & Admin. of Justice, The Challenge of Crime in a Free Society 100–04, 112–13 (1967), https://www.ncjrs.gov/pdffiles1/nij/42.pdf [https://perma.cc/EAF2-6FMB] [hereinafter Challenge of Crime] (recommending improving community relations and communication efforts, increasing minority recruitment, and creating appropriate procedures for processing civilian complaints, clearer departmental policies regarding police exercise discretion, and improved officer training).

For many commentators, the persistence of the same problems (and solutions) is just a sign that reforms are incomplete. 15 See Rachel Harmon, The Problem of Policing, 110 Mich. L. Rev. 761, 784–85 (2012) (“[L]egal scholars write about constitutional law, and according to legal scholars, the Constitution continues to be the primary means for regulating the police.”). Legal commentators continue to conceive of the police as a species of executive power susceptible to the traditional forms of legislative and judicial control that “legality” presupposes. 16 See Challenge of Crime, supra note 14, at 91–93; Friedman & Ponomarenko, supra note 14, at 1831–32 (“[I]t has largely been left to the courts to govern the police.”). Legality suggests that conduct rules specified in advance and enforced post hoc can restrain the police. 17 This is the textbook definition associated with criminal law’s “legality principle.” See Joshua Dressler, Cases and Materials on Criminal Law 92 (5th ed. 2009). Constitutional criminal procedure’s restraints on the police can be understood as “a species of substantive criminal law for the police.” Carol S. Steiker, Counter-Revolution in Constitutional Criminal Procedure? Two Audiences, Two Answers, 94 Mich. L. Rev. 2466, 2534 (1996). But after generations of persistent “lawlessness in law enforcement,” 18 See generally Nat’l Comm’n on Law Observance & Enf’t, Report on Lawlessness in Law Enforcement (1931) [hereinafter Wickersham Commission] (analyzing the extent of “lawlessness in law enforcement”). we should question whether legislatures and courts are even capable of regulating the police as legality presupposes. 19 See Robert Weisberg, Criminal Law, Criminology, and the Small World of Legal Scholars, 63 U. Colo. L. Rev. 521, 532 (1992) (suggesting a scarcity of new theoretical insights by Fourth Amendment scholars since Herbert Packer).

Courts and legal scholars get off on the wrong conceptual foot by assuming that the police are, first and foremost, agents of criminal law enforcement. 20 See Egon Bittner, The Functions of Police in Modern Society 42 (1970) (“Because the idea that the police are basically a crimefighting agency has never been challenged in the past, no one has troubled to sort out the remaining priorities.”). For an expression of the dominant view of the policing function, see Challenge of Crime, supra note 14, at 92 (“In society’s day-to-day efforts to protect its citizens from . . . crime, the policeman occupies the front line.”). Legal scholars have recognized that patrol officers spend much of their time on noncriminal matters. See Debra Livingston, Police, Community Caretaking, and the Fourth Amendment, 1998 U. Chi. Legal F. 261, 263. But, even when they do, they tend to treat the noncriminal work as a less significant adjunct to criminal work. See infra section I.A.1. Constitutional criminal procedure, for example, frames the central problem in policing as that of officers getting carried away in the pursuit of criminals and thereby running roughshod over civilians’ rights. Justice Jackson famously expressed the idea:

The point of the Fourth Amendment . . . often is not grasped by zealous officers . . . . Its protection consists in requiring that [factual inferences regarding crime’s occurrence] be drawn by a neutral and detached magistrate instead of being judged by the officer engaged in the often competitive enterprise of ferreting out crime. 21 Johnson v. United States, 333 U.S. 10, 13–14 (1948). This language is regularly cited in more recent opinions. See, e.g., Illinois v. Gates, 462 U.S. 213, 240 (1983); Mincey v. Arizona, 437 U.S. 385, 395 (1978).

Courts correct for this zealousness by suppressing illegally obtained evidence after the fact, which is supposed to regulate police behavior going forward. If measured by the volume of opinions and scholarly literature, 22 See William J. Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 Yale L.J. 1, 17 (1997) (“With respect to police misconduct, constitutional criminal procedure occupies the field.”). constitutional criminal procedure would seem to be a success. Many of its rules penetrate deeply into the moment of contact between police officers and civilians. For example, an officer’s simple act of turning a stereo to look at a serial number, 23 See Arizona v. Hicks, 480 U.S. 321, 324–25 (1987). prodding luggage’s outer layer to feel for drugs, 24 See Bond v. United States, 529 U.S. 334, 338–39 (2000). or walking up to someone’s front door with a drug dog 25 See Florida v. Jardines, 569 U.S. 1, 9 (2013). triggers constitutional scrutiny.

These reams of cases pertain to a relatively narrow band of police conduct. 26 Others have noted this point. See Harmon, supra note 15, at 784–85 (“[S]cholars do not write much about police conduct that does not end up at issue in criminal cases, such as the misuse of force.”). And the thin and missing patches in constitutional criminal procedure’s regulatory quilt correspond to American policing’s gravest ills. The bulk of a typical constitutional criminal procedure course is devoted to the Fourth Amendment’s regulation of police searches. 27 In one of the most popular criminal procedure textbooks, eighty-six of the 374 pages devoted to the Fourth Amendment relate to exceptions to the warrant requirement, which is only one aspect of police searches. See Ronald Jay Allen, William J. Stuntz, Joseph L. Hoffmann, Debra A. Livingston & Andrew D. Leipold, Comprehensive Criminal Procedure 449–535 (3d ed. 2011). A student might expect that there would be a comparably long module on what police can and cannot do to civilians’ bodies. But the cases about bruises, broken bones, and other police-inflicted wounds never materialize. 28 See id. at 630–40 (devoting only eleven of 374 pages on the Fourth Amendment to police use of force). There is no remedy for excessive force in criminal cases, 29 See id. the procedural context in which most Fourth Amendment issues are litigated, 30 See Harmon, supra note 15, at 773–74. Civil remedies are available for police violations of constitutional rights, but formal and practical constraints limit their availability. See id. at 784–85 (discussing how civil litigation of police violence is often prohibitively costly); see also infra notes 83, 235–256 and accompanying text. because there is no evidence to suppress. 31 See Nirej Sekhon, Dangerous Warrants, 93 Wash. L. Rev. 967, 993–97 (2017) [hereinafter Sekhon, Dangerous Warrants] (“As a practical matter, the fruits of a search conducted following arrest represent the only vehicle for challenging many arrests’ constitutionality.”); cf. Kenneth W. Starr & Audrey L. Maness, Reasonable Remedies and (or?) the Exclusionary Rule, 43 Tex. Tech L. Rev. 373, 374–75 (2010) (discussing how the suppression of evidence provides numerous benefits to criminal defendants).

There are also important swathes of policing about which constitutional criminal procedure is almost totally hushed. 32 See, e.g., Nirej Sekhon, Redistributive Policing, 101 J. Crim. L. & Criminology 1171, 1172–73 (2011) [hereinafter Sekhon, Redistributive Policing] (explaining how criminal procedure does not regulate departmental discretion). For example, it does not address tactics like plainclothes policing, which has been used aggressively against street crime since the 1970s. 33 See Gary Marx, Undercover: Police Surveillance in America 46 (1988) (noting the extent to which criminal procedure rules regarding arrests incentivized greater use of deception in American policing); see also Street Crime in America (The Police Response): Hearings Before the H. Select Comm. on Crime, 93d Cong. 22 (1973) [hereinafter Street Crime in America] (statement of Anthony M. Voelker, Deputy Chief Inspector, New York City Police Department) (advocating for the use of plainclothes police); Charles Beene, Decoy Ops: Fighting Street Crime Undercover 29 (1992) (“New York City began to experiment with a new type of undercover police unit in 1971.”). Plainclothes officers function more as spies than uniformed police, gathering intelligence and using trickery to induce crime. 34 See Marx, supra note 33, at 61–88 (describing types of undercover operations). It is rife with opportunities for “zealous officers” to get carried away by perpetrating crimes themselves, 35 See id. at 126–27 (suggesting that fencing sting operations may create more crime); see also Elizabeth Joh, Breaking the Law to Enforce It: Undercover Police Participation in Crime, 62 Stan. L. Rev. 155, 165–68 (2009) (discussing how undercover officers may engage in crime to avoid suspicion and how some “‘rogue cops’ leave the bounds of authorized criminality and become mere criminals themselves”). creating risk of injury to civilians and police, 36 See Marx, supra note 33, at 177–79. and unfairly targeting minority groups. 37 See J. Kelly Strader & Lindsey Hay, Lewd Stings: Extending Lawrence v. Texas to Discriminatory Enforcement, 56 Am. Crim. L. Rev. 465, 469–70 (2019) (describing police stings and decoys used against LGBTQ persons); Jason Meisner, Federal Judge Finds ATF Drug Stash House Stings Distasteful but Not Racially Biased, Chi. Trib. (Mar. 13, 2018), https://www.chicagotribune.com/news/breaking/ct-met-atf-drug-stash-house-ruling-20180309-story.html [https://perma.cc/84ZV-3GME] (explaining how a court ruled that the ATF’s sting operations had “an ugly racial component and should be discontinued”).

That constitutional courts’ eyes are averted from a good bit, if not most, of what the police do reflects the vast gap separating legality from the history and sociology of American policing. Municipal policing in the United States was not conceived as a response to crime. The police were conceived as a tool for managing those segments of the lower classes that the upper and middle classes found threatening. 38 See infra section I.B. It was not until the mid-twentieth century that the notion of police as crime-control agents gained serious traction. 39 See Sklansky, supra note 9, at 44–46 (discussing scholars promoting the idea that the police identify with the “rule of law”). Police policymakers advanced this account to allay public anxieties about police abuses that made the American police seem a little too similar to the totalitarian governments just defeated in World War II. 40 See id. at 18 (“Contrasting democracy with the ‘police state’ therefore placed at the heart of pluralism certain ideas about the police, and certain implications about how the police should be ‘reconciled’ with democracy.”); see also Michael J. Klarman, The Racial Origins of Modern Criminal Procedure, 99 Mich. L. Rev. 48, 83 (2000) [hereinafter Klarman, Racial Origins] (“Popular revulsion against Nazi and Stalinist law enforcement methods further contributed to the demise of such practices by the mid-1930s, as did the work of the Commission on Interracial Cooperation and other southern liberal organizations . . . .”). The idea of American police as primarily crime fighters was a palliative that police officers and the public embraced, 41 Bittner, supra note 20, at 40–42 (describing the overwhelming view that police are a “crimefighting agency”). as did courts and commentators. No matter that it was and remains an inaccurate gloss on the police.

A defining feature of the municipal police in the United States is that they serve as the public agencies of first and last resort for a range of social problems, few of which are criminal law violations. 42 See id. at 43 (“Many, perhaps most [police activities], consist of addressing situations in which people simply do not seem to be able to manage their own lives adequately.”); see also Peter K. Manning, Police Work: The Social Organization of Policing 26 (2d ed. 1997) (explaining that police work is, in the main, neither “crime- nor law- relevant, but represent[s] various kinds of public and private troubles about which the individual citizen can do nothing”). Whether the exigency is imagined or real, 43 See, e.g., Elizabeth Dunbar, Authorities Say They Can’t Second-Guess a 911 Call, Minn. Pub. Radio (July 13, 2018), https://www.mprnews.org/story/2018/07/13/authorities-say-they-cant-second-guess-a-911-call [https://perma.cc/V374-LCU6] (describing police officers responding to a fake 911 call). human or animal, 44 See, e.g., Zolan Kanno-Youngs & Ben Kesling, Look Out—It’s New York City’s Bee Cop! And He’s Got a Vacuum Cleaner, Wall St. J. (Aug. 1, 2017), https://www.wsj.com/articles/swarmed-and-dangerous-new-york-citys-bee-cop-collars-perps-with-a-vacuum-cleaner-1501600272 (on file with the Columbia Law Review); Daniel Uria, Pennsylvania Police Remove 3-Foot Alligator from Resident’s Yard, United Press Int’l (July 1, 2018), https://www.upi.com/Odd_News/2018/07/01/Pennsylvania-police-remove-3-foot-alligator-from-residents-yard/2181530484994/ [https://perma.cc/SYB6-GRL9]. or act of nature, 45 See Daphne E. Whitmer, Valerie K. Sims, Shannon K.T. Bailey & Bradford L. Schroeder, Time to Decide: To Call or Not to Call 911 During Weather Crises, 60 Procs. Hum. Factors & Ergonomics Soc’y Ann. Meeting 1160, 1162 (2016) (finding that survey participants were likely to call 911 during a tornado). Americans “call the cops.” 46 See Bittner, supra note 20, at 43. The “catchall tradition” in American policing means that the police serve as the public face of the state in a wide variety of contexts, but particularly on the streets. 47 See Robert Fogelson, Big-City Police 108 (1977) (describing the persistence of the “catchall tradition” in American policing). The ironic implication of this is that the official who shows up to prepare an accident report, do a welfare check, or resolve a street dispute is legally authorized to kill. That power helps ensure that the police’s assessment and resolution of problems is final, including when they use violence. 48 See infra section I.B. Most police–civilian contacts do not generate criminal cases in which prosecutors and courts get involved, nor are most police abuses susceptible to meaningful post hoc review by any entity other than perhaps the police themselves. 49 See infra section I.B.

If the history and sociology of policing do not line up with our liberal precepts of divided government and legality, how should we make sense of the municipal police in the United States? The police are ubiquitous symbols of the state 50 Manning, supra note 42, at 20–21 (“[T]he police, to many audiences, represent the presence of the civil body politic in everyday life . . . .”). but do not themselves seem wholly subject to the liberal constraints that are supposed to define the state. Liberalism’s critics have long noted its failure to describe the nature of coercive power in modern states. 51 See infra section II.C.1. In this vein, Giorgio Agamben, relying on the earlier work of Carl Schmitt, has advanced a theory of “sovereign power.” 52 Giorgio Agamben, Homo Sacer: Sovereign Power and Bare Life 83 (Werner Hamacher & David E. Wellbery eds., Daniel Heller-Roazen trans., Stanford Univ. Press 1998) (1995) [hereinafter Agamben, Homo Sacer]. Carl Schmitt defined sovereign power as “the highest, legally independent, underived power” which is wielded by the actor who “decides whether there is an extreme emergency as well as what must be done to eliminate it.” 53 Carl Schmitt, Political Theology: Four Chapters on the Concept of Sovereignty 7, 12–17 (George Schwab trans., Univ. of Chi. Press 1985) (1922) [hereinafter Schmitt, Political Theology]. Sovereignty is the power to “decide[] on the exception,” or to decide when the rule of law must be suspended in the face of an extreme emergency. 54 Id. at 5 n.1 (translator’s commentary). Necessity (rather than law) governs in the state of exception. 55 Id. “Sovereign power” elucidates the role of municipal police in the United States and the failure of our legality-based paradigm to constrain as liberalism suggests it should.

Municipal police are best understood as sovereigns of the street. The “thin blue line” metaphor used by police and their boosters is a colloquial shorthand for this idea: The police’s power underwrites, but cannot be subsumed by, our sociopolitical order. 56 See Alice Ristroph, The Thin Blue Line from Crime to Punishment, 108 J. Crim. L. & Criminology 305, 305–06 & n.1 (2018) (“Policing is central to the operation of the modern criminal law, and yet, it has long been almost entirely ignored by criminal law theorists.”). Criminologist Egon Bittner, for example, understood the state to have invested the police with coercive power to resolve problems on the street with finality and based less on law than personal intuition and occupational experience. 57 Bittner, supra note 20, at 46 (“[T]he role of the police is best understood as a mechanism for the distribution of non-negotiably coercive force employed in accordance with the dictates of an intuitive grasp of situational exigencies.” (emphasis omitted)). The law gives way to whatever the exigency requires and it is the police’s understanding of what is required that is usually dispositive.

By this account, criminal procedure’s failure to adequately constrain police is not a correctable defect but rather a design feature. Agamben has argued that liberal states make gestures toward restraining sovereign power, but can only do so ambivalently. 58 See Giorgio Agamben, State of Exception 50–51 (Kevin Attell trans., Univ. of Chi. Press 2005) (2003) [hereinafter Agamben, State of Exception]. Purporting to extend legality’s reach to the police creates a “fiction[] through which law attempts to encompass its own absence and to appropriate the state of exception, or at least to assure itself a relation with it.” 59 Id. at 51. For example, consider the way that ordinary rule of law gives way to sovereign prerogative in “martial law” or a “war zone”—the legal order is suspended for the sake (however ironic) of its own preservation. It is no coincidence that the expression “war zone” is often invoked to describe poor minority neighborhoods in the United States. 60 See, e.g., Sarah Favot, Homicides Hard to Solve in this South Los Angeles ZIP Code Deemed ‘War Zone,’ Pasadena Star-News (Mar. 28, 2015), https://www.pasadenastarnews.com/2015/03/28/homicides-hard-to-solve-in-this-south-los-angeles-zip-code-deemed-war-zone/ [https://perma.cc/Y5FN-CXG3] (last updated Aug. 28, 2017); John Kass, ER Workers on Chicago Gang Violence: ‘We’re in a War Zone Too,’ Chi. Trib. (May 8, 2018), https://www.chicagotribune.com/columns/john-kass/ct-met-chicago-violence-kass-0509-story.html [https://perma.cc/BWQ7-TADJ]; Robert Snell, The Red Zone: Inside Detroit’s Deadly Gang Wars, Det. News (Apr. 22, 2018), https://www.detroitnews.com/story/story-series/death-by-instagram/2018/04/22/detroit-gang-war-red-zone/432776002/ [https://perma.cc/SS3S-Y6JC] (last updated Apr. 23, 2018).

Constitutional criminal procedure’s ambivalent relation to the police is exemplified by excessive force jurisprudence. The authority to deploy deadly violence is the defining feature of modern policing and underwrites the police’s capacity to resolve problems “non-negotiably.” 61 See Bittner, supra note 20, at 46 (emphasis omitted). Constitutional doctrine purports to regulate this power but that “regulation” functions as little more than a restatement of the police’s power to decide the exception. 62 See supra notes 21–26 and accompanying text. It allows the police to suspend the norm prohibiting their use of extrajudicial violence against a civilian when, from the police perspective, it is reasonably necessary to prevent the civilian from engaging in extrajudicial violence. 63 See Graham v. Connor, 490 U.S. 386, 397 (1989) (holding that an officer can constitutionally use excessive force in certain circumstances under the Fourth Amendment’s objective reasonableness standard). More importantly, police violence is rarely challenged in court for both formal and practical reasons. 64 See infra section II.A. This means that, for the vast majority of cases in which police use force, the law of excessive force has virtually no consequence. Such irony is characteristic of the state of exception. 65 See Agamben, State of Exception, supra note 58, at 60.

The police’s sovereign power is also suggested by the relative ease with which they can coopt legality’s strictures while ostensibly acting within those very strictures. Plainclothes policing exemplifies this dynamic. 66 For background information on plainclothes policing, see infra section II.D.1. In the 1970s, police were confronted with political pressure to generate more arrests leading to convictions. 67 See infra section II.D.1. Departments responded by simply inducing crime themselves. Plainclothes officers used strategic deceptions, sometimes violating criminal laws themselves, to lure suspects to commit crimes in the officers’ presence. 68 See Marx, supra note 33, at 65–67, 126–27 (detailing ways in which police engage in criminal activity when conducting undercover operations). This generated so-called “high-quality arrests” likely to yield convictions. 69 See Street Crime in America, supra note 33, at 22. Courts have little to say about such policing tactics, with the subconstitutional doctrine of entrapment as their primary tool. 70 See infra section II.D.2. Entrapment, like “excessive force,” is more a restatement of the police’s sovereign power than it is a restriction of it. It authorizes the use of decoys and stings if the targets are inclined toward criminality, a trait that is often attributed to the ranks of racial and other disfavored groups. 71 See infra notes 467–470 and accompanying text. The discussion below reveals that the police’s role as street sovereigns is most starkly realized in relation to marginal groups but is not limited to them.

Part I demonstrates that the related liberal ideals of legality and divided government, upon which constitutional criminal procedures rest, do not line up with the history and sociology of municipal policing. Part II advances a new descriptive theory of the municipal police that accounts for how and why police use violence. Part II also illustrates the theory using the example of plainclothes policing. Part III concludes by sketching the consequences of a sovereignty-based theory for law’s role in challenging policing abuses.

I. Legality, Crime Control, and the Police

Courts and commentators conceive of the police as wielding executive power in the manner suggested by our traditional notions of divided government. 72 See Friedman & Ponomarenko, supra note 6, at 1844 (noting that “[t]he typical enabling statute of a policing agency simply authorizes it to enforce the substantive criminal law”). This casts the police as amenable to legality-based restraints—that is, judicially-enforced conduct rules that have been pre-specified by the other two branches of government. 73 See supra note 17 and accompanying text. Section I.A below identifies the core assumptions of the dominant legality paradigm. That paradigm takes the police’s central mission as enforcing criminal laws, 74 Law scholars seem to view as obvious the fact that police exist primarily (if not exclusively) to control crime. See, e.g., Friedman & Ponomarenko, supra note 6, at 1831– 32 (stating that police are given free rein because of their role in keeping crime “in check”); David M. Jaros, Preempting the Police, 55 B.C. L. Rev. 1149, 1149 (2014) (“The practicalities of fighting crime require that we vest the police with extensive discretion . . . .”); Tom R. Tyler & Je rey Fagan, Legitimacy and Cooperation: Why Do People Help the Police Fight Crime in Their Communities?, 6 Ohio St. J. Crim. L. 231, 233 (2008) (“To be e ective in lowering crime and creating secure communities, the police must be able to elicit cooperation from community residents.”). which sometimes leads them to run roughshod over civil rights by, for example, “thrust[ing] themselves into a home.” 75 Johnson v. United States, 333 U.S. 10, 14 (1948). Courts then must balance the harm of such intrusions against the value of crime control. Our traditional legality-based paradigm assumes that there is a system of criminal justice that police and courts co-inhabit that allows the latter to speak authoritatively to the former.

The problem is that the paradigm does not line up with the history and sociology of municipal policing in the United States. Section I.B shows that the municipal police were created and continue to exist for the purpose of controlling socially disfavored groups, not to fight crime. The latter rationale evolved only in the mid-twentieth century as a calculated political strategy to bolster and rationalize police legitimacy. 76 See infra notes 176–180 and accompanying text. This occurred at the same time that police agencies became more bureaucratic and insulated from political and judicial actors. 77 See infra notes 183–186 and accompanying text. This history suggests that there is no unified criminal justice system as we ordinarily think of it, nor do courts speak authoritatively to the police.

A.  Legality’s Three Core Assumptions

For all their criticism of constitutional criminal procedure, most scholars leave legality’s basic premises intact. The dominant view among scholars of why criminal procedure has failed to restrain the police is that an increasingly conservative Supreme Court put the brakes on the Warren Court’s “revolution in criminal procedure.” 78 See, e.g., Eric J. Miller, The Warren Court’s Regulatory Revolution in Criminal Procedure, 43 Conn. L. Rev. 1, 3, 9–10 (2010) (characterizing this as a “standard” scholarly narrative). Some, such as the late William Stuntz, have argued that the fault lies with the Warren Court for tying criminal procedure to the Bill of Rights, which is not expansive enough to regulate the police. 79 William J. Stuntz, The Collapse of American Criminal Justice 227–28 (2011) [hereinafter Stuntz, The Collapse]. Others have argued that giving full effect to the literal meaning of the Fourth Amendment’s text would do the trick. For example, Jed Rubenfeld contends that the Court should do more with the word “security” to remedy policing’s ills. 80 Jed Rubenfeld, The End of Privacy, 61 Stan. L. Rev. 101, 119–22 (2008). Even for legal scholars who study the municipal police, there is an abiding faith in legality, even if not of a specifically constitutional flavor. 81 See, e.g., Friedman & Ponomarenko, supra note 6, at 1877–78 (asserting that rule-based regulation of police is so readily possible that it is “parochial” to think otherwise); Erik Luna, Principled Enforcement of Penal Codes, 4 Bu . Crim. L. Rev. 515, 590 (2000) (recommending “the open adoption of sub-statutory, sub-constitutional principles to limit and guide executive discretion”); Christopher Slobogin, Policing as Administration, 165 U. Pa. L. Rev. 91, 134–35 (2016) (advocating for an administrative law based approach for certain kinds of searches); see also Harmon, supra note 15, at 785–86 (noting that criminal procedure scholars are not naïve about courts’ limitations but still embrace a traditional court-centered approach to police).

This Essay argues that legality itself is part of the problem. Before turning a critical eye on legality, we should first identify its basic assumptions about policing and criminal justice: first, that there is a crime-control system composed of interlocking institutions with well-defined relations of authority; second, that the competing values at play in that system are commensurable; and third, that courts can induce police to behave as “street magistrates.”

1.  There Is a Crime-Control “System.” — Constitutional criminal procedure’s conduct rules are supposed to be judicially enforceable. But what compels the police to listen? In civil cases, it is the threat of damages or a contempt sanction. 82 See 42 U.S.C. § 1983 (2012); Bivens v. Six Unknown Named Agents of Fed. Bureau of Narcotics, 403 U.S. 388, 395 (1971) (“[D]amages may be obtained for injuries consequent upon a violation of the Fourth Amendment by federal officials. . . .”). Neither of these is a possibility in criminal cases and that is the context in which the vast majority of police misconduct is litigated. 83 Cf. Starr & Maness, supra note 31, at 375 (describing the overwhelming number of criminal cases addressing the exclusionary rule). While civil remedies are in principle available for violations of Fourth Amendment rights, there is little financial incentive for those whose rights have been violated to bring suit. For example, the privacy and liberty injuries associated with a brief, unconstitutional street stop will not typically justify the costs of bringing a civil suit. Nirej Sekhon, Mass Suppression: Aggregation and the Fourth Amendment, 51 Ga. L. Rev. 429, 454–57 (2017) [hereinafter Sekhon, Mass Suppression]. Because police are not obliged to explain the bases for a stop to the target, it may not be obvious to those stopped and searched that their rights were violated. See id. at 454 (“For most searches and seizures, the individual’s first formal opportunity to learn of its legal basis will come during litigation.”). The formal and practical constraints on bringing civil suits against the police for more egregious constitutional violations resulting in physical injury are described at length in section II.A.

Legality assumes that courts and police intimately coexist in a “system” of criminal justice that enables the former to speak authoritatively to the latter. Courts and commentators cast the police as the criminal justice system’s front line. 84 See, e.g., Allen et al., supra note 27, at 5 (“The initial enforcement responsibility rests with police agencies . . . .”). Because the police generate a substantial portion of the “inputs” to criminal courts, it seems like a reasonable inference that this is the police’s main function. 85 See Bittner, supra note 20, at 42 (explaining how this view leads police to characterize and justify activities that are not related to crime control as if they were). The police themselves have cultivated that impression with great public effect. 86 Cf. Sklansky, supra note 9, at 50–51 (noting police marketing efforts). Most observers view the police as the quintessential agents of criminal justice. 87 See John P. Crank, Institutional Theory of the Police: A Review of the State of the Art, 26 Policing 186, 189 (2003) (describing the “professionalism movement”); see also Bittner, supra note 20, at 42 (“[T]he idea that the police are basically a crimefighting agency has never been challenged in the past . . . .”). Sarah Mayeux recently critiqued the notion of a “criminal justice system” by tracing the history of the concept. 88 Sara Mayeux, The Idea of the Criminal Justice System, 45 Am. J. Crim. L. 55, 58–60 (2018). The term “system” denotes capacities for self-regulation and the production of an output. 89 Id. at 66, 69 (citing sociologist Talcott Parsons’s systems theory and describing the capacities of systems to act as self-regulating entities). It was in the 1960s that systems became vernacular in social science and public policy for making sense of the world bureaucratization had created. 90 Id. at 59 (“[T]he phrase ‘the criminal justice system’ spread wildly in the late 1960s when it was introduced to a generation of lawyers, policymakers, jurists, and social scientists . . . and helped to shape[] their world—as one grand system of systems.”). By the 1960s, courts, prosecutors, and police had become professionalized and were perceived as the key actors in producing criminal justice. 91 Id. at 73 (“As career police and prosecutors replaced part-time amateurs and states developed ever-more complex penal bureaucracies, every part of the process became ‘professionalized,’ and professionals, as they are wont to do, formed communities of pedagogy and practice . . . .”). It was these actors’ shared goal of controlling crime that supposedly unified them as a system. 92 Id. at 75–79 (discussing Challenge of Crime, supra note 14).

Although courts are not formally responsible for directing the municipal police, legality assumes that the shared institutional mission and vocabulary of crime control permits the former to speak authoritatively to the latter. 93 In justifying incorporation of the Fourth Amendment exclusionary rule against the states, the Supreme Court simply assumed that the judiciary’s regulatory reach extended to the police: “[W]e can no longer permit [the Fourth Amendment] to be revocable at the whim of any police officer who, in the name of law enforcement itself, chooses to suspend its enjoyment.” Mapp v. Ohio, 367 U.S. 643, 660 (1961). The Supreme Court’s assumption was elaborated upon in subsequent exclusionary rule cases. See infra notes 94–101 and accompanying text. The Supreme Court’s Fourth Amendment exclusionary rule jurisprudence reflects this understanding. Exclusion deprives the state of unconstitutionally seized evidence in the criminal case against a defendant. The defendant’s reward does not directly remedy the constitutional harm suffered, 94 See Davis v. United States, 564 U.S. 229, 236–37 (2011). but it represents an incidental cost of minimizing future constitutional externalities to the public. 95 See Sekhon, Mass Suppression, supra note 83, at 439 (arguing that “[t]he criminal defendant is simply a vehicle for ensuring that unnamed community members’ rights are not violated in the future”). “The [exclusionary] rule’s sole purpose . . . is to deter future Fourth Amendment violations.” 96 Davis, 564 U.S. at 236–37. Excluding wrongfully seized evidence in a criminal case is supposed to make a teachable moment of constitutional misconduct, serving as an astringent rebuke to the police rather than a formal punishment. 97 See id. at 237–38 (“When the police exhibit ‘deliberate,’ ‘reckless,’ or ‘grossly negligent’ disregard for Fourth Amendment rights, the deterrent value of exclusion is strong and tends to outweigh the resulting costs.” (quoting Herring v. United States, 555 U.S. 135, 144 (2009)). There are empirical debates about how effectively courts can influence police behavior and whether trying to do so is worth exclusion’s social cost in lost convictions. 98 See Christopher Slobogin, Comparative Empiricism and Police Investigative Practices, 37 N.C. J. Int’l L. & Com. Reg. 321, 332–35 (2011) (providing a summary of relevant comparative empirical studies and identifying key research areas for future scholarship on the exclusionary rule). But the theory is that because police are, first and foremost, crime-control agents, they perceive lost convictions as a professional rebuke to be avoided.

Legality’s emphasis on criminal evidence misses a lot. Most of the shiftwork by patrol officers yields no criminal evidence or arrests, let alone convictions. 99 Cf. Stephen D. Mastrofski & James J. Willis, Police Organization Continuity and Change: Into the Twenty-First Century, 39 Crime & Just. 55, 85 (2010) (arguing that standard patrol methods are best designed not to control crime but to “distribute fairly equitably a wide range of services” that the public values). Among other noncriminal work, police perform informal dispute resolution, aid the injured, and keep traffic moving. Twenty years ago, Professor and current-Judge Debra Livingston clumped these diverse functions together under the rubric of “community caretaking.” 100 See Livingston, supra note 20, at 271–72. She argued that the Fourth Amendment ought to regulate these activities less stringently because police are trying to help rather than harm the target. 101 Id. at 271–78. The Supreme Court took her suggestion in Brigham City v. Stuart. 102 See 547 U.S. 398, 406 (2006). Whereas probable cause is ordinarily required before an officer may enter a home to search for criminal evidence, an officer can enter to perform a welfare check if they reasonably think that someone inside is hurt and needs help. 103 Id. at 404. In Stuart, upon entering the premises, the officers discovered evidence of criminal wrongdoing in plain view. 104 Id. at 401. The Court deemed that evidence admissible against Stuart because the officers’ entry was a permissible community caretaking search. 105 See id. at 406–07.

Even though Stuart was about the police’s noncriminal functions, the Court still spoke to the police in terms of excluding criminal evidence. 106 See id. at 401, 406–07. This remedy does no good in the many cases in which unlawful community caretaking yields no criminal evidence. The premise in Stuart must be that exclusion of evidence will deter future violations (including those in which no criminal evidence is discovered) because exclusion prompts police to internalize the relevant constitutional standard. By using the admissibility of criminal evidence to regulate police behavior in other contexts, the Court implicitly casts “community caretaking,” however consuming, 107 See Livingston, supra note 20, at 263 (noting that municipal police spend “a good deal of time” on noncriminal matters). as incidental to the core function of crime control.

2.  Courts Make Crime-Control Tradeoffs. — Our legality paradigm takes it as courts’ role to trade off crime control with other values and generate legal norms for the criminal justice system. 108 See Challenge of Crime, supra note 14, at 93 (“The process of striking this balance is complex and delicate.”). This is to assume that the competing values to be “traded off” by courts are commensurable. For this to be true, there must be a conceptual register within which the exchange value of crime control can be fixed against the libertarian and process values embedded in the Constitution. This idea underwrites Herbert Packer’s iconic “crime control”–“due process” dualism. 109 Herbert L. Packer, The Limits of the Criminal Sanction 157 (1968) (“[C]ommon ground . . . should not be overlooked, because it is, by definition, what permits partial resolutions of the tension between the two models to take place.”).

The easy instances of commensurability are those in which due process values are coextensive with crime-control values—where constitutional rules promote the “repression of criminal conduct” by encouraging accurate processing of criminal cases. 110 See id. at 158. Packer tells us that the crime-control model favors the quick screening of suspects and defendants, relying heavily on the professional judgments of police and prosecutors. 111 Id. at 159–60. Crime control tolerates the mistaken conviction of factually innocent persons but only to the extent consistent with its underlying purpose of deterring crime. 112 Id. at 164–65. People might stop obeying criminal law if they thought the criminal process so unreliable that complying with criminal laws had little bearing on the likelihood of arrest and conviction. 113 Id. Procedural protections that promote accuracy in arrests and convictions can thus be consistent with the crime-control model. 114 See id. Commensurability is straightforward here because there is a shared unit of exchange: mistaken convictions.

When civil liberties and equality values unrelated to accuracy are folded into the juridical mix, commensurability becomes trickier. There must be a shared unit of exchange if tradeoffs can be coherently calculated. Packer’s “due process model” assumes that efficient crime control has the potential to generate harms beyond just factually incorrect convictions. 115 Id. at 165–66. Criminal law enforcement’s aggressively coercive bent creates authoritarian dangers for disfavored groups, if not everyone. 116 See id. at 166, 170–71, 180 (“Unease may be stirred simply by reflection on the variety of uses to which the criminal sanction is put and by a judgment that an increasingly large proportion of those uses may represent an unwise invocation of so extreme a sanction.”). Accordingly, the due process model suggests that courts should not defer to police and prosecutors. These actors will systematically discount privacy, liberty, and equality norms in favor of crime control. 117 See id. at 179–80 (“A totally efficient system of crime control would be a totally repressive one.”). The largest and most bitter forms of intrusion would be foisted upon society’s lowest socioeconomic strata, where policing is most intensive. 118 See id. at 168–69, 180 (explaining how those “who are least able to draw attention to their plight” typically suffer the most from police intrusions). Proponents of the crime-control model would likely argue that these are the very spaces within which the worst crimes are concentrated. 119 Cf. id. at 158 (suggesting that, in this model, law enforcement’s focus would be on those places where criminal misconduct most threatens to lead “to the breakdown of public order”). So how many crime-control units should be sacrificed for how many privacy, liberty, or equality units?

Packer, like most of the rest of us, understands it to be within courts’ prerogative to authoritatively reconcile due process with crime control values. 120 See id. at 173 (stating that “the authoritative force at work is the judicial power”). The problem is that there are no readily quantifiable proxies for the complex moral content that the words “privacy,” “liberty,” or “equality” purport to describe. The determinacy and precision suggested by the notion of judicial balancing is belied by the Court’s reliance on impressionistic, brushstroke characterizations rather than precise, mathematical exchange when making tradeoffs between due process and crime control. 121 See Scott E. Sundby, “Everyman” ’s Fourth Amendment: Privacy or Mutual Trust Between Government and Citizen?, 94 Colum. L. Rev. 1751, 1770 (1994) (“[T]he vast majority of cases will involve both legitimate government and privacy interests, yielding no black or white ‘right’ answers but only questions of which shade of grey is better.”); see also David Wolitz, Indeterminacy, Value Pluralism, and Tragic Cases, 62 Buff. L. Rev. 529, 566–71 (2014) (discussing Isaiah Berlin’s theory that it is impossible to assign relative weights to some moral values in the context of constitutional law). Perhaps for this very reason it is best that courts perform this role because it is ineluctably necessary, 122 See Wolitz, supra note 121, at 591–97 (“[J]udges can and should recognize the valid claims of both values and seek to ensure that both values remain vital . . . .”). and courts, unlike police or prosecutors, are obliged to try to render a logical reconciliation in public view.

Legality suggests that courts reconcile competing values into operational principles that apply as precedent in subsequent cases and, perhaps more importantly, guide how police are to wield their discretionary power in the field. 123 See Packer, supra note 109, at 180 (arguing that sanctions for illegal arrests “should be located within the criminal process itself[] because it is the efficiency of that process that they seek so mistakenly to promote”). Police, in other words, are not left to make these value tradeoffs themselves but to apply a judicially sanctioned framework.

3. Police as “Street Magistrates.” — Legality assumes that the central problem for judicial regulation of the police is regulating discretion. 124 See id. at 176–80 (describing how exclusionary remedies are ill-suited to regulating police discretion). Criminal procedure’s answer has been to conceive of police officers as street magistrates, who can be induced to check their own discretion. 125 See id. at 177–79 (“It is enough of a check on police discretion to let the dictates of police efficiency determine under what circumstances and for how long a person may be stopped . . . .”). This assumption is built into the myriad exceptions to the Fourth Amendment’s so-called “warrant preference” and exclusionary rule.

The Fourth Amendment’s warrant preference requires that inferences of criminal wrongdoing “be drawn by a neutral and detached magistrate instead of being judged by the officer engaged in the often competitive enterprise of ferreting out crime.” 126 Johnson v. United States, 333 U.S. 10, 14 (1948). Unlike evenhanded magistrates, the “zealous officer[]” will be inclined to substantially discount constitutional interests. 127 Id. at 13. Thus, the Court developed the Fourth Amendment’s warrant preference. This does not mean that police must always obtain a warrant before conducting a search or seizure. From early on, the Court has acknowledged that it can be impractical for officers to halt an enforcement action and seek a warrant—for example, when in hot pursuit of a quickly fleeing suspect. 128 See Warden, Md. Penitentiary v. Hayden, 387 U.S. 294, 298–99 (1967). Over the years, such exceptions have proliferated and, as many have noted, swallowed the rule. 129 See, e.g., California v. Acevedo, 500 U.S. 565, 581–82 (1991) (Scalia, J., concurring) (“Even before today’s decision, the ‘warrant requirement’ had become so riddled with exceptions that it was basically unrecognizable.”).

Underlying the exceptions is the implicit notion that courts can induce police officers to behave as magistrates in the field. 130 See Nirej Sekhon, Purpose, Policing, and the Fourth Amendment, 107 J. Crim. L. & Criminology 65, 101–02 (2017). The officer need not get the precise facts or law right so long as her ultimate conclusion regarding the permissibility of the search or seizure is correct. Id. That is, with proper standards and incentives, officers will make choices that a magistrate would have approved ex ante had a warrant been applied for. 131 See id. For example, many of the exceptions require that officers must have had the same “quantum of individualized suspicion” that would have been required to obtain a warrant. 132 See Whren v. United States, 517 U.S. 806, 817–18 (1996). The Supreme Court has taken pains to avoid regulating police discretion regarding substantive criminal offenses. See id. at 817–19 (describing “the traditional common-law rule that probable cause justifies a search and seizure”). These choices raise profound issues of race and class equality. The Court has tacitly recognized this but stated that those concerns are best formulated in the language of equal protection, not the Fourth Amendment. Id. at 813 (“We of course agree with petitioners that the Constitution prohibits selective enforcement of the law based on considerations such as race. But the constitutional basis for objecting to intentionally discriminatory application of laws is the Equal Protection Clause, not the Fourth Amendment.”). Ironically, the Court made it nearly impossible for criminal defendants to obtain discovery, let alone succeed on, selective enforcement claims under the Equal Protection Clause. See United States v. Armstrong, 517 U.S. 456, 470 (1996) (requiring “a credible showing of different treatment of similarly situated persons” to make a prima facie case). In the case of exigent circumstances, like an automobile stop, the officer need have had probable cause that a crime occurred to lawfully seize or search. 133 See, e.g., Whren, 517 U.S. at 819 (finding that the officers could search the petitioners because they “had probable cause to believe that petitioners had violated the traffic code”). This requirement posits that officer decisionmaking in the field is equivalent to magistrate decisionmaking in the courtroom. Officers are imagined as surveying those around them for cues of criminality and then calculating whether the cues satisfy the constitutional standard, mimicking the juridical gaze. 134 Id. Of course, police officers’ choices are subject to judicial review ex post in all cases that are prosecuted, with suppression as the remedy for a constitutional violation. As described above in section I.A.1., the threat of exclusion is supposed to induce officers to internalize constitutional conduct rules and, in effect, self-police. 135 See supra notes 93–105 and accompanying text.

One sees a similar street magistrate construction of the police at play in Fourth Amendment excessive force doctrine. As described more fully in Part II, the Constitution permits physical violence to the extent reasonably necessary to deal with an immediate criminal law exigency. 136 See infra notes 235–246 and accompanying text. The doctrine does not demand that police exercise precise judgment regarding what is necessary—that is, the kind of judgment that a magistrate would exercise from the bench. 137 See Graham v. Connor, 490 U.S. 386, 396–97 (1989) (“The calculus of reasonableness must embody allowance for the fact that police officers are often forced to make split-second judgments—in circumstances that are tense, uncertain, and rapidly evolving—about the amount of force that is necessary in a particular situation.”). But the species of judgment that Fourth Amendment doctrine understands police to be capable of is that with which courts are charged: carefully calibrated coercion used against an individual to achieve a legally authorized result. 138 See id. at 396 (“Determining whether the force used to effect a particular seizure is ‘reasonable’ under the Fourth Amendment requires a careful balancing of ‘the nature and quality of the intrusion on the individual’s Fourth Amendment interests’ against the countervailing governmental interests at stake.” (quoting Tennessee v. Garner, 471 U.S. 1, 8 (1985)). The street-magistrate conception of police flows easily from the assumption that criminal justice is an integrated system. 139 See supra notes 84–92 and accompanying text. Police officers are the system’s vanguard attuned to the juridical endgame of producing convictions. 140 See supra notes 84–92 and accompanying text. The line connecting officer to magistrate is sinuous and continuous, even if not always drawn taut. As discussed next, there is little in the history and sociology of municipal policing to validate legality’s three assumptions and much to undermine them.

B. A Short History of Municipal Policing from the Nineteenth Century Until the Present

The problem with legality’s core assumptions is that they are not based upon an accurate portrait of the municipal police. Section I.B.1 describes how, from its advent in the nineteenth century, the municipal police’s primary function has been to control lower socioeconomic classes at the behest of privileged classes. A crime control–based account of policing’s purpose emerged in the United States only in the twentieth century. 141 See infra notes 175–179 and accompanying text. Despite that gloss’s success in shaping public and legal perceptions of the police, section I.B.2 reveals that the municipal police function is still best understood in terms of the kinds of social control that account for the municipal police’s genesis in the nineteenth century.

1.  “Crime Control,” After the Fact. — The municipal police arose in the nineteenth century in tandem with the rise of the modern, industrial metropolis. 142 See Allan Silver, The Demand for Order in Civil Society: A Review of Some Themes in the History of Urban Crime, Police, and Riot, in The Police: Six Sociological Essays 1, 3–4, 7 (David J. Bordua ed., 1967). Both Britain and, through inheritance, the United States had social and legal conventions antagonistic to individuals without fixed vocations and residences. 143 See Lawrence M. Friedman, Crime and Punishment in American History 196, 201 (1993) (“At the very core of the system of criminal justice was a profound distrust of men without settled connections.”). Those conventions came under enormous pressure from nineteenth-century industrialization and urbanization. 144 See id. at 193–94 (describing how urbanization and industrialization led to greater mobility); Silver, supra note 142, at 3–4 (explaining how European countries struggled to deal with the new urban poor). Even more terrifying to the dominant classes than living in a sea of unattached working-class strangers were workers’ regular, semiorganized mob protests. 145 See Silver, supra note 142, at 15–16 (describing how “riots and mobs . . . were . . . often means of protest” that contributed to “the very real fears of privileged and propertied people” in eighteenth- and nineteenth-century London); see also Friedman, supra note 143, at 68–69 (discussing race riots in 19th century American cities).

Protests in industrialized London precipitated the creation of the police agency that would be the progenitor for municipal police agencies in the United States. Robert Peel conceived of the London Metropolitan Police as a professionalized and diffuse force capable of penetrating working-class communities and quickly identifying prospective agitators. 146 See Alex Vitale, The End of Policing 35 (2018); see also Silver, supra note 142, at 11–13 (outlining the ways in which the police were used to “penetrate civil society”). Uniformed police patrol was supposed to blend into the rhythms of urban life. Not coincidentally, Peel’s thinking about the municipal police had taken shape as a tool of British rule in early nineteenth-century Ireland. 147 See Galen Broeker, Rural Disorder and Police Reform in Ireland, 1812–36, at 101–04 (1970) (describing how Peel’s police force ruled Ireland). Peasant uprisings were a major challenge to British rule in nineteenth-century Ireland. 148 Id. at 7. The British government was concerned that local uprisings would metastasize into mass revolt. 149 See id. (noting that “outbreaks of rural violence” were thought to indicate “the beginning of a revolution”). Full-scale military response was both expensive 150 See id. at 37–38 (describing the prohibitive expense of maintaining a military force to quell riots). and blunt, tending to inflame the very rural sentiments it was designed to suppress. 151 See id. at 32. As Chief Secretary of Ireland, Peel took the first steps toward creating a system of professionalized, civil police to interdict and suppress rural disorder. 152 See id. at 102–03 (describing the establishment of the Peace Preservation Force). He would leave Ireland before that vision of a police force could be fully realized, 153 See id. at 104. but his thinking about policing found expression in London’s municipal police. 154 See Vitale, supra note 146, at 35–36; see also Manning, supra note 42, at 76–78 (describing how Peel advocated for and helped establish a full-time police force in London).

Within a generation, Peel’s model had found traction in American industrial cities such as New York, Boston, and Philadelphia, 155 See Vitale, supra note 146, at 36; see also Friedman, supra note 143, at 68–69 (explaining how the London police inspired “American experiments with a standing army of professional law-enforcers”). which were even more chaotic and diverse than London. Large-scale Catholic immigration in the nineteenth century alarmed native Protestant elites. 156 Josh Zeitz, When America Hated Catholics, Politico Mag. (Sept. 23, 2015), https://www.politico.com/magazine/story/2015/09/when-america-hated-catholics-213177 [https://perma.cc/95MZ-MDJX] (describing how “immigrant Catholics faced the brunt of Protestant America’s rage”). They viewed the newcomers as racially inferior and inclined to criminality. 157 See Nicole Hahn Rafter, Creating Born Criminals 118–19 (1997) (describing how the language of “criminality” was used generically to describe new, laboring-class immigrants). But the labor pull created by America’s industrial boom made continued immigration inevitable. 158 See Douglas S. Massey, The Social and Economic Origins of Immigration, 510 Annals Am. Acad. Pol. & Soc. Sci. 60, 61 (1990). This impelled demand for new techniques of social control to address the new arrivals’ threat (perceived and actual) to dominant social interests. 159 See Samuel Walker, Governing the American Police: Wrestling with the Problems of Democracy, 2016 U. Chi. Legal F. 615, 625 (describing how the police served the will of the majority at the expense of the minority). The story was more explicitly tied to racial dominance in the South where the precursor to policing was the slave patrol. See Edward L. Ayers, Vengeance and Justice 83–86 (1984) (explaining how “urban slave patrols” operated in Savannah, Georgia). In a country deeply skeptical of centralized authority, municipal police presented an attractive model. 160 See Fogelson, supra note 47, at 15 (“From the start most Americans had a strong conviction that the police should have an essentially civilian orientation . . . .”); Marx, supra note 33, at 22–23 (“The first municipal police force in the new nation consisted of non-uniformed and unarmed men, reflecting the country’s antimilitary attitudes.”).

Protestant elites’ ability to control the police eroded as Catholic immigrants’ increasing numbers found electoral expression in large cities. 161 See Zeitz, supra note 156 (“Catholics quickly came to forge demographic majorities and supermajorities in American cities . . . .”). Within a generation of the police’s advent, Catholic immigrants and their progeny would capture municipal governments in Boston, New York, and Philadelphia, all of which were tightly bound up with their respective police departments. 162 See Fogelson, supra note 47, at 36–38 (detailing how a change in party led to a change in the police force with “newcomers [securing] a disproportionately large share of the available jobs”).

After gaining traction in northeastern cities, municipal police became a modular form. 163 See Eric H. Monkonnen, Police in Urban America 1860–1920, at 55 (2004) (explaining that “the conspicuously successful Metropolitan Police of London served as a policing model to be adopted when any one of several precipitants occurred”). Police departments quickly sprang up in small- and medium-sized cities throughout the country. 164 See id. This was not in response to a crime wave 165 See id. at 75 (countering the traditional narrative “about crime and the growth of industrial cities”). but part of a broad reform movement focused on the rationalization of municipal government and services. 166 See id. at 55–56 (“The growth of uniformed urban police forces should be seen simply as a part of the growth of urban service bureaucracies.”). A police department became a hallmark of municipal status.

Early police departments were intimately bound up with party politics, and often served at the behest of political machines. Political bosses viewed police department jobs as patronage spoils to be doled out to supporters. 167 Fogelson, supra note 47, at 36–39. There was no training or even minimum qualifications for officers save for allegiance to the local political boss. 168 See id. This encouraged the graft for which police departments in these and other cities became notorious. It was understood that officers would supplement their income with payouts for performing various services—for example, a store owner might pay the beat cop for protection from thieves. 169 See id. (describing the need for regular payoffs to the police). Crime control was not part of the police officer’s job, so much as an occasional consequence of it. 170 See Marx, supra note 33, at 23 (stating that “much of the assistance [the police] rendered had nothing to do with crime”); Walker, supra note 159, at 624–26 (“With respect to fighting crime . . . police in the nineteenth century were utterly ineffective.”).

Beyond “keeping the peace,” and its attendant notion of keeping the dangerous classes in check, little public consensus existed on the police’s specific functions until well into the twentieth century. 171 See Fogelson, supra note 47, at 16 (“From the outset most Americans had only a few vague ideas about what the police should do besides maintain public order.”). Keeping the peace included welfare functions such as sheltering the destitute, which was as much about their welfare as it was about keeping track of them. 172 See Monkonnen, supra note 163, at 86. The police were largely tolerant of criminal misconduct, particularly when they themselves stood to profit from it. 173 See Fogelson, supra note 47, at 37–39 (discussing how police would accept payoffs in exchange for ignoring violations of vice laws). Police were as likely to administer rough justice on the street as they were to make an arrest. 174 See Walker, supra note 159, at 626–27 (“Physical brutality was routine and unpunished.”). Such violence was pervasive and largely unchecked. 175 See id. Occasionally news would percolate out of the working classes and provoke response. From the nineteenth century on, commissions have been periodically convened to investigate police abuses and documented the unhinged and open-ended nature of police violence. See Kerner Report, supra note 7; Report of the Special Committee Appointed to Investigate the Police Department of the City of New York (1895) [hereinafter Lexow Committee Report], reprinted in 1 New York City Police Corruption Investigation Commissions, 1894–1994 (Gabriel J. Chin ed., 1997); Wickersham Commission, supra note 18. The Lexow Committee Report, the first such investigation of the New York City Police Department completed in 1895, referred to such violence as pervasive and brutal. See Lexow Committee Report, supra, at 30. The report documented hearings in which “a stream of witnesses” reported that officers and their superiors used violence to “gratify personal spite and brutal instincts, and to reduce their victims to a condition of servility.” Id. Officers faced no serious repercussions for those abuses. Id. at 31.

The idea of a “crime control” mission developed in the twentieth century to help rationalize the police function. 176 See Fogelson, supra note 47, at 187–88 (discussing reformers who thought the police could not solve all social problems because they needed to spend their time deterring criminal activity); see also Walker, supra note 159, at 628–29 (describing how the police reform movement caused police to stop focusing on “serving political bosses”). This idea would be enormously successful in shaping modern perceptions of the police, securing deep purchase in the public consciousness and police officers’ self-awareness. 177 See Bittner, supra note 20, at 2, 42 (stating that it is commonly thought that police are “a crimefighting agency”); John Van Maanen, Working the Street: A Developmental View of Police Behavior 21, 42–43 (Mass. Inst. of Tech., Working Paper No. 681-73, 1973), https://dspace.mit.edu/bitstream/handle/1721.1/1873/SWP-0681-14451100.pdf [https://perma.cc/E4UJ-MU8E] (describing how officers consider crime control to be “‘real’ police work”). In his canonical treatment of municipal police in the United States, Robert Fogelson posits two broad waves of police reform, both of which emphasized crime control as the police’s central mission. 178 See Fogelson, supra note 47, at 11, 84–88, 190–91. The first was during the Progressive Era and was spearheaded by civic organizations, academics, and the media, who tended to be privileged Protestants leery of Catholic immigrants. 179 See id. at 11, 41, 91–92. These reformers eschewed machine politics in favor of rationalized bureaucracy managed by experts. 180 See id. at 91 (describing how reforms advocated for shifting control “from the precincts to the headquarters”). If policing were to fit that mold, it would need a field in which to have expertise and some institutional goal that it could efficiently pursue. Enter “crime” and “crime control.”

With the second wave of police reform that, according to Fogelson, began in the 1930s and continued until the 1970s, the crime-control rationale for the police function became more deeply entrenched. 181 See id. at 11. Second-wave reformers were police themselves, typically police chiefs with training in police science who reflexively embraced the notion that police were, first and foremost, crime-control agents. 182 See id. at 187–88.

The rise of a crime-control rationale for the police was coeval with its political insulation. In preindustrial America, enforcement and adjudication authority were often coterminous—the constable worked for the justice of the peace. 183 See Monkonnen, supra note 163, at 31; Thomas Y. Davies, Recovering the Original Fourth Amendment, 98 Mich. L. Rev. 547, 623–24 (1999) (describing the close relationship between the constable and the justice of the peace). In some places this arrangement persisted into the twentieth century. See Michael Willrich, City of Courts: Socializing Justice in Progressive Era Chicago 10–12 (2003). Second-wave reformers sought to break the remunerative relationships between police and other branches of government. They pushed political insulation as a way of controlling corruption and increasing the police administration’s control over officers. 184 See Fogelson, supra note 47, at 175–76. These reforms sought to remove political actors from officer hiring and retention and to “strengthen the chief’s hand” in relation to political actors in the management of personnel and other policy choices. 185 See id. Chiefs would enjoy greater protection from termination by mayors, and officer units would be reorganized in favor of centralized, bureaucratic controls as opposed to a territorial, precinct-based orientation that favored capture by ward bosses. 186 See id. at 176–78.

Police departments became more regimented, but also more inward in their orientation. Even though crime control became a more salient feature of police officers’ identity, this did not make them agents of the criminal justice system in the way legality presupposes. Officers became—and remain—preoccupied with avoiding rebuke from police administrators, much less so, if at all, from criminal court judges. 187 See Van Maanen, supra note 177, at 52 (explaining how patrolmen typically “react primarily to departmental direction”). Although much has changed in American policing since the late 1970s, departments’ commitment to uniformed street patrol whose orientation is primarily reactive—that is, responding to quotidian exigencies that arise in their assigned territory—has not. 188 See Mastrofski & Willis, supra note 99, at 83 (explaining how “traditional patrol . . . tend[s] to dominate the work” of the police). American policing’s catchall tradition means that patrol officers will spend their shifts using workaday rules to solve a range of problems, most of which are unanticipated by criminal laws. 189 Bittner, supra note 20, at 42 (“[T]he police have always been forced to justify activities that did not involve law enforcement in the direct sense . . . .”). Ironically, crime control’s conceptual prominence was and continues to be untethered from the realities of routine patrol work. Very little of it can be defined in terms of crime control. 190 See Van Maanen, supra note 177, at 42–43 (describing how patrolmen spend “little of [their] time on the street” performing crime control functions); see also Bittner, supra note 20, at 2, 41–42 (“[A]pproximately one-third of available manpower resources of the police are at any time committed to dealing with crimes and criminals.”). Arrests are only a small part of policework, 191 See, e.g., Van Maanen, supra note 177, at 42–43. so there is little incentive for most officers to imagine their work in terms of a criminal case’s life cycle.

2. The Dangerous Classes as Continuing Preoccupation. — The notion that police are crime-control professionals has found enormous traction in the public consciousness. 192 Manning, supra note 42, at 29–33 (describing how police claim the mandate of preventing crime, which has come to legitimate their social function). Counterintuitively, this can work to the police’s disadvantage. The police have relatively little ability to broadly affect rates of criminal misconduct or victimization. 193 David H. Bayley, Police for the Future 3 (1994) (“The police do not prevent crime. This is one of the best kept secrets of modern life.”). Public perceptions of threat and safety are also susceptible to exogenous shock, such as by an isolated but particularly repugnant crime. 194 Cf. Mark Warr, Public Perceptions and Reactions to Violent Offending and Victimization, in 4 Understanding and Preventing Violence 1, 28–34 (Albert J. Reiss, Jr. & Jeffery A. Roth eds., 1994) (describing how media influences people’s perception of crime). Such shocks are impossible to predict and difficult to tamp down. 195 See id. Policing innovation over the last generation has sought to manage these intractable problems and in this vein has developed techniques of preventive policing—that is, intervening before serious crime occurs. 196 See Manuel A. Utset, Digital Surveillance and Preventive Policing, 49 Conn. L. Rev. 1453, 1457 (2017) (describing how digital policing allows for more preventive policing). Police have used the latest in computer technology to improve their ability to track criminal activity, conduct surveillance, and communicate with one another. 197 See Mastrofski & Willis, supra note 99, at 79–80. Preventive policing has the administrative benefit of being arrest-intensive and of supplying deliverables for increasingly data-oriented police bureaucracies eager to demonstrate their productivity. 198 See Marx, supra note 33, at 8–9, 44–45 (discussing the way preventive techniques became subject to “systematic empirical study and questioning”).

The municipal police’s commitment to preventive policing underscores its continued preoccupation with containing the dangerous classes. One should not take the crime-control gloss on such tactics too literally. Two examples are suggestive: first, police departments’ reliance on specialized units to generate intelligence and arrests; and second, the embrace of quality-of-life policing.

Large municipal departments now have many officers who focus on the use of specific tactics, specific crimes, specific groups of people, or all of the above. 199 See Charles M. Katz & Vincent J. Webb, Policing Gangs in America 12 (2006) (“[S]pecialized units are said to be created in order to focus departmental resources, energy, and skills on specific community problems.”); Vitale, supra note 146, at 50–54, 90, 156–57 (underscoring how municipal police departments have developed different strategies to address drug crimes, homelessness, and gang affiliation). Anti-narcotics and gang-suppression units are examples. Creating anti-gang units has become a common response to a community’s perception that gangs are a local problem. 200 See Katz & Webb, supra note 199, at 9–13 (describing how police departments created gang units in response to public pressure); see also Crank, supra note 87, at 194 (describing how gang units were “created in response to constituency pressures”). These units are supposed to generate intelligence as to gang members’ identities and activities to forestall serious crimes. 201 See Vitale, supra note 146, at 157 (“Gang units tend to take on two main functions: intelligence gathering and street suppression.”). They might, for instance, surveil and photograph known gang members and their associates. 202 See Lynn Langton, Bureau of Justice Statistics, DOJ, Gang Units in Large Local Law Enforcement Agencies, 2007, at 4–6 (2010), https://www.bjs.gov/content/pub/pdf/gulllea07.pdf [https://perma.cc/3444-F8BR] (summarizing a survey of gang surveillance techniques). Once identified, gang members are subject to stricter surveillance and enforcement of all criminal laws on the premise that this preempts serious crimes later on. 203 See Vitale, supra note 146, at 158–61 (describing raids conducted on suspected gang members). The problem is that such units may well be created in the absence of an actual gang problem. 204 See Katz & Webb, supra note 199, at 16–18, 268 (summarizing criminology studies arguing that anti-gang units were created in response to “moral panics” as opposed to actual threats). But see id. at 268 (concluding that actual threats did contribute to the creation of anti-gang units in the authors’ four case studies presented in the book). In one study, researchers concluded that the police department created a gang unit to respond to unwarranted community anxiety regarding gang activity. 205 See Crank, supra note 87, at 193–94. Once created, the unit generated self-reinforcing “intelligence,” tagging minority youth as gang members not because they were gang members but because they looked the part. 206 See id. The unit, in other words, validated the community’s racially-informed anxieties to justify the unit’s own existence.

So-called “broken windows” policing refracts racially informed anxieties into crime-control policy. Broken windows policing has been written about extensively, so only a short account is required here. 207 See Sekhon, Redistributive Policing, supra note 32, at 1203–06 (summarizing literature). The theory is that disorder is criminogenic, signaling public disinterest in a space and thus enabling criminal victimization. 208 See id. Proponents assert that policing low-level crimes that generate “disorder” prevents more serious crimes. 209 See id. There was little empirical evidence supporting the theory when New York City implemented an aggressive version beginning in the 1990s. 210 See Bernard E. Harcourt & Jens Ludwig, Broken Windows: New Evidence from New York City and a Five-City Social Experiment, 73 U. Chi. L. Rev. 271, 272–73 (2006) (noting the lack of evidence supporting broken windows policing). Other cities followed suit. 211 See id. (recognizing that Chicago and Los Angeles also implemented broken windows policing).

The problem is that “disorder” is not an objective fact but heavily shaped by race and class presuppositions. 212 Sekhon, Redistributive Policing, supra note 32, at 1220 (arguing that subjective determinations of “disorderliness” are particularly vulnerable to bias). Broken windows policing focused largely on young people of color for minor pedestrian infractions, having alcohol or marijuana in public, or for nothing at all. 213 See id. at 1207–08. The policing strategy generated huge numbers of stops, arrests, and outstanding warrants for young people of color. 214 Id. (describing an increase in misdemeanor arrests of minorities). Experts are skeptical that the campaign has had meaningful effect on serious crime rates in New York City or elsewhere. 215 See, e.g., id. at 1205.

C. History and Law, Disjointed

This Part has shown the disjunction between criminal procedure’s legality paradigm and the history of municipal police. The former takes police to be crime-control professionals who are part of a system of criminal justice. 216 See supra section I.A.1. Their discretion must be contained because they overvalue crime-control values at the expense of civil liberties. 217 See supra section I.A.2. History suggests that police’s raison d’être is not crime control but rather containing those groups generically believed to be “dangerous.” 218 See supra notes 148–166 and accompanying text; supra section I.B.2. Crime control emerged as a post hoc rationale for the police and has never accurately described the police’s core function. 219 See supra notes 176–180 and accompanying text. Nor are police an integral part of a criminal justice system. Police departments are bureaucratically insulated and sustain an occupational culture that is often insensitive to judicial pronouncement. 220 See supra notes 183–191 and accompanying text; see also Radley Balko, The Increasing Isolation of America’s Police, Wash. Post (May 11, 2015), https://www.washingtonpost.com/news/the-watch/wp/2015/05/11/the-increasing-isolation-of-americas-police/ (on file with the Columbia Law Review) (“It’s hard to think of a profession more sensitive, psychologically isolated, and protective of its own than law enforcement.”).

The Supreme Court is not blind to the disjuncture. In Terry v. Ohio, for example, the Court explicitly noted “the limitations of the judicial function in controlling the myriad daily situations in which policemen and citizens confront each other on the street.” 221 392 U.S. 1, 12 (1968). The exclusionary remedy “is powerless to deter invasions of constitutionally guaranteed rights where the police . . . have no interest in prosecuting.” 222 Id. at 14. These self-chastening observations notwithstanding, the Court extended the exclusionary remedy to evidence seized following an unconstitutional stop and frisk. 223 Id. at 18–19. It also crafted a standard more generous to the police than that used in other contexts. 224 See id. at 19–20 (adopting a reasonableness standard, rather than a probable cause standard). This “reasonable suspicion” standard would, consistent with the Court’s anxieties, impose little serious restraint on the police in the street. 225 Id.; see also id. at 37 (Douglas, J., dissenting) (“The term ‘probable cause’ rings a bell of certainty that is not sounded by phrases such as ‘reasonable suspicion.’”). The Court was correct that many of those interactions do not culminate in arrest. 226 See Eric T. Schneiderman, N.Y. State Office of the Attorney Gen., A Report on Arrests Arising from the New York City Police Department’s Stop-and-Frisk Practices 1, 6 (2013), https://www.ag.ny.gov/pdfs/OAG_REPORT_ON_SQF_PRACTICES_NOV_2013.pdf [https://perma.cc/9JFF-VJBA] (showing that only six percent of stops by the NYPD between 2009 and 2013 resulted in arrests). But, even for those that do, it is relatively easy for officers to manufacture facts following an arrest that satisfy the relaxed standard. 227 See Jeffrey Fagan & Amanda Geller, Following the Script: Narratives of Suspicion in Terry Stops in Street Policing, 82 U. Chi. L. Rev. 51, 73–74, 86 (2015) (“[W]e suspect that police constructed scripts of suspicion that could be tailored and invoked to fit the cosmetic or epidemiological circumstances of a stop.”). Terry should thus be thought of as an ambivalent regulatory gesture at best. In the next section, I argue that this brand of ambivalence is an essential ingredient of constitutional criminal procedure.

II. Sovereignty and the Police

Legality and its attendant notion of crime control, the default theory of municipal policing in legal scholarship, are anemic. The project of describing what the police do in the streets and how it fits into a broader conception of the polity has been neglected. Criminologists have constructed detailed portraits of officers’ occupational realities but have typically left the second part of the question unaddressed. 228 See, e.g., Jonathan Rubinstein, City Police, at ix–xii (1973) (providing a “study of policemen at work”); Van Maanen, supra note 177, at 5–6 (describing the author’s “participant-observation study” of a police department). The work that does address the relationship between street policing and democracy dates from the 1950s, 1960s, and 1970s. See Sklansky, supra note 9, at 39–44. David Sklansky’s Democracy and the Police is the shining exception. Its focus, however, is not police conduct on the streets. Rather, the book tracks the relationship between judicial and scholarly accounts of the police and theories of democracy. See id. at 7–8. Much of the writing done about the municipal police and the state was generated when constitutional criminal procedure was in its incipient stages. See id. at 39–44 (characterizing the work of Jerome Hall, William Westley, Jerome Skolnick, James Q. Wilson, and Herbert Packer). If legality and its presuppositions regarding crime control cannot link policing on the street to a broader account of the state, then what theory can?

Below, this Essay argues that the theoretical framework of “sovereignty” offers a more accurate descriptive account of the municipal police than a legality-based account can. Sovereignty also explains why legality cannot achieve more than an ambivalent relation with the municipal police. Section II.A leads into a theoretical exposition on sovereignty by first illustrating how Fourth Amendment doctrine on excessive force reveals legality’s ambivalence toward the police. Police violence is constitutional if it is reasonably in the service of a discrete crime-control goal. 229 See infra notes 235–244 and accompanying text. By casting violence as incidental to crime control, constitutional doctrine obfuscates violence’s centrality to defining police authority. It is the threat of violence that may be inflicted without an external check that defines police status and authority, 230 See Bittner, supra note 20, at 46. particularly in low-income, minority communities. 231 See infra section II.B.

Section II.B draws on police sociology and urban ethnography to construct a typology of police violence that reveals both its intensity and pervasiveness. Much of this violence is, technically speaking, unconstitutional. 232 See infra notes 261–268 and accompanying text for a discussion of the types of police violence and their constitutionality. But formal and practical constraints ensure that aggrieved individuals cannot bring constitutional claims. 233 See infra notes 247–259 and accompanying text. This means that even though constitutional excessive force principles may apply, they have little effect. This situation describes what constitutional and political theorist Carl Schmitt termed a “state of exception.” 234 See Schmitt, Political Theology, supra note 53, at 12–15. Section II.C argues that the police operate in a state of exception, functioning as “street sovereigns” whose authority can neither be subsumed by legality nor, surprisingly, be wholly divorced from it either. Section II.D uses the street sovereign theory to make sense of plainclothes policing and, in so doing, anticipates criticism of the theory.

A. Excessive Force

The law of excessive force casts police violence as legitimate when it is necessary to eliminate a crime-control exigency and proportional to the emergency presented. Ambivalence arises for two reasons. First, the police usually define the crime-control exigency against which reasonableness is measured. Second, even if the standard could be formally satisfied, there is little practical opportunity to obtain a remedy.

In Graham v. Connor, the Supreme Court held that the Fourth Amendment standard for excessive force permits violence reasonably necessary under the circumstances, “including the severity of the crime at issue, whether the suspect poses an immediate threat to the safety of the officers or others, and whether he is actively resisting arrest or attempting to evade arrest by flight.” 235 490 U.S. 386, 396 (1989) (citing Tennessee v. Garner, 471 U.S. 1, 8–9 (1985)). Graham was exceptional not because of the alleged police violence but because the police did not arrest Graham for committing any alleged offense. 236 Id. at 388–89. After the police beat the diabetic Graham for erratic behavior triggered by an insulin spike, they concluded that he had not violated any law and dropped him off at home. 237 See id. The police might have thought better of this if they had anticipated that Graham would bring a federal case against them. Commentators have long noted that police are wont to allege criminal law violations to protect themselves against discipline and civil liability. More than a generation ago, Paul Chevigny noted the pervasiveness of this practice in New York City. 238 Paul Chevigny, Police Power: Police Abuses in New York City 25–27, 62, 143, 248 (1969) (“An arrest, together with the necessary testimony, is used to cover almost all street-corner abuses.”). The New York City Police Department officers favored charges of “disorderly conduct” and “resisting arrest” to shield themselves against charges of physical abuse. 239 Id. at 25. Not coincidentally, proving such charges typically depends heavily on police officer testimony. 240 See id. at 248–49 (“[Police] testimony is usually effective in covering the abuse for perfectly natural reasons—for example, because there is no one in court to contradict it except another policeman, and he will not do so.”). Since Chevigny’s account, others have noted similar police charging practices in various places. 241 See, e.g., Scott Holmes, Resisting Arrest and Racism—The Crime of “Disrespect,” 85 UMKC L. Rev. 625, 629–31 (2017) (describing reports analyzing the connection between charges of resisting arrest and police misconduct); Stephen M. Ryals, Prosecution of Excessive Force Cases: Practical Considerations, 18 Touro L. Rev. 713, 716–17 (2002) (discussing the “classic pattern” of officers using excessive force and issuing resisting arrest charges “to cover for their misconduct”).

As is true for our legality-based constitutional paradigm generally, Graham v. Connor presupposes that police are, first and foremost, crime-control agents. Effectively performing that function sometimes requires physical force. 242 Graham, 490 U.S. at 396 (citing Terry v. Ohio, 392 U.S. 1, 22–27 (1968)) (recognizing that the authority to use force flows from the authority to make stops). The law thus tolerates police violence 243 See id. at 396 (quoting Johnson v. Glick, 481 F.2d 1028, 1033 (1973)) (noting that not all police violence violates the Fourth Amendment). but requires that it not exceed what a reasonable officer would have used in the same circumstances. 244 Id. at 396–97. But this “totality of the circumstances” inquiry leaves much responsibility with the police, not least because the fact of arrest militates strongly against the juridical machinery taking up an excessive force claim. 245 See supra notes 238–241 and accompanying text.

Chevigny and others paint a picture of crime control (and arrest) as operating in service of police violence, rather than the other way around, as legality presupposes. Arrests for resisting-type offenses insulate police officers in a couple of ways. Because of their symbiotic relationship with the police, 246 See Sekhon, Mass Suppression, supra note 83, at 466. Prosecutors depend on police to bring them work and police depend on prosecutors to vindicate their work. See id.; see also David A. Harris, The Interaction and Relationship Between Prosecutors and Police Officers in the U.S., and How This Affects Police Reform Efforts 2–3 (Univ. of Pittsburgh Sch. of Law Legal Studies Research Paper Series, Working Paper No. 2011-19, 2011), https://www.nlg-npap.org/sites/default/files/DavidHarrisProsecutorandPolice.pdf [https://perma.cc/D8SS-KWYH] (“Police officers conduct the investigation and make arrests, but they are not able to end the case on their own; prosecutors must accept the case and move it from its investigatory phase to a conclusion.”). prosecutors are likely to take charges alleging resistance to officer authority seriously. 247 See, e.g., Alli Maloney, Henry Green, The Columbus Police, and the Jumpout Boys: How Racism Becomes Institutionalized in Police Departments, Pac. Standard Mag. (Feb. 7, 2018), https://psmag.com/social-justice/racism-institutionalized-cpd [https://perma.cc/VEH9-TJNJ] (last updated Oct. 31, 2018) (quoting Columbus, Ohio prosecutors voicing their support for the police in police violence cases); see also Harris, supra note 246, at 7–9 (explaining the variety of factors that prevent zealous prosecution of police misconduct). When an individual has been physically harmed by the police, a criminal charge that ostensibly explains why the police inflicted that harm will militate against administrative or judicial reprimand for the officer. Police departments tend to be reluctant to discipline officers to begin with; 248 See Rachel Moran, Ending the Internal Affairs Farce, 64 Buff. L. Rev. 837, 861–68 (2016) (reviewing discipline practices of various urban police departments). an outstanding criminal case against a complainant is likely to further discourage such discipline. 249 See Chevigny, supra note 238, at 144–45 (explaining that the New York City Review Board “does not hold a hearing until after criminal charges against the complainant have been disposed of”). It will also undercut the complainants’ credibility in any subsequent civil litigation.

An outstanding criminal complaint may even formally extinguish the possibility of bringing civil suit. A federal court must abstain from deciding any case when doing so would undermine the validity of a state criminal case. 250 See Heck v. Humphrey, 512 U.S. 477, 486–87 (1994). The state may also demand waiver of civil claims as a condition of a plea agreement on the underlying criminal charge. 251 See Town of Newton v. Rumery, 480 U.S. 386, 394 (1987). Assuming no such waiver, it is theoretically possible to evade abstention by alleging that the force used was in excess of that required to overcome the individual’s resistance or disorderliness. 252 See G. Todd Butler & Nicholas F. Morisani, Heck, Excessive Force, and the Fifth Circuit, 29 Miss. C. L. Rev. 529, 530–35 (2010) (explaining that Heck does not apply when officers “use more force than necessary in executing the arrest”). But it will be difficult to succeed absent some compelling evidence beyond just the arrestee’s testimony regarding the police violence. 253 Taylor Dolven, Shot by Cops, Smeared in Court: Why It’s So Hard for Victims of Police Abuse to Sue and Win, Vice News (Oct. 30, 2017), https://news.vice.com/en_us/article/pazq57/police-shootings-rule-609 [https://perma.cc/75Z6-FXY9] (describing the challenges faced by plaintiffs with criminal convictions in excessive force cases because “in trials involving police misconduct, an entire case can rest on a victim’s word versus a police officer’s”). And this is the least of the practical challenges preventing a criminal defendant from obtaining a civil remedy for excessive force. 254 See Ryals, supra note 241, at 716–17 (detailing the many challenges that may deter people from succeeding on a civil excessive force charge). The majority of criminal defendants are poor. 255 See Erica J. Hashimoto, Class Matters, 101 J. Crim. L. & Criminology 31, 58 (2011) (“[T]he criminal justice system prosecutes and incarcerates poor people at a much higher rate than non-poor people.”). Finding and paying a lawyer to bring a civil suit against the police will be difficult, if not impossible, particularly for the most common forms of violence, which do not typically result in death or serious injury. 256 Incentives to litigate turn on the likelihood of securing damages and the anticipated size of the award. Accordingly, much of the civil litigation around excessive force involves death and serious injuries. See Darren M. Gelber, Civil Rights Attorney’s Fees: Qualifying as a Prevailing Party Offers No Guarantees, N.J. Law., Feb. 2014, at 30, 31–32 (“If an attorney expects to be paid at the end of a civil rights case, he or she better be pretty certain the fact finder can be convinced to award actual damages in an amount that will justify the full award of fees.”).

Criminal cases, of course, create an opportunity to vindicate Fourth Amendment claims against the police, but only to the extent that evidence is seized as a direct result of the constitutional violation. 257 See Sekhon, Mass Suppression, supra note 83, at 430–31 (explaining how “the Fourth Amendment is ineffective at regulating police” because it is primarily limited to excluding evidence in individual criminal cases). Exclusion of the ill-gotten evidence is the remedy. Instances of excessive force do not generally yield evidence. 258 See Sekhon, Dangerous Warrants, supra note 31, at 993–97. And courts will not exclude an arrested person for whom there is probable cause to make the arrest, even if the seizure was egregiously unconstitutional. 259 See id. (“The simple . . . answer is that exclusion is available for unconstitutionally seized evidence but not for an unconstitutionally seized suspect . . . .”). It is not obvious why courts would exclude ill-gotten evidence, but not an ill-gotten defendant. But neither the Warren Court nor lower courts took the latter prospect seriously. 260 See e.g., United States v. Crews, 445 U.S. 463, 474 (1980) (explaining that a criminal defendant is not “suppressible ‘fruit’”); Brown v. Doe, 2 F.3d 1236, 1243 (2d Cir. 1993) (finding “no authority . . . barring the prosecution of a defendant who was illegally taken into custody”); Matta-Ballesteros v. Henman, 896 F.2d 255, 260 (7th Cir. 1990) (“For the past 100 years, the Supreme Court has consistently held that the manner in which a defendant is brought to trial does not affect the ability of the government to try him.”).

Because constitutional remedies are scarce, particularly when the police allege that the person they injured committed a crime, the police themselves will most often be the final arbiters of violence’s propriety. Before drawing significant conclusions from this, we must identify the ends to which police use violence.

B. The Sociology of Police Violence

This subsection presents a sociologically informed account of how and why police use violence. Police violence can be typed into three overlapping categories that I will label as follows: juridical violence, status-oriented violence, and anomic violence. “Juridical violence,” the dominant conception in legal discourse, finds expression in excessive force doctrine: This is police violence in the service of a discrete crime-control goal. 261 See supra section II.A. “Status-oriented violence” describes the quotidian violence that attends policing in poor minority communities. 262 See infra notes 278–287 and accompanying text. It describes violence meted out for “disrespecting” the police. 263 See infra notes 269–276 and accompanying text. Respect is a complex social dynamic that often has little to do with a discrete crime-control exigency. 264 See infra notes 272–275 and accompanying text. But police understand respect as essential to achieving crime control in the more diffuse sense of “maintaining order” or “keeping the peace.” 265 See infra notes 287–292 and accompanying text. The occupational norms of policing tolerate, and sometimes encourage, status-oriented violence. Finally, “anomic violence” describes violence that does not serve any function beyond the psychological gratification of the officers who inflict it. 266 See infra notes 298–301 and accompanying text.

Fourth Amendment excessive force doctrine is built upon the paradigm of juridical violence. The Constitution authorizes use of force that is reasonably proportional to a crime-control exigency. 267 See supra notes 235–237 and accompanying text. This casts police violence in bureaucratic terms: a tool that trained experts use to accomplish legitimate ends. The law of excessive force casts everything else, likely most police violence, as unconstitutional (even if courts are structurally incapable of taking cognizance of that reality, let alone providing a remedy). 268 See supra notes 238–256 and accompanying text. “Juridical violence” cannot provide, and does not aspire to provide, a sociologically complete account of police violence.

If “crime control” is the dominant currency in courts, “respect” is the dominant currency on the street, and its value is underwritten by violence. 269 See Butler, supra note 4, at 137–38 (discussing the importance of violence as a form of image building). For police officers assigned to poor minority neighborhoods, as for many of the young people who live there, respect buys social standing and personal security, at least in the short term. 270 See Elijah Anderson, Code of the Street: Decency, Violence, and the Moral Life of the Inner City 66–67 (1999); Van Maanen, supra note 177, at 68 (quoting a veteran police officer describing his tendency to retaliate physically to “teach[] [people] a little respect on the street”). Respect is a fiercely rivalrous good, locking the police and the policed in competition. 271 See Anderson, supra note 270, at 66–67 (“In the inner-city environment respect on the street may be viewed as a form of social capital that is very valuable, especially when other forms of capital have been denied or are unavailable.”). Police officers take respect as a core commodity for maintaining order and aggressively tamp down any perceived disrespect lobbed at them. 272 See, e.g., Van Maanen, supra note 177, at 62–64 (“The patrolman realizes . . . that he must delicately manage his appearance on the street, maintaining, as best he can, the respect and fear he feels necessary for him to do his job.”). For young men in poor communities of color, disrespecting the police can be a powerful way of generating respect from their peers on the street. 273 See Victor M. Rios, Punished: Policing the Lives of Black and Latino Boys 125 (2011) (describing a young man challenging police authority “to prove himself to his peers and . . . to intimidate the police officers and avoid future conflict”). While the culture of respect is not the exclusive province of young men, see Anderson, supra note 270, at 67–68, much of the sociological literature focuses on male behavior, as suggested by Victor Rios’s book title. Even Elijah Anderson’s field notes tend to make clear that he was focused on male behavior. See id. at 74, 80, 88, 91. References in this section accordingly reflect the literature’s gender specificity. More so if one then endures physical reprisal by police with a stiff upper lip.

Status-oriented violence reflects the zero-sum nature of respect on the streets. Among the young men who congregate in the street, the willingness to dispense violence and endure it are defining features of the space they inhabit. 274 See Anderson, supra note 270, at 68, 80–91 (narrating the story of a young man proving himself through partaking in violent altercations). In his classic treatment, sociologist Elijah Anderson explained these dynamics in instrumental terms, as techniques for managing the pervasive lack of personal security in poor communities of color. 275 See id. A young man’s willingness to dispense violence, even when the provocation is trifling, may deter future attacks. 276 See id. at 68–79 (“For [violence] to happen, the young man’s life does not always have to be in danger; pride, how he feels about his homies, low feelings, or [getting] the bad end of an altercation may be enough for him . . . to try to avenge the offense.”). Ironically, for that very reason, it may be status enhancing to antagonize someone precisely because he has a reputation for not backing down. 277 See id. at 86–87 (explaining how a young man gained entry to a group by fighting a member with a “strong reputation”). Such status competition need not always end in physical violence—truces are possible 278 See id. (describing the creation of mutual understanding between young men following a fight); David Simon & Ed Burns, The Corner 158–59 (1999) (explaining how “paper bag” diplomacy allows people to violate public alcohol laws without disrespecting the police). —but the possibility of violence always remains just beneath the surface of social interactions. The police operate in this respect-driven street culture with the state at their back.

Police demand respect and, like the young men they police, are willing to inflict violence to secure it. Failure to pay obeisance can prompt a harsh toll. 279 See Rios, supra note 273, at 104 (explaining how police, and other adults, grew frustrated with “everyday acts of resistance” that led them to “abandon empathy for the boys and to apply the toughest sanctions on them”); see also Butler, supra note 4, at 205–07 (suggesting how to avoid formal arrest if stopped on the street). David Simon and Ed Burns, having spent a year observing a poor black neighborhood in West Baltimore, noted the routine ways in which street denizens paid obeisance to the police with a bow of the head or an averted gaze or by simply ceding space when the police claimed it. 280 See Simon & Burns, supra note 278, at 8. These gestures communicate submission to the police. 281 See id. This staved off police violence, but, in turn, marked the individual’s low status in the street hierarchy. In Simon and Burns’s account, it was older, weathered addicts who reflexively deferred to the police, not the younger men who jockeyed aggressively for status. 282 Id. at 66 (“At times, the younger ones senselessly provoke the charge through pride and bluster as no old-timer would . . . until [the officer] is out of the cruiser and swinging the nightstick hard, enraged at being called a bitch by some seventeen-year-old . . . .”). Younger men sometimes went out of their way to disrespect the police, provoking violent police reprisal. 283 Id. at 66, 150–52 (describing various individuals’ violent interactions with the police).

For the police, maintaining respect is essential to maintaining order, which is in turn understood as essential to achieving crime control generally. 284 See Edward Conlon, Blue Blood 80 (2004) (explaining how police see respect as a prerequisite to safety); see also Simon & Burns, supra note 278, at 150 (describing how police would stick to a strict schedule of “sweeping the corners clean” during which time “a routine insult would often result in a mighty ass-kicking”); Van Maanen, supra note 177, at 63 (explaining how officers must maintain a careful balance of “respect and fear”). Poor minority neighborhoods strike the police (and most outside observers) as disorderly and dangerous—without respect, police are unable to quell the disorder and prevent it from metastasizing into complete chaos. 285 See generally Robert J. Sampson & Stephen Raudenbush, Seeing Disorder: Neighborhood Stigma and the Social Construction of “Broken Windows,” 67 Soc. Psychol. Q. 319, 323 (2004) (describing generally the relationship between disorder, race, and social stratification in lower class neighborhoods). This is the “thin blue line” conception of police authority. To be “punked” by a young man on the street is to lose face and thereby incur a significant status injury. 286 See Butler, supra note 4, at 138 (“Being respected is a big deal. Being disrespected is an even bigger deal.”). Police contend that losing face makes “future interactions much more difficult and dangerous” because an emboldened “punk” is less likely to heed police authority. 287 See Conlon, supra note 284, at 80–81; Peter Moskos, Cop in the Hood: My Year Policing Baltimore’s Eastern District 105 (2009). If “swagger and bark” are not enough for an officer to “maintain the edge” in a street confrontation, police occupational norms favor physical force or arrest. 288 Van Maanen, supra note 177, at 64 (“In most threatening situations, the officer attempts to maintain his edge by managing his appearance such that others will believe he is ready . . . for action.”).

Status-oriented violence may secure short-term submission, but at the cost of triggering negative feedback. Police’s willingness to use violence makes them ever more salient foils against which young men on the street enact status-oriented performances of disrespect by talking back, taunting, or otherwise trying to catch the police flatfooted. 289 See Rios, supra note 273, at 125 (describing a young man adjusting his posture and murmuring curse words to show his peers he disrespected the police); Simon & Burns, supra note 278, at 367. Young men lob disrespect even when they know police reprisal is inevitable. 290 See, e.g., Rios, supra note 273, at 106–08, 125 (describing incidents of youth acting recklessly to gain respect); Simon & Burns, supra note 277, at 66 (describing young men challenging police knowing they will be physically attacked). This dynamic can quickly become deadly, as suggested by the number of officer-involved shootings of individuals who have run from the police. 291 See Nirej Sekhon, Blue on Black: An Empirical Analysis of Police Shootings, 54 Am. Crim. L. Rev. 189, 207 (2017) [hereinafter Sekhon, Blue on Black] (“More than a third of the shootings that occurred during or immediately after a foot chase began as provocative encounters.”).

When it comes to disrespect, running is an unambiguously frontal challenge to police authority, particularly if it follows an express command to remain in place. 292 See Simon & Burns, supra note 278, at 85, 119 (describing police beating a person “who had tried to run and had to be punished on principle”). The imperative for the police to give chase is accordingly high, even in the absence of any obvious crime-control exigency. 293 See id. Running is also a way for young men to display their effrontery and street wiles, 294 See Alice Goffman, On the Run: Fugitive Life in an American City 23–26 (2014). even if unrelated to escaping the consequences of having committed a specific criminal offense. The sociological complexity of this dance between police and young men in poor minority neighborhoods is lost when juridical violence is the only frame of reference. And violence is often the outcome when someone runs from the police. 295 See Sekhon, Blue on Black, supra note 291, at 189, 205, 207 (describing the percentage of police shootings stemming from foot chases). My earlier research suggests that foot chases are often a precursor to officer-involved shootings and other nonlethal police violence. 296 Id. This owes something to the effects of a quickened pulse and fear on decisionmaking. 297 Judith P. Andersen & Harri Gustafsberg, A Training Method to Improve Police Use of Force Decision Making: A Randomized Controlled Trial, SAGE Open, Apr.–June 2016, at 1, 1–3 (on file with the Columbia Law Review) (describing how stress negatively affects police decisionmaking around use of force). But disrespect also makes officers angry and that can take on a life of its own, suggesting a third category of violence.

Anomic violence, unlike juridical and status-oriented violence, cannot be characterized as advancing any crime-control goal, even on the police’s own terms. Anomic violence describes blows that are delivered for their own sake, or, at best, to sate an appetite born of frustration, anger, or sadism. Simon and Burns described the pitched frustration that police officers experienced following an anticipated warrant search for narcotics in a West Baltimore stash house that came up dry—the officers vented their frustration by beating up the people in the house. 298 See Simon & Burns, supra note 278, at 367. Jon Gould and Stephen Mastrofski described officers performing a strip search and invasive cavity check on the street after an unconstitutional stop and search failed to yield narcotics evidence. 299 See Jon B. Gould & Stephen D. Mastrofski, Suspect Searches: Assessing Police Behavior Under the U.S. Constitution, 3 Criminology & Pub. Pol’y 315, 350–52 (2004). Simon and Burns also describe a similar public strip search of a bystander who just happened to be present after police officers got “punked” by someone on the street. 300 See Simon & Burns, supra note 278, at 367. Anomic violence evinces a loss of control, a surrender to darker impulses that exceed the ordinary bounds of occupationally sanctioned violence. Of course such violence is unconstitutional. 301 Anomic violence would be impossible to justify under the prevailing reasonableness standard. See Graham v. Connor, 490 U.S. 386, 397 (1989). Most police violence probably is. But for reasons identified in section II.A, the law does not have much effect.

C. The Thin Blue Paradox

Juxtaposing sociological and legal accounts of police violence reveals that legality-based regulation cannot constrain the police consistently with our liberal theories of divided government. The Fourth Amendment prohibits most (if not all) status-oriented violence and all anomic violence, but that prohibition is rarely vindicated. 302 See supra section II.A; see also Graham, 490 U.S. at 397 (describing the contours of the reasonableness standard used to determine use of force violations). The liberal response is to double down on legality, calling for new juridical carrots and sticks to compel police compliance. 303 See supra notes 78–81 and accompanying text. But such proposals inevitably crash on the same rocky shoals: There is no state actor available to police the police, other than the police themselves. 304 See Balko, supra note 220 (describing the entrenched isolation of police).

It is not that police have no relation whatsoever to law. The police themselves (along with most others) understand crime control to be their raison d’etre along with maintaining order more generally. 305 See supra section I.B.1. The “thin blue line” metaphor is typically deployed to make sense of the latter police function. 306 See Victor E. Kappeler, Richard D. Sluder & Geoffrey P. Alpert, Forces of Deviance: Understanding the Dark Side of Policing 95 (2d ed. 1998) (characterizing the “police worldview” as one anchored in notions of a “thin blue line”). The thin blue line is the figurative boundary separating basic order from a kind of primal chaos; the former is, by liberal accounts, a fundamental prerequisite for legislatures, courts, and civil society to function. 307 See id. (“[T]he police are compelled to view disorder, law-breaking, and lack of respect for police authority as enemies of a civilized society.”); see also Schmitt, Political Theology, supra note 53, at 13 (“For a legal order to make sense, a normal situation must exist.”). The “thin blue line” metaphor casts the police as underwriters of that foundational order—the “normal situation,” as it were. 308 Schmitt, Political Theology, supra note 53, at 13. Criminologist Peter Manning has written that “the police role conveys a sense of sacredness or awesome power that lies at the root of political order, and authority, the claims a state makes upon its people for deference to rules, laws, and norms.” Manning, supra note 42, at 21, 106. Peter Manning has accordingly attributed “sacred” and “mythical” qualities to police authority. 309 See Manning, supra note 42, at 45, 94, 106, 280. When considered in light of the discussion of police violence above, the thin blue line suggests a paradox: The police guarantee the possibility of legal order but cannot themselves be made to submit to it.

This paradox is central to the police’s role in the American state. Liberal theories of divided government and legality cannot resolve it, thus requiring we look elsewhere. Schmitt’s and Agamben’s ostensibly illiberal theories of “sovereignty” and the “state of exception” offer a better account of the police than legality. Section II.C.1 describes the theories and section II.C.2 explains how they clarify the nature of police authority and violence.

1. Sovereignty and the State of Exception. — The liberal ideal of divided government takes the state’s sovereign power as divisible with the organs capable of containing one another’s authority. 310 See Martin S. Flaherty, The Most Dangerous Branch, 105 Yale L.J. 1725, 1764 (1996) (discussing the history of American conceptions of divided government and Locke’s and Montesquieu’s contributions). This finds expression in the familiar tropes of checks and balances, legislative supremacy, and judicial review. 311 See Michael J. Teter, Gridlock, Legislative Supremacy, and the Problem of Arbitrary Inaction, 88 Notre Dame L. Rev. 2217, 2217, 2221–22 (2013) (identifying checks and balances as an element of separation of powers). But the discussion of excessive force and police violence above suggests that the police wield a form of extreme discretion that cannot be readily checked by other government actors. 312 See supra sections II.A–.B. That discretion includes the power to suspend legal rights and protections for those seized—a situation akin to a hyperlocalized, street declaration of martial law. 313 See Bittner, supra note 20, at 41 (“[P]olice procedure is defined by the feature that it may not be opposed in its course, and that force can be used if it is opposed.”). In contrast to political liberalism, which tends to view such assertions of power as aberrant, political theology views them as a defining feature of the modern state. 314 But see Nomi Maya Stolzenberg, Political Theology with a Difference, 4 U.C. Irvine L. Rev. 407, 413–14 (2014) (arguing that political theology and political liberalism are not “mutually antagonistic”).

German constitutional theorist Carl Schmitt, writing between World Wars I and II, used the metaphor of “the exception” to describe the power to suspend the legal order. 315 Schmitt, Political Theology, supra note 53, at 5, 12 (“Sovereign is he who decides the exception.”). Schmitt understood the modern state to have a monopoly on the power “to decide,” not to “coerce” or “rule” per se. 316 Id. at 13. In this vein, the essence of “sovereignty” is the power to “decide the exception”—that is, to decide whether there is an extreme emergency that requires suspension of the juridical order. 317 Id. at 12 (“What characterizes an exception is principally unlimited authority, which means the suspension of the entire existing order.”). The sovereign is the actor who decides when such an emergency exists, how to address it, and when the emergency is over. 318 Tracy B. Strong, Foreword to Schmitt, Political Theology, supra note 53, at vii, xiv (2005) (“The nature of the sovereign . . . is the making of a ‘genuine decision.’”). These decisions cannot be predetermined by law in any specific way. 319 Schmitt, Political Theology, supra note 53, at 6–7. Law might, for example, identify who has the power to declare a state of emergency, but emergencies are by definition situational, unexpected, and therefore not amenable to prespecified rules. 320 See id. (explaining that “the precise details of an emergency cannot be anticipated”).

Schmitt critiqued modern states’ obfuscation of sovereign power’s essential nature through “rule of law” tropes. 321 See id. at 38, 48–51 (“[T]here appears a huge cloak-and-dagger drama in which the state acts in many disguises but always the same invisible person.”). Echoing the legal realists who were writing contemporaneously with him, 322 See Or Bassok, The Mysterious Meeting Between Carl Schmitt and Josef Redlich 2–3 (Mar. 1, 2019) (unpublished manuscript) (on file with the Columbia Law Review) (describing Schmitt meeting with an American legal realist); see also David Kennedy & William Fisher III, Canons of American Legal Thought 7 (2006) (describing the tenets of realist thinking). Schmitt noted that norms cannot resolve specific problems without an actor making a decision. 323 Schmitt, Political Theology, supra note 53, at 30. Schmitt believed that sovereign power must have personalistic essence. See id. at 60–61 (critiquing liberalism for ignoring the “personal element”). This reads darkly in light of Schmitt’s affiliation with the Nazi Party in Germany. See Stolzenberg, supra note 314, at 407. Notwithstanding liberalism’s efforts to purge the sovereign of this essence in favor of “rule of law,” 324 See Schmitt, Political Theology, supra note 53, at 21–23, 29 (explaining that liberalism values the law’s power rather than the state’s power). Schmitt believed that the sovereign would inevitably reveal its true nature in the “state of exception.” 325 See id. at 30–31 (describing how an authoritative decisionmaker emerges when decisions are needed).

Using Schmitt’s language of “sovereignty” and “exception,” Giorgio Agamben has more recently argued that both are organizing principles for politics and law in liberal democracies. His understanding of sovereignty turns on the state’s power over human bodies in space—“biopolitics.” 326 Agamben, Homo Sacer, supra note 52, at 174 (understanding the state of exception to entail the “creation of a space in which bare life and the juridical rule enter into a threshold of indistinction”). The account is apt for making sense of the power municipal police wield and why legality never gains more than an ambivalent hold on the police. 327 See id. (inviting the extension of his theory to different bodies and spaces).

Agamben argues that the state of exception is a ubiquitous technique of control in liberal democracies. 328 Agamben, State of Exception, supra note 58, at 2–3 (describing the “creation of a permanent state of emergency” as “one of the essential practices of contemporary states, including so-called democratic ones”). Emergency-occasioned states of exception are the norm and the “‘force’ of law consists in [the] capacity of law to maintain itself in relation to” such exceptions. 329 Agamben, Homo Sacer, supra note 52, at 18. Rule of law requires a zone—a physical and sociopolitical space—within which the order it creates “can have validity.” 330 Id. at 19. Schmitt articulated a crude version of this point when he argued that, to secure its very existence, the state must posit in-group friendship in opposition to an out-group enemy. 331 Carl Schmitt, The Concept of the Political 26–37 (George Schwab trans., 2007) (1932). The enemy may lurk outside or inside the state’s formal borders, but it serves as the foil against which a cohesive and orderly politico-legal community can be posited. 332 See id. (“The phenomenon of the political can be understood only in the context of . . . friend-and-enemy groupings, regardless of the aspects which this possibility implies for morality, aesthetics, and economics.”). Agamben casts the binary as a more nuanced and abstract opposition of exteriority and interiority. Exteriority, and all the chaos that lurks there, is a necessary foil for an interiority defined by juridico-political order. 333 Agamben, Homo Sacer, supra note 52, at 37 (defining “state of exception” as “a complex topological figure in which not only the exception and the rule but also the state of nature and law, outside and inside, pass through one another”). Guarding against that exteriority requires constant vigilance, for it always threatens to penetrate and disrupt the state’s order. 334 This is why Agamben suggests that the state of exception defines modern governance. See Agamben, State of Exception, supra note 58, at 2. That order is forever tethered to its antithesis, owing its very existence to the relation. Agamben defines the “state of exception” as the porous and undulating border that purports to effect the impossible task of separating interiority from exteriority. 335 Agamben, Homo Sacer, supra note 52, at 24–26.

The state of exception is a “border concept” describing a “zone of indifference,” neither within nor without the law. 336 Agamben, State of Exception, supra note 58, at 22–23. The traditional features of legality, law and discretion, do not help make sense of it. It is a zone of extreme discretion recognized by law, but operating outside of it—a zone in which the sovereign acts in response to necessity as defined by it. 337 Agamben, Homo Sacer, supra note 52, at 38; see also Agamben, State of Exception, supra note 58, at 25 (“Necessity is not a source of law, nor does it properly suspend the law; it merely releases a particular case from the literal application of the norm . . . .”). In the state of exception, the law is in force but is not applied. 338 Agamben, State of Exception, supra note 58, at 36–40.

Martial law presents the most intuitive and obvious example of a state of exception. The Third Reich, for example, was constructed based on the Fuhrer’s order suspending the Weimar Constitution pursuant to an express article that allowed for martial law. 339 Id. at 14–15. Agamben notes that democratic states have also come to rely on “permanent state[s] of emergency (though perhaps not declared in the technical sense)” for the purposes of containing “entire categories of citizens who . . . cannot be integrated into the political system.” 340 Id. at 2; see also Agamben, Homo Sacer, supra note 52, at 123 (describing concentration camps as the archetypal modern instantiation of the state of exception and “the hidden paradigm of the political space of modernity”). These are “anomic spaces” within which transgressing the law cannot be distinguished from its enforcement. 341 Agamben, Homo Sacer, supra note 52, at 57. But the state of exception does not just purport to describe the sovereign’s relation to physical space—sovereign power acts on the people within it. 342 Id. at 83–84 (describing the “sovereign sphere” as a place in which “it is permitted to kill without committing homicide”). A defining feature of sovereign power is its capacity to reduce those people to what Agamben calls “bare life.” 343 Id. at 83–84, 88.

The concepts of “bare life” and “the state of exception” are coeval and mutually constitutive. 344 Id. at 85 (“[H]omo sacer names something like the originary ‘political’ relation, which is to say, bare life insofar as it operates in an inclusive exclusion as the referent of the sovereign decision.”). “Bare life” describes the paradoxical figure of a politically constituted, prepolitical human being—a feeling body and sentient consciousness that is without national, deontological, or other status markers of belonging to a political community. 345 Id. at 83, 100. Liberal political theory often takes prepolitical life as a starting point, positing a mythical state of nature (rife with brutalities in the Hobbesian formulation) from which a sovereign power arises. 346 See id. at 37, 105–06. Agamben turns this mythology on its head, arguing that the very notion of prepolitical human life is not just political, but first occurs in a state of exception. 347 Id. at 83, 105–06, 111. Bare life is constituted within the state of exception. 348 See id. at 83 (“[T]he production of bare life is the originary activity of sovereignty.”). Relying on Roman sources, Agamben identifies homo sacer as the ancient prototype of “bare life.” 349 Id. at 99–100. This described a human being who could not be ritually sacrificed, but could nonetheless be killed. 350 Id. at 71–74. In modern parlance, this is a human being stripped of rights and the other markers of citizenship (or comparable status). The paradox lies in the fact that the “stripping” is a juridico–political act requiring sovereign agency. Bare life is simultaneously abandoned by the juridico-political order and tied to it. 351 See id. at 85 (describing “the originary exception in which human life is included in the political order in being exposed to an unconditional capacity to be killed”). Agamben argues that those corralled in “camps” typify bare life: prison camps, refugee camps, and of course, concentration camps. 352 See id. at 174 (explaining how “camps” are created whenever a space is created “in which bare life and juridical rule enter into a threshold of indistinction”). The juridico–political act of corralling people in camps strips them of civic and other status markers, rendering them bare life to be sustained, or not. 353 See id.

In Agamben’s account, the camp and bare life it creates are not aberrations; rather, he argues that they reflect the hidden paradigm of modern governance 354 Id. at 122–23. —bare life finds expression in all corners of the modern state. 355 See id. at 174–75 (explaining that the “line . . . marking the point at which the decision on life becomes a decision on death . . . is now in motion and gradually moving into areas other than political life”). The possibility that any citizen may be reduced to bare life is a foundational condition for the existence of our politico–juridical institutions. 356 See id. at 111, 115 (“The banishment of sacred life is the sovereign nomos that conditions every rule . . . .”). That possibility is more insistently and brutally present in some spaces than in others but is not limited to any of them. 357 See id. at 174 (explaining the creation of camps where “bare life and the juridical rule enter into a threshold of indistinction”). This leads back to police and the thin blue line.

2.  Street Sovereigns. — On the streets, police exercise the de facto power to declare martial law, if only briefly, for specific individuals. 358 See Bittner, supra note 20, at 39–40 (describing how “every conceivable police intervention projects the message that force may be, and may have to be, used to achieve a desired object”). This power is realized most starkly and revealingly in cases of police violence described in section II.A above. Sovereignty, the state of exception, and bare life help elucidate the nature of this authority and constitutional criminal procedure’s ambivalent engagement with it.

The victims of officer-involved killings are an example of Agamben’s notion of “bare life.” These victims could not have been put to death following judicial ritual, but they could be killed. The typical police-shooting victim is not shot while engaging in a capital crime, or perhaps any crime at all. 359 See Sekhon, Blue on Black, supra note 291, at 205 (explaining that perceived gun threats are the most common instigator of police shootings). Officer-involved killings tend to be concentrated in poor minority neighborhoods, 360 See id. at 201. which, in and of themselves, could be considered “states of exception.” 361 See Agamben, Homo Sacer, supra note 52, at 175 (describing spaces that “enter[] into a lasting crisis” as “dislocating localization[s]” that are continuously within a state of exception). In midwestern and northeastern cities, these areas were developed less as neighborhoods than as enclosures for the waves of black refugees escaping the violence of the Jim Crow South. 362 See Kerner Report, supra note 7, at 237–40 (discussing rates of African American migration to northern cities); see also Stuntz, The Collapse, supra note 79, at 15–16 (same). Sociologist Loic Wacquant has argued that these neighborhoods function as tools of social isolation that anticipated mass incarceration. See Loic Wacquant, Urban Outcasts 3–4 (2008) (describing how these spaces “necessitat[e] and elicit[] the . . . deployment of an intrusive and omnipresent police and penal apparatus”). These neighborhoods exist at a distant remove from the rest of the city, despite lying near its geographic center. 363 See Mitchell Duneier, Ghetto: The Invention of a Place 35 (2016). The earlier expression “ghetto” and then “inner city” pithily captured the spatial irony. 364 See id. at x–xii, 70–74 (discussing Horace Cayton and St. Clair Drake’s use of the term). Since the nineteenth century, it has consistently been the municipal police’s task to manage the others who live in a city’s poorest sections, 365 See supra section I.B. whether termed “the ghetto,” or in today’s no-less-controversial parlance, “high-crime areas.” 366 Donald F. Tibbs, From Black Power to Hip Hop: Discussing Race, Policing, and the Fourth Amendment Through the “War on” Paradigm, 15 J. Gender, Race & Just. 47, 67 (2012) (“[O]ne should note that ‘high crime neighborhood’ is code for poor Black ghetto.”).

In high-crime areas, necessity trumps law as the basis for police action. “Necessity” is both the source of and the constraint upon the sovereign’s authority in the state of exception. 367 See Strong, supra note 318, at xi–xvii (“[I]t is the essence of sovereignty both to decide what is an exception and to make the decision appropriate to that exception.”). What defines the high-crime area is not the state of exception’s intensity, but its extensiveness. There, the state of exception is unremitting—the high-crime area, like the ghetto before it, is cast as a recrudescent Hobbesian state of nature, which threatens to rupture its enclosure and penetrate the well-ordered society in which it is embedded (and from which it is simultaneously excluded). 368 Cf. Wacquant, supra note 362, at 51–57 (discussing the formation of violent African American ghettos on Chicago’s South Side). No wonder that war metaphors are so often used to describe these places, which are usually poor minority neighborhoods. 369 See, e.g., Editorial, Chicago’s Great Shame, Chicago’s Crisis: Blood on the Streets, Chi. Trib. (Aug. 6, 2018), https://www.chicagotribune.com/opinion/editorials/ct-edit-violence-chicago-gangs-police-20180806-story.html [https://perma.cc/8R7A-ZHN9](comparing shooting figures to those of a “war zone”); Rory Carroll, In the Breaking Bad City, Trust in the Trigger-Happy Police Has Broken Down, The Guardian (May 10, 2014), https://www.theguardian.com/world/2014/may/10/albuquerque-police-shootings-violence-breaking-bad [https://perma.cc/35CX-XMQN] (describing a part of Albuquerque nicknamed the “War Zone” because of “decades of domestic beatings, gang feuds and drunken brawls”); Michael D’Onofrio, ‘It’s Almost like You’re Living in a War Zone’ After South Philly Slaying, Phila. Trib. (June 19, 2018), https://www.phillytrib.com/news/it-s-almost-like-you-re-living-in-a-war/article_4ab34260-8254-5416-8e4f-ba6b87d87736.html [https://perma.cc/RQ7V-J8U5]; Paige St. John & Adam Ehlmahrek, ‘Nothing Seems to Change’: In the Wake of Stephon Clark Killing, a Demand for Police Reform, L.A. Times (Apr. 1, 2018), https://www.latimes.com/local/lanow/la-me-stephon-clark-peace-20180401-story.html [https://perma.cc/C4PN-TN7D] (“We are living in communities that look like war zones.”).

It was in the mid-twentieth century that crime and war were elided. 370 See Elizabeth Hinton, From the War on Poverty to the War on Crime: The Making of Mass Incarceration in America 55–62 (2016) (discussing the origin of the Johnson Administration’s “War on Crime”). The police have been fighting a “war on crime” for nearly two generations. During the 1960s, the government formally declared this war. 371 See id.              Not coincidentally, the war in Vietnam became a cipher for the lawless desperation of America’s cities. 372 See Michael W. Flamm, Law and Order 51–52 (2005) (“By promoting the idea of victory and hailing the policeman as the ‘frontline soldier in our war against crime,’ Johnson had staked a great deal of his political credibility on a domestic struggle no less dangerous than the foreign conflict in South Vietnam.”). “The ghetto” became shorthand for, among other things, a domestic warzone. 373 See id. at 64.

The police’s use of status-oriented and anomic violence are best understood as techniques of power in a state of exception. The sovereign must assert (and defend) its prerogative to decide above all else. 374 Schmitt, Political Theology, supra note 53, at 5 (defining sovereignty based on an ability to “decide the exception”). Status-oriented violence, as described above, 375 See supra section II.B. is designed to suppress any challenge to that prerogative. The hierarchy of concerns that are reflected in formal criminal law has little practical consequence: Egregious crimes often go overlooked, while small transgressions are subject to harsh treatment. 376 See Jill Leovy, Ghettoside 9 (2015) (arguing that our criminal justice system “is at once oppressive and inadequate”).

Systematized investigation of violent and other serious crimes is often neglected in high-crime areas. 377 See id. Police often favor enforcement of narcotics and other less serious crimes in poor minority neighborhoods, leading Stuntz to observe that these neighborhoods receive too much of the wrong kind of policing and too little of the right kind. See, e.g., William J. Stuntz, Unequal Justice, 121 Harv. L. Rev. 1969, 2033 (2008). But what to an outside observer might seem like impudence is met with fierce police reprisal. Simon and Burns, for example, describe the criminal misadventures of various protagonists in a poor, largely black West Baltimore neighborhood. 378 See generally Simon & Burns, supra note 278 (providing this account). Provided that they remained in their section of the neighborhood, police paid little heed to people with heroin addictions conspicuously carting piles of stolen materials from one place to another. 379 Id. at 187. Even homicides were met with shrugs from the police. 380 Id. at 169, 249; see also Leovy, supra note 376, at 65–66 (recognizing the same phenomenon in South Los Angeles). But staring at the police in the wrong way, talking back (by, say, asserting one’s constitutional rights), or outwardly displaying disrespect could get one killed. 381 See Simon & Burns, supra note 278, at 251 (recounting an incident in which police beat a young man to death for jumping on the hood of a police cruiser). It might not even be a question of doing anything. Just bearing witness to the police stumble or embarrass themselves could precipitate physical violence. 382 See id. at 367 (describing an incident in which police strip-searched a man because their target escaped). This is what it is like to live in a state of exception, where even “the most innocent gesture or the smallest forgetfulness can have the most extreme consequences.” 383 Agamben, Homo Sacer, supra note 52, at 52.

While the police’s sovereign power is realized most starkly in poor minority neighborhoods, it extends to other places. As noted, Bittner defined the police as wielding nonnegotiable coercion. 384 Bittner, supra note 20, at 46. When the police act against a target, that individual has no formal recourse, at least not until some later moment if ever. 385 See id. at 40–42 (“[W]hatever the substance of the task at hand . . . police intervention means above all making use of the capacity and authority to overpower resistance to an attempted solution in the native habitat of the problem.”). According to Bittner, this is just as true in the routine traffic stop as it is in a serious criminal investigation. 386 Id. Bittner’s account dovetails with Agamben’s theory of state power. It is not just that police exercise nonnegotiable coercion but that they wield the power to decide when, where, and in what measure to deploy that coercion. Bittner’s account suggests that individual officers exercise the power to decide according to workaday rules and intuitions that lead them to focus on people who seem “out of place”—who do not conform to what an officer considers normal for the beat. 387 See Rubinstein, supra note 228, at 226 (describing how police identify people who are “out of place”); Megan Stroshine, Geoffrey Alpert & Roger Dunham, The Influence of “Working Rules” on Police Suspicion and Discretionary Decision Making, 11 Police Q. 315, 322 (2008) (analyzing “working rules” based on “identifying persons who would not normally be found in a particular area”); Van Maanen, supra note 177, at 67. For example, race will often be a salient marker of being out of place: Black people in white neighborhoods can expect to be stopped by police, just as white people in black neighborhoods can be. 388 See Stroshine et al., supra note 387, at 322–23.

Agamben noted that “the decision can never be derived from the content of a norm without a remainder.” 389 Agamben, State of Exception, supra note 58, at 36. Bittner’s observations about police occupational culture bear this out. Criminal law and procedure do not determine the heuristics police use to make decisions about whom to target and how to then deal with those individuals. 390 See Bittner, supra note 20, at 11–12 (describing how police rely on discriminatory heuristics “[b]ecause the preponderant majority of police interventions are based on mere suspicion or on merely tentative indications of risk”). Nor can a police target typically use criminal law or procedure norms to limit police power during these encounters. 391 See id. at 33–35 (“To make arrests the police need the kind of discretionary freedom that is difficult to reconcile with those observances of civil rights that are guaranteed in the courtroom.”). One does not have the right to hear, let alone challenge, an officer’s decisions in the field. 392 See Butler, supra note 4, at 205. It can be dangerous even to try. 393 See id. at 206. The most immediate police goal in an encounter is to secure control over the individuals targeted and the immediate surroundings. 394 Bittner, supra note 20, at 40, 46 (proposing that one of the primary activities of the police is using force to gain control). Even when arrest and prosecution are future possibilities, the police’s immediate goal is to bring about a “normal” situation. 395 See Agamben, State of Exception, supra note 58, at 36 (explaining Schmitt’s theory that states of exception are created to restore normalcy). There is a vast temporal and normative gap separating the moment on the street from the hypothetical, future moment when a judge considers the defendant and police’s conduct. In cases of officer-involved homicide, the gap can be wholly subsuming. 396 See Kate Levine, How We Prosecute the Police, 104 Geo. L.J. 745, 768–69 (2016) (discussing how the police are “given the greatest chance of never facing charges”). Again, the “state of exception” is the name for a gap like this—one that “separates the norm from its application in order to make its application possible.” 397 See Agamben, State of Exception, supra note 58, at 36.

Because police violence (and policing generally) occurs in a state of exception, the law purporting to regulate police can only do so ambivalently. Agamben and Schmitt suggest that this is a structural feature of law, not a remediable defect in the law. 398 See supra section II.C.1. American courts cannot turn away from the most salient questions about police power—that would be an abdication. But neither can courts police the police in the way legality presupposes. And so, the constitutional law of excessive force exists in a permanent state of ironic disjuncture from the sociology of police violence. It is in force, but without effect.

D. Plainclothes Policing and Legality’s Subversion

It is not just that constitutional criminal procedure’s legality paradigm fails to describe the police. The police have the power to subvert legality from within; a case study of plainclothes policing as a street enforcement tactic illustrates how. Police have the capacity to induce offenses in their presence, sometimes by committing crimes themselves, and then arrest the civilian offenders. 399 See Joh, supra note 35, at 164–66 (describing how police can be authorized to facilitate and participate in crime to secure arrests). Imagine that because of an actual or perceived wave of muggings, there is political pressure to arrest and convict muggers. One approach would be to investigate past muggings aggressively and to be vigilant for muggings in progress—these reactive approaches typify a street magistrate concept of policing. 400 See supra notes 130–131 and accompanying text. But investigations are time consuming and often fruitless; it is similarly difficult to come upon crimes in progress. 401 See Beene, supra note 33, at 57–58. A department will generate more arrests by deploying plainclothes officers to feign illness with a carelessly exposed wallet peeking out of a pocket or bag. 402 See Street Crime in America, supra note 33, at 11 (statement of Anthony M. Voelker); Beene, supra note 33, at 71–72. Anyone who takes the bait is arrested. In some operations, the police themselves violate substantive criminal laws in order to induce the target. The police might arrange to purchase or sell narcotics or buy stolen goods. 403 See Joh, supra note 35, at 177 (describing the tactics allowed under FBI guidelines). Because the suspect is enticed to approach the disguised officer, there are no search and seizure issues. This makes for a tidy evidentiary record and quick path to conviction.

Inducing a criminal offense to prosecute it has a Kafkaesque quality that suggests a state of exception. The most significant legal constraint on this power to decide is the law of entrapment. Entrapment has not been constitutionalized (or even proceduralized). 404 United States v. Russell, 411 U.S. 423, 430 (1973) (rejecting an argument to constitutionalize entrapment). It is a substantive criminal law defense, 405 See id. at 433 (“Since the defense is not of a constitutional dimension, Congress may address itself to the question and adopt any substantive definition of the defense that it may find desirable.”). and provides broad license to engage in deception so long as the targets are members of groups perceived as criminally predisposed. 406 See infra notes 467–471 and accompanying text. Entrapment brings this Essay full circle. It stitches the sovereign prerogative together with the historical purpose of the municipal police: containing the restive elements of the lower classes. 407 See supra section I.B.2.

This section is as much a response to anticipated criticism as it is a case study. One might object to the argument in the preceding sections as follows: The police are at least sometimes motivated to secure a criminal conviction such that criminal law and procedure’s strictures constrain their behavior as legality presupposes. If this is true enough of the time, police will act less like street sovereigns and more like street magistrates, methodically piecing together facts they understand will later be reviewed by a court. But the example of plainclothes policing suggests the opposite.

1. Plainclothes Policing. — It is ironic that police would doff the most potent symbols of their crime-control mission—badge and uniform—to achieve that mission. Likewise that they would “break[] the law to enforce it.” 408 See generally Joh, supra note 35 (writing about the phenomenon). But both ironies define plainclothes policing.

The use of plainclothes officers for street policing is a relatively recent innovation. 409 Marx, supra note 33, at 7. Before its invention, municipal police long used nonuniformed officers in more circumscribed ways. For example, spying was an important function for those departments in cities, such as New York City in the early twentieth century, where the working classes were organized. 410 Id. at 25. Civilian clothes are supposed to help facilitate rapport with civilian witnesses and allow officers to move undetected in the social milieus in which crimes occur. 411 See, e.g., id. at 18–19 (describing an early example of a police department in Paris using spies to infiltrate criminal environments). Between these two modalities—spying and solving past crimes—were “vice squads” and other forms of undercover work that required infiltration of criminal networks in which transactions were consensual. 412 Id. at 7–8. Such tactics were not, by and large, used against ordinary street crime. Not, at least, until the 1960s. 413 Id. at 7, 14.

During the 1960s, the municipal police were called upon to subdue the fires that raged across American cities. Ironically, the police themselves played a salient role in precipitating much of the unrest that generated those fires. The Kerner Commission convened by President Lyndon Johnson to investigate the pervasive and highly visible civil unrest in poor minority communities concluded that police violence often sparked such incidents. 414 See Kerner Report, supra note 7, at 299–305 (“Police misconduct . . . contributes directly to the risk of civil disorder.”). But this official finding did not gain traction among middle-class whites. 415 See Flamm, supra note 372, at 110 (discussing dissatisfaction with the Kerner Report). As historians have documented, many middle-class whites believed that American cities, and perhaps the entire country, were in a state of emergency. 416 See id. at 102 (describing “white fear of racial violence” and “the atmosphere of anarchy perceived by many middle-class whites”). Whites had been fleeing cities since the 1950s, particularly in the Northeast and Midwest, leaving behind a poorer black population. 417 Myron Orfield, Milliken, Meredith, and Metropolitan Segregation, 62 UCLA L. Rev. 364, 377–78 (2015) (“Between 1950 and 1970, as white flight to the suburbs reached its highest levels, the percentage of blacks more than doubled in most cities.”). The Kerner Commission concluded that the poverty and discrimination that defined blacks’ status in these cities was the kindling for unrest. 418 See Kerner Report, supra note 7, at 203–06. The Civil Rights Movement made the situation more combustible by raising black city dwellers’ expectations without producing significant material change outside the South. 419 See id. at 204, 226–27 (“The expectations aroused by the great judicial and legislative victories of the civil rights movement have led to frustration, hostility and cynicism in the face of the persistent gap between promise and fulfillment.”). Suburban whites viewed the ensuing protests as evidence that the Civil Rights Movement had gone too far. 420 See Flamm, supra note 372, at 101–03 (“The disorders . . . strengthened the white backlash and reduced white support for civil rights and the Great Society.”). Consequently, political support for President Johnson’s ambitious antipoverty programs eroded. 421 See id. at 96 (“The riots . . . dashed the administration’s flagging hopes that the Great Society would generate both social justice and social peace.”). He had pitched this legislation along with civil rights legislation to the public as, in part, cures for crime. 422 See Hinton, supra note 370, at 13–14. But here were the ostensible beneficiaries of those programs destroying their own neighborhoods, in what appeared to whites as wonton ingratitude. 423 See Flamm, supra note 372, at 102 (describing how whites believed that the “unrest was at least in part the product of the welfare state”). It was not whites alone who feared crime and urban unrest. Many black civic leaders in urban settings advocated for more aggressive crime control as well. 424 See James Forman Jr., Locking Up Our Own: Crime and Punishment in Black America 10–11 (2017) (describing how “[s]ome [black officials] displayed tremendous hostility towards perpetrators of crime”); Michael Javen Fortner, The Black Silent Majority: The Rockefeller Drug Laws and the Politics of Punishment 9 (2015) (discussing how “the black silent majority” focused attention on individuals who committed crimes). Street crime was a particularly salient concern for these leaders. 425 See Fortner, supra note 424, at 9.

It was in this volatile context that municipal police departments developed the brand of street-level plainclothes policing that remains prevalent today. 426 See Marx, supra note 33, at 36–37, 40. The goal of generating high-quality arrests required building an evidentiary record that satisfied the essential elements of a crime without running afoul of any constitutional rules that could result in suppression. 427 See id. at 45–48 (explaining how increased protections for criminal defendants helped spur the growth of undercover work). By the late 1960s, the Supreme Court had expanded that set of rules and applied them to street policing. For example, in Terry v. Ohio, the Court held that police must have individualized suspicion that a crime is afoot or may occur before stopping an individual on the street. 428 392 U.S. 1, 21 (1968). Catching someone in the act of committing a street crime requires patience and luck. Decoy and sting operations were designed to solve these problems.

The New York Police Department was among the first to experiment with plainclothes policing as a response to street crime. 429 See Beene, supra note 33, at 29. The department created a specialized plainclothes unit with citywide jurisdiction to interdict street crimes. 430 See Street Crime in America, supra note 33, at 10–11 (statement of Anthony M. Voelker). It also authorized precinct captains to redirect patrol resources to arrest-intensive plainclothes operations. 431 See id. at 11–13 (explaining how police are assigned to specific locations); see also id. at 542 (statement of Emil E. Peters, Chief, Police Department, San Antonio, Texas) (highlighting the creation of a task force aimed at increasing arrests through “inconspicuous surveillance and stakeouts in high crime areas”). The officers themselves devised the specific tactics and ruses they would use to ensnare would-be criminals. 432 See id. at 11 (statement of Anthony M. Voelker). For example, officers came up with the costumes that would make them seem like attractive targets for thieves in a particular location, whether inebriated, elderly, or otherwise vulnerable. 433 See id. (describing tactics used “to blend and . . . decoy”); see also Beene, supra note 33, at 57–65, 69–73 (discussing the “unlimited” number of scenarios police could design). Assignment to one of these units offered far more in the way of adrenaline-fueled adventure than uniformed patrol. 434 See Street Crime in America, supra note 33, at 542 (statement of Emil E. Peters) (“The morale of this squad is exceptional . . . .”). They made more arrests and had higher conviction rates. 435 See Select Comm. on Crime, Street Crime: Reduction Through Positive Criminal Justice Response, H.R. Rep. No. 93-358, at 13–14 (1973) (describing the high rates of arrest and conviction).

Police departments that relied on plainclothes units to conduct decoy operations and stings enthusiastically reported that the tactic successfully reduced street crimes. In testimony before Congress on street crime in 1973, representatives from various urban police departments attested to plainclothes policing’s positive impact on crime reduction. 436 See, e.g., Street Crime in America, supra note 33, at 13 (statement of Anthony M. Voelker); id. at 125 (statement of Warren G. Woodfork, Sergeant, New Orleans, Louisiana Police Department) (describing the successful arrests carried out by plainclothes officers); id. at 543 (statement of Emil E. Peters) (noting a drop in robberies after the initiation of plainclothes operations). But the only real evidence for this appears to have been arrest and (sometimes) conviction data. 437 See id. at 13 (statement of Anthony M. Voelker). The conclusion turned on the assumption that these defendants would necessarily have victimized civilians had they not been caught in decoy operations. It was impossible to know whether that was true for any defendant, and there was good reason to think that the opposite was true. 438 See Marx, supra note 33, at 126–27 (discussing how undercover operations may amplify crime by generating “the idea and motive for the crime” and intimidating or tricking a person to commit an offense they would not otherwise be predisposed to commit); see also Rubinstein, supra note 228, at 367 (suggesting that plainclothes operations merely “drive criminals elsewhere” or “redistribute crime over a wider area”).

That the police would simply generate crime to produce arrests has a tautological quality that suggests a state of exception. The irony is most acute when, in Elizabeth Joh’s words, the police actually “break[] the law to enforce it.” 439 See Joh, supra note 35, at 155, 165–67 (“Undercover officers participate in authorized crimes for a number of different reasons.”). Following the success of street-level decoy operations in the early 1970s, plainclothes units spun more elaborate ruses to generate greater numbers of arrests. For example, police departments began using plainclothes units to carry out fencing decoys to ensnare thieves. 440 See Marx, supra note 33, at 75, 110–11 (describing how police fencing operations flourished). These operations yielded “high-quality” arrests, but they may have increased overall crime. Some have suggested that fencing decoys created a larger resale market for stolen goods and thus induced some to steal who otherwise would not have. 441 See id. at 119–20 (suggesting that fencing operations may actually create a fencing system where one did not previously exist).         

Breaking the law to enforce it has been a defining feature of plainclothes policing in the narcotics context. But as Agamben predicts, the techniques of social control born in states of exception tend to become normalized and permanent. 442 See supra note 329 and accompanying text. And so it has been with plainclothes policing. Buy-busts and reverse buy-busts have been and continue to be an effective technique for generating copious arrests of low-level offenders. 443 See Gail M. Greaney, Crossing the Constitutional Line: Due Process and the Law Enforcement Justification, 67 Notre Dame L. Rev. 745, 745 (1992) (“Undercover operations such as reverse stings contribute substantially to the detection, investigation, and prosecution of crime.” (footnote omitted)). For a description of such an operation, see Bruce A. Jacobs, Dealing Crack 13–14 (1999). As with other decoy operations, these tactics lead to high-quality arrests, but also require that the police violate criminal laws prohibiting the purchase or sale of narcotics—that is, the very laws being enforced. 444 See Joh, supra note 35, at 165–67 (describing the role undercover officers play in facilitating crimes, such as drug trafficking). The law is in full force, but has no effect, at least not on the police. 445 See Agamben, State of Exception, supra note 58, at 39–40 (describing how “a force of law without law” exists in a state of exception).

High-quality arrests and police violence go hand in hand. Plainclothes units not only generate more arrests per capita than patrol units, they also generate more instances of violence. 446 See Hinton, supra note 370, at 191–96; Sekhon, Blue on Black, supra note 291, at 223–24 (analyzing statistics finding that “[n]early 40% of the on-duty shootings [in the reports] involved plainclothes officers”). Studies have suggested that plainclothes officers rack up more shootings and complaints than patrol officers. 447 See Sekhon, Blue on Black, supra note 291, at 223 (describing the large numbers of shootings by plainclothes officers); see also DOJ, Investigation of the Baltimore City Police Department 45 (2016) https://www.justice.gov/crt/file/883296/download [https://perma.cc/NGX4-3LEA] (describing the large number of complaints filed against plainclothes officers). This could be because these officers thrust themselves into confrontational dynamics with civilians. 448 See Sekhon, Blue on Black, supra note 291, at 223–24 (describing how plainclothes officers can participate in “more-aggressive crime interdiction”). Relatedly, the most aggressive officers may self-select for these high-intensity positions. 449 See id. The fearsome nicknames given these squads in poor minority neighborhoods—”jump-out squads,” “knockers” 450 See Fenton & Prudente, supra note 5 (quoting city residents describing plainclothes police as “knockers”); Max Kutner, ‘Jump-Outs’: D.C.’s Scarier Version of ‘Stop-and-Frisk,’ Newsweek (Jan. 16, 2015), https://www.newsweek.com/jump-outs-dcs-scarier-version-stop-and-frisk-300151 [https://perma.cc/HVT6-M38H] (describing D.C. children referring to undercover agents as “jump-out squads”). —attest to the style of policing that prevails in those places. This is policing that exceeds the law’s limits in the law’s name.

2. Entrapment and Sovereign Prerogative. — The law of entrapment is the most direct judicial limitation on plainclothes policing. It, however, ratifies the police’s sovereign power, so long as it is directed at the lowest socioeconomic classes. When using decoy operations to target would-be street thieves, entrapment doctrine suggests police should be careful not to use lures that are too attractive. 451 See, e.g., United States v. Hollingsworth, 27 F.3d 1196, 1200 (7th Cir. 1994) (“[T]he greater the inducement, the weaker the inference that in yielding to it the defendant demonstrated that he was predisposed to commit the crime in question.”). The decoy who dangles $1,000 from a pocket rather than $10 will run a higher risk of an entrapment challenge. 452 See Sheriff v. Hawkins, 752 P.2d 769, 771 (Nev. 1988) (noting that a higher bait amount would increase the likelihood of a successful entrapment defense). Why a prospective miscreant’s “reserve price” for stealing should be of any legal significance is puzzling. That someone might be inclined to pinch $100 or $1,000 but not $10 does not make for a significant moral distinction as much as a class-based one. Wealthier offenders will have a much higher opportunity cost than poor ones.

Entrapment is an affirmative defense, not a constitutionalized procedural rule. 453 See, e.g., United States v. Russell, 411 U.S. 423, 430 (1973) (rejecting an argument to constitutionalize entrapment); Sorrells v. United States, 287 U.S. 435, 448 (1932) (concluding that the entrapment defense is implicitly contained in a federal criminal statute). It is not, in other words, a threshold judicial question of whether police tactics conform to standards of justice. 454 But see Sherman v. United States, 356 U.S. 369, 380 (1958) (Frankfurter, J., concurring in the result) (concluding that entrapment should be a threshold question for federal courts). Rather, it is a factual question whether the defendant’s ostensible guilt is properly attributable to him or to the government. “[T]he government cannot be permitted to contend that [a defendant] is guilty of a crime where the government officials are the instigators of his conduct.” 455 Sorrells, 287 U.S. at 452. A federal entrapment defense exists when the government induced the defendant to break the law and the defendant was not subjectively predisposed to violating the law absent the inducement. 456 See Jacobson v. United States, 503 U.S. 540, 548–49 (1992) (noting that, in cases in which entrapment is raised, it is the government’s burden to “prove beyond reasonable doubt that the defendant was disposed to commit the criminal act” absent the government ruse). Some states use a so-called “objective” standard, requiring that the inducement be objectively unreasonable. 457 See Louis Michael Seidman, The Supreme Court, Entrapment, and Our Criminal Justice Dilemma, 1981 Sup. Ct. Rev. 111, 117–20 [hereinafter Seidman, The Supreme Court] (discussing the Court’s subjective and objective approaches to the entrapment defense). The two approaches are not terribly different, raising problems of circularity that have puzzled legal scholars. 458 Id. But see Richard H. McAdams, The Political Economy of Entrapment, 96 J. Crim. L. & Criminology 107, 118 (2005) (noting that a difference between the subjective and objective approach is the procedural question of who decides the issue, with juries deciding under the former and judges under the latter).

One cannot meaningfully evaluate an individual’s willingness to commit a criminal act without knowing something about the inducement offered to commit the act. 459 Seidman, The Supreme Court, supra note 457, at 118–19. Consider a decoy operation focusing on street theft. Any would-be thief engages in the criminal act to get something. The value of the object determines the would-be thief’s willingness to engage in the criminal act. 460 Id. The objective version of the test does not solve this problem because “reasonableness” requires some socioeconomic reference point. 461 Id. at 119–20 (explaining how the objective version still depends “in large measure on the group to whom the inducement is targeted”). The reserve price for engaging in theft will be different for substance-dependent homeless people than for upper-middle-class teenagers. 462 See id. An individual’s willingness to commit a theft will turn on what inducement they are offered. 463 See id. A would-be thief’s reserve price does not matter when a civilian presents the inducement, 464 See McAdams, supra note 458, at 122–26. so why does it when a police officer presents the inducement? Richard McAdams has argued that it matters for two reasons: 1) because the state has far vaster power to increase inducements than private actors and will use that power against disfavored groups; and 2) because police will likely use decoy tactics inefficiently, maximizing arrests of low-probability offenders. 465 See id. at 153, 162. These concerns would be jurisprudentially salient if the law were actually concerned with deterring those who are likely to commit future crimes against civilians. But current doctrine clearly does not reflect these concerns. 466 See id. at 173–84 (proposing changes to existing law).

Justice Frankfurter accurately characterized the law of entrapment as espousing “the notion that when dealing with the criminal classes anything goes.” 467 Sherman v. United States, 356 U.S. 369, 383 (1958) (Frankfurter, J., concurring in the result). The law of entrapment allows judges and jurors the discretion to grant an acquittal based on their “deeply ingrained presuppositions” about whether a defendant seems like a “criminal.” 468 See Louis Michael Seidman, Entrapment and the “Free Market” for Crime, in Criminal Law Conversations 494, 502 (Paul H. Robinson, Stephen P. Garvey & Kimberly Kessler Ferzan eds., 2009). Entrapment doctrine thus ratifies the historical purpose of the municipal police: controlling the lower socioeconomic classes. 469 See supra section I.B. Entrapment doctrine has little, if anything, to say about the police themselves violating criminal laws to advance that purpose. According to the Supreme Court, it is substantive criminal law that must afford a remedy “[i]f the police engage in illegal activity.” 470 Hampton v. United States, 425 U.S. 484, 490 (1976). There are not likely many prosecutors inclined to pass up “high-quality arrest[s]” to prosecute the police instead. 471 See Street Crime in America, supra note 33, at 22.

Returning to the decoy example with which this section began, the reason that the denomination of the currency matters is because anything too high could induce someone other than a marginal street denizen to commit the theft. Entrapment doctrine authorizes distinctions to be drawn between those who have the “predisposition” to commit crimes and those of upstanding character, despite the fact that such distinctions are ordinarily prohibited by rules of evidence. 472 See Sorrells v. United States, 287 U.S. 435, 451 (1932) (discussing how neither the prosecution nor the defense can object to the introduction of propensity evidence when the entrapment defense is in issue). This suspension of ordinary rules is characteristic of states of exception. 473 See supra section II.C. As Louis Michael Seidman has observed, these judgments will inevitably be informed by broadly shared race and class biases. 474 See Seidman, The Supreme Court, supra note 457, at 151 (“Even if we cannot agree on a theory of blame, the great majority of us can remain secure if we agree on a limited class of people to be blamed.”).

The law of entrapment therefore ratifies the police’s sovereign power so long as this power is exercised within the spaces where the lower classes are understood to congregate. It makes doctrinally explicit what is politically implicit with all forms of sovereign authority: The power to decide the exception has limits, but those limits are tethered to the interests and preferences of dominant groups. 475 See supra section II.C.1.

III. Law and Resistance

The police’s power to skirt and subvert legality does not mean that their power is absolute or impervious to challenge. 476 Agamben, for example, posits the possibility of human activity that “attempt[s] to halt the machine” by revealing “its central fiction,” denying the state of exception as such. See Agamben, State of Exception, supra note 58, at 86–88. Agamben does not, however, develop any systematic program for resistance, but notes that we should be leery of claims that promise return to an original state and questions whether “politics” as conventionally understood can be the vehicle for doing so. Id. at 88. The discussion that follows does not, and really cannot, purport to extrapolate from Agamben’s work. Perhaps that work’s great weakness lies in the great difficulty in imagining (let alone effecting) such extrapolation. Street sovereigns’ power, like that of any other sovereign, can be destabilized, extinguished, and, if not eliminated outright, perhaps remade. 477 See id. at 86–88. And even if the ordinary suppression motion cannot be the vehicle for such challenges, that does not mean that law has no role to play. What one means by “law” has far greater moral range than is captured by the traditional legality paradigm critiqued in preceding sections.

Law is more than just “a system of rules to be observed” and judicial remedies to be sought when those rules are violated. 478 See Robert M. Cover, Nomos and Narrative, 97 Harv. L. Rev. 4, 5 (1983). Law codifies “our visions of alternative futures,” and thus can be a powerful narrative tool for unsettling and transforming repressive institutions. 479 See id. at 9. Robert Cover’s famously expansive view of law places it at the center of social struggles on the streets, not just court cases. By his account, part of law’s value is in enabling principled contestation in various public and private contexts. 480 See id. at 9–10 (“Law is a force, like gravity, through which our worlds exercise an influence upon one another . . . .”). As discussed above, sovereign power often claims law’s authority when acting in derogation of it. 481 See supra section II.C.1. Just confronting the sovereign power with that disjuncture can be a powerful act of resistance. And that need not happen in a courtroom.

Law can enable and empower direct challenges to police injustice on the streets. Writing in 1970, Gary T. Marx and Dane Archer counted over 300 examples of community police patrols taking “the law into their own hands.” 482 See Gary T. Marx & Dane Archer, Citizen Involvement in the Law Enforcement Process: The Case of Community Police Patrols, 15 Am. Behav. Scientist 52, 52 (1971). For a particularly salient example of radical resistance to municipal policing, one might look to the late 1960s when the Black Panthers created armed patrols in Oakland, California. 483 See id. at 54; David Ray Papke, The Black Panther Party’s Narratives of Resistance, 18 Vt. L. Rev. 645, 673–74 (1994); see also Reginald Major, A Panther Is a Black Cat 73 (1971). The Black Panthers embraced an independent vision of racial liberation and self-governance. 484 See Papke, supra note 483, at 673–74 (describing a movement “based upon defending the community against the aggression of the power structure, including the military and the armed might of the police” (internal quotation marks omitted) (quoting Huey P. Newton, Revolutionary Suicide 114 (1973))). It sought to force the municipal police from their community. 485 See id. at 674–76 (“[T]he Panthers were developing plans for community-based policing as an alternative to white policing of the ghetto.”). Armed patrols consisting of neighborhood residents served both as a bulwark against the municipal police and as a substitute for them. 486 See id. Open carry of certain firearms was legal in California at the time. 487 See id. That changed in response to the armed patrols. 488 Thad Morgan, The NRA Supported Gun Control When the Black Panthers Had the Weapons, History (Mar. 22, 2018), https://www.history.com/news/black-panthers-gun-control-nra-support-mulford-act [https://perma.cc/LF96-YSPM]. But in the meantime, the Black Panthers invoked their legal rights to be present in public spaces with firearms to bolster the legitimacy of the armed patrols. 489 See Papke, supra note 483, at 673–74. Just as important as the lawfulness of the Black Panthers’ actions was their claim that the armed patrols were prompted by the lawlessness of the Oakland Police Department, which had a history of brutality in the black community. 490 See id.

Like the Black Panthers, today’s “copwatchers” purport to uphold the rights and values embedded in formal law by observing police–civilian encounters in neighborhoods with histories of police violence. 491 See Jocelyn Simonson, Copwatching, 104 Calif. L. Rev. 391, 392–93 (2016). Copwatcher groups exist across the United States; they typically wear uniforms to distinguish themselves from other civilians and observe police in a manner designed not to interfere with lawful police conduct or civilians’ privacy interests. 492 See id. at 410. The copwatcher’s project is to observe and document police–civilian encounters in order to prevent unlawful police misconduct. 493 See id. at 412. The copwatching movement arose in response to local instances of misconduct, 494 See id. at 409. and as with the Panthers’ armed patrols, its moral anchor for resistance is checking the police’s lawlessness. 495 See id. at 409–412 (“[T]he central idea is to prevent police misconduct . . . .”).

It is impossible to say how many instances of abusive or unlawful policing are averted by such efforts. Intuition suggests that officers who are being observed are more likely to behave lawfully than if they were not, but relatively few police–civilian encounters can actually be observed. 496 See id. at 414–17 (describing the perceived effect copwatching has on police behavior). The long-term political and social consequences of copwatching on the observers themselves and, by extension, the communities to which they belong, are likely more significant. Observers must cultivate their own fluency with constitutional principles and, by so doing, underscore to their fellow citizens the “communal stake” in regulating the police. 497 See id. at 421.

Legal challenges brought in court can have similarly salutary effects on the consciousness of oppressed people and thereby fuel the politics of resistance. In his account of early twentieth-century criminal procedure cases, Michael Klarman noted that the opinions had very little impact on the operation of criminal courts, but they did signal to African Americans in the South that change was possible. 498 Klarman, Racial Origins, supra note 40, at 83. The cases may thus have helped impel the Civil Rights Movement that materialized within a generation of the early due process cases Klarman described. 499 Id.

Klarman’s account suggests litigation can have longer-term consequences beyond the remedy sought in a particular case. This suggests that reformers should be particularly attentive to the possibility of legal cases that might galvanize public opinion against a police practice. The early creators of the municipal police understood them to require the general public’s moral acquiescence. 500 See Silver, supra note 142, at 14 (stating that the police “absolutely required the moral cooperation of civil society”). To the extent that the public accepts the police presence, it may well accept or be indifferent to their abuses. Paul Chevigny noted that “the pattern of police abuses continues because . . . most people in our society do not wish to change the pattern.” 501 Chevigny, supra note 238, at 248; see also Susan Bandes, Patterns of Injustice: Police Brutality in the Courts, 47 Buff. L. Rev. 1275, 1287 (1999) (describing a “desire to avoid knowledge” of police brutality). Legal advocacy can play an instrumental role in disrupting the public’s acquiescence. For example, the civil litigation challenging the New York City Police Department’s aggressive use of stop-and-frisk likely played such a role. 502 See Floyd v. City of New York, 959 F. Supp. 2d 540, 562 (S.D.N.Y. 2013) (finding New York City liable for constitutional violations). The high-profile litigation was years from final resolution when New York City announced that it would cease stop-and-frisk. 503 See Benjamin Weiser & Joseph Goldstein, Mayor Says New York City Will Settle Suits on Stop-and-Frisk Tactics, N.Y. Times (Jan. 30, 2014), https://www.nytimes.com/2014/01/31/nyregion/de-blasio-stop-and-frisk.html (on file with the Columbia Law Review). Mayor-elect Bill de Blasio’s campaign platform included the promise to end it. 504 Id. The litigation likely helped make the issue politically salient for the public and the mayoral candidates. 505 See id.; see also J. David Goodman, As Critics United, Stalled Battle Against Frisking Tactic Took Off, N.Y. Times (Aug. 13, 2013), https://www.nytimes.com/2013/08/14/nyregion/in-a-crescendo-the-citys-crime-policy-changed.html (on file with the Columbia Law Review). For some significant swath of New Yorkers, stop-and-frisk came to appear unjust and they voted accordingly. One should not conclude that this sounded the death knell for stop-and-frisk, let alone for other forms of broken windows policing. 506 See supra notes 207–215 and accompanying text. Much of the data supporting the Floyd litigation were collected by the police themselves. 507 See Floyd v. City of New York, No. 08 Civ. 1034(SAS), 2008 WL 4179210, at *1 (S.D.N.Y. Sept. 10, 2008) (describing UF-250 forms kept in a police database). This, consistent with the argument in Part II, suggests that the police may have the capacity to subvert reform by reporting (or not reporting) selectively.

Similarly, when the Supreme Court held the so-called “fleeing felon” rule unconstitutional in 1985, 508 See Tennessee v. Garner, 471 U.S. 1, 11 (1985). it tapped into a vein of public sentiment that was already critically attuned to police abuses. 509 See Keith Humphreys, How Fatal Shootings by Police Were Cut in Half—And How We Might Do It Again, Wash. Post (July 9, 2018), https://www.washingtonpost.com/news/wonk/wp/2018/07/09/how-fatal-shootings-by-police-were-cut-in-half-and-how-we-might-do-it-again/ (on file with the Columbia Law Review) (describing a “national outcry” around police shootings in the 1980s). Until Tennessee v. Garner was decided, some police departments had authorized officers to shoot any fleeing felony suspect, irrespective of whether that suspect posed any immediate threat to the officers or others. 510 See 471 U.S. at 10–11 (acknowledging that not all “police departments in this country have forbidden the use of deadly force against nonviolent suspects”). Garner held that this blanket authorization to use deadly force violated the Fourth Amendment. 511 See id. In part because of the political salience of police violence at the time, police department policymakers across the country were receptive to operationalizing Garner in their internal rules. 512 See Charles R. Epp, Making Rights Real: Activists, Bureaucrats, and the Creation of the Legalistic State 13–15 (2009) (describing how policy changes in the police’s use of force are achieved through complex interaction among litigation, activists, and police administrators). A wave of police rulemaking that restricted use of force thus followed the Garner opinion. 513 See id. And the number of police shootings decreased, at least for a time. 514 See Humphreys, supra note 509 (describing a dramatic decrease in police killings between 1970 and 1985). That trend has since reversed itself. 515 See id. (describing how police killings began to rise again in the 1990s). The reversal tends to bear out this Essay’s thesis, which suggests that legal victories cannot permanently contain sovereign maneuver, but can only stymie and redirect it for a time.

It is also worth noting the near-misses—litigation that plausibly sought to curtail the use of particular police tactics or circumscribe entire police functions. While these cases did not generate the restrictive rules that police reform advocates might have hoped for, the criminal defendants’ arguments suggest law’s potential to shape alternative futures. For example, had the Supreme Court taken up the call to constitutionalize and expand entrapment principles, 516 See Sherman v. United States, 356 U.S. 369, 380 (1958) (Frankfurter, J., concurring in the result) (concluding that entrapment should be a threshold question for federal courts). decoy and sting tactics would not have seemed like such obvious tools to generate high-quality arrests. 517 See supra notes 426–438 and accompanying text.

Similar examples might be found in litigation that challenged (at least implicitly) American policing’s catchall tradition. 518 See supra notes 42–47 and accompanying text. One can read Brigham City v. Stuart 519 547 U.S. 398 (2006). and New York v. Burger 520 482 U.S. 691 (1987). in this vein. Stuart sought to limit the police’s ability to use community caretaking as a valid rationale for searches that yield evidence of criminal activity. 521 See Megan Pauline Marinos, Comment, Breaking and Entering or Community Caretaking? A Solution to the Overbroad Expansion of the Inventory Search, 22 Geo. Mason U. C.R.L.J. 249, 250–51 (2012) (describing the dispute that predated Stuart). In Stuart, the police claimed that they entered a private home to break up an ongoing fight, and thereby prevent injury. 522 Stuart, 547 U.S. at 400–01. The State (successfully) argued that such searches should be subject to a more relaxed constitutional standard than those involving criminal investigation. 523 See id. at 401–02. The State prevailed notwithstanding that the police arrested Stuart after entering the home and that the State used the evidence the police collected to convict Stuart. 524 See id. at 404–05. Stuart was, in effect, challenging the police’s ability to use their benevolent, public-assistance function as cover for carrying out their coercive function. 525 See id.

Years earlier, Joseph Burger tried to persuade the Court to limit legislatures’ power to authorize such sleights of hand. 526 New York v. Burger, 482 U.S. 691, 693 (1987). Police had searched Burger’s property pursuant to a civil statute authorizing suspicionless searches of junkyards for stolen vehicles and discovered evidence of exactly that. 527 Id. at 693–96. The State then used the evidence to prosecute Burger for criminal law violations. 528 Id. Burger argued that it was unconstitutional for a legislature to equip the police with the authority to conduct suspicionless civil searches subject to less constitutional constraint than criminal searches because police will use the former as cover to do the latter. 529 Id. at 717–18.

Stuart and Burger suggest the extent to which law exists as a vernacular for expressing concepts of justice separate and apart from the outcome in a case. There is nothing to prevent us from thinking and arguing that the Supreme Court decided those cases incorrectly, or from using the rejected arguments to inform and enable resistance in other contexts.


The problems with American policing seem intractable because, at some level, they are. This is not because of a correctible defect in the legality-based regime that purports to regulate American police. Rather, it is because the municipal police wield sovereign power. Courts and scholars have misconceived the municipal police as legality’s agents (and subjects) when history and sociology suggest otherwise. Casting the municipal police in the mold of traditional executive authority has served to obfuscate the quality and quantity of the power that they wield.

Giorgio Agamben’s theory of sovereignty lays the municipal police’s essential nature bare and accounts for why the police can easily evade and coopt legality-based regulation. The police operate in a state of exception that maintains a relationship with law but cannot be restrained by it. This relationship is pithily captured by the notion of a “thin blue line”: Precisely because the police make the juridical order possible, they cannot be entirely subject to it. A theory of the police grounded in sovereignty captures how and why the police deploy violence and generate arrests and situates those practices in a broader account of the American state.

That the police wield sovereign power is at serious odds with our reflexive understanding of the American state as democratic. The account of sovereignty provided here destabilizes any simple opposition between democratic and nondemocratic government, but also suggests law’s importance in enabling and empowering resistance, even if not always in court.